/*
 * Functions necessary to parse a file and transform its content into
 * a deck of slides containing lines. All based on markdown formating
 * rules.
 * Copyright (C) 2016 Michael Goehler
 *
 * This file is part of mdp.
 *
 * 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 3 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, see <http://www.gnu.org/licenses/>.
 *
 */

#include <ctype.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <wchar.h>
#include <wctype.h>
#include <string.h>

#include "parser.h"

deck_t *markdown_load(FILE *input) {

    wchar_t c = L'\0';    // char
    int i = 0;    // increment
    int hc = 0;   // header count
    int lc = 0;   // line count
    int sc = 1;   // slide count
    int bits = 0; // markdown bits
    int prev = 0; // markdown bits of previous line

    deck_t *deck = new_deck();
    slide_t *slide = deck->slide;
    line_t *line = NULL;
    line_t *tmp = NULL;
    cstring_t *text = cstring_init();

    // initialize bits as empty line
    SET_BIT(bits, IS_EMPTY);

    while ((c = fgetwc(input)) != WEOF) {
        if (ferror(input)) {
            fprintf(stderr, "markdown_load() failed to read input: %s\n", strerror(errno));
            exit(EXIT_FAILURE);
        }

        if(c == L'\n') {

            // markdown analyse
            prev = bits;
            bits = markdown_analyse(text, prev);

            // if first line in file is markdown hr
            if(!line && CHECK_BIT(bits, IS_HR)) {

                // clear text
                (text->reset)(text);

            } else if(line && CHECK_BIT(bits, IS_STOP)) {

                // set stop bit on last line
                SET_BIT(line->bits, IS_STOP);

                // clear text
                (text->reset)(text);

            // if text is markdown hr
            } else if(CHECK_BIT(bits, IS_HR) &&
                      CHECK_BIT(line->bits, IS_EMPTY)) {

                slide->lines = lc;

                // clear text
                (text->reset)(text);

                // create next slide
                slide = next_slide(slide);
                sc++;

            } else if(CHECK_BIT(bits, IS_TILDE_CODE) &&
                      CHECK_BIT(bits, IS_EMPTY)) {
                // remove tilde code markers
                (text->reset)(text);

            } else {

                // if slide ! has line
                if(!slide->line || !line) {

                    // create new line
                    line = new_line();
                    slide->line = line;
                    lc = 1;

                } else {

                    // create next line
                    line = next_line(line);
                    lc++;

                }

                // add text to line
                line->text = text;

                // add bits to line
                line->bits = bits;

                // calc offset
                line->offset = next_nonblank(text, 0);

                // adjust line length dynamicaly - excluding markup
                if(line->text->value)
                    adjust_line_length(line);

                // new text
                text = cstring_init();
            }

        } else if(c == L'\t') {

            // expand tab to spaces
            for (i = 0;  i < EXPAND_TABS;  i++) {
                (text->expand)(text, L' ');
            }

        } else if(c == L'\\') {

            // add char to line
            (text->expand)(text, c);

            // if !IS_CODE add next char to line
            // and do not increase line count
            if(next_nonblank(text, 0) < CODE_INDENT) {

                c = fgetwc(input);
                (text->expand)(text, c);
            }

        } else if(iswprint(c) || iswspace(c)) {

            // add char to line
            (text->expand)(text, c);
        }
    }
    (text->delete)(text);

    slide->lines = lc;
    deck->slides = sc;

    // detect header
    line = deck->slide->line;
    if(line && line->text->size > 0 && line->text->value[0] == L'%') {

        // assign header to deck
        deck->header = line;

        // find first non-header line
        while(line && line->text->size > 0 && line->text->value[0] == L'%') {
            hc++;
            line = line->next;
        }

        // only split header if any non-header line is found
        if(line) {

            // split linked list
            line->prev->next = NULL;
            line->prev = NULL;

            // remove header lines from slide
            deck->slide->line = line;

            // adjust counts
            deck->headers += hc;
            deck->slide->lines -= hc;
        } else {

            // remove header from deck
            deck->header = NULL;
        }
    }

    slide = deck->slide;
    while(slide) {
        line = slide->line;

        // ignore mdpress format attributes
        if(line &&
           slide->lines > 1 &&
           !CHECK_BIT(line->bits, IS_EMPTY) &&
           line->text->value[line->offset] == L'=' &&
           line->text->value[line->offset + 1] == L' ') {

            // remove line from linked list
            slide->line = line->next;
            line->next->prev = NULL;

            // maintain loop condition
            tmp = line;
            line = line->next;

            // adjust line count
            slide->lines -= 1;

            // delete line
            (tmp->text->delete)(tmp->text);
            free(tmp);
        }

        while(line) {
            // combine underlined H1/H2 in single line
            if((CHECK_BIT(line->bits, IS_H1) ||
                CHECK_BIT(line->bits, IS_H2)) &&
               CHECK_BIT(line->bits, IS_EMPTY) &&
               line->prev &&
               !CHECK_BIT(line->prev->bits, IS_EMPTY)) {


                // remove line from linked list
                line->prev->next = line->next;
                if(line->next)
                    line->next->prev = line->prev;

                // set bits on previous line
                if(CHECK_BIT(line->bits, IS_H1)) {
                    SET_BIT(line->prev->bits, IS_H1);
                } else {
                    SET_BIT(line->prev->bits, IS_H2);
                }

                // adjust line count
                slide->lines -= 1;

                // maintain loop condition
                tmp = line;
                line = line->prev;

                // delete line
                (tmp->text->delete)(tmp->text);
                free(tmp);

            // pass enclosing flag IS_UNORDERED_LIST_3
            // to nested levels for unordered lists
            } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_3)) {
                tmp = line->next;
                line_t *list_last_level_3 = line;

                while(tmp &&
                      CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
                    if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
                        list_last_level_3 = tmp;
                    }
                    tmp = tmp->next;
                }

                for(tmp = line; tmp != list_last_level_3; tmp = tmp->next) {
                    SET_BIT(tmp->bits, IS_UNORDERED_LIST_3);
                }

            // pass enclosing flag IS_UNORDERED_LIST_2
            // to nested levels for unordered lists
            } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_2)) {
                tmp = line->next;
                line_t *list_last_level_2 = line;

                while(tmp &&
                      (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
                       CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
                    if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2)) {
                        list_last_level_2 = tmp;
                    }
                    tmp = tmp->next;
                }

                for(tmp = line; tmp != list_last_level_2; tmp = tmp->next) {
                    SET_BIT(tmp->bits, IS_UNORDERED_LIST_2);
                }

            // pass enclosing flag IS_UNORDERED_LIST_1
            // to nested levels for unordered lists
            } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_1)) {
                tmp = line->next;
                line_t *list_last_level_1 = line;

                while(tmp &&
                      (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1) ||
                       CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
                       CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
                    if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1)) {
                        list_last_level_1 = tmp;
                    }
                    tmp = tmp->next;
                }

                for(tmp = line; tmp != list_last_level_1; tmp = tmp->next) {
                    SET_BIT(tmp->bits, IS_UNORDERED_LIST_1);
                }
            }

            line = line->next;
        }
        slide = slide->next;
    }

    return deck;
}

int markdown_analyse(cstring_t *text, int prev) {

    // static variables can not be redeclaired, but changed outside of a declaration
    // the program remembers their value on every function calls
    static int unordered_list_level = 0;
    static int unordered_list_level_offset[] = {-1, -1, -1, -1};
    static int num_tilde_characters = 0;

    int i = 0;      // increment
    int bits = 0;   // markdown bits
    int offset = 0; // text offset
    int eol    = 0; // end of line

    int equals = 0, hashes = 0,
        stars  = 0, minus  = 0,
        spaces = 0, other  = 0; // special character counts

    const int unordered_list_offset = unordered_list_level_offset[unordered_list_level];

    // return IS_EMPTY on null pointers
    if(!text || !text->value) {
        SET_BIT(bits, IS_EMPTY);

        // continue fenced code blocks across empty lines
        if(num_tilde_characters > 0)
            SET_BIT(bits, IS_CODE);

        return bits;
    }

    // count leading spaces
    offset = next_nonblank(text, 0);

    // IS_TILDE_CODE
    if (wcsncmp(text->value, L"~~~", 3) == 0) {
        int tildes_in_line = next_nontilde(text, 0);
        if (tildes_in_line >= num_tilde_characters) {
            if (num_tilde_characters > 0) {
                num_tilde_characters = 0;
            } else {
                num_tilde_characters = tildes_in_line;
            }
            SET_BIT(bits, IS_EMPTY);
            SET_BIT(bits, IS_TILDE_CODE);
            return bits;
        }
    }

    if (num_tilde_characters > 0) {
        SET_BIT(bits, IS_CODE);
        SET_BIT(bits, IS_TILDE_CODE);
        return bits;
    }

    // IS_STOP
    if((offset < CODE_INDENT || !CHECK_BIT(prev, IS_CODE)) &&
       (!wcsncmp(&text->value[offset], L"<br>", 4) ||
        !wcsncmp(&text->value[offset], L"<BR>", 4) ||
        !wcsncmp(&text->value[offset], L"^", 1))) {
        SET_BIT(bits, IS_STOP);
        return bits;
    }

    // strip trailing spaces
    for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);

    // IS_UNORDERED_LIST_#
    if(text->size >= offset + 2 &&
       (text->value[offset] == L'*' || text->value[offset] == L'-') &&
       iswspace(text->value[offset + 1])) {

        // if different from last lines offset
        if(offset != unordered_list_offset) {

            // test if offset matches a lower indent level
            for(i = unordered_list_level; i >= 0; i--) {
                if(unordered_list_level_offset[i] == offset) {
                    unordered_list_level = i;
                    break;
                }
            }
            // if offset doesn't match any previously stored indent level
            if(i != unordered_list_level) {
                unordered_list_level = MIN(unordered_list_level + 1, UNORDERED_LIST_MAX_LEVEL);
                // memorize the offset as next bigger indent level
                unordered_list_level_offset[unordered_list_level] = offset;
            }
        }

        // if no previous indent level matches, this must be the first line of the list
        if(unordered_list_level == 0) {
            unordered_list_level = 1;
            unordered_list_level_offset[1] = offset;
        }

        switch(unordered_list_level) {
            case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
            case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
            case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
            default: break;
        }
    }

    if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
       !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
       !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {

        // continue list if indent level is still the same as in previous line
        if ((CHECK_BIT(prev, IS_UNORDERED_LIST_1) ||
             CHECK_BIT(prev, IS_UNORDERED_LIST_2) ||
             CHECK_BIT(prev, IS_UNORDERED_LIST_3)) &&
            offset >= unordered_list_offset) {

            switch(unordered_list_level) {
                case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
                case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
                case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
                default: break;
            }

            // this line extends the previous list item
            SET_BIT(bits, IS_UNORDERED_LIST_EXT);

        // or reset indent level
        } else {
            unordered_list_level = 0;
        }
    }

    if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
       !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
       !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {

        // IS_CODE
        if(offset >= CODE_INDENT &&
           (CHECK_BIT(prev, IS_EMPTY) ||
            CHECK_BIT(prev, IS_CODE)  ||
            CHECK_BIT(prev, IS_STOP))) {
            SET_BIT(bits, IS_CODE);

        } else {

            // IS_QUOTE
            if(text->value[offset] == L'>') {
                SET_BIT(bits, IS_QUOTE);
            }

            // IS_CENTER
            if(text->size >= offset + 3 &&
               text->value[offset] == L'-' &&
               text->value[offset + 1] == L'>' &&
               iswspace(text->value[offset + 2])) {
                SET_BIT(bits, IS_CENTER);

                // remove start tag
                (text->strip)(text, offset, 3);
                eol -= 3;

                if(text->size >= offset + 3 &&
                   text->value[eol - 1] == L'-' &&
                   text->value[eol - 2] == L'<' &&
                   iswspace(text->value[eol - 3])) {

                    // remove end tags
                    (text->strip)(text, eol - 3, 3);

                    // adjust end of line
                    for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);

                }
            }

            for(i = offset; i < eol; i++) {

                if(iswspace(text->value[i])) {
                    spaces++;

                } else {
                    switch(text->value[i]) {
                        case L'=': equals++;  break;
                        case L'#': hashes++;  break;
                        case L'*': stars++;   break;
                        case L'-': minus++;   break;
                        case L'\\': other++; i++; break;
                        default:  other++;   break;
                    }
                }
            }

            // IS_H1
            if(equals > 0 &&
               hashes + stars + minus + spaces + other == 0) {
                SET_BIT(bits, IS_H1);
            }
            if(text->value[offset] == L'#' &&
               iswspace(text->value[offset+1])) {
                SET_BIT(bits, IS_H1);
                SET_BIT(bits, IS_H1_ATX);
            }

            // IS_H2
            if(minus > 0 &&
               equals + hashes + stars + spaces + other == 0) {
                SET_BIT(bits, IS_H2);
            }
            if(text->value[offset] == L'#' &&
               text->value[offset+1] == L'#' &&
               iswspace(text->value[offset+2])) {
                SET_BIT(bits, IS_H2);
                SET_BIT(bits, IS_H2_ATX);
            }

            // IS_HR
            if((minus >= 3 && equals + hashes + stars + other == 0) ||
               (stars >= 3 && equals + hashes + minus + other == 0)) {

                SET_BIT(bits, IS_HR);
            }

            // IS_EMPTY
            if(other == 0) {
                SET_BIT(bits, IS_EMPTY);
            }
        }
    }

    return bits;
}

void markdown_debug(deck_t *deck, int debug) {

    int sc = 0; // slide count
    int lc = 0; // line count

    int offset;
    line_t *header;

    if(debug == 1) {
        fwprintf(stderr, L"headers: %i\nslides: %i\n", deck->headers, deck->slides);

    } else if(debug > 1) {

        // print header to STDERR
        if(deck->header) {
            header = deck->header;
            while(header &&
                header->length > 0 &&
                header->text->value[0] == L'%') {

                // skip descriptor word (e.g. %title:)
                offset = next_blank(header->text, 0) + 1;

                fwprintf(stderr, L"header: %S\n", &header->text->value[offset]);
                header = header->next;
            }
        }
    }

    slide_t *slide = deck->slide;
    line_t *line;

    // print slide/line count to STDERR
    while(slide) {
        sc++;

        if(debug == 1) {
            fwprintf(stderr, L"  slide %i: %i lines\n", sc, slide->lines);

        } else if(debug > 1) {

            // also print bits and line length
            fwprintf(stderr, L"  slide %i:\n", sc);
            line = slide->line;
            lc = 0;
            while(line) {
                lc++;
                fwprintf(stderr, L"    line %i: bits = %i, length = %i\n", lc, line->bits, line->length);
                line = line->next;
            }
        }

        slide = slide->next;
    }
}

void adjust_line_length(line_t *line) {
    int l = 0;
    const static wchar_t *special = L"\\*_`"; // list of interpreted chars
    const wchar_t *c = &line->text->value[0];
    cstack_t *stack = cstack_init();

    // for each char in line
    for(; *c; c++) {
        // if char is in special char list
        if(wcschr(special, *c)) {

            // closing special char (or second backslash)
            if((stack->top)(stack, *c)) {
                if(*c == L'\\') l++;
                (stack->pop)(stack);

            // treat special as regular char
            } else if((stack->top)(stack, L'\\')) {
                l++;
                (stack->pop)(stack);

            // opening special char
            } else {
                (stack->push)(stack, *c);
            }

        } else {
            // remove backslash from stack
            if((stack->top)(stack, L'\\'))
                (stack->pop)(stack);
            l++;
        }
    }

    if(CHECK_BIT(line->bits, IS_H1_ATX))
        l -= 2;
    if(CHECK_BIT(line->bits, IS_H2_ATX))
        l -= 3;

    line->length = l;

    (stack->delete)(stack);
}

int next_nonblank(cstring_t *text, int i) {
    while ((i < text->size) && iswspace((text->value)[i]))
        i++;

    return i;
}

int prev_blank(cstring_t *text, int i) {
    while ((i > 0) && !iswspace((text->value)[i]))
        i--;

    return i;
}

int next_blank(cstring_t *text, int i) {
    while ((i < text->size) && !iswspace((text->value)[i]))
        i++;

    return i;
}

int next_word(cstring_t *text, int i) {
    return next_nonblank(text, next_blank(text, i));
}

int next_nontilde(cstring_t *text, int i) {
    while ((i < text->size) && text->value[i] == L'~')
        i++;

    return i;
}