summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorShawn Landden <shawn@churchofgit.com>2015-12-13 14:26:43 -0800
committerSven Eden <yamakuzure@gmx.net>2017-05-17 15:22:14 +0200
commit3a7af06c3e33eb33e8c177ab0891d10c687aed83 (patch)
treeea140377bdde49ed22fdb5d3cf3f0e6a6d6df014
parentd9a6c68accd606c69cfb92b6ffba3ca8e3cf6c56 (diff)
utf8.[ch] et al: use char32_t and char16_t instead of int, int32_t, int16_t
rework C11 utf8.[ch] to use char32_t instead of uint32_t when referring to unicode chars, to make things more expressive. [ @zonque: * rebased to current master * use AC_CHECK_DECLS to detect availibility of char{16,32}_t * make utf8_encoded_to_unichar() return int ]
-rw-r--r--src/basic/string-util.c13
-rw-r--r--src/basic/utf8.c48
-rw-r--r--src/basic/utf8.h14
3 files changed, 43 insertions, 32 deletions
diff --git a/src/basic/string-util.c b/src/basic/string-util.c
index 6006767da..2c9754db0 100644
--- a/src/basic/string-util.c
+++ b/src/basic/string-util.c
@@ -394,6 +394,7 @@ char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigne
char *e;
const char *i, *j;
unsigned k, len, len2;
+ int r;
assert(s);
assert(percent <= 100);
@@ -413,10 +414,10 @@ char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigne
k = 0;
for (i = s; k < x && i < s + old_length; i = utf8_next_char(i)) {
- int c;
+ char32_t c;
- c = utf8_encoded_to_unichar(i);
- if (c < 0)
+ r = utf8_encoded_to_unichar(i, &c);
+ if (r < 0)
return NULL;
k += unichar_iswide(c) ? 2 : 1;
}
@@ -425,11 +426,11 @@ char *ellipsize_mem(const char *s, size_t old_length, size_t new_length, unsigne
x ++;
for (j = s + old_length; k < new_length && j > i; ) {
- int c;
+ char32_t c;
j = utf8_prev_char(j);
- c = utf8_encoded_to_unichar(j);
- if (c < 0)
+ r = utf8_encoded_to_unichar(j, &c);
+ if (r < 0)
return NULL;
k += unichar_iswide(c) ? 2 : 1;
}
diff --git a/src/basic/utf8.c b/src/basic/utf8.c
index 124effd6d..3f024f7e5 100644
--- a/src/basic/utf8.c
+++ b/src/basic/utf8.c
@@ -53,7 +53,7 @@
#include "macro.h"
#include "utf8.h"
-bool unichar_is_valid(uint32_t ch) {
+bool unichar_is_valid(char32_t ch) {
if (ch >= 0x110000) /* End of unicode space */
return false;
@@ -67,7 +67,7 @@ bool unichar_is_valid(uint32_t ch) {
return true;
}
-static bool unichar_is_control(uint32_t ch) {
+static bool unichar_is_control(char32_t ch) {
/*
0 to ' '-1 is the C0 range.
@@ -103,8 +103,9 @@ static int utf8_encoded_expected_len(const char *str) {
}
/* decode one unicode char */
-int utf8_encoded_to_unichar(const char *str) {
- int unichar, len, i;
+int utf8_encoded_to_unichar(const char *str, char32_t *ret_unichar) {
+ char32_t unichar;
+ int len, i;
assert(str);
@@ -112,34 +113,37 @@ int utf8_encoded_to_unichar(const char *str) {
switch (len) {
case 1:
- return (int)str[0];
+ *ret_unichar = (char32_t)str[0];
+ return 0;
case 2:
unichar = str[0] & 0x1f;
break;
case 3:
- unichar = (int)str[0] & 0x0f;
+ unichar = (char32_t)str[0] & 0x0f;
break;
case 4:
- unichar = (int)str[0] & 0x07;
+ unichar = (char32_t)str[0] & 0x07;
break;
case 5:
- unichar = (int)str[0] & 0x03;
+ unichar = (char32_t)str[0] & 0x03;
break;
case 6:
- unichar = (int)str[0] & 0x01;
+ unichar = (char32_t)str[0] & 0x01;
break;
default:
return -EINVAL;
}
for (i = 1; i < len; i++) {
- if (((int)str[i] & 0xc0) != 0x80)
+ if (((char32_t)str[i] & 0xc0) != 0x80)
return -EINVAL;
unichar <<= 6;
- unichar |= (int)str[i] & 0x3f;
+ unichar |= (char32_t)str[i] & 0x3f;
}
- return unichar;
+ *ret_unichar = unichar;
+
+ return 0;
}
bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
@@ -148,15 +152,16 @@ bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
assert(str);
for (p = str; length;) {
- int encoded_len, val;
+ int encoded_len, r;
+ char32_t val;
encoded_len = utf8_encoded_valid_unichar(p);
if (encoded_len < 0 ||
(size_t) encoded_len > length)
return false;
- val = utf8_encoded_to_unichar(p);
- if (val < 0 ||
+ r = utf8_encoded_to_unichar(p, &val);
+ if (r < 0 ||
unichar_is_control(val) ||
(!newline && val == '\n'))
return false;
@@ -276,7 +281,7 @@ char *ascii_is_valid(const char *str) {
* Returns: The length in bytes that the UTF-8 representation does or would
* occupy.
*/
-size_t utf8_encode_unichar(char *out_utf8, uint32_t g) {
+size_t utf8_encode_unichar(char *out_utf8, char32_t g) {
if (g < (1 << 7)) {
if (out_utf8)
@@ -320,7 +325,7 @@ char *utf16_to_utf8(const void *s, size_t length) {
t = r;
while (f < (const uint8_t*) s + length) {
- uint16_t w1, w2;
+ char16_t w1, w2;
/* see RFC 2781 section 2.2 */
@@ -354,7 +359,7 @@ char *utf16_to_utf8(const void *s, size_t length) {
}
/* expected size used to encode one unicode char */
-static int utf8_unichar_to_encoded_len(int unichar) {
+static int utf8_unichar_to_encoded_len(char32_t unichar) {
if (unichar < 0x80)
return 1;
@@ -372,7 +377,8 @@ static int utf8_unichar_to_encoded_len(int unichar) {
/* validate one encoded unicode char and return its length */
int utf8_encoded_valid_unichar(const char *str) {
- int len, unichar, i;
+ int len, i, r;
+ char32_t unichar;
assert(str);
@@ -389,7 +395,9 @@ int utf8_encoded_valid_unichar(const char *str) {
if ((str[i] & 0x80) != 0x80)
return -EINVAL;
- unichar = utf8_encoded_to_unichar(str);
+ r = utf8_encoded_to_unichar(str, &unichar);
+ if (r < 0)
+ return r;
/* check if encoded length matches encoded value */
if (utf8_unichar_to_encoded_len(unichar) != len)
diff --git a/src/basic/utf8.h b/src/basic/utf8.h
index 16c4b5b55..3e2e35b96 100644
--- a/src/basic/utf8.h
+++ b/src/basic/utf8.h
@@ -24,12 +24,14 @@
#include <stdbool.h>
#include <stddef.h>
#include <stdint.h>
+#include <uchar.h>
#include "macro.h"
+#include "missing.h"
#define UTF8_REPLACEMENT_CHARACTER "\xef\xbf\xbd"
-bool unichar_is_valid(uint32_t c);
+bool unichar_is_valid(char32_t c);
const char *utf8_is_valid(const char *s) _pure_;
char *ascii_is_valid(const char *s) _pure_;
@@ -40,20 +42,20 @@ bool utf8_is_printable_newline(const char* str, size_t length, bool newline) _pu
char *utf8_escape_invalid(const char *s);
char *utf8_escape_non_printable(const char *str);
-size_t utf8_encode_unichar(char *out_utf8, uint32_t g);
+size_t utf8_encode_unichar(char *out_utf8, char32_t g);
char *utf16_to_utf8(const void *s, size_t length);
int utf8_encoded_valid_unichar(const char *str);
-int utf8_encoded_to_unichar(const char *str);
+int utf8_encoded_to_unichar(const char *str, char32_t *ret_unichar);
-static inline bool utf16_is_surrogate(uint16_t c) {
+static inline bool utf16_is_surrogate(char16_t c) {
return (0xd800 <= c && c <= 0xdfff);
}
-static inline bool utf16_is_trailing_surrogate(uint16_t c) {
+static inline bool utf16_is_trailing_surrogate(char16_t c) {
return (0xdc00 <= c && c <= 0xdfff);
}
-static inline uint32_t utf16_surrogate_pair_to_unichar(uint16_t lead, uint16_t trail) {
+static inline char32_t utf16_surrogate_pair_to_unichar(char16_t lead, char16_t trail) {
return ((lead - 0xd800) << 10) + (trail - 0xdc00) + 0x10000;
}