diff options
Diffstat (limited to 'src/utilfuns')
41 files changed, 16414 insertions, 0 deletions
diff --git a/src/utilfuns/Makefile b/src/utilfuns/Makefile new file mode 100644 index 0000000..339f87a --- /dev/null +++ b/src/utilfuns/Makefile @@ -0,0 +1,4 @@ +root := ../.. + +all: + make -C ${root} diff --git a/src/utilfuns/Makefile.am b/src/utilfuns/Makefile.am new file mode 100644 index 0000000..4160db0 --- /dev/null +++ b/src/utilfuns/Makefile.am @@ -0,0 +1,40 @@ + +if HAVE_VSNPRINTF +else +CXXFLAGS += -DNO_VSNPRINTF +endif + +utilfunsdir = $(top_srcdir)/src/utilfuns +libsword_la_SOURCES += $(utilfunsdir)/swobject.cpp +libsword_la_SOURCES += $(utilfunsdir)/utilstr.cpp +libsword_la_SOURCES += $(utilfunsdir)/utilxml.cpp +libsword_la_SOURCES += $(utilfunsdir)/swunicod.cpp +libsword_la_SOURCES += $(utilfunsdir)/swversion.cpp +libsword_la_SOURCES += $(utilfunsdir)/swbuf.cpp +libsword_la_SOURCES += $(utilfunsdir)/ftpparse.c +libsword_la_SOURCES += $(utilfunsdir)/url.cpp + +if INTERNALFTPLIB +ftpsrc = $(utilfunsdir)/ftplib.c +else +ftpsrc = +endif + +if ZLIB +UNTGZ = $(utilfunsdir)/zlib/untgz.c +else +UNTGZ = +endif + +libsword_la_SOURCES += $(UNTGZ) +libsword_la_SOURCES += $(ftpsrc) + +if MINGW +SWREGEX = $(utilfunsdir)/regex.c +else +SWREGEX = +endif +libsword_la_SOURCES += $(SWREGEX) + + +libsword_la_SOURCES += $(utilfunsdir)/roman.cpp diff --git a/src/utilfuns/ftplib.c b/src/utilfuns/ftplib.c new file mode 100644 index 0000000..6f94806 --- /dev/null +++ b/src/utilfuns/ftplib.c @@ -0,0 +1,1355 @@ +/***************************************************************************/ +/* ftplib.c - callable ftp access routines */ +/* Copyright (C) 1996-2000 Thomas Pfau, pfau@cnj.digex.net */ +/* 73 Catherine Street, South Bound Brook, NJ, 08880 */ +/* */ +/* This library is free software; you can redistribute it and/or */ +/* modify it under the terms of the GNU Library General Public */ +/* License as published by the Free Software Foundation; either */ +/* version 2 of the License, or (at your option) any later version. */ +/* */ +/* This library is distributed in the hope that it will be useful, */ +/* but WITHOUT ANY WARRANTY; without even the implied warranty of */ +/* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU */ +/* Library General Public License for more details. */ +/* */ +/* You should have received a copy of the GNU Library General Public */ +/* License along with this progam; if not, write to the */ +/* Free Software Foundation, Inc., 59 Temple Place - Suite 330, */ +/* Boston, MA 02111-1307, USA. */ +/* */ +/***************************************************************************/ +// changes made by Lorn Potter <llornkcor@handhelds.org> +// +#if defined(__unix__) || defined(__VMS) +#include <unistd.h> +#endif +#if defined(_WIN32) +#include <windows.h> +#endif + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <errno.h> +#include <ctype.h> + +#if defined(__unix__) + +#include <sys/types.h> +#include <unistd.h> +#include <fcntl.h> +#include <sys/time.h> +#include <sys/types.h> +#include <sys/socket.h> +#include <netinet/in.h> +#include <netdb.h> +#include <arpa/inet.h> + +#elif defined(VMS) + +#include <types.h> +#include <socket.h> +#include <in.h> +#include <netdb.h> +#include <inet.h> + +#elif defined(_WIN32) + +#include <winsock.h> + +#endif + +#define BUILDING_LIBRARY +#include "ftplib.h" + +#if defined(_WIN32) +#define SETSOCKOPT_OPTVAL_TYPE (const char *) +#else +#define SETSOCKOPT_OPTVAL_TYPE (void *) +#endif + +#define FTPLIB_BUFSIZ 8192 +#define ACCEPT_TIMEOUT 10 + +#define FTPLIB_CONTROL 0 +#define FTPLIB_READ 1 +#define FTPLIB_WRITE 2 + +#if !defined FTPLIB_DEFMODE +#define FTPLIB_DEFMODE FTPLIB_PASSIVE +#endif + +struct NetBuf { + char *cput,*cget; + int handle; + int cavail,cleft; + char *buf; + int dir; + netbuf *ctrl; + netbuf *data; + int cmode; + struct timeval idletime; + FtpCallback idlecb; + void *idlearg; + int xfered; + int cbbytes; + int xfered1; + char response[256]; +}; + +static char *version = +"ftplib Release 3.1-1 9/16/00, copyright 1996-2000 Thomas Pfau"; + +GLOBALDEF int ftplib_debug = 0; + +#if defined(__unix__) || defined(VMS) +#define net_read read +#define net_write write +#define net_close close +#elif defined(_WIN32) +#define net_read(x,y,z) recv(x,y,z,0) +#define net_write(x,y,z) send(x,y,z,0) +#define net_close closesocket +#endif + +#if defined(NEED_MEMCCPY) +/* + * VAX C does not supply a memccpy routine so I provide my own + */ +void *memccpy(void *dest, const void *src, int c, size_t n) +{ + int i=0; + const unsigned char *ip=src; + unsigned char *op=dest; + + while (i < n) + { + if ((*op++ = *ip++) == c) + break; + i++; + } + if (i == n) + return NULL; + return op; +} +#endif +#if defined(NEED_STRDUP) +/* + * strdup - return a malloc'ed copy of a string + */ +char *strdup(const char *src) +{ + int l = strlen(src) + 1; + char *dst = malloc(l); + if (dst) + strcpy(dst,src); + return dst; +} +#endif + +/* + * socket_wait - wait for socket to receive or flush data + * + * return 1 if no user callback, otherwise, return value returned by + * user callback + */ +static int socket_wait(netbuf *ctl) +{ + fd_set fd,*rfd = NULL,*wfd = NULL; + struct timeval tv; + int rv = 0; + if ((ctl->dir == FTPLIB_CONTROL) || (ctl->idlecb == NULL)) + return 1; + if (ctl->dir == FTPLIB_WRITE) + wfd = &fd; + else + rfd = &fd; + FD_ZERO(&fd); + do + { + FD_SET(ctl->handle,&fd); + tv = ctl->idletime; + rv = select(ctl->handle+1, rfd, wfd, NULL, &tv); + if (rv == -1) + { + rv = 0; + strncpy(ctl->ctrl->response, strerror(errno), + sizeof(ctl->ctrl->response)); + break; + } + else if (rv > 0) + { + rv = 1; + break; + } + } + while ((rv = ctl->idlecb(ctl, ctl->xfered, ctl->idlearg))); + return rv; +} + +/* + * read a line of text + * + * return -1 on error or bytecount + */ +static int readline(char *buf,int max,netbuf *ctl) +{ + int x,retval = 0; + char *end,*bp=buf; + int eof = 0; + + if ((ctl->dir != FTPLIB_CONTROL) && (ctl->dir != FTPLIB_READ)) + return -1; + if (max == 0) + return 0; + do + { + if (ctl->cavail > 0) + { + x = (max >= ctl->cavail) ? ctl->cavail : max-1; + end = memccpy(bp,ctl->cget,'\n',x); + if (end != NULL) + x = end - bp; + retval += x; + bp += x; + *bp = '\0'; + max -= x; + ctl->cget += x; + ctl->cavail -= x; + if (end != NULL) + { + bp -= 2; + if (strcmp(bp,"\r\n") == 0) + { + *bp++ = '\n'; + *bp++ = '\0'; + --retval; + } + break; + } + } + if (max == 1) + { + *buf = '\0'; + break; + } + if (ctl->cput == ctl->cget) + { + ctl->cput = ctl->cget = ctl->buf; + ctl->cavail = 0; + ctl->cleft = FTPLIB_BUFSIZ; + } + if (eof) + { + if (retval == 0) + retval = -1; + break; + } + if (!socket_wait(ctl)) + return retval; + if ((x = net_read(ctl->handle,ctl->cput,ctl->cleft)) == -1) + { + perror("read"); + retval = -1; + break; + } + if (x == 0) + eof = 1; + ctl->cleft -= x; + ctl->cavail += x; + ctl->cput += x; + } + while (1); + return retval; +} + +/* + * write lines of text + * + * return -1 on error or bytecount + */ +static int writeline(char *buf, int len, netbuf *nData) +{ + int x, nb=0, w; + char *ubp = buf, *nbp; + char lc=0; + + if (nData->dir != FTPLIB_WRITE) + return -1; + nbp = nData->buf; + for (x=0; x < len; x++) + { + if ((*ubp == '\n') && (lc != '\r')) + { + if (nb == FTPLIB_BUFSIZ) + { + if (!socket_wait(nData)) + return x; + w = net_write(nData->handle, nbp, FTPLIB_BUFSIZ); + if (w != FTPLIB_BUFSIZ) + { + printf("net_write(1) returned %d, errno = %d\n", w, errno); + return(-1); + } + nb = 0; + } + nbp[nb++] = '\r'; + } + if (nb == FTPLIB_BUFSIZ) + { + if (!socket_wait(nData)) + return x; + w = net_write(nData->handle, nbp, FTPLIB_BUFSIZ); + if (w != FTPLIB_BUFSIZ) + { + printf("net_write(2) returned %d, errno = %d\n", w, errno); + return(-1); + } + nb = 0; + } + nbp[nb++] = lc = *ubp++; + } + if (nb) + { + if (!socket_wait(nData)) + return x; + w = net_write(nData->handle, nbp, nb); + if (w != nb) + { + printf("net_write(3) returned %d, errno = %d\n", w, errno); + return(-1); + } + } + return len; +} + +/* + * read a response from the server + * + * return 0 if first char doesn't match + * return 1 if first char matches + */ +static int readresp(char c, netbuf *nControl) +{ + char match[5]; + if (readline(nControl->response,256,nControl) == -1) + { + perror("Control socket read failed"); + return 0; + } + if (ftplib_debug > 1) + fprintf(stderr,"%s",nControl->response); + if (nControl->response[3] == '-') + { + strncpy(match,nControl->response,3); + match[3] = ' '; + match[4] = '\0'; + do + { + if (readline(nControl->response,256,nControl) == -1) + { + perror("Control socket read failed"); + return 0; + } + if (ftplib_debug > 1) + fprintf(stderr,"%s",nControl->response); + } + while (strncmp(nControl->response,match,4)); + } + if (nControl->response[0] == c) + return 1; + return 0; +} + +/* + * FtpInit for stupid operating systems that require it (Windows NT) + */ +GLOBALDEF void FtpInit(void) +{ +#if defined(_WIN32) + WORD wVersionRequested; + WSADATA wsadata; + int err; + wVersionRequested = MAKEWORD(1,1); + if ((err = WSAStartup(wVersionRequested,&wsadata)) != 0) + fprintf(stderr,"Network failed to start: %d\n",err); +#endif +} + +/* + * FtpLastResponse - return a pointer to the last response received + */ +GLOBALDEF char *FtpLastResponse(netbuf *nControl) +{ + if ((nControl) && (nControl->dir == FTPLIB_CONTROL)) + return nControl->response; + return NULL; +} + +/* + * FtpConnect - connect to remote server + * + * return 1 if connected, 0 if not + */ +GLOBALDEF int FtpConnect(const char *host, netbuf **nControl) +{ + int sControl, stat, flags, oldflags; + struct sockaddr_in sin; + struct hostent *phe; + struct servent *pse; + int on=1; + netbuf *ctrl; + char *lhost; + char *pnum; + struct timeval tv; + fd_set wr; + + memset(&sin,0,sizeof(sin)); + sin.sin_family = AF_INET; + lhost = strdup(host); + pnum = strchr(lhost,':'); + if (pnum == NULL) + { +#if defined(VMS) + sin.sin_port = htons(21); +#else + if ((pse = getservbyname("ftp","tcp")) == NULL) + { + perror("getservbyname"); + return 0; + } + sin.sin_port = pse->s_port; +#endif + } + else + { + *pnum++ = '\0'; + if (isdigit(*pnum)) + sin.sin_port = htons(atoi(pnum)); + else + { + pse = getservbyname(pnum,"tcp"); + sin.sin_port = pse->s_port; + } + } + if ((sin.sin_addr.s_addr = inet_addr(lhost)) == -1) + { + if ((phe = gethostbyname(lhost)) == NULL) + { + perror("gethostbyname"); + return 0; + } + + memcpy((char *)&sin.sin_addr, phe->h_addr, phe->h_length); + + } + free(lhost); + + sControl = socket(PF_INET, SOCK_STREAM, IPPROTO_TCP); + if (sControl == -1) + { + perror("socket"); + return 0; + } + + if ( setsockopt(sControl,SOL_SOCKET,SO_REUSEADDR, + SETSOCKOPT_OPTVAL_TYPE &on, sizeof(on)) == -1) + { + perror("setsockopt"); + net_close(sControl); + return 0; + } + +#if defined(_WIN32) + if (connect(sControl, (struct sockaddr *)&sin, sizeof(sin)) == -1) + { + perror("connect"); + net_close(sControl); + return 0; + } +#else + //set nonblocking for connection timeout + flags = fcntl( sControl, F_GETFL,0); + oldflags=flags; + fcntl( sControl, F_SETFL, O_NONBLOCK|flags); + + stat=connect( sControl, (struct sockaddr *)&sin, sizeof(sin)); + if (stat < 0) + { + if (errno != EWOULDBLOCK && errno != EINPROGRESS) + { + perror("connect"); + net_close(sControl); + return 0; + } + } + + FD_ZERO(&wr); + FD_SET( sControl, &wr); + + tv.tv_sec = ACCEPT_TIMEOUT; + tv.tv_usec = 0; + + stat = select(sControl+1, 0, &wr, 0, &tv); + + if (stat < 1) + { + // time out has expired, + // or an error has ocurred + perror("timeout"); + net_close(sControl); + return 0; + } + + printf("connected\n"); + + //set original flags + fcntl( sControl, F_SETFL, oldflags); +#endif + + ctrl = calloc(1,sizeof(netbuf)); + if (ctrl == NULL) + { + perror("calloc"); + net_close(sControl); + return 0; + } + ctrl->buf = malloc(FTPLIB_BUFSIZ); + if (ctrl->buf == NULL) + { + perror("calloc"); + net_close(sControl); + free(ctrl); + return 0; + } + ctrl->handle = sControl; + ctrl->dir = FTPLIB_CONTROL; + ctrl->ctrl = NULL; + ctrl->cmode = FTPLIB_DEFMODE; + ctrl->idlecb = NULL; + ctrl->idletime.tv_sec = ctrl->idletime.tv_usec = 0; + ctrl->idlearg = NULL; + ctrl->xfered = 0; + ctrl->xfered1 = 0; + ctrl->cbbytes = 0; + if (readresp('2', ctrl) == 0) + { + net_close(sControl); + free(ctrl->buf); + free(ctrl); + return 0; + } + *nControl = ctrl; + return 1; +} + +/* + * FtpOptions - change connection options + * + * returns 1 if successful, 0 on error + */ +GLOBALDEF int FtpOptions(int opt, long val, netbuf *nControl) +{ + int v,rv=0; + switch (opt) + { + case FTPLIB_CONNMODE: + v = (int) val; + if ((v == FTPLIB_PASSIVE) || (v == FTPLIB_PORT)) + { + nControl->cmode = v; + rv = 1; + } + break; + case FTPLIB_CALLBACK: + nControl->idlecb = (FtpCallback) val; + rv = 1; + break; + case FTPLIB_IDLETIME: + v = (int) val; + rv = 1; + nControl->idletime.tv_sec = v / 1000; + nControl->idletime.tv_usec = (v % 1000) * 1000; + break; + case FTPLIB_CALLBACKARG: + rv = 1; + nControl->idlearg = (void *) val; + break; + case FTPLIB_CALLBACKBYTES: + rv = 1; + nControl->cbbytes = (int) val; + break; + } + return rv; +} + +/* + * FtpSendCmd - send a command and wait for expected response + * + * return 1 if proper response received, 0 otherwise + */ +static int FtpSendCmd(const char *cmd, char expresp, netbuf *nControl) +{ + char buf[256]; + if (nControl->dir != FTPLIB_CONTROL) + return 0; + if (ftplib_debug > 2) + fprintf(stderr,"%s\n",cmd); + if ((strlen(cmd) + 3) > sizeof(buf)) + return 0; + sprintf(buf,"%s\r\n",cmd); + if (net_write(nControl->handle,buf,strlen(buf)) <= 0) + { + perror("write"); + return 0; + } + return readresp(expresp, nControl); +} + +/* + * FtpLogin - log in to remote server + * + * return 1 if logged in, 0 otherwise + */ +GLOBALDEF int FtpLogin(const char *user, const char *pass, netbuf *nControl) +{ + char tempbuf[64]; + + if (((strlen(user) + 7) > sizeof(tempbuf)) || + ((strlen(pass) + 7) > sizeof(tempbuf))) + return 0; + sprintf(tempbuf,"USER %s",user); + if (!FtpSendCmd(tempbuf,'3',nControl)) + { + if (nControl->response[0] == '2') + return 1; + return 0; + } + sprintf(tempbuf,"PASS %s",pass); + return FtpSendCmd(tempbuf,'2',nControl); +} + +/* + * FtpOpenPort - set up data connection + * + * return 1 if successful, 0 otherwise + */ +static int FtpOpenPort(netbuf *nControl, netbuf **nData, int mode, int dir) +{ + int sData; + union { + struct sockaddr sa; + struct sockaddr_in in; + } sin; + struct linger lng = { 0, 0 }; + unsigned int l; + int on=1; + netbuf *ctrl; + char *cp; + unsigned int v[6]; + char buf[256]; + + if (nControl->dir != FTPLIB_CONTROL) + return -1; + if ((dir != FTPLIB_READ) && (dir != FTPLIB_WRITE)) + { + sprintf(nControl->response, "Invalid direction %d\n", dir); + return -1; + } + if ((mode != FTPLIB_ASCII) && (mode != FTPLIB_IMAGE)) + { + sprintf(nControl->response, "Invalid mode %c\n", mode); + return -1; + } + l = sizeof(sin); + if (nControl->cmode == FTPLIB_PASSIVE) + { + memset(&sin, 0, l); + sin.in.sin_family = AF_INET; + if (!FtpSendCmd("PASV",'2',nControl)) + return -1; + cp = strchr(nControl->response,'('); + if (cp == NULL) + return -1; + cp++; + sscanf(cp,"%u,%u,%u,%u,%u,%u",&v[2],&v[3],&v[4],&v[5],&v[0],&v[1]); + sin.sa.sa_data[2] = v[2]; + sin.sa.sa_data[3] = v[3]; + sin.sa.sa_data[4] = v[4]; + sin.sa.sa_data[5] = v[5]; + sin.sa.sa_data[0] = v[0]; + sin.sa.sa_data[1] = v[1]; + } + else + { + if (getsockname(nControl->handle, &sin.sa, &l) < 0) + { + perror("getsockname"); + return 0; + } + } + sData = socket(PF_INET,SOCK_STREAM,IPPROTO_TCP); + if (sData == -1) + { + perror("socket"); + return -1; + } + if (setsockopt(sData,SOL_SOCKET,SO_REUSEADDR, + SETSOCKOPT_OPTVAL_TYPE &on,sizeof(on)) == -1) + { + perror("setsockopt"); + net_close(sData); + return -1; + } + if (setsockopt(sData,SOL_SOCKET,SO_LINGER, + SETSOCKOPT_OPTVAL_TYPE &lng,sizeof(lng)) == -1) + { + perror("setsockopt"); + net_close(sData); + return -1; + } + if (nControl->cmode == FTPLIB_PASSIVE) + { + if (connect(sData, &sin.sa, sizeof(sin.sa)) == -1) + { + perror("connect"); + net_close(sData); + return -1; + } + } + else + { + sin.in.sin_port = 0; + if (bind(sData, &sin.sa, sizeof(sin)) == -1) + { + perror("bind"); + net_close(sData); + return 0; + } + if (listen(sData, 1) < 0) + { + perror("listen"); + net_close(sData); + return 0; + } + if (getsockname(sData, &sin.sa, &l) < 0) + return 0; + sprintf(buf, "PORT %d,%d,%d,%d,%d,%d", + (unsigned char) sin.sa.sa_data[2], + (unsigned char) sin.sa.sa_data[3], + (unsigned char) sin.sa.sa_data[4], + (unsigned char) sin.sa.sa_data[5], + (unsigned char) sin.sa.sa_data[0], + (unsigned char) sin.sa.sa_data[1]); + if (!FtpSendCmd(buf,'2',nControl)) + { + net_close(sData); + return 0; + } + } + ctrl = calloc(1,sizeof(netbuf)); + if (ctrl == NULL) + { + perror("calloc"); + net_close(sData); + return -1; + } + if ((mode == 'A') && ((ctrl->buf = malloc(FTPLIB_BUFSIZ)) == NULL)) + { + perror("calloc"); + net_close(sData); + free(ctrl); + return -1; + } + ctrl->handle = sData; + ctrl->dir = dir; + ctrl->idletime = nControl->idletime; + ctrl->idlearg = nControl->idlearg; + ctrl->xfered = 0; + ctrl->xfered1 = 0; + ctrl->cbbytes = nControl->cbbytes; + if (ctrl->idletime.tv_sec || ctrl->idletime.tv_usec || ctrl->cbbytes) + ctrl->idlecb = nControl->idlecb; + else + ctrl->idlecb = NULL; + *nData = ctrl; + return 1; +} + +/* + * FtpAcceptConnection - accept connection from server + * + * return 1 if successful, 0 otherwise + */ +static int FtpAcceptConnection(netbuf *nData, netbuf *nControl) +{ + int sData; + struct sockaddr addr; + unsigned int l; + int i; + struct timeval tv; + fd_set mask; + int rv; + + FD_ZERO(&mask); + FD_SET(nControl->handle, &mask); + FD_SET(nData->handle, &mask); + tv.tv_usec = 0; + tv.tv_sec = ACCEPT_TIMEOUT; + printf("<<<<<<<<<<<<<<<<%d\n",ACCEPT_TIMEOUT); + i = nControl->handle; + if (i < nData->handle) + i = nData->handle; + i = select(i+1, &mask, NULL, NULL, &tv); + if (i == -1) + { + strncpy(nControl->response, strerror(errno), + sizeof(nControl->response)); + net_close(nData->handle); + nData->handle = 0; + rv = 0; + } + else if (i == 0) + { + strcpy(nControl->response, "timed out waiting for connection"); + net_close(nData->handle); + nData->handle = 0; + rv = 0; + } + else + { + if (FD_ISSET(nData->handle, &mask)) + { + l = sizeof(addr); + sData = accept(nData->handle, &addr, &l); + i = errno; + net_close(nData->handle); + if (sData > 0) + { + rv = 1; + nData->handle = sData; + } + else + { + strncpy(nControl->response, strerror(i), + sizeof(nControl->response)); + nData->handle = 0; + rv = 0; + } + } + else if (FD_ISSET(nControl->handle, &mask)) + { + net_close(nData->handle); + nData->handle = 0; + readresp('2', nControl); + rv = 0; + } + } + return rv; +} + +/* + * FtpAccess - return a handle for a data stream + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpAccess(const char *path, int typ, int mode, netbuf *nControl, + netbuf **nData) +{ + char buf[256]; + int dir; + if ((path == NULL) && + ((typ == FTPLIB_FILE_WRITE) || (typ == FTPLIB_FILE_READ))) + { + sprintf(nControl->response, + "Missing path argument for file transfer\n"); + return 0; + } + sprintf(buf, "TYPE %c", mode); + if (!FtpSendCmd(buf, '2', nControl)) + return 0; + switch (typ) + { + case FTPLIB_DIR: + strcpy(buf,"NLST"); + dir = FTPLIB_READ; + break; + case FTPLIB_DIR_VERBOSE: + strcpy(buf,"LIST"); + dir = FTPLIB_READ; + break; + case FTPLIB_FILE_READ: + strcpy(buf,"RETR"); + dir = FTPLIB_READ; + break; + case FTPLIB_FILE_WRITE: + strcpy(buf,"STOR"); + dir = FTPLIB_WRITE; + break; + default: + sprintf(nControl->response, "Invalid open type %d\n", typ); + return 0; + } + if (path != NULL) + { + int i = strlen(buf); + buf[i++] = ' '; + if ((strlen(path) + i) >= sizeof(buf)) + return 0; + strcpy(&buf[i],path); + } + if (FtpOpenPort(nControl, nData, mode, dir) == -1) + return 0; + if (!FtpSendCmd(buf, '1', nControl)) + { + FtpClose(*nData); + *nData = NULL; + return 0; + } + (*nData)->ctrl = nControl; + nControl->data = *nData; + if (nControl->cmode == FTPLIB_PORT) + { + if (!FtpAcceptConnection(*nData,nControl)) + { + FtpClose(*nData); + *nData = NULL; + nControl->data = NULL; + return 0; + } + } + return 1; +} + +/* + * FtpRead - read from a data connection + */ +GLOBALDEF int FtpRead(void *buf, int max, netbuf *nData) +{ + int i; + if (nData->dir != FTPLIB_READ) + return 0; + if (nData->buf) + i = readline(buf, max, nData); + else + { + i = socket_wait(nData); + if (i != 1) + return 0; + i = net_read(nData->handle, buf, max); + } + if (i == -1) + return 0; + nData->xfered += i; + if (nData->idlecb && nData->cbbytes) + { + nData->xfered1 += i; + if (nData->xfered1 > nData->cbbytes) + { + if (nData->idlecb(nData, nData->xfered, nData->idlearg) == 0) + return 0; + nData->xfered1 = 0; + } + } + return i; +} + +/* + * FtpWrite - write to a data connection + */ +GLOBALDEF int FtpWrite(void *buf, int len, netbuf *nData) +{ + int i; + if (nData->dir != FTPLIB_WRITE) + return 0; + if (nData->buf) + i = writeline(buf, len, nData); + else + { + socket_wait(nData); + i = net_write(nData->handle, buf, len); + } + if (i == -1) + return 0; + nData->xfered += i; + if (nData->idlecb && nData->cbbytes) + { + nData->xfered1 += i; + if (nData->xfered1 > nData->cbbytes) + { + nData->idlecb(nData, nData->xfered, nData->idlearg); + nData->xfered1 = 0; + } + } + return i; +} + +/* + * FtpClose - close a data connection + */ +GLOBALDEF int FtpClose(netbuf *nData) +{ + netbuf *ctrl; + switch (nData->dir) + { + case FTPLIB_WRITE: + /* potential problem - if buffer flush fails, how to notify user? */ + if (nData->buf != NULL) + writeline(NULL, 0, nData); + case FTPLIB_READ: + if (nData->buf) + free(nData->buf); + shutdown(nData->handle,2); + net_close(nData->handle); + ctrl = nData->ctrl; + free(nData); + if (ctrl) + { + ctrl->data = NULL; + return(readresp('2', ctrl)); + } + return 1; + case FTPLIB_CONTROL: + if (nData->data) + { + nData->ctrl = NULL; + FtpClose(nData); + } + net_close(nData->handle); + free(nData); + return 0; + } + return 1; +} + +/* + * FtpSite - send a SITE command + * + * return 1 if command successful, 0 otherwise + */ +GLOBALDEF int FtpSite(const char *cmd, netbuf *nControl) +{ + char buf[256]; + + if ((strlen(cmd) + 7) > sizeof(buf)) + return 0; + sprintf(buf,"SITE %s",cmd); + if (!FtpSendCmd(buf,'2',nControl)) + return 0; + return 1; +} + +/* + * FtpSysType - send a SYST command + * + * Fills in the user buffer with the remote system type. If more + * information from the response is required, the user can parse + * it out of the response buffer returned by FtpLastResponse(). + * + * return 1 if command successful, 0 otherwise + */ +GLOBALDEF int FtpSysType(char *buf, int max, netbuf *nControl) +{ + int l = max; + char *b = buf; + char *s; + if (!FtpSendCmd("SYST",'2',nControl)) + return 0; + s = &nControl->response[4]; + while ((--l) && (*s != ' ')) + *b++ = *s++; + *b++ = '\0'; + return 1; +} + +/* + * FtpMkdir - create a directory at server + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpMkdir(const char *path, netbuf *nControl) +{ + char buf[256]; + + if ((strlen(path) + 6) > sizeof(buf)) + return 0; + sprintf(buf,"MKD %s",path); + if (!FtpSendCmd(buf,'2', nControl)) + return 0; + return 1; +} + +/* + * FtpChdir - change path at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpChdir(const char *path, netbuf *nControl) +{ + char buf[256]; + + if ((strlen(path) + 6) > sizeof(buf)) + return 0; + sprintf(buf,"CWD %s",path); + if (!FtpSendCmd(buf,'2',nControl)) + return 0; + return 1; +} + +/* + * FtpCDUp - move to parent directory at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpCDUp(netbuf *nControl) +{ + if (!FtpSendCmd("CDUP",'2',nControl)) + return 0; + return 1; +} + +/* + * FtpRmdir - remove directory at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpRmdir(const char *path, netbuf *nControl) +{ + char buf[256]; + + if ((strlen(path) + 6) > sizeof(buf)) + return 0; + sprintf(buf,"RMD %s",path); + if (!FtpSendCmd(buf,'2',nControl)) + return 0; + return 1; +} + +/* + * FtpPwd - get working directory at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpPwd(char *path, int max, netbuf *nControl) +{ + int l = max; + char *b = path; + char *s; + if (!FtpSendCmd("PWD",'2',nControl)) + return 0; + s = strchr(nControl->response, '"'); + if (s == NULL) + return 0; + s++; + while ((--l) && (*s) && (*s != '"')) + *b++ = *s++; + *b++ = '\0'; + return 1; +} + +/* + * FtpXfer - issue a command and transfer data + * + * return 1 if successful, 0 otherwise + */ +static int FtpXfer(const char *localfile, const char *path, + netbuf *nControl, int typ, int mode) +{ + int l,c; + char *dbuf; + FILE *local = NULL; + netbuf *nData; + int rv=1; + + if (localfile != NULL) + { + char ac[4] = "w"; + if (typ == FTPLIB_FILE_WRITE) + ac[0] = 'r'; + if (mode == FTPLIB_IMAGE) + ac[1] = 'b'; + local = fopen(localfile, ac); + if (local == NULL) + { + strncpy(nControl->response, strerror(errno), + sizeof(nControl->response)); + return 0; + } + } + if (local == NULL) + local = (typ == FTPLIB_FILE_WRITE) ? stdin : stdout; + if (!FtpAccess(path, typ, mode, nControl, &nData)) + return 0; + dbuf = malloc(FTPLIB_BUFSIZ); + if (typ == FTPLIB_FILE_WRITE) + { + while ((l = fread(dbuf, 1, FTPLIB_BUFSIZ, local)) > 0) + if ((c = FtpWrite(dbuf, l, nData)) < l) + { + printf("short write: passed %d, wrote %d\n", l, c); + rv = 0; + break; + } + } + else + { + while ((l = FtpRead(dbuf, FTPLIB_BUFSIZ, nData)) > 0) + if (fwrite(dbuf, 1, l, local) <= 0) + { + perror("localfile write"); + rv = 0; + break; + } + } + free(dbuf); + fflush(local); + if (localfile != NULL) + fclose(local); + FtpClose(nData); + return rv; +} + +/* + * FtpNlst - issue an NLST command and write response to output + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpNlst(const char *outputfile, const char *path, + netbuf *nControl) +{ + return FtpXfer(outputfile, path, nControl, FTPLIB_DIR, FTPLIB_ASCII); +} + +/* + * FtpDir - issue a LIST command and write response to output + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpDir(const char *outputfile, const char *path, netbuf *nControl) +{ + return FtpXfer(outputfile, path, nControl, FTPLIB_DIR_VERBOSE, FTPLIB_ASCII); +} + +/* + * FtpSize - determine the size of a remote file + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpSize(const char *path, int *size, char mode, netbuf *nControl) +{ + char cmd[256]; + int resp,sz,rv=1; + + if ((strlen(path) + 7) > sizeof(cmd)) + return 0; + sprintf(cmd, "TYPE %c", mode); + if (!FtpSendCmd(cmd, '2', nControl)) + return 0; + sprintf(cmd,"SIZE %s",path); + if (!FtpSendCmd(cmd,'2',nControl)) + rv = 0; + else + { + if (sscanf(nControl->response, "%d %d", &resp, &sz) == 2) + *size = sz; + else + rv = 0; + } + return rv; +} + +/* + * FtpModDate - determine the modification date of a remote file + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpModDate(const char *path, char *dt, int max, netbuf *nControl) +{ + char buf[256]; + int rv = 1; + + if ((strlen(path) + 7) > sizeof(buf)) + return 0; + sprintf(buf,"MDTM %s",path); + if (!FtpSendCmd(buf,'2',nControl)) + rv = 0; + else + strncpy(dt, &nControl->response[4], max); + return rv; +} + +/* + * FtpGet - issue a GET command and write received data to output + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpGet(const char *outputfile, const char *path, + char mode, netbuf *nControl) +{ + return FtpXfer(outputfile, path, nControl, FTPLIB_FILE_READ, mode); +} + +/* + * FtpPut - issue a PUT command and send data from input + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpPut(const char *inputfile, const char *path, char mode, + netbuf *nControl) +{ + return FtpXfer(inputfile, path, nControl, FTPLIB_FILE_WRITE, mode); +} + +/* + * FtpRename - rename a file at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpRename(const char *src, const char *dst, netbuf *nControl) +{ + char cmd[256]; + + if (((strlen(src) + 7) > sizeof(cmd)) || + ((strlen(dst) + 7) > sizeof(cmd))) + return 0; + sprintf(cmd,"RNFR %s",src); + if (!FtpSendCmd(cmd,'3',nControl)) + return 0; + sprintf(cmd,"RNTO %s",dst); + if (!FtpSendCmd(cmd,'2',nControl)) + return 0; + return 1; +} + +/* + * FtpDelete - delete a file at remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF int FtpDelete(const char *fnm, netbuf *nControl) +{ + char cmd[256]; + + if ((strlen(fnm) + 7) > sizeof(cmd)) + return 0; + sprintf(cmd,"DELE %s",fnm); + if (!FtpSendCmd(cmd,'2', nControl)) + return 0; + return 1; +} + +/* + * FtpQuit - disconnect from remote + * + * return 1 if successful, 0 otherwise + */ +GLOBALDEF void FtpQuit(netbuf *nControl) +{ + if (nControl->dir != FTPLIB_CONTROL) + return; + FtpSendCmd("QUIT",'2',nControl); + net_close(nControl->handle); + free(nControl->buf); + free(nControl); +} diff --git a/src/utilfuns/ftpparse.c b/src/utilfuns/ftpparse.c new file mode 100644 index 0000000..b9bafa1 --- /dev/null +++ b/src/utilfuns/ftpparse.c @@ -0,0 +1,447 @@ +/* ftpparse.c, ftpparse.h: library for parsing FTP LIST responses +20001223 +D. J. Bernstein, djb@cr.yp.to +http://cr.yp.to/ftpparse.html + +Commercial use is fine, if you let me know what programs you're using this in. + +Currently covered formats: +EPLF. +UNIX ls, with or without gid. +Microsoft FTP Service. +Windows NT FTP Server. +VMS. +WFTPD. +NetPresenz (Mac). +NetWare. +MSDOS. + +Definitely not covered: +Long VMS filenames, with information split across two lines. +NCSA Telnet FTP server. Has LIST = NLST (and bad NLST for directories). +*/ + +#include <ftpparse.h> + +static long totai(long year,long month,long mday) +{ + long result; + if (month >= 2) month -= 2; + else { month += 10; --year; } + result = (mday - 1) * 10 + 5 + 306 * month; + result /= 10; + if (result == 365) { year -= 3; result = 1460; } + else result += 365 * (year % 4); + year /= 4; + result += 1461 * (year % 25); + year /= 25; + if (result == 36524) { year -= 3; result = 146096; } + else { result += 36524 * (year % 4); } + year /= 4; + result += 146097 * (year - 5); + result += 11017; + return result * 86400; +} + +static int flagneedbase = 1; +static time_t base; /* time() value on this OS at the beginning of 1970 TAI */ +static long now; /* current time */ +static int flagneedcurrentyear = 1; +static long currentyear; /* approximation to current year */ + +static void initbase(void) +{ + struct tm *t; + if (!flagneedbase) return; + + base = 0; + t = gmtime(&base); + base = -(totai(t->tm_year + 1900,t->tm_mon,t->tm_mday) + t->tm_hour * 3600 + t->tm_min * 60 + t->tm_sec); + /* assumes the right time_t, counting seconds. */ + /* base may be slightly off if time_t counts non-leap seconds. */ + flagneedbase = 0; +} + +static void initnow(void) +{ + long day; + long year; + + initbase(); + now = time((time_t *) 0) - base; + + if (flagneedcurrentyear) { + day = now / 86400; + if ((now % 86400) < 0) --day; + day -= 11017; + year = 5 + day / 146097; + day = day % 146097; + if (day < 0) { day += 146097; --year; } + year *= 4; + if (day == 146096) { year += 3; day = 36524; } + else { year += day / 36524; day %= 36524; } + year *= 25; + year += day / 1461; + day %= 1461; + year *= 4; + if (day == 1460) { year += 3; day = 365; } + else { year += day / 365; day %= 365; } + day *= 10; + if ((day + 5) / 306 >= 10) ++year; + currentyear = year; + flagneedcurrentyear = 0; + } +} + +/* UNIX ls does not show the year for dates in the last six months. */ +/* So we have to guess the year. */ +/* Apparently NetWare uses ``twelve months'' instead of ``six months''; ugh. */ +/* Some versions of ls also fail to show the year for future dates. */ +static long guesstai(long month,long mday) +{ + long year; + long t; + + initnow(); + + for (year = currentyear - 1;year < currentyear + 100;++year) { + t = totai(year,month,mday); + if (now - t < 350 * 86400) + return t; + } + return 0; +} + +static int check(char *buf,char *monthname) +{ + if ((buf[0] != monthname[0]) && (buf[0] != monthname[0] - 32)) return 0; + if ((buf[1] != monthname[1]) && (buf[1] != monthname[1] - 32)) return 0; + if ((buf[2] != monthname[2]) && (buf[2] != monthname[2] - 32)) return 0; + return 1; +} + +static char *months[12] = { + "jan","feb","mar","apr","may","jun","jul","aug","sep","oct","nov","dec" +} ; + +static int getmonth(char *buf,int len) +{ + int i; + if (len == 3) + for (i = 0;i < 12;++i) + if (check(buf,months[i])) return i; + return -1; +} + +static long getlong(char *buf,int len) +{ + long u = 0; + while (len-- > 0) + u = u * 10 + (*buf++ - '0'); + return u; +} + +int ftpparse(struct ftpparse *fp,char *buf,int len) +{ + int i; + int j; + int state; + long size = 0; + long year; + long month = 0; + long mday = 0; + long hour; + long minute; + + fp->name = 0; + fp->namelen = 0; + fp->flagtrycwd = 0; + fp->flagtryretr = 0; + fp->sizetype = FTPPARSE_SIZE_UNKNOWN; + fp->size = 0; + fp->mtimetype = FTPPARSE_MTIME_UNKNOWN; + fp->mtime = 0; + fp->idtype = FTPPARSE_ID_UNKNOWN; + fp->id = 0; + fp->idlen = 0; + + if (len < 2) /* an empty name in EPLF, with no info, could be 2 chars */ + return 0; + + switch(*buf) { + /* see http://pobox.com/~djb/proto/eplf.txt */ + /* "+i8388621.29609,m824255902,/,\tdev" */ + /* "+i8388621.44468,m839956783,r,s10376,\tRFCEPLF" */ + case '+': + i = 1; + for (j = 1;j < len;++j) { + if (buf[j] == 9) { + fp->name = buf + j + 1; + fp->namelen = len - j - 1; + return 1; + } + if (buf[j] == ',') { + switch(buf[i]) { + case '/': + fp->flagtrycwd = 1; + break; + case 'r': + fp->flagtryretr = 1; + break; + case 's': + fp->sizetype = FTPPARSE_SIZE_BINARY; + fp->size = getlong(buf + i + 1,j - i - 1); + break; + case 'm': + fp->mtimetype = FTPPARSE_MTIME_LOCAL; + initbase(); + fp->mtime = base + getlong(buf + i + 1,j - i - 1); + break; + case 'i': + fp->idtype = FTPPARSE_ID_FULL; + fp->id = buf + i + 1; + fp->idlen = j - i - 1; + } + i = j + 1; + } + } + return 0; + + /* UNIX-style listing, without inum and without blocks */ + /* "-rw-r--r-- 1 root other 531 Jan 29 03:26 README" */ + /* "dr-xr-xr-x 2 root other 512 Apr 8 1994 etc" */ + /* "dr-xr-xr-x 2 root 512 Apr 8 1994 etc" */ + /* "lrwxrwxrwx 1 root other 7 Jan 25 00:17 bin -> usr/bin" */ + /* Also produced by Microsoft's FTP servers for Windows: */ + /* "---------- 1 owner group 1803128 Jul 10 10:18 ls-lR.Z" */ + /* "d--------- 1 owner group 0 May 9 19:45 Softlib" */ + /* Also WFTPD for MSDOS: */ + /* "-rwxrwxrwx 1 noone nogroup 322 Aug 19 1996 message.ftp" */ + /* Also NetWare: */ + /* "d [R----F--] supervisor 512 Jan 16 18:53 login" */ + /* "- [R----F--] rhesus 214059 Oct 20 15:27 cx.exe" */ + /* Also NetPresenz for the Mac: */ + /* "-------r-- 326 1391972 1392298 Nov 22 1995 MegaPhone.sit" */ + /* "drwxrwxr-x folder 2 May 10 1996 network" */ + case 'b': + case 'c': + case 'd': + case 'l': + case 'p': + case 's': + case '-': + + if (*buf == 'd') fp->flagtrycwd = 1; + if (*buf == '-') fp->flagtryretr = 1; + if (*buf == 'l') fp->flagtrycwd = fp->flagtryretr = 1; + + state = 1; + i = 0; + for (j = 1;j < len;++j) + if ((buf[j] == ' ') && (buf[j - 1] != ' ')) { + switch(state) { + case 1: /* skipping perm */ + state = 2; + break; + case 2: /* skipping nlink */ + state = 3; + if ((j - i == 6) && (buf[i] == 'f')) /* for NetPresenz */ + state = 4; + break; + case 3: /* skipping uid */ + state = 4; + break; + case 4: /* getting tentative size */ + size = getlong(buf + i,j - i); + state = 5; + break; + case 5: /* searching for month, otherwise getting tentative size */ + month = getmonth(buf + i,j - i); + if (month >= 0) + state = 6; + else + size = getlong(buf + i,j - i); + break; + case 6: /* have size and month */ + mday = getlong(buf + i,j - i); + state = 7; + break; + case 7: /* have size, month, mday */ + if ((j - i == 4) && (buf[i + 1] == ':')) { + hour = getlong(buf + i,1); + minute = getlong(buf + i + 2,2); + fp->mtimetype = FTPPARSE_MTIME_REMOTEMINUTE; + initbase(); + fp->mtime = base + guesstai(month,mday) + hour * 3600 + minute * 60; + } else if ((j - i == 5) && (buf[i + 2] == ':')) { + hour = getlong(buf + i,2); + minute = getlong(buf + i + 3,2); + fp->mtimetype = FTPPARSE_MTIME_REMOTEMINUTE; + initbase(); + fp->mtime = base + guesstai(month,mday) + hour * 3600 + minute * 60; + } + else if (j - i >= 4) { + year = getlong(buf + i,j - i); + fp->mtimetype = FTPPARSE_MTIME_REMOTEDAY; + initbase(); + fp->mtime = base + totai(year,month,mday); + } + else + return 0; + fp->name = buf + j + 1; + fp->namelen = len - j - 1; + state = 8; + break; + case 8: /* twiddling thumbs */ + break; + } + i = j + 1; + while ((i < len) && (buf[i] == ' ')) ++i; + } + + if (state != 8) + return 0; + + fp->size = size; + fp->sizetype = FTPPARSE_SIZE_BINARY; + + if (*buf == 'l') + for (i = 0;i + 3 < fp->namelen;++i) + if (fp->name[i] == ' ') + if (fp->name[i + 1] == '-') + if (fp->name[i + 2] == '>') + if (fp->name[i + 3] == ' ') { + fp->namelen = i; + break; + } + + /* eliminate extra NetWare spaces */ + if ((buf[1] == ' ') || (buf[1] == '[')) + if (fp->namelen > 3) + if (fp->name[0] == ' ') + if (fp->name[1] == ' ') + if (fp->name[2] == ' ') { + fp->name += 3; + fp->namelen -= 3; + } + + return 1; + } + + /* MultiNet (some spaces removed from examples) */ + /* "00README.TXT;1 2 30-DEC-1996 17:44 [SYSTEM] (RWED,RWED,RE,RE)" */ + /* "CORE.DIR;1 1 8-SEP-1996 16:09 [SYSTEM] (RWE,RWE,RE,RE)" */ + /* and non-MutliNet VMS: */ + /* "CII-MANUAL.TEX;1 213/216 29-JAN-1996 03:33:12 [ANONYMOU,ANONYMOUS] (RWED,RWED,,)" */ + for (i = 0;i < len;++i) + if (buf[i] == ';') + break; + if (i < len) { + fp->name = buf; + fp->namelen = i; + if (i > 4) + if (buf[i - 4] == '.') + if (buf[i - 3] == 'D') + if (buf[i - 2] == 'I') + if (buf[i - 1] == 'R') { + fp->namelen -= 4; + fp->flagtrycwd = 1; + } + if (!fp->flagtrycwd) + fp->flagtryretr = 1; + while (buf[i] != ' ') if (++i == len) return 0; + while (buf[i] == ' ') if (++i == len) return 0; + while (buf[i] != ' ') if (++i == len) return 0; + while (buf[i] == ' ') if (++i == len) return 0; + j = i; + while (buf[j] != '-') if (++j == len) return 0; + mday = getlong(buf + i,j - i); + while (buf[j] == '-') if (++j == len) return 0; + i = j; + while (buf[j] != '-') if (++j == len) return 0; + month = getmonth(buf + i,j - i); + if (month < 0) return 0; + while (buf[j] == '-') if (++j == len) return 0; + i = j; + while (buf[j] != ' ') if (++j == len) return 0; + year = getlong(buf + i,j - i); + while (buf[j] == ' ') if (++j == len) return 0; + i = j; + while (buf[j] != ':') if (++j == len) return 0; + hour = getlong(buf + i,j - i); + while (buf[j] == ':') if (++j == len) return 0; + i = j; + while ((buf[j] != ':') && (buf[j] != ' ')) if (++j == len) return 0; + minute = getlong(buf + i,j - i); + + fp->mtimetype = FTPPARSE_MTIME_REMOTEMINUTE; + initbase(); + fp->mtime = base + totai(year,month,mday) + hour * 3600 + minute * 60; + + return 1; + } + + /* MSDOS format */ + /* 04-27-00 09:09PM <DIR> licensed */ + /* 07-18-00 10:16AM <DIR> pub */ + /* 04-14-00 03:47PM 589 readme.htm */ + if ((*buf >= '0') && (*buf <= '9')) { + i = 0; + j = 0; + while (buf[j] != '-') if (++j == len) return 0; + month = getlong(buf + i,j - i) - 1; + while (buf[j] == '-') if (++j == len) return 0; + i = j; + while (buf[j] != '-') if (++j == len) return 0; + mday = getlong(buf + i,j - i); + while (buf[j] == '-') if (++j == len) return 0; + i = j; + while (buf[j] != ' ') if (++j == len) return 0; + year = getlong(buf + i,j - i); + if (year < 50) year += 2000; + if (year < 1000) year += 1900; + while (buf[j] == ' ') if (++j == len) return 0; + i = j; + while (buf[j] != ':') if (++j == len) return 0; + hour = getlong(buf + i,j - i); + while (buf[j] == ':') if (++j == len) return 0; + i = j; + while ((buf[j] != 'A') && (buf[j] != 'P')) if (++j == len) return 0; + minute = getlong(buf + i,j - i); + if (hour == 12) hour = 0; + if (buf[j] == 'A') if (++j == len) return 0; + if (buf[j] == 'P') { hour += 12; if (++j == len) return 0; } + if (buf[j] == 'M') if (++j == len) return 0; + + while (buf[j] == ' ') if (++j == len) return 0; + if (buf[j] == '<') { + fp->flagtrycwd = 1; + while (buf[j] != ' ') if (++j == len) return 0; + } + else { + i = j; + while (buf[j] != ' ') if (++j == len) return 0; + fp->size = getlong(buf + i,j - i); + fp->sizetype = FTPPARSE_SIZE_BINARY; + fp->flagtryretr = 1; + } + while (buf[j] == ' ') if (++j == len) return 0; + + fp->name = buf + j; + fp->namelen = len - j; + + fp->mtimetype = FTPPARSE_MTIME_REMOTEMINUTE; + initbase(); + fp->mtime = base + totai(year,month,mday) + hour * 3600 + minute * 60; + + return 1; + } + + /* Some useless lines, safely ignored: */ + /* "Total of 11 Files, 10966 Blocks." (VMS) */ + /* "total 14786" (UNIX) */ + /* "DISK$ANONFTP:[ANONYMOUS]" (VMS) */ + /* "Directory DISK$PCSA:[ANONYM]" (VMS) */ + + return 0; +} diff --git a/src/utilfuns/regex.c b/src/utilfuns/regex.c new file mode 100644 index 0000000..09743d2 --- /dev/null +++ b/src/utilfuns/regex.c @@ -0,0 +1,5721 @@ +/* Extended regular expression matching and search library, + version 0.12. + (Implements POSIX draft P1003.2/D11.2, except for some of the + internationalization features.) + + Copyright (C) 1993, 1994, 1995, 1996, 1997 Free Software Foundation, Inc. + + the C library, however. The master source lives in /gd/gnu/lib. + +NOTE: The canonical source of this file is maintained with the +GNU C Library. Bugs can be reported to bug-glibc@prep.ai.mit.edu. + +This program is free software; you can redistribute it and/or modify it +under the terms of the GNU General Public License as published by the +Free Software Foundation; either version 2, or (at your option) any +later version. + +This program is distributed in the hope that it will be useful, +but WITHOUT ANY WARRANTY; without even the implied warranty of +MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the +GNU General Public License for more details. + +You should have received a copy of the GNU General Public License +along with this program; if not, write to the Free Software Foundation, +Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ + +/* AIX requires this to be the first thing in the file. */ +#if defined (_AIX) && !defined (REGEX_MALLOC) + #pragma alloca +#endif + +#undef _GNU_SOURCE +#define _GNU_SOURCE + +#ifdef HAVE_CONFIG_H +#include <config.h> +#endif + +#if defined(STDC_HEADERS) && !defined(emacs) +#include <stddef.h> +#else +/* We need this for `regex.h', and perhaps for the Emacs include files. */ +#include <sys/types.h> +#endif +#include <stdlib.h> // sword + +/* For platform which support the ISO C amendement 1 functionality we + support user defined character classes. */ +#if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H) +# include <wctype.h> +# include <wchar.h> +#endif + +/* This is for other GNU distributions with internationalized messages. */ +#if HAVE_LIBINTL_H || defined (_LIBC) +# include <libintl.h> +#else +# define gettext(msgid) (msgid) +#endif + +#ifndef gettext_noop +/* This define is so xgettext can find the internationalizable + strings. */ +#define gettext_noop(String) String +#endif + +/* The `emacs' switch turns on certain matching commands + that make sense only in Emacs. */ +#ifdef emacs + +#include "lisp.h" +#include "buffer.h" +#include "syntax.h" + +#else /* not emacs */ + +/* If we are not linking with Emacs proper, + we can't use the relocating allocator + even if config.h says that we can. */ +#undef REL_ALLOC + +#if defined (STDC_HEADERS) || defined (_LIBC) +#include <stdlib.h> +#else +//sword char *malloc (); +//sword char *realloc (); +#endif + +/* When used in Emacs's lib-src, we need to get bzero and bcopy somehow. + If nothing else has been done, use the method below. */ +#ifdef INHIBIT_STRING_HEADER +#if !(defined (HAVE_BZERO) && defined (HAVE_BCOPY)) +#if !defined (bzero) && !defined (bcopy) +#undef INHIBIT_STRING_HEADER +#endif +#endif +#endif + +/* This is the normal way of making sure we have a bcopy and a bzero. + This is used in most programs--a few other programs avoid this + by defining INHIBIT_STRING_HEADER. */ +#define HAVE_STRING_H // for SWORD Project +#ifndef INHIBIT_STRING_HEADER +#if defined (HAVE_STRING_H) || defined (STDC_HEADERS) || defined (_LIBC) +#include <string.h> +#ifndef bcmp +#define bcmp(s1, s2, n) memcmp ((s1), (s2), (n)) +#endif +#ifndef bcopy +#define bcopy(s, d, n) memcpy ((d), (s), (n)) +#endif +#ifndef bzero +#define bzero(s, n) memset ((s), 0, (n)) +#endif +#else +#include <strings.h> +#endif +#endif + +/* Define the syntax stuff for \<, \>, etc. */ + +/* This must be nonzero for the wordchar and notwordchar pattern + commands in re_match_2. */ +#ifndef Sword +#define Sword 1 +#endif + +#ifdef SWITCH_ENUM_BUG +#define SWITCH_ENUM_CAST(x) ((int)(x)) +#else +#define SWITCH_ENUM_CAST(x) (x) +#endif + +#ifdef SYNTAX_TABLE + +extern char *re_syntax_table; + +#else /* not SYNTAX_TABLE */ + +/* How many characters in the character set. */ +#define CHAR_SET_SIZE 256 + +static char re_syntax_table[CHAR_SET_SIZE]; + +static void +init_syntax_once () +{ + register int c; + static int done = 0; + + if (done) + return; + + bzero (re_syntax_table, sizeof re_syntax_table); + + for (c = 'a'; c <= 'z'; c++) + re_syntax_table[c] = Sword; + + for (c = 'A'; c <= 'Z'; c++) + re_syntax_table[c] = Sword; + + for (c = '0'; c <= '9'; c++) + re_syntax_table[c] = Sword; + + re_syntax_table['_'] = Sword; + + done = 1; +} + +#endif /* not SYNTAX_TABLE */ + +#define SYNTAX(c) re_syntax_table[c] + +#endif /* not emacs */ + +/* Get the interface, including the syntax bits. */ +#include "regex.h" + +/* isalpha etc. are used for the character classes. */ +#include <ctype.h> + +/* Jim Meyering writes: + + "... Some ctype macros are valid only for character codes that + isascii says are ASCII (SGI's IRIX-4.0.5 is one such system --when + using /bin/cc or gcc but without giving an ansi option). So, all + ctype uses should be through macros like ISPRINT... If + STDC_HEADERS is defined, then autoconf has verified that the ctype + macros don't need to be guarded with references to isascii. ... + Defining isascii to 1 should let any compiler worth its salt + eliminate the && through constant folding." */ + +#if defined (STDC_HEADERS) || (!defined (isascii) && !defined (HAVE_ISASCII)) +#define ISASCII(c) 1 +#else +#define ISASCII(c) isascii(c) +#endif + +#ifdef isblank +#define ISBLANK(c) (ISASCII (c) && isblank (c)) +#else +#define ISBLANK(c) ((c) == ' ' || (c) == '\t') +#endif +#ifdef isgraph +#define ISGRAPH(c) (ISASCII (c) && isgraph (c)) +#else +#define ISGRAPH(c) (ISASCII (c) && isprint (c) && !isspace (c)) +#endif + +#define ISPRINT(c) (ISASCII (c) && isprint (c)) +#define ISDIGIT(c) (ISASCII (c) && isdigit (c)) +#define ISALNUM(c) (ISASCII (c) && isalnum (c)) +#define ISALPHA(c) (ISASCII (c) && isalpha (c)) +#define ISCNTRL(c) (ISASCII (c) && iscntrl (c)) +#define ISLOWER(c) (ISASCII (c) && islower (c)) +#define ISPUNCT(c) (ISASCII (c) && ispunct (c)) +#define ISSPACE(c) (ISASCII (c) && isspace (c)) +#define ISUPPER(c) (ISASCII (c) && isupper (c)) +#define ISXDIGIT(c) (ISASCII (c) && isxdigit (c)) + +#ifndef NULL +#define NULL (void *)0 +#endif + +/* We remove any previous definition of `SIGN_EXTEND_CHAR', + since ours (we hope) works properly with all combinations of + machines, compilers, `char' and `unsigned char' argument types. + (Per Bothner suggested the basic approach.) */ +#undef SIGN_EXTEND_CHAR +#if __STDC__ +#define SIGN_EXTEND_CHAR(c) ((signed char) (c)) +#else /* not __STDC__ */ +/* As in Harbison and Steele. */ +#define SIGN_EXTEND_CHAR(c) ((((unsigned char) (c)) ^ 128) - 128) +#endif + +/* Should we use malloc or alloca? If REGEX_MALLOC is not defined, we + use `alloca' instead of `malloc'. This is because using malloc in + re_search* or re_match* could cause memory leaks when C-g is used in + Emacs; also, malloc is slower and causes storage fragmentation. On + the other hand, malloc is more portable, and easier to debug. + + Because we sometimes use alloca, some routines have to be macros, + not functions -- `alloca'-allocated space disappears at the end of the + function it is called in. */ + +#ifdef REGEX_MALLOC + +#define REGEX_ALLOCATE malloc +#define REGEX_REALLOCATE(source, osize, nsize) realloc (source, nsize) +#define REGEX_FREE free + +#else /* not REGEX_MALLOC */ + +/* Emacs already defines alloca, sometimes. */ +#ifndef alloca + +/* Make alloca work the best possible way. */ +#ifdef __GNUC__ +#define alloca __builtin_alloca +#else /* not __GNUC__ */ +#if HAVE_ALLOCA_H +#include <alloca.h> +#else /* not __GNUC__ or HAVE_ALLOCA_H */ +#if 0 /* It is a bad idea to declare alloca. We always cast the result. */ +#ifndef _AIX /* Already did AIX, up at the top. */ +char *alloca (); +#endif /* not _AIX */ +#endif +#endif /* not HAVE_ALLOCA_H */ +#endif /* not __GNUC__ */ + +#endif /* not alloca */ + +#define REGEX_ALLOCATE alloca + +/* Assumes a `char *destination' variable. */ +#define REGEX_REALLOCATE(source, osize, nsize) \ + (destination = (char *) alloca (nsize), \ + bcopy (source, destination, osize), \ + destination) + +/* No need to do anything to free, after alloca. */ +#define REGEX_FREE(arg) ((void)0) /* Do nothing! But inhibit gcc warning. */ + +#endif /* not REGEX_MALLOC */ + +/* Define how to allocate the failure stack. */ + +#if defined (REL_ALLOC) && defined (REGEX_MALLOC) + +#define REGEX_ALLOCATE_STACK(size) \ + r_alloc (&failure_stack_ptr, (size)) +#define REGEX_REALLOCATE_STACK(source, osize, nsize) \ + r_re_alloc (&failure_stack_ptr, (nsize)) +#define REGEX_FREE_STACK(ptr) \ + r_alloc_free (&failure_stack_ptr) + +#else /* not using relocating allocator */ + +#ifdef REGEX_MALLOC + +#define REGEX_ALLOCATE_STACK malloc +#define REGEX_REALLOCATE_STACK(source, osize, nsize) realloc (source, nsize) +#define REGEX_FREE_STACK free + +#else /* not REGEX_MALLOC */ + +#define REGEX_ALLOCATE_STACK alloca + +#define REGEX_REALLOCATE_STACK(source, osize, nsize) \ + REGEX_REALLOCATE (source, osize, nsize) +/* No need to explicitly free anything. */ +#define REGEX_FREE_STACK(arg) + +#endif /* not REGEX_MALLOC */ +#endif /* not using relocating allocator */ + + +/* True if `size1' is non-NULL and PTR is pointing anywhere inside + `string1' or just past its end. This works if PTR is NULL, which is + a good thing. */ +#define FIRST_STRING_P(ptr) \ + (size1 && string1 <= (ptr) && (ptr) <= string1 + size1) + +/* (Re)Allocate N items of type T using malloc, or fail. */ +#define TALLOC(n, t) ((t *) malloc ((n) * sizeof (t))) +#define RETALLOC(addr, n, t) ((addr) = (t *) realloc (addr, (n) * sizeof (t))) +#define RETALLOC_IF(addr, n, t) \ + if (addr) RETALLOC((addr), (n), t); else (addr) = TALLOC ((n), t) +#define REGEX_TALLOC(n, t) ((t *) REGEX_ALLOCATE ((n) * sizeof (t))) + +#define BYTEWIDTH 8 /* In bits. */ + +#define STREQ(s1, s2) ((strcmp (s1, s2) == 0)) + +#undef MAX +#undef MIN +#define MAX(a, b) ((a) > (b) ? (a) : (b)) +#define MIN(a, b) ((a) < (b) ? (a) : (b)) + +typedef char boolean; +#define false 0 +#define true 1 + +static int re_match_2_internal (); + +/* These are the command codes that appear in compiled regular + expressions. Some opcodes are followed by argument bytes. A + command code can specify any interpretation whatsoever for its + arguments. Zero bytes may appear in the compiled regular expression. */ + +typedef enum +{ + no_op = 0, + + /* Succeed right away--no more backtracking. */ + succeed, + + /* Followed by one byte giving n, then by n literal bytes. */ + exactn, + + /* Matches any (more or less) character. */ + anychar, + + /* Matches any one char belonging to specified set. First + following byte is number of bitmap bytes. Then come bytes + for a bitmap saying which chars are in. Bits in each byte + are ordered low-bit-first. A character is in the set if its + bit is 1. A character too large to have a bit in the map is + automatically not in the set. */ + charset, + + /* Same parameters as charset, but match any character that is + not one of those specified. */ + charset_not, + + /* Start remembering the text that is matched, for storing in a + register. Followed by one byte with the register number, in + the range 0 to one less than the pattern buffer's re_nsub + field. Then followed by one byte with the number of groups + inner to this one. (This last has to be part of the + start_memory only because we need it in the on_failure_jump + of re_match_2.) */ + start_memory, + + /* Stop remembering the text that is matched and store it in a + memory register. Followed by one byte with the register + number, in the range 0 to one less than `re_nsub' in the + pattern buffer, and one byte with the number of inner groups, + just like `start_memory'. (We need the number of inner + groups here because we don't have any easy way of finding the + corresponding start_memory when we're at a stop_memory.) */ + stop_memory, + + /* Match a duplicate of something remembered. Followed by one + byte containing the register number. */ + duplicate, + + /* Fail unless at beginning of line. */ + begline, + + /* Fail unless at end of line. */ + endline, + + /* Succeeds if at beginning of buffer (if emacs) or at beginning + of string to be matched (if not). */ + begbuf, + + /* Analogously, for end of buffer/string. */ + endbuf, + + /* Followed by two byte relative address to which to jump. */ + jump, + + /* Same as jump, but marks the end of an alternative. */ + jump_past_alt, + + /* Followed by two-byte relative address of place to resume at + in case of failure. */ + on_failure_jump, + + /* Like on_failure_jump, but pushes a placeholder instead of the + current string position when executed. */ + on_failure_keep_string_jump, + + /* Throw away latest failure point and then jump to following + two-byte relative address. */ + pop_failure_jump, + + /* Change to pop_failure_jump if know won't have to backtrack to + match; otherwise change to jump. This is used to jump + back to the beginning of a repeat. If what follows this jump + clearly won't match what the repeat does, such that we can be + sure that there is no use backtracking out of repetitions + already matched, then we change it to a pop_failure_jump. + Followed by two-byte address. */ + maybe_pop_jump, + + /* Jump to following two-byte address, and push a dummy failure + point. This failure point will be thrown away if an attempt + is made to use it for a failure. A `+' construct makes this + before the first repeat. Also used as an intermediary kind + of jump when compiling an alternative. */ + dummy_failure_jump, + + /* Push a dummy failure point and continue. Used at the end of + alternatives. */ + push_dummy_failure, + + /* Followed by two-byte relative address and two-byte number n. + After matching N times, jump to the address upon failure. */ + succeed_n, + + /* Followed by two-byte relative address, and two-byte number n. + Jump to the address N times, then fail. */ + jump_n, + + /* Set the following two-byte relative address to the + subsequent two-byte number. The address *includes* the two + bytes of number. */ + set_number_at, + + wordchar, /* Matches any word-constituent character. */ + notwordchar, /* Matches any char that is not a word-constituent. */ + + wordbeg, /* Succeeds if at word beginning. */ + wordend, /* Succeeds if at word end. */ + + wordbound, /* Succeeds if at a word boundary. */ + notwordbound /* Succeeds if not at a word boundary. */ + +#ifdef emacs + ,before_dot, /* Succeeds if before point. */ + at_dot, /* Succeeds if at point. */ + after_dot, /* Succeeds if after point. */ + + /* Matches any character whose syntax is specified. Followed by + a byte which contains a syntax code, e.g., Sword. */ + syntaxspec, + + /* Matches any character whose syntax is not that specified. */ + notsyntaxspec +#endif /* emacs */ +} re_opcode_t; + +/* Common operations on the compiled pattern. */ + +/* Store NUMBER in two contiguous bytes starting at DESTINATION. */ + +#define STORE_NUMBER(destination, number) \ + do { \ + (destination)[0] = (number) & 0377; \ + (destination)[1] = (number) >> 8; \ + } while (0) + +/* Same as STORE_NUMBER, except increment DESTINATION to + the byte after where the number is stored. Therefore, DESTINATION + must be an lvalue. */ + +#define STORE_NUMBER_AND_INCR(destination, number) \ + do { \ + STORE_NUMBER (destination, number); \ + (destination) += 2; \ + } while (0) + +/* Put into DESTINATION a number stored in two contiguous bytes starting + at SOURCE. */ + +#define EXTRACT_NUMBER(destination, source) \ + do { \ + (destination) = *(source) & 0377; \ + (destination) += SIGN_EXTEND_CHAR (*((source) + 1)) << 8; \ + } while (0) + +#ifdef DEBUG +static void extract_number _RE_ARGS ((int *dest, unsigned char *source)); +static void +extract_number (dest, source) + int *dest; + unsigned char *source; +{ + int temp = SIGN_EXTEND_CHAR (*(source + 1)); + *dest = *source & 0377; + *dest += temp << 8; +} + +#ifndef EXTRACT_MACROS /* To debug the macros. */ +#undef EXTRACT_NUMBER +#define EXTRACT_NUMBER(dest, src) extract_number (&dest, src) +#endif /* not EXTRACT_MACROS */ + +#endif /* DEBUG */ + +/* Same as EXTRACT_NUMBER, except increment SOURCE to after the number. + SOURCE must be an lvalue. */ + +#define EXTRACT_NUMBER_AND_INCR(destination, source) \ + do { \ + EXTRACT_NUMBER (destination, source); \ + (source) += 2; \ + } while (0) + +#ifdef DEBUG +static void extract_number_and_incr _RE_ARGS ((int *destination, + unsigned char **source)); +static void +extract_number_and_incr (destination, source) + int *destination; + unsigned char **source; +{ + extract_number (destination, *source); + *source += 2; +} + +#ifndef EXTRACT_MACROS +#undef EXTRACT_NUMBER_AND_INCR +#define EXTRACT_NUMBER_AND_INCR(dest, src) \ + extract_number_and_incr (&dest, &src) +#endif /* not EXTRACT_MACROS */ + +#endif /* DEBUG */ + +/* If DEBUG is defined, Regex prints many voluminous messages about what + it is doing (if the variable `debug' is nonzero). If linked with the + main program in `iregex.c', you can enter patterns and strings + interactively. And if linked with the main program in `main.c' and + the other test files, you can run the already-written tests. */ + +#ifdef DEBUG + +/* We use standard I/O for debugging. */ +#include <stdio.h> + +/* It is useful to test things that ``must'' be true when debugging. */ +#include <assert.h> + +static int debug = 0; + +#define DEBUG_STATEMENT(e) e +#define DEBUG_PRINT1(x) if (debug) printf (x) +#define DEBUG_PRINT2(x1, x2) if (debug) printf (x1, x2) +#define DEBUG_PRINT3(x1, x2, x3) if (debug) printf (x1, x2, x3) +#define DEBUG_PRINT4(x1, x2, x3, x4) if (debug) printf (x1, x2, x3, x4) +#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) \ + if (debug) print_partial_compiled_pattern (s, e) +#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) \ + if (debug) print_double_string (w, s1, sz1, s2, sz2) + + +/* Print the fastmap in human-readable form. */ + +void +print_fastmap (fastmap) + char *fastmap; +{ + unsigned was_a_range = 0; + unsigned i = 0; + + while (i < (1 << BYTEWIDTH)) + { + if (fastmap[i++]) + { + was_a_range = 0; + putchar (i - 1); + while (i < (1 << BYTEWIDTH) && fastmap[i]) + { + was_a_range = 1; + i++; + } + if (was_a_range) + { + printf ("-"); + putchar (i - 1); + } + } + } + putchar ('\n'); +} + + +/* Print a compiled pattern string in human-readable form, starting at + the START pointer into it and ending just before the pointer END. */ + +void +print_partial_compiled_pattern (start, end) + unsigned char *start; + unsigned char *end; +{ + int mcnt, mcnt2; + unsigned char *p1; + unsigned char *p = start; + unsigned char *pend = end; + + if (start == NULL) + { + printf ("(null)\n"); + return; + } + + /* Loop over pattern commands. */ + while (p < pend) + { + printf ("%d:\t", p - start); + + switch ((re_opcode_t) *p++) + { + case no_op: + printf ("/no_op"); + break; + + case exactn: + mcnt = *p++; + printf ("/exactn/%d", mcnt); + do + { + putchar ('/'); + putchar (*p++); + } + while (--mcnt); + break; + + case start_memory: + mcnt = *p++; + printf ("/start_memory/%d/%d", mcnt, *p++); + break; + + case stop_memory: + mcnt = *p++; + printf ("/stop_memory/%d/%d", mcnt, *p++); + break; + + case duplicate: + printf ("/duplicate/%d", *p++); + break; + + case anychar: + printf ("/anychar"); + break; + + case charset: + case charset_not: + { + register int c, last = -100; + register int in_range = 0; + + printf ("/charset [%s", + (re_opcode_t) *(p - 1) == charset_not ? "^" : ""); + + assert (p + *p < pend); + + for (c = 0; c < 256; c++) + if (c / 8 < *p + && (p[1 + (c/8)] & (1 << (c % 8)))) + { + /* Are we starting a range? */ + if (last + 1 == c && ! in_range) + { + putchar ('-'); + in_range = 1; + } + /* Have we broken a range? */ + else if (last + 1 != c && in_range) + { + putchar (last); + in_range = 0; + } + + if (! in_range) + putchar (c); + + last = c; + } + + if (in_range) + putchar (last); + + putchar (']'); + + p += 1 + *p; + } + break; + + case begline: + printf ("/begline"); + break; + + case endline: + printf ("/endline"); + break; + + case on_failure_jump: + extract_number_and_incr (&mcnt, &p); + printf ("/on_failure_jump to %d", p + mcnt - start); + break; + + case on_failure_keep_string_jump: + extract_number_and_incr (&mcnt, &p); + printf ("/on_failure_keep_string_jump to %d", p + mcnt - start); + break; + + case dummy_failure_jump: + extract_number_and_incr (&mcnt, &p); + printf ("/dummy_failure_jump to %d", p + mcnt - start); + break; + + case push_dummy_failure: + printf ("/push_dummy_failure"); + break; + + case maybe_pop_jump: + extract_number_and_incr (&mcnt, &p); + printf ("/maybe_pop_jump to %d", p + mcnt - start); + break; + + case pop_failure_jump: + extract_number_and_incr (&mcnt, &p); + printf ("/pop_failure_jump to %d", p + mcnt - start); + break; + + case jump_past_alt: + extract_number_and_incr (&mcnt, &p); + printf ("/jump_past_alt to %d", p + mcnt - start); + break; + + case jump: + extract_number_and_incr (&mcnt, &p); + printf ("/jump to %d", p + mcnt - start); + break; + + case succeed_n: + extract_number_and_incr (&mcnt, &p); + p1 = p + mcnt; + extract_number_and_incr (&mcnt2, &p); + printf ("/succeed_n to %d, %d times", p1 - start, mcnt2); + break; + + case jump_n: + extract_number_and_incr (&mcnt, &p); + p1 = p + mcnt; + extract_number_and_incr (&mcnt2, &p); + printf ("/jump_n to %d, %d times", p1 - start, mcnt2); + break; + + case set_number_at: + extract_number_and_incr (&mcnt, &p); + p1 = p + mcnt; + extract_number_and_incr (&mcnt2, &p); + printf ("/set_number_at location %d to %d", p1 - start, mcnt2); + break; + + case wordbound: + printf ("/wordbound"); + break; + + case notwordbound: + printf ("/notwordbound"); + break; + + case wordbeg: + printf ("/wordbeg"); + break; + + case wordend: + printf ("/wordend"); + +#ifdef emacs + case before_dot: + printf ("/before_dot"); + break; + + case at_dot: + printf ("/at_dot"); + break; + + case after_dot: + printf ("/after_dot"); + break; + + case syntaxspec: + printf ("/syntaxspec"); + mcnt = *p++; + printf ("/%d", mcnt); + break; + + case notsyntaxspec: + printf ("/notsyntaxspec"); + mcnt = *p++; + printf ("/%d", mcnt); + break; +#endif /* emacs */ + + case wordchar: + printf ("/wordchar"); + break; + + case notwordchar: + printf ("/notwordchar"); + break; + + case begbuf: + printf ("/begbuf"); + break; + + case endbuf: + printf ("/endbuf"); + break; + + default: + printf ("?%d", *(p-1)); + } + + putchar ('\n'); + } + + printf ("%d:\tend of pattern.\n", p - start); +} + + +void +print_compiled_pattern (bufp) + struct re_pattern_buffer *bufp; +{ + unsigned char *buffer = bufp->buffer; + + print_partial_compiled_pattern (buffer, buffer + bufp->used); + printf ("%ld bytes used/%ld bytes allocated.\n", + bufp->used, bufp->allocated); + + if (bufp->fastmap_accurate && bufp->fastmap) + { + printf ("fastmap: "); + print_fastmap (bufp->fastmap); + } + + printf ("re_nsub: %d\t", bufp->re_nsub); + printf ("regs_alloc: %d\t", bufp->regs_allocated); + printf ("can_be_null: %d\t", bufp->can_be_null); + printf ("newline_anchor: %d\n", bufp->newline_anchor); + printf ("no_sub: %d\t", bufp->no_sub); + printf ("not_bol: %d\t", bufp->not_bol); + printf ("not_eol: %d\t", bufp->not_eol); + printf ("syntax: %lx\n", bufp->syntax); + /* Perhaps we should print the translate table? */ +} + + +void +print_double_string (where, string1, size1, string2, size2) + const char *where; + const char *string1; + const char *string2; + int size1; + int size2; +{ + int this_char; + + if (where == NULL) + printf ("(null)"); + else + { + if (FIRST_STRING_P (where)) + { + for (this_char = where - string1; this_char < size1; this_char++) + putchar (string1[this_char]); + + where = string2; + } + + for (this_char = where - string2; this_char < size2; this_char++) + putchar (string2[this_char]); + } +} + +void +printchar (c) + int c; +{ + putc (c, stderr); +} + +#else /* not DEBUG */ + +#undef assert +#define assert(e) + +#define DEBUG_STATEMENT(e) +#define DEBUG_PRINT1(x) +#define DEBUG_PRINT2(x1, x2) +#define DEBUG_PRINT3(x1, x2, x3) +#define DEBUG_PRINT4(x1, x2, x3, x4) +#define DEBUG_PRINT_COMPILED_PATTERN(p, s, e) +#define DEBUG_PRINT_DOUBLE_STRING(w, s1, sz1, s2, sz2) + +#endif /* not DEBUG */ + +/* Set by `re_set_syntax' to the current regexp syntax to recognize. Can + also be assigned to arbitrarily: each pattern buffer stores its own + syntax, so it can be changed between regex compilations. */ +/* This has no initializer because initialized variables in Emacs + become read-only after dumping. */ +reg_syntax_t re_syntax_options; + + +/* Specify the precise syntax of regexps for compilation. This provides + for compatibility for various utilities which historically have + different, incompatible syntaxes. + + The argument SYNTAX is a bit mask comprised of the various bits + defined in regex.h. We return the old syntax. */ + +reg_syntax_t +re_set_syntax (syntax) + reg_syntax_t syntax; +{ + reg_syntax_t ret = re_syntax_options; + + re_syntax_options = syntax; +#ifdef DEBUG + if (syntax & RE_DEBUG) + debug = 1; + else if (debug) /* was on but now is not */ + debug = 0; +#endif /* DEBUG */ + return ret; +} + +/* This table gives an error message for each of the error codes listed + in regex.h. Obviously the order here has to be same as there. + POSIX doesn't require that we do anything for REG_NOERROR, + but why not be nice? */ + +static const char *re_error_msgid[] = + { + gettext_noop ("Success"), /* REG_NOERROR */ + gettext_noop ("No match"), /* REG_NOMATCH */ + gettext_noop ("Invalid regular expression"), /* REG_BADPAT */ + gettext_noop ("Invalid collation character"), /* REG_ECOLLATE */ + gettext_noop ("Invalid character class name"), /* REG_ECTYPE */ + gettext_noop ("Trailing backslash"), /* REG_EESCAPE */ + gettext_noop ("Invalid back reference"), /* REG_ESUBREG */ + gettext_noop ("Unmatched [ or [^"), /* REG_EBRACK */ + gettext_noop ("Unmatched ( or \\("), /* REG_EPAREN */ + gettext_noop ("Unmatched \\{"), /* REG_EBRACE */ + gettext_noop ("Invalid content of \\{\\}"), /* REG_BADBR */ + gettext_noop ("Invalid range end"), /* REG_ERANGE */ + gettext_noop ("Memory exhausted"), /* REG_ESPACE */ + gettext_noop ("Invalid preceding regular expression"), /* REG_BADRPT */ + gettext_noop ("Premature end of regular expression"), /* REG_EEND */ + gettext_noop ("Regular expression too big"), /* REG_ESIZE */ + gettext_noop ("Unmatched ) or \\)"), /* REG_ERPAREN */ + }; + +/* Avoiding alloca during matching, to placate r_alloc. */ + +/* Define MATCH_MAY_ALLOCATE unless we need to make sure that the + searching and matching functions should not call alloca. On some + systems, alloca is implemented in terms of malloc, and if we're + using the relocating allocator routines, then malloc could cause a + relocation, which might (if the strings being searched are in the + ralloc heap) shift the data out from underneath the regexp + routines. + + Here's another reason to avoid allocation: Emacs + processes input from X in a signal handler; processing X input may + call malloc; if input arrives while a matching routine is calling + malloc, then we're scrod. But Emacs can't just block input while + calling matching routines; then we don't notice interrupts when + they come in. So, Emacs blocks input around all regexp calls + except the matching calls, which it leaves unprotected, in the + faith that they will not malloc. */ + +/* Normally, this is fine. */ +#define MATCH_MAY_ALLOCATE + +/* When using GNU C, we are not REALLY using the C alloca, no matter + what config.h may say. So don't take precautions for it. */ +#ifdef __GNUC__ +#undef C_ALLOCA +#endif + +/* The match routines may not allocate if (1) they would do it with malloc + and (2) it's not safe for them to use malloc. + Note that if REL_ALLOC is defined, matching would not use malloc for the + failure stack, but we would still use it for the register vectors; + so REL_ALLOC should not affect this. */ +#if (defined (C_ALLOCA) || defined (REGEX_MALLOC)) && defined (emacs) +#undef MATCH_MAY_ALLOCATE +#endif + + +/* Failure stack declarations and macros; both re_compile_fastmap and + re_match_2 use a failure stack. These have to be macros because of + REGEX_ALLOCATE_STACK. */ + + +/* Number of failure points for which to initially allocate space + when matching. If this number is exceeded, we allocate more + space, so it is not a hard limit. */ +#ifndef INIT_FAILURE_ALLOC +#define INIT_FAILURE_ALLOC 5 +#endif + +/* Roughly the maximum number of failure points on the stack. Would be + exactly that if always used MAX_FAILURE_ITEMS items each time we failed. + This is a variable only so users of regex can assign to it; we never + change it ourselves. */ + +#ifdef INT_IS_16BIT + +#if defined (MATCH_MAY_ALLOCATE) +/* 4400 was enough to cause a crash on Alpha OSF/1, + whose default stack limit is 2mb. */ +long int re_max_failures = 4000; +#else +long int re_max_failures = 2000; +#endif + +union fail_stack_elt +{ + unsigned char *pointer; + long int integer; +}; + +typedef union fail_stack_elt fail_stack_elt_t; + +typedef struct +{ + fail_stack_elt_t *stack; + unsigned long int size; + unsigned long int avail; /* Offset of next open position. */ +} fail_stack_type; + +#else /* not INT_IS_16BIT */ + +#if defined (MATCH_MAY_ALLOCATE) +/* 4400 was enough to cause a crash on Alpha OSF/1, + whose default stack limit is 2mb. */ +int re_max_failures = 20000; +#else +int re_max_failures = 2000; +#endif + +union fail_stack_elt +{ + unsigned char *pointer; + int integer; +}; + +typedef union fail_stack_elt fail_stack_elt_t; + +typedef struct +{ + fail_stack_elt_t *stack; + unsigned size; + unsigned avail; /* Offset of next open position. */ +} fail_stack_type; + +#endif /* INT_IS_16BIT */ + +#define FAIL_STACK_EMPTY() (fail_stack.avail == 0) +#define FAIL_STACK_PTR_EMPTY() (fail_stack_ptr->avail == 0) +#define FAIL_STACK_FULL() (fail_stack.avail == fail_stack.size) + + +/* Define macros to initialize and free the failure stack. + Do `return -2' if the alloc fails. */ + +#ifdef MATCH_MAY_ALLOCATE +#define INIT_FAIL_STACK() \ + do { \ + fail_stack.stack = (fail_stack_elt_t *) \ + REGEX_ALLOCATE_STACK (INIT_FAILURE_ALLOC * sizeof (fail_stack_elt_t)); \ + \ + if (fail_stack.stack == NULL) \ + return -2; \ + \ + fail_stack.size = INIT_FAILURE_ALLOC; \ + fail_stack.avail = 0; \ + } while (0) + +#define RESET_FAIL_STACK() REGEX_FREE_STACK (fail_stack.stack) +#else +#define INIT_FAIL_STACK() \ + do { \ + fail_stack.avail = 0; \ + } while (0) + +#define RESET_FAIL_STACK() +#endif + + +/* Double the size of FAIL_STACK, up to approximately `re_max_failures' items. + + Return 1 if succeeds, and 0 if either ran out of memory + allocating space for it or it was already too large. + + REGEX_REALLOCATE_STACK requires `destination' be declared. */ + +#define DOUBLE_FAIL_STACK(fail_stack) \ + ((fail_stack).size > (unsigned) (re_max_failures * MAX_FAILURE_ITEMS) \ + ? 0 \ + : ((fail_stack).stack = (fail_stack_elt_t *) \ + REGEX_REALLOCATE_STACK ((fail_stack).stack, \ + (fail_stack).size * sizeof (fail_stack_elt_t), \ + ((fail_stack).size << 1) * sizeof (fail_stack_elt_t)), \ + \ + (fail_stack).stack == NULL \ + ? 0 \ + : ((fail_stack).size <<= 1, \ + 1))) + + +/* Push pointer POINTER on FAIL_STACK. + Return 1 if was able to do so and 0 if ran out of memory allocating + space to do so. */ +#define PUSH_PATTERN_OP(POINTER, FAIL_STACK) \ + ((FAIL_STACK_FULL () \ + && !DOUBLE_FAIL_STACK (FAIL_STACK)) \ + ? 0 \ + : ((FAIL_STACK).stack[(FAIL_STACK).avail++].pointer = POINTER, \ + 1)) + +/* Push a pointer value onto the failure stack. + Assumes the variable `fail_stack'. Probably should only + be called from within `PUSH_FAILURE_POINT'. */ +#define PUSH_FAILURE_POINTER(item) \ + fail_stack.stack[fail_stack.avail++].pointer = (unsigned char *) (item) + +/* This pushes an integer-valued item onto the failure stack. + Assumes the variable `fail_stack'. Probably should only + be called from within `PUSH_FAILURE_POINT'. */ +#define PUSH_FAILURE_INT(item) \ + fail_stack.stack[fail_stack.avail++].integer = (item) + +/* Push a fail_stack_elt_t value onto the failure stack. + Assumes the variable `fail_stack'. Probably should only + be called from within `PUSH_FAILURE_POINT'. */ +#define PUSH_FAILURE_ELT(item) \ + fail_stack.stack[fail_stack.avail++] = (item) + +/* These three POP... operations complement the three PUSH... operations. + All assume that `fail_stack' is nonempty. */ +#define POP_FAILURE_POINTER() fail_stack.stack[--fail_stack.avail].pointer +#define POP_FAILURE_INT() fail_stack.stack[--fail_stack.avail].integer +#define POP_FAILURE_ELT() fail_stack.stack[--fail_stack.avail] + +/* Used to omit pushing failure point id's when we're not debugging. */ +#ifdef DEBUG +#define DEBUG_PUSH PUSH_FAILURE_INT +#define DEBUG_POP(item_addr) (item_addr)->integer = POP_FAILURE_INT () +#else +#define DEBUG_PUSH(item) +#define DEBUG_POP(item_addr) +#endif + + +/* Push the information about the state we will need + if we ever fail back to it. + + Requires variables fail_stack, regstart, regend, reg_info, and + num_regs be declared. DOUBLE_FAIL_STACK requires `destination' be + declared. + + Does `return FAILURE_CODE' if runs out of memory. */ + +#define PUSH_FAILURE_POINT(pattern_place, string_place, failure_code) \ + do { \ + char *destination; \ + /* Must be int, so when we don't save any registers, the arithmetic \ + of 0 + -1 isn't done as unsigned. */ \ + /* Can't be int, since there is not a shred of a guarantee that int \ + is wide enough to hold a value of something to which pointer can \ + be assigned */ \ + s_reg_t this_reg; \ + \ + DEBUG_STATEMENT (failure_id++); \ + DEBUG_STATEMENT (nfailure_points_pushed++); \ + DEBUG_PRINT2 ("\nPUSH_FAILURE_POINT #%u:\n", failure_id); \ + DEBUG_PRINT2 (" Before push, next avail: %d\n", (fail_stack).avail);\ + DEBUG_PRINT2 (" size: %d\n", (fail_stack).size);\ + \ + DEBUG_PRINT2 (" slots needed: %d\n", NUM_FAILURE_ITEMS); \ + DEBUG_PRINT2 (" available: %d\n", REMAINING_AVAIL_SLOTS); \ + \ + /* Ensure we have enough space allocated for what we will push. */ \ + while (REMAINING_AVAIL_SLOTS < NUM_FAILURE_ITEMS) \ + { \ + if (!DOUBLE_FAIL_STACK (fail_stack)) \ + return failure_code; \ + \ + DEBUG_PRINT2 ("\n Doubled stack; size now: %d\n", \ + (fail_stack).size); \ + DEBUG_PRINT2 (" slots available: %d\n", REMAINING_AVAIL_SLOTS);\ + } \ + \ + /* Push the info, starting with the registers. */ \ + DEBUG_PRINT1 ("\n"); \ + \ + if (1) \ + for (this_reg = lowest_active_reg; this_reg <= highest_active_reg; \ + this_reg++) \ + { \ + DEBUG_PRINT2 (" Pushing reg: %d\n", this_reg); \ + DEBUG_STATEMENT (num_regs_pushed++); \ + \ + DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \ + PUSH_FAILURE_POINTER (regstart[this_reg]); \ + \ + DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \ + PUSH_FAILURE_POINTER (regend[this_reg]); \ + \ + DEBUG_PRINT2 (" info: 0x%x\n ", reg_info[this_reg]); \ + DEBUG_PRINT2 (" match_null=%d", \ + REG_MATCH_NULL_STRING_P (reg_info[this_reg])); \ + DEBUG_PRINT2 (" active=%d", IS_ACTIVE (reg_info[this_reg])); \ + DEBUG_PRINT2 (" matched_something=%d", \ + MATCHED_SOMETHING (reg_info[this_reg])); \ + DEBUG_PRINT2 (" ever_matched=%d", \ + EVER_MATCHED_SOMETHING (reg_info[this_reg])); \ + DEBUG_PRINT1 ("\n"); \ + PUSH_FAILURE_ELT (reg_info[this_reg].word); \ + } \ + \ + DEBUG_PRINT2 (" Pushing low active reg: %d\n", lowest_active_reg);\ + PUSH_FAILURE_INT (lowest_active_reg); \ + \ + DEBUG_PRINT2 (" Pushing high active reg: %d\n", highest_active_reg);\ + PUSH_FAILURE_INT (highest_active_reg); \ + \ + DEBUG_PRINT2 (" Pushing pattern 0x%x:\n", pattern_place); \ + DEBUG_PRINT_COMPILED_PATTERN (bufp, pattern_place, pend); \ + PUSH_FAILURE_POINTER (pattern_place); \ + \ + DEBUG_PRINT2 (" Pushing string 0x%x: `", string_place); \ + DEBUG_PRINT_DOUBLE_STRING (string_place, string1, size1, string2, \ + size2); \ + DEBUG_PRINT1 ("'\n"); \ + PUSH_FAILURE_POINTER (string_place); \ + \ + DEBUG_PRINT2 (" Pushing failure id: %u\n", failure_id); \ + DEBUG_PUSH (failure_id); \ + } while (0) + +/* This is the number of items that are pushed and popped on the stack + for each register. */ +#define NUM_REG_ITEMS 3 + +/* Individual items aside from the registers. */ +#ifdef DEBUG +#define NUM_NONREG_ITEMS 5 /* Includes failure point id. */ +#else +#define NUM_NONREG_ITEMS 4 +#endif + +/* We push at most this many items on the stack. */ +/* We used to use (num_regs - 1), which is the number of registers + this regexp will save; but that was changed to 5 + to avoid stack overflow for a regexp with lots of parens. */ +#define MAX_FAILURE_ITEMS (5 * NUM_REG_ITEMS + NUM_NONREG_ITEMS) + +/* We actually push this many items. */ +#define NUM_FAILURE_ITEMS \ + (((0 \ + ? 0 : highest_active_reg - lowest_active_reg + 1) \ + * NUM_REG_ITEMS) \ + + NUM_NONREG_ITEMS) + +/* How many items can still be added to the stack without overflowing it. */ +#define REMAINING_AVAIL_SLOTS ((fail_stack).size - (fail_stack).avail) + + +/* Pops what PUSH_FAIL_STACK pushes. + + We restore into the parameters, all of which should be lvalues: + STR -- the saved data position. + PAT -- the saved pattern position. + LOW_REG, HIGH_REG -- the highest and lowest active registers. + REGSTART, REGEND -- arrays of string positions. + REG_INFO -- array of information about each subexpression. + + Also assumes the variables `fail_stack' and (if debugging), `bufp', + `pend', `string1', `size1', `string2', and `size2'. */ + +#define POP_FAILURE_POINT(str, pat, low_reg, high_reg, regstart, regend, reg_info)\ +{ \ + DEBUG_STATEMENT (fail_stack_elt_t failure_id;) \ + s_reg_t this_reg; \ + const unsigned char *string_temp; \ + \ + assert (!FAIL_STACK_EMPTY ()); \ + \ + /* Remove failure points and point to how many regs pushed. */ \ + DEBUG_PRINT1 ("POP_FAILURE_POINT:\n"); \ + DEBUG_PRINT2 (" Before pop, next avail: %d\n", fail_stack.avail); \ + DEBUG_PRINT2 (" size: %d\n", fail_stack.size); \ + \ + assert (fail_stack.avail >= NUM_NONREG_ITEMS); \ + \ + DEBUG_POP (&failure_id); \ + DEBUG_PRINT2 (" Popping failure id: %u\n", failure_id); \ + \ + /* If the saved string location is NULL, it came from an \ + on_failure_keep_string_jump opcode, and we want to throw away the \ + saved NULL, thus retaining our current position in the string. */ \ + string_temp = POP_FAILURE_POINTER (); \ + if (string_temp != NULL) \ + str = (const char *) string_temp; \ + \ + DEBUG_PRINT2 (" Popping string 0x%x: `", str); \ + DEBUG_PRINT_DOUBLE_STRING (str, string1, size1, string2, size2); \ + DEBUG_PRINT1 ("'\n"); \ + \ + pat = (unsigned char *) POP_FAILURE_POINTER (); \ + DEBUG_PRINT2 (" Popping pattern 0x%x:\n", pat); \ + DEBUG_PRINT_COMPILED_PATTERN (bufp, pat, pend); \ + \ + /* Restore register info. */ \ + high_reg = (active_reg_t) POP_FAILURE_INT (); \ + DEBUG_PRINT2 (" Popping high active reg: %d\n", high_reg); \ + \ + low_reg = (active_reg_t) POP_FAILURE_INT (); \ + DEBUG_PRINT2 (" Popping low active reg: %d\n", low_reg); \ + \ + if (1) \ + for (this_reg = high_reg; this_reg >= low_reg; this_reg--) \ + { \ + DEBUG_PRINT2 (" Popping reg: %d\n", this_reg); \ + \ + reg_info[this_reg].word = POP_FAILURE_ELT (); \ + DEBUG_PRINT2 (" info: 0x%x\n", reg_info[this_reg]); \ + \ + regend[this_reg] = (const char *) POP_FAILURE_POINTER (); \ + DEBUG_PRINT2 (" end: 0x%x\n", regend[this_reg]); \ + \ + regstart[this_reg] = (const char *) POP_FAILURE_POINTER (); \ + DEBUG_PRINT2 (" start: 0x%x\n", regstart[this_reg]); \ + } \ + else \ + { \ + for (this_reg = highest_active_reg; this_reg > high_reg; this_reg--) \ + { \ + reg_info[this_reg].word.integer = 0; \ + regend[this_reg] = 0; \ + regstart[this_reg] = 0; \ + } \ + highest_active_reg = high_reg; \ + } \ + \ + set_regs_matched_done = 0; \ + DEBUG_STATEMENT (nfailure_points_popped++); \ +} /* POP_FAILURE_POINT */ + + + +/* Structure for per-register (a.k.a. per-group) information. + Other register information, such as the + starting and ending positions (which are addresses), and the list of + inner groups (which is a bits list) are maintained in separate + variables. + + We are making a (strictly speaking) nonportable assumption here: that + the compiler will pack our bit fields into something that fits into + the type of `word', i.e., is something that fits into one item on the + failure stack. */ + + +/* Declarations and macros for re_match_2. */ + +typedef union +{ + fail_stack_elt_t word; + struct + { + /* This field is one if this group can match the empty string, + zero if not. If not yet determined, `MATCH_NULL_UNSET_VALUE'. */ +#define MATCH_NULL_UNSET_VALUE 3 + unsigned match_null_string_p : 2; + unsigned is_active : 1; + unsigned matched_something : 1; + unsigned ever_matched_something : 1; + } bits; +} register_info_type; + +#define REG_MATCH_NULL_STRING_P(R) ((R).bits.match_null_string_p) +#define IS_ACTIVE(R) ((R).bits.is_active) +#define MATCHED_SOMETHING(R) ((R).bits.matched_something) +#define EVER_MATCHED_SOMETHING(R) ((R).bits.ever_matched_something) + + +/* Call this when have matched a real character; it sets `matched' flags + for the subexpressions which we are currently inside. Also records + that those subexprs have matched. */ +#define SET_REGS_MATCHED() \ + do \ + { \ + if (!set_regs_matched_done) \ + { \ + active_reg_t r; \ + set_regs_matched_done = 1; \ + for (r = lowest_active_reg; r <= highest_active_reg; r++) \ + { \ + MATCHED_SOMETHING (reg_info[r]) \ + = EVER_MATCHED_SOMETHING (reg_info[r]) \ + = 1; \ + } \ + } \ + } \ + while (0) + +/* Registers are set to a sentinel when they haven't yet matched. */ +static char reg_unset_dummy; +#define REG_UNSET_VALUE (®_unset_dummy) +#define REG_UNSET(e) ((e) == REG_UNSET_VALUE) + +/* Subroutine declarations and macros for regex_compile. */ + +static reg_errcode_t regex_compile _RE_ARGS ((const char *pattern, size_t size, + reg_syntax_t syntax, + struct re_pattern_buffer *bufp)); +static void store_op1 _RE_ARGS ((re_opcode_t op, unsigned char *loc, int arg)); +static void store_op2 _RE_ARGS ((re_opcode_t op, unsigned char *loc, + int arg1, int arg2)); +static void insert_op1 _RE_ARGS ((re_opcode_t op, unsigned char *loc, + int arg, unsigned char *end)); +static void insert_op2 _RE_ARGS ((re_opcode_t op, unsigned char *loc, + int arg1, int arg2, unsigned char *end)); +static boolean at_begline_loc_p _RE_ARGS ((const char *pattern, const char *p, + reg_syntax_t syntax)); +static boolean at_endline_loc_p _RE_ARGS ((const char *p, const char *pend, + reg_syntax_t syntax)); +static reg_errcode_t compile_range _RE_ARGS ((const char **p_ptr, + const char *pend, + char *translate, + reg_syntax_t syntax, + unsigned char *b)); + +/* Fetch the next character in the uncompiled pattern---translating it + if necessary. Also cast from a signed character in the constant + string passed to us by the user to an unsigned char that we can use + as an array index (in, e.g., `translate'). */ +#ifndef PATFETCH +#define PATFETCH(c) \ + do {if (p == pend) return REG_EEND; \ + c = (unsigned char) *p++; \ + if (translate) c = (unsigned char) translate[c]; \ + } while (0) +#endif + +/* Fetch the next character in the uncompiled pattern, with no + translation. */ +#define PATFETCH_RAW(c) \ + do {if (p == pend) return REG_EEND; \ + c = (unsigned char) *p++; \ + } while (0) + +/* Go backwards one character in the pattern. */ +#define PATUNFETCH p-- + + +/* If `translate' is non-null, return translate[D], else just D. We + cast the subscript to translate because some data is declared as + `char *', to avoid warnings when a string constant is passed. But + when we use a character as a subscript we must make it unsigned. */ +#ifndef TRANSLATE +#define TRANSLATE(d) \ + (translate ? (char) translate[(unsigned char) (d)] : (d)) +#endif + + +/* Macros for outputting the compiled pattern into `buffer'. */ + +/* If the buffer isn't allocated when it comes in, use this. */ +#define INIT_BUF_SIZE 32 + +/* Make sure we have at least N more bytes of space in buffer. */ +#define GET_BUFFER_SPACE(n) \ + while ((unsigned long) (b - bufp->buffer + (n)) > bufp->allocated) \ + EXTEND_BUFFER () + +/* Make sure we have one more byte of buffer space and then add C to it. */ +#define BUF_PUSH(c) \ + do { \ + GET_BUFFER_SPACE (1); \ + *b++ = (unsigned char) (c); \ + } while (0) + + +/* Ensure we have two more bytes of buffer space and then append C1 and C2. */ +#define BUF_PUSH_2(c1, c2) \ + do { \ + GET_BUFFER_SPACE (2); \ + *b++ = (unsigned char) (c1); \ + *b++ = (unsigned char) (c2); \ + } while (0) + + +/* As with BUF_PUSH_2, except for three bytes. */ +#define BUF_PUSH_3(c1, c2, c3) \ + do { \ + GET_BUFFER_SPACE (3); \ + *b++ = (unsigned char) (c1); \ + *b++ = (unsigned char) (c2); \ + *b++ = (unsigned char) (c3); \ + } while (0) + + +/* Store a jump with opcode OP at LOC to location TO. We store a + relative address offset by the three bytes the jump itself occupies. */ +#define STORE_JUMP(op, loc, to) \ + store_op1 (op, loc, (int) ((to) - (loc) - 3)) + +/* Likewise, for a two-argument jump. */ +#define STORE_JUMP2(op, loc, to, arg) \ + store_op2 (op, loc, (int) ((to) - (loc) - 3), arg) + +/* Like `STORE_JUMP', but for inserting. Assume `b' is the buffer end. */ +#define INSERT_JUMP(op, loc, to) \ + insert_op1 (op, loc, (int) ((to) - (loc) - 3), b) + +/* Like `STORE_JUMP2', but for inserting. Assume `b' is the buffer end. */ +#define INSERT_JUMP2(op, loc, to, arg) \ + insert_op2 (op, loc, (int) ((to) - (loc) - 3), arg, b) + + +/* This is not an arbitrary limit: the arguments which represent offsets + into the pattern are two bytes long. So if 2^16 bytes turns out to + be too small, many things would have to change. */ +/* Any other compiler which, like MSC, has allocation limit below 2^16 + bytes will have to use approach similar to what was done below for + MSC and drop MAX_BUF_SIZE a bit. Otherwise you may end up + reallocating to 0 bytes. Such thing is not going to work too well. + You have been warned!! */ +#if defined(_MSC_VER) && !defined(WIN32) +/* Microsoft C 16-bit versions limit malloc to approx 65512 bytes. + The REALLOC define eliminates a flurry of conversion warnings, + but is not required. */ +#define MAX_BUF_SIZE 65500L +#define REALLOC(p,s) realloc ((p), (size_t) (s)) +#else +#define MAX_BUF_SIZE (1L << 16) +#define REALLOC(p,s) realloc ((p), (s)) +#endif + +/* Extend the buffer by twice its current size via realloc and + reset the pointers that pointed into the old block to point to the + correct places in the new one. If extending the buffer results in it + being larger than MAX_BUF_SIZE, then flag memory exhausted. */ +#define EXTEND_BUFFER() \ + do { \ + unsigned char *old_buffer = bufp->buffer; \ + if (bufp->allocated == MAX_BUF_SIZE) \ + return REG_ESIZE; \ + bufp->allocated <<= 1; \ + if (bufp->allocated > MAX_BUF_SIZE) \ + bufp->allocated = MAX_BUF_SIZE; \ + bufp->buffer = (unsigned char *) REALLOC (bufp->buffer, bufp->allocated);\ + if (bufp->buffer == NULL) \ + return REG_ESPACE; \ + /* If the buffer moved, move all the pointers into it. */ \ + if (old_buffer != bufp->buffer) \ + { \ + b = (b - old_buffer) + bufp->buffer; \ + begalt = (begalt - old_buffer) + bufp->buffer; \ + if (fixup_alt_jump) \ + fixup_alt_jump = (fixup_alt_jump - old_buffer) + bufp->buffer;\ + if (laststart) \ + laststart = (laststart - old_buffer) + bufp->buffer; \ + if (pending_exact) \ + pending_exact = (pending_exact - old_buffer) + bufp->buffer; \ + } \ + } while (0) + + +/* Since we have one byte reserved for the register number argument to + {start,stop}_memory, the maximum number of groups we can report + things about is what fits in that byte. */ +#define MAX_REGNUM 255 + +/* But patterns can have more than `MAX_REGNUM' registers. We just + ignore the excess. */ +typedef unsigned regnum_t; + + +/* Macros for the compile stack. */ + +/* Since offsets can go either forwards or backwards, this type needs to + be able to hold values from -(MAX_BUF_SIZE - 1) to MAX_BUF_SIZE - 1. */ +/* int may be not enough when sizeof(int) == 2. */ +typedef long pattern_offset_t; + +typedef struct +{ + pattern_offset_t begalt_offset; + pattern_offset_t fixup_alt_jump; + pattern_offset_t inner_group_offset; + pattern_offset_t laststart_offset; + regnum_t regnum; +} compile_stack_elt_t; + + +typedef struct +{ + compile_stack_elt_t *stack; + unsigned size; + unsigned avail; /* Offset of next open position. */ +} compile_stack_type; + + +#define INIT_COMPILE_STACK_SIZE 32 + +#define COMPILE_STACK_EMPTY (compile_stack.avail == 0) +#define COMPILE_STACK_FULL (compile_stack.avail == compile_stack.size) + +/* The next available element. */ +#define COMPILE_STACK_TOP (compile_stack.stack[compile_stack.avail]) + + +/* Set the bit for character C in a list. */ +#define SET_LIST_BIT(c) \ + (b[((unsigned char) (c)) / BYTEWIDTH] \ + |= 1 << (((unsigned char) c) % BYTEWIDTH)) + + +/* Get the next unsigned number in the uncompiled pattern. */ +#define GET_UNSIGNED_NUMBER(num) \ + { if (p != pend) \ + { \ + PATFETCH (c); \ + while (ISDIGIT (c)) \ + { \ + if (num < 0) \ + num = 0; \ + num = num * 10 + c - '0'; \ + if (p == pend) \ + break; \ + PATFETCH (c); \ + } \ + } \ + } + +#if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H) +/* The GNU C library provides support for user-defined character classes + and the functions from ISO C amendement 1. */ +# ifdef CHARCLASS_NAME_MAX +# define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX +# else +/* This shouldn't happen but some implementation might still have this + problem. Use a reasonable default value. */ +# define CHAR_CLASS_MAX_LENGTH 256 +# endif + +# define IS_CHAR_CLASS(string) wctype (string) +#else +# define CHAR_CLASS_MAX_LENGTH 6 /* Namely, `xdigit'. */ + +# define IS_CHAR_CLASS(string) \ + (STREQ (string, "alpha") || STREQ (string, "upper") \ + || STREQ (string, "lower") || STREQ (string, "digit") \ + || STREQ (string, "alnum") || STREQ (string, "xdigit") \ + || STREQ (string, "space") || STREQ (string, "print") \ + || STREQ (string, "punct") || STREQ (string, "graph") \ + || STREQ (string, "cntrl") || STREQ (string, "blank")) +#endif + +#ifndef MATCH_MAY_ALLOCATE + +/* If we cannot allocate large objects within re_match_2_internal, + we make the fail stack and register vectors global. + The fail stack, we grow to the maximum size when a regexp + is compiled. + The register vectors, we adjust in size each time we + compile a regexp, according to the number of registers it needs. */ + +static fail_stack_type fail_stack; + +/* Size with which the following vectors are currently allocated. + That is so we can make them bigger as needed, + but never make them smaller. */ +static int regs_allocated_size; + +static const char ** regstart, ** regend; +static const char ** old_regstart, ** old_regend; +static const char **best_regstart, **best_regend; +static register_info_type *reg_info; +static const char **reg_dummy; +static register_info_type *reg_info_dummy; + +/* Make the register vectors big enough for NUM_REGS registers, + but don't make them smaller. */ + +static +regex_grow_registers (num_regs) + int num_regs; +{ + if (num_regs > regs_allocated_size) + { + RETALLOC_IF (regstart, num_regs, const char *); + RETALLOC_IF (regend, num_regs, const char *); + RETALLOC_IF (old_regstart, num_regs, const char *); + RETALLOC_IF (old_regend, num_regs, const char *); + RETALLOC_IF (best_regstart, num_regs, const char *); + RETALLOC_IF (best_regend, num_regs, const char *); + RETALLOC_IF (reg_info, num_regs, register_info_type); + RETALLOC_IF (reg_dummy, num_regs, const char *); + RETALLOC_IF (reg_info_dummy, num_regs, register_info_type); + + regs_allocated_size = num_regs; + } +} + +#endif /* not MATCH_MAY_ALLOCATE */ + +static boolean group_in_compile_stack _RE_ARGS ((compile_stack_type + compile_stack, + regnum_t regnum)); + +/* `regex_compile' compiles PATTERN (of length SIZE) according to SYNTAX. + Returns one of error codes defined in `regex.h', or zero for success. + + Assumes the `allocated' (and perhaps `buffer') and `translate' + fields are set in BUFP on entry. + + If it succeeds, results are put in BUFP (if it returns an error, the + contents of BUFP are undefined): + `buffer' is the compiled pattern; + `syntax' is set to SYNTAX; + `used' is set to the length of the compiled pattern; + `fastmap_accurate' is zero; + `re_nsub' is the number of subexpressions in PATTERN; + `not_bol' and `not_eol' are zero; + + The `fastmap' and `newline_anchor' fields are neither + examined nor set. */ + +/* Return, freeing storage we allocated. */ +#define FREE_STACK_RETURN(value) \ + return (free (compile_stack.stack), value) + +static reg_errcode_t +regex_compile (pattern, size, syntax, bufp) + const char *pattern; + size_t size; + reg_syntax_t syntax; + struct re_pattern_buffer *bufp; +{ + /* We fetch characters from PATTERN here. Even though PATTERN is + `char *' (i.e., signed), we declare these variables as unsigned, so + they can be reliably used as array indices. */ + register unsigned char c, c1; + + /* A random temporary spot in PATTERN. */ + const char *p1; + + /* Points to the end of the buffer, where we should append. */ + register unsigned char *b; + + /* Keeps track of unclosed groups. */ + compile_stack_type compile_stack; + + /* Points to the current (ending) position in the pattern. */ + const char *p = pattern; + const char *pend = pattern + size; + + /* How to translate the characters in the pattern. */ + RE_TRANSLATE_TYPE translate = bufp->translate; + + /* Address of the count-byte of the most recently inserted `exactn' + command. This makes it possible to tell if a new exact-match + character can be added to that command or if the character requires + a new `exactn' command. */ + unsigned char *pending_exact = 0; + + /* Address of start of the most recently finished expression. + This tells, e.g., postfix * where to find the start of its + operand. Reset at the beginning of groups and alternatives. */ + unsigned char *laststart = 0; + + /* Address of beginning of regexp, or inside of last group. */ + unsigned char *begalt; + + /* Place in the uncompiled pattern (i.e., the {) to + which to go back if the interval is invalid. */ + const char *beg_interval; + + /* Address of the place where a forward jump should go to the end of + the containing expression. Each alternative of an `or' -- except the + last -- ends with a forward jump of this sort. */ + unsigned char *fixup_alt_jump = 0; + + /* Counts open-groups as they are encountered. Remembered for the + matching close-group on the compile stack, so the same register + number is put in the stop_memory as the start_memory. */ + regnum_t regnum = 0; + +#ifdef DEBUG + DEBUG_PRINT1 ("\nCompiling pattern: "); + if (debug) + { + unsigned debug_count; + + for (debug_count = 0; debug_count < size; debug_count++) + putchar (pattern[debug_count]); + putchar ('\n'); + } +#endif /* DEBUG */ + + /* Initialize the compile stack. */ + compile_stack.stack = TALLOC (INIT_COMPILE_STACK_SIZE, compile_stack_elt_t); + if (compile_stack.stack == NULL) + return REG_ESPACE; + + compile_stack.size = INIT_COMPILE_STACK_SIZE; + compile_stack.avail = 0; + + /* Initialize the pattern buffer. */ + bufp->syntax = syntax; + bufp->fastmap_accurate = 0; + bufp->not_bol = bufp->not_eol = 0; + + /* Set `used' to zero, so that if we return an error, the pattern + printer (for debugging) will think there's no pattern. We reset it + at the end. */ + bufp->used = 0; + + /* Always count groups, whether or not bufp->no_sub is set. */ + bufp->re_nsub = 0; + +#if !defined (emacs) && !defined (SYNTAX_TABLE) + /* Initialize the syntax table. */ + init_syntax_once (); +#endif + + if (bufp->allocated == 0) + { + if (bufp->buffer) + { /* If zero allocated, but buffer is non-null, try to realloc + enough space. This loses if buffer's address is bogus, but + that is the user's responsibility. */ + RETALLOC (bufp->buffer, INIT_BUF_SIZE, unsigned char); + } + else + { /* Caller did not allocate a buffer. Do it for them. */ + bufp->buffer = TALLOC (INIT_BUF_SIZE, unsigned char); + } + if (!bufp->buffer) FREE_STACK_RETURN (REG_ESPACE); + + bufp->allocated = INIT_BUF_SIZE; + } + + begalt = b = bufp->buffer; + + /* Loop through the uncompiled pattern until we're at the end. */ + while (p != pend) + { + PATFETCH (c); + + switch (c) + { + case '^': + { + if ( /* If at start of pattern, it's an operator. */ + p == pattern + 1 + /* If context independent, it's an operator. */ + || syntax & RE_CONTEXT_INDEP_ANCHORS + /* Otherwise, depends on what's come before. */ + || at_begline_loc_p (pattern, p, syntax)) + BUF_PUSH (begline); + else + goto normal_char; + } + break; + + + case '$': + { + if ( /* If at end of pattern, it's an operator. */ + p == pend + /* If context independent, it's an operator. */ + || syntax & RE_CONTEXT_INDEP_ANCHORS + /* Otherwise, depends on what's next. */ + || at_endline_loc_p (p, pend, syntax)) + BUF_PUSH (endline); + else + goto normal_char; + } + break; + + + case '+': + case '?': + if ((syntax & RE_BK_PLUS_QM) + || (syntax & RE_LIMITED_OPS)) + goto normal_char; + handle_plus: + case '*': + /* If there is no previous pattern... */ + if (!laststart) + { + if (syntax & RE_CONTEXT_INVALID_OPS) + FREE_STACK_RETURN (REG_BADRPT); + else if (!(syntax & RE_CONTEXT_INDEP_OPS)) + goto normal_char; + } + + { + /* Are we optimizing this jump? */ + boolean keep_string_p = false; + + /* 1 means zero (many) matches is allowed. */ + char zero_times_ok = 0, many_times_ok = 0; + + /* If there is a sequence of repetition chars, collapse it + down to just one (the right one). We can't combine + interval operators with these because of, e.g., `a{2}*', + which should only match an even number of `a's. */ + + for (;;) + { + zero_times_ok |= c != '+'; + many_times_ok |= c != '?'; + + if (p == pend) + break; + + PATFETCH (c); + + if (c == '*' + || (!(syntax & RE_BK_PLUS_QM) && (c == '+' || c == '?'))) + ; + + else if (syntax & RE_BK_PLUS_QM && c == '\\') + { + if (p == pend) FREE_STACK_RETURN (REG_EESCAPE); + + PATFETCH (c1); + if (!(c1 == '+' || c1 == '?')) + { + PATUNFETCH; + PATUNFETCH; + break; + } + + c = c1; + } + else + { + PATUNFETCH; + break; + } + + /* If we get here, we found another repeat character. */ + } + + /* Star, etc. applied to an empty pattern is equivalent + to an empty pattern. */ + if (!laststart) + break; + + /* Now we know whether or not zero matches is allowed + and also whether or not two or more matches is allowed. */ + if (many_times_ok) + { /* More than one repetition is allowed, so put in at the + end a backward relative jump from `b' to before the next + jump we're going to put in below (which jumps from + laststart to after this jump). + + But if we are at the `*' in the exact sequence `.*\n', + insert an unconditional jump backwards to the ., + instead of the beginning of the loop. This way we only + push a failure point once, instead of every time + through the loop. */ + assert (p - 1 > pattern); + + /* Allocate the space for the jump. */ + GET_BUFFER_SPACE (3); + + /* We know we are not at the first character of the pattern, + because laststart was nonzero. And we've already + incremented `p', by the way, to be the character after + the `*'. Do we have to do something analogous here + for null bytes, because of RE_DOT_NOT_NULL? */ + if (TRANSLATE (*(p - 2)) == TRANSLATE ('.') + && zero_times_ok + && p < pend && TRANSLATE (*p) == TRANSLATE ('\n') + && !(syntax & RE_DOT_NEWLINE)) + { /* We have .*\n. */ + STORE_JUMP (jump, b, laststart); + keep_string_p = true; + } + else + /* Anything else. */ + STORE_JUMP (maybe_pop_jump, b, laststart - 3); + + /* We've added more stuff to the buffer. */ + b += 3; + } + + /* On failure, jump from laststart to b + 3, which will be the + end of the buffer after this jump is inserted. */ + GET_BUFFER_SPACE (3); + INSERT_JUMP (keep_string_p ? on_failure_keep_string_jump + : on_failure_jump, + laststart, b + 3); + pending_exact = 0; + b += 3; + + if (!zero_times_ok) + { + /* At least one repetition is required, so insert a + `dummy_failure_jump' before the initial + `on_failure_jump' instruction of the loop. This + effects a skip over that instruction the first time + we hit that loop. */ + GET_BUFFER_SPACE (3); + INSERT_JUMP (dummy_failure_jump, laststart, laststart + 6); + b += 3; + } + } + break; + + + case '.': + laststart = b; + BUF_PUSH (anychar); + break; + + + case '[': + { + boolean had_char_class = false; + + if (p == pend) FREE_STACK_RETURN (REG_EBRACK); + + /* Ensure that we have enough space to push a charset: the + opcode, the length count, and the bitset; 34 bytes in all. */ + GET_BUFFER_SPACE (34); + + laststart = b; + + /* We test `*p == '^' twice, instead of using an if + statement, so we only need one BUF_PUSH. */ + BUF_PUSH (*p == '^' ? charset_not : charset); + if (*p == '^') + p++; + + /* Remember the first position in the bracket expression. */ + p1 = p; + + /* Push the number of bytes in the bitmap. */ + BUF_PUSH ((1 << BYTEWIDTH) / BYTEWIDTH); + + /* Clear the whole map. */ + bzero (b, (1 << BYTEWIDTH) / BYTEWIDTH); + + /* charset_not matches newline according to a syntax bit. */ + if ((re_opcode_t) b[-2] == charset_not + && (syntax & RE_HAT_LISTS_NOT_NEWLINE)) + SET_LIST_BIT ('\n'); + + /* Read in characters and ranges, setting map bits. */ + for (;;) + { + if (p == pend) FREE_STACK_RETURN (REG_EBRACK); + + PATFETCH (c); + + /* \ might escape characters inside [...] and [^...]. */ + if ((syntax & RE_BACKSLASH_ESCAPE_IN_LISTS) && c == '\\') + { + if (p == pend) FREE_STACK_RETURN (REG_EESCAPE); + + PATFETCH (c1); + SET_LIST_BIT (c1); + continue; + } + + /* Could be the end of the bracket expression. If it's + not (i.e., when the bracket expression is `[]' so + far), the ']' character bit gets set way below. */ + if (c == ']' && p != p1 + 1) + break; + + /* Look ahead to see if it's a range when the last thing + was a character class. */ + if (had_char_class && c == '-' && *p != ']') + FREE_STACK_RETURN (REG_ERANGE); + + /* Look ahead to see if it's a range when the last thing + was a character: if this is a hyphen not at the + beginning or the end of a list, then it's the range + operator. */ + if (c == '-' + && !(p - 2 >= pattern && p[-2] == '[') + && !(p - 3 >= pattern && p[-3] == '[' && p[-2] == '^') + && *p != ']') + { + reg_errcode_t ret + = compile_range (&p, pend, translate, syntax, b); + if (ret != REG_NOERROR) FREE_STACK_RETURN (ret); + } + + else if (p[0] == '-' && p[1] != ']') + { /* This handles ranges made up of characters only. */ + reg_errcode_t ret; + + /* Move past the `-'. */ + PATFETCH (c1); + + ret = compile_range (&p, pend, translate, syntax, b); + if (ret != REG_NOERROR) FREE_STACK_RETURN (ret); + } + + /* See if we're at the beginning of a possible character + class. */ + + else if (syntax & RE_CHAR_CLASSES && c == '[' && *p == ':') + { /* Leave room for the null. */ + char str[CHAR_CLASS_MAX_LENGTH + 1]; + + PATFETCH (c); + c1 = 0; + + /* If pattern is `[[:'. */ + if (p == pend) FREE_STACK_RETURN (REG_EBRACK); + + for (;;) + { + PATFETCH (c); + if (c == ':' || c == ']' || p == pend + || c1 == CHAR_CLASS_MAX_LENGTH) + break; + str[c1++] = c; + } + str[c1] = '\0'; + + /* If isn't a word bracketed by `[:' and:`]': + undo the ending character, the letters, and leave + the leading `:' and `[' (but set bits for them). */ + if (c == ':' && *p == ']') + { +#if defined _LIBC || (defined HAVE_WCTYPE_H && defined HAVE_WCHAR_H) + boolean is_lower = STREQ (str, "lower"); + boolean is_upper = STREQ (str, "upper"); + wctype_t wt; + int ch; + + wt = wctype (str); + if (wt == 0) + FREE_STACK_RETURN (REG_ECTYPE); + + /* Throw away the ] at the end of the character + class. */ + PATFETCH (c); + + if (p == pend) FREE_STACK_RETURN (REG_EBRACK); + + for (ch = 0; ch < 1 << BYTEWIDTH; ++ch) + { + if (iswctype (btowc (ch), wt)) + SET_LIST_BIT (ch); + + if (translate && (is_upper || is_lower) + && (ISUPPER (ch) || ISLOWER (ch))) + SET_LIST_BIT (ch); + } + + had_char_class = true; +#else + int ch; + boolean is_alnum = STREQ (str, "alnum"); + boolean is_alpha = STREQ (str, "alpha"); + boolean is_blank = STREQ (str, "blank"); + boolean is_cntrl = STREQ (str, "cntrl"); + boolean is_digit = STREQ (str, "digit"); + boolean is_graph = STREQ (str, "graph"); + boolean is_lower = STREQ (str, "lower"); + boolean is_print = STREQ (str, "print"); + boolean is_punct = STREQ (str, "punct"); + boolean is_space = STREQ (str, "space"); + boolean is_upper = STREQ (str, "upper"); + boolean is_xdigit = STREQ (str, "xdigit"); + + if (!IS_CHAR_CLASS (str)) + FREE_STACK_RETURN (REG_ECTYPE); + + /* Throw away the ] at the end of the character + class. */ + PATFETCH (c); + + if (p == pend) FREE_STACK_RETURN (REG_EBRACK); + + for (ch = 0; ch < (1 << BYTEWIDTH); ch++) + { + /* This was split into 3 if's to + avoid an arbitrary limit in some compiler. */ + if ( (is_alnum && ISALNUM (ch)) + || (is_alpha && ISALPHA (ch)) + || (is_blank && ISBLANK (ch)) + || (is_cntrl && ISCNTRL (ch))) + SET_LIST_BIT (ch); + if ( (is_digit && ISDIGIT (ch)) + || (is_graph && ISGRAPH (ch)) + || (is_lower && ISLOWER (ch)) + || (is_print && ISPRINT (ch))) + SET_LIST_BIT (ch); + if ( (is_punct && ISPUNCT (ch)) + || (is_space && ISSPACE (ch)) + || (is_upper && ISUPPER (ch)) + || (is_xdigit && ISXDIGIT (ch))) + SET_LIST_BIT (ch); + if ( translate && (is_upper || is_lower) + && (ISUPPER (ch) || ISLOWER (ch))) + SET_LIST_BIT (ch); + } + had_char_class = true; +#endif /* libc || wctype.h */ + } + else + { + c1++; + while (c1--) + PATUNFETCH; + SET_LIST_BIT ('['); + SET_LIST_BIT (':'); + had_char_class = false; + } + } + else + { + had_char_class = false; + SET_LIST_BIT (c); + } + } + + /* Discard any (non)matching list bytes that are all 0 at the + end of the map. Decrease the map-length byte too. */ + while ((int) b[-1] > 0 && b[b[-1] - 1] == 0) + b[-1]--; + b += b[-1]; + } + break; + + + case '(': + if (syntax & RE_NO_BK_PARENS) + goto handle_open; + else + goto normal_char; + + + case ')': + if (syntax & RE_NO_BK_PARENS) + goto handle_close; + else + goto normal_char; + + + case '\n': + if (syntax & RE_NEWLINE_ALT) + goto handle_alt; + else + goto normal_char; + + + case '|': + if (syntax & RE_NO_BK_VBAR) + goto handle_alt; + else + goto normal_char; + + + case '{': + if (syntax & RE_INTERVALS && syntax & RE_NO_BK_BRACES) + goto handle_interval; + else + goto normal_char; + + + case '\\': + if (p == pend) FREE_STACK_RETURN (REG_EESCAPE); + + /* Do not translate the character after the \, so that we can + distinguish, e.g., \B from \b, even if we normally would + translate, e.g., B to b. */ + PATFETCH_RAW (c); + + switch (c) + { + case '(': + if (syntax & RE_NO_BK_PARENS) + goto normal_backslash; + + handle_open: + bufp->re_nsub++; + regnum++; + + if (COMPILE_STACK_FULL) + { + RETALLOC (compile_stack.stack, compile_stack.size << 1, + compile_stack_elt_t); + if (compile_stack.stack == NULL) return REG_ESPACE; + + compile_stack.size <<= 1; + } + + /* These are the values to restore when we hit end of this + group. They are all relative offsets, so that if the + whole pattern moves because of realloc, they will still + be valid. */ + COMPILE_STACK_TOP.begalt_offset = begalt - bufp->buffer; + COMPILE_STACK_TOP.fixup_alt_jump + = fixup_alt_jump ? fixup_alt_jump - bufp->buffer + 1 : 0; + COMPILE_STACK_TOP.laststart_offset = b - bufp->buffer; + COMPILE_STACK_TOP.regnum = regnum; + + /* We will eventually replace the 0 with the number of + groups inner to this one. But do not push a + start_memory for groups beyond the last one we can + represent in the compiled pattern. */ + if (regnum <= MAX_REGNUM) + { + COMPILE_STACK_TOP.inner_group_offset = b - bufp->buffer + 2; + BUF_PUSH_3 (start_memory, regnum, 0); + } + + compile_stack.avail++; + + fixup_alt_jump = 0; + laststart = 0; + begalt = b; + /* If we've reached MAX_REGNUM groups, then this open + won't actually generate any code, so we'll have to + clear pending_exact explicitly. */ + pending_exact = 0; + break; + + + case ')': + if (syntax & RE_NO_BK_PARENS) goto normal_backslash; + + if (COMPILE_STACK_EMPTY) { + if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD) + goto normal_backslash; + } + else FREE_STACK_RETURN (REG_ERPAREN); + + handle_close: + if (fixup_alt_jump) + { /* Push a dummy failure point at the end of the + alternative for a possible future + `pop_failure_jump' to pop. See comments at + `push_dummy_failure' in `re_match_2'. */ + BUF_PUSH (push_dummy_failure); + + /* We allocated space for this jump when we assigned + to `fixup_alt_jump', in the `handle_alt' case below. */ + STORE_JUMP (jump_past_alt, fixup_alt_jump, b - 1); + } + + /* See similar code for backslashed left paren above. */ + if (COMPILE_STACK_EMPTY) { + if (syntax & RE_UNMATCHED_RIGHT_PAREN_ORD) + goto normal_char; + } + else FREE_STACK_RETURN (REG_ERPAREN); + + /* Since we just checked for an empty stack above, this + ``can't happen''. */ + assert (compile_stack.avail != 0); + { + /* We don't just want to restore into `regnum', because + later groups should continue to be numbered higher, + as in `(ab)c(de)' -- the second group is #2. */ + regnum_t this_group_regnum; + + compile_stack.avail--; + begalt = bufp->buffer + COMPILE_STACK_TOP.begalt_offset; + fixup_alt_jump + = COMPILE_STACK_TOP.fixup_alt_jump + ? bufp->buffer + COMPILE_STACK_TOP.fixup_alt_jump - 1 + : 0; + laststart = bufp->buffer + COMPILE_STACK_TOP.laststart_offset; + this_group_regnum = COMPILE_STACK_TOP.regnum; + /* If we've reached MAX_REGNUM groups, then this open + won't actually generate any code, so we'll have to + clear pending_exact explicitly. */ + pending_exact = 0; + + /* We're at the end of the group, so now we know how many + groups were inside this one. */ + if (this_group_regnum <= MAX_REGNUM) + { + unsigned char *inner_group_loc + = bufp->buffer + COMPILE_STACK_TOP.inner_group_offset; + + *inner_group_loc = regnum - this_group_regnum; + BUF_PUSH_3 (stop_memory, this_group_regnum, + regnum - this_group_regnum); + } + } + break; + + + case '|': /* `\|'. */ + if (syntax & RE_LIMITED_OPS || syntax & RE_NO_BK_VBAR) + goto normal_backslash; + handle_alt: + if (syntax & RE_LIMITED_OPS) + goto normal_char; + + /* Insert before the previous alternative a jump which + jumps to this alternative if the former fails. */ + GET_BUFFER_SPACE (3); + INSERT_JUMP (on_failure_jump, begalt, b + 6); + pending_exact = 0; + b += 3; + + /* The alternative before this one has a jump after it + which gets executed if it gets matched. Adjust that + jump so it will jump to this alternative's analogous + jump (put in below, which in turn will jump to the next + (if any) alternative's such jump, etc.). The last such + jump jumps to the correct final destination. A picture: + _____ _____ + | | | | + | v | v + a | b | c + + If we are at `b', then fixup_alt_jump right now points to a + three-byte space after `a'. We'll put in the jump, set + fixup_alt_jump to right after `b', and leave behind three + bytes which we'll fill in when we get to after `c'. */ + + if (fixup_alt_jump) + STORE_JUMP (jump_past_alt, fixup_alt_jump, b); + + /* Mark and leave space for a jump after this alternative, + to be filled in later either by next alternative or + when know we're at the end of a series of alternatives. */ + fixup_alt_jump = b; + GET_BUFFER_SPACE (3); + b += 3; + + laststart = 0; + begalt = b; + break; + + + case '{': + /* If \{ is a literal. */ + if (!(syntax & RE_INTERVALS) + /* If we're at `\{' and it's not the open-interval + operator. */ + || ((syntax & RE_INTERVALS) && (syntax & RE_NO_BK_BRACES)) + || (p - 2 == pattern && p == pend)) + goto normal_backslash; + + handle_interval: + { + /* If got here, then the syntax allows intervals. */ + + /* At least (most) this many matches must be made. */ + int lower_bound = -1, upper_bound = -1; + + beg_interval = p - 1; + + if (p == pend) + { + if (syntax & RE_NO_BK_BRACES) + goto unfetch_interval; + else + FREE_STACK_RETURN (REG_EBRACE); + } + + GET_UNSIGNED_NUMBER (lower_bound); + + if (c == ',') + { + GET_UNSIGNED_NUMBER (upper_bound); + if (upper_bound < 0) upper_bound = RE_DUP_MAX; + } + else + /* Interval such as `{1}' => match exactly once. */ + upper_bound = lower_bound; + + if (lower_bound < 0 || upper_bound > RE_DUP_MAX + || lower_bound > upper_bound) + { + if (syntax & RE_NO_BK_BRACES) + goto unfetch_interval; + else + FREE_STACK_RETURN (REG_BADBR); + } + + if (!(syntax & RE_NO_BK_BRACES)) + { + if (c != '\\') FREE_STACK_RETURN (REG_EBRACE); + + PATFETCH (c); + } + + if (c != '}') + { + if (syntax & RE_NO_BK_BRACES) + goto unfetch_interval; + else + FREE_STACK_RETURN (REG_BADBR); + } + + /* We just parsed a valid interval. */ + + /* If it's invalid to have no preceding re. */ + if (!laststart) + { + if (syntax & RE_CONTEXT_INVALID_OPS) + FREE_STACK_RETURN (REG_BADRPT); + else if (syntax & RE_CONTEXT_INDEP_OPS) + laststart = b; + else + goto unfetch_interval; + } + + /* If the upper bound is zero, don't want to succeed at + all; jump from `laststart' to `b + 3', which will be + the end of the buffer after we insert the jump. */ + if (upper_bound == 0) + { + GET_BUFFER_SPACE (3); + INSERT_JUMP (jump, laststart, b + 3); + b += 3; + } + + /* Otherwise, we have a nontrivial interval. When + we're all done, the pattern will look like: + set_number_at <jump count> <upper bound> + set_number_at <succeed_n count> <lower bound> + succeed_n <after jump addr> <succeed_n count> + <body of loop> + jump_n <succeed_n addr> <jump count> + (The upper bound and `jump_n' are omitted if + `upper_bound' is 1, though.) */ + else + { /* If the upper bound is > 1, we need to insert + more at the end of the loop. */ + unsigned nbytes = 10 + (upper_bound > 1) * 10; + + GET_BUFFER_SPACE (nbytes); + + /* Initialize lower bound of the `succeed_n', even + though it will be set during matching by its + attendant `set_number_at' (inserted next), + because `re_compile_fastmap' needs to know. + Jump to the `jump_n' we might insert below. */ + INSERT_JUMP2 (succeed_n, laststart, + b + 5 + (upper_bound > 1) * 5, + lower_bound); + b += 5; + + /* Code to initialize the lower bound. Insert + before the `succeed_n'. The `5' is the last two + bytes of this `set_number_at', plus 3 bytes of + the following `succeed_n'. */ + insert_op2 (set_number_at, laststart, 5, lower_bound, b); + b += 5; + + if (upper_bound > 1) + { /* More than one repetition is allowed, so + append a backward jump to the `succeed_n' + that starts this interval. + + When we've reached this during matching, + we'll have matched the interval once, so + jump back only `upper_bound - 1' times. */ + STORE_JUMP2 (jump_n, b, laststart + 5, + upper_bound - 1); + b += 5; + + /* The location we want to set is the second + parameter of the `jump_n'; that is `b-2' as + an absolute address. `laststart' will be + the `set_number_at' we're about to insert; + `laststart+3' the number to set, the source + for the relative address. But we are + inserting into the middle of the pattern -- + so everything is getting moved up by 5. + Conclusion: (b - 2) - (laststart + 3) + 5, + i.e., b - laststart. + + We insert this at the beginning of the loop + so that if we fail during matching, we'll + reinitialize the bounds. */ + insert_op2 (set_number_at, laststart, b - laststart, + upper_bound - 1, b); + b += 5; + } + } + pending_exact = 0; + beg_interval = NULL; + } + break; + + unfetch_interval: + /* If an invalid interval, match the characters as literals. */ + assert (beg_interval); + p = beg_interval; + beg_interval = NULL; + + /* normal_char and normal_backslash need `c'. */ + PATFETCH (c); + + if (!(syntax & RE_NO_BK_BRACES)) + { + if (p > pattern && p[-1] == '\\') + goto normal_backslash; + } + goto normal_char; + +#ifdef emacs + /* There is no way to specify the before_dot and after_dot + operators. rms says this is ok. --karl */ + case '=': + BUF_PUSH (at_dot); + break; + + case 's': + laststart = b; + PATFETCH (c); + BUF_PUSH_2 (syntaxspec, syntax_spec_code[c]); + break; + + case 'S': + laststart = b; + PATFETCH (c); + BUF_PUSH_2 (notsyntaxspec, syntax_spec_code[c]); + break; +#endif /* emacs */ + + + case 'w': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + laststart = b; + BUF_PUSH (wordchar); + break; + + + case 'W': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + laststart = b; + BUF_PUSH (notwordchar); + break; + + + case '<': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (wordbeg); + break; + + case '>': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (wordend); + break; + + case 'b': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (wordbound); + break; + + case 'B': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (notwordbound); + break; + + case '`': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (begbuf); + break; + + case '\'': + if (re_syntax_options & RE_NO_GNU_OPS) + goto normal_char; + BUF_PUSH (endbuf); + break; + + case '1': case '2': case '3': case '4': case '5': + case '6': case '7': case '8': case '9': + if (syntax & RE_NO_BK_REFS) + goto normal_char; + + c1 = c - '0'; + + if (c1 > regnum) + FREE_STACK_RETURN (REG_ESUBREG); + + /* Can't back reference to a subexpression if inside of it. */ + if (group_in_compile_stack (compile_stack, (regnum_t) c1)) + goto normal_char; + + laststart = b; + BUF_PUSH_2 (duplicate, c1); + break; + + + case '+': + case '?': + if (syntax & RE_BK_PLUS_QM) + goto handle_plus; + else + goto normal_backslash; + + default: + normal_backslash: + /* You might think it would be useful for \ to mean + not to translate; but if we don't translate it + it will never match anything. */ + c = TRANSLATE (c); + goto normal_char; + } + break; + + + default: + /* Expects the character in `c'. */ + normal_char: + /* If no exactn currently being built. */ + if (!pending_exact + + /* If last exactn not at current position. */ + || pending_exact + *pending_exact + 1 != b + + /* We have only one byte following the exactn for the count. */ + || *pending_exact == (1 << BYTEWIDTH) - 1 + + /* If followed by a repetition operator. */ + || *p == '*' || *p == '^' + || ((syntax & RE_BK_PLUS_QM) + ? *p == '\\' && (p[1] == '+' || p[1] == '?') + : (*p == '+' || *p == '?')) + || ((syntax & RE_INTERVALS) + && ((syntax & RE_NO_BK_BRACES) + ? *p == '{' + : (p[0] == '\\' && p[1] == '{')))) + { + /* Start building a new exactn. */ + + laststart = b; + + BUF_PUSH_2 (exactn, 0); + pending_exact = b - 1; + } + + BUF_PUSH (c); + (*pending_exact)++; + break; + } /* switch (c) */ + } /* while p != pend */ + + + /* Through the pattern now. */ + + if (fixup_alt_jump) + STORE_JUMP (jump_past_alt, fixup_alt_jump, b); + + if (!COMPILE_STACK_EMPTY) + FREE_STACK_RETURN (REG_EPAREN); + + /* If we don't want backtracking, force success + the first time we reach the end of the compiled pattern. */ + if (syntax & RE_NO_POSIX_BACKTRACKING) + BUF_PUSH (succeed); + + free (compile_stack.stack); + + /* We have succeeded; set the length of the buffer. */ + bufp->used = b - bufp->buffer; + +#ifdef DEBUG + if (debug) + { + DEBUG_PRINT1 ("\nCompiled pattern: \n"); + print_compiled_pattern (bufp); + } +#endif /* DEBUG */ + +#ifndef MATCH_MAY_ALLOCATE + /* Initialize the failure stack to the largest possible stack. This + isn't necessary unless we're trying to avoid calling alloca in + the search and match routines. */ + { + int num_regs = bufp->re_nsub + 1; + + /* Since DOUBLE_FAIL_STACK refuses to double only if the current size + is strictly greater than re_max_failures, the largest possible stack + is 2 * re_max_failures failure points. */ + if (fail_stack.size < (2 * re_max_failures * MAX_FAILURE_ITEMS)) + { + fail_stack.size = (2 * re_max_failures * MAX_FAILURE_ITEMS); + +#ifdef emacs + if (! fail_stack.stack) + fail_stack.stack + = (fail_stack_elt_t *) xmalloc (fail_stack.size + * sizeof (fail_stack_elt_t)); + else + fail_stack.stack + = (fail_stack_elt_t *) xrealloc (fail_stack.stack, + (fail_stack.size + * sizeof (fail_stack_elt_t))); +#else /* not emacs */ + if (! fail_stack.stack) + fail_stack.stack + = (fail_stack_elt_t *) malloc (fail_stack.size + * sizeof (fail_stack_elt_t)); + else + fail_stack.stack + = (fail_stack_elt_t *) realloc (fail_stack.stack, + (fail_stack.size + * sizeof (fail_stack_elt_t))); +#endif /* not emacs */ + } + + regex_grow_registers (num_regs); + } +#endif /* not MATCH_MAY_ALLOCATE */ + + return REG_NOERROR; +} /* regex_compile */ + +/* Subroutines for `regex_compile'. */ + +/* Store OP at LOC followed by two-byte integer parameter ARG. */ + +static void +store_op1 (op, loc, arg) + re_opcode_t op; + unsigned char *loc; + int arg; +{ + *loc = (unsigned char) op; + STORE_NUMBER (loc + 1, arg); +} + + +/* Like `store_op1', but for two two-byte parameters ARG1 and ARG2. */ + +static void +store_op2 (op, loc, arg1, arg2) + re_opcode_t op; + unsigned char *loc; + int arg1, arg2; +{ + *loc = (unsigned char) op; + STORE_NUMBER (loc + 1, arg1); + STORE_NUMBER (loc + 3, arg2); +} + + +/* Copy the bytes from LOC to END to open up three bytes of space at LOC + for OP followed by two-byte integer parameter ARG. */ + +static void +insert_op1 (op, loc, arg, end) + re_opcode_t op; + unsigned char *loc; + int arg; + unsigned char *end; +{ + register unsigned char *pfrom = end; + register unsigned char *pto = end + 3; + + while (pfrom != loc) + *--pto = *--pfrom; + + store_op1 (op, loc, arg); +} + + +/* Like `insert_op1', but for two two-byte parameters ARG1 and ARG2. */ + +static void +insert_op2 (op, loc, arg1, arg2, end) + re_opcode_t op; + unsigned char *loc; + int arg1, arg2; + unsigned char *end; +{ + register unsigned char *pfrom = end; + register unsigned char *pto = end + 5; + + while (pfrom != loc) + *--pto = *--pfrom; + + store_op2 (op, loc, arg1, arg2); +} + + +/* P points to just after a ^ in PATTERN. Return true if that ^ comes + after an alternative or a begin-subexpression. We assume there is at + least one character before the ^. */ + +static boolean +at_begline_loc_p (pattern, p, syntax) + const char *pattern, *p; + reg_syntax_t syntax; +{ + const char *prev = p - 2; + boolean prev_prev_backslash = prev > pattern && prev[-1] == '\\'; + + return + /* After a subexpression? */ + (*prev == '(' && (syntax & RE_NO_BK_PARENS || prev_prev_backslash)) + /* After an alternative? */ + || (*prev == '|' && (syntax & RE_NO_BK_VBAR || prev_prev_backslash)); +} + + +/* The dual of at_begline_loc_p. This one is for $. We assume there is + at least one character after the $, i.e., `P < PEND'. */ + +static boolean +at_endline_loc_p (p, pend, syntax) + const char *p, *pend; + reg_syntax_t syntax; +{ + const char *next = p; + boolean next_backslash = *next == '\\'; + const char *next_next = p + 1 < pend ? p + 1 : 0; + + return + /* Before a subexpression? */ + (syntax & RE_NO_BK_PARENS ? *next == ')' + : next_backslash && next_next && *next_next == ')') + /* Before an alternative? */ + || (syntax & RE_NO_BK_VBAR ? *next == '|' + : next_backslash && next_next && *next_next == '|'); +} + + +/* Returns true if REGNUM is in one of COMPILE_STACK's elements and + false if it's not. */ + +static boolean +group_in_compile_stack (compile_stack, regnum) + compile_stack_type compile_stack; + regnum_t regnum; +{ + int this_element; + + for (this_element = compile_stack.avail - 1; + this_element >= 0; + this_element--) + if (compile_stack.stack[this_element].regnum == regnum) + return true; + + return false; +} + + +/* Read the ending character of a range (in a bracket expression) from the + uncompiled pattern *P_PTR (which ends at PEND). We assume the + starting character is in `P[-2]'. (`P[-1]' is the character `-'.) + Then we set the translation of all bits between the starting and + ending characters (inclusive) in the compiled pattern B. + + Return an error code. + + We use these short variable names so we can use the same macros as + `regex_compile' itself. */ + +static reg_errcode_t +compile_range (p_ptr, pend, translate, syntax, b) + const char **p_ptr, *pend; + RE_TRANSLATE_TYPE translate; + reg_syntax_t syntax; + unsigned char *b; +{ + unsigned this_char; + + const char *p = *p_ptr; + unsigned int range_start, range_end; + + if (p == pend) + return REG_ERANGE; + + /* Even though the pattern is a signed `char *', we need to fetch + with unsigned char *'s; if the high bit of the pattern character + is set, the range endpoints will be negative if we fetch using a + signed char *. + + We also want to fetch the endpoints without translating them; the + appropriate translation is done in the bit-setting loop below. */ + /* The SVR4 compiler on the 3B2 had trouble with unsigned const char *. */ + range_start = ((const unsigned char *) p)[-2]; + range_end = ((const unsigned char *) p)[0]; + + /* Have to increment the pointer into the pattern string, so the + caller isn't still at the ending character. */ + (*p_ptr)++; + + /* If the start is after the end, the range is empty. */ + if (range_start > range_end) + return syntax & RE_NO_EMPTY_RANGES ? REG_ERANGE : REG_NOERROR; + + /* Here we see why `this_char' has to be larger than an `unsigned + char' -- the range is inclusive, so if `range_end' == 0xff + (assuming 8-bit characters), we would otherwise go into an infinite + loop, since all characters <= 0xff. */ + for (this_char = range_start; this_char <= range_end; this_char++) + { + SET_LIST_BIT (TRANSLATE (this_char)); + } + + return REG_NOERROR; +} + +/* re_compile_fastmap computes a ``fastmap'' for the compiled pattern in + BUFP. A fastmap records which of the (1 << BYTEWIDTH) possible + characters can start a string that matches the pattern. This fastmap + is used by re_search to skip quickly over impossible starting points. + + The caller must supply the address of a (1 << BYTEWIDTH)-byte data + area as BUFP->fastmap. + + We set the `fastmap', `fastmap_accurate', and `can_be_null' fields in + the pattern buffer. + + Returns 0 if we succeed, -2 if an internal error. */ + +int +re_compile_fastmap (bufp) + struct re_pattern_buffer *bufp; +{ + int j, k; +#ifdef MATCH_MAY_ALLOCATE + fail_stack_type fail_stack; +#endif +#ifndef REGEX_MALLOC + char *destination; +#endif + /* We don't push any register information onto the failure stack. */ +//sword unsigned num_regs = 0; + + register char *fastmap = bufp->fastmap; + unsigned char *pattern = bufp->buffer; + unsigned char *p = pattern; + register unsigned char *pend = pattern + bufp->used; + +#ifdef REL_ALLOC + /* This holds the pointer to the failure stack, when + it is allocated relocatably. */ + fail_stack_elt_t *failure_stack_ptr; +#endif + + /* Assume that each path through the pattern can be null until + proven otherwise. We set this false at the bottom of switch + statement, to which we get only if a particular path doesn't + match the empty string. */ + boolean path_can_be_null = true; + + /* We aren't doing a `succeed_n' to begin with. */ + boolean succeed_n_p = false; + + assert (fastmap != NULL && p != NULL); + + INIT_FAIL_STACK (); + bzero (fastmap, 1 << BYTEWIDTH); /* Assume nothing's valid. */ + bufp->fastmap_accurate = 1; /* It will be when we're done. */ + bufp->can_be_null = 0; + + while (1) + { + if (p == pend || *p == succeed) + { + /* We have reached the (effective) end of pattern. */ + if (!FAIL_STACK_EMPTY ()) + { + bufp->can_be_null |= path_can_be_null; + + /* Reset for next path. */ + path_can_be_null = true; + + p = fail_stack.stack[--fail_stack.avail].pointer; + + continue; + } + else + break; + } + + /* We should never be about to go beyond the end of the pattern. */ + assert (p < pend); + + switch (SWITCH_ENUM_CAST ((re_opcode_t) *p++)) + { + + /* I guess the idea here is to simply not bother with a fastmap + if a backreference is used, since it's too hard to figure out + the fastmap for the corresponding group. Setting + `can_be_null' stops `re_search_2' from using the fastmap, so + that is all we do. */ + case duplicate: + bufp->can_be_null = 1; + goto done; + + + /* Following are the cases which match a character. These end + with `break'. */ + + case exactn: + fastmap[p[1]] = 1; + break; + + + case charset: + for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--) + if (p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH))) + fastmap[j] = 1; + break; + + + case charset_not: + /* Chars beyond end of map must be allowed. */ + for (j = *p * BYTEWIDTH; j < (1 << BYTEWIDTH); j++) + fastmap[j] = 1; + + for (j = *p++ * BYTEWIDTH - 1; j >= 0; j--) + if (!(p[j / BYTEWIDTH] & (1 << (j % BYTEWIDTH)))) + fastmap[j] = 1; + break; + + + case wordchar: + for (j = 0; j < (1 << BYTEWIDTH); j++) + if (SYNTAX (j) == Sword) + fastmap[j] = 1; + break; + + + case notwordchar: + for (j = 0; j < (1 << BYTEWIDTH); j++) + if (SYNTAX (j) != Sword) + fastmap[j] = 1; + break; + + + case anychar: + { + int fastmap_newline = fastmap['\n']; + + /* `.' matches anything ... */ + for (j = 0; j < (1 << BYTEWIDTH); j++) + fastmap[j] = 1; + + /* ... except perhaps newline. */ + if (!(bufp->syntax & RE_DOT_NEWLINE)) + fastmap['\n'] = fastmap_newline; + + /* Return if we have already set `can_be_null'; if we have, + then the fastmap is irrelevant. Something's wrong here. */ + else if (bufp->can_be_null) + goto done; + + /* Otherwise, have to check alternative paths. */ + break; + } + +#ifdef emacs + case syntaxspec: + k = *p++; + for (j = 0; j < (1 << BYTEWIDTH); j++) + if (SYNTAX (j) == (enum syntaxcode) k) + fastmap[j] = 1; + break; + + + case notsyntaxspec: + k = *p++; + for (j = 0; j < (1 << BYTEWIDTH); j++) + if (SYNTAX (j) != (enum syntaxcode) k) + fastmap[j] = 1; + break; + + + /* All cases after this match the empty string. These end with + `continue'. */ + + + case before_dot: + case at_dot: + case after_dot: + continue; +#endif /* emacs */ + + + case no_op: + case begline: + case endline: + case begbuf: + case endbuf: + case wordbound: + case notwordbound: + case wordbeg: + case wordend: + case push_dummy_failure: + continue; + + + case jump_n: + case pop_failure_jump: + case maybe_pop_jump: + case jump: + case jump_past_alt: + case dummy_failure_jump: + EXTRACT_NUMBER_AND_INCR (j, p); + p += j; + if (j > 0) + continue; + + /* Jump backward implies we just went through the body of a + loop and matched nothing. Opcode jumped to should be + `on_failure_jump' or `succeed_n'. Just treat it like an + ordinary jump. For a * loop, it has pushed its failure + point already; if so, discard that as redundant. */ + if ((re_opcode_t) *p != on_failure_jump + && (re_opcode_t) *p != succeed_n) + continue; + + p++; + EXTRACT_NUMBER_AND_INCR (j, p); + p += j; + + /* If what's on the stack is where we are now, pop it. */ + if (!FAIL_STACK_EMPTY () + && fail_stack.stack[fail_stack.avail - 1].pointer == p) + fail_stack.avail--; + + continue; + + + case on_failure_jump: + case on_failure_keep_string_jump: + handle_on_failure_jump: + EXTRACT_NUMBER_AND_INCR (j, p); + + /* For some patterns, e.g., `(a?)?', `p+j' here points to the + end of the pattern. We don't want to push such a point, + since when we restore it above, entering the switch will + increment `p' past the end of the pattern. We don't need + to push such a point since we obviously won't find any more + fastmap entries beyond `pend'. Such a pattern can match + the null string, though. */ + if (p + j < pend) + { + if (!PUSH_PATTERN_OP (p + j, fail_stack)) + { + RESET_FAIL_STACK (); + return -2; + } + } + else + bufp->can_be_null = 1; + + if (succeed_n_p) + { + EXTRACT_NUMBER_AND_INCR (k, p); /* Skip the n. */ + succeed_n_p = false; + } + + continue; + + + case succeed_n: + /* Get to the number of times to succeed. */ + p += 2; + + /* Increment p past the n for when k != 0. */ + EXTRACT_NUMBER_AND_INCR (k, p); + if (k == 0) + { + p -= 4; + succeed_n_p = true; /* Spaghetti code alert. */ + goto handle_on_failure_jump; + } + continue; + + + case set_number_at: + p += 4; + continue; + + + case start_memory: + case stop_memory: + p += 2; + continue; + + + default: + abort (); /* We have listed all the cases. */ + } /* switch *p++ */ + + /* Getting here means we have found the possible starting + characters for one path of the pattern -- and that the empty + string does not match. We need not follow this path further. + Instead, look at the next alternative (remembered on the + stack), or quit if no more. The test at the top of the loop + does these things. */ + path_can_be_null = false; + p = pend; + } /* while p */ + + /* Set `can_be_null' for the last path (also the first path, if the + pattern is empty). */ + bufp->can_be_null |= path_can_be_null; + + done: + RESET_FAIL_STACK (); + return 0; +} /* re_compile_fastmap */ + +/* Set REGS to hold NUM_REGS registers, storing them in STARTS and + ENDS. Subsequent matches using PATTERN_BUFFER and REGS will use + this memory for recording register information. STARTS and ENDS + must be allocated using the malloc library routine, and must each + be at least NUM_REGS * sizeof (regoff_t) bytes long. + + If NUM_REGS == 0, then subsequent matches should allocate their own + register data. + + Unless this function is called, the first search or match using + PATTERN_BUFFER will allocate its own register data, without + freeing the old data. */ + +void +re_set_registers (bufp, regs, num_regs, starts, ends) + struct re_pattern_buffer *bufp; + struct re_registers *regs; + unsigned num_regs; + regoff_t *starts, *ends; +{ + if (num_regs) + { + bufp->regs_allocated = REGS_REALLOCATE; + regs->num_regs = num_regs; + regs->start = starts; + regs->end = ends; + } + else + { + bufp->regs_allocated = REGS_UNALLOCATED; + regs->num_regs = 0; + regs->start = regs->end = (regoff_t *) 0; + } +} + +/* Searching routines. */ + +/* Like re_search_2, below, but only one string is specified, and + doesn't let you say where to stop matching. */ + +int +re_search (bufp, string, size, startpos, range, regs) + struct re_pattern_buffer *bufp; + const char *string; + int size, startpos, range; + struct re_registers *regs; +{ + return re_search_2 (bufp, NULL, 0, string, size, startpos, range, + regs, size); +} + + +/* Using the compiled pattern in BUFP->buffer, first tries to match the + virtual concatenation of STRING1 and STRING2, starting first at index + STARTPOS, then at STARTPOS + 1, and so on. + + STRING1 and STRING2 have length SIZE1 and SIZE2, respectively. + + RANGE is how far to scan while trying to match. RANGE = 0 means try + only at STARTPOS; in general, the last start tried is STARTPOS + + RANGE. + + In REGS, return the indices of the virtual concatenation of STRING1 + and STRING2 that matched the entire BUFP->buffer and its contained + subexpressions. + + Do not consider matching one past the index STOP in the virtual + concatenation of STRING1 and STRING2. + + We return either the position in the strings at which the match was + found, -1 if no match, or -2 if error (such as failure + stack overflow). */ + +int +re_search_2 (bufp, string1, size1, string2, size2, startpos, range, regs, stop) + struct re_pattern_buffer *bufp; + const char *string1, *string2; + int size1, size2; + int startpos; + int range; + struct re_registers *regs; + int stop; +{ + int val; + register char *fastmap = bufp->fastmap; + register RE_TRANSLATE_TYPE translate = bufp->translate; + int total_size = size1 + size2; + int endpos = startpos + range; + + /* Check for out-of-range STARTPOS. */ + if (startpos < 0 || startpos > total_size) + return -1; + + /* Fix up RANGE if it might eventually take us outside + the virtual concatenation of STRING1 and STRING2. + Make sure we won't move STARTPOS below 0 or above TOTAL_SIZE. */ + if (endpos < 0) + range = 0 - startpos; + else if (endpos > total_size) + range = total_size - startpos; + + /* If the search isn't to be a backwards one, don't waste time in a + search for a pattern that must be anchored. */ + if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == begbuf && range > 0) + { + if (startpos > 0) + return -1; + else + range = 1; + } + +#ifdef emacs + /* In a forward search for something that starts with \=. + don't keep searching past point. */ + if (bufp->used > 0 && (re_opcode_t) bufp->buffer[0] == at_dot && range > 0) + { + range = PT - startpos; + if (range <= 0) + return -1; + } +#endif /* emacs */ + + /* Update the fastmap now if not correct already. */ + if (fastmap && !bufp->fastmap_accurate) + if (re_compile_fastmap (bufp) == -2) + return -2; + + /* Loop through the string, looking for a place to start matching. */ + for (;;) + { + /* If a fastmap is supplied, skip quickly over characters that + cannot be the start of a match. If the pattern can match the + null string, however, we don't need to skip characters; we want + the first null string. */ + if (fastmap && startpos < total_size && !bufp->can_be_null) + { + if (range > 0) /* Searching forwards. */ + { + register const char *d; + register int lim = 0; + int irange = range; + + if (startpos < size1 && startpos + range >= size1) + lim = range - (size1 - startpos); + + d = (startpos >= size1 ? string2 - size1 : string1) + startpos; + + /* Written out as an if-else to avoid testing `translate' + inside the loop. */ + if (translate) + while (range > lim + && !fastmap[(unsigned char) + translate[(unsigned char) *d++]]) + range--; + else + while (range > lim && !fastmap[(unsigned char) *d++]) + range--; + + startpos += irange - range; + } + else /* Searching backwards. */ + { + register char c = (size1 == 0 || startpos >= size1 + ? string2[startpos - size1] + : string1[startpos]); + + if (!fastmap[(unsigned char) TRANSLATE (c)]) + goto advance; + } + } + + /* If can't match the null string, and that's all we have left, fail. */ + if (range >= 0 && startpos == total_size && fastmap + && !bufp->can_be_null) + return -1; + + val = re_match_2_internal (bufp, string1, size1, string2, size2, + startpos, regs, stop); +#ifndef REGEX_MALLOC +#ifdef C_ALLOCA + alloca (0); +#endif +#endif + + if (val >= 0) + return startpos; + + if (val == -2) + return -2; + + advance: + if (!range) + break; + else if (range > 0) + { + range--; + startpos++; + } + else + { + range++; + startpos--; + } + } + return -1; +} /* re_search_2 */ + +/* This converts PTR, a pointer into one of the search strings `string1' + and `string2' into an offset from the beginning of that string. */ +#define POINTER_TO_OFFSET(ptr) \ + (FIRST_STRING_P (ptr) \ + ? ((regoff_t) ((ptr) - string1)) \ + : ((regoff_t) ((ptr) - string2 + size1))) + +/* Macros for dealing with the split strings in re_match_2. */ + +#define MATCHING_IN_FIRST_STRING (dend == end_match_1) + +/* Call before fetching a character with *d. This switches over to + string2 if necessary. */ +#define PREFETCH() \ + while (d == dend) \ + { \ + /* End of string2 => fail. */ \ + if (dend == end_match_2) \ + goto fail; \ + /* End of string1 => advance to string2. */ \ + d = string2; \ + dend = end_match_2; \ + } + + +/* Test if at very beginning or at very end of the virtual concatenation + of `string1' and `string2'. If only one string, it's `string2'. */ +#define AT_STRINGS_BEG(d) ((d) == (size1 ? string1 : string2) || !size2) +#define AT_STRINGS_END(d) ((d) == end2) + + +/* Test if D points to a character which is word-constituent. We have + two special cases to check for: if past the end of string1, look at + the first character in string2; and if before the beginning of + string2, look at the last character in string1. */ +#define WORDCHAR_P(d) \ + (SYNTAX ((d) == end1 ? *string2 \ + : (d) == string2 - 1 ? *(end1 - 1) : *(d)) \ + == Sword) + +/* Disabled due to a compiler bug -- see comment at case wordbound */ +#if 0 +/* Test if the character before D and the one at D differ with respect + to being word-constituent. */ +#define AT_WORD_BOUNDARY(d) \ + (AT_STRINGS_BEG (d) || AT_STRINGS_END (d) \ + || WORDCHAR_P (d - 1) != WORDCHAR_P (d)) +#endif + +/* Free everything we malloc. */ +#ifdef MATCH_MAY_ALLOCATE +#define FREE_VAR(var) if (var) REGEX_FREE (var); var = NULL +#define FREE_VARIABLES() \ + do { \ + REGEX_FREE_STACK (fail_stack.stack); \ + FREE_VAR (regstart); \ + FREE_VAR (regend); \ + FREE_VAR (old_regstart); \ + FREE_VAR (old_regend); \ + FREE_VAR (best_regstart); \ + FREE_VAR (best_regend); \ + FREE_VAR (reg_info); \ + FREE_VAR (reg_dummy); \ + FREE_VAR (reg_info_dummy); \ + } while (0) +#else +#define FREE_VARIABLES() ((void)0) /* Do nothing! But inhibit gcc warning. */ +#endif /* not MATCH_MAY_ALLOCATE */ + +/* These values must meet several constraints. They must not be valid + register values; since we have a limit of 255 registers (because + we use only one byte in the pattern for the register number), we can + use numbers larger than 255. They must differ by 1, because of + NUM_FAILURE_ITEMS above. And the value for the lowest register must + be larger than the value for the highest register, so we do not try + to actually save any registers when none are active. */ +#define NO_HIGHEST_ACTIVE_REG (1 << BYTEWIDTH) +#define NO_LOWEST_ACTIVE_REG (NO_HIGHEST_ACTIVE_REG + 1) + +/* Matching routines. */ + +#ifndef emacs /* Emacs never uses this. */ +/* re_match is like re_match_2 except it takes only a single string. */ + +int +re_match (bufp, string, size, pos, regs) + struct re_pattern_buffer *bufp; + const char *string; + int size, pos; + struct re_registers *regs; +{ + int result = re_match_2_internal (bufp, NULL, 0, string, size, + pos, regs, size); +#ifndef REGEX_MALLOC +#ifdef C_ALLOCA + alloca (0); +#endif +#endif + return result; +} +#endif /* not emacs */ + +static boolean group_match_null_string_p _RE_ARGS ((unsigned char **p, + unsigned char *end, + register_info_type *reg_info)); +static boolean alt_match_null_string_p _RE_ARGS ((unsigned char *p, + unsigned char *end, + register_info_type *reg_info)); +static boolean common_op_match_null_string_p _RE_ARGS ((unsigned char **p, + unsigned char *end, + register_info_type *reg_info)); +static int bcmp_translate _RE_ARGS ((const char *s1, const char *s2, + int len, char *translate)); + +/* re_match_2 matches the compiled pattern in BUFP against the + the (virtual) concatenation of STRING1 and STRING2 (of length SIZE1 + and SIZE2, respectively). We start matching at POS, and stop + matching at STOP. + + If REGS is non-null and the `no_sub' field of BUFP is nonzero, we + store offsets for the substring each group matched in REGS. See the + documentation for exactly how many groups we fill. + + We return -1 if no match, -2 if an internal error (such as the + failure stack overflowing). Otherwise, we return the length of the + matched substring. */ + +int +re_match_2 (bufp, string1, size1, string2, size2, pos, regs, stop) + struct re_pattern_buffer *bufp; + const char *string1, *string2; + int size1, size2; + int pos; + struct re_registers *regs; + int stop; +{ + int result = re_match_2_internal (bufp, string1, size1, string2, size2, + pos, regs, stop); +#ifndef REGEX_MALLOC +#ifdef C_ALLOCA + alloca (0); +#endif +#endif + return result; +} + +/* This is a separate function so that we can force an alloca cleanup + afterwards. */ +static int +re_match_2_internal (bufp, string1, size1, string2, size2, pos, regs, stop) + struct re_pattern_buffer *bufp; + const char *string1, *string2; + int size1, size2; + int pos; + struct re_registers *regs; + int stop; +{ + /* General temporaries. */ + int mcnt; + unsigned char *p1; + + /* Just past the end of the corresponding string. */ + const char *end1, *end2; + + /* Pointers into string1 and string2, just past the last characters in + each to consider matching. */ + const char *end_match_1, *end_match_2; + + /* Where we are in the data, and the end of the current string. */ + const char *d, *dend; + + /* Where we are in the pattern, and the end of the pattern. */ + unsigned char *p = bufp->buffer; + register unsigned char *pend = p + bufp->used; + + /* Mark the opcode just after a start_memory, so we can test for an + empty subpattern when we get to the stop_memory. */ + unsigned char *just_past_start_mem = 0; + + /* We use this to map every character in the string. */ + RE_TRANSLATE_TYPE translate = bufp->translate; + + /* Failure point stack. Each place that can handle a failure further + down the line pushes a failure point on this stack. It consists of + restart, regend, and reg_info for all registers corresponding to + the subexpressions we're currently inside, plus the number of such + registers, and, finally, two char *'s. The first char * is where + to resume scanning the pattern; the second one is where to resume + scanning the strings. If the latter is zero, the failure point is + a ``dummy''; if a failure happens and the failure point is a dummy, + it gets discarded and the next next one is tried. */ +#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */ + fail_stack_type fail_stack; +#endif +#ifdef DEBUG + static unsigned failure_id = 0; + unsigned nfailure_points_pushed = 0, nfailure_points_popped = 0; +#endif + +#ifdef REL_ALLOC + /* This holds the pointer to the failure stack, when + it is allocated relocatably. */ + fail_stack_elt_t *failure_stack_ptr; +#endif + + /* We fill all the registers internally, independent of what we + return, for use in backreferences. The number here includes + an element for register zero. */ + size_t num_regs = bufp->re_nsub + 1; + + /* The currently active registers. */ + active_reg_t lowest_active_reg = NO_LOWEST_ACTIVE_REG; + active_reg_t highest_active_reg = NO_HIGHEST_ACTIVE_REG; + + /* Information on the contents of registers. These are pointers into + the input strings; they record just what was matched (on this + attempt) by a subexpression part of the pattern, that is, the + regnum-th regstart pointer points to where in the pattern we began + matching and the regnum-th regend points to right after where we + stopped matching the regnum-th subexpression. (The zeroth register + keeps track of what the whole pattern matches.) */ +#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */ + const char **regstart, **regend; +#endif + + /* If a group that's operated upon by a repetition operator fails to + match anything, then the register for its start will need to be + restored because it will have been set to wherever in the string we + are when we last see its open-group operator. Similarly for a + register's end. */ +#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */ + const char **old_regstart, **old_regend; +#endif + + /* The is_active field of reg_info helps us keep track of which (possibly + nested) subexpressions we are currently in. The matched_something + field of reg_info[reg_num] helps us tell whether or not we have + matched any of the pattern so far this time through the reg_num-th + subexpression. These two fields get reset each time through any + loop their register is in. */ +#ifdef MATCH_MAY_ALLOCATE /* otherwise, this is global. */ + register_info_type *reg_info; +#endif + + /* The following record the register info as found in the above + variables when we find a match better than any we've seen before. + This happens as we backtrack through the failure points, which in + turn happens only if we have not yet matched the entire string. */ + unsigned best_regs_set = false; +#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */ + const char **best_regstart, **best_regend; +#endif + + /* Logically, this is `best_regend[0]'. But we don't want to have to + allocate space for that if we're not allocating space for anything + else (see below). Also, we never need info about register 0 for + any of the other register vectors, and it seems rather a kludge to + treat `best_regend' differently than the rest. So we keep track of + the end of the best match so far in a separate variable. We + initialize this to NULL so that when we backtrack the first time + and need to test it, it's not garbage. */ + const char *match_end = NULL; + + /* This helps SET_REGS_MATCHED avoid doing redundant work. */ + int set_regs_matched_done = 0; + + /* Used when we pop values we don't care about. */ +#ifdef MATCH_MAY_ALLOCATE /* otherwise, these are global. */ + const char **reg_dummy; + register_info_type *reg_info_dummy; +#endif + +#ifdef DEBUG + /* Counts the total number of registers pushed. */ + unsigned num_regs_pushed = 0; +#endif + + DEBUG_PRINT1 ("\n\nEntering re_match_2.\n"); + + INIT_FAIL_STACK (); + +#ifdef MATCH_MAY_ALLOCATE + /* Do not bother to initialize all the register variables if there are + no groups in the pattern, as it takes a fair amount of time. If + there are groups, we include space for register 0 (the whole + pattern), even though we never use it, since it simplifies the + array indexing. We should fix this. */ + if (bufp->re_nsub) + { + regstart = REGEX_TALLOC (num_regs, const char *); + regend = REGEX_TALLOC (num_regs, const char *); + old_regstart = REGEX_TALLOC (num_regs, const char *); + old_regend = REGEX_TALLOC (num_regs, const char *); + best_regstart = REGEX_TALLOC (num_regs, const char *); + best_regend = REGEX_TALLOC (num_regs, const char *); + reg_info = REGEX_TALLOC (num_regs, register_info_type); + reg_dummy = REGEX_TALLOC (num_regs, const char *); + reg_info_dummy = REGEX_TALLOC (num_regs, register_info_type); + + if (!(regstart && regend && old_regstart && old_regend && reg_info + && best_regstart && best_regend && reg_dummy && reg_info_dummy)) + { + FREE_VARIABLES (); + return -2; + } + } + else + { + /* We must initialize all our variables to NULL, so that + `FREE_VARIABLES' doesn't try to free them. */ + regstart = regend = old_regstart = old_regend = best_regstart + = best_regend = reg_dummy = NULL; + reg_info = reg_info_dummy = (register_info_type *) NULL; + } +#endif /* MATCH_MAY_ALLOCATE */ + + /* The starting position is bogus. */ + if (pos < 0 || pos > size1 + size2) + { + FREE_VARIABLES (); + return -1; + } + + /* Initialize subexpression text positions to -1 to mark ones that no + start_memory/stop_memory has been seen for. Also initialize the + register information struct. */ + for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++) + { + regstart[mcnt] = regend[mcnt] + = old_regstart[mcnt] = old_regend[mcnt] = REG_UNSET_VALUE; + + REG_MATCH_NULL_STRING_P (reg_info[mcnt]) = MATCH_NULL_UNSET_VALUE; + IS_ACTIVE (reg_info[mcnt]) = 0; + MATCHED_SOMETHING (reg_info[mcnt]) = 0; + EVER_MATCHED_SOMETHING (reg_info[mcnt]) = 0; + } + + /* We move `string1' into `string2' if the latter's empty -- but not if + `string1' is null. */ + if (size2 == 0 && string1 != NULL) + { + string2 = string1; + size2 = size1; + string1 = 0; + size1 = 0; + } + end1 = string1 + size1; + end2 = string2 + size2; + + /* Compute where to stop matching, within the two strings. */ + if (stop <= size1) + { + end_match_1 = string1 + stop; + end_match_2 = string2; + } + else + { + end_match_1 = end1; + end_match_2 = string2 + stop - size1; + } + + /* `p' scans through the pattern as `d' scans through the data. + `dend' is the end of the input string that `d' points within. `d' + is advanced into the following input string whenever necessary, but + this happens before fetching; therefore, at the beginning of the + loop, `d' can be pointing at the end of a string, but it cannot + equal `string2'. */ + if (size1 > 0 && pos <= size1) + { + d = string1 + pos; + dend = end_match_1; + } + else + { + d = string2 + pos - size1; + dend = end_match_2; + } + + DEBUG_PRINT1 ("The compiled pattern is:\n"); + DEBUG_PRINT_COMPILED_PATTERN (bufp, p, pend); + DEBUG_PRINT1 ("The string to match is: `"); + DEBUG_PRINT_DOUBLE_STRING (d, string1, size1, string2, size2); + DEBUG_PRINT1 ("'\n"); + + /* This loops over pattern commands. It exits by returning from the + function if the match is complete, or it drops through if the match + fails at this starting point in the input data. */ + for (;;) + { +#ifdef _LIBC + DEBUG_PRINT2 ("\n%p: ", p); +#else + DEBUG_PRINT2 ("\n0x%x: ", p); +#endif + + if (p == pend) + { /* End of pattern means we might have succeeded. */ + DEBUG_PRINT1 ("end of pattern ... "); + + /* If we haven't matched the entire string, and we want the + longest match, try backtracking. */ + if (d != end_match_2) + { + /* 1 if this match ends in the same string (string1 or string2) + as the best previous match. */ + boolean same_str_p = (FIRST_STRING_P (match_end) + == MATCHING_IN_FIRST_STRING); + /* 1 if this match is the best seen so far. */ + boolean best_match_p; + + /* AIX compiler got confused when this was combined + with the previous declaration. */ + if (same_str_p) + best_match_p = d > match_end; + else + best_match_p = !MATCHING_IN_FIRST_STRING; + + DEBUG_PRINT1 ("backtracking.\n"); + + if (!FAIL_STACK_EMPTY ()) + { /* More failure points to try. */ + + /* If exceeds best match so far, save it. */ + if (!best_regs_set || best_match_p) + { + best_regs_set = true; + match_end = d; + + DEBUG_PRINT1 ("\nSAVING match as best so far.\n"); + + for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++) + { + best_regstart[mcnt] = regstart[mcnt]; + best_regend[mcnt] = regend[mcnt]; + } + } + goto fail; + } + + /* If no failure points, don't restore garbage. And if + last match is real best match, don't restore second + best one. */ + else if (best_regs_set && !best_match_p) + { + restore_best_regs: + /* Restore best match. It may happen that `dend == + end_match_1' while the restored d is in string2. + For example, the pattern `x.*y.*z' against the + strings `x-' and `y-z-', if the two strings are + not consecutive in memory. */ + DEBUG_PRINT1 ("Restoring best registers.\n"); + + d = match_end; + dend = ((d >= string1 && d <= end1) + ? end_match_1 : end_match_2); + + for (mcnt = 1; (unsigned) mcnt < num_regs; mcnt++) + { + regstart[mcnt] = best_regstart[mcnt]; + regend[mcnt] = best_regend[mcnt]; + } + } + } /* d != end_match_2 */ + + succeed_label: + DEBUG_PRINT1 ("Accepting match.\n"); + + /* If caller wants register contents data back, do it. */ + if (regs && !bufp->no_sub) + { + /* Have the register data arrays been allocated? */ + if (bufp->regs_allocated == REGS_UNALLOCATED) + { /* No. So allocate them with malloc. We need one + extra element beyond `num_regs' for the `-1' marker + GNU code uses. */ + regs->num_regs = MAX (RE_NREGS, num_regs + 1); + regs->start = TALLOC (regs->num_regs, regoff_t); + regs->end = TALLOC (regs->num_regs, regoff_t); + if (regs->start == NULL || regs->end == NULL) + { + FREE_VARIABLES (); + return -2; + } + bufp->regs_allocated = REGS_REALLOCATE; + } + else if (bufp->regs_allocated == REGS_REALLOCATE) + { /* Yes. If we need more elements than were already + allocated, reallocate them. If we need fewer, just + leave it alone. */ + if (regs->num_regs < num_regs + 1) + { + regs->num_regs = num_regs + 1; + RETALLOC (regs->start, regs->num_regs, regoff_t); + RETALLOC (regs->end, regs->num_regs, regoff_t); + if (regs->start == NULL || regs->end == NULL) + { + FREE_VARIABLES (); + return -2; + } + } + } + else + { + /* These braces fend off a "empty body in an else-statement" + warning under GCC when assert expands to nothing. */ + assert (bufp->regs_allocated == REGS_FIXED); + } + + /* Convert the pointer data in `regstart' and `regend' to + indices. Register zero has to be set differently, + since we haven't kept track of any info for it. */ + if (regs->num_regs > 0) + { + regs->start[0] = pos; + regs->end[0] = (MATCHING_IN_FIRST_STRING + ? ((regoff_t) (d - string1)) + : ((regoff_t) (d - string2 + size1))); + } + + /* Go through the first `min (num_regs, regs->num_regs)' + registers, since that is all we initialized. */ + for (mcnt = 1; (unsigned) mcnt < MIN (num_regs, regs->num_regs); + mcnt++) + { + if (REG_UNSET (regstart[mcnt]) || REG_UNSET (regend[mcnt])) + regs->start[mcnt] = regs->end[mcnt] = -1; + else + { + regs->start[mcnt] + = (regoff_t) POINTER_TO_OFFSET (regstart[mcnt]); + regs->end[mcnt] + = (regoff_t) POINTER_TO_OFFSET (regend[mcnt]); + } + } + + /* If the regs structure we return has more elements than + were in the pattern, set the extra elements to -1. If + we (re)allocated the registers, this is the case, + because we always allocate enough to have at least one + -1 at the end. */ + for (mcnt = num_regs; (unsigned) mcnt < regs->num_regs; mcnt++) + regs->start[mcnt] = regs->end[mcnt] = -1; + } /* regs && !bufp->no_sub */ + + DEBUG_PRINT4 ("%u failure points pushed, %u popped (%u remain).\n", + nfailure_points_pushed, nfailure_points_popped, + nfailure_points_pushed - nfailure_points_popped); + DEBUG_PRINT2 ("%u registers pushed.\n", num_regs_pushed); + + mcnt = d - pos - (MATCHING_IN_FIRST_STRING + ? string1 + : string2 - size1); + + DEBUG_PRINT2 ("Returning %d from re_match_2.\n", mcnt); + + FREE_VARIABLES (); + return mcnt; + } + + /* Otherwise match next pattern command. */ + switch (SWITCH_ENUM_CAST ((re_opcode_t) *p++)) + { + /* Ignore these. Used to ignore the n of succeed_n's which + currently have n == 0. */ + case no_op: + DEBUG_PRINT1 ("EXECUTING no_op.\n"); + break; + + case succeed: + DEBUG_PRINT1 ("EXECUTING succeed.\n"); + goto succeed_label; + + /* Match the next n pattern characters exactly. The following + byte in the pattern defines n, and the n bytes after that + are the characters to match. */ + case exactn: + mcnt = *p++; + DEBUG_PRINT2 ("EXECUTING exactn %d.\n", mcnt); + + /* This is written out as an if-else so we don't waste time + testing `translate' inside the loop. */ + if (translate) + { + do + { + PREFETCH (); + if ((unsigned char) translate[(unsigned char) *d++] + != (unsigned char) *p++) + goto fail; + } + while (--mcnt); + } + else + { + do + { + PREFETCH (); + if (*d++ != (char) *p++) goto fail; + } + while (--mcnt); + } + SET_REGS_MATCHED (); + break; + + + /* Match any character except possibly a newline or a null. */ + case anychar: + DEBUG_PRINT1 ("EXECUTING anychar.\n"); + + PREFETCH (); + + if ((!(bufp->syntax & RE_DOT_NEWLINE) && TRANSLATE (*d) == '\n') + || (bufp->syntax & RE_DOT_NOT_NULL && TRANSLATE (*d) == '\000')) + goto fail; + + SET_REGS_MATCHED (); + DEBUG_PRINT2 (" Matched `%d'.\n", *d); + d++; + break; + + + case charset: + case charset_not: + { + register unsigned char c; + boolean not = (re_opcode_t) *(p - 1) == charset_not; + + DEBUG_PRINT2 ("EXECUTING charset%s.\n", not ? "_not" : ""); + + PREFETCH (); + c = TRANSLATE (*d); /* The character to match. */ + + /* Cast to `unsigned' instead of `unsigned char' in case the + bit list is a full 32 bytes long. */ + if (c < (unsigned) (*p * BYTEWIDTH) + && p[1 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH))) + not = !not; + + p += 1 + *p; + + if (!not) goto fail; + + SET_REGS_MATCHED (); + d++; + break; + } + + + /* The beginning of a group is represented by start_memory. + The arguments are the register number in the next byte, and the + number of groups inner to this one in the next. The text + matched within the group is recorded (in the internal + registers data structure) under the register number. */ + case start_memory: + DEBUG_PRINT3 ("EXECUTING start_memory %d (%d):\n", *p, p[1]); + + /* Find out if this group can match the empty string. */ + p1 = p; /* To send to group_match_null_string_p. */ + + if (REG_MATCH_NULL_STRING_P (reg_info[*p]) == MATCH_NULL_UNSET_VALUE) + REG_MATCH_NULL_STRING_P (reg_info[*p]) + = group_match_null_string_p (&p1, pend, reg_info); + + /* Save the position in the string where we were the last time + we were at this open-group operator in case the group is + operated upon by a repetition operator, e.g., with `(a*)*b' + against `ab'; then we want to ignore where we are now in + the string in case this attempt to match fails. */ + old_regstart[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p]) + ? REG_UNSET (regstart[*p]) ? d : regstart[*p] + : regstart[*p]; + DEBUG_PRINT2 (" old_regstart: %d\n", + POINTER_TO_OFFSET (old_regstart[*p])); + + regstart[*p] = d; + DEBUG_PRINT2 (" regstart: %d\n", POINTER_TO_OFFSET (regstart[*p])); + + IS_ACTIVE (reg_info[*p]) = 1; + MATCHED_SOMETHING (reg_info[*p]) = 0; + + /* Clear this whenever we change the register activity status. */ + set_regs_matched_done = 0; + + /* This is the new highest active register. */ + highest_active_reg = *p; + + /* If nothing was active before, this is the new lowest active + register. */ + if (lowest_active_reg == NO_LOWEST_ACTIVE_REG) + lowest_active_reg = *p; + + /* Move past the register number and inner group count. */ + p += 2; + just_past_start_mem = p; + + break; + + + /* The stop_memory opcode represents the end of a group. Its + arguments are the same as start_memory's: the register + number, and the number of inner groups. */ + case stop_memory: + DEBUG_PRINT3 ("EXECUTING stop_memory %d (%d):\n", *p, p[1]); + + /* We need to save the string position the last time we were at + this close-group operator in case the group is operated + upon by a repetition operator, e.g., with `((a*)*(b*)*)*' + against `aba'; then we want to ignore where we are now in + the string in case this attempt to match fails. */ + old_regend[*p] = REG_MATCH_NULL_STRING_P (reg_info[*p]) + ? REG_UNSET (regend[*p]) ? d : regend[*p] + : regend[*p]; + DEBUG_PRINT2 (" old_regend: %d\n", + POINTER_TO_OFFSET (old_regend[*p])); + + regend[*p] = d; + DEBUG_PRINT2 (" regend: %d\n", POINTER_TO_OFFSET (regend[*p])); + + /* This register isn't active anymore. */ + IS_ACTIVE (reg_info[*p]) = 0; + + /* Clear this whenever we change the register activity status. */ + set_regs_matched_done = 0; + + /* If this was the only register active, nothing is active + anymore. */ + if (lowest_active_reg == highest_active_reg) + { + lowest_active_reg = NO_LOWEST_ACTIVE_REG; + highest_active_reg = NO_HIGHEST_ACTIVE_REG; + } + else + { /* We must scan for the new highest active register, since + it isn't necessarily one less than now: consider + (a(b)c(d(e)f)g). When group 3 ends, after the f), the + new highest active register is 1. */ + unsigned char r = *p - 1; + while (r > 0 && !IS_ACTIVE (reg_info[r])) + r--; + + /* If we end up at register zero, that means that we saved + the registers as the result of an `on_failure_jump', not + a `start_memory', and we jumped to past the innermost + `stop_memory'. For example, in ((.)*) we save + registers 1 and 2 as a result of the *, but when we pop + back to the second ), we are at the stop_memory 1. + Thus, nothing is active. */ + if (r == 0) + { + lowest_active_reg = NO_LOWEST_ACTIVE_REG; + highest_active_reg = NO_HIGHEST_ACTIVE_REG; + } + else + highest_active_reg = r; + } + + /* If just failed to match something this time around with a + group that's operated on by a repetition operator, try to + force exit from the ``loop'', and restore the register + information for this group that we had before trying this + last match. */ + if ((!MATCHED_SOMETHING (reg_info[*p]) + || just_past_start_mem == p - 1) + && (p + 2) < pend) + { + boolean is_a_jump_n = false; + + p1 = p + 2; + mcnt = 0; + switch ((re_opcode_t) *p1++) + { + case jump_n: + is_a_jump_n = true; + case pop_failure_jump: + case maybe_pop_jump: + case jump: + case dummy_failure_jump: + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + if (is_a_jump_n) + p1 += 2; + break; + + default: + /* do nothing */ ; + } + p1 += mcnt; + + /* If the next operation is a jump backwards in the pattern + to an on_failure_jump right before the start_memory + corresponding to this stop_memory, exit from the loop + by forcing a failure after pushing on the stack the + on_failure_jump's jump in the pattern, and d. */ + if (mcnt < 0 && (re_opcode_t) *p1 == on_failure_jump + && (re_opcode_t) p1[3] == start_memory && p1[4] == *p) + { + /* If this group ever matched anything, then restore + what its registers were before trying this last + failed match, e.g., with `(a*)*b' against `ab' for + regstart[1], and, e.g., with `((a*)*(b*)*)*' + against `aba' for regend[3]. + + Also restore the registers for inner groups for, + e.g., `((a*)(b*))*' against `aba' (register 3 would + otherwise get trashed). */ + + if (EVER_MATCHED_SOMETHING (reg_info[*p])) + { + unsigned r; + + EVER_MATCHED_SOMETHING (reg_info[*p]) = 0; + + /* Restore this and inner groups' (if any) registers. */ + for (r = *p; r < (unsigned) *p + (unsigned) *(p + 1); + r++) + { + regstart[r] = old_regstart[r]; + + /* xx why this test? */ + if (old_regend[r] >= regstart[r]) + regend[r] = old_regend[r]; + } + } + p1++; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + PUSH_FAILURE_POINT (p1 + mcnt, d, -2); + + goto fail; + } + } + + /* Move past the register number and the inner group count. */ + p += 2; + break; + + + /* \<digit> has been turned into a `duplicate' command which is + followed by the numeric value of <digit> as the register number. */ + case duplicate: + { + register const char *d2, *dend2; + int regno = *p++; /* Get which register to match against. */ + DEBUG_PRINT2 ("EXECUTING duplicate %d.\n", regno); + + /* Can't back reference a group which we've never matched. */ + if (REG_UNSET (regstart[regno]) || REG_UNSET (regend[regno])) + goto fail; + + /* Where in input to try to start matching. */ + d2 = regstart[regno]; + + /* Where to stop matching; if both the place to start and + the place to stop matching are in the same string, then + set to the place to stop, otherwise, for now have to use + the end of the first string. */ + + dend2 = ((FIRST_STRING_P (regstart[regno]) + == FIRST_STRING_P (regend[regno])) + ? regend[regno] : end_match_1); + for (;;) + { + /* If necessary, advance to next segment in register + contents. */ + while (d2 == dend2) + { + if (dend2 == end_match_2) break; + if (dend2 == regend[regno]) break; + + /* End of string1 => advance to string2. */ + d2 = string2; + dend2 = regend[regno]; + } + /* At end of register contents => success */ + if (d2 == dend2) break; + + /* If necessary, advance to next segment in data. */ + PREFETCH (); + + /* How many characters left in this segment to match. */ + mcnt = dend - d; + + /* Want how many consecutive characters we can match in + one shot, so, if necessary, adjust the count. */ + if (mcnt > dend2 - d2) + mcnt = dend2 - d2; + + /* Compare that many; failure if mismatch, else move + past them. */ + if (translate + ? bcmp_translate (d, d2, mcnt, translate) + : bcmp (d, d2, mcnt)) + goto fail; + d += mcnt, d2 += mcnt; + + /* Do this because we've match some characters. */ + SET_REGS_MATCHED (); + } + } + break; + + + /* begline matches the empty string at the beginning of the string + (unless `not_bol' is set in `bufp'), and, if + `newline_anchor' is set, after newlines. */ + case begline: + DEBUG_PRINT1 ("EXECUTING begline.\n"); + + if (AT_STRINGS_BEG (d)) + { + if (!bufp->not_bol) break; + } + else if (d[-1] == '\n' && bufp->newline_anchor) + { + break; + } + /* In all other cases, we fail. */ + goto fail; + + + /* endline is the dual of begline. */ + case endline: + DEBUG_PRINT1 ("EXECUTING endline.\n"); + + if (AT_STRINGS_END (d)) + { + if (!bufp->not_eol) break; + } + + /* We have to ``prefetch'' the next character. */ + else if ((d == end1 ? *string2 : *d) == '\n' + && bufp->newline_anchor) + { + break; + } + goto fail; + + + /* Match at the very beginning of the data. */ + case begbuf: + DEBUG_PRINT1 ("EXECUTING begbuf.\n"); + if (AT_STRINGS_BEG (d)) + break; + goto fail; + + + /* Match at the very end of the data. */ + case endbuf: + DEBUG_PRINT1 ("EXECUTING endbuf.\n"); + if (AT_STRINGS_END (d)) + break; + goto fail; + + + /* on_failure_keep_string_jump is used to optimize `.*\n'. It + pushes NULL as the value for the string on the stack. Then + `pop_failure_point' will keep the current value for the + string, instead of restoring it. To see why, consider + matching `foo\nbar' against `.*\n'. The .* matches the foo; + then the . fails against the \n. But the next thing we want + to do is match the \n against the \n; if we restored the + string value, we would be back at the foo. + + Because this is used only in specific cases, we don't need to + check all the things that `on_failure_jump' does, to make + sure the right things get saved on the stack. Hence we don't + share its code. The only reason to push anything on the + stack at all is that otherwise we would have to change + `anychar's code to do something besides goto fail in this + case; that seems worse than this. */ + case on_failure_keep_string_jump: + DEBUG_PRINT1 ("EXECUTING on_failure_keep_string_jump"); + + EXTRACT_NUMBER_AND_INCR (mcnt, p); +#ifdef _LIBC + DEBUG_PRINT3 (" %d (to %p):\n", mcnt, p + mcnt); +#else + DEBUG_PRINT3 (" %d (to 0x%x):\n", mcnt, p + mcnt); +#endif + + PUSH_FAILURE_POINT (p + mcnt, NULL, -2); + break; + + + /* Uses of on_failure_jump: + + Each alternative starts with an on_failure_jump that points + to the beginning of the next alternative. Each alternative + except the last ends with a jump that in effect jumps past + the rest of the alternatives. (They really jump to the + ending jump of the following alternative, because tensioning + these jumps is a hassle.) + + Repeats start with an on_failure_jump that points past both + the repetition text and either the following jump or + pop_failure_jump back to this on_failure_jump. */ + case on_failure_jump: + on_failure: + DEBUG_PRINT1 ("EXECUTING on_failure_jump"); + + EXTRACT_NUMBER_AND_INCR (mcnt, p); +#ifdef _LIBC + DEBUG_PRINT3 (" %d (to %p)", mcnt, p + mcnt); +#else + DEBUG_PRINT3 (" %d (to 0x%x)", mcnt, p + mcnt); +#endif + + /* If this on_failure_jump comes right before a group (i.e., + the original * applied to a group), save the information + for that group and all inner ones, so that if we fail back + to this point, the group's information will be correct. + For example, in \(a*\)*\1, we need the preceding group, + and in \(zz\(a*\)b*\)\2, we need the inner group. */ + + /* We can't use `p' to check ahead because we push + a failure point to `p + mcnt' after we do this. */ + p1 = p; + + /* We need to skip no_op's before we look for the + start_memory in case this on_failure_jump is happening as + the result of a completed succeed_n, as in \(a\)\{1,3\}b\1 + against aba. */ + while (p1 < pend && (re_opcode_t) *p1 == no_op) + p1++; + + if (p1 < pend && (re_opcode_t) *p1 == start_memory) + { + /* We have a new highest active register now. This will + get reset at the start_memory we are about to get to, + but we will have saved all the registers relevant to + this repetition op, as described above. */ + highest_active_reg = *(p1 + 1) + *(p1 + 2); + if (lowest_active_reg == NO_LOWEST_ACTIVE_REG) + lowest_active_reg = *(p1 + 1); + } + + DEBUG_PRINT1 (":\n"); + PUSH_FAILURE_POINT (p + mcnt, d, -2); + break; + + + /* A smart repeat ends with `maybe_pop_jump'. + We change it to either `pop_failure_jump' or `jump'. */ + case maybe_pop_jump: + EXTRACT_NUMBER_AND_INCR (mcnt, p); + DEBUG_PRINT2 ("EXECUTING maybe_pop_jump %d.\n", mcnt); + { + register unsigned char *p2 = p; + + /* Compare the beginning of the repeat with what in the + pattern follows its end. If we can establish that there + is nothing that they would both match, i.e., that we + would have to backtrack because of (as in, e.g., `a*a') + then we can change to pop_failure_jump, because we'll + never have to backtrack. + + This is not true in the case of alternatives: in + `(a|ab)*' we do need to backtrack to the `ab' alternative + (e.g., if the string was `ab'). But instead of trying to + detect that here, the alternative has put on a dummy + failure point which is what we will end up popping. */ + + /* Skip over open/close-group commands. + If what follows this loop is a ...+ construct, + look at what begins its body, since we will have to + match at least one of that. */ + while (1) + { + if (p2 + 2 < pend + && ((re_opcode_t) *p2 == stop_memory + || (re_opcode_t) *p2 == start_memory)) + p2 += 3; + else if (p2 + 6 < pend + && (re_opcode_t) *p2 == dummy_failure_jump) + p2 += 6; + else + break; + } + + p1 = p + mcnt; + /* p1[0] ... p1[2] are the `on_failure_jump' corresponding + to the `maybe_finalize_jump' of this case. Examine what + follows. */ + + /* If we're at the end of the pattern, we can change. */ + if (p2 == pend) + { + /* Consider what happens when matching ":\(.*\)" + against ":/". I don't really understand this code + yet. */ + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT1 + (" End of pattern: change to `pop_failure_jump'.\n"); + } + + else if ((re_opcode_t) *p2 == exactn + || (bufp->newline_anchor && (re_opcode_t) *p2 == endline)) + { + register unsigned char c + = *p2 == (unsigned char) endline ? '\n' : p2[2]; + + if ((re_opcode_t) p1[3] == exactn && p1[5] != c) + { + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT3 (" %c != %c => pop_failure_jump.\n", + c, p1[5]); + } + + else if ((re_opcode_t) p1[3] == charset + || (re_opcode_t) p1[3] == charset_not) + { + int not = (re_opcode_t) p1[3] == charset_not; + + if (c < (unsigned char) (p1[4] * BYTEWIDTH) + && p1[5 + c / BYTEWIDTH] & (1 << (c % BYTEWIDTH))) + not = !not; + + /* `not' is equal to 1 if c would match, which means + that we can't change to pop_failure_jump. */ + if (!not) + { + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT1 (" No match => pop_failure_jump.\n"); + } + } + } + else if ((re_opcode_t) *p2 == charset) + { +#ifdef DEBUG + register unsigned char c + = *p2 == (unsigned char) endline ? '\n' : p2[2]; +#endif + +#if 0 + if ((re_opcode_t) p1[3] == exactn + && ! ((int) p2[1] * BYTEWIDTH > (int) p1[5] + && (p2[2 + p1[5] / BYTEWIDTH] + & (1 << (p1[5] % BYTEWIDTH))))) +#else + if ((re_opcode_t) p1[3] == exactn + && ! ((int) p2[1] * BYTEWIDTH > (int) p1[4] + && (p2[2 + p1[4] / BYTEWIDTH] + & (1 << (p1[4] % BYTEWIDTH))))) +#endif + { + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT3 (" %c != %c => pop_failure_jump.\n", + c, p1[5]); + } + + else if ((re_opcode_t) p1[3] == charset_not) + { + int idx; + /* We win if the charset_not inside the loop + lists every character listed in the charset after. */ + for (idx = 0; idx < (int) p2[1]; idx++) + if (! (p2[2 + idx] == 0 + || (idx < (int) p1[4] + && ((p2[2 + idx] & ~ p1[5 + idx]) == 0)))) + break; + + if (idx == p2[1]) + { + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT1 (" No match => pop_failure_jump.\n"); + } + } + else if ((re_opcode_t) p1[3] == charset) + { + int idx; + /* We win if the charset inside the loop + has no overlap with the one after the loop. */ + for (idx = 0; + idx < (int) p2[1] && idx < (int) p1[4]; + idx++) + if ((p2[2 + idx] & p1[5 + idx]) != 0) + break; + + if (idx == p2[1] || idx == p1[4]) + { + p[-3] = (unsigned char) pop_failure_jump; + DEBUG_PRINT1 (" No match => pop_failure_jump.\n"); + } + } + } + } + p -= 2; /* Point at relative address again. */ + if ((re_opcode_t) p[-1] != pop_failure_jump) + { + p[-1] = (unsigned char) jump; + DEBUG_PRINT1 (" Match => jump.\n"); + goto unconditional_jump; + } + /* Note fall through. */ + + + /* The end of a simple repeat has a pop_failure_jump back to + its matching on_failure_jump, where the latter will push a + failure point. The pop_failure_jump takes off failure + points put on by this pop_failure_jump's matching + on_failure_jump; we got through the pattern to here from the + matching on_failure_jump, so didn't fail. */ + case pop_failure_jump: + { + /* We need to pass separate storage for the lowest and + highest registers, even though we don't care about the + actual values. Otherwise, we will restore only one + register from the stack, since lowest will == highest in + `pop_failure_point'. */ + active_reg_t dummy_low_reg, dummy_high_reg; + unsigned char *pdummy; + const char *sdummy; + + DEBUG_PRINT1 ("EXECUTING pop_failure_jump.\n"); + POP_FAILURE_POINT (sdummy, pdummy, + dummy_low_reg, dummy_high_reg, + reg_dummy, reg_dummy, reg_info_dummy); + } + /* Note fall through. */ + + unconditional_jump: +#ifdef _LIBC + DEBUG_PRINT2 ("\n%p: ", p); +#else + DEBUG_PRINT2 ("\n0x%x: ", p); +#endif + /* Note fall through. */ + + /* Unconditionally jump (without popping any failure points). */ + case jump: + EXTRACT_NUMBER_AND_INCR (mcnt, p); /* Get the amount to jump. */ + DEBUG_PRINT2 ("EXECUTING jump %d ", mcnt); + p += mcnt; /* Do the jump. */ +#ifdef _LIBC + DEBUG_PRINT2 ("(to %p).\n", p); +#else + DEBUG_PRINT2 ("(to 0x%x).\n", p); +#endif + break; + + + /* We need this opcode so we can detect where alternatives end + in `group_match_null_string_p' et al. */ + case jump_past_alt: + DEBUG_PRINT1 ("EXECUTING jump_past_alt.\n"); + goto unconditional_jump; + + + /* Normally, the on_failure_jump pushes a failure point, which + then gets popped at pop_failure_jump. We will end up at + pop_failure_jump, also, and with a pattern of, say, `a+', we + are skipping over the on_failure_jump, so we have to push + something meaningless for pop_failure_jump to pop. */ + case dummy_failure_jump: + DEBUG_PRINT1 ("EXECUTING dummy_failure_jump.\n"); + /* It doesn't matter what we push for the string here. What + the code at `fail' tests is the value for the pattern. */ + PUSH_FAILURE_POINT (0, 0, -2); + goto unconditional_jump; + + + /* At the end of an alternative, we need to push a dummy failure + point in case we are followed by a `pop_failure_jump', because + we don't want the failure point for the alternative to be + popped. For example, matching `(a|ab)*' against `aab' + requires that we match the `ab' alternative. */ + case push_dummy_failure: + DEBUG_PRINT1 ("EXECUTING push_dummy_failure.\n"); + /* See comments just above at `dummy_failure_jump' about the + two zeroes. */ + PUSH_FAILURE_POINT (0, 0, -2); + break; + + /* Have to succeed matching what follows at least n times. + After that, handle like `on_failure_jump'. */ + case succeed_n: + EXTRACT_NUMBER (mcnt, p + 2); + DEBUG_PRINT2 ("EXECUTING succeed_n %d.\n", mcnt); + + assert (mcnt >= 0); + /* Originally, this is how many times we HAVE to succeed. */ + if (mcnt > 0) + { + mcnt--; + p += 2; + STORE_NUMBER_AND_INCR (p, mcnt); +#ifdef _LIBC + DEBUG_PRINT3 (" Setting %p to %d.\n", p - 2, mcnt); +#else + DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p - 2, mcnt); +#endif + } + else if (mcnt == 0) + { +#ifdef _LIBC + DEBUG_PRINT2 (" Setting two bytes from %p to no_op.\n", p+2); +#else + DEBUG_PRINT2 (" Setting two bytes from 0x%x to no_op.\n", p+2); +#endif + p[2] = (unsigned char) no_op; + p[3] = (unsigned char) no_op; + goto on_failure; + } + break; + + case jump_n: + EXTRACT_NUMBER (mcnt, p + 2); + DEBUG_PRINT2 ("EXECUTING jump_n %d.\n", mcnt); + + /* Originally, this is how many times we CAN jump. */ + if (mcnt) + { + mcnt--; + STORE_NUMBER (p + 2, mcnt); +#ifdef _LIBC + DEBUG_PRINT3 (" Setting %p to %d.\n", p + 2, mcnt); +#else + DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p + 2, mcnt); +#endif + goto unconditional_jump; + } + /* If don't have to jump any more, skip over the rest of command. */ + else + p += 4; + break; + + case set_number_at: + { + DEBUG_PRINT1 ("EXECUTING set_number_at.\n"); + + EXTRACT_NUMBER_AND_INCR (mcnt, p); + p1 = p + mcnt; + EXTRACT_NUMBER_AND_INCR (mcnt, p); +#ifdef _LIBC + DEBUG_PRINT3 (" Setting %p to %d.\n", p1, mcnt); +#else + DEBUG_PRINT3 (" Setting 0x%x to %d.\n", p1, mcnt); +#endif + STORE_NUMBER (p1, mcnt); + break; + } + +#if 0 + /* The DEC Alpha C compiler 3.x generates incorrect code for the + test WORDCHAR_P (d - 1) != WORDCHAR_P (d) in the expansion of + AT_WORD_BOUNDARY, so this code is disabled. Expanding the + macro and introducing temporary variables works around the bug. */ + + case wordbound: + DEBUG_PRINT1 ("EXECUTING wordbound.\n"); + if (AT_WORD_BOUNDARY (d)) + break; + goto fail; + + case notwordbound: + DEBUG_PRINT1 ("EXECUTING notwordbound.\n"); + if (AT_WORD_BOUNDARY (d)) + goto fail; + break; +#else + case wordbound: + { + boolean prevchar, thischar; + + DEBUG_PRINT1 ("EXECUTING wordbound.\n"); + if (AT_STRINGS_BEG (d) || AT_STRINGS_END (d)) + break; + + prevchar = WORDCHAR_P (d - 1); + thischar = WORDCHAR_P (d); + if (prevchar != thischar) + break; + goto fail; + } + + case notwordbound: + { + boolean prevchar, thischar; + + DEBUG_PRINT1 ("EXECUTING notwordbound.\n"); + if (AT_STRINGS_BEG (d) || AT_STRINGS_END (d)) + goto fail; + + prevchar = WORDCHAR_P (d - 1); + thischar = WORDCHAR_P (d); + if (prevchar != thischar) + goto fail; + break; + } +#endif + + case wordbeg: + DEBUG_PRINT1 ("EXECUTING wordbeg.\n"); + if (WORDCHAR_P (d) && (AT_STRINGS_BEG (d) || !WORDCHAR_P (d - 1))) + break; + goto fail; + + case wordend: + DEBUG_PRINT1 ("EXECUTING wordend.\n"); + if (!AT_STRINGS_BEG (d) && WORDCHAR_P (d - 1) + && (!WORDCHAR_P (d) || AT_STRINGS_END (d))) + break; + goto fail; + +#ifdef emacs + case before_dot: + DEBUG_PRINT1 ("EXECUTING before_dot.\n"); + if (PTR_CHAR_POS ((unsigned char *) d) >= point) + goto fail; + break; + + case at_dot: + DEBUG_PRINT1 ("EXECUTING at_dot.\n"); + if (PTR_CHAR_POS ((unsigned char *) d) != point) + goto fail; + break; + + case after_dot: + DEBUG_PRINT1 ("EXECUTING after_dot.\n"); + if (PTR_CHAR_POS ((unsigned char *) d) <= point) + goto fail; + break; + + case syntaxspec: + DEBUG_PRINT2 ("EXECUTING syntaxspec %d.\n", mcnt); + mcnt = *p++; + goto matchsyntax; + + case wordchar: + DEBUG_PRINT1 ("EXECUTING Emacs wordchar.\n"); + mcnt = (int) Sword; + matchsyntax: + PREFETCH (); + /* Can't use *d++ here; SYNTAX may be an unsafe macro. */ + d++; + if (SYNTAX (d[-1]) != (enum syntaxcode) mcnt) + goto fail; + SET_REGS_MATCHED (); + break; + + case notsyntaxspec: + DEBUG_PRINT2 ("EXECUTING notsyntaxspec %d.\n", mcnt); + mcnt = *p++; + goto matchnotsyntax; + + case notwordchar: + DEBUG_PRINT1 ("EXECUTING Emacs notwordchar.\n"); + mcnt = (int) Sword; + matchnotsyntax: + PREFETCH (); + /* Can't use *d++ here; SYNTAX may be an unsafe macro. */ + d++; + if (SYNTAX (d[-1]) == (enum syntaxcode) mcnt) + goto fail; + SET_REGS_MATCHED (); + break; + +#else /* not emacs */ + case wordchar: + DEBUG_PRINT1 ("EXECUTING non-Emacs wordchar.\n"); + PREFETCH (); + if (!WORDCHAR_P (d)) + goto fail; + SET_REGS_MATCHED (); + d++; + break; + + case notwordchar: + DEBUG_PRINT1 ("EXECUTING non-Emacs notwordchar.\n"); + PREFETCH (); + if (WORDCHAR_P (d)) + goto fail; + SET_REGS_MATCHED (); + d++; + break; +#endif /* not emacs */ + + default: + abort (); + } + continue; /* Successfully executed one pattern command; keep going. */ + + + /* We goto here if a matching operation fails. */ + fail: + if (!FAIL_STACK_EMPTY ()) + { /* A restart point is known. Restore to that state. */ + DEBUG_PRINT1 ("\nFAIL:\n"); + POP_FAILURE_POINT (d, p, + lowest_active_reg, highest_active_reg, + regstart, regend, reg_info); + + /* If this failure point is a dummy, try the next one. */ + if (!p) + goto fail; + + /* If we failed to the end of the pattern, don't examine *p. */ + assert (p <= pend); + if (p < pend) + { + boolean is_a_jump_n = false; + + /* If failed to a backwards jump that's part of a repetition + loop, need to pop this failure point and use the next one. */ + switch ((re_opcode_t) *p) + { + case jump_n: + is_a_jump_n = true; + case maybe_pop_jump: + case pop_failure_jump: + case jump: + p1 = p + 1; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + p1 += mcnt; + + if ((is_a_jump_n && (re_opcode_t) *p1 == succeed_n) + || (!is_a_jump_n + && (re_opcode_t) *p1 == on_failure_jump)) + goto fail; + break; + default: + /* do nothing */ ; + } + } + + if (d >= string1 && d <= end1) + dend = end_match_1; + } + else + break; /* Matching at this starting point really fails. */ + } /* for (;;) */ + + if (best_regs_set) + goto restore_best_regs; + + FREE_VARIABLES (); + + return -1; /* Failure to match. */ +} /* re_match_2 */ + +/* Subroutine definitions for re_match_2. */ + + +/* We are passed P pointing to a register number after a start_memory. + + Return true if the pattern up to the corresponding stop_memory can + match the empty string, and false otherwise. + + If we find the matching stop_memory, sets P to point to one past its number. + Otherwise, sets P to an undefined byte less than or equal to END. + + We don't handle duplicates properly (yet). */ + +static boolean +group_match_null_string_p (p, end, reg_info) + unsigned char **p, *end; + register_info_type *reg_info; +{ + int mcnt; + /* Point to after the args to the start_memory. */ + unsigned char *p1 = *p + 2; + + while (p1 < end) + { + /* Skip over opcodes that can match nothing, and return true or + false, as appropriate, when we get to one that can't, or to the + matching stop_memory. */ + + switch ((re_opcode_t) *p1) + { + /* Could be either a loop or a series of alternatives. */ + case on_failure_jump: + p1++; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + + /* If the next operation is not a jump backwards in the + pattern. */ + + if (mcnt >= 0) + { + /* Go through the on_failure_jumps of the alternatives, + seeing if any of the alternatives cannot match nothing. + The last alternative starts with only a jump, + whereas the rest start with on_failure_jump and end + with a jump, e.g., here is the pattern for `a|b|c': + + /on_failure_jump/0/6/exactn/1/a/jump_past_alt/0/6 + /on_failure_jump/0/6/exactn/1/b/jump_past_alt/0/3 + /exactn/1/c + + So, we have to first go through the first (n-1) + alternatives and then deal with the last one separately. */ + + + /* Deal with the first (n-1) alternatives, which start + with an on_failure_jump (see above) that jumps to right + past a jump_past_alt. */ + + while ((re_opcode_t) p1[mcnt-3] == jump_past_alt) + { + /* `mcnt' holds how many bytes long the alternative + is, including the ending `jump_past_alt' and + its number. */ + + if (!alt_match_null_string_p (p1, p1 + mcnt - 3, + reg_info)) + return false; + + /* Move to right after this alternative, including the + jump_past_alt. */ + p1 += mcnt; + + /* Break if it's the beginning of an n-th alternative + that doesn't begin with an on_failure_jump. */ + if ((re_opcode_t) *p1 != on_failure_jump) + break; + + /* Still have to check that it's not an n-th + alternative that starts with an on_failure_jump. */ + p1++; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + if ((re_opcode_t) p1[mcnt-3] != jump_past_alt) + { + /* Get to the beginning of the n-th alternative. */ + p1 -= 3; + break; + } + } + + /* Deal with the last alternative: go back and get number + of the `jump_past_alt' just before it. `mcnt' contains + the length of the alternative. */ + EXTRACT_NUMBER (mcnt, p1 - 2); + + if (!alt_match_null_string_p (p1, p1 + mcnt, reg_info)) + return false; + + p1 += mcnt; /* Get past the n-th alternative. */ + } /* if mcnt > 0 */ + break; + + + case stop_memory: + assert (p1[1] == **p); + *p = p1 + 2; + return true; + + + default: + if (!common_op_match_null_string_p (&p1, end, reg_info)) + return false; + } + } /* while p1 < end */ + + return false; +} /* group_match_null_string_p */ + + +/* Similar to group_match_null_string_p, but doesn't deal with alternatives: + It expects P to be the first byte of a single alternative and END one + byte past the last. The alternative can contain groups. */ + +static boolean +alt_match_null_string_p (p, end, reg_info) + unsigned char *p, *end; + register_info_type *reg_info; +{ + int mcnt; + unsigned char *p1 = p; + + while (p1 < end) + { + /* Skip over opcodes that can match nothing, and break when we get + to one that can't. */ + + switch ((re_opcode_t) *p1) + { + /* It's a loop. */ + case on_failure_jump: + p1++; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + p1 += mcnt; + break; + + default: + if (!common_op_match_null_string_p (&p1, end, reg_info)) + return false; + } + } /* while p1 < end */ + + return true; +} /* alt_match_null_string_p */ + + +/* Deals with the ops common to group_match_null_string_p and + alt_match_null_string_p. + + Sets P to one after the op and its arguments, if any. */ + +static boolean +common_op_match_null_string_p (p, end, reg_info) + unsigned char **p, *end; + register_info_type *reg_info; +{ + int mcnt; + boolean ret; + int reg_no; + unsigned char *p1 = *p; + + switch ((re_opcode_t) *p1++) + { + case no_op: + case begline: + case endline: + case begbuf: + case endbuf: + case wordbeg: + case wordend: + case wordbound: + case notwordbound: +#ifdef emacs + case before_dot: + case at_dot: + case after_dot: +#endif + break; + + case start_memory: + reg_no = *p1; + assert (reg_no > 0 && reg_no <= MAX_REGNUM); + ret = group_match_null_string_p (&p1, end, reg_info); + + /* Have to set this here in case we're checking a group which + contains a group and a back reference to it. */ + + if (REG_MATCH_NULL_STRING_P (reg_info[reg_no]) == MATCH_NULL_UNSET_VALUE) + REG_MATCH_NULL_STRING_P (reg_info[reg_no]) = ret; + + if (!ret) + return false; + break; + + /* If this is an optimized succeed_n for zero times, make the jump. */ + case jump: + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + if (mcnt >= 0) + p1 += mcnt; + else + return false; + break; + + case succeed_n: + /* Get to the number of times to succeed. */ + p1 += 2; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + + if (mcnt == 0) + { + p1 -= 4; + EXTRACT_NUMBER_AND_INCR (mcnt, p1); + p1 += mcnt; + } + else + return false; + break; + + case duplicate: + if (!REG_MATCH_NULL_STRING_P (reg_info[*p1])) + return false; + break; + + case set_number_at: + p1 += 4; + + default: + /* All other opcodes mean we cannot match the empty string. */ + return false; + } + + *p = p1; + return true; +} /* common_op_match_null_string_p */ + + +/* Return zero if TRANSLATE[S1] and TRANSLATE[S2] are identical for LEN + bytes; nonzero otherwise. */ + +static int +bcmp_translate (s1, s2, len, translate) + const char *s1, *s2; + register int len; + RE_TRANSLATE_TYPE translate; +{ + register const unsigned char *p1 = (const unsigned char *) s1; + register const unsigned char *p2 = (const unsigned char *) s2; + while (len) + { + if (translate[*p1++] != translate[*p2++]) return 1; + len--; + } + return 0; +} + +/* Entry points for GNU code. */ + +/* re_compile_pattern is the GNU regular expression compiler: it + compiles PATTERN (of length SIZE) and puts the result in BUFP. + Returns 0 if the pattern was valid, otherwise an error string. + + Assumes the `allocated' (and perhaps `buffer') and `translate' fields + are set in BUFP on entry. + + We call regex_compile to do the actual compilation. */ + +const char * +re_compile_pattern (pattern, length, bufp) + const char *pattern; + size_t length; + struct re_pattern_buffer *bufp; +{ + reg_errcode_t ret; + + /* GNU code is written to assume at least RE_NREGS registers will be set + (and at least one extra will be -1). */ + bufp->regs_allocated = REGS_UNALLOCATED; + + /* And GNU code determines whether or not to get register information + by passing null for the REGS argument to re_match, etc., not by + setting no_sub. */ + bufp->no_sub = 0; + + /* Match anchors at newline. */ + bufp->newline_anchor = 1; + + ret = regex_compile (pattern, length, re_syntax_options, bufp); + + if (!ret) + return NULL; + return gettext (re_error_msgid[(int) ret]); +} + +/* Entry points compatible with 4.2 BSD regex library. We don't define + them unless specifically requested. */ + +#if defined (_REGEX_RE_COMP) || defined (_LIBC) + +/* BSD has one and only one pattern buffer. */ +static struct re_pattern_buffer re_comp_buf; + +char * +#ifdef _LIBC +/* Make these definitions weak in libc, so POSIX programs can redefine + these names if they don't use our functions, and still use + regcomp/regexec below without link errors. */ +weak_function +#endif +re_comp (s) + const char *s; +{ + reg_errcode_t ret; + + if (!s) + { + if (!re_comp_buf.buffer) + return gettext ("No previous regular expression"); + return 0; + } + + if (!re_comp_buf.buffer) + { + re_comp_buf.buffer = (unsigned char *) malloc (200); + if (re_comp_buf.buffer == NULL) + return gettext (re_error_msgid[(int) REG_ESPACE]); + re_comp_buf.allocated = 200; + + re_comp_buf.fastmap = (char *) malloc (1 << BYTEWIDTH); + if (re_comp_buf.fastmap == NULL) + return gettext (re_error_msgid[(int) REG_ESPACE]); + } + + /* Since `re_exec' always passes NULL for the `regs' argument, we + don't need to initialize the pattern buffer fields which affect it. */ + + /* Match anchors at newlines. */ + re_comp_buf.newline_anchor = 1; + + ret = regex_compile (s, strlen (s), re_syntax_options, &re_comp_buf); + + if (!ret) + return NULL; + + /* Yes, we're discarding `const' here if !HAVE_LIBINTL. */ + return (char *) gettext (re_error_msgid[(int) ret]); +} + + +int +#ifdef _LIBC +weak_function +#endif +re_exec (s) + const char *s; +{ + const int len = strlen (s); + return + 0 <= re_search (&re_comp_buf, s, len, 0, len, (struct re_registers *) 0); +} + +#endif /* _REGEX_RE_COMP */ + +/* POSIX.2 functions. Don't define these for Emacs. */ + +#ifndef emacs + +/* regcomp takes a regular expression as a string and compiles it. + + PREG is a regex_t *. We do not expect any fields to be initialized, + since POSIX says we shouldn't. Thus, we set + + `buffer' to the compiled pattern; + `used' to the length of the compiled pattern; + `syntax' to RE_SYNTAX_POSIX_EXTENDED if the + REG_EXTENDED bit in CFLAGS is set; otherwise, to + RE_SYNTAX_POSIX_BASIC; + `newline_anchor' to REG_NEWLINE being set in CFLAGS; + `fastmap' and `fastmap_accurate' to zero; + `re_nsub' to the number of subexpressions in PATTERN. + + PATTERN is the address of the pattern string. + + CFLAGS is a series of bits which affect compilation. + + If REG_EXTENDED is set, we use POSIX extended syntax; otherwise, we + use POSIX basic syntax. + + If REG_NEWLINE is set, then . and [^...] don't match newline. + Also, regexec will try a match beginning after every newline. + + If REG_ICASE is set, then we considers upper- and lowercase + versions of letters to be equivalent when matching. + + If REG_NOSUB is set, then when PREG is passed to regexec, that + routine will report only success or failure, and nothing about the + registers. + + It returns 0 if it succeeds, nonzero if it doesn't. (See regex.h for + the return codes and their meanings.) */ + +int +regcomp (preg, pattern, cflags) + regex_t *preg; + const char *pattern; + int cflags; +{ + reg_errcode_t ret; + reg_syntax_t syntax + = (cflags & REG_EXTENDED) ? + RE_SYNTAX_POSIX_EXTENDED : RE_SYNTAX_POSIX_BASIC; + + /* regex_compile will allocate the space for the compiled pattern. */ + preg->buffer = 0; + preg->allocated = 0; + preg->used = 0; + + /* Don't bother to use a fastmap when searching. This simplifies the + REG_NEWLINE case: if we used a fastmap, we'd have to put all the + characters after newlines into the fastmap. This way, we just try + every character. */ + preg->fastmap = 0; + + if (cflags & REG_ICASE) + { + unsigned i; + + preg->translate + = (RE_TRANSLATE_TYPE) malloc (CHAR_SET_SIZE + * sizeof (*(RE_TRANSLATE_TYPE)0)); + if (preg->translate == NULL) + return (int) REG_ESPACE; + + /* Map uppercase characters to corresponding lowercase ones. */ + for (i = 0; i < CHAR_SET_SIZE; i++) + preg->translate[i] = ISUPPER (i) ? tolower (i) : i; + } + else + preg->translate = NULL; + + /* If REG_NEWLINE is set, newlines are treated differently. */ + if (cflags & REG_NEWLINE) + { /* REG_NEWLINE implies neither . nor [^...] match newline. */ + syntax &= ~RE_DOT_NEWLINE; + syntax |= RE_HAT_LISTS_NOT_NEWLINE; + /* It also changes the matching behavior. */ + preg->newline_anchor = 1; + } + else + preg->newline_anchor = 0; + + preg->no_sub = !!(cflags & REG_NOSUB); + + /* POSIX says a null character in the pattern terminates it, so we + can use strlen here in compiling the pattern. */ + ret = regex_compile (pattern, strlen (pattern), syntax, preg); + + /* POSIX doesn't distinguish between an unmatched open-group and an + unmatched close-group: both are REG_EPAREN. */ + if (ret == REG_ERPAREN) ret = REG_EPAREN; + + return (int) ret; +} + + +/* regexec searches for a given pattern, specified by PREG, in the + string STRING. + + If NMATCH is zero or REG_NOSUB was set in the cflags argument to + `regcomp', we ignore PMATCH. Otherwise, we assume PMATCH has at + least NMATCH elements, and we set them to the offsets of the + corresponding matched substrings. + + EFLAGS specifies `execution flags' which affect matching: if + REG_NOTBOL is set, then ^ does not match at the beginning of the + string; if REG_NOTEOL is set, then $ does not match at the end. + + We return 0 if we find a match and REG_NOMATCH if not. */ + +int +regexec (preg, string, nmatch, pmatch, eflags) + const regex_t *preg; + const char *string; + size_t nmatch; + regmatch_t pmatch[]; + int eflags; +{ + int ret; + struct re_registers regs; + regex_t private_preg; + int len = strlen (string); + boolean want_reg_info = !preg->no_sub && nmatch > 0; + + private_preg = *preg; + + private_preg.not_bol = !!(eflags & REG_NOTBOL); + private_preg.not_eol = !!(eflags & REG_NOTEOL); + + /* The user has told us exactly how many registers to return + information about, via `nmatch'. We have to pass that on to the + matching routines. */ + private_preg.regs_allocated = REGS_FIXED; + + if (want_reg_info) + { + regs.num_regs = nmatch; + regs.start = TALLOC (nmatch, regoff_t); + regs.end = TALLOC (nmatch, regoff_t); + if (regs.start == NULL || regs.end == NULL) + return (int) REG_NOMATCH; + } + + /* Perform the searching operation. */ + ret = re_search (&private_preg, string, len, + /* start: */ 0, /* range: */ len, + want_reg_info ? ®s : (struct re_registers *) 0); + + /* Copy the register information to the POSIX structure. */ + if (want_reg_info) + { + if (ret >= 0) + { + unsigned r; + + for (r = 0; r < nmatch; r++) + { + pmatch[r].rm_so = regs.start[r]; + pmatch[r].rm_eo = regs.end[r]; + } + } + + /* If we needed the temporary register info, free the space now. */ + free (regs.start); + free (regs.end); + } + + /* We want zero return to mean success, unlike `re_search'. */ + return ret >= 0 ? (int) REG_NOERROR : (int) REG_NOMATCH; +} + + +/* Returns a message corresponding to an error code, ERRCODE, returned + from either regcomp or regexec. We don't use PREG here. */ + +size_t +regerror (int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size) +{ + const char *msg; + size_t msg_size; + + if (errcode < 0 + || errcode >= (int) (sizeof (re_error_msgid) + / sizeof (re_error_msgid[0]))) + /* Only error codes returned by the rest of the code should be passed + to this routine. If we are given anything else, or if other regex + code generates an invalid error code, then the program has a bug. + Dump core so we can fix it. */ + abort (); + + msg = gettext (re_error_msgid[errcode]); + + msg_size = strlen (msg) + 1; /* Includes the null. */ + + if (errbuf_size != 0) + { + if (msg_size > errbuf_size) + { + strncpy (errbuf, msg, errbuf_size - 1); + errbuf[errbuf_size - 1] = 0; + } + else + strcpy (errbuf, msg); + } + + return msg_size; +} + + +/* Free dynamically allocated space used by PREG. */ + +void +regfree (preg) + regex_t *preg; +{ + if (preg->buffer != NULL) + free (preg->buffer); + preg->buffer = NULL; + + preg->allocated = 0; + preg->used = 0; + + if (preg->fastmap != NULL) + free (preg->fastmap); + preg->fastmap = NULL; + preg->fastmap_accurate = 0; + + if (preg->translate != NULL) + free (preg->translate); + preg->translate = NULL; +} + +#endif /* not emacs */ diff --git a/src/utilfuns/roman.cpp b/src/utilfuns/roman.cpp new file mode 100644 index 0000000..85f361d --- /dev/null +++ b/src/utilfuns/roman.cpp @@ -0,0 +1,87 @@ +/* + * roman.c + * Copyright 2001 by CrossWire Bible Society + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, + * MA 02111-1307, USA. + * + */ + +#include <stdlib.h> +#include <string.h> +#include <roman.h> + +SWORD_NAMESPACE_START + +char isroman (const char* str) { + char * ch = (char*)str; + for (; *ch; ch++) + if (!strchr("IVXLCDMivxlcdm ", *ch)) + return 0; + return 1; +} + +int from_rom(const char* str) { + int i, n = strlen(str); + short * num= (short *) calloc(n, sizeof(short)); + for (i = 0; str[i]; i++) { + switch(str[i]) { + case 'i': + case 'I': + num[i] = 1; + break; + case 'v': + case 'V': + num[i] = 5; + break; + case 'x': + case 'X': + num[i] = 10; + break; + case 'l': + case 'L': + num[i] = 50; + break; + case 'c': + case 'C': + num[i] = 100; + break; + case 'd': + case 'D': + num[i] = 500; + break; + case 'm': + case 'M': + num[i] = 1000; + break; + default: + num[i] = 0; + } + } + for (i = 1; str[i]; i++) { + if (num[i] > num[i-1]) { + num[i] -= num[i-1]; + num[i-1] = 0; + } + } + n = 0; + for (i = 0; str[i]; i++) { + n += num[i]; + } + free(num); + return n; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/sub.c b/src/utilfuns/sub.c new file mode 100644 index 0000000..fcea6e2 --- /dev/null +++ b/src/utilfuns/sub.c @@ -0,0 +1,36 @@ + +#include <stdio.h> +#include <stdlib.h> + +main(int argc, char **argv) +{ + FILE *fp; + char *buf; + int size; + + if ((argc < 3) || (argc > 4)) { + fprintf(stderr, "usage: %s <string> <substitute string> [filename]\n", *argv); + exit(-1); + } + + if (argc > 3) + fp = fopen(argv[3], "r"); + else fp = stdin; + + size = strlen(argv[1]); + buf = (char *)calloc(size + 1, 1); + + while ((buf[size - 1] = fgetc(fp)) != EOF) { + if (!strcmp(buf, argv[1])) { + printf("\n%s", argv[2]); + memset(buf, 0, size); + continue; + } + if (*buf) { + printf("%c", *buf); + } + memmove(buf, &buf[1], size); + } + buf[size - 1] = 0; + printf("%s", buf); +} diff --git a/src/utilfuns/swbuf.cpp b/src/utilfuns/swbuf.cpp new file mode 100644 index 0000000..430a705 --- /dev/null +++ b/src/utilfuns/swbuf.cpp @@ -0,0 +1,167 @@ +/****************************************************************************** +* swbuf.cpp - code for SWBuf used as a transport and utility for data buffers +* +* $Id: swbuf.cpp 2169 2008-05-18 02:50:53Z scribe $ +* +* Copyright 2003 CrossWire Bible Society (http://www.crosswire.org) +* CrossWire Bible Society +* P. O. Box 2528 +* Tempe, AZ 85280-2528 +* +* This program is free software; you can redistribute it and/or modify it +* under the terms of the GNU General Public License as published by the +* Free Software Foundation version 2. +* +* This program is distributed in the hope that it will be useful, but +* WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +* General Public License for more details. +* +*/ + +#include <swbuf.h> + +#include <stdlib.h> +#include <stdarg.h> +#include <stdio.h> + +SWORD_NAMESPACE_START + +char *SWBuf::nullStr = (char *)""; +char SWBuf::junkBuf[JUNKBUFSIZE]; + +/****************************************************************************** +* SWBuf Constructor - Creates an empty SWBuf object or an SWBuf initialized +* to a value from a const char * +* +*/ +SWBuf::SWBuf(const char *initVal, unsigned long initSize) { + init(initSize); + if (initVal) + set(initVal); +} + +/****************************************************************************** +* SWBuf Constructor - Creates an SWBuf initialized +* to a value from another SWBuf +* +*/ +SWBuf::SWBuf(const SWBuf &other, unsigned long initSize) { + init(initSize); + set(other); +} + +/****************************************************************************** +* SWBuf Constructor - Creates an SWBuf initialized +* to a value from a char +* +*/ +SWBuf::SWBuf(char initVal, unsigned long initSize) { + init(initSize+1); + *buf = initVal; + end = buf+1; + *end = 0; +} + +/* +SWBuf::SWBuf(unsigned long initSize) { + init(initSize); + set((const char *)0); +} +*/ + + +/****************************************************************************** +* SWBuf::setFormatted - sets this buf to a formatted string +* WARNING: This function can only write at most +* JUNKBUFSIZE to the string per call. +*/ +void SWBuf::setFormatted(const char *format, ...) { + va_list argptr; + + va_start(argptr, format); +#ifdef NO_VSNPRINTF + int len = vsprintf(junkBuf, format, argptr)+1; +#else + int len = vsnprintf(0, 0, format, argptr)+1; +#endif + va_end(argptr); + assureSize(len); + va_start(argptr, format); + end = vsprintf(buf, format, argptr) + buf; + va_end(argptr); +} + +/****************************************************************************** +* SWBuf::append - appends a value to the current value of this SWBuf +* +*/ +void SWBuf::append(const char *str, long max) { +// if (!str) //A null string was passed +// return; + if (max < 0) + max = strlen(str); + assureMore(max+1); + for (;((max)&&(*str));max--) + *end++ = *str++; + *end = 0; +} + +/****************************************************************************** +* SWBuf::setSize - Size this buffer to a specific length +*/ +void SWBuf::setSize(unsigned long len) { + assureSize(len+1); + if ((unsigned)(end - buf) < len) + memset(end, fillByte, len - (end-buf)); + end = buf + len; + *end = 0; +} + +/****************************************************************************** +* SWBuf::appendFormatted - appends formatted strings to the current value of this SWBuf +* WARNING: This function can only write at most +* JUNKBUFSIZE to the string per call. +*/ +void SWBuf::appendFormatted(const char *format, ...) { + va_list argptr; + + va_start(argptr, format); +#ifdef NO_VSNPRINTF + int len = vsprintf(junkBuf, format, argptr)+1; +#else + int len = vsnprintf(0, 0, format, argptr)+1; +#endif + va_end(argptr); + assureMore(len); + va_start(argptr, format); + end += vsprintf(end, format, argptr); + va_end(argptr); +} + +void SWBuf::insert(unsigned long pos, const char* str, unsigned long start, signed long max) { +// if (!str) //A null string was passed +// return; + + str += start; + int len = (max > -1) ? max : strlen(str); + + if (!len || (pos > length())) //nothing to do, return + return; + + // pos==length(), so we can call append in this case + if (pos == length()) { //append is more efficient + append(str, max); + return; + } + + assureMore( len ); + + memmove(buf + pos + len, buf + pos, (end - buf) - pos); //make a gap of "len" bytes + memcpy(buf+pos, str, len); + + end += len; + *end = 0; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/swobject.cpp b/src/utilfuns/swobject.cpp new file mode 100644 index 0000000..227911b --- /dev/null +++ b/src/utilfuns/swobject.cpp @@ -0,0 +1,46 @@ +/****************************************************************************** +* swobject.cpp - code for SWObject used as lowest base class for many +* SWORD objects +* +* $Id: swobject.cpp 1785 2005-04-30 19:11:12Z scribe $ +* +* Copyright 2005 CrossWire Bible Society (http://www.crosswire.org) +* CrossWire Bible Society +* P. O. Box 2528 +* Tempe, AZ 85280-2528 +* +* This program is free software; you can redistribute it and/or modify it +* under the terms of the GNU General Public License as published by the +* Free Software Foundation version 2. +* +* This program is distributed in the hope that it will be useful, but +* WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +* General Public License for more details. +* +*/ + +#include <swobject.h> + +// hack. remove this when we figure out our link error +#ifndef __CYGWIN__ +#include <utilstr.h> +#else +#include <string.h> +#endif + +SWORD_NAMESPACE_START + +bool SWClass::isAssignableFrom(const char *className) const { + for (int i = 0; descends[i]; i++) { +#ifndef __CYGWIN__ + if (!sword::stricmp(descends[i], className)) +#else + if (!stricmp(descends[i], className)) +#endif + return true; + } + return false; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/swunicod.cpp b/src/utilfuns/swunicod.cpp new file mode 100644 index 0000000..4b338eb --- /dev/null +++ b/src/utilfuns/swunicod.cpp @@ -0,0 +1,138 @@ +/* + * + * Copyright 1998 CrossWire Bible Society (http://www.crosswire.org) + * CrossWire Bible Society + * P. O. Box 2528 + * Tempe, AZ 85280-2528 + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + */ + +#include "swunicod.h" +SWORD_NAMESPACE_START +unsigned char* UTF32to8 (unsigned long utf32, unsigned char * utf8) { + unsigned int i; + for (i = 0; i < 6; i++) utf8[i] = 0; + + if (utf32 < 0x80) { + utf8[0] = (char)utf32; + } + else if (utf32 < 0x800) { + i = utf32 & 0x3f; + utf8[1] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x1f; + utf8[0] = 0xc0 | i; + } + else if (utf32 < 0x10000) { + i = utf32 & 0x3f; + utf8[2] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[1] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x0f; + utf8[0] = 0xe0 | i; + } + else if (utf32 < 0x200000) { + i = utf32 & 0x3f; + utf8[3] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[2] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[1] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x07; + utf8[0] = 0xf0 | i; + } + else if (utf32 < 0x4000000) { + i = utf32 & 0x3f; + utf8[4] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[3] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[2] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[1] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x03; + utf8[0] = 0xf8 | i; + } + else if (utf32 < 0x80000000) { + i = utf32 & 0x3f; + utf8[5] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[4] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[3] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[2] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x3f; + utf8[1] = 0x80 | i; + utf32 >>= 6; + + i = utf32 & 0x01; + utf8[0] = 0xfc | i; + } + return utf8; +} + +unsigned long UTF8to32 (unsigned char * utf8) { + + unsigned char i = utf8[0]; + unsigned char count; + unsigned long utf32 = 0; + + for (count = 0; i & 0x80; count++) i <<= 1; + if (!count) { + return utf8[0]; + } + else if (count == 1) { + return 0xffff; + } + else { + count--; + utf32 = i >> count; + for (i = 1; i <= count; i++) { + if ((0xc0 & utf8[i]) != 0x80) { + return 0xffff; + } + utf32 <<= 6; + utf32 |= (utf8[i] & 0x3f); + } + } + return utf32; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/swversion.cpp b/src/utilfuns/swversion.cpp new file mode 100644 index 0000000..9e61f06 --- /dev/null +++ b/src/utilfuns/swversion.cpp @@ -0,0 +1,83 @@ +#include <config.h> + +#include <swversion.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> + +SWORD_NAMESPACE_START + +SWVersion SWVersion::currentVersion( VERSION ); + +/****************************************************************************** + * SWVersion c-tor - Constructs a new SWVersion + * + * ENT: version - const version string + */ + +SWVersion::SWVersion(const char *version) { + char *buf = new char[ strlen(version) + 1 ]; + char *tok; + major = minor = minor2 = minor3 = -1; + + strcpy(buf, version); + tok = strtok(buf, "."); + if (tok) + major = atoi(tok); + tok = strtok(0, "."); + if (tok) + minor = atoi(tok); + tok = strtok(0, "."); + if (tok) + minor2 = atoi(tok); + tok = strtok(0, "."); + if (tok) + minor3 = atoi(tok); + delete [] buf; +} + + +/****************************************************************************** + * compare - compares this version to another version + * + * ENT: vi - other version with which to compare + * + * RET: = 0 if equal; + * < 0 if this version is less than other version; + * > 0 if this version is greater than other version + */ + +int SWVersion::compare(const SWVersion &vi) const { + if (major == vi.major) + if (minor == vi.minor) + if (minor2 == vi.minor2) + if (minor3 == vi.minor3) + return 0; + else return minor3 - vi.minor3; + else return minor2 - vi.minor2; + else return minor - vi.minor; + else return major - vi.major; +} + + +const char *SWVersion::getText() const { + + // 255 is safe because there is no way 4 integers (plus 3 '.'s) can have + // a string representation that will overrun this buffer + static char buf[255]; + + if (minor > -1) { + if (minor2 > -1) { + if (minor3 > -1) { + sprintf(buf, "%d.%d.%d.%d", major, minor, minor2, minor3); + } + else sprintf(buf, "%d.%d.%d", major, minor, minor2); + } + else sprintf(buf, "%d.%d", major, minor); + } + else sprintf(buf, "%d", major); + + return buf; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/url.cpp b/src/utilfuns/url.cpp new file mode 100644 index 0000000..3a50e2c --- /dev/null +++ b/src/utilfuns/url.cpp @@ -0,0 +1,269 @@ +/****************************************************************************** +* url.cpp - code for an URL parser utility class +* +* $Id: url.cpp 2013 2006-11-25 22:01:04Z dglassey $ +* +* Copyright 2003 CrossWire Bible Society (http://www.crosswire.org) +* CrossWire Bible Society +* P. O. Box 2528 +* Tempe, AZ 85280-2528 +* +* This program is free software; you can redistribute it and/or modify it +* under the terms of the GNU General Public License as published by the +* Free Software Foundation version 2. +* +* This program is distributed in the hope that it will be useful, but +* WITHOUT ANY WARRANTY; without even the implied warranty of +* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +* General Public License for more details. +* +*/ + +//Sword includes +#include <url.h> +#include <swlog.h> + +//system includes +#include <ctype.h> +#include <map> +#include <stdio.h> +#include <iostream> + +SWORD_NAMESPACE_START + +/** + * Constructors/Destructors + */ +URL::URL(const char* url) + : url(""), + protocol(""), + hostname(""), + path("") +{ + if (url && strlen(url)) { + this->url = url; + parse(); + } +} + +const char* URL::getProtocol() const { + return protocol.c_str(); +} + +const char* URL::getHostName () const { + return hostname.c_str(); +} + +const char* URL::getPath() const { + return path.c_str(); +} + +const URL::ParameterMap &URL::getParameters() const { + return parameterMap; +} + +/** + * Returns the value of an URL parameter. For the URL "http://www.crosswire.org/index.jsp?page=test&user=nobody" the value of the parameter "page" would be "test". + * If the parameter is not set an empty string is returned. + */ +const char* URL::getParameterValue(const char* name) const { + static SWBuf emptyStr(""); + + ParameterMap::const_iterator it = parameterMap.find(name); + static SWBuf retVal; + + if (it != parameterMap.end()) + retVal = it->second.c_str(); + else + retVal = emptyStr.c_str(); + + return retVal.c_str(); +} + + +/** Parse the URL. + * Parse the URL into the protocol, the hostname, the path and the paramters with their values + * + */ +void URL::parse () { + /* format example protocol://hostname/path/path/path.pl?param1=value1&param2=value2 + * we include the script name in the path, so the path would be /path/path/path.pl in this example + * & could also be & + */ + + //1. Init + const char* urlPtr = url.c_str(); + + protocol = ""; + hostname = ""; + path = ""; + parameterMap.clear(); + + // 2. Get the protocol, which is from the begining to the first :// + const char* end = strchr( urlPtr, ':' ); + if (end) { //protocol was found + protocol.append(urlPtr, end-urlPtr); + urlPtr = end + 1; + + //find the end of the protocol separator (e.g. "://") + for (; (*urlPtr == ':') || (*urlPtr == '/'); urlPtr++); + } + + //3.Get the hostname part. This is the part from pos up to the first slash + bool checkPath = true; + bool checkParams = true; + bool checkAnchor = true; + + end = strchr(urlPtr, '/'); + if (!end) { + checkPath = false; + end = strchr(urlPtr, '?'); + } + if (!end) { + checkParams = false; + end = strchr(urlPtr, '#'); + } + if (!end) { + checkAnchor = false; + end = urlPtr+strlen(urlPtr); + } + + hostname.append(urlPtr, end-urlPtr); + + urlPtr = end + ((*end)? 1 : 0); + + if (checkPath) { + end = strchr(urlPtr, '?'); + if (!end) { + checkParams = false; + end = strchr(urlPtr, '#'); + } + if (!end) { + checkAnchor = false; + end = urlPtr+strlen(urlPtr); + } + + path.append(urlPtr, end-urlPtr); + + urlPtr = end + ((*end)? 1 : 0); + } + + if (checkParams) { + //5. Fill the map with the parameters and their values + SWBuf paramName; + SWBuf paramValue; + +/* + end = strchr(urlPtr, '#'); + if (!end) { + checkAnchor = false; + end = urlPtr+strlen(urlPtr); + } +*/ + //end = (start && strchr(start, '?')) ? strchr(start, '?')+1 :0; + end = urlPtr; + while (end) { + paramName = ""; + paramValue = ""; + + //search for the equal sign to find the value part + const char* valueStart = strchr(end, '='); + if (valueStart) { + const char* valueEnd = strstr(valueStart, "&") ? strstr(valueStart, "&") : strstr(valueStart, "&"); //try to find a new paramter part + + if (valueEnd) { + paramName.append(end, valueStart-end); + paramValue.append(valueStart+1, valueEnd-(valueStart+1)); + } + else { //this is the last paramter of the URL + paramName.append(end, valueStart-end); + paramValue.append(valueStart+1); + } + + if (paramName.length() && paramValue.length()) {//insert the param into the map if it's valid + paramName = decode(paramName.c_str()); + paramValue = decode(paramValue.c_str()); + + parameterMap[ paramName ] = paramValue; + } + } + else { + break; //no valid parameter in the url + } + + const char *start = end+1; + end = strstr(start, "&") ? strstr(start, "&")+5 : (strstr(start, "&") ? strstr(start, "&")+1 : 0); //try to find a new paramter part + } + } +} + +const SWBuf URL::encode(const char *urlText) { + /*static*/ SWBuf url; + url = urlText; + + typedef std::map< unsigned char, SWBuf > DataMap; + DataMap m; + for (unsigned short int c = 32; c <= 255; ++c) { //first set all encoding chars + if ( (c >= 'A' && c <= 'Z') || (c >= 'a' && c <= 'z') || (c >= '0' && c <= '9') || strchr("-_.!~*'()", c)) { + continue; //we don't need an encoding for this char + } + + SWBuf buf; + buf.setFormatted("%%%-.2X", c); + m[c] = buf; + } + //the special encodings for certain chars + m[' '] = '+'; + + SWBuf buf; + const int length = url.length(); + for (int i = 0; i < length; i++) { //fill "buf" + const char& c = url[i]; + buf.append( ((m[c].length()) ? m[c] : SWBuf(c)) ); + } + + url = buf; + return url; +} + +const SWBuf URL::decode(const char *encoded) { + /*static*/ SWBuf text; + text = encoded; + + SWBuf decoded; + const int length = text.length(); + int i = 0; + + while (i < length) { + char a = text[i]; + + if ( a == '+' ) { //handle special cases + decoded.append(' '); + } + else if ( (a == '%') && (i+2 < length)) { //decode the %ab hex encoded char + const char b = toupper( text[i+1] ); + const char c = toupper( text[i+2] ); + + if (isxdigit(b) && isxdigit(c)) { //valid %ab part + unsigned int dec = 16 * ((b >= 'A' && b <= 'F') ? (b - 'A' + 10) : (b - '0')); //dec value of the most left digit (b) + dec += (c >= 'A' && c <= 'F') ? (c - 'A' + 10) : (c - '0'); //dec value of the right digit (c) + + decoded.append((char)dec); //append the decoded char + + i += 2; //we jump over the %ab part; we have to leave out three, but the while loop adds one, too + } + } + else { //just append the char + decoded.append(a); + } + + i++; + } + + if (decoded.length()) { + text = decoded; + } + return text; +} + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/utilstr.cpp b/src/utilfuns/utilstr.cpp new file mode 100644 index 0000000..52dc286 --- /dev/null +++ b/src/utilfuns/utilstr.cpp @@ -0,0 +1,241 @@ +#include <utilstr.h> +#include <ctype.h> +#include <string.h> + +#include <localemgr.h> + + +#ifdef _ICU_ +#include <unicode/utypes.h> +#include <unicode/ucnv.h> +#include <unicode/ustring.h> +#include <unicode/uchar.h> + +#include <unicode/unistr.h> +#include <unicode/translit.h> + +#endif + +SWORD_NAMESPACE_START + +const unsigned char SW_toupper_array[256] = + { + 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, + 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f, + 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, + 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f, + 0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27, + 0x28, 0x29, 0x2a, 0x2b, 0x2c, 0x2d, 0x2e, 0x2f, + 0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37, + 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x3f, + 0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, + 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x59, 0x5a, 0x5b, 0x5c, 0x5d, 0x5e, 0x5f, + 0x60, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47, + 0x48, 0x49, 0x4a, 0x4b, 0x4c, 0x4d, 0x4e, 0x4f, + 0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57, + 0x58, 0x59, 0x5a, 0x7b, 0x7c, 0x7d, 0x7e, 0x7f, + 0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, + 0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f, + 0x90, 0x91, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, + 0x98, 0x99, 0x9a, 0x9b, 0x9c, 0x9d, 0x9e, 0x9f, + 0xa0, 0xa1, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, + 0xa8, 0xa9, 0xaa, 0xab, 0xac, 0xad, 0xae, 0xaf, + 0xb0, 0xb1, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, 0xb7, + 0xb8, 0xb9, 0xba, 0xbb, 0xbc, 0xbd, 0xbe, 0xbf, + 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, + 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, + 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, + 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xdf, + 0xc0, 0xc1, 0xc2, 0xc3, 0xc4, 0xc5, 0xc6, 0xc7, + 0xc8, 0xc9, 0xca, 0xcb, 0xcc, 0xcd, 0xce, 0xcf, + 0xd0, 0xd1, 0xd2, 0xd3, 0xd4, 0xd5, 0xd6, 0xf7, + 0xd8, 0xd9, 0xda, 0xdb, 0xdc, 0xdd, 0xde, 0xff + }; + + + +/****************************************************************************** + * stdstr - clones a string + * + * ENT: ipstr - pointer to a string pointer to set if necessary + * istr - string to set to *ipstr + * 0 - only get + * + * RET: *ipstr + */ + +char *stdstr(char **ipstr, const char *istr, unsigned int memPadFactor) { + if (istr) { + if (*ipstr) + delete [] *ipstr; + int len = strlen(istr) + 1; + *ipstr = new char [ len * memPadFactor ]; + memcpy(*ipstr, istr, len); + } + return *ipstr; +} + + +/****************************************************************************** + * strstrip - Removes leading and trailing spaces from a string + * + * ENT: istr - string pointer to strip + * + * RET: *istr + */ + +char *strstrip(char *istr) { + char *tmp = istr; + char *rtmp; + + int len = strlen(istr); + if (len < 1) + return istr; + rtmp = istr + (len - 1); + + while ((rtmp > istr)&&((*rtmp == ' ')||(*rtmp == '\t')||(*rtmp == 10)||(*rtmp == 13))) *(rtmp--) = 0; + while ((*tmp == ' ')||(*tmp == '\t')||(*tmp == 10)||(*tmp == 13)) tmp++; + memmove(istr, tmp, (rtmp - tmp) + 1); + istr[(rtmp - tmp) + 1] = 0; + + return istr; +} + + +/****************************************************************************** + * stristr - Scans a string for the occurrence of a given substring, no case + * + * ENT: scans s1 for the first occurrence of the substring s2, ingnoring case + * + * RET: a pointer to the element in s1, where s2 begins (points to s2 in s1). + * If s2 does not occur in s1, returns null. + */ + +const char *stristr(const char *s1, const char *s2) { + int tLen = strlen(s2); + int cLen = strlen(s1); + char *target = new char [ tLen + 1 ]; + int i, j; + const char *retVal = 0; + + strcpy(target, s2); + for (i = 0; i < tLen; i++) + target[i] = SW_toupper(target[i]); + + for (i = 0; i < (cLen - tLen)+1; i++) { + if (SW_toupper(s1[i]) == (unsigned char)*target) { + for (j = 1; j < tLen; j++) { + if (SW_toupper(s1[i+j]) != (unsigned char)target[j]) + break; + } + if (j == tLen) { + retVal = s1+i; + break; + } + } + } + delete [] target; + return retVal; +} + +/****************************************************************************** + * strnicmp - compares the first n bytes of 2 strings ignoring case + * + * ENT: compares s1 to s2 comparing the first n byte ingnoring case + * + * RET: same as strcmp + */ + +int strnicmp(const char *s1, const char *s2, int len) { + int tLen = strlen(s2); + int cLen = strlen(s1); + char diff; + int i; + for (i = 0; ((i < len) && (i < tLen) && (i < cLen)); i++) { + if ((diff = SW_toupper(*s1) - SW_toupper(*s2))) + return diff; + s1++; + s2++; + } + return (i < len) ? cLen - tLen : 0; +} + +int stricmp(const char *s1, const char *s2) { +#if defined(__GNUC__) + return ::strcasecmp(s1, s2); +#else + #if defined(_WIN32_WCE) + return ::_stricmp(s1, s2); + #else + return ::stricmp(s1, s2); + #endif +#endif +} + +/****************************************************************************** + * toupperstr - converts a string to uppercase string + * + * ENT: target - string to convert + * + * RET: target + */ + +// char *toupperstr(char *buf) { +// char *ret = buf; +// +// /*if (StringHelper::getSystemStringHelper()) { +// StringHelper::getSystemStringHelper()->upperStringLatin1( ret ); +// } +// else*/ { +// while (*buf) { +// *buf++ = SW_toupper(*buf); +// } +// // } +// return ret; +// } + + +/****************************************************************************** + * toupperstr - converts a string to uppercase string + * + * ENT: target - string to convert + * + * RET: target + */ + +// char *toupperstr_utf8(char *buf, unsigned int max) { +// char *ret = buf; +// +// /* if (StringHelper::getSystemStringHelper()) { +// StringHelper::getSystemStringHelper()->upperStringUtf8( ret ); +// return ret; +// }*/ +// +// #ifndef _ICU_ +// // try to decide if it's worth trying to toupper. Do we have more +// // characters that are probably lower latin than not? +// long performOp = 0; +// for (const char *ch = buf; *ch; ch++) +// performOp += (*ch > 0) ? 1 : -1; +// +// if (performOp > 0) { +// while (*buf) +// *buf = SW_toupper(*buf++); +// } +// #else +// if (!max) +// max = strlen(ret); +// UErrorCode err = U_ZERO_ERROR; +// UConverter *conv = ucnv_open("UTF-8", &err); +// UnicodeString str(buf, -1, conv, err); +// UnicodeString ustr = str.toUpper(); +// ustr.extract(ret, max, conv, err); +// ucnv_close(conv); +// #endif +// +// return ret; +// } + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/utilxml.cpp b/src/utilfuns/utilxml.cpp new file mode 100644 index 0000000..9184102 --- /dev/null +++ b/src/utilfuns/utilxml.cpp @@ -0,0 +1,296 @@ +/****************************************************************************** + * utilxml.cpp - implementaion of utility classes to handle XML processing + * + * $Id: utilxml.cpp 2096 2007-10-07 00:40:00Z scribe $ + * + * Copyright 1998 CrossWire Bible Society (http://www.crosswire.org) + * CrossWire Bible Society + * P. O. Box 2528 + * Tempe, AZ 85280-2528 + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation version 2. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + */ + +#include <utilxml.h> +#include <ctype.h> +#include <utilstr.h> + +SWORD_NAMESPACE_START + +const char *XMLTag::nullstr = ""; +void XMLTag::parse() const { + int i; + int start; + char *name = 0; + char *value = 0; + attributes.clear(); + + if (!buf) + return; + for (i = 0; ((buf[i]) && (!isalpha(buf[i]))); i++); + for (; buf[i]; i++) { + if (strchr("\t\r\n ", buf[i])) { + // Convert newlines, carriage returns and tabs to spaces + buf[i] = ' '; + + for (; ((buf[i]) && (!isalpha(buf[i]))); i++); + if (buf[i]) { // we have an attribute name + start = i; + // Deprecated: check for following whitespacee + // Should be: for (; (buf[i] && buf[i] != '='; i++); + for (; ((buf[i]) && (!strchr(" =", buf[i]))); i++); + + if (i-start) { + if (name) + delete [] name; + name = new char [ (i-start) + 1 ]; + strncpy(name, buf+start, i-start); + name[i-start] = 0; + } + + // The following does not allow for empty attributes + //for (; ((buf[i]) && (strchr(" =\"\'", buf[i]))); i++); + + // skip space preceding the = sign + // Deprecated: this is not part of the xml spec + for (; buf[i] == ' '; i++) ; + + // skip the = sign + if (buf[i]) + i++; + + // skip space following the = sign + // Deprecated: this is not part of the xml spec + for (; buf[i] == ' '; i++) ; + + // remember and skip the quote sign + char quoteChar = buf[i]; + if (quoteChar) + i++; + + if (buf[i]) { // we have attribute value + start = i; + // Skip until matching quote character + for (; ((buf[i]) && (buf[i] != quoteChar)); i++); + + // Allow for empty quotes + //if (i-start) { + if (value) + delete [] value; + value = new char [ (i-start) + 1 ]; + if (i-start) { + strncpy(value, buf+start, i-start); + } + value[i-start] = 0; + attributes[name] = value; + //} + } + } + } + + // if there are no more characters left then quit + if (!buf[i]) + break; + + } + for (;i;i--) { + if (buf[i] == '/') + empty = true; + if (!strchr(" \t\r\n>\t", buf[i])) + break; + } + + parsed = true; + if (name) delete [] name; + if (value) delete [] value; +} + + +XMLTag::XMLTag(const char *tagString) { + + name = 0; + buf = 0; + setText(tagString); +} + +void XMLTag::setText(const char *tagString) { + parsed = false; + empty = false; + endTag = false; + + if (buf) { + delete [] buf; + buf = 0; + } + + if (!tagString) // assert tagString before proceeding + return; + + stdstr(&buf, tagString); + + int start = 0; + int i; + + // skip beginning silliness + for (i = 0; ((tagString[i]) && (!isalpha(tagString[i]))); i++) { + if (tagString[i] == '/') + endTag = true; + } + start = i; + for (; ((tagString[i]) && (!strchr("\t\r\n />", tagString[i]))); i++); + if (i-start) { + if (name) + delete [] name; + name = new char [ (i-start) + 1 ]; + strncpy(name, tagString+start, i-start); + name[i-start] = 0; + if (tagString[i] == '/') + empty = true; + } +} + +XMLTag::~XMLTag() { + if (buf) + delete [] buf; + if (name) + delete [] name; +} + +const StringList XMLTag::getAttributeNames() const { + StringList retVal; + + if (!parsed) + parse(); + + for (StringPairMap::const_iterator it = attributes.begin(); it != attributes.end(); it++) + retVal.push_back(it->first.c_str()); + + return retVal; +} + + +const char *XMLTag::getPart(const char *buf, int partNum, char partSplit) const { + for (; (buf && partNum); partNum--) { + buf = strchr(buf, partSplit); + if (buf) + buf++; + } + if (buf) { + const char *end = strchr(buf, partSplit); + junkBuf = buf; + if (end) + junkBuf.setSize(end - buf); + return junkBuf.c_str(); + } + return 0; +} + + +int XMLTag::getAttributePartCount(const char *attribName, char partSplit) const { + int count; + const char *buf = getAttribute(attribName); + for (count = 0; buf; count++) { + buf = strchr(buf, partSplit); + if (buf) + buf++; + } + return count; +} + + +const char *XMLTag::getAttribute(const char *attribName, int partNum, char partSplit) const { + + if (!parsed) + parse(); + + StringPairMap::const_iterator it = attributes.find(attribName); + + const char *retVal = 0; + if (it != attributes.end()) + retVal = it->second.c_str(); + + if ((retVal) && (partNum > -1)) + retVal = getPart(retVal, partNum, partSplit); + + return retVal; +} + + +const char *XMLTag::setAttribute(const char *attribName, const char *attribValue, int partNum, char partSplit) { + if (!parsed) + parse(); + + SWBuf newVal = ""; + // set part of an attribute + if (partNum > -1) { + const char *wholeAttr = getAttribute(attribName); + int attrCount = getAttributePartCount(attribName, partSplit); + for (int i = 0; i < attrCount; i++) { + if (i == partNum) { + if (attribValue) { + newVal += attribValue; + newVal += partSplit; + } + else { + // discard this part per null attribValue + } + } + else { + newVal += getPart(wholeAttr, i, partSplit); + newVal += partSplit; + } + } + if (newVal.length()) newVal--; // discard the last partSplit + attribValue = (!attribValue && !newVal.length()) ? 0 : newVal.c_str(); + } + + // perform the actual set + if (attribValue) + attributes[attribName] = attribValue; + else attributes.erase(attribName); + + return attribValue; +} + +const char *XMLTag::toString() const { + SWBuf tag = "<"; + if (!parsed) + parse(); + + if (isEndTag()) + tag.append('/'); + + tag.append(getName()); + for (StringPairMap::iterator it = attributes.begin(); it != attributes.end(); it++) { + //tag.appendFormatted(" %s=\"%s\"", it->first.c_str(), it->second.c_str()); + tag.append(' '); + tag.append(it->first.c_str()); + tag.append((strchr(it->second.c_str(), '\"')) ? "=\'" : "=\""); + tag.append(it->second.c_str()); + tag.append((strchr(it->second.c_str(), '\"'))? '\'' : '\"'); + } + + if (isEmpty()) + tag.append('/'); + + tag.append('>'); + + + if (buf) + delete [] buf; + buf = new char [ tag.length() + 1 ]; + strcpy(buf, tag.c_str()); + + return buf; +} + + +SWORD_NAMESPACE_END diff --git a/src/utilfuns/win32/dirent.cpp b/src/utilfuns/win32/dirent.cpp new file mode 100644 index 0000000..150bf10 --- /dev/null +++ b/src/utilfuns/win32/dirent.cpp @@ -0,0 +1,131 @@ +/* + + Implementation of POSIX directory browsing functions and types for Win32. + + Kevlin Henney (mailto:kevlin@acm.org), March 1997. + + Copyright Kevlin Henney, 1997. All rights reserved. + + Permission to use, copy, modify, and distribute this software and its + documentation for any purpose is hereby granted without fee, provided + that this copyright and permissions notice appear in all copies and + derivatives, and that no charge may be made for the software and its + documentation except to cover cost of distribution. + + This software is supplied "as is" without express or implied warranty. + + But that said, if there are any problems please get in touch. + +*/ + +#include <dirent.h> +#include <errno.h> +#include <io.h> +#include <stdlib.h> +#include <string.h> + +struct DIR +{ + long handle; /* -1 for failed rewind */ + struct _finddata_t info; + struct dirent result; /* d_name null iff first time */ + char *name; /* NTBS */ +}; + +DIR *opendir(const char *name) +{ + DIR *dir = 0; + + if(name && name[0]) + { + size_t base_length = strlen(name); + const char *all = /* the root directory is a special case... */ + strchr("/\\", name[base_length - 1]) ? "*" : "/*"; + + if((dir = (DIR *) malloc(sizeof *dir)) != 0 && + (dir->name = (char *) malloc(base_length + strlen(all) + 1)) != 0) + { + strcat(strcpy(dir->name, name), all); + + if((dir->handle = _findfirst(dir->name, &dir->info)) != -1) + { + dir->result.d_name = 0; + } + else /* rollback */ + { + free(dir->name); + free(dir); + dir = 0; + } + } + else /* rollback */ + { + free(dir); + dir = 0; + errno = ENOMEM; + } + } + else + { + errno = EINVAL; + } + + return dir; +} + +int closedir(DIR *dir) +{ + int result = -1; + + if(dir) + { + if(dir->handle != -1) + { + result = _findclose(dir->handle); + } + + free(dir->name); + free(dir); + } + + if(result == -1) /* map all errors to EBADF */ + { + errno = EBADF; + } + + return result; +} + +struct dirent *readdir(DIR *dir) +{ + struct dirent *result = 0; + + if(dir && dir->handle != -1) + { + if(!dir->result.d_name || _findnext(dir->handle, &dir->info) != -1) + { + result = &dir->result; + result->d_name = dir->info.name; + } + } + else + { + errno = EBADF; + } + + return result; +} + +void rewinddir(DIR *dir) +{ + if(dir && dir->handle != -1) + { + _findclose(dir->handle); + dir->handle = _findfirst(dir->name, &dir->info); + dir->result.d_name = 0; + } + else + { + errno = EBADF; + } +} diff --git a/src/utilfuns/win32/dirent.h b/src/utilfuns/win32/dirent.h new file mode 100644 index 0000000..ba22833 --- /dev/null +++ b/src/utilfuns/win32/dirent.h @@ -0,0 +1,32 @@ +/* + + Declaration of POSIX directory browsing functions and types for Win32. + + Kevlin Henney (mailto:kevlin@acm.org), March 1997. + + Copyright Kevlin Henney, 1997. All rights reserved. + + Permission to use, copy, modify, and distribute this software and its + documentation for any purpose is hereby granted without fee, provided + that this copyright and permissions notice appear in all copies and + derivatives, and that no charge may be made for the software and its + documentation except to cover cost of distribution. + +*/ + +#ifndef DIRENT_INCLUDED +#define DIRENT_INCLUDED + +typedef struct DIR DIR; + +struct dirent +{ + char *d_name; +}; + +DIR *opendir(const char *); +int closedir(DIR *); +struct dirent *readdir(DIR *); +void rewinddir(DIR *); + +#endif diff --git a/src/utilfuns/zlib/adler32.c b/src/utilfuns/zlib/adler32.c new file mode 100644 index 0000000..c786e9a --- /dev/null +++ b/src/utilfuns/zlib/adler32.c @@ -0,0 +1,48 @@ +/* adler32.c -- compute the Adler-32 checksum of a data stream + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id: adler32.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "zlib.h" + +#define BASE 65521L /* largest prime smaller than 65536 */ +#define NMAX 5552 +/* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1 */ + +#define DO1(buf,i) {s1 += buf[i]; s2 += s1;} +#define DO2(buf,i) DO1(buf,i); DO1(buf,i+1); +#define DO4(buf,i) DO2(buf,i); DO2(buf,i+2); +#define DO8(buf,i) DO4(buf,i); DO4(buf,i+4); +#define DO16(buf) DO8(buf,0); DO8(buf,8); + +/* ========================================================================= */ +uLong ZEXPORT adler32(adler, buf, len) + uLong adler; + const Bytef *buf; + uInt len; +{ + unsigned long s1 = adler & 0xffff; + unsigned long s2 = (adler >> 16) & 0xffff; + int k; + + if (buf == Z_NULL) return 1L; + + while (len > 0) { + k = len < NMAX ? len : NMAX; + len -= k; + while (k >= 16) { + DO16(buf); + buf += 16; + k -= 16; + } + if (k != 0) do { + s1 += *buf++; + s2 += s1; + } while (--k); + s1 %= BASE; + s2 %= BASE; + } + return (s2 << 16) | s1; +} diff --git a/src/utilfuns/zlib/compress.c b/src/utilfuns/zlib/compress.c new file mode 100644 index 0000000..2b1f048 --- /dev/null +++ b/src/utilfuns/zlib/compress.c @@ -0,0 +1,68 @@ +/* compress.c -- compress a memory buffer + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id: compress.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "zlib.h" + +/* =========================================================================== + Compresses the source buffer into the destination buffer. The level + parameter has the same meaning as in deflateInit. sourceLen is the byte + length of the source buffer. Upon entry, destLen is the total size of the + destination buffer, which must be at least 0.1% larger than sourceLen plus + 12 bytes. Upon exit, destLen is the actual size of the compressed buffer. + + compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough + memory, Z_BUF_ERROR if there was not enough room in the output buffer, + Z_STREAM_ERROR if the level parameter is invalid. +*/ +int ZEXPORT compress2 (dest, destLen, source, sourceLen, level) + Bytef *dest; + uLongf *destLen; + const Bytef *source; + uLong sourceLen; + int level; +{ + z_stream stream; + int err; + + stream.next_in = (Bytef*)source; + stream.avail_in = (uInt)sourceLen; +#ifdef MAXSEG_64K + /* Check for source > 64K on 16-bit machine: */ + if ((uLong)stream.avail_in != sourceLen) return Z_BUF_ERROR; +#endif + stream.next_out = dest; + stream.avail_out = (uInt)*destLen; + if ((uLong)stream.avail_out != *destLen) return Z_BUF_ERROR; + + stream.zalloc = (alloc_func)0; + stream.zfree = (free_func)0; + stream.opaque = (voidpf)0; + + err = deflateInit(&stream, level); + if (err != Z_OK) return err; + + err = deflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) { + deflateEnd(&stream); + return err == Z_OK ? Z_BUF_ERROR : err; + } + *destLen = stream.total_out; + + err = deflateEnd(&stream); + return err; +} + +/* =========================================================================== + */ +int ZEXPORT compress (dest, destLen, source, sourceLen) + Bytef *dest; + uLongf *destLen; + const Bytef *source; + uLong sourceLen; +{ + return compress2(dest, destLen, source, sourceLen, Z_DEFAULT_COMPRESSION); +} diff --git a/src/utilfuns/zlib/crc32.c b/src/utilfuns/zlib/crc32.c new file mode 100644 index 0000000..e44440c --- /dev/null +++ b/src/utilfuns/zlib/crc32.c @@ -0,0 +1,162 @@ +/* crc32.c -- compute the CRC-32 of a data stream + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id: crc32.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "zlib.h" + +#define local static + +#ifdef DYNAMIC_CRC_TABLE + +local int crc_table_empty = 1; +local uLongf crc_table[256]; +local void make_crc_table OF((void)); + +/* + Generate a table for a byte-wise 32-bit CRC calculation on the polynomial: + x^32+x^26+x^23+x^22+x^16+x^12+x^11+x^10+x^8+x^7+x^5+x^4+x^2+x+1. + + Polynomials over GF(2) are represented in binary, one bit per coefficient, + with the lowest powers in the most significant bit. Then adding polynomials + is just exclusive-or, and multiplying a polynomial by x is a right shift by + one. If we call the above polynomial p, and represent a byte as the + polynomial q, also with the lowest power in the most significant bit (so the + byte 0xb1 is the polynomial x^7+x^3+x+1), then the CRC is (q*x^32) mod p, + where a mod b means the remainder after dividing a by b. + + This calculation is done using the shift-register method of multiplying and + taking the remainder. The register is initialized to zero, and for each + incoming bit, x^32 is added mod p to the register if the bit is a one (where + x^32 mod p is p+x^32 = x^26+...+1), and the register is multiplied mod p by + x (which is shifting right by one and adding x^32 mod p if the bit shifted + out is a one). We start with the highest power (least significant bit) of + q and repeat for all eight bits of q. + + The table is simply the CRC of all possible eight bit values. This is all + the information needed to generate CRC's on data a byte at a time for all + combinations of CRC register values and incoming bytes. +*/ +local void make_crc_table() +{ + uLong c; + int n, k; + uLong poly; /* polynomial exclusive-or pattern */ + /* terms of polynomial defining this crc (except x^32): */ + static const Byte p[] = {0,1,2,4,5,7,8,10,11,12,16,22,23,26}; + + /* make exclusive-or pattern from polynomial (0xedb88320L) */ + poly = 0L; + for (n = 0; n < sizeof(p)/sizeof(Byte); n++) + poly |= 1L << (31 - p[n]); + + for (n = 0; n < 256; n++) + { + c = (uLong)n; + for (k = 0; k < 8; k++) + c = c & 1 ? poly ^ (c >> 1) : c >> 1; + crc_table[n] = c; + } + crc_table_empty = 0; +} +#else +/* ======================================================================== + * Table of CRC-32's of all single-byte values (made by make_crc_table) + */ +local const uLongf crc_table[256] = { + 0x00000000L, 0x77073096L, 0xee0e612cL, 0x990951baL, 0x076dc419L, + 0x706af48fL, 0xe963a535L, 0x9e6495a3L, 0x0edb8832L, 0x79dcb8a4L, + 0xe0d5e91eL, 0x97d2d988L, 0x09b64c2bL, 0x7eb17cbdL, 0xe7b82d07L, + 0x90bf1d91L, 0x1db71064L, 0x6ab020f2L, 0xf3b97148L, 0x84be41deL, + 0x1adad47dL, 0x6ddde4ebL, 0xf4d4b551L, 0x83d385c7L, 0x136c9856L, + 0x646ba8c0L, 0xfd62f97aL, 0x8a65c9ecL, 0x14015c4fL, 0x63066cd9L, + 0xfa0f3d63L, 0x8d080df5L, 0x3b6e20c8L, 0x4c69105eL, 0xd56041e4L, + 0xa2677172L, 0x3c03e4d1L, 0x4b04d447L, 0xd20d85fdL, 0xa50ab56bL, + 0x35b5a8faL, 0x42b2986cL, 0xdbbbc9d6L, 0xacbcf940L, 0x32d86ce3L, + 0x45df5c75L, 0xdcd60dcfL, 0xabd13d59L, 0x26d930acL, 0x51de003aL, + 0xc8d75180L, 0xbfd06116L, 0x21b4f4b5L, 0x56b3c423L, 0xcfba9599L, + 0xb8bda50fL, 0x2802b89eL, 0x5f058808L, 0xc60cd9b2L, 0xb10be924L, + 0x2f6f7c87L, 0x58684c11L, 0xc1611dabL, 0xb6662d3dL, 0x76dc4190L, + 0x01db7106L, 0x98d220bcL, 0xefd5102aL, 0x71b18589L, 0x06b6b51fL, + 0x9fbfe4a5L, 0xe8b8d433L, 0x7807c9a2L, 0x0f00f934L, 0x9609a88eL, + 0xe10e9818L, 0x7f6a0dbbL, 0x086d3d2dL, 0x91646c97L, 0xe6635c01L, + 0x6b6b51f4L, 0x1c6c6162L, 0x856530d8L, 0xf262004eL, 0x6c0695edL, + 0x1b01a57bL, 0x8208f4c1L, 0xf50fc457L, 0x65b0d9c6L, 0x12b7e950L, + 0x8bbeb8eaL, 0xfcb9887cL, 0x62dd1ddfL, 0x15da2d49L, 0x8cd37cf3L, + 0xfbd44c65L, 0x4db26158L, 0x3ab551ceL, 0xa3bc0074L, 0xd4bb30e2L, + 0x4adfa541L, 0x3dd895d7L, 0xa4d1c46dL, 0xd3d6f4fbL, 0x4369e96aL, + 0x346ed9fcL, 0xad678846L, 0xda60b8d0L, 0x44042d73L, 0x33031de5L, + 0xaa0a4c5fL, 0xdd0d7cc9L, 0x5005713cL, 0x270241aaL, 0xbe0b1010L, + 0xc90c2086L, 0x5768b525L, 0x206f85b3L, 0xb966d409L, 0xce61e49fL, + 0x5edef90eL, 0x29d9c998L, 0xb0d09822L, 0xc7d7a8b4L, 0x59b33d17L, + 0x2eb40d81L, 0xb7bd5c3bL, 0xc0ba6cadL, 0xedb88320L, 0x9abfb3b6L, + 0x03b6e20cL, 0x74b1d29aL, 0xead54739L, 0x9dd277afL, 0x04db2615L, + 0x73dc1683L, 0xe3630b12L, 0x94643b84L, 0x0d6d6a3eL, 0x7a6a5aa8L, + 0xe40ecf0bL, 0x9309ff9dL, 0x0a00ae27L, 0x7d079eb1L, 0xf00f9344L, + 0x8708a3d2L, 0x1e01f268L, 0x6906c2feL, 0xf762575dL, 0x806567cbL, + 0x196c3671L, 0x6e6b06e7L, 0xfed41b76L, 0x89d32be0L, 0x10da7a5aL, + 0x67dd4accL, 0xf9b9df6fL, 0x8ebeeff9L, 0x17b7be43L, 0x60b08ed5L, + 0xd6d6a3e8L, 0xa1d1937eL, 0x38d8c2c4L, 0x4fdff252L, 0xd1bb67f1L, + 0xa6bc5767L, 0x3fb506ddL, 0x48b2364bL, 0xd80d2bdaL, 0xaf0a1b4cL, + 0x36034af6L, 0x41047a60L, 0xdf60efc3L, 0xa867df55L, 0x316e8eefL, + 0x4669be79L, 0xcb61b38cL, 0xbc66831aL, 0x256fd2a0L, 0x5268e236L, + 0xcc0c7795L, 0xbb0b4703L, 0x220216b9L, 0x5505262fL, 0xc5ba3bbeL, + 0xb2bd0b28L, 0x2bb45a92L, 0x5cb36a04L, 0xc2d7ffa7L, 0xb5d0cf31L, + 0x2cd99e8bL, 0x5bdeae1dL, 0x9b64c2b0L, 0xec63f226L, 0x756aa39cL, + 0x026d930aL, 0x9c0906a9L, 0xeb0e363fL, 0x72076785L, 0x05005713L, + 0x95bf4a82L, 0xe2b87a14L, 0x7bb12baeL, 0x0cb61b38L, 0x92d28e9bL, + 0xe5d5be0dL, 0x7cdcefb7L, 0x0bdbdf21L, 0x86d3d2d4L, 0xf1d4e242L, + 0x68ddb3f8L, 0x1fda836eL, 0x81be16cdL, 0xf6b9265bL, 0x6fb077e1L, + 0x18b74777L, 0x88085ae6L, 0xff0f6a70L, 0x66063bcaL, 0x11010b5cL, + 0x8f659effL, 0xf862ae69L, 0x616bffd3L, 0x166ccf45L, 0xa00ae278L, + 0xd70dd2eeL, 0x4e048354L, 0x3903b3c2L, 0xa7672661L, 0xd06016f7L, + 0x4969474dL, 0x3e6e77dbL, 0xaed16a4aL, 0xd9d65adcL, 0x40df0b66L, + 0x37d83bf0L, 0xa9bcae53L, 0xdebb9ec5L, 0x47b2cf7fL, 0x30b5ffe9L, + 0xbdbdf21cL, 0xcabac28aL, 0x53b39330L, 0x24b4a3a6L, 0xbad03605L, + 0xcdd70693L, 0x54de5729L, 0x23d967bfL, 0xb3667a2eL, 0xc4614ab8L, + 0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL, + 0x2d02ef8dL +}; +#endif + +/* ========================================================================= + * This function can be used by asm versions of crc32() + */ +const uLongf * ZEXPORT get_crc_table() +{ +#ifdef DYNAMIC_CRC_TABLE + if (crc_table_empty) make_crc_table(); +#endif + return (const uLongf *)crc_table; +} + +/* ========================================================================= */ +#define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8); +#define DO2(buf) DO1(buf); DO1(buf); +#define DO4(buf) DO2(buf); DO2(buf); +#define DO8(buf) DO4(buf); DO4(buf); + +/* ========================================================================= */ +uLong ZEXPORT crc32(crc, buf, len) + uLong crc; + const Bytef *buf; + uInt len; +{ + if (buf == Z_NULL) return 0L; +#ifdef DYNAMIC_CRC_TABLE + if (crc_table_empty) + make_crc_table(); +#endif + crc = crc ^ 0xffffffffL; + while (len >= 8) + { + DO8(buf); + len -= 8; + } + if (len) do { + DO1(buf); + } while (--len); + return crc ^ 0xffffffffL; +} diff --git a/src/utilfuns/zlib/deflate.c b/src/utilfuns/zlib/deflate.c new file mode 100644 index 0000000..43ea588 --- /dev/null +++ b/src/utilfuns/zlib/deflate.c @@ -0,0 +1,1350 @@ +/* deflate.c -- compress data using the deflation algorithm + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* + * ALGORITHM + * + * The "deflation" process depends on being able to identify portions + * of the input text which are identical to earlier input (within a + * sliding window trailing behind the input currently being processed). + * + * The most straightforward technique turns out to be the fastest for + * most input files: try all possible matches and select the longest. + * The key feature of this algorithm is that insertions into the string + * dictionary are very simple and thus fast, and deletions are avoided + * completely. Insertions are performed at each input character, whereas + * string matches are performed only when the previous match ends. So it + * is preferable to spend more time in matches to allow very fast string + * insertions and avoid deletions. The matching algorithm for small + * strings is inspired from that of Rabin & Karp. A brute force approach + * is used to find longer strings when a small match has been found. + * A similar algorithm is used in comic (by Jan-Mark Wams) and freeze + * (by Leonid Broukhis). + * A previous version of this file used a more sophisticated algorithm + * (by Fiala and Greene) which is guaranteed to run in linear amortized + * time, but has a larger average cost, uses more memory and is patented. + * However the F&G algorithm may be faster for some highly redundant + * files if the parameter max_chain_length (described below) is too large. + * + * ACKNOWLEDGEMENTS + * + * The idea of lazy evaluation of matches is due to Jan-Mark Wams, and + * I found it in 'freeze' written by Leonid Broukhis. + * Thanks to many people for bug reports and testing. + * + * REFERENCES + * + * Deutsch, L.P.,"DEFLATE Compressed Data Format Specification". + * Available in ftp://ds.internic.net/rfc/rfc1951.txt + * + * A description of the Rabin and Karp algorithm is given in the book + * "Algorithms" by R. Sedgewick, Addison-Wesley, p252. + * + * Fiala,E.R., and Greene,D.H. + * Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595 + * + */ + +/* @(#) $Id: deflate.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "deflate.h" + +const char deflate_copyright[] = + " deflate 1.1.4 Copyright 1995-2002 Jean-loup Gailly "; +/* + If you use the zlib library in a product, an acknowledgment is welcome + in the documentation of your product. If for some reason you cannot + include such an acknowledgment, I would appreciate that you keep this + copyright string in the executable of your product. + */ + +/* =========================================================================== + * Function prototypes. + */ +typedef enum { + need_more, /* block not completed, need more input or more output */ + block_done, /* block flush performed */ + finish_started, /* finish started, need only more output at next deflate */ + finish_done /* finish done, accept no more input or output */ +} block_state; + +typedef block_state (*compress_func) OF((deflate_state *s, int flush)); +/* Compression function. Returns the block state after the call. */ + +local void fill_window OF((deflate_state *s)); +local block_state deflate_stored OF((deflate_state *s, int flush)); +local block_state deflate_fast OF((deflate_state *s, int flush)); +local block_state deflate_slow OF((deflate_state *s, int flush)); +local void lm_init OF((deflate_state *s)); +local void putShortMSB OF((deflate_state *s, uInt b)); +local void flush_pending OF((z_streamp strm)); +local int read_buf OF((z_streamp strm, Bytef *buf, unsigned size)); +#ifdef ASMV + void match_init OF((void)); /* asm code initialization */ + uInt longest_match OF((deflate_state *s, IPos cur_match)); +#else +local uInt longest_match OF((deflate_state *s, IPos cur_match)); +#endif + +#ifdef DEBUG +local void check_match OF((deflate_state *s, IPos start, IPos match, + int length)); +#endif + +/* =========================================================================== + * Local data + */ + +#define NIL 0 +/* Tail of hash chains */ + +#ifndef TOO_FAR +# define TOO_FAR 4096 +#endif +/* Matches of length 3 are discarded if their distance exceeds TOO_FAR */ + +#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1) +/* Minimum amount of lookahead, except at the end of the input file. + * See deflate.c for comments about the MIN_MATCH+1. + */ + +/* Values for max_lazy_match, good_match and max_chain_length, depending on + * the desired pack level (0..9). The values given below have been tuned to + * exclude worst case performance for pathological files. Better values may be + * found for specific files. + */ +typedef struct config_s { + ush good_length; /* reduce lazy search above this match length */ + ush max_lazy; /* do not perform lazy search above this match length */ + ush nice_length; /* quit search above this match length */ + ush max_chain; + compress_func func; +} config; + +local const config configuration_table[10] = { +/* good lazy nice chain */ +/* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */ +/* 1 */ {4, 4, 8, 4, deflate_fast}, /* maximum speed, no lazy matches */ +/* 2 */ {4, 5, 16, 8, deflate_fast}, +/* 3 */ {4, 6, 32, 32, deflate_fast}, + +/* 4 */ {4, 4, 16, 16, deflate_slow}, /* lazy matches */ +/* 5 */ {8, 16, 32, 32, deflate_slow}, +/* 6 */ {8, 16, 128, 128, deflate_slow}, +/* 7 */ {8, 32, 128, 256, deflate_slow}, +/* 8 */ {32, 128, 258, 1024, deflate_slow}, +/* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* maximum compression */ + +/* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4 + * For deflate_fast() (levels <= 3) good is ignored and lazy has a different + * meaning. + */ + +#define EQUAL 0 +/* result of memcmp for equal strings */ + +struct static_tree_desc_s {int dummy;}; /* for buggy compilers */ + +/* =========================================================================== + * Update a hash value with the given input byte + * IN assertion: all calls to to UPDATE_HASH are made with consecutive + * input characters, so that a running hash key can be computed from the + * previous key instead of complete recalculation each time. + */ +#define UPDATE_HASH(s,h,c) (h = (((h)<<s->hash_shift) ^ (c)) & s->hash_mask) + + +/* =========================================================================== + * Insert string str in the dictionary and set match_head to the previous head + * of the hash chain (the most recent string with same hash key). Return + * the previous length of the hash chain. + * If this file is compiled with -DFASTEST, the compression level is forced + * to 1, and no hash chains are maintained. + * IN assertion: all calls to to INSERT_STRING are made with consecutive + * input characters and the first MIN_MATCH bytes of str are valid + * (except for the last MIN_MATCH-1 bytes of the input file). + */ +#ifdef FASTEST +#define INSERT_STRING(s, str, match_head) \ + (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ + match_head = s->head[s->ins_h], \ + s->head[s->ins_h] = (Pos)(str)) +#else +#define INSERT_STRING(s, str, match_head) \ + (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ + s->prev[(str) & s->w_mask] = match_head = s->head[s->ins_h], \ + s->head[s->ins_h] = (Pos)(str)) +#endif + +/* =========================================================================== + * Initialize the hash table (avoiding 64K overflow for 16 bit systems). + * prev[] will be initialized on the fly. + */ +#define CLEAR_HASH(s) \ + s->head[s->hash_size-1] = NIL; \ + zmemzero((Bytef *)s->head, (unsigned)(s->hash_size-1)*sizeof(*s->head)); + +/* ========================================================================= */ +int ZEXPORT deflateInit_(strm, level, version, stream_size) + z_streamp strm; + int level; + const char *version; + int stream_size; +{ + return deflateInit2_(strm, level, Z_DEFLATED, MAX_WBITS, DEF_MEM_LEVEL, + Z_DEFAULT_STRATEGY, version, stream_size); + /* To do: ignore strm->next_in if we use it as window */ +} + +/* ========================================================================= */ +int ZEXPORT deflateInit2_(strm, level, method, windowBits, memLevel, strategy, + version, stream_size) + z_streamp strm; + int level; + int method; + int windowBits; + int memLevel; + int strategy; + const char *version; + int stream_size; +{ + deflate_state *s; + int noheader = 0; + static const char* my_version = ZLIB_VERSION; + + ushf *overlay; + /* We overlay pending_buf and d_buf+l_buf. This works since the average + * output size for (length,distance) codes is <= 24 bits. + */ + + if (version == Z_NULL || version[0] != my_version[0] || + stream_size != sizeof(z_stream)) { + return Z_VERSION_ERROR; + } + if (strm == Z_NULL) return Z_STREAM_ERROR; + + strm->msg = Z_NULL; + if (strm->zalloc == Z_NULL) { + strm->zalloc = zcalloc; + strm->opaque = (voidpf)0; + } + if (strm->zfree == Z_NULL) strm->zfree = zcfree; + + if (level == Z_DEFAULT_COMPRESSION) level = 6; +#ifdef FASTEST + level = 1; +#endif + + if (windowBits < 0) { /* undocumented feature: suppress zlib header */ + noheader = 1; + windowBits = -windowBits; + } + if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED || + windowBits < 9 || windowBits > 15 || level < 0 || level > 9 || + strategy < 0 || strategy > Z_HUFFMAN_ONLY) { + return Z_STREAM_ERROR; + } + s = (deflate_state *) ZALLOC(strm, 1, sizeof(deflate_state)); + if (s == Z_NULL) return Z_MEM_ERROR; + strm->state = (struct internal_state FAR *)s; + s->strm = strm; + + s->noheader = noheader; + s->w_bits = windowBits; + s->w_size = 1 << s->w_bits; + s->w_mask = s->w_size - 1; + + s->hash_bits = memLevel + 7; + s->hash_size = 1 << s->hash_bits; + s->hash_mask = s->hash_size - 1; + s->hash_shift = ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH); + + s->window = (Bytef *) ZALLOC(strm, s->w_size, 2*sizeof(Byte)); + s->prev = (Posf *) ZALLOC(strm, s->w_size, sizeof(Pos)); + s->head = (Posf *) ZALLOC(strm, s->hash_size, sizeof(Pos)); + + s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */ + + overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2); + s->pending_buf = (uchf *) overlay; + s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L); + + if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL || + s->pending_buf == Z_NULL) { + strm->msg = (char*)ERR_MSG(Z_MEM_ERROR); + deflateEnd (strm); + return Z_MEM_ERROR; + } + s->d_buf = overlay + s->lit_bufsize/sizeof(ush); + s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize; + + s->level = level; + s->strategy = strategy; + s->method = (Byte)method; + + return deflateReset(strm); +} + +/* ========================================================================= */ +int ZEXPORT deflateSetDictionary (strm, dictionary, dictLength) + z_streamp strm; + const Bytef *dictionary; + uInt dictLength; +{ + deflate_state *s; + uInt length = dictLength; + uInt n; + IPos hash_head = 0; + + if (strm == Z_NULL || strm->state == Z_NULL || dictionary == Z_NULL || + strm->state->status != INIT_STATE) return Z_STREAM_ERROR; + + s = strm->state; + strm->adler = adler32(strm->adler, dictionary, dictLength); + + if (length < MIN_MATCH) return Z_OK; + if (length > MAX_DIST(s)) { + length = MAX_DIST(s); +#ifndef USE_DICT_HEAD + dictionary += dictLength - length; /* use the tail of the dictionary */ +#endif + } + zmemcpy(s->window, dictionary, length); + s->strstart = length; + s->block_start = (long)length; + + /* Insert all strings in the hash table (except for the last two bytes). + * s->lookahead stays null, so s->ins_h will be recomputed at the next + * call of fill_window. + */ + s->ins_h = s->window[0]; + UPDATE_HASH(s, s->ins_h, s->window[1]); + for (n = 0; n <= length - MIN_MATCH; n++) { + INSERT_STRING(s, n, hash_head); + } + if (hash_head) hash_head = 0; /* to make compiler happy */ + return Z_OK; +} + +/* ========================================================================= */ +int ZEXPORT deflateReset (strm) + z_streamp strm; +{ + deflate_state *s; + + if (strm == Z_NULL || strm->state == Z_NULL || + strm->zalloc == Z_NULL || strm->zfree == Z_NULL) return Z_STREAM_ERROR; + + strm->total_in = strm->total_out = 0; + strm->msg = Z_NULL; /* use zfree if we ever allocate msg dynamically */ + strm->data_type = Z_UNKNOWN; + + s = (deflate_state *)strm->state; + s->pending = 0; + s->pending_out = s->pending_buf; + + if (s->noheader < 0) { + s->noheader = 0; /* was set to -1 by deflate(..., Z_FINISH); */ + } + s->status = s->noheader ? BUSY_STATE : INIT_STATE; + strm->adler = 1; + s->last_flush = Z_NO_FLUSH; + + _tr_init(s); + lm_init(s); + + return Z_OK; +} + +/* ========================================================================= */ +int ZEXPORT deflateParams(strm, level, strategy) + z_streamp strm; + int level; + int strategy; +{ + deflate_state *s; + compress_func func; + int err = Z_OK; + + if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR; + s = strm->state; + + if (level == Z_DEFAULT_COMPRESSION) { + level = 6; + } + if (level < 0 || level > 9 || strategy < 0 || strategy > Z_HUFFMAN_ONLY) { + return Z_STREAM_ERROR; + } + func = configuration_table[s->level].func; + + if (func != configuration_table[level].func && strm->total_in != 0) { + /* Flush the last buffer: */ + err = deflate(strm, Z_PARTIAL_FLUSH); + } + if (s->level != level) { + s->level = level; + s->max_lazy_match = configuration_table[level].max_lazy; + s->good_match = configuration_table[level].good_length; + s->nice_match = configuration_table[level].nice_length; + s->max_chain_length = configuration_table[level].max_chain; + } + s->strategy = strategy; + return err; +} + +/* ========================================================================= + * Put a short in the pending buffer. The 16-bit value is put in MSB order. + * IN assertion: the stream state is correct and there is enough room in + * pending_buf. + */ +local void putShortMSB (s, b) + deflate_state *s; + uInt b; +{ + put_byte(s, (Byte)(b >> 8)); + put_byte(s, (Byte)(b & 0xff)); +} + +/* ========================================================================= + * Flush as much pending output as possible. All deflate() output goes + * through this function so some applications may wish to modify it + * to avoid allocating a large strm->next_out buffer and copying into it. + * (See also read_buf()). + */ +local void flush_pending(strm) + z_streamp strm; +{ + unsigned len = strm->state->pending; + + if (len > strm->avail_out) len = strm->avail_out; + if (len == 0) return; + + zmemcpy(strm->next_out, strm->state->pending_out, len); + strm->next_out += len; + strm->state->pending_out += len; + strm->total_out += len; + strm->avail_out -= len; + strm->state->pending -= len; + if (strm->state->pending == 0) { + strm->state->pending_out = strm->state->pending_buf; + } +} + +/* ========================================================================= */ +int ZEXPORT deflate (strm, flush) + z_streamp strm; + int flush; +{ + int old_flush; /* value of flush param for previous deflate call */ + deflate_state *s; + + if (strm == Z_NULL || strm->state == Z_NULL || + flush > Z_FINISH || flush < 0) { + return Z_STREAM_ERROR; + } + s = strm->state; + + if (strm->next_out == Z_NULL || + (strm->next_in == Z_NULL && strm->avail_in != 0) || + (s->status == FINISH_STATE && flush != Z_FINISH)) { + ERR_RETURN(strm, Z_STREAM_ERROR); + } + if (strm->avail_out == 0) ERR_RETURN(strm, Z_BUF_ERROR); + + s->strm = strm; /* just in case */ + old_flush = s->last_flush; + s->last_flush = flush; + + /* Write the zlib header */ + if (s->status == INIT_STATE) { + + uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8; + uInt level_flags = (s->level-1) >> 1; + + if (level_flags > 3) level_flags = 3; + header |= (level_flags << 6); + if (s->strstart != 0) header |= PRESET_DICT; + header += 31 - (header % 31); + + s->status = BUSY_STATE; + putShortMSB(s, header); + + /* Save the adler32 of the preset dictionary: */ + if (s->strstart != 0) { + putShortMSB(s, (uInt)(strm->adler >> 16)); + putShortMSB(s, (uInt)(strm->adler & 0xffff)); + } + strm->adler = 1L; + } + + /* Flush as much pending output as possible */ + if (s->pending != 0) { + flush_pending(strm); + if (strm->avail_out == 0) { + /* Since avail_out is 0, deflate will be called again with + * more output space, but possibly with both pending and + * avail_in equal to zero. There won't be anything to do, + * but this is not an error situation so make sure we + * return OK instead of BUF_ERROR at next call of deflate: + */ + s->last_flush = -1; + return Z_OK; + } + + /* Make sure there is something to do and avoid duplicate consecutive + * flushes. For repeated and useless calls with Z_FINISH, we keep + * returning Z_STREAM_END instead of Z_BUFF_ERROR. + */ + } else if (strm->avail_in == 0 && flush <= old_flush && + flush != Z_FINISH) { + ERR_RETURN(strm, Z_BUF_ERROR); + } + + /* User must not provide more input after the first FINISH: */ + if (s->status == FINISH_STATE && strm->avail_in != 0) { + ERR_RETURN(strm, Z_BUF_ERROR); + } + + /* Start a new block or continue the current one. + */ + if (strm->avail_in != 0 || s->lookahead != 0 || + (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) { + block_state bstate; + + bstate = (*(configuration_table[s->level].func))(s, flush); + + if (bstate == finish_started || bstate == finish_done) { + s->status = FINISH_STATE; + } + if (bstate == need_more || bstate == finish_started) { + if (strm->avail_out == 0) { + s->last_flush = -1; /* avoid BUF_ERROR next call, see above */ + } + return Z_OK; + /* If flush != Z_NO_FLUSH && avail_out == 0, the next call + * of deflate should use the same flush parameter to make sure + * that the flush is complete. So we don't have to output an + * empty block here, this will be done at next call. This also + * ensures that for a very small output buffer, we emit at most + * one empty block. + */ + } + if (bstate == block_done) { + if (flush == Z_PARTIAL_FLUSH) { + _tr_align(s); + } else { /* FULL_FLUSH or SYNC_FLUSH */ + _tr_stored_block(s, (char*)0, 0L, 0); + /* For a full flush, this empty block will be recognized + * as a special marker by inflate_sync(). + */ + if (flush == Z_FULL_FLUSH) { + CLEAR_HASH(s); /* forget history */ + } + } + flush_pending(strm); + if (strm->avail_out == 0) { + s->last_flush = -1; /* avoid BUF_ERROR at next call, see above */ + return Z_OK; + } + } + } + Assert(strm->avail_out > 0, "bug2"); + + if (flush != Z_FINISH) return Z_OK; + if (s->noheader) return Z_STREAM_END; + + /* Write the zlib trailer (adler32) */ + putShortMSB(s, (uInt)(strm->adler >> 16)); + putShortMSB(s, (uInt)(strm->adler & 0xffff)); + flush_pending(strm); + /* If avail_out is zero, the application will call deflate again + * to flush the rest. + */ + s->noheader = -1; /* write the trailer only once! */ + return s->pending != 0 ? Z_OK : Z_STREAM_END; +} + +/* ========================================================================= */ +int ZEXPORT deflateEnd (strm) + z_streamp strm; +{ + int status; + + if (strm == Z_NULL || strm->state == Z_NULL) return Z_STREAM_ERROR; + + status = strm->state->status; + if (status != INIT_STATE && status != BUSY_STATE && + status != FINISH_STATE) { + return Z_STREAM_ERROR; + } + + /* Deallocate in reverse order of allocations: */ + TRY_FREE(strm, strm->state->pending_buf); + TRY_FREE(strm, strm->state->head); + TRY_FREE(strm, strm->state->prev); + TRY_FREE(strm, strm->state->window); + + ZFREE(strm, strm->state); + strm->state = Z_NULL; + + return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK; +} + +/* ========================================================================= + * Copy the source state to the destination state. + * To simplify the source, this is not supported for 16-bit MSDOS (which + * doesn't have enough memory anyway to duplicate compression states). + */ +int ZEXPORT deflateCopy (dest, source) + z_streamp dest; + z_streamp source; +{ +#ifdef MAXSEG_64K + return Z_STREAM_ERROR; +#else + deflate_state *ds; + deflate_state *ss; + ushf *overlay; + + + if (source == Z_NULL || dest == Z_NULL || source->state == Z_NULL) { + return Z_STREAM_ERROR; + } + + ss = source->state; + + *dest = *source; + + ds = (deflate_state *) ZALLOC(dest, 1, sizeof(deflate_state)); + if (ds == Z_NULL) return Z_MEM_ERROR; + dest->state = (struct internal_state FAR *) ds; + *ds = *ss; + ds->strm = dest; + + ds->window = (Bytef *) ZALLOC(dest, ds->w_size, 2*sizeof(Byte)); + ds->prev = (Posf *) ZALLOC(dest, ds->w_size, sizeof(Pos)); + ds->head = (Posf *) ZALLOC(dest, ds->hash_size, sizeof(Pos)); + overlay = (ushf *) ZALLOC(dest, ds->lit_bufsize, sizeof(ush)+2); + ds->pending_buf = (uchf *) overlay; + + if (ds->window == Z_NULL || ds->prev == Z_NULL || ds->head == Z_NULL || + ds->pending_buf == Z_NULL) { + deflateEnd (dest); + return Z_MEM_ERROR; + } + /* following zmemcpy do not work for 16-bit MSDOS */ + zmemcpy(ds->window, ss->window, ds->w_size * 2 * sizeof(Byte)); + zmemcpy(ds->prev, ss->prev, ds->w_size * sizeof(Pos)); + zmemcpy(ds->head, ss->head, ds->hash_size * sizeof(Pos)); + zmemcpy(ds->pending_buf, ss->pending_buf, (uInt)ds->pending_buf_size); + + ds->pending_out = ds->pending_buf + (ss->pending_out - ss->pending_buf); + ds->d_buf = overlay + ds->lit_bufsize/sizeof(ush); + ds->l_buf = ds->pending_buf + (1+sizeof(ush))*ds->lit_bufsize; + + ds->l_desc.dyn_tree = ds->dyn_ltree; + ds->d_desc.dyn_tree = ds->dyn_dtree; + ds->bl_desc.dyn_tree = ds->bl_tree; + + return Z_OK; +#endif +} + +/* =========================================================================== + * Read a new buffer from the current input stream, update the adler32 + * and total number of bytes read. All deflate() input goes through + * this function so some applications may wish to modify it to avoid + * allocating a large strm->next_in buffer and copying from it. + * (See also flush_pending()). + */ +local int read_buf(strm, buf, size) + z_streamp strm; + Bytef *buf; + unsigned size; +{ + unsigned len = strm->avail_in; + + if (len > size) len = size; + if (len == 0) return 0; + + strm->avail_in -= len; + + if (!strm->state->noheader) { + strm->adler = adler32(strm->adler, strm->next_in, len); + } + zmemcpy(buf, strm->next_in, len); + strm->next_in += len; + strm->total_in += len; + + return (int)len; +} + +/* =========================================================================== + * Initialize the "longest match" routines for a new zlib stream + */ +local void lm_init (s) + deflate_state *s; +{ + s->window_size = (ulg)2L*s->w_size; + + CLEAR_HASH(s); + + /* Set the default configuration parameters: + */ + s->max_lazy_match = configuration_table[s->level].max_lazy; + s->good_match = configuration_table[s->level].good_length; + s->nice_match = configuration_table[s->level].nice_length; + s->max_chain_length = configuration_table[s->level].max_chain; + + s->strstart = 0; + s->block_start = 0L; + s->lookahead = 0; + s->match_length = s->prev_length = MIN_MATCH-1; + s->match_available = 0; + s->ins_h = 0; +#ifdef ASMV + match_init(); /* initialize the asm code */ +#endif +} + +/* =========================================================================== + * Set match_start to the longest match starting at the given string and + * return its length. Matches shorter or equal to prev_length are discarded, + * in which case the result is equal to prev_length and match_start is + * garbage. + * IN assertions: cur_match is the head of the hash chain for the current + * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1 + * OUT assertion: the match length is not greater than s->lookahead. + */ +#ifndef ASMV +/* For 80x86 and 680x0, an optimized version will be provided in match.asm or + * match.S. The code will be functionally equivalent. + */ +#ifndef FASTEST +local uInt longest_match(s, cur_match) + deflate_state *s; + IPos cur_match; /* current match */ +{ + unsigned chain_length = s->max_chain_length;/* max hash chain length */ + register Bytef *scan = s->window + s->strstart; /* current string */ + register Bytef *match; /* matched string */ + register int len; /* length of current match */ + int best_len = s->prev_length; /* best match length so far */ + int nice_match = s->nice_match; /* stop if match long enough */ + IPos limit = s->strstart > (IPos)MAX_DIST(s) ? + s->strstart - (IPos)MAX_DIST(s) : NIL; + /* Stop when cur_match becomes <= limit. To simplify the code, + * we prevent matches with the string of window index 0. + */ + Posf *prev = s->prev; + uInt wmask = s->w_mask; + +#ifdef UNALIGNED_OK + /* Compare two bytes at a time. Note: this is not always beneficial. + * Try with and without -DUNALIGNED_OK to check. + */ + register Bytef *strend = s->window + s->strstart + MAX_MATCH - 1; + register ush scan_start = *(ushf*)scan; + register ush scan_end = *(ushf*)(scan+best_len-1); +#else + register Bytef *strend = s->window + s->strstart + MAX_MATCH; + register Byte scan_end1 = scan[best_len-1]; + register Byte scan_end = scan[best_len]; +#endif + + /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16. + * It is easy to get rid of this optimization if necessary. + */ + Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever"); + + /* Do not waste too much time if we already have a good match: */ + if (s->prev_length >= s->good_match) { + chain_length >>= 2; + } + /* Do not look for matches beyond the end of the input. This is necessary + * to make deflate deterministic. + */ + if ((uInt)nice_match > s->lookahead) nice_match = s->lookahead; + + Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead"); + + do { + Assert(cur_match < s->strstart, "no future"); + match = s->window + cur_match; + + /* Skip to next match if the match length cannot increase + * or if the match length is less than 2: + */ +#if (defined(UNALIGNED_OK) && MAX_MATCH == 258) + /* This code assumes sizeof(unsigned short) == 2. Do not use + * UNALIGNED_OK if your compiler uses a different size. + */ + if (*(ushf*)(match+best_len-1) != scan_end || + *(ushf*)match != scan_start) continue; + + /* It is not necessary to compare scan[2] and match[2] since they are + * always equal when the other bytes match, given that the hash keys + * are equal and that HASH_BITS >= 8. Compare 2 bytes at a time at + * strstart+3, +5, ... up to strstart+257. We check for insufficient + * lookahead only every 4th comparison; the 128th check will be made + * at strstart+257. If MAX_MATCH-2 is not a multiple of 8, it is + * necessary to put more guard bytes at the end of the window, or + * to check more often for insufficient lookahead. + */ + Assert(scan[2] == match[2], "scan[2]?"); + scan++, match++; + do { + } while (*(ushf*)(scan+=2) == *(ushf*)(match+=2) && + *(ushf*)(scan+=2) == *(ushf*)(match+=2) && + *(ushf*)(scan+=2) == *(ushf*)(match+=2) && + *(ushf*)(scan+=2) == *(ushf*)(match+=2) && + scan < strend); + /* The funny "do {}" generates better code on most compilers */ + + /* Here, scan <= window+strstart+257 */ + Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); + if (*scan == *match) scan++; + + len = (MAX_MATCH - 1) - (int)(strend-scan); + scan = strend - (MAX_MATCH-1); + +#else /* UNALIGNED_OK */ + + if (match[best_len] != scan_end || + match[best_len-1] != scan_end1 || + *match != *scan || + *++match != scan[1]) continue; + + /* The check at best_len-1 can be removed because it will be made + * again later. (This heuristic is not always a win.) + * It is not necessary to compare scan[2] and match[2] since they + * are always equal when the other bytes match, given that + * the hash keys are equal and that HASH_BITS >= 8. + */ + scan += 2, match++; + Assert(*scan == *match, "match[2]?"); + + /* We check for insufficient lookahead only every 8th comparison; + * the 256th check will be made at strstart+258. + */ + do { + } while (*++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + scan < strend); + + Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); + + len = MAX_MATCH - (int)(strend - scan); + scan = strend - MAX_MATCH; + +#endif /* UNALIGNED_OK */ + + if (len > best_len) { + s->match_start = cur_match; + best_len = len; + if (len >= nice_match) break; +#ifdef UNALIGNED_OK + scan_end = *(ushf*)(scan+best_len-1); +#else + scan_end1 = scan[best_len-1]; + scan_end = scan[best_len]; +#endif + } + } while ((cur_match = prev[cur_match & wmask]) > limit + && --chain_length != 0); + + if ((uInt)best_len <= s->lookahead) return (uInt)best_len; + return s->lookahead; +} + +#else /* FASTEST */ +/* --------------------------------------------------------------------------- + * Optimized version for level == 1 only + */ +local uInt longest_match(s, cur_match) + deflate_state *s; + IPos cur_match; /* current match */ +{ + register Bytef *scan = s->window + s->strstart; /* current string */ + register Bytef *match; /* matched string */ + register int len; /* length of current match */ + register Bytef *strend = s->window + s->strstart + MAX_MATCH; + + /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16. + * It is easy to get rid of this optimization if necessary. + */ + Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever"); + + Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead"); + + Assert(cur_match < s->strstart, "no future"); + + match = s->window + cur_match; + + /* Return failure if the match length is less than 2: + */ + if (match[0] != scan[0] || match[1] != scan[1]) return MIN_MATCH-1; + + /* The check at best_len-1 can be removed because it will be made + * again later. (This heuristic is not always a win.) + * It is not necessary to compare scan[2] and match[2] since they + * are always equal when the other bytes match, given that + * the hash keys are equal and that HASH_BITS >= 8. + */ + scan += 2, match += 2; + Assert(*scan == *match, "match[2]?"); + + /* We check for insufficient lookahead only every 8th comparison; + * the 256th check will be made at strstart+258. + */ + do { + } while (*++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + *++scan == *++match && *++scan == *++match && + scan < strend); + + Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); + + len = MAX_MATCH - (int)(strend - scan); + + if (len < MIN_MATCH) return MIN_MATCH - 1; + + s->match_start = cur_match; + return len <= s->lookahead ? len : s->lookahead; +} +#endif /* FASTEST */ +#endif /* ASMV */ + +#ifdef DEBUG +/* =========================================================================== + * Check that the match at match_start is indeed a match. + */ +local void check_match(s, start, match, length) + deflate_state *s; + IPos start, match; + int length; +{ + /* check that the match is indeed a match */ + if (zmemcmp(s->window + match, + s->window + start, length) != EQUAL) { + fprintf(stderr, " start %u, match %u, length %d\n", + start, match, length); + do { + fprintf(stderr, "%c%c", s->window[match++], s->window[start++]); + } while (--length != 0); + z_error("invalid match"); + } + if (z_verbose > 1) { + fprintf(stderr,"\\[%d,%d]", start-match, length); + do { putc(s->window[start++], stderr); } while (--length != 0); + } +} +#else +# define check_match(s, start, match, length) +#endif + +/* =========================================================================== + * Fill the window when the lookahead becomes insufficient. + * Updates strstart and lookahead. + * + * IN assertion: lookahead < MIN_LOOKAHEAD + * OUT assertions: strstart <= window_size-MIN_LOOKAHEAD + * At least one byte has been read, or avail_in == 0; reads are + * performed for at least two bytes (required for the zip translate_eol + * option -- not supported here). + */ +local void fill_window(s) + deflate_state *s; +{ + register unsigned n, m; + register Posf *p; + unsigned more; /* Amount of free space at the end of the window. */ + uInt wsize = s->w_size; + + do { + more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart); + + /* Deal with !@#$% 64K limit: */ + if (more == 0 && s->strstart == 0 && s->lookahead == 0) { + more = wsize; + + } else if (more == (unsigned)(-1)) { + /* Very unlikely, but possible on 16 bit machine if strstart == 0 + * and lookahead == 1 (input done one byte at time) + */ + more--; + + /* If the window is almost full and there is insufficient lookahead, + * move the upper half to the lower one to make room in the upper half. + */ + } else if (s->strstart >= wsize+MAX_DIST(s)) { + + zmemcpy(s->window, s->window+wsize, (unsigned)wsize); + s->match_start -= wsize; + s->strstart -= wsize; /* we now have strstart >= MAX_DIST */ + s->block_start -= (long) wsize; + + /* Slide the hash table (could be avoided with 32 bit values + at the expense of memory usage). We slide even when level == 0 + to keep the hash table consistent if we switch back to level > 0 + later. (Using level 0 permanently is not an optimal usage of + zlib, so we don't care about this pathological case.) + */ + n = s->hash_size; + p = &s->head[n]; + do { + m = *--p; + *p = (Pos)(m >= wsize ? m-wsize : NIL); + } while (--n); + + n = wsize; +#ifndef FASTEST + p = &s->prev[n]; + do { + m = *--p; + *p = (Pos)(m >= wsize ? m-wsize : NIL); + /* If n is not on any hash chain, prev[n] is garbage but + * its value will never be used. + */ + } while (--n); +#endif + more += wsize; + } + if (s->strm->avail_in == 0) return; + + /* If there was no sliding: + * strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 && + * more == window_size - lookahead - strstart + * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1) + * => more >= window_size - 2*WSIZE + 2 + * In the BIG_MEM or MMAP case (not yet supported), + * window_size == input_size + MIN_LOOKAHEAD && + * strstart + s->lookahead <= input_size => more >= MIN_LOOKAHEAD. + * Otherwise, window_size == 2*WSIZE so more >= 2. + * If there was sliding, more >= WSIZE. So in all cases, more >= 2. + */ + Assert(more >= 2, "more < 2"); + + n = read_buf(s->strm, s->window + s->strstart + s->lookahead, more); + s->lookahead += n; + + /* Initialize the hash value now that we have some input: */ + if (s->lookahead >= MIN_MATCH) { + s->ins_h = s->window[s->strstart]; + UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); +#if MIN_MATCH != 3 + Call UPDATE_HASH() MIN_MATCH-3 more times +#endif + } + /* If the whole input has less than MIN_MATCH bytes, ins_h is garbage, + * but this is not important since only literal bytes will be emitted. + */ + + } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0); +} + +/* =========================================================================== + * Flush the current block, with given end-of-file flag. + * IN assertion: strstart is set to the end of the current match. + */ +#define FLUSH_BLOCK_ONLY(s, eof) { \ + _tr_flush_block(s, (s->block_start >= 0L ? \ + (charf *)&s->window[(unsigned)s->block_start] : \ + (charf *)Z_NULL), \ + (ulg)((long)s->strstart - s->block_start), \ + (eof)); \ + s->block_start = s->strstart; \ + flush_pending(s->strm); \ + Tracev((stderr,"[FLUSH]")); \ +} + +/* Same but force premature exit if necessary. */ +#define FLUSH_BLOCK(s, eof) { \ + FLUSH_BLOCK_ONLY(s, eof); \ + if (s->strm->avail_out == 0) return (eof) ? finish_started : need_more; \ +} + +/* =========================================================================== + * Copy without compression as much as possible from the input stream, return + * the current block state. + * This function does not insert new strings in the dictionary since + * uncompressible data is probably not useful. This function is used + * only for the level=0 compression option. + * NOTE: this function should be optimized to avoid extra copying from + * window to pending_buf. + */ +local block_state deflate_stored(s, flush) + deflate_state *s; + int flush; +{ + /* Stored blocks are limited to 0xffff bytes, pending_buf is limited + * to pending_buf_size, and each stored block has a 5 byte header: + */ + ulg max_block_size = 0xffff; + ulg max_start; + + if (max_block_size > s->pending_buf_size - 5) { + max_block_size = s->pending_buf_size - 5; + } + + /* Copy as much as possible from input to output: */ + for (;;) { + /* Fill the window as much as possible: */ + if (s->lookahead <= 1) { + + Assert(s->strstart < s->w_size+MAX_DIST(s) || + s->block_start >= (long)s->w_size, "slide too late"); + + fill_window(s); + if (s->lookahead == 0 && flush == Z_NO_FLUSH) return need_more; + + if (s->lookahead == 0) break; /* flush the current block */ + } + Assert(s->block_start >= 0L, "block gone"); + + s->strstart += s->lookahead; + s->lookahead = 0; + + /* Emit a stored block if pending_buf will be full: */ + max_start = s->block_start + max_block_size; + if (s->strstart == 0 || (ulg)s->strstart >= max_start) { + /* strstart == 0 is possible when wraparound on 16-bit machine */ + s->lookahead = (uInt)(s->strstart - max_start); + s->strstart = (uInt)max_start; + FLUSH_BLOCK(s, 0); + } + /* Flush if we may have to slide, otherwise block_start may become + * negative and the data will be gone: + */ + if (s->strstart - (uInt)s->block_start >= MAX_DIST(s)) { + FLUSH_BLOCK(s, 0); + } + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} + +/* =========================================================================== + * Compress as much as possible from the input stream, return the current + * block state. + * This function does not perform lazy evaluation of matches and inserts + * new strings in the dictionary only for unmatched strings or for short + * matches. It is used only for the fast compression options. + */ +local block_state deflate_fast(s, flush) + deflate_state *s; + int flush; +{ + IPos hash_head = NIL; /* head of the hash chain */ + int bflush; /* set if current block must be flushed */ + + for (;;) { + /* Make sure that we always have enough lookahead, except + * at the end of the input file. We need MAX_MATCH bytes + * for the next match, plus MIN_MATCH bytes to insert the + * string following the next match. + */ + if (s->lookahead < MIN_LOOKAHEAD) { + fill_window(s); + if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { + return need_more; + } + if (s->lookahead == 0) break; /* flush the current block */ + } + + /* Insert the string window[strstart .. strstart+2] in the + * dictionary, and set hash_head to the head of the hash chain: + */ + if (s->lookahead >= MIN_MATCH) { + INSERT_STRING(s, s->strstart, hash_head); + } + + /* Find the longest match, discarding those <= prev_length. + * At this point we have always match_length < MIN_MATCH + */ + if (hash_head != NIL && s->strstart - hash_head <= MAX_DIST(s)) { + /* To simplify the code, we prevent matches with the string + * of window index 0 (in particular we have to avoid a match + * of the string with itself at the start of the input file). + */ + if (s->strategy != Z_HUFFMAN_ONLY) { + s->match_length = longest_match (s, hash_head); + } + /* longest_match() sets match_start */ + } + if (s->match_length >= MIN_MATCH) { + check_match(s, s->strstart, s->match_start, s->match_length); + + _tr_tally_dist(s, s->strstart - s->match_start, + s->match_length - MIN_MATCH, bflush); + + s->lookahead -= s->match_length; + + /* Insert new strings in the hash table only if the match length + * is not too large. This saves time but degrades compression. + */ +#ifndef FASTEST + if (s->match_length <= s->max_insert_length && + s->lookahead >= MIN_MATCH) { + s->match_length--; /* string at strstart already in hash table */ + do { + s->strstart++; + INSERT_STRING(s, s->strstart, hash_head); + /* strstart never exceeds WSIZE-MAX_MATCH, so there are + * always MIN_MATCH bytes ahead. + */ + } while (--s->match_length != 0); + s->strstart++; + } else +#endif + { + s->strstart += s->match_length; + s->match_length = 0; + s->ins_h = s->window[s->strstart]; + UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); +#if MIN_MATCH != 3 + Call UPDATE_HASH() MIN_MATCH-3 more times +#endif + /* If lookahead < MIN_MATCH, ins_h is garbage, but it does not + * matter since it will be recomputed at next deflate call. + */ + } + } else { + /* No match, output a literal byte */ + Tracevv((stderr,"%c", s->window[s->strstart])); + _tr_tally_lit (s, s->window[s->strstart], bflush); + s->lookahead--; + s->strstart++; + } + if (bflush) FLUSH_BLOCK(s, 0); + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} + +/* =========================================================================== + * Same as above, but achieves better compression. We use a lazy + * evaluation for matches: a match is finally adopted only if there is + * no better match at the next window position. + */ +local block_state deflate_slow(s, flush) + deflate_state *s; + int flush; +{ + IPos hash_head = NIL; /* head of hash chain */ + int bflush; /* set if current block must be flushed */ + + /* Process the input block. */ + for (;;) { + /* Make sure that we always have enough lookahead, except + * at the end of the input file. We need MAX_MATCH bytes + * for the next match, plus MIN_MATCH bytes to insert the + * string following the next match. + */ + if (s->lookahead < MIN_LOOKAHEAD) { + fill_window(s); + if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { + return need_more; + } + if (s->lookahead == 0) break; /* flush the current block */ + } + + /* Insert the string window[strstart .. strstart+2] in the + * dictionary, and set hash_head to the head of the hash chain: + */ + if (s->lookahead >= MIN_MATCH) { + INSERT_STRING(s, s->strstart, hash_head); + } + + /* Find the longest match, discarding those <= prev_length. + */ + s->prev_length = s->match_length, s->prev_match = s->match_start; + s->match_length = MIN_MATCH-1; + + if (hash_head != NIL && s->prev_length < s->max_lazy_match && + s->strstart - hash_head <= MAX_DIST(s)) { + /* To simplify the code, we prevent matches with the string + * of window index 0 (in particular we have to avoid a match + * of the string with itself at the start of the input file). + */ + if (s->strategy != Z_HUFFMAN_ONLY) { + s->match_length = longest_match (s, hash_head); + } + /* longest_match() sets match_start */ + + if (s->match_length <= 5 && (s->strategy == Z_FILTERED || + (s->match_length == MIN_MATCH && + s->strstart - s->match_start > TOO_FAR))) { + + /* If prev_match is also MIN_MATCH, match_start is garbage + * but we will ignore the current match anyway. + */ + s->match_length = MIN_MATCH-1; + } + } + /* If there was a match at the previous step and the current + * match is not better, output the previous match: + */ + if (s->prev_length >= MIN_MATCH && s->match_length <= s->prev_length) { + uInt max_insert = s->strstart + s->lookahead - MIN_MATCH; + /* Do not insert strings in hash table beyond this. */ + + check_match(s, s->strstart-1, s->prev_match, s->prev_length); + + _tr_tally_dist(s, s->strstart -1 - s->prev_match, + s->prev_length - MIN_MATCH, bflush); + + /* Insert in hash table all strings up to the end of the match. + * strstart-1 and strstart are already inserted. If there is not + * enough lookahead, the last two strings are not inserted in + * the hash table. + */ + s->lookahead -= s->prev_length-1; + s->prev_length -= 2; + do { + if (++s->strstart <= max_insert) { + INSERT_STRING(s, s->strstart, hash_head); + } + } while (--s->prev_length != 0); + s->match_available = 0; + s->match_length = MIN_MATCH-1; + s->strstart++; + + if (bflush) FLUSH_BLOCK(s, 0); + + } else if (s->match_available) { + /* If there was no match at the previous position, output a + * single literal. If there was a match but the current match + * is longer, truncate the previous match to a single literal. + */ + Tracevv((stderr,"%c", s->window[s->strstart-1])); + _tr_tally_lit(s, s->window[s->strstart-1], bflush); + if (bflush) { + FLUSH_BLOCK_ONLY(s, 0); + } + s->strstart++; + s->lookahead--; + if (s->strm->avail_out == 0) return need_more; + } else { + /* There is no previous match to compare with, wait for + * the next step to decide. + */ + s->match_available = 1; + s->strstart++; + s->lookahead--; + } + } + Assert (flush != Z_NO_FLUSH, "no flush?"); + if (s->match_available) { + Tracevv((stderr,"%c", s->window[s->strstart-1])); + _tr_tally_lit(s, s->window[s->strstart-1], bflush); + s->match_available = 0; + } + FLUSH_BLOCK(s, flush == Z_FINISH); + return flush == Z_FINISH ? finish_done : block_done; +} diff --git a/src/utilfuns/zlib/deflate.h b/src/utilfuns/zlib/deflate.h new file mode 100644 index 0000000..965e0a3 --- /dev/null +++ b/src/utilfuns/zlib/deflate.h @@ -0,0 +1,318 @@ +/* deflate.h -- internal compression state + * Copyright (C) 1995-2002 Jean-loup Gailly + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* @(#) $Id: deflate.h 1688 2005-01-01 04:42:26Z scribe $ */ + +#ifndef _DEFLATE_H +#define _DEFLATE_H + +#include "zutil.h" + +/* =========================================================================== + * Internal compression state. + */ + +#define LENGTH_CODES 29 +/* number of length codes, not counting the special END_BLOCK code */ + +#define LITERALS 256 +/* number of literal bytes 0..255 */ + +#define L_CODES (LITERALS+1+LENGTH_CODES) +/* number of Literal or Length codes, including the END_BLOCK code */ + +#define D_CODES 30 +/* number of distance codes */ + +#define BL_CODES 19 +/* number of codes used to transfer the bit lengths */ + +#define HEAP_SIZE (2*L_CODES+1) +/* maximum heap size */ + +#define MAX_BITS 15 +/* All codes must not exceed MAX_BITS bits */ + +#define INIT_STATE 42 +#define BUSY_STATE 113 +#define FINISH_STATE 666 +/* Stream status */ + + +/* Data structure describing a single value and its code string. */ +typedef struct ct_data_s { + union { + ush freq; /* frequency count */ + ush code; /* bit string */ + } fc; + union { + ush dad; /* father node in Huffman tree */ + ush len; /* length of bit string */ + } dl; +} FAR ct_data; + +#define Freq fc.freq +#define Code fc.code +#define Dad dl.dad +#define Len dl.len + +typedef struct static_tree_desc_s static_tree_desc; + +typedef struct tree_desc_s { + ct_data *dyn_tree; /* the dynamic tree */ + int max_code; /* largest code with non zero frequency */ + static_tree_desc *stat_desc; /* the corresponding static tree */ +} FAR tree_desc; + +typedef ush Pos; +typedef Pos FAR Posf; +typedef unsigned IPos; + +/* A Pos is an index in the character window. We use short instead of int to + * save space in the various tables. IPos is used only for parameter passing. + */ + +typedef struct internal_state { + z_streamp strm; /* pointer back to this zlib stream */ + int status; /* as the name implies */ + Bytef *pending_buf; /* output still pending */ + ulg pending_buf_size; /* size of pending_buf */ + Bytef *pending_out; /* next pending byte to output to the stream */ + int pending; /* nb of bytes in the pending buffer */ + int noheader; /* suppress zlib header and adler32 */ + Byte data_type; /* UNKNOWN, BINARY or ASCII */ + Byte method; /* STORED (for zip only) or DEFLATED */ + int last_flush; /* value of flush param for previous deflate call */ + + /* used by deflate.c: */ + + uInt w_size; /* LZ77 window size (32K by default) */ + uInt w_bits; /* log2(w_size) (8..16) */ + uInt w_mask; /* w_size - 1 */ + + Bytef *window; + /* Sliding window. Input bytes are read into the second half of the window, + * and move to the first half later to keep a dictionary of at least wSize + * bytes. With this organization, matches are limited to a distance of + * wSize-MAX_MATCH bytes, but this ensures that IO is always + * performed with a length multiple of the block size. Also, it limits + * the window size to 64K, which is quite useful on MSDOS. + * To do: use the user input buffer as sliding window. + */ + + ulg window_size; + /* Actual size of window: 2*wSize, except when the user input buffer + * is directly used as sliding window. + */ + + Posf *prev; + /* Link to older string with same hash index. To limit the size of this + * array to 64K, this link is maintained only for the last 32K strings. + * An index in this array is thus a window index modulo 32K. + */ + + Posf *head; /* Heads of the hash chains or NIL. */ + + uInt ins_h; /* hash index of string to be inserted */ + uInt hash_size; /* number of elements in hash table */ + uInt hash_bits; /* log2(hash_size) */ + uInt hash_mask; /* hash_size-1 */ + + uInt hash_shift; + /* Number of bits by which ins_h must be shifted at each input + * step. It must be such that after MIN_MATCH steps, the oldest + * byte no longer takes part in the hash key, that is: + * hash_shift * MIN_MATCH >= hash_bits + */ + + long block_start; + /* Window position at the beginning of the current output block. Gets + * negative when the window is moved backwards. + */ + + uInt match_length; /* length of best match */ + IPos prev_match; /* previous match */ + int match_available; /* set if previous match exists */ + uInt strstart; /* start of string to insert */ + uInt match_start; /* start of matching string */ + uInt lookahead; /* number of valid bytes ahead in window */ + + uInt prev_length; + /* Length of the best match at previous step. Matches not greater than this + * are discarded. This is used in the lazy match evaluation. + */ + + uInt max_chain_length; + /* To speed up deflation, hash chains are never searched beyond this + * length. A higher limit improves compression ratio but degrades the + * speed. + */ + + uInt max_lazy_match; + /* Attempt to find a better match only when the current match is strictly + * smaller than this value. This mechanism is used only for compression + * levels >= 4. + */ +# define max_insert_length max_lazy_match + /* Insert new strings in the hash table only if the match length is not + * greater than this length. This saves time but degrades compression. + * max_insert_length is used only for compression levels <= 3. + */ + + int level; /* compression level (1..9) */ + int strategy; /* favor or force Huffman coding*/ + + uInt good_match; + /* Use a faster search when the previous match is longer than this */ + + int nice_match; /* Stop searching when current match exceeds this */ + + /* used by trees.c: */ + /* Didn't use ct_data typedef below to supress compiler warning */ + struct ct_data_s dyn_ltree[HEAP_SIZE]; /* literal and length tree */ + struct ct_data_s dyn_dtree[2*D_CODES+1]; /* distance tree */ + struct ct_data_s bl_tree[2*BL_CODES+1]; /* Huffman tree for bit lengths */ + + struct tree_desc_s l_desc; /* desc. for literal tree */ + struct tree_desc_s d_desc; /* desc. for distance tree */ + struct tree_desc_s bl_desc; /* desc. for bit length tree */ + + ush bl_count[MAX_BITS+1]; + /* number of codes at each bit length for an optimal tree */ + + int heap[2*L_CODES+1]; /* heap used to build the Huffman trees */ + int heap_len; /* number of elements in the heap */ + int heap_max; /* element of largest frequency */ + /* The sons of heap[n] are heap[2*n] and heap[2*n+1]. heap[0] is not used. + * The same heap array is used to build all trees. + */ + + uch depth[2*L_CODES+1]; + /* Depth of each subtree used as tie breaker for trees of equal frequency + */ + + uchf *l_buf; /* buffer for literals or lengths */ + + uInt lit_bufsize; + /* Size of match buffer for literals/lengths. There are 4 reasons for + * limiting lit_bufsize to 64K: + * - frequencies can be kept in 16 bit counters + * - if compression is not successful for the first block, all input + * data is still in the window so we can still emit a stored block even + * when input comes from standard input. (This can also be done for + * all blocks if lit_bufsize is not greater than 32K.) + * - if compression is not successful for a file smaller than 64K, we can + * even emit a stored file instead of a stored block (saving 5 bytes). + * This is applicable only for zip (not gzip or zlib). + * - creating new Huffman trees less frequently may not provide fast + * adaptation to changes in the input data statistics. (Take for + * example a binary file with poorly compressible code followed by + * a highly compressible string table.) Smaller buffer sizes give + * fast adaptation but have of course the overhead of transmitting + * trees more frequently. + * - I can't count above 4 + */ + + uInt last_lit; /* running index in l_buf */ + + ushf *d_buf; + /* Buffer for distances. To simplify the code, d_buf and l_buf have + * the same number of elements. To use different lengths, an extra flag + * array would be necessary. + */ + + ulg opt_len; /* bit length of current block with optimal trees */ + ulg static_len; /* bit length of current block with static trees */ + uInt matches; /* number of string matches in current block */ + int last_eob_len; /* bit length of EOB code for last block */ + +#ifdef DEBUG + ulg compressed_len; /* total bit length of compressed file mod 2^32 */ + ulg bits_sent; /* bit length of compressed data sent mod 2^32 */ +#endif + + ush bi_buf; + /* Output buffer. bits are inserted starting at the bottom (least + * significant bits). + */ + int bi_valid; + /* Number of valid bits in bi_buf. All bits above the last valid bit + * are always zero. + */ + +} FAR deflate_state; + +/* Output a byte on the stream. + * IN assertion: there is enough room in pending_buf. + */ +#define put_byte(s, c) {s->pending_buf[s->pending++] = (c);} + + +#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1) +/* Minimum amount of lookahead, except at the end of the input file. + * See deflate.c for comments about the MIN_MATCH+1. + */ + +#define MAX_DIST(s) ((s)->w_size-MIN_LOOKAHEAD) +/* In order to simplify the code, particularly on 16 bit machines, match + * distances are limited to MAX_DIST instead of WSIZE. + */ + + /* in trees.c */ +void _tr_init OF((deflate_state *s)); +int _tr_tally OF((deflate_state *s, unsigned dist, unsigned lc)); +void _tr_flush_block OF((deflate_state *s, charf *buf, ulg stored_len, + int eof)); +void _tr_align OF((deflate_state *s)); +void _tr_stored_block OF((deflate_state *s, charf *buf, ulg stored_len, + int eof)); + +#define d_code(dist) \ + ((dist) < 256 ? _dist_code[dist] : _dist_code[256+((dist)>>7)]) +/* Mapping from a distance to a distance code. dist is the distance - 1 and + * must not have side effects. _dist_code[256] and _dist_code[257] are never + * used. + */ + +#ifndef DEBUG +/* Inline versions of _tr_tally for speed: */ + +#if defined(GEN_TREES_H) || !defined(STDC) + extern uch _length_code[]; + extern uch _dist_code[]; +#else + extern const uch _length_code[]; + extern const uch _dist_code[]; +#endif + +# define _tr_tally_lit(s, c, flush) \ + { uch cc = (c); \ + s->d_buf[s->last_lit] = 0; \ + s->l_buf[s->last_lit++] = cc; \ + s->dyn_ltree[cc].Freq++; \ + flush = (s->last_lit == s->lit_bufsize-1); \ + } +# define _tr_tally_dist(s, distance, length, flush) \ + { uch len = (length); \ + ush dist = (distance); \ + s->d_buf[s->last_lit] = dist; \ + s->l_buf[s->last_lit++] = len; \ + dist--; \ + s->dyn_ltree[_length_code[len]+LITERALS+1].Freq++; \ + s->dyn_dtree[d_code(dist)].Freq++; \ + flush = (s->last_lit == s->lit_bufsize-1); \ + } +#else +# define _tr_tally_lit(s, c, flush) flush = _tr_tally(s, 0, c) +# define _tr_tally_dist(s, distance, length, flush) \ + flush = _tr_tally(s, distance, length) +#endif + +#endif diff --git a/src/utilfuns/zlib/gzio.c b/src/utilfuns/zlib/gzio.c new file mode 100644 index 0000000..612c154 --- /dev/null +++ b/src/utilfuns/zlib/gzio.c @@ -0,0 +1,875 @@ +/* gzio.c -- IO on .gz files + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + * + * Compile this file with -DNO_DEFLATE to avoid the compression code. + */ + +/* @(#) $Id: gzio.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include <stdio.h> + +#include "zutil.h" + +struct internal_state {int dummy;}; /* for buggy compilers */ + +#ifndef Z_BUFSIZE +# ifdef MAXSEG_64K +# define Z_BUFSIZE 4096 /* minimize memory usage for 16-bit DOS */ +# else +# define Z_BUFSIZE 16384 +# endif +#endif +#ifndef Z_PRINTF_BUFSIZE +# define Z_PRINTF_BUFSIZE 4096 +#endif + +#define ALLOC(size) malloc(size) +#define TRYFREE(p) {if (p) free(p);} + +static int gz_magic[2] = {0x1f, 0x8b}; /* gzip magic header */ + +/* gzip flag byte */ +#define ASCII_FLAG 0x01 /* bit 0 set: file probably ascii text */ +#define HEAD_CRC 0x02 /* bit 1 set: header CRC present */ +#define EXTRA_FIELD 0x04 /* bit 2 set: extra field present */ +#define ORIG_NAME 0x08 /* bit 3 set: original file name present */ +#define COMMENT 0x10 /* bit 4 set: file comment present */ +#define RESERVED 0xE0 /* bits 5..7: reserved */ + +typedef struct gz_stream { + z_stream stream; + int z_err; /* error code for last stream operation */ + int z_eof; /* set if end of input file */ + FILE *file; /* .gz file */ + Byte *inbuf; /* input buffer */ + Byte *outbuf; /* output buffer */ + uLong crc; /* crc32 of uncompressed data */ + char *msg; /* error message */ + char *path; /* path name for debugging only */ + int transparent; /* 1 if input file is not a .gz file */ + char mode; /* 'w' or 'r' */ + long startpos; /* start of compressed data in file (header skipped) */ +} gz_stream; + + +local gzFile gz_open OF((const char *path, const char *mode, int fd)); +local int do_flush OF((gzFile file, int flush)); +local int get_byte OF((gz_stream *s)); +local void check_header OF((gz_stream *s)); +local int destroy OF((gz_stream *s)); +local void putLong OF((FILE *file, uLong x)); +local uLong getLong OF((gz_stream *s)); + +/* =========================================================================== + Opens a gzip (.gz) file for reading or writing. The mode parameter + is as in fopen ("rb" or "wb"). The file is given either by file descriptor + or path name (if fd == -1). + gz_open return NULL if the file could not be opened or if there was + insufficient memory to allocate the (de)compression state; errno + can be checked to distinguish the two cases (if errno is zero, the + zlib error is Z_MEM_ERROR). +*/ +local gzFile gz_open (path, mode, fd) + const char *path; + const char *mode; + int fd; +{ + int err; + int level = Z_DEFAULT_COMPRESSION; /* compression level */ + int strategy = Z_DEFAULT_STRATEGY; /* compression strategy */ + char *p = (char*)mode; + gz_stream *s; + char fmode[80]; /* copy of mode, without the compression level */ + char *m = fmode; + + if (!path || !mode) return Z_NULL; + + s = (gz_stream *)ALLOC(sizeof(gz_stream)); + if (!s) return Z_NULL; + + s->stream.zalloc = (alloc_func)0; + s->stream.zfree = (free_func)0; + s->stream.opaque = (voidpf)0; + s->stream.next_in = s->inbuf = Z_NULL; + s->stream.next_out = s->outbuf = Z_NULL; + s->stream.avail_in = s->stream.avail_out = 0; + s->file = NULL; + s->z_err = Z_OK; + s->z_eof = 0; + s->crc = crc32(0L, Z_NULL, 0); + s->msg = NULL; + s->transparent = 0; + + s->path = (char*)ALLOC(strlen(path)+1); + if (s->path == NULL) { + return destroy(s), (gzFile)Z_NULL; + } + strcpy(s->path, path); /* do this early for debugging */ + + s->mode = '\0'; + do { + if (*p == 'r') s->mode = 'r'; + if (*p == 'w' || *p == 'a') s->mode = 'w'; + if (*p >= '0' && *p <= '9') { + level = *p - '0'; + } else if (*p == 'f') { + strategy = Z_FILTERED; + } else if (*p == 'h') { + strategy = Z_HUFFMAN_ONLY; + } else { + *m++ = *p; /* copy the mode */ + } + } while (*p++ && m != fmode + sizeof(fmode)); + if (s->mode == '\0') return destroy(s), (gzFile)Z_NULL; + + if (s->mode == 'w') { +#ifdef NO_DEFLATE + err = Z_STREAM_ERROR; +#else + err = deflateInit2(&(s->stream), level, + Z_DEFLATED, -MAX_WBITS, DEF_MEM_LEVEL, strategy); + /* windowBits is passed < 0 to suppress zlib header */ + + s->stream.next_out = s->outbuf = (Byte*)ALLOC(Z_BUFSIZE); +#endif + if (err != Z_OK || s->outbuf == Z_NULL) { + return destroy(s), (gzFile)Z_NULL; + } + } else { + s->stream.next_in = s->inbuf = (Byte*)ALLOC(Z_BUFSIZE); + + err = inflateInit2(&(s->stream), -MAX_WBITS); + /* windowBits is passed < 0 to tell that there is no zlib header. + * Note that in this case inflate *requires* an extra "dummy" byte + * after the compressed stream in order to complete decompression and + * return Z_STREAM_END. Here the gzip CRC32 ensures that 4 bytes are + * present after the compressed stream. + */ + if (err != Z_OK || s->inbuf == Z_NULL) { + return destroy(s), (gzFile)Z_NULL; + } + } + s->stream.avail_out = Z_BUFSIZE; + + errno = 0; + s->file = fd < 0 ? F_OPEN(path, fmode) : (FILE*)fdopen(fd, fmode); + + if (s->file == NULL) { + return destroy(s), (gzFile)Z_NULL; + } + if (s->mode == 'w') { + /* Write a very simple .gz header: + */ + fprintf(s->file, "%c%c%c%c%c%c%c%c%c%c", gz_magic[0], gz_magic[1], + Z_DEFLATED, 0 /*flags*/, 0,0,0,0 /*time*/, 0 /*xflags*/, OS_CODE); + s->startpos = 10L; + /* We use 10L instead of ftell(s->file) to because ftell causes an + * fflush on some systems. This version of the library doesn't use + * startpos anyway in write mode, so this initialization is not + * necessary. + */ + } else { + check_header(s); /* skip the .gz header */ + s->startpos = (ftell(s->file) - s->stream.avail_in); + } + + return (gzFile)s; +} + +/* =========================================================================== + Opens a gzip (.gz) file for reading or writing. +*/ +gzFile ZEXPORT gzopen (path, mode) + const char *path; + const char *mode; +{ + return gz_open (path, mode, -1); +} + +/* =========================================================================== + Associate a gzFile with the file descriptor fd. fd is not dup'ed here + to mimic the behavio(u)r of fdopen. +*/ +gzFile ZEXPORT gzdopen (fd, mode) + int fd; + const char *mode; +{ + char name[20]; + + if (fd < 0) return (gzFile)Z_NULL; + sprintf(name, "<fd:%d>", fd); /* for debugging */ + + return gz_open (name, mode, fd); +} + +/* =========================================================================== + * Update the compression level and strategy + */ +int ZEXPORT gzsetparams (file, level, strategy) + gzFile file; + int level; + int strategy; +{ + gz_stream *s = (gz_stream*)file; + + if (s == NULL || s->mode != 'w') return Z_STREAM_ERROR; + + /* Make room to allow flushing */ + if (s->stream.avail_out == 0) { + + s->stream.next_out = s->outbuf; + if (fwrite(s->outbuf, 1, Z_BUFSIZE, s->file) != Z_BUFSIZE) { + s->z_err = Z_ERRNO; + } + s->stream.avail_out = Z_BUFSIZE; + } + + return deflateParams (&(s->stream), level, strategy); +} + +/* =========================================================================== + Read a byte from a gz_stream; update next_in and avail_in. Return EOF + for end of file. + IN assertion: the stream s has been sucessfully opened for reading. +*/ +local int get_byte(s) + gz_stream *s; +{ + if (s->z_eof) return EOF; + if (s->stream.avail_in == 0) { + errno = 0; + s->stream.avail_in = fread(s->inbuf, 1, Z_BUFSIZE, s->file); + if (s->stream.avail_in == 0) { + s->z_eof = 1; + if (ferror(s->file)) s->z_err = Z_ERRNO; + return EOF; + } + s->stream.next_in = s->inbuf; + } + s->stream.avail_in--; + return *(s->stream.next_in)++; +} + +/* =========================================================================== + Check the gzip header of a gz_stream opened for reading. Set the stream + mode to transparent if the gzip magic header is not present; set s->err + to Z_DATA_ERROR if the magic header is present but the rest of the header + is incorrect. + IN assertion: the stream s has already been created sucessfully; + s->stream.avail_in is zero for the first time, but may be non-zero + for concatenated .gz files. +*/ +local void check_header(s) + gz_stream *s; +{ + int method; /* method byte */ + int flags; /* flags byte */ + uInt len; + int c; + + /* Check the gzip magic header */ + for (len = 0; len < 2; len++) { + c = get_byte(s); + if (c != gz_magic[len]) { + if (len != 0) s->stream.avail_in++, s->stream.next_in--; + if (c != EOF) { + s->stream.avail_in++, s->stream.next_in--; + s->transparent = 1; + } + s->z_err = s->stream.avail_in != 0 ? Z_OK : Z_STREAM_END; + return; + } + } + method = get_byte(s); + flags = get_byte(s); + if (method != Z_DEFLATED || (flags & RESERVED) != 0) { + s->z_err = Z_DATA_ERROR; + return; + } + + /* Discard time, xflags and OS code: */ + for (len = 0; len < 6; len++) (void)get_byte(s); + + if ((flags & EXTRA_FIELD) != 0) { /* skip the extra field */ + len = (uInt)get_byte(s); + len += ((uInt)get_byte(s))<<8; + /* len is garbage if EOF but the loop below will quit anyway */ + while (len-- != 0 && get_byte(s) != EOF) ; + } + if ((flags & ORIG_NAME) != 0) { /* skip the original file name */ + while ((c = get_byte(s)) != 0 && c != EOF) ; + } + if ((flags & COMMENT) != 0) { /* skip the .gz file comment */ + while ((c = get_byte(s)) != 0 && c != EOF) ; + } + if ((flags & HEAD_CRC) != 0) { /* skip the header crc */ + for (len = 0; len < 2; len++) (void)get_byte(s); + } + s->z_err = s->z_eof ? Z_DATA_ERROR : Z_OK; +} + + /* =========================================================================== + * Cleanup then free the given gz_stream. Return a zlib error code. + Try freeing in the reverse order of allocations. + */ +local int destroy (s) + gz_stream *s; +{ + int err = Z_OK; + + if (!s) return Z_STREAM_ERROR; + + TRYFREE(s->msg); + + if (s->stream.state != NULL) { + if (s->mode == 'w') { +#ifdef NO_DEFLATE + err = Z_STREAM_ERROR; +#else + err = deflateEnd(&(s->stream)); +#endif + } else if (s->mode == 'r') { + err = inflateEnd(&(s->stream)); + } + } + if (s->file != NULL && fclose(s->file)) { +#ifdef ESPIPE + if (errno != ESPIPE) /* fclose is broken for pipes in HP/UX */ +#endif + err = Z_ERRNO; + } + if (s->z_err < 0) err = s->z_err; + + TRYFREE(s->inbuf); + TRYFREE(s->outbuf); + TRYFREE(s->path); + TRYFREE(s); + return err; +} + +/* =========================================================================== + Reads the given number of uncompressed bytes from the compressed file. + gzread returns the number of bytes actually read (0 for end of file). +*/ +int ZEXPORT gzread (file, buf, len) + gzFile file; + voidp buf; + unsigned len; +{ + gz_stream *s = (gz_stream*)file; + Bytef *start = (Bytef*)buf; /* starting point for crc computation */ + Byte *next_out; /* == stream.next_out but not forced far (for MSDOS) */ + + if (s == NULL || s->mode != 'r') return Z_STREAM_ERROR; + + if (s->z_err == Z_DATA_ERROR || s->z_err == Z_ERRNO) return -1; + if (s->z_err == Z_STREAM_END) return 0; /* EOF */ + + next_out = (Byte*)buf; + s->stream.next_out = (Bytef*)buf; + s->stream.avail_out = len; + + while (s->stream.avail_out != 0) { + + if (s->transparent) { + /* Copy first the lookahead bytes: */ + uInt n = s->stream.avail_in; + if (n > s->stream.avail_out) n = s->stream.avail_out; + if (n > 0) { + zmemcpy(s->stream.next_out, s->stream.next_in, n); + next_out += n; + s->stream.next_out = next_out; + s->stream.next_in += n; + s->stream.avail_out -= n; + s->stream.avail_in -= n; + } + if (s->stream.avail_out > 0) { + s->stream.avail_out -= fread(next_out, 1, s->stream.avail_out, + s->file); + } + len -= s->stream.avail_out; + s->stream.total_in += (uLong)len; + s->stream.total_out += (uLong)len; + if (len == 0) s->z_eof = 1; + return (int)len; + } + if (s->stream.avail_in == 0 && !s->z_eof) { + + errno = 0; + s->stream.avail_in = fread(s->inbuf, 1, Z_BUFSIZE, s->file); + if (s->stream.avail_in == 0) { + s->z_eof = 1; + if (ferror(s->file)) { + s->z_err = Z_ERRNO; + break; + } + } + s->stream.next_in = s->inbuf; + } + s->z_err = inflate(&(s->stream), Z_NO_FLUSH); + + if (s->z_err == Z_STREAM_END) { + /* Check CRC and original size */ + s->crc = crc32(s->crc, start, (uInt)(s->stream.next_out - start)); + start = s->stream.next_out; + + if (getLong(s) != s->crc) { + s->z_err = Z_DATA_ERROR; + } else { + (void)getLong(s); + /* The uncompressed length returned by above getlong() may + * be different from s->stream.total_out) in case of + * concatenated .gz files. Check for such files: + */ + check_header(s); + if (s->z_err == Z_OK) { + uLong total_in = s->stream.total_in; + uLong total_out = s->stream.total_out; + + inflateReset(&(s->stream)); + s->stream.total_in = total_in; + s->stream.total_out = total_out; + s->crc = crc32(0L, Z_NULL, 0); + } + } + } + if (s->z_err != Z_OK || s->z_eof) break; + } + s->crc = crc32(s->crc, start, (uInt)(s->stream.next_out - start)); + + return (int)(len - s->stream.avail_out); +} + + +/* =========================================================================== + Reads one byte from the compressed file. gzgetc returns this byte + or -1 in case of end of file or error. +*/ +int ZEXPORT gzgetc(file) + gzFile file; +{ + unsigned char c; + + return gzread(file, &c, 1) == 1 ? c : -1; +} + + +/* =========================================================================== + Reads bytes from the compressed file until len-1 characters are + read, or a newline character is read and transferred to buf, or an + end-of-file condition is encountered. The string is then terminated + with a null character. + gzgets returns buf, or Z_NULL in case of error. + + The current implementation is not optimized at all. +*/ +char * ZEXPORT gzgets(file, buf, len) + gzFile file; + char *buf; + int len; +{ + char *b = buf; + if (buf == Z_NULL || len <= 0) return Z_NULL; + + while (--len > 0 && gzread(file, buf, 1) == 1 && *buf++ != '\n') ; + *buf = '\0'; + return b == buf && len > 0 ? Z_NULL : b; +} + + +#ifndef NO_DEFLATE +/* =========================================================================== + Writes the given number of uncompressed bytes into the compressed file. + gzwrite returns the number of bytes actually written (0 in case of error). +*/ +int ZEXPORT gzwrite (file, buf, len) + gzFile file; + const voidp buf; + unsigned len; +{ + gz_stream *s = (gz_stream*)file; + + if (s == NULL || s->mode != 'w') return Z_STREAM_ERROR; + + s->stream.next_in = (Bytef*)buf; + s->stream.avail_in = len; + + while (s->stream.avail_in != 0) { + + if (s->stream.avail_out == 0) { + + s->stream.next_out = s->outbuf; + if (fwrite(s->outbuf, 1, Z_BUFSIZE, s->file) != Z_BUFSIZE) { + s->z_err = Z_ERRNO; + break; + } + s->stream.avail_out = Z_BUFSIZE; + } + s->z_err = deflate(&(s->stream), Z_NO_FLUSH); + if (s->z_err != Z_OK) break; + } + s->crc = crc32(s->crc, (const Bytef *)buf, len); + + return (int)(len - s->stream.avail_in); +} + +/* =========================================================================== + Converts, formats, and writes the args to the compressed file under + control of the format string, as in fprintf. gzprintf returns the number of + uncompressed bytes actually written (0 in case of error). +*/ +#ifdef STDC +#include <stdarg.h> + +int ZEXPORTVA gzprintf (gzFile file, const char *format, /* args */ ...) +{ + char buf[Z_PRINTF_BUFSIZE]; + va_list va; + int len; + + va_start(va, format); +#ifdef HAS_vsnprintf + (void)vsnprintf(buf, sizeof(buf), format, va); +#else + (void)vsprintf(buf, format, va); +#endif + va_end(va); + len = strlen(buf); /* some *sprintf don't return the nb of bytes written */ + if (len <= 0) return 0; + + return gzwrite(file, buf, (unsigned)len); +} +#else /* not ANSI C */ + +int ZEXPORTVA gzprintf (file, format, a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, + a11, a12, a13, a14, a15, a16, a17, a18, a19, a20) + gzFile file; + const char *format; + int a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, + a11, a12, a13, a14, a15, a16, a17, a18, a19, a20; +{ + char buf[Z_PRINTF_BUFSIZE]; + int len; + +#ifdef HAS_snprintf + snprintf(buf, sizeof(buf), format, a1, a2, a3, a4, a5, a6, a7, a8, + a9, a10, a11, a12, a13, a14, a15, a16, a17, a18, a19, a20); +#else + sprintf(buf, format, a1, a2, a3, a4, a5, a6, a7, a8, + a9, a10, a11, a12, a13, a14, a15, a16, a17, a18, a19, a20); +#endif + len = strlen(buf); /* old sprintf doesn't return the nb of bytes written */ + if (len <= 0) return 0; + + return gzwrite(file, buf, len); +} +#endif + +/* =========================================================================== + Writes c, converted to an unsigned char, into the compressed file. + gzputc returns the value that was written, or -1 in case of error. +*/ +int ZEXPORT gzputc(file, c) + gzFile file; + int c; +{ + unsigned char cc = (unsigned char) c; /* required for big endian systems */ + + return gzwrite(file, &cc, 1) == 1 ? (int)cc : -1; +} + + +/* =========================================================================== + Writes the given null-terminated string to the compressed file, excluding + the terminating null character. + gzputs returns the number of characters written, or -1 in case of error. +*/ +int ZEXPORT gzputs(file, s) + gzFile file; + const char *s; +{ + return gzwrite(file, (char*)s, (unsigned)strlen(s)); +} + + +/* =========================================================================== + Flushes all pending output into the compressed file. The parameter + flush is as in the deflate() function. +*/ +local int do_flush (file, flush) + gzFile file; + int flush; +{ + uInt len; + int done = 0; + gz_stream *s = (gz_stream*)file; + + if (s == NULL || s->mode != 'w') return Z_STREAM_ERROR; + + s->stream.avail_in = 0; /* should be zero already anyway */ + + for (;;) { + len = Z_BUFSIZE - s->stream.avail_out; + + if (len != 0) { + if ((uInt)fwrite(s->outbuf, 1, len, s->file) != len) { + s->z_err = Z_ERRNO; + return Z_ERRNO; + } + s->stream.next_out = s->outbuf; + s->stream.avail_out = Z_BUFSIZE; + } + if (done) break; + s->z_err = deflate(&(s->stream), flush); + + /* Ignore the second of two consecutive flushes: */ + if (len == 0 && s->z_err == Z_BUF_ERROR) s->z_err = Z_OK; + + /* deflate has finished flushing only when it hasn't used up + * all the available space in the output buffer: + */ + done = (s->stream.avail_out != 0 || s->z_err == Z_STREAM_END); + + if (s->z_err != Z_OK && s->z_err != Z_STREAM_END) break; + } + return s->z_err == Z_STREAM_END ? Z_OK : s->z_err; +} + +int ZEXPORT gzflush (file, flush) + gzFile file; + int flush; +{ + gz_stream *s = (gz_stream*)file; + int err = do_flush (file, flush); + + if (err) return err; + fflush(s->file); + return s->z_err == Z_STREAM_END ? Z_OK : s->z_err; +} +#endif /* NO_DEFLATE */ + +/* =========================================================================== + Sets the starting position for the next gzread or gzwrite on the given + compressed file. The offset represents a number of bytes in the + gzseek returns the resulting offset location as measured in bytes from + the beginning of the uncompressed stream, or -1 in case of error. + SEEK_END is not implemented, returns error. + In this version of the library, gzseek can be extremely slow. +*/ +z_off_t ZEXPORT gzseek (file, offset, whence) + gzFile file; + z_off_t offset; + int whence; +{ + gz_stream *s = (gz_stream*)file; + + if (s == NULL || whence == SEEK_END || + s->z_err == Z_ERRNO || s->z_err == Z_DATA_ERROR) { + return -1L; + } + + if (s->mode == 'w') { +#ifdef NO_DEFLATE + return -1L; +#else + if (whence == SEEK_SET) { + offset -= s->stream.total_in; + } + if (offset < 0) return -1L; + + /* At this point, offset is the number of zero bytes to write. */ + if (s->inbuf == Z_NULL) { + s->inbuf = (Byte*)ALLOC(Z_BUFSIZE); /* for seeking */ + zmemzero(s->inbuf, Z_BUFSIZE); + } + while (offset > 0) { + uInt size = Z_BUFSIZE; + if (offset < Z_BUFSIZE) size = (uInt)offset; + + size = gzwrite(file, s->inbuf, size); + if (size == 0) return -1L; + + offset -= size; + } + return (z_off_t)s->stream.total_in; +#endif + } + /* Rest of function is for reading only */ + + /* compute absolute position */ + if (whence == SEEK_CUR) { + offset += s->stream.total_out; + } + if (offset < 0) return -1L; + + if (s->transparent) { + /* map to fseek */ + s->stream.avail_in = 0; + s->stream.next_in = s->inbuf; + if (fseek(s->file, offset, SEEK_SET) < 0) return -1L; + + s->stream.total_in = s->stream.total_out = (uLong)offset; + return offset; + } + + /* For a negative seek, rewind and use positive seek */ + if ((uLong)offset >= s->stream.total_out) { + offset -= s->stream.total_out; + } else if (gzrewind(file) < 0) { + return -1L; + } + /* offset is now the number of bytes to skip. */ + + if (offset != 0 && s->outbuf == Z_NULL) { + s->outbuf = (Byte*)ALLOC(Z_BUFSIZE); + } + while (offset > 0) { + int size = Z_BUFSIZE; + if (offset < Z_BUFSIZE) size = (int)offset; + + size = gzread(file, s->outbuf, (uInt)size); + if (size <= 0) return -1L; + offset -= size; + } + return (z_off_t)s->stream.total_out; +} + +/* =========================================================================== + Rewinds input file. +*/ +int ZEXPORT gzrewind (file) + gzFile file; +{ + gz_stream *s = (gz_stream*)file; + + if (s == NULL || s->mode != 'r') return -1; + + s->z_err = Z_OK; + s->z_eof = 0; + s->stream.avail_in = 0; + s->stream.next_in = s->inbuf; + s->crc = crc32(0L, Z_NULL, 0); + + if (s->startpos == 0) { /* not a compressed file */ + rewind(s->file); + return 0; + } + + (void) inflateReset(&s->stream); + return fseek(s->file, s->startpos, SEEK_SET); +} + +/* =========================================================================== + Returns the starting position for the next gzread or gzwrite on the + given compressed file. This position represents a number of bytes in the + uncompressed data stream. +*/ +z_off_t ZEXPORT gztell (file) + gzFile file; +{ + return gzseek(file, 0L, SEEK_CUR); +} + +/* =========================================================================== + Returns 1 when EOF has previously been detected reading the given + input stream, otherwise zero. +*/ +int ZEXPORT gzeof (file) + gzFile file; +{ + gz_stream *s = (gz_stream*)file; + + return (s == NULL || s->mode != 'r') ? 0 : s->z_eof; +} + +/* =========================================================================== + Outputs a long in LSB order to the given file +*/ +local void putLong (file, x) + FILE *file; + uLong x; +{ + int n; + for (n = 0; n < 4; n++) { + fputc((int)(x & 0xff), file); + x >>= 8; + } +} + +/* =========================================================================== + Reads a long in LSB order from the given gz_stream. Sets z_err in case + of error. +*/ +local uLong getLong (s) + gz_stream *s; +{ + uLong x = (uLong)get_byte(s); + int c; + + x += ((uLong)get_byte(s))<<8; + x += ((uLong)get_byte(s))<<16; + c = get_byte(s); + if (c == EOF) s->z_err = Z_DATA_ERROR; + x += ((uLong)c)<<24; + return x; +} + +/* =========================================================================== + Flushes all pending output if necessary, closes the compressed file + and deallocates all the (de)compression state. +*/ +int ZEXPORT gzclose (file) + gzFile file; +{ + int err; + gz_stream *s = (gz_stream*)file; + + if (s == NULL) return Z_STREAM_ERROR; + + if (s->mode == 'w') { +#ifdef NO_DEFLATE + return Z_STREAM_ERROR; +#else + err = do_flush (file, Z_FINISH); + if (err != Z_OK) return destroy((gz_stream*)file); + + putLong (s->file, s->crc); + putLong (s->file, s->stream.total_in); +#endif + } + return destroy((gz_stream*)file); +} + +/* =========================================================================== + Returns the error message for the last error which occured on the + given compressed file. errnum is set to zlib error number. If an + error occured in the file system and not in the compression library, + errnum is set to Z_ERRNO and the application may consult errno + to get the exact error code. +*/ +const char* ZEXPORT gzerror (file, errnum) + gzFile file; + int *errnum; +{ + char *m; + gz_stream *s = (gz_stream*)file; + + if (s == NULL) { + *errnum = Z_STREAM_ERROR; + return (const char*)ERR_MSG(Z_STREAM_ERROR); + } + *errnum = s->z_err; + if (*errnum == Z_OK) return (const char*)""; + + m = (char*)(*errnum == Z_ERRNO ? zstrerror(errno) : s->stream.msg); + + if (m == NULL || *m == '\0') m = (char*)ERR_MSG(s->z_err); + + TRYFREE(s->msg); + s->msg = (char*)ALLOC(strlen(s->path) + strlen(m) + 3); + strcpy(s->msg, s->path); + strcat(s->msg, ": "); + strcat(s->msg, m); + return (const char*)s->msg; +} diff --git a/src/utilfuns/zlib/infblock.c b/src/utilfuns/zlib/infblock.c new file mode 100644 index 0000000..dd7a6d4 --- /dev/null +++ b/src/utilfuns/zlib/infblock.c @@ -0,0 +1,403 @@ +/* infblock.c -- interpret and process block types to last block + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "infblock.h" +#include "inftrees.h" +#include "infcodes.h" +#include "infutil.h" + +struct inflate_codes_state {int dummy;}; /* for buggy compilers */ + +/* simplify the use of the inflate_huft type with some defines */ +#define exop word.what.Exop +#define bits word.what.Bits + +/* Table for deflate from PKZIP's appnote.txt. */ +local const uInt border[] = { /* Order of the bit length code lengths */ + 16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; + +/* + Notes beyond the 1.93a appnote.txt: + + 1. Distance pointers never point before the beginning of the output + stream. + 2. Distance pointers can point back across blocks, up to 32k away. + 3. There is an implied maximum of 7 bits for the bit length table and + 15 bits for the actual data. + 4. If only one code exists, then it is encoded using one bit. (Zero + would be more efficient, but perhaps a little confusing.) If two + codes exist, they are coded using one bit each (0 and 1). + 5. There is no way of sending zero distance codes--a dummy must be + sent if there are none. (History: a pre 2.0 version of PKZIP would + store blocks with no distance codes, but this was discovered to be + too harsh a criterion.) Valid only for 1.93a. 2.04c does allow + zero distance codes, which is sent as one code of zero bits in + length. + 6. There are up to 286 literal/length codes. Code 256 represents the + end-of-block. Note however that the static length tree defines + 288 codes just to fill out the Huffman codes. Codes 286 and 287 + cannot be used though, since there is no length base or extra bits + defined for them. Similarily, there are up to 30 distance codes. + However, static trees define 32 codes (all 5 bits) to fill out the + Huffman codes, but the last two had better not show up in the data. + 7. Unzip can check dynamic Huffman blocks for complete code sets. + The exception is that a single code would not be complete (see #4). + 8. The five bits following the block type is really the number of + literal codes sent minus 257. + 9. Length codes 8,16,16 are interpreted as 13 length codes of 8 bits + (1+6+6). Therefore, to output three times the length, you output + three codes (1+1+1), whereas to output four times the same length, + you only need two codes (1+3). Hmm. + 10. In the tree reconstruction algorithm, Code = Code + Increment + only if BitLength(i) is not zero. (Pretty obvious.) + 11. Correction: 4 Bits: # of Bit Length codes - 4 (4 - 19) + 12. Note: length code 284 can represent 227-258, but length code 285 + really is 258. The last length deserves its own, short code + since it gets used a lot in very redundant files. The length + 258 is special since 258 - 3 (the min match length) is 255. + 13. The literal/length and distance code bit lengths are read as a + single stream of lengths. It is possible (and advantageous) for + a repeat code (16, 17, or 18) to go across the boundary between + the two sets of lengths. + */ + + +void inflate_blocks_reset(s, z, c) +inflate_blocks_statef *s; +z_streamp z; +uLongf *c; +{ + if (c != Z_NULL) + *c = s->check; + if (s->mode == BTREE || s->mode == DTREE) + ZFREE(z, s->sub.trees.blens); + if (s->mode == CODES) + inflate_codes_free(s->sub.decode.codes, z); + s->mode = TYPE; + s->bitk = 0; + s->bitb = 0; + s->read = s->write = s->window; + if (s->checkfn != Z_NULL) + z->adler = s->check = (*s->checkfn)(0L, (const Bytef *)Z_NULL, 0); + Tracev((stderr, "inflate: blocks reset\n")); +} + + +inflate_blocks_statef *inflate_blocks_new(z, c, w) +z_streamp z; +check_func c; +uInt w; +{ + inflate_blocks_statef *s; + + if ((s = (inflate_blocks_statef *)ZALLOC + (z,1,sizeof(struct inflate_blocks_state))) == Z_NULL) + return s; + if ((s->hufts = + (inflate_huft *)ZALLOC(z, sizeof(inflate_huft), MANY)) == Z_NULL) + { + ZFREE(z, s); + return Z_NULL; + } + if ((s->window = (Bytef *)ZALLOC(z, 1, w)) == Z_NULL) + { + ZFREE(z, s->hufts); + ZFREE(z, s); + return Z_NULL; + } + s->end = s->window + w; + s->checkfn = c; + s->mode = TYPE; + Tracev((stderr, "inflate: blocks allocated\n")); + inflate_blocks_reset(s, z, Z_NULL); + return s; +} + + +int inflate_blocks(s, z, r) +inflate_blocks_statef *s; +z_streamp z; +int r; +{ + uInt t; /* temporary storage */ + uLong b; /* bit buffer */ + uInt k; /* bits in bit buffer */ + Bytef *p; /* input data pointer */ + uInt n; /* bytes available there */ + Bytef *q; /* output window write pointer */ + uInt m; /* bytes to end of window or read pointer */ + + /* copy input/output information to locals (UPDATE macro restores) */ + LOAD + + /* process input based on current state */ + while (1) switch (s->mode) + { + case TYPE: + NEEDBITS(3) + t = (uInt)b & 7; + s->last = t & 1; + switch (t >> 1) + { + case 0: /* stored */ + Tracev((stderr, "inflate: stored block%s\n", + s->last ? " (last)" : "")); + DUMPBITS(3) + t = k & 7; /* go to byte boundary */ + DUMPBITS(t) + s->mode = LENS; /* get length of stored block */ + break; + case 1: /* fixed */ + Tracev((stderr, "inflate: fixed codes block%s\n", + s->last ? " (last)" : "")); + { + uInt bl, bd; + inflate_huft *tl, *td; + + inflate_trees_fixed(&bl, &bd, &tl, &td, z); + s->sub.decode.codes = inflate_codes_new(bl, bd, tl, td, z); + if (s->sub.decode.codes == Z_NULL) + { + r = Z_MEM_ERROR; + LEAVE + } + } + DUMPBITS(3) + s->mode = CODES; + break; + case 2: /* dynamic */ + Tracev((stderr, "inflate: dynamic codes block%s\n", + s->last ? " (last)" : "")); + DUMPBITS(3) + s->mode = TABLE; + break; + case 3: /* illegal */ + DUMPBITS(3) + s->mode = BAD; + z->msg = (char*)"invalid block type"; + r = Z_DATA_ERROR; + LEAVE + } + break; + case LENS: + NEEDBITS(32) + if ((((~b) >> 16) & 0xffff) != (b & 0xffff)) + { + s->mode = BAD; + z->msg = (char*)"invalid stored block lengths"; + r = Z_DATA_ERROR; + LEAVE + } + s->sub.left = (uInt)b & 0xffff; + b = k = 0; /* dump bits */ + Tracev((stderr, "inflate: stored length %u\n", s->sub.left)); + s->mode = s->sub.left ? STORED : (s->last ? DRY : TYPE); + break; + case STORED: + if (n == 0) + LEAVE + NEEDOUT + t = s->sub.left; + if (t > n) t = n; + if (t > m) t = m; + zmemcpy(q, p, t); + p += t; n -= t; + q += t; m -= t; + if ((s->sub.left -= t) != 0) + break; + Tracev((stderr, "inflate: stored end, %lu total out\n", + z->total_out + (q >= s->read ? q - s->read : + (s->end - s->read) + (q - s->window)))); + s->mode = s->last ? DRY : TYPE; + break; + case TABLE: + NEEDBITS(14) + s->sub.trees.table = t = (uInt)b & 0x3fff; +#ifndef PKZIP_BUG_WORKAROUND + if ((t & 0x1f) > 29 || ((t >> 5) & 0x1f) > 29) + { + s->mode = BAD; + z->msg = (char*)"too many length or distance symbols"; + r = Z_DATA_ERROR; + LEAVE + } +#endif + t = 258 + (t & 0x1f) + ((t >> 5) & 0x1f); + if ((s->sub.trees.blens = (uIntf*)ZALLOC(z, t, sizeof(uInt))) == Z_NULL) + { + r = Z_MEM_ERROR; + LEAVE + } + DUMPBITS(14) + s->sub.trees.index = 0; + Tracev((stderr, "inflate: table sizes ok\n")); + s->mode = BTREE; + case BTREE: + while (s->sub.trees.index < 4 + (s->sub.trees.table >> 10)) + { + NEEDBITS(3) + s->sub.trees.blens[border[s->sub.trees.index++]] = (uInt)b & 7; + DUMPBITS(3) + } + while (s->sub.trees.index < 19) + s->sub.trees.blens[border[s->sub.trees.index++]] = 0; + s->sub.trees.bb = 7; + t = inflate_trees_bits(s->sub.trees.blens, &s->sub.trees.bb, + &s->sub.trees.tb, s->hufts, z); + if (t != Z_OK) + { + r = t; + if (r == Z_DATA_ERROR) + { + ZFREE(z, s->sub.trees.blens); + s->mode = BAD; + } + LEAVE + } + s->sub.trees.index = 0; + Tracev((stderr, "inflate: bits tree ok\n")); + s->mode = DTREE; + case DTREE: + while (t = s->sub.trees.table, + s->sub.trees.index < 258 + (t & 0x1f) + ((t >> 5) & 0x1f)) + { + inflate_huft *h; + uInt i, j, c; + + t = s->sub.trees.bb; + NEEDBITS(t) + h = s->sub.trees.tb + ((uInt)b & inflate_mask[t]); + t = h->bits; + c = h->base; + if (c < 16) + { + DUMPBITS(t) + s->sub.trees.blens[s->sub.trees.index++] = c; + } + else /* c == 16..18 */ + { + i = c == 18 ? 7 : c - 14; + j = c == 18 ? 11 : 3; + NEEDBITS(t + i) + DUMPBITS(t) + j += (uInt)b & inflate_mask[i]; + DUMPBITS(i) + i = s->sub.trees.index; + t = s->sub.trees.table; + if (i + j > 258 + (t & 0x1f) + ((t >> 5) & 0x1f) || + (c == 16 && i < 1)) + { + ZFREE(z, s->sub.trees.blens); + s->mode = BAD; + z->msg = (char*)"invalid bit length repeat"; + r = Z_DATA_ERROR; + LEAVE + } + c = c == 16 ? s->sub.trees.blens[i - 1] : 0; + do { + s->sub.trees.blens[i++] = c; + } while (--j); + s->sub.trees.index = i; + } + } + s->sub.trees.tb = Z_NULL; + { + uInt bl, bd; + inflate_huft *tl, *td; + inflate_codes_statef *c; + + bl = 9; /* must be <= 9 for lookahead assumptions */ + bd = 6; /* must be <= 9 for lookahead assumptions */ + t = s->sub.trees.table; + t = inflate_trees_dynamic(257 + (t & 0x1f), 1 + ((t >> 5) & 0x1f), + s->sub.trees.blens, &bl, &bd, &tl, &td, + s->hufts, z); + if (t != Z_OK) + { + if (t == (uInt)Z_DATA_ERROR) + { + ZFREE(z, s->sub.trees.blens); + s->mode = BAD; + } + r = t; + LEAVE + } + Tracev((stderr, "inflate: trees ok\n")); + if ((c = inflate_codes_new(bl, bd, tl, td, z)) == Z_NULL) + { + r = Z_MEM_ERROR; + LEAVE + } + s->sub.decode.codes = c; + } + ZFREE(z, s->sub.trees.blens); + s->mode = CODES; + case CODES: + UPDATE + if ((r = inflate_codes(s, z, r)) != Z_STREAM_END) + return inflate_flush(s, z, r); + r = Z_OK; + inflate_codes_free(s->sub.decode.codes, z); + LOAD + Tracev((stderr, "inflate: codes end, %lu total out\n", + z->total_out + (q >= s->read ? q - s->read : + (s->end - s->read) + (q - s->window)))); + if (!s->last) + { + s->mode = TYPE; + break; + } + s->mode = DRY; + case DRY: + FLUSH + if (s->read != s->write) + LEAVE + s->mode = DONE; + case DONE: + r = Z_STREAM_END; + LEAVE + case BAD: + r = Z_DATA_ERROR; + LEAVE + default: + r = Z_STREAM_ERROR; + LEAVE + } +} + + +int inflate_blocks_free(s, z) +inflate_blocks_statef *s; +z_streamp z; +{ + inflate_blocks_reset(s, z, Z_NULL); + ZFREE(z, s->window); + ZFREE(z, s->hufts); + ZFREE(z, s); + Tracev((stderr, "inflate: blocks freed\n")); + return Z_OK; +} + + +void inflate_set_dictionary(s, d, n) +inflate_blocks_statef *s; +const Bytef *d; +uInt n; +{ + zmemcpy(s->window, d, n); + s->read = s->write = s->window + n; +} + + +/* Returns true if inflate is currently at the end of a block generated + * by Z_SYNC_FLUSH or Z_FULL_FLUSH. + * IN assertion: s != Z_NULL + */ +int inflate_blocks_sync_point(s) +inflate_blocks_statef *s; +{ + return s->mode == LENS; +} diff --git a/src/utilfuns/zlib/infblock.h b/src/utilfuns/zlib/infblock.h new file mode 100644 index 0000000..173b226 --- /dev/null +++ b/src/utilfuns/zlib/infblock.h @@ -0,0 +1,39 @@ +/* infblock.h -- header to use infblock.c + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +struct inflate_blocks_state; +typedef struct inflate_blocks_state FAR inflate_blocks_statef; + +extern inflate_blocks_statef * inflate_blocks_new OF(( + z_streamp z, + check_func c, /* check function */ + uInt w)); /* window size */ + +extern int inflate_blocks OF(( + inflate_blocks_statef *, + z_streamp , + int)); /* initial return code */ + +extern void inflate_blocks_reset OF(( + inflate_blocks_statef *, + z_streamp , + uLongf *)); /* check value on output */ + +extern int inflate_blocks_free OF(( + inflate_blocks_statef *, + z_streamp)); + +extern void inflate_set_dictionary OF(( + inflate_blocks_statef *s, + const Bytef *d, /* dictionary */ + uInt n)); /* dictionary length */ + +extern int inflate_blocks_sync_point OF(( + inflate_blocks_statef *s)); diff --git a/src/utilfuns/zlib/infcodes.c b/src/utilfuns/zlib/infcodes.c new file mode 100644 index 0000000..0d54ea7 --- /dev/null +++ b/src/utilfuns/zlib/infcodes.c @@ -0,0 +1,251 @@ +/* infcodes.c -- process literals and length/distance pairs + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "inftrees.h" +#include "infblock.h" +#include "infcodes.h" +#include "infutil.h" +#include "inffast.h" + +/* simplify the use of the inflate_huft type with some defines */ +#define exop word.what.Exop +#define bits word.what.Bits + +typedef enum { /* waiting for "i:"=input, "o:"=output, "x:"=nothing */ + START, /* x: set up for LEN */ + LEN, /* i: get length/literal/eob next */ + LENEXT, /* i: getting length extra (have base) */ + DIST, /* i: get distance next */ + DISTEXT, /* i: getting distance extra */ + COPY, /* o: copying bytes in window, waiting for space */ + LIT, /* o: got literal, waiting for output space */ + WASH, /* o: got eob, possibly still output waiting */ + END, /* x: got eob and all data flushed */ + BADCODE} /* x: got error */ +inflate_codes_mode; + +/* inflate codes private state */ +struct inflate_codes_state { + + /* mode */ + inflate_codes_mode mode; /* current inflate_codes mode */ + + /* mode dependent information */ + uInt len; + union { + struct { + inflate_huft *tree; /* pointer into tree */ + uInt need; /* bits needed */ + } code; /* if LEN or DIST, where in tree */ + uInt lit; /* if LIT, literal */ + struct { + uInt get; /* bits to get for extra */ + uInt dist; /* distance back to copy from */ + } copy; /* if EXT or COPY, where and how much */ + } sub; /* submode */ + + /* mode independent information */ + Byte lbits; /* ltree bits decoded per branch */ + Byte dbits; /* dtree bits decoder per branch */ + inflate_huft *ltree; /* literal/length/eob tree */ + inflate_huft *dtree; /* distance tree */ + +}; + + +inflate_codes_statef *inflate_codes_new(bl, bd, tl, td, z) +uInt bl, bd; +inflate_huft *tl; +inflate_huft *td; /* need separate declaration for Borland C++ */ +z_streamp z; +{ + inflate_codes_statef *c; + + if ((c = (inflate_codes_statef *) + ZALLOC(z,1,sizeof(struct inflate_codes_state))) != Z_NULL) + { + c->mode = START; + c->lbits = (Byte)bl; + c->dbits = (Byte)bd; + c->ltree = tl; + c->dtree = td; + Tracev((stderr, "inflate: codes new\n")); + } + return c; +} + + +int inflate_codes(s, z, r) +inflate_blocks_statef *s; +z_streamp z; +int r; +{ + uInt j; /* temporary storage */ + inflate_huft *t; /* temporary pointer */ + uInt e; /* extra bits or operation */ + uLong b; /* bit buffer */ + uInt k; /* bits in bit buffer */ + Bytef *p; /* input data pointer */ + uInt n; /* bytes available there */ + Bytef *q; /* output window write pointer */ + uInt m; /* bytes to end of window or read pointer */ + unsigned long csf; /* pointer to copy strings from */ + inflate_codes_statef *c = s->sub.decode.codes; /* codes state */ + + /* copy input/output information to locals (UPDATE macro restores) */ + LOAD + + /* process input and output based on current state */ + while (1) switch (c->mode) + { /* waiting for "i:"=input, "o:"=output, "x:"=nothing */ + case START: /* x: set up for LEN */ +#ifndef SLOW + if (m >= 258 && n >= 10) + { + UPDATE + r = inflate_fast(c->lbits, c->dbits, c->ltree, c->dtree, s, z); + LOAD + if (r != Z_OK) + { + c->mode = r == Z_STREAM_END ? WASH : BADCODE; + break; + } + } +#endif /* !SLOW */ + c->sub.code.need = c->lbits; + c->sub.code.tree = c->ltree; + c->mode = LEN; + case LEN: /* i: get length/literal/eob next */ + j = c->sub.code.need; + NEEDBITS(j) + t = c->sub.code.tree + ((uInt)b & inflate_mask[j]); + DUMPBITS(t->bits) + e = (uInt)(t->exop); + if (e == 0) /* literal */ + { + c->sub.lit = t->base; + Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ? + "inflate: literal '%c'\n" : + "inflate: literal 0x%02x\n", t->base)); + c->mode = LIT; + break; + } + if (e & 16) /* length */ + { + c->sub.copy.get = e & 15; + c->len = t->base; + c->mode = LENEXT; + break; + } + if ((e & 64) == 0) /* next table */ + { + c->sub.code.need = e; + c->sub.code.tree = t + t->base; + break; + } + if (e & 32) /* end of block */ + { + Tracevv((stderr, "inflate: end of block\n")); + c->mode = WASH; + break; + } + c->mode = BADCODE; /* invalid code */ + z->msg = (char*)"invalid literal/length code"; + r = Z_DATA_ERROR; + LEAVE + case LENEXT: /* i: getting length extra (have base) */ + j = c->sub.copy.get; + NEEDBITS(j) + c->len += (uInt)b & inflate_mask[j]; + DUMPBITS(j) + c->sub.code.need = c->dbits; + c->sub.code.tree = c->dtree; + Tracevv((stderr, "inflate: length %u\n", c->len)); + c->mode = DIST; + case DIST: /* i: get distance next */ + j = c->sub.code.need; + NEEDBITS(j) + t = c->sub.code.tree + ((uInt)b & inflate_mask[j]); + DUMPBITS(t->bits) + e = (uInt)(t->exop); + if (e & 16) /* distance */ + { + c->sub.copy.get = e & 15; + c->sub.copy.dist = t->base; + c->mode = DISTEXT; + break; + } + if ((e & 64) == 0) /* next table */ + { + c->sub.code.need = e; + c->sub.code.tree = t + t->base; + break; + } + c->mode = BADCODE; /* invalid code */ + z->msg = (char*)"invalid distance code"; + r = Z_DATA_ERROR; + LEAVE + case DISTEXT: /* i: getting distance extra */ + j = c->sub.copy.get; + NEEDBITS(j) + c->sub.copy.dist += (uInt)b & inflate_mask[j]; + DUMPBITS(j) + Tracevv((stderr, "inflate: distance %u\n", c->sub.copy.dist)); + c->mode = COPY; + case COPY: /* o: copying bytes in window, waiting for space */ + csf = (unsigned long)q - c->sub.copy.dist; + while (csf < (unsigned long)s->window) /* modulo window size-"while" instead */ + csf += (unsigned long)(s->end - s->window); /* of "if" handles invalid distances */ + while (c->len) + { + NEEDOUT + OUTBYTE(*(Bytef *)csf++) + if (csf == (unsigned long)s->end) + csf = (unsigned long)s->window; + c->len--; + } + c->mode = START; + break; + case LIT: /* o: got literal, waiting for output space */ + NEEDOUT + OUTBYTE(c->sub.lit) + c->mode = START; + break; + case WASH: /* o: got eob, possibly more output */ + if (k > 7) /* return unused byte, if any */ + { + Assert(k < 16, "inflate_codes grabbed too many bytes") + k -= 8; + n++; + p--; /* can always return one */ + } + FLUSH + if (s->read != s->write) + LEAVE + c->mode = END; + case END: + r = Z_STREAM_END; + LEAVE + case BADCODE: /* x: got error */ + r = Z_DATA_ERROR; + LEAVE + default: + r = Z_STREAM_ERROR; + LEAVE + } +#ifdef NEED_DUMMY_RETURN + return Z_STREAM_ERROR; /* Some dumb compilers complain without this */ +#endif +} + + +void inflate_codes_free(c, z) +inflate_codes_statef *c; +z_streamp z; +{ + ZFREE(z, c); + Tracev((stderr, "inflate: codes free\n")); +} diff --git a/src/utilfuns/zlib/infcodes.h b/src/utilfuns/zlib/infcodes.h new file mode 100644 index 0000000..46821a0 --- /dev/null +++ b/src/utilfuns/zlib/infcodes.h @@ -0,0 +1,27 @@ +/* infcodes.h -- header to use infcodes.c + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +struct inflate_codes_state; +typedef struct inflate_codes_state FAR inflate_codes_statef; + +extern inflate_codes_statef *inflate_codes_new OF(( + uInt, uInt, + inflate_huft *, inflate_huft *, + z_streamp )); + +extern int inflate_codes OF(( + inflate_blocks_statef *, + z_streamp , + int)); + +extern void inflate_codes_free OF(( + inflate_codes_statef *, + z_streamp )); + diff --git a/src/utilfuns/zlib/inffast.c b/src/utilfuns/zlib/inffast.c new file mode 100644 index 0000000..a7ebc64 --- /dev/null +++ b/src/utilfuns/zlib/inffast.c @@ -0,0 +1,183 @@ +/* inffast.c -- process literals and length/distance pairs fast + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "inftrees.h" +#include "infblock.h" +#include "infcodes.h" +#include "infutil.h" +#include "inffast.h" + +struct inflate_codes_state {int dummy;}; /* for buggy compilers */ + +/* simplify the use of the inflate_huft type with some defines */ +#define exop word.what.Exop +#define bits word.what.Bits + +/* macros for bit input with no checking and for returning unused bytes */ +#define GRABBITS(j) {while(k<(j)){b|=((uLong)NEXTBYTE)<<k;k+=8;}} +#define UNGRAB {c=z->avail_in-n;c=(k>>3)<c?k>>3:c;n+=c;p-=c;k-=c<<3;} + +/* Called with number of bytes left to write in window at least 258 + (the maximum string length) and number of input bytes available + at least ten. The ten bytes are six bytes for the longest length/ + distance pair plus four bytes for overloading the bit buffer. */ + +int inflate_fast(bl, bd, tl, td, s, z) +uInt bl, bd; +inflate_huft *tl; +inflate_huft *td; /* need separate declaration for Borland C++ */ +inflate_blocks_statef *s; +z_streamp z; +{ + inflate_huft *t; /* temporary pointer */ + uInt e; /* extra bits or operation */ + uLong b; /* bit buffer */ + uInt k; /* bits in bit buffer */ + Bytef *p; /* input data pointer */ + uInt n; /* bytes available there */ + Bytef *q; /* output window write pointer */ + uInt m; /* bytes to end of window or read pointer */ + uInt ml; /* mask for literal/length tree */ + uInt md; /* mask for distance tree */ + uInt c; /* bytes to copy */ + uInt d; /* distance back to copy from */ + unsigned long csp; /* copy source pointer */ + + /* load input, output, bit values */ + LOAD + + /* initialize masks */ + ml = inflate_mask[bl]; + md = inflate_mask[bd]; + + /* do until not enough input or output space for fast loop */ + do { /* assume called with m >= 258 && n >= 10 */ + /* get literal/length code */ + GRABBITS(20) /* max bits for literal/length code */ + if ((e = (t = tl + ((uInt)b & ml))->exop) == 0) + { + DUMPBITS(t->bits) + Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ? + "inflate: * literal '%c'\n" : + "inflate: * literal 0x%02x\n", t->base)); + *q++ = (Byte)t->base; + m--; + continue; + } + do { + DUMPBITS(t->bits) + if (e & 16) + { + /* get extra bits for length */ + e &= 15; + c = t->base + ((uInt)b & inflate_mask[e]); + DUMPBITS(e) + Tracevv((stderr, "inflate: * length %u\n", c)); + + /* decode distance base of block to copy */ + GRABBITS(15); /* max bits for distance code */ + e = (t = td + ((uInt)b & md))->exop; + do { + DUMPBITS(t->bits) + if (e & 16) + { + /* get extra bits to add to distance base */ + e &= 15; + GRABBITS(e) /* get extra bits (up to 13) */ + d = t->base + ((uInt)b & inflate_mask[e]); + DUMPBITS(e) + Tracevv((stderr, "inflate: * distance %u\n", d)); + + /* do the copy */ + m -= c; + csp = (unsigned long)q - d; + if (csp < (unsigned long)s->window) /* wrap if needed */ + { + do { + csp += (unsigned long)(s->end - s->window); /* force pointer in window */ + } while (csp < (unsigned long)s->window); /* covers invalid distances */ + e = (unsigned long)s->end - csp; + if (c > e) + { + c -= e; /* wrapped copy */ + do { + *q++ = *(Bytef *)csp++; + } while (--e); + csp = s->window; + do { + *q++ = *(Bytef *)csp++; + } while (--c); + } + else /* normal copy */ + { + *q++ = *(Bytef *)csp++; c--; + *q++ = *(Bytef *)csp++; c--; + do { + *q++ = *(Bytef *)csp++; + } while (--c); + } + } + else /* normal copy */ + { + *q++ = *(Bytef *)csp++; c--; + *q++ = *(Bytef *)csp++; c--; + do { + *q++ = *(Bytef *)csp++; + } while (--c); + } + break; + } + else if ((e & 64) == 0) + { + t += t->base; + e = (t += ((uInt)b & inflate_mask[e]))->exop; + } + else + { + z->msg = (char*)"invalid distance code"; + UNGRAB + UPDATE + return Z_DATA_ERROR; + } + } while (1); + break; + } + if ((e & 64) == 0) + { + t += t->base; + if ((e = (t += ((uInt)b & inflate_mask[e]))->exop) == 0) + { + DUMPBITS(t->bits) + Tracevv((stderr, t->base >= 0x20 && t->base < 0x7f ? + "inflate: * literal '%c'\n" : + "inflate: * literal 0x%02x\n", t->base)); + *q++ = (Byte)t->base; + m--; + break; + } + } + else if (e & 32) + { + Tracevv((stderr, "inflate: * end of block\n")); + UNGRAB + UPDATE + return Z_STREAM_END; + } + else + { + z->msg = (char*)"invalid literal/length code"; + UNGRAB + UPDATE + return Z_DATA_ERROR; + } + } while (1); + } while (m >= 258 && n >= 10); + + /* not enough input or output--restore pointers and return */ + UNGRAB + UPDATE + return Z_OK; +} diff --git a/src/utilfuns/zlib/inffast.h b/src/utilfuns/zlib/inffast.h new file mode 100644 index 0000000..a31a4bb --- /dev/null +++ b/src/utilfuns/zlib/inffast.h @@ -0,0 +1,17 @@ +/* inffast.h -- header to use inffast.c + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +extern int inflate_fast OF(( + uInt, + uInt, + inflate_huft *, + inflate_huft *, + inflate_blocks_statef *, + z_streamp )); diff --git a/src/utilfuns/zlib/inffixed.h b/src/utilfuns/zlib/inffixed.h new file mode 100644 index 0000000..77f7e76 --- /dev/null +++ b/src/utilfuns/zlib/inffixed.h @@ -0,0 +1,151 @@ +/* inffixed.h -- table for decoding fixed codes + * Generated automatically by the maketree.c program + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +local uInt fixed_bl = 9; +local uInt fixed_bd = 5; +local inflate_huft fixed_tl[] = { + {{{96,7}},256}, {{{0,8}},80}, {{{0,8}},16}, {{{84,8}},115}, + {{{82,7}},31}, {{{0,8}},112}, {{{0,8}},48}, {{{0,9}},192}, + {{{80,7}},10}, {{{0,8}},96}, {{{0,8}},32}, {{{0,9}},160}, + {{{0,8}},0}, {{{0,8}},128}, {{{0,8}},64}, {{{0,9}},224}, + {{{80,7}},6}, {{{0,8}},88}, {{{0,8}},24}, {{{0,9}},144}, + {{{83,7}},59}, {{{0,8}},120}, {{{0,8}},56}, {{{0,9}},208}, + {{{81,7}},17}, {{{0,8}},104}, {{{0,8}},40}, {{{0,9}},176}, + {{{0,8}},8}, {{{0,8}},136}, {{{0,8}},72}, {{{0,9}},240}, + {{{80,7}},4}, {{{0,8}},84}, {{{0,8}},20}, {{{85,8}},227}, + {{{83,7}},43}, {{{0,8}},116}, {{{0,8}},52}, {{{0,9}},200}, + {{{81,7}},13}, {{{0,8}},100}, {{{0,8}},36}, {{{0,9}},168}, + {{{0,8}},4}, {{{0,8}},132}, {{{0,8}},68}, {{{0,9}},232}, + {{{80,7}},8}, {{{0,8}},92}, {{{0,8}},28}, {{{0,9}},152}, + {{{84,7}},83}, {{{0,8}},124}, {{{0,8}},60}, {{{0,9}},216}, + {{{82,7}},23}, {{{0,8}},108}, {{{0,8}},44}, {{{0,9}},184}, + {{{0,8}},12}, {{{0,8}},140}, {{{0,8}},76}, {{{0,9}},248}, + {{{80,7}},3}, {{{0,8}},82}, {{{0,8}},18}, {{{85,8}},163}, + {{{83,7}},35}, {{{0,8}},114}, {{{0,8}},50}, {{{0,9}},196}, + {{{81,7}},11}, {{{0,8}},98}, {{{0,8}},34}, {{{0,9}},164}, + {{{0,8}},2}, {{{0,8}},130}, {{{0,8}},66}, {{{0,9}},228}, + {{{80,7}},7}, {{{0,8}},90}, {{{0,8}},26}, {{{0,9}},148}, + {{{84,7}},67}, {{{0,8}},122}, {{{0,8}},58}, {{{0,9}},212}, + {{{82,7}},19}, {{{0,8}},106}, {{{0,8}},42}, {{{0,9}},180}, + {{{0,8}},10}, {{{0,8}},138}, {{{0,8}},74}, {{{0,9}},244}, + {{{80,7}},5}, {{{0,8}},86}, {{{0,8}},22}, {{{192,8}},0}, + {{{83,7}},51}, {{{0,8}},118}, {{{0,8}},54}, {{{0,9}},204}, + {{{81,7}},15}, {{{0,8}},102}, {{{0,8}},38}, {{{0,9}},172}, + {{{0,8}},6}, {{{0,8}},134}, {{{0,8}},70}, {{{0,9}},236}, + {{{80,7}},9}, {{{0,8}},94}, {{{0,8}},30}, {{{0,9}},156}, + {{{84,7}},99}, {{{0,8}},126}, {{{0,8}},62}, {{{0,9}},220}, + {{{82,7}},27}, {{{0,8}},110}, {{{0,8}},46}, {{{0,9}},188}, + {{{0,8}},14}, {{{0,8}},142}, {{{0,8}},78}, {{{0,9}},252}, + {{{96,7}},256}, {{{0,8}},81}, {{{0,8}},17}, {{{85,8}},131}, + {{{82,7}},31}, {{{0,8}},113}, {{{0,8}},49}, {{{0,9}},194}, + {{{80,7}},10}, {{{0,8}},97}, {{{0,8}},33}, {{{0,9}},162}, + {{{0,8}},1}, {{{0,8}},129}, {{{0,8}},65}, {{{0,9}},226}, + {{{80,7}},6}, {{{0,8}},89}, {{{0,8}},25}, {{{0,9}},146}, + {{{83,7}},59}, {{{0,8}},121}, {{{0,8}},57}, {{{0,9}},210}, + {{{81,7}},17}, {{{0,8}},105}, {{{0,8}},41}, {{{0,9}},178}, + {{{0,8}},9}, {{{0,8}},137}, {{{0,8}},73}, {{{0,9}},242}, + {{{80,7}},4}, {{{0,8}},85}, {{{0,8}},21}, {{{80,8}},258}, + {{{83,7}},43}, {{{0,8}},117}, {{{0,8}},53}, {{{0,9}},202}, + {{{81,7}},13}, {{{0,8}},101}, {{{0,8}},37}, {{{0,9}},170}, + {{{0,8}},5}, {{{0,8}},133}, {{{0,8}},69}, {{{0,9}},234}, + {{{80,7}},8}, {{{0,8}},93}, {{{0,8}},29}, {{{0,9}},154}, + {{{84,7}},83}, {{{0,8}},125}, {{{0,8}},61}, {{{0,9}},218}, + {{{82,7}},23}, {{{0,8}},109}, {{{0,8}},45}, {{{0,9}},186}, + {{{0,8}},13}, {{{0,8}},141}, {{{0,8}},77}, {{{0,9}},250}, + {{{80,7}},3}, {{{0,8}},83}, {{{0,8}},19}, {{{85,8}},195}, + {{{83,7}},35}, {{{0,8}},115}, {{{0,8}},51}, {{{0,9}},198}, + {{{81,7}},11}, {{{0,8}},99}, {{{0,8}},35}, {{{0,9}},166}, + {{{0,8}},3}, {{{0,8}},131}, {{{0,8}},67}, {{{0,9}},230}, + {{{80,7}},7}, {{{0,8}},91}, {{{0,8}},27}, {{{0,9}},150}, + {{{84,7}},67}, {{{0,8}},123}, {{{0,8}},59}, {{{0,9}},214}, + {{{82,7}},19}, {{{0,8}},107}, {{{0,8}},43}, {{{0,9}},182}, + {{{0,8}},11}, {{{0,8}},139}, {{{0,8}},75}, {{{0,9}},246}, + {{{80,7}},5}, {{{0,8}},87}, {{{0,8}},23}, {{{192,8}},0}, + {{{83,7}},51}, {{{0,8}},119}, {{{0,8}},55}, {{{0,9}},206}, + {{{81,7}},15}, {{{0,8}},103}, {{{0,8}},39}, {{{0,9}},174}, + {{{0,8}},7}, {{{0,8}},135}, {{{0,8}},71}, {{{0,9}},238}, + {{{80,7}},9}, {{{0,8}},95}, {{{0,8}},31}, {{{0,9}},158}, + {{{84,7}},99}, {{{0,8}},127}, {{{0,8}},63}, {{{0,9}},222}, + {{{82,7}},27}, {{{0,8}},111}, {{{0,8}},47}, {{{0,9}},190}, + {{{0,8}},15}, {{{0,8}},143}, {{{0,8}},79}, {{{0,9}},254}, + {{{96,7}},256}, {{{0,8}},80}, {{{0,8}},16}, {{{84,8}},115}, + {{{82,7}},31}, {{{0,8}},112}, {{{0,8}},48}, {{{0,9}},193}, + {{{80,7}},10}, {{{0,8}},96}, {{{0,8}},32}, {{{0,9}},161}, + {{{0,8}},0}, {{{0,8}},128}, {{{0,8}},64}, {{{0,9}},225}, + {{{80,7}},6}, {{{0,8}},88}, {{{0,8}},24}, {{{0,9}},145}, + {{{83,7}},59}, {{{0,8}},120}, {{{0,8}},56}, {{{0,9}},209}, + {{{81,7}},17}, {{{0,8}},104}, {{{0,8}},40}, {{{0,9}},177}, + {{{0,8}},8}, {{{0,8}},136}, {{{0,8}},72}, {{{0,9}},241}, + {{{80,7}},4}, {{{0,8}},84}, {{{0,8}},20}, {{{85,8}},227}, + {{{83,7}},43}, {{{0,8}},116}, {{{0,8}},52}, {{{0,9}},201}, + {{{81,7}},13}, {{{0,8}},100}, {{{0,8}},36}, {{{0,9}},169}, + {{{0,8}},4}, {{{0,8}},132}, {{{0,8}},68}, {{{0,9}},233}, + {{{80,7}},8}, {{{0,8}},92}, {{{0,8}},28}, {{{0,9}},153}, + {{{84,7}},83}, {{{0,8}},124}, {{{0,8}},60}, {{{0,9}},217}, + {{{82,7}},23}, {{{0,8}},108}, {{{0,8}},44}, {{{0,9}},185}, + {{{0,8}},12}, {{{0,8}},140}, {{{0,8}},76}, {{{0,9}},249}, + {{{80,7}},3}, {{{0,8}},82}, {{{0,8}},18}, {{{85,8}},163}, + {{{83,7}},35}, {{{0,8}},114}, {{{0,8}},50}, {{{0,9}},197}, + {{{81,7}},11}, {{{0,8}},98}, {{{0,8}},34}, {{{0,9}},165}, + {{{0,8}},2}, {{{0,8}},130}, {{{0,8}},66}, {{{0,9}},229}, + {{{80,7}},7}, {{{0,8}},90}, {{{0,8}},26}, {{{0,9}},149}, + {{{84,7}},67}, {{{0,8}},122}, {{{0,8}},58}, {{{0,9}},213}, + {{{82,7}},19}, {{{0,8}},106}, {{{0,8}},42}, {{{0,9}},181}, + {{{0,8}},10}, {{{0,8}},138}, {{{0,8}},74}, {{{0,9}},245}, + {{{80,7}},5}, {{{0,8}},86}, {{{0,8}},22}, {{{192,8}},0}, + {{{83,7}},51}, {{{0,8}},118}, {{{0,8}},54}, {{{0,9}},205}, + {{{81,7}},15}, {{{0,8}},102}, {{{0,8}},38}, {{{0,9}},173}, + {{{0,8}},6}, {{{0,8}},134}, {{{0,8}},70}, {{{0,9}},237}, + {{{80,7}},9}, {{{0,8}},94}, {{{0,8}},30}, {{{0,9}},157}, + {{{84,7}},99}, {{{0,8}},126}, {{{0,8}},62}, {{{0,9}},221}, + {{{82,7}},27}, {{{0,8}},110}, {{{0,8}},46}, {{{0,9}},189}, + {{{0,8}},14}, {{{0,8}},142}, {{{0,8}},78}, {{{0,9}},253}, + {{{96,7}},256}, {{{0,8}},81}, {{{0,8}},17}, {{{85,8}},131}, + {{{82,7}},31}, {{{0,8}},113}, {{{0,8}},49}, {{{0,9}},195}, + {{{80,7}},10}, {{{0,8}},97}, {{{0,8}},33}, {{{0,9}},163}, + {{{0,8}},1}, {{{0,8}},129}, {{{0,8}},65}, {{{0,9}},227}, + {{{80,7}},6}, {{{0,8}},89}, {{{0,8}},25}, {{{0,9}},147}, + {{{83,7}},59}, {{{0,8}},121}, {{{0,8}},57}, {{{0,9}},211}, + {{{81,7}},17}, {{{0,8}},105}, {{{0,8}},41}, {{{0,9}},179}, + {{{0,8}},9}, {{{0,8}},137}, {{{0,8}},73}, {{{0,9}},243}, + {{{80,7}},4}, {{{0,8}},85}, {{{0,8}},21}, {{{80,8}},258}, + {{{83,7}},43}, {{{0,8}},117}, {{{0,8}},53}, {{{0,9}},203}, + {{{81,7}},13}, {{{0,8}},101}, {{{0,8}},37}, {{{0,9}},171}, + {{{0,8}},5}, {{{0,8}},133}, {{{0,8}},69}, {{{0,9}},235}, + {{{80,7}},8}, {{{0,8}},93}, {{{0,8}},29}, {{{0,9}},155}, + {{{84,7}},83}, {{{0,8}},125}, {{{0,8}},61}, {{{0,9}},219}, + {{{82,7}},23}, {{{0,8}},109}, {{{0,8}},45}, {{{0,9}},187}, + {{{0,8}},13}, {{{0,8}},141}, {{{0,8}},77}, {{{0,9}},251}, + {{{80,7}},3}, {{{0,8}},83}, {{{0,8}},19}, {{{85,8}},195}, + {{{83,7}},35}, {{{0,8}},115}, {{{0,8}},51}, {{{0,9}},199}, + {{{81,7}},11}, {{{0,8}},99}, {{{0,8}},35}, {{{0,9}},167}, + {{{0,8}},3}, {{{0,8}},131}, {{{0,8}},67}, {{{0,9}},231}, + {{{80,7}},7}, {{{0,8}},91}, {{{0,8}},27}, {{{0,9}},151}, + {{{84,7}},67}, {{{0,8}},123}, {{{0,8}},59}, {{{0,9}},215}, + {{{82,7}},19}, {{{0,8}},107}, {{{0,8}},43}, {{{0,9}},183}, + {{{0,8}},11}, {{{0,8}},139}, {{{0,8}},75}, {{{0,9}},247}, + {{{80,7}},5}, {{{0,8}},87}, {{{0,8}},23}, {{{192,8}},0}, + {{{83,7}},51}, {{{0,8}},119}, {{{0,8}},55}, {{{0,9}},207}, + {{{81,7}},15}, {{{0,8}},103}, {{{0,8}},39}, {{{0,9}},175}, + {{{0,8}},7}, {{{0,8}},135}, {{{0,8}},71}, {{{0,9}},239}, + {{{80,7}},9}, {{{0,8}},95}, {{{0,8}},31}, {{{0,9}},159}, + {{{84,7}},99}, {{{0,8}},127}, {{{0,8}},63}, {{{0,9}},223}, + {{{82,7}},27}, {{{0,8}},111}, {{{0,8}},47}, {{{0,9}},191}, + {{{0,8}},15}, {{{0,8}},143}, {{{0,8}},79}, {{{0,9}},255} + }; +local inflate_huft fixed_td[] = { + {{{80,5}},1}, {{{87,5}},257}, {{{83,5}},17}, {{{91,5}},4097}, + {{{81,5}},5}, {{{89,5}},1025}, {{{85,5}},65}, {{{93,5}},16385}, + {{{80,5}},3}, {{{88,5}},513}, {{{84,5}},33}, {{{92,5}},8193}, + {{{82,5}},9}, {{{90,5}},2049}, {{{86,5}},129}, {{{192,5}},24577}, + {{{80,5}},2}, {{{87,5}},385}, {{{83,5}},25}, {{{91,5}},6145}, + {{{81,5}},7}, {{{89,5}},1537}, {{{85,5}},97}, {{{93,5}},24577}, + {{{80,5}},4}, {{{88,5}},769}, {{{84,5}},49}, {{{92,5}},12289}, + {{{82,5}},13}, {{{90,5}},3073}, {{{86,5}},193}, {{{192,5}},24577} + }; diff --git a/src/utilfuns/zlib/inflate.c b/src/utilfuns/zlib/inflate.c new file mode 100644 index 0000000..dfb2e86 --- /dev/null +++ b/src/utilfuns/zlib/inflate.c @@ -0,0 +1,366 @@ +/* inflate.c -- zlib interface to inflate modules + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "infblock.h" + +struct inflate_blocks_state {int dummy;}; /* for buggy compilers */ + +typedef enum { + METHOD, /* waiting for method byte */ + FLAG, /* waiting for flag byte */ + DICT4, /* four dictionary check bytes to go */ + DICT3, /* three dictionary check bytes to go */ + DICT2, /* two dictionary check bytes to go */ + DICT1, /* one dictionary check byte to go */ + DICT0, /* waiting for inflateSetDictionary */ + BLOCKS, /* decompressing blocks */ + CHECK4, /* four check bytes to go */ + CHECK3, /* three check bytes to go */ + CHECK2, /* two check bytes to go */ + CHECK1, /* one check byte to go */ + DONE, /* finished check, done */ + BAD} /* got an error--stay here */ +inflate_mode; + +/* inflate private state */ +struct internal_state { + + /* mode */ + inflate_mode mode; /* current inflate mode */ + + /* mode dependent information */ + union { + uInt method; /* if FLAGS, method byte */ + struct { + uLong was; /* computed check value */ + uLong need; /* stream check value */ + } check; /* if CHECK, check values to compare */ + uInt marker; /* if BAD, inflateSync's marker bytes count */ + } sub; /* submode */ + + /* mode independent information */ + int nowrap; /* flag for no wrapper */ + uInt wbits; /* log2(window size) (8..15, defaults to 15) */ + inflate_blocks_statef + *blocks; /* current inflate_blocks state */ + +}; + + +int ZEXPORT inflateReset(z) +z_streamp z; +{ + if (z == Z_NULL || z->state == Z_NULL) + return Z_STREAM_ERROR; + z->total_in = z->total_out = 0; + z->msg = Z_NULL; + z->state->mode = z->state->nowrap ? BLOCKS : METHOD; + inflate_blocks_reset(z->state->blocks, z, Z_NULL); + Tracev((stderr, "inflate: reset\n")); + return Z_OK; +} + + +int ZEXPORT inflateEnd(z) +z_streamp z; +{ + if (z == Z_NULL || z->state == Z_NULL || z->zfree == Z_NULL) + return Z_STREAM_ERROR; + if (z->state->blocks != Z_NULL) + inflate_blocks_free(z->state->blocks, z); + ZFREE(z, z->state); + z->state = Z_NULL; + Tracev((stderr, "inflate: end\n")); + return Z_OK; +} + + +int ZEXPORT inflateInit2_(z, w, version, stream_size) +z_streamp z; +int w; +const char *version; +int stream_size; +{ + if (version == Z_NULL || version[0] != ZLIB_VERSION[0] || + stream_size != sizeof(z_stream)) + return Z_VERSION_ERROR; + + /* initialize state */ + if (z == Z_NULL) + return Z_STREAM_ERROR; + z->msg = Z_NULL; + if (z->zalloc == Z_NULL) + { + z->zalloc = zcalloc; + z->opaque = (voidpf)0; + } + if (z->zfree == Z_NULL) z->zfree = zcfree; + if ((z->state = (struct internal_state FAR *) + ZALLOC(z,1,sizeof(struct internal_state))) == Z_NULL) + return Z_MEM_ERROR; + z->state->blocks = Z_NULL; + + /* handle undocumented nowrap option (no zlib header or check) */ + z->state->nowrap = 0; + if (w < 0) + { + w = - w; + z->state->nowrap = 1; + } + + /* set window size */ + if (w < 8 || w > 15) + { + inflateEnd(z); + return Z_STREAM_ERROR; + } + z->state->wbits = (uInt)w; + + /* create inflate_blocks state */ + if ((z->state->blocks = + inflate_blocks_new(z, z->state->nowrap ? Z_NULL : adler32, (uInt)1 << w)) + == Z_NULL) + { + inflateEnd(z); + return Z_MEM_ERROR; + } + Tracev((stderr, "inflate: allocated\n")); + + /* reset state */ + inflateReset(z); + return Z_OK; +} + + +int ZEXPORT inflateInit_(z, version, stream_size) +z_streamp z; +const char *version; +int stream_size; +{ + return inflateInit2_(z, DEF_WBITS, version, stream_size); +} + + +#define NEEDBYTE {if(z->avail_in==0)return r;r=f;} +#define NEXTBYTE (z->avail_in--,z->total_in++,*z->next_in++) + +int ZEXPORT inflate(z, f) +z_streamp z; +int f; +{ + int r; + uInt b; + + if (z == Z_NULL || z->state == Z_NULL || z->next_in == Z_NULL) + return Z_STREAM_ERROR; + f = f == Z_FINISH ? Z_BUF_ERROR : Z_OK; + r = Z_BUF_ERROR; + while (1) switch (z->state->mode) + { + case METHOD: + NEEDBYTE + if (((z->state->sub.method = NEXTBYTE) & 0xf) != Z_DEFLATED) + { + z->state->mode = BAD; + z->msg = (char*)"unknown compression method"; + z->state->sub.marker = 5; /* can't try inflateSync */ + break; + } + if ((z->state->sub.method >> 4) + 8 > z->state->wbits) + { + z->state->mode = BAD; + z->msg = (char*)"invalid window size"; + z->state->sub.marker = 5; /* can't try inflateSync */ + break; + } + z->state->mode = FLAG; + case FLAG: + NEEDBYTE + b = NEXTBYTE; + if (((z->state->sub.method << 8) + b) % 31) + { + z->state->mode = BAD; + z->msg = (char*)"incorrect header check"; + z->state->sub.marker = 5; /* can't try inflateSync */ + break; + } + Tracev((stderr, "inflate: zlib header ok\n")); + if (!(b & PRESET_DICT)) + { + z->state->mode = BLOCKS; + break; + } + z->state->mode = DICT4; + case DICT4: + NEEDBYTE + z->state->sub.check.need = (uLong)NEXTBYTE << 24; + z->state->mode = DICT3; + case DICT3: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE << 16; + z->state->mode = DICT2; + case DICT2: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE << 8; + z->state->mode = DICT1; + case DICT1: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE; + z->adler = z->state->sub.check.need; + z->state->mode = DICT0; + return Z_NEED_DICT; + case DICT0: + z->state->mode = BAD; + z->msg = (char*)"need dictionary"; + z->state->sub.marker = 0; /* can try inflateSync */ + return Z_STREAM_ERROR; + case BLOCKS: + r = inflate_blocks(z->state->blocks, z, r); + if (r == Z_DATA_ERROR) + { + z->state->mode = BAD; + z->state->sub.marker = 0; /* can try inflateSync */ + break; + } + if (r == Z_OK) + r = f; + if (r != Z_STREAM_END) + return r; + r = f; + inflate_blocks_reset(z->state->blocks, z, &z->state->sub.check.was); + if (z->state->nowrap) + { + z->state->mode = DONE; + break; + } + z->state->mode = CHECK4; + case CHECK4: + NEEDBYTE + z->state->sub.check.need = (uLong)NEXTBYTE << 24; + z->state->mode = CHECK3; + case CHECK3: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE << 16; + z->state->mode = CHECK2; + case CHECK2: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE << 8; + z->state->mode = CHECK1; + case CHECK1: + NEEDBYTE + z->state->sub.check.need += (uLong)NEXTBYTE; + + if (z->state->sub.check.was != z->state->sub.check.need) + { + z->state->mode = BAD; + z->msg = (char*)"incorrect data check"; + z->state->sub.marker = 5; /* can't try inflateSync */ + break; + } + Tracev((stderr, "inflate: zlib check ok\n")); + z->state->mode = DONE; + case DONE: + return Z_STREAM_END; + case BAD: + return Z_DATA_ERROR; + default: + return Z_STREAM_ERROR; + } +#ifdef NEED_DUMMY_RETURN + return Z_STREAM_ERROR; /* Some dumb compilers complain without this */ +#endif +} + + +int ZEXPORT inflateSetDictionary(z, dictionary, dictLength) +z_streamp z; +const Bytef *dictionary; +uInt dictLength; +{ + uInt length = dictLength; + + if (z == Z_NULL || z->state == Z_NULL || z->state->mode != DICT0) + return Z_STREAM_ERROR; + + if (adler32(1L, dictionary, dictLength) != z->adler) return Z_DATA_ERROR; + z->adler = 1L; + + if (length >= ((uInt)1<<z->state->wbits)) + { + length = (1<<z->state->wbits)-1; + dictionary += dictLength - length; + } + inflate_set_dictionary(z->state->blocks, dictionary, length); + z->state->mode = BLOCKS; + return Z_OK; +} + + +int ZEXPORT inflateSync(z) +z_streamp z; +{ + uInt n; /* number of bytes to look at */ + Bytef *p; /* pointer to bytes */ + uInt m; /* number of marker bytes found in a row */ + uLong r, w; /* temporaries to save total_in and total_out */ + + /* set up */ + if (z == Z_NULL || z->state == Z_NULL) + return Z_STREAM_ERROR; + if (z->state->mode != BAD) + { + z->state->mode = BAD; + z->state->sub.marker = 0; + } + if ((n = z->avail_in) == 0) + return Z_BUF_ERROR; + p = z->next_in; + m = z->state->sub.marker; + + /* search */ + while (n && m < 4) + { + static const Byte mark[4] = {0, 0, 0xff, 0xff}; + if (*p == mark[m]) + m++; + else if (*p) + m = 0; + else + m = 4 - m; + p++, n--; + } + + /* restore */ + z->total_in += p - z->next_in; + z->next_in = p; + z->avail_in = n; + z->state->sub.marker = m; + + /* return no joy or set up to restart on a new block */ + if (m != 4) + return Z_DATA_ERROR; + r = z->total_in; w = z->total_out; + inflateReset(z); + z->total_in = r; z->total_out = w; + z->state->mode = BLOCKS; + return Z_OK; +} + + +/* Returns true if inflate is currently at the end of a block generated + * by Z_SYNC_FLUSH or Z_FULL_FLUSH. This function is used by one PPP + * implementation to provide an additional safety check. PPP uses Z_SYNC_FLUSH + * but removes the length bytes of the resulting empty stored block. When + * decompressing, PPP checks that at the end of input packet, inflate is + * waiting for these length bytes. + */ +int ZEXPORT inflateSyncPoint(z) +z_streamp z; +{ + if (z == Z_NULL || z->state == Z_NULL || z->state->blocks == Z_NULL) + return Z_STREAM_ERROR; + return inflate_blocks_sync_point(z->state->blocks); +} diff --git a/src/utilfuns/zlib/inftrees.c b/src/utilfuns/zlib/inftrees.c new file mode 100644 index 0000000..4c32ca3 --- /dev/null +++ b/src/utilfuns/zlib/inftrees.c @@ -0,0 +1,454 @@ +/* inftrees.c -- generate Huffman trees for efficient decoding + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "inftrees.h" + +#if !defined(BUILDFIXED) && !defined(STDC) +# define BUILDFIXED /* non ANSI compilers may not accept inffixed.h */ +#endif + +const char inflate_copyright[] = + " inflate 1.1.4 Copyright 1995-2002 Mark Adler "; +/* + If you use the zlib library in a product, an acknowledgment is welcome + in the documentation of your product. If for some reason you cannot + include such an acknowledgment, I would appreciate that you keep this + copyright string in the executable of your product. + */ +struct internal_state {int dummy;}; /* for buggy compilers */ + +/* simplify the use of the inflate_huft type with some defines */ +#define exop word.what.Exop +#define bits word.what.Bits + + +local int huft_build OF(( + uIntf *, /* code lengths in bits */ + uInt, /* number of codes */ + uInt, /* number of "simple" codes */ + const uIntf *, /* list of base values for non-simple codes */ + const uIntf *, /* list of extra bits for non-simple codes */ + inflate_huft * FAR*,/* result: starting table */ + uIntf *, /* maximum lookup bits (returns actual) */ + inflate_huft *, /* space for trees */ + uInt *, /* hufts used in space */ + uIntf * )); /* space for values */ + +/* Tables for deflate from PKZIP's appnote.txt. */ +local const uInt cplens[31] = { /* Copy lengths for literal codes 257..285 */ + 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, + 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; + /* see note #13 above about 258 */ +local const uInt cplext[31] = { /* Extra bits for literal codes 257..285 */ + 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 2, + 3, 3, 3, 3, 4, 4, 4, 4, 5, 5, 5, 5, 0, 112, 112}; /* 112==invalid */ +local const uInt cpdist[30] = { /* Copy offsets for distance codes 0..29 */ + 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, + 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, + 8193, 12289, 16385, 24577}; +local const uInt cpdext[30] = { /* Extra bits for distance codes */ + 0, 0, 0, 0, 1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, + 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, + 12, 12, 13, 13}; + +/* + Huffman code decoding is performed using a multi-level table lookup. + The fastest way to decode is to simply build a lookup table whose + size is determined by the longest code. However, the time it takes + to build this table can also be a factor if the data being decoded + is not very long. The most common codes are necessarily the + shortest codes, so those codes dominate the decoding time, and hence + the speed. The idea is you can have a shorter table that decodes the + shorter, more probable codes, and then point to subsidiary tables for + the longer codes. The time it costs to decode the longer codes is + then traded against the time it takes to make longer tables. + + This results of this trade are in the variables lbits and dbits + below. lbits is the number of bits the first level table for literal/ + length codes can decode in one step, and dbits is the same thing for + the distance codes. Subsequent tables are also less than or equal to + those sizes. These values may be adjusted either when all of the + codes are shorter than that, in which case the longest code length in + bits is used, or when the shortest code is *longer* than the requested + table size, in which case the length of the shortest code in bits is + used. + + There are two different values for the two tables, since they code a + different number of possibilities each. The literal/length table + codes 286 possible values, or in a flat code, a little over eight + bits. The distance table codes 30 possible values, or a little less + than five bits, flat. The optimum values for speed end up being + about one bit more than those, so lbits is 8+1 and dbits is 5+1. + The optimum values may differ though from machine to machine, and + possibly even between compilers. Your mileage may vary. + */ + + +/* If BMAX needs to be larger than 16, then h and x[] should be uLong. */ +#define BMAX 15 /* maximum bit length of any code */ + +local int huft_build(b, n, s, d, e, t, m, hp, hn, v) +uIntf *b; /* code lengths in bits (all assumed <= BMAX) */ +uInt n; /* number of codes (assumed <= 288) */ +uInt s; /* number of simple-valued codes (0..s-1) */ +const uIntf *d; /* list of base values for non-simple codes */ +const uIntf *e; /* list of extra bits for non-simple codes */ +inflate_huft * FAR *t; /* result: starting table */ +uIntf *m; /* maximum lookup bits, returns actual */ +inflate_huft *hp; /* space for trees */ +uInt *hn; /* hufts used in space */ +uIntf *v; /* working area: values in order of bit length */ +/* Given a list of code lengths and a maximum table size, make a set of + tables to decode that set of codes. Return Z_OK on success, Z_BUF_ERROR + if the given code set is incomplete (the tables are still built in this + case), or Z_DATA_ERROR if the input is invalid. */ +{ + + uInt a; /* counter for codes of length k */ + uInt c[BMAX+1]; /* bit length count table */ + uInt f; /* i repeats in table every f entries */ + int g; /* maximum code length */ + int h; /* table level */ + register uInt i; /* counter, current code */ + register uInt j; /* counter */ + register int k; /* number of bits in current code */ + int l; /* bits per table (returned in m) */ + uInt mask; /* (1 << w) - 1, to avoid cc -O bug on HP */ + register uIntf *p; /* pointer into c[], b[], or v[] */ + inflate_huft *q; /* points to current table */ + struct inflate_huft_s r; /* table entry for structure assignment */ + inflate_huft *u[BMAX]; /* table stack */ + register int w; /* bits before this table == (l * h) */ + uInt x[BMAX+1]; /* bit offsets, then code stack */ + uIntf *xp; /* pointer into x */ + int y; /* number of dummy codes added */ + uInt z; /* number of entries in current table */ + + + /* Generate counts for each bit length */ + p = c; +#define C0 *p++ = 0; +#define C2 C0 C0 C0 C0 +#define C4 C2 C2 C2 C2 + C4 /* clear c[]--assume BMAX+1 is 16 */ + p = b; i = n; + do { + c[*p++]++; /* assume all entries <= BMAX */ + } while (--i); + if (c[0] == n) /* null input--all zero length codes */ + { + *t = (inflate_huft *)Z_NULL; + *m = 0; + return Z_OK; + } + + + /* Find minimum and maximum length, bound *m by those */ + l = *m; + for (j = 1; j <= BMAX; j++) + if (c[j]) + break; + k = j; /* minimum code length */ + if ((uInt)l < j) + l = j; + for (i = BMAX; i; i--) + if (c[i]) + break; + g = i; /* maximum code length */ + if ((uInt)l > i) + l = i; + *m = l; + + + /* Adjust last length count to fill out codes, if needed */ + for (y = 1 << j; j < i; j++, y <<= 1) + if ((y -= c[j]) < 0) + return Z_DATA_ERROR; + if ((y -= c[i]) < 0) + return Z_DATA_ERROR; + c[i] += y; + + + /* Generate starting offsets into the value table for each length */ + x[1] = j = 0; + p = c + 1; xp = x + 2; + while (--i) { /* note that i == g from above */ + *xp++ = (j += *p++); + } + + + /* Make a table of values in order of bit lengths */ + p = b; i = 0; + do { + if ((j = *p++) != 0) + v[x[j]++] = i; + } while (++i < n); + n = x[g]; /* set n to length of v */ + + + /* Generate the Huffman codes and for each, make the table entries */ + x[0] = i = 0; /* first Huffman code is zero */ + p = v; /* grab values in bit order */ + h = -1; /* no tables yet--level -1 */ + w = -l; /* bits decoded == (l * h) */ + u[0] = (inflate_huft *)Z_NULL; /* just to keep compilers happy */ + q = (inflate_huft *)Z_NULL; /* ditto */ + z = 0; /* ditto */ + + /* go through the bit lengths (k already is bits in shortest code) */ + for (; k <= g; k++) + { + a = c[k]; + while (a--) + { + /* here i is the Huffman code of length k bits for value *p */ + /* make tables up to required level */ + while (k > w + l) + { + h++; + w += l; /* previous table always l bits */ + + /* compute minimum size table less than or equal to l bits */ + z = g - w; + z = z > (uInt)l ? l : z; /* table size upper limit */ + if ((f = 1 << (j = k - w)) > a + 1) /* try a k-w bit table */ + { /* too few codes for k-w bit table */ + f -= a + 1; /* deduct codes from patterns left */ + xp = c + k; + if (j < z) + while (++j < z) /* try smaller tables up to z bits */ + { + if ((f <<= 1) <= *++xp) + break; /* enough codes to use up j bits */ + f -= *xp; /* else deduct codes from patterns */ + } + } + z = 1 << j; /* table entries for j-bit table */ + + /* allocate new table */ + if (*hn + z > MANY) /* (note: doesn't matter for fixed) */ + return Z_DATA_ERROR; /* overflow of MANY */ + u[h] = q = hp + *hn; + *hn += z; + + /* connect to last table, if there is one */ + if (h) + { + x[h] = i; /* save pattern for backing up */ + r.bits = (Byte)l; /* bits to dump before this table */ + r.exop = (Byte)j; /* bits in this table */ + j = i >> (w - l); + r.base = (uInt)(q - u[h-1] - j); /* offset to this table */ + u[h-1][j] = r; /* connect to last table */ + } + else + *t = q; /* first table is returned result */ + } + + /* set up table entry in r */ + r.bits = (Byte)(k - w); + if (p >= v + n) + r.exop = 128 + 64; /* out of values--invalid code */ + else if (*p < s) + { + r.exop = (Byte)(*p < 256 ? 0 : 32 + 64); /* 256 is end-of-block */ + r.base = *p++; /* simple code is just the value */ + } + else + { + r.exop = (Byte)(e[*p - s] + 16 + 64);/* non-simple--look up in lists */ + r.base = d[*p++ - s]; + } + + /* fill code-like entries with r */ + f = 1 << (k - w); + for (j = i >> w; j < z; j += f) + q[j] = r; + + /* backwards increment the k-bit code i */ + for (j = 1 << (k - 1); i & j; j >>= 1) + i ^= j; + i ^= j; + + /* backup over finished tables */ + mask = (1 << w) - 1; /* needed on HP, cc -O bug */ + while ((i & mask) != x[h]) + { + h--; /* don't need to update q */ + w -= l; + mask = (1 << w) - 1; + } + } + } + + + /* Return Z_BUF_ERROR if we were given an incomplete table */ + return y != 0 && g != 1 ? Z_BUF_ERROR : Z_OK; +} + + +int inflate_trees_bits(c, bb, tb, hp, z) +uIntf *c; /* 19 code lengths */ +uIntf *bb; /* bits tree desired/actual depth */ +inflate_huft * FAR *tb; /* bits tree result */ +inflate_huft *hp; /* space for trees */ +z_streamp z; /* for messages */ +{ + int r; + uInt hn = 0; /* hufts used in space */ + uIntf *v; /* work area for huft_build */ + + if ((v = (uIntf*)ZALLOC(z, 19, sizeof(uInt))) == Z_NULL) + return Z_MEM_ERROR; + r = huft_build(c, 19, 19, (uIntf*)Z_NULL, (uIntf*)Z_NULL, + tb, bb, hp, &hn, v); + if (r == Z_DATA_ERROR) + z->msg = (char*)"oversubscribed dynamic bit lengths tree"; + else if (r == Z_BUF_ERROR || *bb == 0) + { + z->msg = (char*)"incomplete dynamic bit lengths tree"; + r = Z_DATA_ERROR; + } + ZFREE(z, v); + return r; +} + + +int inflate_trees_dynamic(nl, nd, c, bl, bd, tl, td, hp, z) +uInt nl; /* number of literal/length codes */ +uInt nd; /* number of distance codes */ +uIntf *c; /* that many (total) code lengths */ +uIntf *bl; /* literal desired/actual bit depth */ +uIntf *bd; /* distance desired/actual bit depth */ +inflate_huft * FAR *tl; /* literal/length tree result */ +inflate_huft * FAR *td; /* distance tree result */ +inflate_huft *hp; /* space for trees */ +z_streamp z; /* for messages */ +{ + int r; + uInt hn = 0; /* hufts used in space */ + uIntf *v; /* work area for huft_build */ + + /* allocate work area */ + if ((v = (uIntf*)ZALLOC(z, 288, sizeof(uInt))) == Z_NULL) + return Z_MEM_ERROR; + + /* build literal/length tree */ + r = huft_build(c, nl, 257, cplens, cplext, tl, bl, hp, &hn, v); + if (r != Z_OK || *bl == 0) + { + if (r == Z_DATA_ERROR) + z->msg = (char*)"oversubscribed literal/length tree"; + else if (r != Z_MEM_ERROR) + { + z->msg = (char*)"incomplete literal/length tree"; + r = Z_DATA_ERROR; + } + ZFREE(z, v); + return r; + } + + /* build distance tree */ + r = huft_build(c + nl, nd, 0, cpdist, cpdext, td, bd, hp, &hn, v); + if (r != Z_OK || (*bd == 0 && nl > 257)) + { + if (r == Z_DATA_ERROR) + z->msg = (char*)"oversubscribed distance tree"; + else if (r == Z_BUF_ERROR) { +#ifdef PKZIP_BUG_WORKAROUND + r = Z_OK; + } +#else + z->msg = (char*)"incomplete distance tree"; + r = Z_DATA_ERROR; + } + else if (r != Z_MEM_ERROR) + { + z->msg = (char*)"empty distance tree with lengths"; + r = Z_DATA_ERROR; + } + ZFREE(z, v); + return r; +#endif + } + + /* done */ + ZFREE(z, v); + return Z_OK; +} + + +/* build fixed tables only once--keep them here */ +#ifdef BUILDFIXED +local int fixed_built = 0; +#define FIXEDH 544 /* number of hufts used by fixed tables */ +local inflate_huft fixed_mem[FIXEDH]; +local uInt fixed_bl; +local uInt fixed_bd; +local inflate_huft *fixed_tl; +local inflate_huft *fixed_td; +#else +#include "inffixed.h" +#endif + + +int inflate_trees_fixed(bl, bd, tl, td, z) +uIntf *bl; /* literal desired/actual bit depth */ +uIntf *bd; /* distance desired/actual bit depth */ +inflate_huft * FAR *tl; /* literal/length tree result */ +inflate_huft * FAR *td; /* distance tree result */ +z_streamp z; /* for memory allocation */ +{ +#ifdef BUILDFIXED + /* build fixed tables if not already */ + if (!fixed_built) + { + int k; /* temporary variable */ + uInt f = 0; /* number of hufts used in fixed_mem */ + uIntf *c; /* length list for huft_build */ + uIntf *v; /* work area for huft_build */ + + /* allocate memory */ + if ((c = (uIntf*)ZALLOC(z, 288, sizeof(uInt))) == Z_NULL) + return Z_MEM_ERROR; + if ((v = (uIntf*)ZALLOC(z, 288, sizeof(uInt))) == Z_NULL) + { + ZFREE(z, c); + return Z_MEM_ERROR; + } + + /* literal table */ + for (k = 0; k < 144; k++) + c[k] = 8; + for (; k < 256; k++) + c[k] = 9; + for (; k < 280; k++) + c[k] = 7; + for (; k < 288; k++) + c[k] = 8; + fixed_bl = 9; + huft_build(c, 288, 257, cplens, cplext, &fixed_tl, &fixed_bl, + fixed_mem, &f, v); + + /* distance table */ + for (k = 0; k < 30; k++) + c[k] = 5; + fixed_bd = 5; + huft_build(c, 30, 0, cpdist, cpdext, &fixed_td, &fixed_bd, + fixed_mem, &f, v); + + /* done */ + ZFREE(z, v); + ZFREE(z, c); + fixed_built = 1; + } +#endif + *bl = fixed_bl; + *bd = fixed_bd; + *tl = fixed_tl; + *td = fixed_td; + return Z_OK; +} diff --git a/src/utilfuns/zlib/inftrees.h b/src/utilfuns/zlib/inftrees.h new file mode 100644 index 0000000..04b73b7 --- /dev/null +++ b/src/utilfuns/zlib/inftrees.h @@ -0,0 +1,58 @@ +/* inftrees.h -- header to use inftrees.c + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* Huffman code lookup table entry--this entry is four bytes for machines + that have 16-bit pointers (e.g. PC's in the small or medium model). */ + +typedef struct inflate_huft_s FAR inflate_huft; + +struct inflate_huft_s { + union { + struct { + Byte Exop; /* number of extra bits or operation */ + Byte Bits; /* number of bits in this code or subcode */ + } what; + uInt pad; /* pad structure to a power of 2 (4 bytes for */ + } word; /* 16-bit, 8 bytes for 32-bit int's) */ + uInt base; /* literal, length base, distance base, + or table offset */ +}; + +/* Maximum size of dynamic tree. The maximum found in a long but non- + exhaustive search was 1004 huft structures (850 for length/literals + and 154 for distances, the latter actually the result of an + exhaustive search). The actual maximum is not known, but the + value below is more than safe. */ +#define MANY 1440 + +extern int inflate_trees_bits OF(( + uIntf *, /* 19 code lengths */ + uIntf *, /* bits tree desired/actual depth */ + inflate_huft * FAR *, /* bits tree result */ + inflate_huft *, /* space for trees */ + z_streamp)); /* for messages */ + +extern int inflate_trees_dynamic OF(( + uInt, /* number of literal/length codes */ + uInt, /* number of distance codes */ + uIntf *, /* that many (total) code lengths */ + uIntf *, /* literal desired/actual bit depth */ + uIntf *, /* distance desired/actual bit depth */ + inflate_huft * FAR *, /* literal/length tree result */ + inflate_huft * FAR *, /* distance tree result */ + inflate_huft *, /* space for trees */ + z_streamp)); /* for messages */ + +extern int inflate_trees_fixed OF(( + uIntf *, /* literal desired/actual bit depth */ + uIntf *, /* distance desired/actual bit depth */ + inflate_huft * FAR *, /* literal/length tree result */ + inflate_huft * FAR *, /* distance tree result */ + z_streamp)); /* for memory allocation */ diff --git a/src/utilfuns/zlib/infutil.c b/src/utilfuns/zlib/infutil.c new file mode 100644 index 0000000..9a07622 --- /dev/null +++ b/src/utilfuns/zlib/infutil.c @@ -0,0 +1,87 @@ +/* inflate_util.c -- data and routines common to blocks and codes + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +#include "zutil.h" +#include "infblock.h" +#include "inftrees.h" +#include "infcodes.h" +#include "infutil.h" + +struct inflate_codes_state {int dummy;}; /* for buggy compilers */ + +/* And'ing with mask[n] masks the lower n bits */ +uInt inflate_mask[17] = { + 0x0000, + 0x0001, 0x0003, 0x0007, 0x000f, 0x001f, 0x003f, 0x007f, 0x00ff, + 0x01ff, 0x03ff, 0x07ff, 0x0fff, 0x1fff, 0x3fff, 0x7fff, 0xffff +}; + + +/* copy as much as possible from the sliding window to the output area */ +int inflate_flush(s, z, r) +inflate_blocks_statef *s; +z_streamp z; +int r; +{ + uInt n; + Bytef *p; + Bytef *q; + + /* local copies of source and destination pointers */ + p = z->next_out; + q = s->read; + + /* compute number of bytes to copy as far as end of window */ + n = (uInt)((q <= s->write ? s->write : s->end) - q); + if (n > z->avail_out) n = z->avail_out; + if (n && r == Z_BUF_ERROR) r = Z_OK; + + /* update counters */ + z->avail_out -= n; + z->total_out += n; + + /* update check information */ + if (s->checkfn != Z_NULL) + z->adler = s->check = (*s->checkfn)(s->check, q, n); + + /* copy as far as end of window */ + zmemcpy(p, q, n); + p += n; + q += n; + + /* see if more to copy at beginning of window */ + if (q == s->end) + { + /* wrap pointers */ + q = s->window; + if (s->write == s->end) + s->write = s->window; + + /* compute bytes to copy */ + n = (uInt)(s->write - q); + if (n > z->avail_out) n = z->avail_out; + if (n && r == Z_BUF_ERROR) r = Z_OK; + + /* update counters */ + z->avail_out -= n; + z->total_out += n; + + /* update check information */ + if (s->checkfn != Z_NULL) + z->adler = s->check = (*s->checkfn)(s->check, q, n); + + /* copy */ + zmemcpy(p, q, n); + p += n; + q += n; + } + + /* update pointers */ + z->next_out = p; + s->read = q; + + /* done */ + return r; +} diff --git a/src/utilfuns/zlib/infutil.h b/src/utilfuns/zlib/infutil.h new file mode 100644 index 0000000..4401df8 --- /dev/null +++ b/src/utilfuns/zlib/infutil.h @@ -0,0 +1,98 @@ +/* infutil.h -- types and macros common to blocks and codes + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +#ifndef _INFUTIL_H +#define _INFUTIL_H + +typedef enum { + TYPE, /* get type bits (3, including end bit) */ + LENS, /* get lengths for stored */ + STORED, /* processing stored block */ + TABLE, /* get table lengths */ + BTREE, /* get bit lengths tree for a dynamic block */ + DTREE, /* get length, distance trees for a dynamic block */ + CODES, /* processing fixed or dynamic block */ + DRY, /* output remaining window bytes */ + DONE, /* finished last block, done */ + BAD} /* got a data error--stuck here */ +inflate_block_mode; + +/* inflate blocks semi-private state */ +struct inflate_blocks_state { + + /* mode */ + inflate_block_mode mode; /* current inflate_block mode */ + + /* mode dependent information */ + union { + uInt left; /* if STORED, bytes left to copy */ + struct { + uInt table; /* table lengths (14 bits) */ + uInt index; /* index into blens (or border) */ + uIntf *blens; /* bit lengths of codes */ + uInt bb; /* bit length tree depth */ + inflate_huft *tb; /* bit length decoding tree */ + } trees; /* if DTREE, decoding info for trees */ + struct { + inflate_codes_statef + *codes; + } decode; /* if CODES, current state */ + } sub; /* submode */ + uInt last; /* true if this block is the last block */ + + /* mode independent information */ + uInt bitk; /* bits in bit buffer */ + uLong bitb; /* bit buffer */ + inflate_huft *hufts; /* single malloc for tree space */ + Bytef *window; /* sliding window */ + Bytef *end; /* one byte after sliding window */ + Bytef *read; /* window read pointer */ + Bytef *write; /* window write pointer */ + check_func checkfn; /* check function */ + uLong check; /* check on output */ + +}; + + +/* defines for inflate input/output */ +/* update pointers and return */ +#define UPDBITS {s->bitb=b;s->bitk=k;} +#define UPDIN {z->avail_in=n;z->total_in+=p-z->next_in;z->next_in=p;} +#define UPDOUT {s->write=q;} +#define UPDATE {UPDBITS UPDIN UPDOUT} +#define LEAVE {UPDATE return inflate_flush(s,z,r);} +/* get bytes and bits */ +#define LOADIN {p=z->next_in;n=z->avail_in;b=s->bitb;k=s->bitk;} +#define NEEDBYTE {if(n)r=Z_OK;else LEAVE} +#define NEXTBYTE (n--,*p++) +#define NEEDBITS(j) {while(k<(j)){NEEDBYTE;b|=((uLong)NEXTBYTE)<<k;k+=8;}} +#define DUMPBITS(j) {b>>=(j);k-=(j);} +/* output bytes */ +#define WAVAIL (uInt)(q<s->read?s->read-q-1:s->end-q) +#define LOADOUT {q=s->write;m=(uInt)WAVAIL;} +#define WRAP {if(q==s->end&&s->read!=s->window){q=s->window;m=(uInt)WAVAIL;}} +#define FLUSH {UPDOUT r=inflate_flush(s,z,r); LOADOUT} +#define NEEDOUT {if(m==0){WRAP if(m==0){FLUSH WRAP if(m==0) LEAVE}}r=Z_OK;} +#define OUTBYTE(a) {*q++=(Byte)(a);m--;} +/* load local pointers */ +#define LOAD {LOADIN LOADOUT} + +/* masks for lower bits (size given to avoid silly warnings with Visual C++) */ +extern uInt inflate_mask[17]; + +/* copy as much as possible from the sliding window to the output area */ +extern int inflate_flush OF(( + inflate_blocks_statef *, + z_streamp , + int)); + +struct internal_state {int dummy;}; /* for buggy compilers */ + +#endif diff --git a/src/utilfuns/zlib/maketree.c b/src/utilfuns/zlib/maketree.c new file mode 100644 index 0000000..a16d4b1 --- /dev/null +++ b/src/utilfuns/zlib/maketree.c @@ -0,0 +1,85 @@ +/* maketree.c -- make inffixed.h table for decoding fixed codes + * Copyright (C) 1995-2002 Mark Adler + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* This program is included in the distribution for completeness. + You do not need to compile or run this program since inffixed.h + is already included in the distribution. To use this program + you need to compile zlib with BUILDFIXED defined and then compile + and link this program with the zlib library. Then the output of + this program can be piped to inffixed.h. */ + +#include <stdio.h> +#include <stdlib.h> +#include "zutil.h" +#include "inftrees.h" + +/* simplify the use of the inflate_huft type with some defines */ +#define exop word.what.Exop +#define bits word.what.Bits + +/* generate initialization table for an inflate_huft structure array */ +void maketree(uInt b, inflate_huft *t) +{ + int i, e; + + i = 0; + while (1) + { + e = t[i].exop; + if (e && (e & (16+64)) == 0) /* table pointer */ + { + fprintf(stderr, "maketree: cannot initialize sub-tables!\n"); + exit(1); + } + if (i % 4 == 0) + printf("\n "); + printf(" {{{%u,%u}},%u}", t[i].exop, t[i].bits, t[i].base); + if (++i == (1<<b)) + break; + putchar(','); + } + puts(""); +} + +/* create the fixed tables in C initialization syntax */ +void main(void) +{ + int r; + uInt bl, bd; + inflate_huft *tl, *td; + z_stream z; + + z.zalloc = zcalloc; + z.opaque = (voidpf)0; + z.zfree = zcfree; + r = inflate_trees_fixed(&bl, &bd, &tl, &td, &z); + if (r) + { + fprintf(stderr, "inflate_trees_fixed error %d\n", r); + return; + } + puts("/* inffixed.h -- table for decoding fixed codes"); + puts(" * Generated automatically by the maketree.c program"); + puts(" */"); + puts(""); + puts("/* WARNING: this file should *not* be used by applications. It is"); + puts(" part of the implementation of the compression library and is"); + puts(" subject to change. Applications should only use zlib.h."); + puts(" */"); + puts(""); + printf("local uInt fixed_bl = %d;\n", bl); + printf("local uInt fixed_bd = %d;\n", bd); + printf("local inflate_huft fixed_tl[] = {"); + maketree(bl, tl); + puts(" };"); + printf("local inflate_huft fixed_td[] = {"); + maketree(bd, td); + puts(" };"); +} diff --git a/src/utilfuns/zlib/trees.c b/src/utilfuns/zlib/trees.c new file mode 100644 index 0000000..b3117b3 --- /dev/null +++ b/src/utilfuns/zlib/trees.c @@ -0,0 +1,1214 @@ +/* trees.c -- output deflated data using Huffman coding + * Copyright (C) 1995-2002 Jean-loup Gailly + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* + * ALGORITHM + * + * The "deflation" process uses several Huffman trees. The more + * common source values are represented by shorter bit sequences. + * + * Each code tree is stored in a compressed form which is itself + * a Huffman encoding of the lengths of all the code strings (in + * ascending order by source values). The actual code strings are + * reconstructed from the lengths in the inflate process, as described + * in the deflate specification. + * + * REFERENCES + * + * Deutsch, L.P.,"'Deflate' Compressed Data Format Specification". + * Available in ftp.uu.net:/pub/archiving/zip/doc/deflate-1.1.doc + * + * Storer, James A. + * Data Compression: Methods and Theory, pp. 49-50. + * Computer Science Press, 1988. ISBN 0-7167-8156-5. + * + * Sedgewick, R. + * Algorithms, p290. + * Addison-Wesley, 1983. ISBN 0-201-06672-6. + */ + +/* @(#) $Id: trees.c 1688 2005-01-01 04:42:26Z scribe $ */ + +/* #define GEN_TREES_H */ + +#include "deflate.h" + +#ifdef DEBUG +# include <ctype.h> +#endif + +/* =========================================================================== + * Constants + */ + +#define MAX_BL_BITS 7 +/* Bit length codes must not exceed MAX_BL_BITS bits */ + +#define END_BLOCK 256 +/* end of block literal code */ + +#define REP_3_6 16 +/* repeat previous bit length 3-6 times (2 bits of repeat count) */ + +#define REPZ_3_10 17 +/* repeat a zero length 3-10 times (3 bits of repeat count) */ + +#define REPZ_11_138 18 +/* repeat a zero length 11-138 times (7 bits of repeat count) */ + +local const int extra_lbits[LENGTH_CODES] /* extra bits for each length code */ + = {0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0}; + +local const int extra_dbits[D_CODES] /* extra bits for each distance code */ + = {0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; + +local const int extra_blbits[BL_CODES]/* extra bits for each bit length code */ + = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,3,7}; + +local const uch bl_order[BL_CODES] + = {16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15}; +/* The lengths of the bit length codes are sent in order of decreasing + * probability, to avoid transmitting the lengths for unused bit length codes. + */ + +#define Buf_size (8 * 2*sizeof(char)) +/* Number of bits used within bi_buf. (bi_buf might be implemented on + * more than 16 bits on some systems.) + */ + +/* =========================================================================== + * Local data. These are initialized only once. + */ + +#define DIST_CODE_LEN 512 /* see definition of array dist_code below */ + +#if defined(GEN_TREES_H) || !defined(STDC) +/* non ANSI compilers may not accept trees.h */ + +local ct_data static_ltree[L_CODES+2]; +/* The static literal tree. Since the bit lengths are imposed, there is no + * need for the L_CODES extra codes used during heap construction. However + * The codes 286 and 287 are needed to build a canonical tree (see _tr_init + * below). + */ + +local ct_data static_dtree[D_CODES]; +/* The static distance tree. (Actually a trivial tree since all codes use + * 5 bits.) + */ + +uch _dist_code[DIST_CODE_LEN]; +/* Distance codes. The first 256 values correspond to the distances + * 3 .. 258, the last 256 values correspond to the top 8 bits of + * the 15 bit distances. + */ + +uch _length_code[MAX_MATCH-MIN_MATCH+1]; +/* length code for each normalized match length (0 == MIN_MATCH) */ + +local int base_length[LENGTH_CODES]; +/* First normalized length for each code (0 = MIN_MATCH) */ + +local int base_dist[D_CODES]; +/* First normalized distance for each code (0 = distance of 1) */ + +#else +# include "trees.h" +#endif /* GEN_TREES_H */ + +struct static_tree_desc_s { + const ct_data *static_tree; /* static tree or NULL */ + const intf *extra_bits; /* extra bits for each code or NULL */ + int extra_base; /* base index for extra_bits */ + int elems; /* max number of elements in the tree */ + int max_length; /* max bit length for the codes */ +}; + +local static_tree_desc static_l_desc = +{static_ltree, extra_lbits, LITERALS+1, L_CODES, MAX_BITS}; + +local static_tree_desc static_d_desc = +{static_dtree, extra_dbits, 0, D_CODES, MAX_BITS}; + +local static_tree_desc static_bl_desc = +{(const ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS}; + +/* =========================================================================== + * Local (static) routines in this file. + */ + +local void tr_static_init OF((void)); +local void init_block OF((deflate_state *s)); +local void pqdownheap OF((deflate_state *s, ct_data *tree, int k)); +local void gen_bitlen OF((deflate_state *s, tree_desc *desc)); +local void gen_codes OF((ct_data *tree, int max_code, ushf *bl_count)); +local void build_tree OF((deflate_state *s, tree_desc *desc)); +local void scan_tree OF((deflate_state *s, ct_data *tree, int max_code)); +local void send_tree OF((deflate_state *s, ct_data *tree, int max_code)); +local int build_bl_tree OF((deflate_state *s)); +local void send_all_trees OF((deflate_state *s, int lcodes, int dcodes, + int blcodes)); +local void compress_block OF((deflate_state *s, ct_data *ltree, + ct_data *dtree)); +local void set_data_type OF((deflate_state *s)); +local unsigned bi_reverse OF((unsigned value, int length)); +local void bi_windup OF((deflate_state *s)); +local void bi_flush OF((deflate_state *s)); +local void copy_block OF((deflate_state *s, charf *buf, unsigned len, + int header)); + +#ifdef GEN_TREES_H +local void gen_trees_header OF((void)); +#endif + +#ifndef DEBUG +# define send_code(s, c, tree) send_bits(s, tree[c].Code, tree[c].Len) + /* Send a code of the given tree. c and tree must not have side effects */ + +#else /* DEBUG */ +# define send_code(s, c, tree) \ + { if (z_verbose>2) fprintf(stderr,"\ncd %3d ",(c)); \ + send_bits(s, tree[c].Code, tree[c].Len); } +#endif + +/* =========================================================================== + * Output a short LSB first on the stream. + * IN assertion: there is enough room in pendingBuf. + */ +#define put_short(s, w) { \ + put_byte(s, (uch)((w) & 0xff)); \ + put_byte(s, (uch)((ush)(w) >> 8)); \ +} + +/* =========================================================================== + * Send a value on a given number of bits. + * IN assertion: length <= 16 and value fits in length bits. + */ +#ifdef DEBUG +local void send_bits OF((deflate_state *s, int value, int length)); + +local void send_bits(s, value, length) + deflate_state *s; + int value; /* value to send */ + int length; /* number of bits */ +{ + Tracevv((stderr," l %2d v %4x ", length, value)); + Assert(length > 0 && length <= 15, "invalid length"); + s->bits_sent += (ulg)length; + + /* If not enough room in bi_buf, use (valid) bits from bi_buf and + * (16 - bi_valid) bits from value, leaving (width - (16-bi_valid)) + * unused bits in value. + */ + if (s->bi_valid > (int)Buf_size - length) { + s->bi_buf |= (value << s->bi_valid); + put_short(s, s->bi_buf); + s->bi_buf = (ush)value >> (Buf_size - s->bi_valid); + s->bi_valid += length - Buf_size; + } else { + s->bi_buf |= value << s->bi_valid; + s->bi_valid += length; + } +} +#else /* !DEBUG */ + +#define send_bits(s, value, length) \ +{ int len = length;\ + if (s->bi_valid > (int)Buf_size - len) {\ + int val = value;\ + s->bi_buf |= (val << s->bi_valid);\ + put_short(s, s->bi_buf);\ + s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\ + s->bi_valid += len - Buf_size;\ + } else {\ + s->bi_buf |= (value) << s->bi_valid;\ + s->bi_valid += len;\ + }\ +} +#endif /* DEBUG */ + + +#define MAX(a,b) (a >= b ? a : b) +/* the arguments must not have side effects */ + +/* =========================================================================== + * Initialize the various 'constant' tables. + */ +local void tr_static_init() +{ +#if defined(GEN_TREES_H) || !defined(STDC) + static int static_init_done = 0; + int n; /* iterates over tree elements */ + int bits; /* bit counter */ + int length; /* length value */ + int code; /* code value */ + int dist; /* distance index */ + ush bl_count[MAX_BITS+1]; + /* number of codes at each bit length for an optimal tree */ + + if (static_init_done) return; + + /* For some embedded targets, global variables are not initialized: */ + static_l_desc.static_tree = static_ltree; + static_l_desc.extra_bits = extra_lbits; + static_d_desc.static_tree = static_dtree; + static_d_desc.extra_bits = extra_dbits; + static_bl_desc.extra_bits = extra_blbits; + + /* Initialize the mapping length (0..255) -> length code (0..28) */ + length = 0; + for (code = 0; code < LENGTH_CODES-1; code++) { + base_length[code] = length; + for (n = 0; n < (1<<extra_lbits[code]); n++) { + _length_code[length++] = (uch)code; + } + } + Assert (length == 256, "tr_static_init: length != 256"); + /* Note that the length 255 (match length 258) can be represented + * in two different ways: code 284 + 5 bits or code 285, so we + * overwrite length_code[255] to use the best encoding: + */ + _length_code[length-1] = (uch)code; + + /* Initialize the mapping dist (0..32K) -> dist code (0..29) */ + dist = 0; + for (code = 0 ; code < 16; code++) { + base_dist[code] = dist; + for (n = 0; n < (1<<extra_dbits[code]); n++) { + _dist_code[dist++] = (uch)code; + } + } + Assert (dist == 256, "tr_static_init: dist != 256"); + dist >>= 7; /* from now on, all distances are divided by 128 */ + for ( ; code < D_CODES; code++) { + base_dist[code] = dist << 7; + for (n = 0; n < (1<<(extra_dbits[code]-7)); n++) { + _dist_code[256 + dist++] = (uch)code; + } + } + Assert (dist == 256, "tr_static_init: 256+dist != 512"); + + /* Construct the codes of the static literal tree */ + for (bits = 0; bits <= MAX_BITS; bits++) bl_count[bits] = 0; + n = 0; + while (n <= 143) static_ltree[n++].Len = 8, bl_count[8]++; + while (n <= 255) static_ltree[n++].Len = 9, bl_count[9]++; + while (n <= 279) static_ltree[n++].Len = 7, bl_count[7]++; + while (n <= 287) static_ltree[n++].Len = 8, bl_count[8]++; + /* Codes 286 and 287 do not exist, but we must include them in the + * tree construction to get a canonical Huffman tree (longest code + * all ones) + */ + gen_codes((ct_data *)static_ltree, L_CODES+1, bl_count); + + /* The static distance tree is trivial: */ + for (n = 0; n < D_CODES; n++) { + static_dtree[n].Len = 5; + static_dtree[n].Code = bi_reverse((unsigned)n, 5); + } + static_init_done = 1; + +# ifdef GEN_TREES_H + gen_trees_header(); +# endif +#endif /* defined(GEN_TREES_H) || !defined(STDC) */ +} + +/* =========================================================================== + * Genererate the file trees.h describing the static trees. + */ +#ifdef GEN_TREES_H +# ifndef DEBUG +# include <stdio.h> +# endif + +# define SEPARATOR(i, last, width) \ + ((i) == (last)? "\n};\n\n" : \ + ((i) % (width) == (width)-1 ? ",\n" : ", ")) + +void gen_trees_header() +{ + FILE *header = fopen("trees.h", "w"); + int i; + + Assert (header != NULL, "Can't open trees.h"); + fprintf(header, + "/* header created automatically with -DGEN_TREES_H */\n\n"); + + fprintf(header, "local const ct_data static_ltree[L_CODES+2] = {\n"); + for (i = 0; i < L_CODES+2; i++) { + fprintf(header, "{{%3u},{%3u}}%s", static_ltree[i].Code, + static_ltree[i].Len, SEPARATOR(i, L_CODES+1, 5)); + } + + fprintf(header, "local const ct_data static_dtree[D_CODES] = {\n"); + for (i = 0; i < D_CODES; i++) { + fprintf(header, "{{%2u},{%2u}}%s", static_dtree[i].Code, + static_dtree[i].Len, SEPARATOR(i, D_CODES-1, 5)); + } + + fprintf(header, "const uch _dist_code[DIST_CODE_LEN] = {\n"); + for (i = 0; i < DIST_CODE_LEN; i++) { + fprintf(header, "%2u%s", _dist_code[i], + SEPARATOR(i, DIST_CODE_LEN-1, 20)); + } + + fprintf(header, "const uch _length_code[MAX_MATCH-MIN_MATCH+1]= {\n"); + for (i = 0; i < MAX_MATCH-MIN_MATCH+1; i++) { + fprintf(header, "%2u%s", _length_code[i], + SEPARATOR(i, MAX_MATCH-MIN_MATCH, 20)); + } + + fprintf(header, "local const int base_length[LENGTH_CODES] = {\n"); + for (i = 0; i < LENGTH_CODES; i++) { + fprintf(header, "%1u%s", base_length[i], + SEPARATOR(i, LENGTH_CODES-1, 20)); + } + + fprintf(header, "local const int base_dist[D_CODES] = {\n"); + for (i = 0; i < D_CODES; i++) { + fprintf(header, "%5u%s", base_dist[i], + SEPARATOR(i, D_CODES-1, 10)); + } + + fclose(header); +} +#endif /* GEN_TREES_H */ + +/* =========================================================================== + * Initialize the tree data structures for a new zlib stream. + */ +void _tr_init(s) + deflate_state *s; +{ + tr_static_init(); + + s->l_desc.dyn_tree = s->dyn_ltree; + s->l_desc.stat_desc = &static_l_desc; + + s->d_desc.dyn_tree = s->dyn_dtree; + s->d_desc.stat_desc = &static_d_desc; + + s->bl_desc.dyn_tree = s->bl_tree; + s->bl_desc.stat_desc = &static_bl_desc; + + s->bi_buf = 0; + s->bi_valid = 0; + s->last_eob_len = 8; /* enough lookahead for inflate */ +#ifdef DEBUG + s->compressed_len = 0L; + s->bits_sent = 0L; +#endif + + /* Initialize the first block of the first file: */ + init_block(s); +} + +/* =========================================================================== + * Initialize a new block. + */ +local void init_block(s) + deflate_state *s; +{ + int n; /* iterates over tree elements */ + + /* Initialize the trees. */ + for (n = 0; n < L_CODES; n++) s->dyn_ltree[n].Freq = 0; + for (n = 0; n < D_CODES; n++) s->dyn_dtree[n].Freq = 0; + for (n = 0; n < BL_CODES; n++) s->bl_tree[n].Freq = 0; + + s->dyn_ltree[END_BLOCK].Freq = 1; + s->opt_len = s->static_len = 0L; + s->last_lit = s->matches = 0; +} + +#define SMALLEST 1 +/* Index within the heap array of least frequent node in the Huffman tree */ + + +/* =========================================================================== + * Remove the smallest element from the heap and recreate the heap with + * one less element. Updates heap and heap_len. + */ +#define pqremove(s, tree, top) \ +{\ + top = s->heap[SMALLEST]; \ + s->heap[SMALLEST] = s->heap[s->heap_len--]; \ + pqdownheap(s, tree, SMALLEST); \ +} + +/* =========================================================================== + * Compares to subtrees, using the tree depth as tie breaker when + * the subtrees have equal frequency. This minimizes the worst case length. + */ +#define smaller(tree, n, m, depth) \ + (tree[n].Freq < tree[m].Freq || \ + (tree[n].Freq == tree[m].Freq && depth[n] <= depth[m])) + +/* =========================================================================== + * Restore the heap property by moving down the tree starting at node k, + * exchanging a node with the smallest of its two sons if necessary, stopping + * when the heap property is re-established (each father smaller than its + * two sons). + */ +local void pqdownheap(s, tree, k) + deflate_state *s; + ct_data *tree; /* the tree to restore */ + int k; /* node to move down */ +{ + int v = s->heap[k]; + int j = k << 1; /* left son of k */ + while (j <= s->heap_len) { + /* Set j to the smallest of the two sons: */ + if (j < s->heap_len && + smaller(tree, s->heap[j+1], s->heap[j], s->depth)) { + j++; + } + /* Exit if v is smaller than both sons */ + if (smaller(tree, v, s->heap[j], s->depth)) break; + + /* Exchange v with the smallest son */ + s->heap[k] = s->heap[j]; k = j; + + /* And continue down the tree, setting j to the left son of k */ + j <<= 1; + } + s->heap[k] = v; +} + +/* =========================================================================== + * Compute the optimal bit lengths for a tree and update the total bit length + * for the current block. + * IN assertion: the fields freq and dad are set, heap[heap_max] and + * above are the tree nodes sorted by increasing frequency. + * OUT assertions: the field len is set to the optimal bit length, the + * array bl_count contains the frequencies for each bit length. + * The length opt_len is updated; static_len is also updated if stree is + * not null. + */ +local void gen_bitlen(s, desc) + deflate_state *s; + tree_desc *desc; /* the tree descriptor */ +{ + ct_data *tree = desc->dyn_tree; + int max_code = desc->max_code; + const ct_data *stree = desc->stat_desc->static_tree; + const intf *extra = desc->stat_desc->extra_bits; + int base = desc->stat_desc->extra_base; + int max_length = desc->stat_desc->max_length; + int h; /* heap index */ + int n, m; /* iterate over the tree elements */ + int bits; /* bit length */ + int xbits; /* extra bits */ + ush f; /* frequency */ + int overflow = 0; /* number of elements with bit length too large */ + + for (bits = 0; bits <= MAX_BITS; bits++) s->bl_count[bits] = 0; + + /* In a first pass, compute the optimal bit lengths (which may + * overflow in the case of the bit length tree). + */ + tree[s->heap[s->heap_max]].Len = 0; /* root of the heap */ + + for (h = s->heap_max+1; h < HEAP_SIZE; h++) { + n = s->heap[h]; + bits = tree[tree[n].Dad].Len + 1; + if (bits > max_length) bits = max_length, overflow++; + tree[n].Len = (ush)bits; + /* We overwrite tree[n].Dad which is no longer needed */ + + if (n > max_code) continue; /* not a leaf node */ + + s->bl_count[bits]++; + xbits = 0; + if (n >= base) xbits = extra[n-base]; + f = tree[n].Freq; + s->opt_len += (ulg)f * (bits + xbits); + if (stree) s->static_len += (ulg)f * (stree[n].Len + xbits); + } + if (overflow == 0) return; + + Trace((stderr,"\nbit length overflow\n")); + /* This happens for example on obj2 and pic of the Calgary corpus */ + + /* Find the first bit length which could increase: */ + do { + bits = max_length-1; + while (s->bl_count[bits] == 0) bits--; + s->bl_count[bits]--; /* move one leaf down the tree */ + s->bl_count[bits+1] += 2; /* move one overflow item as its brother */ + s->bl_count[max_length]--; + /* The brother of the overflow item also moves one step up, + * but this does not affect bl_count[max_length] + */ + overflow -= 2; + } while (overflow > 0); + + /* Now recompute all bit lengths, scanning in increasing frequency. + * h is still equal to HEAP_SIZE. (It is simpler to reconstruct all + * lengths instead of fixing only the wrong ones. This idea is taken + * from 'ar' written by Haruhiko Okumura.) + */ + for (bits = max_length; bits != 0; bits--) { + n = s->bl_count[bits]; + while (n != 0) { + m = s->heap[--h]; + if (m > max_code) continue; + if (tree[m].Len != (unsigned) bits) { + Trace((stderr,"code %d bits %d->%d\n", m, tree[m].Len, bits)); + s->opt_len += ((long)bits - (long)tree[m].Len) + *(long)tree[m].Freq; + tree[m].Len = (ush)bits; + } + n--; + } + } +} + +/* =========================================================================== + * Generate the codes for a given tree and bit counts (which need not be + * optimal). + * IN assertion: the array bl_count contains the bit length statistics for + * the given tree and the field len is set for all tree elements. + * OUT assertion: the field code is set for all tree elements of non + * zero code length. + */ +local void gen_codes (tree, max_code, bl_count) + ct_data *tree; /* the tree to decorate */ + int max_code; /* largest code with non zero frequency */ + ushf *bl_count; /* number of codes at each bit length */ +{ + ush next_code[MAX_BITS+1]; /* next code value for each bit length */ + ush code = 0; /* running code value */ + int bits; /* bit index */ + int n; /* code index */ + + /* The distribution counts are first used to generate the code values + * without bit reversal. + */ + for (bits = 1; bits <= MAX_BITS; bits++) { + next_code[bits] = code = (code + bl_count[bits-1]) << 1; + } + /* Check that the bit counts in bl_count are consistent. The last code + * must be all ones. + */ + Assert (code + bl_count[MAX_BITS]-1 == (1<<MAX_BITS)-1, + "inconsistent bit counts"); + Tracev((stderr,"\ngen_codes: max_code %d ", max_code)); + + for (n = 0; n <= max_code; n++) { + int len = tree[n].Len; + if (len == 0) continue; + /* Now reverse the bits */ + tree[n].Code = bi_reverse(next_code[len]++, len); + + Tracecv(tree != static_ltree, (stderr,"\nn %3d %c l %2d c %4x (%x) ", + n, (isgraph(n) ? n : ' '), len, tree[n].Code, next_code[len]-1)); + } +} + +/* =========================================================================== + * Construct one Huffman tree and assigns the code bit strings and lengths. + * Update the total bit length for the current block. + * IN assertion: the field freq is set for all tree elements. + * OUT assertions: the fields len and code are set to the optimal bit length + * and corresponding code. The length opt_len is updated; static_len is + * also updated if stree is not null. The field max_code is set. + */ +local void build_tree(s, desc) + deflate_state *s; + tree_desc *desc; /* the tree descriptor */ +{ + ct_data *tree = desc->dyn_tree; + const ct_data *stree = desc->stat_desc->static_tree; + int elems = desc->stat_desc->elems; + int n, m; /* iterate over heap elements */ + int max_code = -1; /* largest code with non zero frequency */ + int node; /* new node being created */ + + /* Construct the initial heap, with least frequent element in + * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n+1]. + * heap[0] is not used. + */ + s->heap_len = 0, s->heap_max = HEAP_SIZE; + + for (n = 0; n < elems; n++) { + if (tree[n].Freq != 0) { + s->heap[++(s->heap_len)] = max_code = n; + s->depth[n] = 0; + } else { + tree[n].Len = 0; + } + } + + /* The pkzip format requires that at least one distance code exists, + * and that at least one bit should be sent even if there is only one + * possible code. So to avoid special checks later on we force at least + * two codes of non zero frequency. + */ + while (s->heap_len < 2) { + node = s->heap[++(s->heap_len)] = (max_code < 2 ? ++max_code : 0); + tree[node].Freq = 1; + s->depth[node] = 0; + s->opt_len--; if (stree) s->static_len -= stree[node].Len; + /* node is 0 or 1 so it does not have extra bits */ + } + desc->max_code = max_code; + + /* The elements heap[heap_len/2+1 .. heap_len] are leaves of the tree, + * establish sub-heaps of increasing lengths: + */ + for (n = s->heap_len/2; n >= 1; n--) pqdownheap(s, tree, n); + + /* Construct the Huffman tree by repeatedly combining the least two + * frequent nodes. + */ + node = elems; /* next internal node of the tree */ + do { + pqremove(s, tree, n); /* n = node of least frequency */ + m = s->heap[SMALLEST]; /* m = node of next least frequency */ + + s->heap[--(s->heap_max)] = n; /* keep the nodes sorted by frequency */ + s->heap[--(s->heap_max)] = m; + + /* Create a new node father of n and m */ + tree[node].Freq = tree[n].Freq + tree[m].Freq; + s->depth[node] = (uch) (MAX(s->depth[n], s->depth[m]) + 1); + tree[n].Dad = tree[m].Dad = (ush)node; +#ifdef DUMP_BL_TREE + if (tree == s->bl_tree) { + fprintf(stderr,"\nnode %d(%d), sons %d(%d) %d(%d)", + node, tree[node].Freq, n, tree[n].Freq, m, tree[m].Freq); + } +#endif + /* and insert the new node in the heap */ + s->heap[SMALLEST] = node++; + pqdownheap(s, tree, SMALLEST); + + } while (s->heap_len >= 2); + + s->heap[--(s->heap_max)] = s->heap[SMALLEST]; + + /* At this point, the fields freq and dad are set. We can now + * generate the bit lengths. + */ + gen_bitlen(s, (tree_desc *)desc); + + /* The field len is now set, we can generate the bit codes */ + gen_codes ((ct_data *)tree, max_code, s->bl_count); +} + +/* =========================================================================== + * Scan a literal or distance tree to determine the frequencies of the codes + * in the bit length tree. + */ +local void scan_tree (s, tree, max_code) + deflate_state *s; + ct_data *tree; /* the tree to be scanned */ + int max_code; /* and its largest code of non zero frequency */ +{ + int n; /* iterates over all tree elements */ + int prevlen = -1; /* last emitted length */ + int curlen; /* length of current code */ + int nextlen = tree[0].Len; /* length of next code */ + int count = 0; /* repeat count of the current code */ + int max_count = 7; /* max repeat count */ + int min_count = 4; /* min repeat count */ + + if (nextlen == 0) max_count = 138, min_count = 3; + tree[max_code+1].Len = (ush)0xffff; /* guard */ + + for (n = 0; n <= max_code; n++) { + curlen = nextlen; nextlen = tree[n+1].Len; + if (++count < max_count && curlen == nextlen) { + continue; + } else if (count < min_count) { + s->bl_tree[curlen].Freq += count; + } else if (curlen != 0) { + if (curlen != prevlen) s->bl_tree[curlen].Freq++; + s->bl_tree[REP_3_6].Freq++; + } else if (count <= 10) { + s->bl_tree[REPZ_3_10].Freq++; + } else { + s->bl_tree[REPZ_11_138].Freq++; + } + count = 0; prevlen = curlen; + if (nextlen == 0) { + max_count = 138, min_count = 3; + } else if (curlen == nextlen) { + max_count = 6, min_count = 3; + } else { + max_count = 7, min_count = 4; + } + } +} + +/* =========================================================================== + * Send a literal or distance tree in compressed form, using the codes in + * bl_tree. + */ +local void send_tree (s, tree, max_code) + deflate_state *s; + ct_data *tree; /* the tree to be scanned */ + int max_code; /* and its largest code of non zero frequency */ +{ + int n; /* iterates over all tree elements */ + int prevlen = -1; /* last emitted length */ + int curlen; /* length of current code */ + int nextlen = tree[0].Len; /* length of next code */ + int count = 0; /* repeat count of the current code */ + int max_count = 7; /* max repeat count */ + int min_count = 4; /* min repeat count */ + + /* tree[max_code+1].Len = -1; */ /* guard already set */ + if (nextlen == 0) max_count = 138, min_count = 3; + + for (n = 0; n <= max_code; n++) { + curlen = nextlen; nextlen = tree[n+1].Len; + if (++count < max_count && curlen == nextlen) { + continue; + } else if (count < min_count) { + do { send_code(s, curlen, s->bl_tree); } while (--count != 0); + + } else if (curlen != 0) { + if (curlen != prevlen) { + send_code(s, curlen, s->bl_tree); count--; + } + Assert(count >= 3 && count <= 6, " 3_6?"); + send_code(s, REP_3_6, s->bl_tree); send_bits(s, count-3, 2); + + } else if (count <= 10) { + send_code(s, REPZ_3_10, s->bl_tree); send_bits(s, count-3, 3); + + } else { + send_code(s, REPZ_11_138, s->bl_tree); send_bits(s, count-11, 7); + } + count = 0; prevlen = curlen; + if (nextlen == 0) { + max_count = 138, min_count = 3; + } else if (curlen == nextlen) { + max_count = 6, min_count = 3; + } else { + max_count = 7, min_count = 4; + } + } +} + +/* =========================================================================== + * Construct the Huffman tree for the bit lengths and return the index in + * bl_order of the last bit length code to send. + */ +local int build_bl_tree(s) + deflate_state *s; +{ + int max_blindex; /* index of last bit length code of non zero freq */ + + /* Determine the bit length frequencies for literal and distance trees */ + scan_tree(s, (ct_data *)s->dyn_ltree, s->l_desc.max_code); + scan_tree(s, (ct_data *)s->dyn_dtree, s->d_desc.max_code); + + /* Build the bit length tree: */ + build_tree(s, (tree_desc *)(&(s->bl_desc))); + /* opt_len now includes the length of the tree representations, except + * the lengths of the bit lengths codes and the 5+5+4 bits for the counts. + */ + + /* Determine the number of bit length codes to send. The pkzip format + * requires that at least 4 bit length codes be sent. (appnote.txt says + * 3 but the actual value used is 4.) + */ + for (max_blindex = BL_CODES-1; max_blindex >= 3; max_blindex--) { + if (s->bl_tree[bl_order[max_blindex]].Len != 0) break; + } + /* Update opt_len to include the bit length tree and counts */ + s->opt_len += 3*(max_blindex+1) + 5+5+4; + Tracev((stderr, "\ndyn trees: dyn %ld, stat %ld", + s->opt_len, s->static_len)); + + return max_blindex; +} + +/* =========================================================================== + * Send the header for a block using dynamic Huffman trees: the counts, the + * lengths of the bit length codes, the literal tree and the distance tree. + * IN assertion: lcodes >= 257, dcodes >= 1, blcodes >= 4. + */ +local void send_all_trees(s, lcodes, dcodes, blcodes) + deflate_state *s; + int lcodes, dcodes, blcodes; /* number of codes for each tree */ +{ + int rank; /* index in bl_order */ + + Assert (lcodes >= 257 && dcodes >= 1 && blcodes >= 4, "not enough codes"); + Assert (lcodes <= L_CODES && dcodes <= D_CODES && blcodes <= BL_CODES, + "too many codes"); + Tracev((stderr, "\nbl counts: ")); + send_bits(s, lcodes-257, 5); /* not +255 as stated in appnote.txt */ + send_bits(s, dcodes-1, 5); + send_bits(s, blcodes-4, 4); /* not -3 as stated in appnote.txt */ + for (rank = 0; rank < blcodes; rank++) { + Tracev((stderr, "\nbl code %2d ", bl_order[rank])); + send_bits(s, s->bl_tree[bl_order[rank]].Len, 3); + } + Tracev((stderr, "\nbl tree: sent %ld", s->bits_sent)); + + send_tree(s, (ct_data *)s->dyn_ltree, lcodes-1); /* literal tree */ + Tracev((stderr, "\nlit tree: sent %ld", s->bits_sent)); + + send_tree(s, (ct_data *)s->dyn_dtree, dcodes-1); /* distance tree */ + Tracev((stderr, "\ndist tree: sent %ld", s->bits_sent)); +} + +/* =========================================================================== + * Send a stored block + */ +void _tr_stored_block(s, buf, stored_len, eof) + deflate_state *s; + charf *buf; /* input block */ + ulg stored_len; /* length of input block */ + int eof; /* true if this is the last block for a file */ +{ + send_bits(s, (STORED_BLOCK<<1)+eof, 3); /* send block type */ +#ifdef DEBUG + s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L; + s->compressed_len += (stored_len + 4) << 3; +#endif + copy_block(s, buf, (unsigned)stored_len, 1); /* with header */ +} + +/* =========================================================================== + * Send one empty static block to give enough lookahead for inflate. + * This takes 10 bits, of which 7 may remain in the bit buffer. + * The current inflate code requires 9 bits of lookahead. If the + * last two codes for the previous block (real code plus EOB) were coded + * on 5 bits or less, inflate may have only 5+3 bits of lookahead to decode + * the last real code. In this case we send two empty static blocks instead + * of one. (There are no problems if the previous block is stored or fixed.) + * To simplify the code, we assume the worst case of last real code encoded + * on one bit only. + */ +void _tr_align(s) + deflate_state *s; +{ + send_bits(s, STATIC_TREES<<1, 3); + send_code(s, END_BLOCK, static_ltree); +#ifdef DEBUG + s->compressed_len += 10L; /* 3 for block type, 7 for EOB */ +#endif + bi_flush(s); + /* Of the 10 bits for the empty block, we have already sent + * (10 - bi_valid) bits. The lookahead for the last real code (before + * the EOB of the previous block) was thus at least one plus the length + * of the EOB plus what we have just sent of the empty static block. + */ + if (1 + s->last_eob_len + 10 - s->bi_valid < 9) { + send_bits(s, STATIC_TREES<<1, 3); + send_code(s, END_BLOCK, static_ltree); +#ifdef DEBUG + s->compressed_len += 10L; +#endif + bi_flush(s); + } + s->last_eob_len = 7; +} + +/* =========================================================================== + * Determine the best encoding for the current block: dynamic trees, static + * trees or store, and output the encoded block to the zip file. + */ +void _tr_flush_block(s, buf, stored_len, eof) + deflate_state *s; + charf *buf; /* input block, or NULL if too old */ + ulg stored_len; /* length of input block */ + int eof; /* true if this is the last block for a file */ +{ + ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */ + int max_blindex = 0; /* index of last bit length code of non zero freq */ + + /* Build the Huffman trees unless a stored block is forced */ + if (s->level > 0) { + + /* Check if the file is ascii or binary */ + if (s->data_type == Z_UNKNOWN) set_data_type(s); + + /* Construct the literal and distance trees */ + build_tree(s, (tree_desc *)(&(s->l_desc))); + Tracev((stderr, "\nlit data: dyn %ld, stat %ld", s->opt_len, + s->static_len)); + + build_tree(s, (tree_desc *)(&(s->d_desc))); + Tracev((stderr, "\ndist data: dyn %ld, stat %ld", s->opt_len, + s->static_len)); + /* At this point, opt_len and static_len are the total bit lengths of + * the compressed block data, excluding the tree representations. + */ + + /* Build the bit length tree for the above two trees, and get the index + * in bl_order of the last bit length code to send. + */ + max_blindex = build_bl_tree(s); + + /* Determine the best encoding. Compute first the block length in bytes*/ + opt_lenb = (s->opt_len+3+7)>>3; + static_lenb = (s->static_len+3+7)>>3; + + Tracev((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u ", + opt_lenb, s->opt_len, static_lenb, s->static_len, stored_len, + s->last_lit)); + + if (static_lenb <= opt_lenb) opt_lenb = static_lenb; + + } else { + Assert(buf != (char*)0, "lost buf"); + opt_lenb = static_lenb = stored_len + 5; /* force a stored block */ + } + +#ifdef FORCE_STORED + if (buf != (char*)0) { /* force stored block */ +#else + if (stored_len+4 <= opt_lenb && buf != (char*)0) { + /* 4: two words for the lengths */ +#endif + /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE. + * Otherwise we can't have processed more than WSIZE input bytes since + * the last block flush, because compression would have been + * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to + * transform a block into a stored block. + */ + _tr_stored_block(s, buf, stored_len, eof); + +#ifdef FORCE_STATIC + } else if (static_lenb >= 0) { /* force static trees */ +#else + } else if (static_lenb == opt_lenb) { +#endif + send_bits(s, (STATIC_TREES<<1)+eof, 3); + compress_block(s, (ct_data *)static_ltree, (ct_data *)static_dtree); +#ifdef DEBUG + s->compressed_len += 3 + s->static_len; +#endif + } else { + send_bits(s, (DYN_TREES<<1)+eof, 3); + send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1, + max_blindex+1); + compress_block(s, (ct_data *)s->dyn_ltree, (ct_data *)s->dyn_dtree); +#ifdef DEBUG + s->compressed_len += 3 + s->opt_len; +#endif + } + Assert (s->compressed_len == s->bits_sent, "bad compressed size"); + /* The above check is made mod 2^32, for files larger than 512 MB + * and uLong implemented on 32 bits. + */ + init_block(s); + + if (eof) { + bi_windup(s); +#ifdef DEBUG + s->compressed_len += 7; /* align on byte boundary */ +#endif + } + Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3, + s->compressed_len-7*eof)); +} + +/* =========================================================================== + * Save the match info and tally the frequency counts. Return true if + * the current block must be flushed. + */ +int _tr_tally (s, dist, lc) + deflate_state *s; + unsigned dist; /* distance of matched string */ + unsigned lc; /* match length-MIN_MATCH or unmatched char (if dist==0) */ +{ + s->d_buf[s->last_lit] = (ush)dist; + s->l_buf[s->last_lit++] = (uch)lc; + if (dist == 0) { + /* lc is the unmatched char */ + s->dyn_ltree[lc].Freq++; + } else { + s->matches++; + /* Here, lc is the match length - MIN_MATCH */ + dist--; /* dist = match distance - 1 */ + Assert((ush)dist < (ush)MAX_DIST(s) && + (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) && + (ush)d_code(dist) < (ush)D_CODES, "_tr_tally: bad match"); + + s->dyn_ltree[_length_code[lc]+LITERALS+1].Freq++; + s->dyn_dtree[d_code(dist)].Freq++; + } + +#ifdef TRUNCATE_BLOCK + /* Try to guess if it is profitable to stop the current block here */ + if ((s->last_lit & 0x1fff) == 0 && s->level > 2) { + /* Compute an upper bound for the compressed length */ + ulg out_length = (ulg)s->last_lit*8L; + ulg in_length = (ulg)((long)s->strstart - s->block_start); + int dcode; + for (dcode = 0; dcode < D_CODES; dcode++) { + out_length += (ulg)s->dyn_dtree[dcode].Freq * + (5L+extra_dbits[dcode]); + } + out_length >>= 3; + Tracev((stderr,"\nlast_lit %u, in %ld, out ~%ld(%ld%%) ", + s->last_lit, in_length, out_length, + 100L - out_length*100L/in_length)); + if (s->matches < s->last_lit/2 && out_length < in_length/2) return 1; + } +#endif + return (s->last_lit == s->lit_bufsize-1); + /* We avoid equality with lit_bufsize because of wraparound at 64K + * on 16 bit machines and because stored blocks are restricted to + * 64K-1 bytes. + */ +} + +/* =========================================================================== + * Send the block data compressed using the given Huffman trees + */ +local void compress_block(s, ltree, dtree) + deflate_state *s; + ct_data *ltree; /* literal tree */ + ct_data *dtree; /* distance tree */ +{ + unsigned dist; /* distance of matched string */ + int lc; /* match length or unmatched char (if dist == 0) */ + unsigned lx = 0; /* running index in l_buf */ + unsigned code; /* the code to send */ + int extra; /* number of extra bits to send */ + + if (s->last_lit != 0) do { + dist = s->d_buf[lx]; + lc = s->l_buf[lx++]; + if (dist == 0) { + send_code(s, lc, ltree); /* send a literal byte */ + Tracecv(isgraph(lc), (stderr," '%c' ", lc)); + } else { + /* Here, lc is the match length - MIN_MATCH */ + code = _length_code[lc]; + send_code(s, code+LITERALS+1, ltree); /* send the length code */ + extra = extra_lbits[code]; + if (extra != 0) { + lc -= base_length[code]; + send_bits(s, lc, extra); /* send the extra length bits */ + } + dist--; /* dist is now the match distance - 1 */ + code = d_code(dist); + Assert (code < D_CODES, "bad d_code"); + + send_code(s, code, dtree); /* send the distance code */ + extra = extra_dbits[code]; + if (extra != 0) { + dist -= base_dist[code]; + send_bits(s, dist, extra); /* send the extra distance bits */ + } + } /* literal or match pair ? */ + + /* Check that the overlay between pending_buf and d_buf+l_buf is ok: */ + Assert(s->pending < s->lit_bufsize + 2*lx, "pendingBuf overflow"); + + } while (lx < s->last_lit); + + send_code(s, END_BLOCK, ltree); + s->last_eob_len = ltree[END_BLOCK].Len; +} + +/* =========================================================================== + * Set the data type to ASCII or BINARY, using a crude approximation: + * binary if more than 20% of the bytes are <= 6 or >= 128, ascii otherwise. + * IN assertion: the fields freq of dyn_ltree are set and the total of all + * frequencies does not exceed 64K (to fit in an int on 16 bit machines). + */ +local void set_data_type(s) + deflate_state *s; +{ + int n = 0; + unsigned ascii_freq = 0; + unsigned bin_freq = 0; + while (n < 7) bin_freq += s->dyn_ltree[n++].Freq; + while (n < 128) ascii_freq += s->dyn_ltree[n++].Freq; + while (n < LITERALS) bin_freq += s->dyn_ltree[n++].Freq; + s->data_type = (Byte)(bin_freq > (ascii_freq >> 2) ? Z_BINARY : Z_ASCII); +} + +/* =========================================================================== + * Reverse the first len bits of a code, using straightforward code (a faster + * method would use a table) + * IN assertion: 1 <= len <= 15 + */ +local unsigned bi_reverse(code, len) + unsigned code; /* the value to invert */ + int len; /* its bit length */ +{ + register unsigned res = 0; + do { + res |= code & 1; + code >>= 1, res <<= 1; + } while (--len > 0); + return res >> 1; +} + +/* =========================================================================== + * Flush the bit buffer, keeping at most 7 bits in it. + */ +local void bi_flush(s) + deflate_state *s; +{ + if (s->bi_valid == 16) { + put_short(s, s->bi_buf); + s->bi_buf = 0; + s->bi_valid = 0; + } else if (s->bi_valid >= 8) { + put_byte(s, (Byte)s->bi_buf); + s->bi_buf >>= 8; + s->bi_valid -= 8; + } +} + +/* =========================================================================== + * Flush the bit buffer and align the output on a byte boundary + */ +local void bi_windup(s) + deflate_state *s; +{ + if (s->bi_valid > 8) { + put_short(s, s->bi_buf); + } else if (s->bi_valid > 0) { + put_byte(s, (Byte)s->bi_buf); + } + s->bi_buf = 0; + s->bi_valid = 0; +#ifdef DEBUG + s->bits_sent = (s->bits_sent+7) & ~7; +#endif +} + +/* =========================================================================== + * Copy a stored block, storing first the length and its + * one's complement if requested. + */ +local void copy_block(s, buf, len, header) + deflate_state *s; + charf *buf; /* the input data */ + unsigned len; /* its length */ + int header; /* true if block header must be written */ +{ + bi_windup(s); /* align on byte boundary */ + s->last_eob_len = 8; /* enough lookahead for inflate */ + + if (header) { + put_short(s, (ush)len); + put_short(s, (ush)~len); +#ifdef DEBUG + s->bits_sent += 2*16; +#endif + } +#ifdef DEBUG + s->bits_sent += (ulg)len<<3; +#endif + while (len--) { + put_byte(s, *buf++); + } +} diff --git a/src/utilfuns/zlib/trees.h b/src/utilfuns/zlib/trees.h new file mode 100644 index 0000000..72facf9 --- /dev/null +++ b/src/utilfuns/zlib/trees.h @@ -0,0 +1,128 @@ +/* header created automatically with -DGEN_TREES_H */ + +local const ct_data static_ltree[L_CODES+2] = { +{{ 12},{ 8}}, {{140},{ 8}}, {{ 76},{ 8}}, {{204},{ 8}}, {{ 44},{ 8}}, +{{172},{ 8}}, {{108},{ 8}}, {{236},{ 8}}, {{ 28},{ 8}}, {{156},{ 8}}, +{{ 92},{ 8}}, {{220},{ 8}}, {{ 60},{ 8}}, {{188},{ 8}}, {{124},{ 8}}, +{{252},{ 8}}, {{ 2},{ 8}}, {{130},{ 8}}, {{ 66},{ 8}}, {{194},{ 8}}, +{{ 34},{ 8}}, {{162},{ 8}}, {{ 98},{ 8}}, {{226},{ 8}}, {{ 18},{ 8}}, +{{146},{ 8}}, {{ 82},{ 8}}, {{210},{ 8}}, {{ 50},{ 8}}, {{178},{ 8}}, +{{114},{ 8}}, {{242},{ 8}}, {{ 10},{ 8}}, {{138},{ 8}}, {{ 74},{ 8}}, +{{202},{ 8}}, {{ 42},{ 8}}, {{170},{ 8}}, {{106},{ 8}}, {{234},{ 8}}, +{{ 26},{ 8}}, {{154},{ 8}}, {{ 90},{ 8}}, {{218},{ 8}}, {{ 58},{ 8}}, +{{186},{ 8}}, {{122},{ 8}}, {{250},{ 8}}, {{ 6},{ 8}}, {{134},{ 8}}, +{{ 70},{ 8}}, {{198},{ 8}}, {{ 38},{ 8}}, {{166},{ 8}}, {{102},{ 8}}, +{{230},{ 8}}, {{ 22},{ 8}}, {{150},{ 8}}, {{ 86},{ 8}}, {{214},{ 8}}, +{{ 54},{ 8}}, {{182},{ 8}}, {{118},{ 8}}, {{246},{ 8}}, {{ 14},{ 8}}, +{{142},{ 8}}, {{ 78},{ 8}}, {{206},{ 8}}, {{ 46},{ 8}}, {{174},{ 8}}, +{{110},{ 8}}, {{238},{ 8}}, {{ 30},{ 8}}, {{158},{ 8}}, {{ 94},{ 8}}, +{{222},{ 8}}, {{ 62},{ 8}}, {{190},{ 8}}, {{126},{ 8}}, {{254},{ 8}}, +{{ 1},{ 8}}, {{129},{ 8}}, {{ 65},{ 8}}, {{193},{ 8}}, {{ 33},{ 8}}, +{{161},{ 8}}, {{ 97},{ 8}}, {{225},{ 8}}, {{ 17},{ 8}}, {{145},{ 8}}, +{{ 81},{ 8}}, {{209},{ 8}}, {{ 49},{ 8}}, {{177},{ 8}}, {{113},{ 8}}, +{{241},{ 8}}, {{ 9},{ 8}}, {{137},{ 8}}, {{ 73},{ 8}}, {{201},{ 8}}, +{{ 41},{ 8}}, {{169},{ 8}}, {{105},{ 8}}, {{233},{ 8}}, {{ 25},{ 8}}, +{{153},{ 8}}, {{ 89},{ 8}}, {{217},{ 8}}, {{ 57},{ 8}}, {{185},{ 8}}, +{{121},{ 8}}, {{249},{ 8}}, {{ 5},{ 8}}, {{133},{ 8}}, {{ 69},{ 8}}, +{{197},{ 8}}, {{ 37},{ 8}}, {{165},{ 8}}, {{101},{ 8}}, {{229},{ 8}}, +{{ 21},{ 8}}, {{149},{ 8}}, {{ 85},{ 8}}, {{213},{ 8}}, {{ 53},{ 8}}, +{{181},{ 8}}, {{117},{ 8}}, {{245},{ 8}}, {{ 13},{ 8}}, {{141},{ 8}}, +{{ 77},{ 8}}, {{205},{ 8}}, {{ 45},{ 8}}, {{173},{ 8}}, {{109},{ 8}}, +{{237},{ 8}}, {{ 29},{ 8}}, {{157},{ 8}}, {{ 93},{ 8}}, {{221},{ 8}}, +{{ 61},{ 8}}, {{189},{ 8}}, {{125},{ 8}}, {{253},{ 8}}, {{ 19},{ 9}}, +{{275},{ 9}}, {{147},{ 9}}, {{403},{ 9}}, {{ 83},{ 9}}, {{339},{ 9}}, +{{211},{ 9}}, {{467},{ 9}}, {{ 51},{ 9}}, {{307},{ 9}}, {{179},{ 9}}, +{{435},{ 9}}, {{115},{ 9}}, {{371},{ 9}}, {{243},{ 9}}, {{499},{ 9}}, +{{ 11},{ 9}}, {{267},{ 9}}, {{139},{ 9}}, {{395},{ 9}}, {{ 75},{ 9}}, +{{331},{ 9}}, {{203},{ 9}}, {{459},{ 9}}, {{ 43},{ 9}}, {{299},{ 9}}, +{{171},{ 9}}, {{427},{ 9}}, {{107},{ 9}}, {{363},{ 9}}, {{235},{ 9}}, +{{491},{ 9}}, {{ 27},{ 9}}, {{283},{ 9}}, {{155},{ 9}}, {{411},{ 9}}, +{{ 91},{ 9}}, {{347},{ 9}}, {{219},{ 9}}, {{475},{ 9}}, {{ 59},{ 9}}, +{{315},{ 9}}, {{187},{ 9}}, {{443},{ 9}}, {{123},{ 9}}, {{379},{ 9}}, +{{251},{ 9}}, {{507},{ 9}}, {{ 7},{ 9}}, {{263},{ 9}}, {{135},{ 9}}, +{{391},{ 9}}, {{ 71},{ 9}}, {{327},{ 9}}, {{199},{ 9}}, {{455},{ 9}}, +{{ 39},{ 9}}, {{295},{ 9}}, {{167},{ 9}}, {{423},{ 9}}, {{103},{ 9}}, +{{359},{ 9}}, {{231},{ 9}}, {{487},{ 9}}, {{ 23},{ 9}}, {{279},{ 9}}, +{{151},{ 9}}, {{407},{ 9}}, {{ 87},{ 9}}, {{343},{ 9}}, {{215},{ 9}}, +{{471},{ 9}}, {{ 55},{ 9}}, {{311},{ 9}}, {{183},{ 9}}, {{439},{ 9}}, +{{119},{ 9}}, {{375},{ 9}}, {{247},{ 9}}, {{503},{ 9}}, {{ 15},{ 9}}, +{{271},{ 9}}, {{143},{ 9}}, {{399},{ 9}}, {{ 79},{ 9}}, {{335},{ 9}}, +{{207},{ 9}}, {{463},{ 9}}, {{ 47},{ 9}}, {{303},{ 9}}, {{175},{ 9}}, +{{431},{ 9}}, {{111},{ 9}}, {{367},{ 9}}, {{239},{ 9}}, {{495},{ 9}}, +{{ 31},{ 9}}, {{287},{ 9}}, {{159},{ 9}}, {{415},{ 9}}, {{ 95},{ 9}}, +{{351},{ 9}}, {{223},{ 9}}, {{479},{ 9}}, {{ 63},{ 9}}, {{319},{ 9}}, +{{191},{ 9}}, {{447},{ 9}}, {{127},{ 9}}, {{383},{ 9}}, {{255},{ 9}}, +{{511},{ 9}}, {{ 0},{ 7}}, {{ 64},{ 7}}, {{ 32},{ 7}}, {{ 96},{ 7}}, +{{ 16},{ 7}}, {{ 80},{ 7}}, {{ 48},{ 7}}, {{112},{ 7}}, {{ 8},{ 7}}, +{{ 72},{ 7}}, {{ 40},{ 7}}, {{104},{ 7}}, {{ 24},{ 7}}, {{ 88},{ 7}}, +{{ 56},{ 7}}, {{120},{ 7}}, {{ 4},{ 7}}, {{ 68},{ 7}}, {{ 36},{ 7}}, +{{100},{ 7}}, {{ 20},{ 7}}, {{ 84},{ 7}}, {{ 52},{ 7}}, {{116},{ 7}}, +{{ 3},{ 8}}, {{131},{ 8}}, {{ 67},{ 8}}, {{195},{ 8}}, {{ 35},{ 8}}, +{{163},{ 8}}, {{ 99},{ 8}}, {{227},{ 8}} +}; + +local const ct_data static_dtree[D_CODES] = { +{{ 0},{ 5}}, {{16},{ 5}}, {{ 8},{ 5}}, {{24},{ 5}}, {{ 4},{ 5}}, +{{20},{ 5}}, {{12},{ 5}}, {{28},{ 5}}, {{ 2},{ 5}}, {{18},{ 5}}, +{{10},{ 5}}, {{26},{ 5}}, {{ 6},{ 5}}, {{22},{ 5}}, {{14},{ 5}}, +{{30},{ 5}}, {{ 1},{ 5}}, {{17},{ 5}}, {{ 9},{ 5}}, {{25},{ 5}}, +{{ 5},{ 5}}, {{21},{ 5}}, {{13},{ 5}}, {{29},{ 5}}, {{ 3},{ 5}}, +{{19},{ 5}}, {{11},{ 5}}, {{27},{ 5}}, {{ 7},{ 5}}, {{23},{ 5}} +}; + +const uch _dist_code[DIST_CODE_LEN] = { + 0, 1, 2, 3, 4, 4, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7, 8, 8, 8, 8, + 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9, 10, 10, 10, 10, 10, 10, 10, 10, +10, 10, 10, 10, 10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, +11, 11, 11, 11, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, +12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 13, 13, 13, 13, +13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, +13, 13, 13, 13, 13, 13, 13, 13, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, +14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, +14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, +14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 15, 15, 15, 15, 15, 15, 15, 15, +15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, +15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, +15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 0, 0, 16, 17, +18, 18, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, +23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, +24, 24, 24, 24, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, +26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, +26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 27, 27, 27, 27, 27, 27, 27, 27, +27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, +27, 27, 27, 27, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, +28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, +28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, +28, 28, 28, 28, 28, 28, 28, 28, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, +29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, +29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, +29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29 +}; + +const uch _length_code[MAX_MATCH-MIN_MATCH+1]= { + 0, 1, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 12, 12, +13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 16, 16, 16, 16, 16, +17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, +19, 19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, +21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 22, 22, 22, 22, +22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 23, 23, 23, 23, 23, 23, 23, 23, +23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, +24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, +25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, +25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 26, 26, 26, 26, 26, 26, 26, 26, +26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, +26, 26, 26, 26, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, +27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 28 +}; + +local const int base_length[LENGTH_CODES] = { +0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, 28, 32, 40, 48, 56, +64, 80, 96, 112, 128, 160, 192, 224, 0 +}; + +local const int base_dist[D_CODES] = { + 0, 1, 2, 3, 4, 6, 8, 12, 16, 24, + 32, 48, 64, 96, 128, 192, 256, 384, 512, 768, + 1024, 1536, 2048, 3072, 4096, 6144, 8192, 12288, 16384, 24576 +}; + diff --git a/src/utilfuns/zlib/uncompr.c b/src/utilfuns/zlib/uncompr.c new file mode 100644 index 0000000..45a4687 --- /dev/null +++ b/src/utilfuns/zlib/uncompr.c @@ -0,0 +1,58 @@ +/* uncompr.c -- decompress a memory buffer + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id: uncompr.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "zlib.h" + +/* =========================================================================== + Decompresses the source buffer into the destination buffer. sourceLen is + the byte length of the source buffer. Upon entry, destLen is the total + size of the destination buffer, which must be large enough to hold the + entire uncompressed data. (The size of the uncompressed data must have + been saved previously by the compressor and transmitted to the decompressor + by some mechanism outside the scope of this compression library.) + Upon exit, destLen is the actual size of the compressed buffer. + This function can be used to decompress a whole file at once if the + input file is mmap'ed. + + uncompress returns Z_OK if success, Z_MEM_ERROR if there was not + enough memory, Z_BUF_ERROR if there was not enough room in the output + buffer, or Z_DATA_ERROR if the input data was corrupted. +*/ +int ZEXPORT uncompress (dest, destLen, source, sourceLen) + Bytef *dest; + uLongf *destLen; + const Bytef *source; + uLong sourceLen; +{ + z_stream stream; + int err; + + stream.next_in = (Bytef*)source; + stream.avail_in = (uInt)sourceLen; + /* Check for source > 64K on 16-bit machine: */ + if ((uLong)stream.avail_in != sourceLen) return Z_BUF_ERROR; + + stream.next_out = dest; + stream.avail_out = (uInt)*destLen; + if ((uLong)stream.avail_out != *destLen) return Z_BUF_ERROR; + + stream.zalloc = (alloc_func)0; + stream.zfree = (free_func)0; + + err = inflateInit(&stream); + if (err != Z_OK) return err; + + err = inflate(&stream, Z_FINISH); + if (err != Z_STREAM_END) { + inflateEnd(&stream); + return err == Z_OK ? Z_BUF_ERROR : err; + } + *destLen = stream.total_out; + + err = inflateEnd(&stream); + return err; +} diff --git a/src/utilfuns/zlib/untgz.c b/src/utilfuns/zlib/untgz.c new file mode 100644 index 0000000..2919656 --- /dev/null +++ b/src/utilfuns/zlib/untgz.c @@ -0,0 +1,436 @@ +/* + * untgz.c -- Display contents and/or extract file from + * a gzip'd TAR file + * written by "Pedro A. Aranda Guti\irrez" <paag@tid.es> + * adaptation to Unix by Jean-loup Gailly <jloup@gzip.org> + */ + +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <time.h> +#include <errno.h> +#include <fcntl.h> +#ifdef unix +# include <unistd.h> +#else +# include <direct.h> +# include <io.h> +#endif + +#include "zlib.h" + +#ifdef WIN32 +# include <windows.h> +# ifndef F_OK +# define F_OK (0) +# endif +# ifdef _MSC_VER +# define mkdir(dirname,mode) _mkdir(dirname) +# define strdup(str) _strdup(str) +# define unlink(fn) _unlink(fn) +# define access(path,mode) _access(path,mode) +# else +# define mkdir(dirname,mode) _mkdir(dirname) +# endif +#else +# include <utime.h> +#endif + + +/* Values used in typeflag field. */ + +#define REGTYPE '0' /* regular file */ +#define AREGTYPE '\0' /* regular file */ +#define LNKTYPE '1' /* link */ +#define SYMTYPE '2' /* reserved */ +#define CHRTYPE '3' /* character special */ +#define BLKTYPE '4' /* block special */ +#define DIRTYPE '5' /* directory */ +#define FIFOTYPE '6' /* FIFO special */ +#define CONTTYPE '7' /* reserved */ + +#define BLOCKSIZE 512 + +struct tar_header +{ /* byte offset */ + char name[100]; /* 0 */ + char mode[8]; /* 100 */ + char uid[8]; /* 108 */ + char gid[8]; /* 116 */ + char size[12]; /* 124 */ + char mtime[12]; /* 136 */ + char chksum[8]; /* 148 */ + char typeflag; /* 156 */ + char linkname[100]; /* 157 */ + char magic[6]; /* 257 */ + char version[2]; /* 263 */ + char uname[32]; /* 265 */ + char gname[32]; /* 297 */ + char devmajor[8]; /* 329 */ + char devminor[8]; /* 337 */ + char prefix[155]; /* 345 */ + /* 500 */ +}; + +union tar_buffer { + char buffer[BLOCKSIZE]; + struct tar_header header; +}; + +enum { TGZ_EXTRACT = 0, TGZ_LIST }; + +static char *TGZfname OF((const char *)); +void TGZnotfound OF((const char *)); + +int getoct OF((char *, int)); +char *strtime OF((time_t *)); +int ExprMatch OF((char *,char *)); + +int makedir OF((char *)); +int matchname OF((int,int,char **,char *)); + +void error OF((const char *)); +int tar OF((gzFile, int, int, int, char **)); + +void help OF((int)); +int main OF((int, char **)); + +char *prog; + +/* This will give a benign warning */ + +static char *TGZprefix[] = { "\0", ".tgz", ".tar.gz", ".tar", NULL }; + +/* Return the real name of the TGZ archive */ +/* or NULL if it does not exist. */ + +static char *TGZfname OF((const char *fname)) +{ + static char buffer[1024]; + int origlen,i; + + strcpy(buffer,fname); + origlen = strlen(buffer); + + for (i=0; TGZprefix[i]; i++) + { + strcpy(buffer+origlen,TGZprefix[i]); + if (access(buffer,F_OK) == 0) + return buffer; + } + return NULL; +} + +/* error message for the filename */ + +void TGZnotfound OF((const char *fname)) +{ + int i; + + fprintf(stderr,"%s : couldn't find ",prog); + for (i=0;TGZprefix[i];i++) + fprintf(stderr,(TGZprefix[i+1]) ? "%s%s, " : "or %s%s\n", + fname, + TGZprefix[i]); + exit(1); +} + + +/* help functions */ + +int getoct(char *p,int width) +{ + int result = 0; + char c; + + while (width --) + { + c = *p++; + if (c == ' ') + continue; + if (c == 0) + break; + result = result * 8 + (c - '0'); + } + return result; +} + +char *strtime (time_t *t) +{ + struct tm *local; + static char result[32]; + + local = localtime(t); + sprintf(result,"%2d/%02d/%4d %02d:%02d:%02d", + local->tm_mday, local->tm_mon+1, local->tm_year+1900, + local->tm_hour, local->tm_min, local->tm_sec); + return result; +} + + +/* regular expression matching */ + +#define ISSPECIAL(c) (((c) == '*') || ((c) == '/')) + +int ExprMatch(char *string,char *expr) +{ + while (1) + { + if (ISSPECIAL(*expr)) + { + if (*expr == '/') + { + if (*string != '\\' && *string != '/') + return 0; + string ++; expr++; + } + else if (*expr == '*') + { + if (*expr ++ == 0) + return 1; + while (*++string != *expr) + if (*string == 0) + return 0; + } + } + else + { + if (*string != *expr) + return 0; + if (*expr++ == 0) + return 1; + string++; + } + } +} + +/* recursive make directory */ +/* abort if you get an ENOENT errno somewhere in the middle */ +/* e.g. ignore error "mkdir on existing directory" */ +/* */ +/* return 1 if OK */ +/* 0 on error */ + +int makedir (char *newdir) +{ + char *buffer = strdup(newdir); + char *p; + int len = strlen(buffer); + + if (len <= 0) { + free(buffer); + return 0; + } + if (buffer[len-1] == '/') { + buffer[len-1] = '\0'; + } + if (mkdir(buffer, 0775) == 0) + { + free(buffer); + return 1; + } + + p = buffer+1; + while (1) + { + char hold; + + while(*p && *p != '\\' && *p != '/') + p++; + hold = *p; + *p = 0; + if ((mkdir(buffer, 0775) == -1) && (errno == ENOENT)) + { + fprintf(stderr,"%s: couldn't create directory %s\n",prog,buffer); + free(buffer); + return 0; + } + if (hold == 0) + break; + *p++ = hold; + } + free(buffer); + return 1; +} + +int matchname (int arg,int argc,char **argv,char *fname) +{ + if (arg == argc) /* no arguments given (untgz tgzarchive) */ + return 1; + + while (arg < argc) + if (ExprMatch(fname,argv[arg++])) + return 1; + + return 0; /* ignore this for the moment being */ +} + + +/* Tar file list or extract */ + +int untar (gzFile in, const char *dest) { + union tar_buffer buffer; + int len; + int err; + int getheader = 1; + int remaining = 0; + FILE *outfile = NULL; + char fname[BLOCKSIZE]; + time_t tartime; + + while (1) { + len = gzread(in, &buffer, BLOCKSIZE); + if (len < 0) + error (gzerror(in, &err)); + /* + * Always expect complete blocks to process + * the tar information. + */ + if (len != BLOCKSIZE) + error("gzread: incomplete block read"); + + /* + * If we have to get a tar header + */ + if (getheader == 1) { + /* + * if we met the end of the tar + * or the end-of-tar block, + * we are done + */ + if ((len == 0) || (buffer.header.name[0]== 0)) break; + + tartime = (time_t)getoct(buffer.header.mtime,12); + strcpy(fname, dest); + if ((fname[strlen(fname)-1] != '/') && (fname[strlen(fname)-1] != '\\')) + strcat(fname, "/"); + strcat(fname, buffer.header.name); + + switch (buffer.header.typeflag) { + case DIRTYPE: + makedir(fname); + break; + case REGTYPE: + case AREGTYPE: + remaining = getoct(buffer.header.size,12); + if (remaining) { + outfile = fopen(fname,"wb"); + if (outfile == NULL) { + // try creating directory + char *p = strrchr(fname, '/'); + if (p != NULL) { + *p = '\0'; + makedir(fname); + *p = '/'; + outfile = fopen(fname,"wb"); + } + } +/* + fprintf(stderr, + "%s %s\n", + (outfile) ? "Extracting" : "Couldn't create", + fname); +*/ + } + else + outfile = NULL; + /* + * could have no contents + */ + getheader = (remaining) ? 0 : 1; + break; + default: + break; + } + } + else { + unsigned int bytes = (remaining > BLOCKSIZE) ? BLOCKSIZE : remaining; + + if (outfile != NULL) { + if (fwrite(&buffer,sizeof(char),bytes,outfile) != bytes) { + fprintf(stderr,"%s : error writing %s skipping...\n",prog,fname); + fclose(outfile); + unlink(fname); + } + } + remaining -= bytes; + if (remaining == 0) { + getheader = 1; + if (outfile != NULL) { +#ifdef WIN32 + HANDLE hFile; + FILETIME ftm,ftLocal; + SYSTEMTIME st; + struct tm localt; + + fclose(outfile); + + localt = *localtime(&tartime); + + hFile = CreateFile(fname, GENERIC_READ | GENERIC_WRITE, + 0, NULL, OPEN_EXISTING, 0, NULL); + + st.wYear = (WORD)localt.tm_year+1900; + st.wMonth = (WORD)localt.tm_mon; + st.wDayOfWeek = (WORD)localt.tm_wday; + st.wDay = (WORD)localt.tm_mday; + st.wHour = (WORD)localt.tm_hour; + st.wMinute = (WORD)localt.tm_min; + st.wSecond = (WORD)localt.tm_sec; + st.wMilliseconds = 0; + SystemTimeToFileTime(&st,&ftLocal); + LocalFileTimeToFileTime(&ftLocal,&ftm); + SetFileTime(hFile,&ftm,NULL,&ftm); + CloseHandle(hFile); + + outfile = NULL; +#else + struct utimbuf settime; + + settime.actime = settime.modtime = tartime; + + fclose(outfile); + outfile = NULL; + utime(fname,&settime); +#endif + } + } + } + } + return 0; +} + + +/* =========================================================== */ + +void help(int exitval) +{ + fprintf(stderr, + "untgz v 0.1\n" + " an sample application of zlib 1.0.4\n\n" + "Usage : untgz TGZfile to extract all files\n" + " untgz TGZfile fname ... to extract selected files\n" + " untgz -l TGZfile to list archive contents\n" + " untgz -h to display this help\n\n"); + exit(exitval); +} + +void error(const char *msg) +{ + fprintf(stderr, "%s: %s\n", prog, msg); + exit(1); +} + + +int untargz(int fd, const char *dest) { + gzFile *f; + + f = gzdopen(fd, "rb"); + if (f == NULL) { + fprintf(stderr,"%s: Couldn't gzopen file\n", prog); + return 1; + } + + return untar(f, dest); +} diff --git a/src/utilfuns/zlib/zutil.c b/src/utilfuns/zlib/zutil.c new file mode 100644 index 0000000..55e047c --- /dev/null +++ b/src/utilfuns/zlib/zutil.c @@ -0,0 +1,225 @@ +/* zutil.c -- target dependent utility functions for the compression library + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* @(#) $Id: zutil.c 1688 2005-01-01 04:42:26Z scribe $ */ + +#include "zutil.h" + +struct internal_state {int dummy;}; /* for buggy compilers */ + +#ifndef STDC +extern void exit OF((int)); +#endif + +const char *z_errmsg[10] = { +"need dictionary", /* Z_NEED_DICT 2 */ +"stream end", /* Z_STREAM_END 1 */ +"", /* Z_OK 0 */ +"file error", /* Z_ERRNO (-1) */ +"stream error", /* Z_STREAM_ERROR (-2) */ +"data error", /* Z_DATA_ERROR (-3) */ +"insufficient memory", /* Z_MEM_ERROR (-4) */ +"buffer error", /* Z_BUF_ERROR (-5) */ +"incompatible version",/* Z_VERSION_ERROR (-6) */ +""}; + + +const char * ZEXPORT zlibVersion() +{ + return ZLIB_VERSION; +} + +#ifdef DEBUG + +# ifndef verbose +# define verbose 0 +# endif +int z_verbose = verbose; + +void z_error (m) + char *m; +{ + fprintf(stderr, "%s\n", m); + exit(1); +} +#endif + +/* exported to allow conversion of error code to string for compress() and + * uncompress() + */ +const char * ZEXPORT zError(err) + int err; +{ + return ERR_MSG(err); +} + + +#ifndef HAVE_MEMCPY + +void zmemcpy(dest, source, len) + Bytef* dest; + const Bytef* source; + uInt len; +{ + if (len == 0) return; + do { + *dest++ = *source++; /* ??? to be unrolled */ + } while (--len != 0); +} + +int zmemcmp(s1, s2, len) + const Bytef* s1; + const Bytef* s2; + uInt len; +{ + uInt j; + + for (j = 0; j < len; j++) { + if (s1[j] != s2[j]) return 2*(s1[j] > s2[j])-1; + } + return 0; +} + +void zmemzero(dest, len) + Bytef* dest; + uInt len; +{ + if (len == 0) return; + do { + *dest++ = 0; /* ??? to be unrolled */ + } while (--len != 0); +} +#endif + +#ifdef __TURBOC__ +#if (defined( __BORLANDC__) || !defined(SMALL_MEDIUM)) && !defined(__32BIT__) +/* Small and medium model in Turbo C are for now limited to near allocation + * with reduced MAX_WBITS and MAX_MEM_LEVEL + */ +# define MY_ZCALLOC + +/* Turbo C malloc() does not allow dynamic allocation of 64K bytes + * and farmalloc(64K) returns a pointer with an offset of 8, so we + * must fix the pointer. Warning: the pointer must be put back to its + * original form in order to free it, use zcfree(). + */ + +#define MAX_PTR 10 +/* 10*64K = 640K */ + +local int next_ptr = 0; + +typedef struct ptr_table_s { + voidpf org_ptr; + voidpf new_ptr; +} ptr_table; + +local ptr_table table[MAX_PTR]; +/* This table is used to remember the original form of pointers + * to large buffers (64K). Such pointers are normalized with a zero offset. + * Since MSDOS is not a preemptive multitasking OS, this table is not + * protected from concurrent access. This hack doesn't work anyway on + * a protected system like OS/2. Use Microsoft C instead. + */ + +voidpf zcalloc (voidpf opaque, unsigned items, unsigned size) +{ + voidpf buf = opaque; /* just to make some compilers happy */ + ulg bsize = (ulg)items*size; + + /* If we allocate less than 65520 bytes, we assume that farmalloc + * will return a usable pointer which doesn't have to be normalized. + */ + if (bsize < 65520L) { + buf = farmalloc(bsize); + if (*(ush*)&buf != 0) return buf; + } else { + buf = farmalloc(bsize + 16L); + } + if (buf == NULL || next_ptr >= MAX_PTR) return NULL; + table[next_ptr].org_ptr = buf; + + /* Normalize the pointer to seg:0 */ + *((ush*)&buf+1) += ((ush)((uch*)buf-0) + 15) >> 4; + *(ush*)&buf = 0; + table[next_ptr++].new_ptr = buf; + return buf; +} + +void zcfree (voidpf opaque, voidpf ptr) +{ + int n; + if (*(ush*)&ptr != 0) { /* object < 64K */ + farfree(ptr); + return; + } + /* Find the original pointer */ + for (n = 0; n < next_ptr; n++) { + if (ptr != table[n].new_ptr) continue; + + farfree(table[n].org_ptr); + while (++n < next_ptr) { + table[n-1] = table[n]; + } + next_ptr--; + return; + } + ptr = opaque; /* just to make some compilers happy */ + Assert(0, "zcfree: ptr not found"); +} +#endif +#endif /* __TURBOC__ */ + + +#if defined(M_I86) && !defined(__32BIT__) +/* Microsoft C in 16-bit mode */ + +# define MY_ZCALLOC + +#if (!defined(_MSC_VER) || (_MSC_VER <= 600)) +# define _halloc halloc +# define _hfree hfree +#endif + +voidpf zcalloc (voidpf opaque, unsigned items, unsigned size) +{ + if (opaque) opaque = 0; /* to make compiler happy */ + return _halloc((long)items, size); +} + +void zcfree (voidpf opaque, voidpf ptr) +{ + if (opaque) opaque = 0; /* to make compiler happy */ + _hfree(ptr); +} + +#endif /* MSC */ + + +#ifndef MY_ZCALLOC /* Any system without a special alloc function */ + +#ifndef STDC +extern voidp calloc OF((uInt items, uInt size)); +extern void free OF((voidpf ptr)); +#endif + +voidpf zcalloc (opaque, items, size) + voidpf opaque; + unsigned items; + unsigned size; +{ + if (opaque) items += size - size; /* make compiler happy */ + return (voidpf)calloc(items, size); +} + +void zcfree (opaque, ptr) + voidpf opaque; + voidpf ptr; +{ + free(ptr); + if (opaque) return; /* make compiler happy */ +} + +#endif /* MY_ZCALLOC */ diff --git a/src/utilfuns/zlib/zutil.h b/src/utilfuns/zlib/zutil.h new file mode 100644 index 0000000..538fc57 --- /dev/null +++ b/src/utilfuns/zlib/zutil.h @@ -0,0 +1,220 @@ +/* zutil.h -- internal interface and configuration of the compression library + * Copyright (C) 1995-2002 Jean-loup Gailly. + * For conditions of distribution and use, see copyright notice in zlib.h + */ + +/* WARNING: this file should *not* be used by applications. It is + part of the implementation of the compression library and is + subject to change. Applications should only use zlib.h. + */ + +/* @(#) $Id: zutil.h 1688 2005-01-01 04:42:26Z scribe $ */ + +#ifndef _Z_UTIL_H +#define _Z_UTIL_H + +#include "zlib.h" + +#ifdef STDC +# include <stddef.h> +# include <string.h> +# include <stdlib.h> +#endif +#ifdef NO_ERRNO_H + extern int errno; +#else +# include <errno.h> +#endif + +#ifndef local +# define local static +#endif +/* compile with -Dlocal if your debugger can't find static symbols */ + +typedef unsigned char uch; +typedef uch FAR uchf; +typedef unsigned short ush; +typedef ush FAR ushf; +typedef unsigned long ulg; + +extern const char *z_errmsg[10]; /* indexed by 2-zlib_error */ +/* (size given to avoid silly warnings with Visual C++) */ + +#define ERR_MSG(err) z_errmsg[Z_NEED_DICT-(err)] + +#define ERR_RETURN(strm,err) \ + return (strm->msg = (char*)ERR_MSG(err), (err)) +/* To be used only when the state is known to be valid */ + + /* common constants */ + +#ifndef DEF_WBITS +# define DEF_WBITS MAX_WBITS +#endif +/* default windowBits for decompression. MAX_WBITS is for compression only */ + +#if MAX_MEM_LEVEL >= 8 +# define DEF_MEM_LEVEL 8 +#else +# define DEF_MEM_LEVEL MAX_MEM_LEVEL +#endif +/* default memLevel */ + +#define STORED_BLOCK 0 +#define STATIC_TREES 1 +#define DYN_TREES 2 +/* The three kinds of block type */ + +#define MIN_MATCH 3 +#define MAX_MATCH 258 +/* The minimum and maximum match lengths */ + +#define PRESET_DICT 0x20 /* preset dictionary flag in zlib header */ + + /* target dependencies */ + +#ifdef MSDOS +# define OS_CODE 0x00 +# if defined(__TURBOC__) || defined(__BORLANDC__) +# if(__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__)) + /* Allow compilation with ANSI keywords only enabled */ + void _Cdecl farfree( void *block ); + void *_Cdecl farmalloc( unsigned long nbytes ); +# else +# include <alloc.h> +# endif +# else /* MSC or DJGPP */ +# include <malloc.h> +# endif +#endif + +#ifdef OS2 +# define OS_CODE 0x06 +#endif + +#ifdef WIN32 /* Window 95 & Windows NT */ +# define OS_CODE 0x0b +#endif + +#if defined(VAXC) || defined(VMS) +# define OS_CODE 0x02 +# define F_OPEN(name, mode) \ + fopen((name), (mode), "mbc=60", "ctx=stm", "rfm=fix", "mrs=512") +#endif + +#ifdef AMIGA +# define OS_CODE 0x01 +#endif + +#if defined(ATARI) || defined(atarist) +# define OS_CODE 0x05 +#endif + +#if defined(MACOS) || defined(TARGET_OS_MAC) +# define OS_CODE 0x07 +# if defined(__MWERKS__) && __dest_os != __be_os && __dest_os != __win32_os +# include <unix.h> /* for fdopen */ +# else +# ifndef fdopen +# define fdopen(fd,mode) NULL /* No fdopen() */ +# endif +# endif +#endif + +#ifdef __50SERIES /* Prime/PRIMOS */ +# define OS_CODE 0x0F +#endif + +#ifdef TOPS20 +# define OS_CODE 0x0a +#endif + +#if defined(_BEOS_) || defined(RISCOS) +# define fdopen(fd,mode) NULL /* No fdopen() */ +#endif + +#if (defined(_MSC_VER) && (_MSC_VER > 600)) +# define fdopen(fd,type) _fdopen(fd,type) +#endif + + + /* Common defaults */ + +#ifndef OS_CODE +# define OS_CODE 0x03 /* assume Unix */ +#endif + +#ifndef F_OPEN +# define F_OPEN(name, mode) fopen((name), (mode)) +#endif + + /* functions */ + +#ifdef HAVE_STRERROR + extern char *strerror OF((int)); +# define zstrerror(errnum) strerror(errnum) +#else +# define zstrerror(errnum) "" +#endif + +#if defined(pyr) +# define NO_MEMCPY +#endif +#if defined(SMALL_MEDIUM) && !defined(_MSC_VER) && !defined(__SC__) + /* Use our own functions for small and medium model with MSC <= 5.0. + * You may have to use the same strategy for Borland C (untested). + * The __SC__ check is for Symantec. + */ +# define NO_MEMCPY +#endif +#if defined(STDC) && !defined(HAVE_MEMCPY) && !defined(NO_MEMCPY) +# define HAVE_MEMCPY +#endif +#ifdef HAVE_MEMCPY +# ifdef SMALL_MEDIUM /* MSDOS small or medium model */ +# define zmemcpy _fmemcpy +# define zmemcmp _fmemcmp +# define zmemzero(dest, len) _fmemset(dest, 0, len) +# else +# define zmemcpy memcpy +# define zmemcmp memcmp +# define zmemzero(dest, len) memset(dest, 0, len) +# endif +#else + extern void zmemcpy OF((Bytef* dest, const Bytef* source, uInt len)); + extern int zmemcmp OF((const Bytef* s1, const Bytef* s2, uInt len)); + extern void zmemzero OF((Bytef* dest, uInt len)); +#endif + +/* Diagnostic functions */ +#ifdef DEBUG +# include <stdio.h> + extern int z_verbose; + extern void z_error OF((char *m)); +# define Assert(cond,msg) {if(!(cond)) z_error(msg);} +# define Trace(x) {if (z_verbose>=0) fprintf x ;} +# define Tracev(x) {if (z_verbose>0) fprintf x ;} +# define Tracevv(x) {if (z_verbose>1) fprintf x ;} +# define Tracec(c,x) {if (z_verbose>0 && (c)) fprintf x ;} +# define Tracecv(c,x) {if (z_verbose>1 && (c)) fprintf x ;} +#else +# define Assert(cond,msg) +# define Trace(x) +# define Tracev(x) +# define Tracevv(x) +# define Tracec(c,x) +# define Tracecv(c,x) +#endif + + +typedef uLong (ZEXPORT *check_func) OF((uLong check, const Bytef *buf, + uInt len)); +voidpf zcalloc OF((voidpf opaque, unsigned items, unsigned size)); +void zcfree OF((voidpf opaque, voidpf ptr)); + +#define ZALLOC(strm, items, size) \ + (*((strm)->zalloc))((strm)->opaque, (items), (size)) +#define ZFREE(strm, addr) (*((strm)->zfree))((strm)->opaque, (voidpf)(addr)) +#define TRY_FREE(s, p) {if (p) ZFREE(s, p);} + +#endif /* _Z_UTIL_H */ |