2 * Functions necessary to parse a file and transform its content into
3 * a deck of slides containing lines. All based on markdown formating
5 * Copyright (C) 2014 Michael Goehler
7 * This file is part of mdp.
9 * This program is free software: you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation, either version 3 of the License, or
12 * (at your option) any later version.
14 * This program is distributed in the hope that it will be useful,
15 * but WITHOUT ANY WARRANTY; without even the implied warranty of
16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 * GNU General Public License for more details.
19 * You should have received a copy of the GNU General Public License
20 * along with this program. If not, see <http://www.gnu.org/licenses/>.
32 deck_t *markdown_load(FILE *input) {
35 int i = 0; // increment
36 int hc = 0; // header count
37 int lc = 0; // line count
38 int sc = 1; // slide count
39 int bits = 0; // markdown bits
40 int prev = 0; // markdown bits of previous line
42 deck_t *deck = new_deck();
43 slide_t *slide = deck->slide;
46 cstring_t *text = cstring_init();
48 // initialize bits as empty line
49 SET_BIT(bits, IS_EMPTY);
51 while ((c = fgetc(input)) != EOF) {
53 fprintf(stderr, "markdown_load() failed to read input: %s\n", strerror(errno));
61 bits = markdown_analyse(text, prev);
63 // if first line in file is markdown hr
64 if(!line && CHECK_BIT(bits, IS_HR)) {
69 // if text is markdown hr
70 } else if(CHECK_BIT(bits, IS_HR) &&
71 CHECK_BIT(line->bits, IS_EMPTY)) {
79 slide = next_slide(slide);
84 // if slide ! has line
95 line = next_line(line);
107 line->offset = next_nonblank(text, 0);
109 // adjust line length dynamicaly - excluding markup
111 adjust_line_length(line);
114 text = cstring_init();
117 } else if(c == '\t') {
119 // expand tab to spaces
120 for (i = 0; i < EXPAND_TABS; i++) {
121 (text->expand)(text, ' ');
124 } else if(c == '\\') {
127 (text->expand)(text, c);
129 // if !IS_CODE add next char to line
130 // and do not increase line count
131 if(next_nonblank(text, 0) < CODE_INDENT) {
134 (text->expand)(text, c);
138 // if utf-8 char > 1 byte add remaing to line
139 for(i = 0; i < length_utf8(c) - 1; i++) {
141 (text->expand)(text, c);
147 } else if(isprint(c) || isspace((unsigned char) c)) {
150 (text->expand)(text, c);
152 } else if(is_utf8(c)) {
155 (text->expand)(text, c);
157 // if utf-8 char > 1 byte add remaing to line
158 for(i = 0; i < length_utf8(c) - 1; i++) {
160 (text->expand)(text, c);
164 (text->delete)(text);
170 line = deck->slide->line;
171 if(line && line->text->size > 0 && line->text->text[0] == '%') {
173 // assign header to deck
176 // find first non-header line
177 while(line && line->text->size > 0 && line->text->text[0] == '%') {
182 // only split header if any non-header line is found
186 line->prev->next = NULL;
189 // remove header lines from slide
190 deck->slide->line = line;
194 deck->slide->lines -= hc;
197 // remove header from deck
206 // combine underlined H1/H2 in single line
207 if((CHECK_BIT(line->bits, IS_H1) ||
208 CHECK_BIT(line->bits, IS_H2)) &&
209 CHECK_BIT(line->bits, IS_EMPTY) &&
211 !CHECK_BIT(line->prev->bits, IS_EMPTY)) {
214 // remove line from linked list
215 line->prev->next = line->next;
217 line->next->prev = line->prev;
219 // set bits on previous line
220 if(CHECK_BIT(line->bits, IS_H1)) {
221 SET_BIT(line->prev->bits, IS_H1);
223 SET_BIT(line->prev->bits, IS_H2);
229 // maintain loop condition
234 (tmp->text->delete)(tmp->text);
237 // pass enclosing flag IS_UNORDERED_LIST_3
238 // to nested levels for unordered lists
239 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_3)) {
241 line_t *list_last_level_3 = line;
244 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
245 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
246 list_last_level_3 = tmp;
251 for(tmp = line; tmp != list_last_level_3; tmp = tmp->next) {
252 SET_BIT(tmp->bits, IS_UNORDERED_LIST_3);
255 // pass enclosing flag IS_UNORDERED_LIST_2
256 // to nested levels for unordered lists
257 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_2)) {
259 line_t *list_last_level_2 = line;
262 (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
263 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
264 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2)) {
265 list_last_level_2 = tmp;
270 for(tmp = line; tmp != list_last_level_2; tmp = tmp->next) {
271 SET_BIT(tmp->bits, IS_UNORDERED_LIST_2);
274 // pass enclosing flag IS_UNORDERED_LIST_1
275 // to nested levels for unordered lists
276 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_1)) {
278 line_t *list_last_level_1 = line;
281 (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1) ||
282 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
283 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
284 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1)) {
285 list_last_level_1 = tmp;
290 for(tmp = line; tmp != list_last_level_1; tmp = tmp->next) {
291 SET_BIT(tmp->bits, IS_UNORDERED_LIST_1);
303 int markdown_analyse(cstring_t *text, int prev) {
305 // static variables can not be redeclaired, but changed outside of a declaration
306 // the program remembers their value on every function calls
307 static int unordered_list_level = 0;
308 static int unordered_list_level_offset[] = {-1, -1, -1, -1};
310 int i = 0; // increment
311 int bits = 0; // markdown bits
312 int offset = 0; // text offset
313 int eol = 0; // end of line
315 int equals = 0, hashes = 0,
316 stars = 0, minus = 0,
317 spaces = 0, other = 0; // special character counts
319 const int unordered_list_offset = unordered_list_level_offset[unordered_list_level];
321 // return IS_EMPTY on null pointers
322 if(!text || !text->text) {
323 SET_BIT(bits, IS_EMPTY);
327 // count leading spaces
328 offset = next_nonblank(text, 0);
330 // strip trailing spaces
331 for(eol = text->size; eol > offset && isspace((unsigned char) text->text[eol - 1]); eol--);
333 // IS_UNORDERED_LIST_#
334 if(text->size >= offset + 2 &&
335 (text->text[offset] == '*' || text->text[offset] == '-') &&
336 text->text[offset + 1] == ' ') {
338 // if different from last lines offset
339 if(offset != unordered_list_offset) {
341 // test if offset matches a lower indent level
342 for(i = unordered_list_level; i >= 0; i--) {
343 if(unordered_list_level_offset[i] == offset) {
344 unordered_list_level = i;
348 // if offset doesn't match any previously stored indent level
349 if(i != unordered_list_level) {
350 unordered_list_level = MIN(unordered_list_level + 1, UNORDERED_LIST_MAX_LEVEL);
351 // memorize the offset as next bigger indent level
352 unordered_list_level_offset[unordered_list_level] = offset;
356 // if no previous indent level matches, this must be the first line of the list
357 if(unordered_list_level == 0) {
358 unordered_list_level = 1;
359 unordered_list_level_offset[1] = offset;
362 switch(unordered_list_level) {
363 case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
364 case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
365 case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
370 if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
371 !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
372 !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
374 // continue list if indent level is still the same as in previous line
375 if ((CHECK_BIT(prev, IS_UNORDERED_LIST_1) ||
376 CHECK_BIT(prev, IS_UNORDERED_LIST_2) ||
377 CHECK_BIT(prev, IS_UNORDERED_LIST_3)) &&
378 offset >= unordered_list_offset) {
380 switch(unordered_list_level) {
381 case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
382 case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
383 case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
387 // this line extends the previous list item
388 SET_BIT(bits, IS_UNORDERED_LIST_EXT);
390 // or reset indent level
392 unordered_list_level = 0;
396 if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
397 !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
398 !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
401 if(offset >= CODE_INDENT &&
402 (CHECK_BIT(prev, IS_EMPTY) ||
403 CHECK_BIT(prev, IS_CODE))) {
404 SET_BIT(bits, IS_CODE);
409 if(text->text[offset] == '>') {
410 SET_BIT(bits, IS_QUOTE);
414 if(text->size >= offset + 3 &&
415 text->text[offset] == '-' &&
416 text->text[offset + 1] == '>' &&
417 text->text[offset + 2] == ' ') {
418 SET_BIT(bits, IS_CENTER);
421 (text->strip)(text, offset, 3);
424 if(text->size >= offset + 3 &&
425 text->text[eol - 1] == '-' &&
426 text->text[eol - 2] == '<' &&
427 text->text[eol - 3] == ' ') {
430 (text->strip)(text, eol - 3, 3);
432 // adjust end of line
433 for(eol = text->size; eol > offset && isspace((unsigned char) text->text[eol - 1]); eol--);
438 for(i = offset; i < eol; i++) {
440 if(text->text[i] == ' ') {
444 switch(text->text[i]) {
445 case '=': equals++; break;
446 case '#': hashes++; break;
447 case '*': stars++; break;
448 case '-': minus++; break;
449 case '\\': other++; i++; break;
450 default: other++; break;
457 hashes + stars + minus + spaces + other == 0) {
458 SET_BIT(bits, IS_H1);
460 if(text->text[offset] == '#' &&
461 text->text[offset+1] == ' ') {
462 SET_BIT(bits, IS_H1);
463 SET_BIT(bits, IS_H1_ATX);
468 equals + hashes + stars + spaces + other == 0) {
469 SET_BIT(bits, IS_H2);
471 if(text->text[offset] == '#' &&
472 text->text[offset+1] == '#' &&
473 text->text[offset+2] == ' ') {
474 SET_BIT(bits, IS_H2);
475 SET_BIT(bits, IS_H2_ATX);
479 if((minus >= 3 && equals + hashes + stars + other == 0) ||
480 (stars >= 3 && equals + hashes + minus + other == 0)) {
482 SET_BIT(bits, IS_HR);
487 SET_BIT(bits, IS_EMPTY);
495 void markdown_debug(deck_t *deck, int debug) {
497 int sc = 0; // slide count
498 int lc = 0; // line count
504 fprintf(stderr, "headers: %i\nslides: %i\n", deck->headers, deck->slides);
506 } else if(debug > 1) {
508 // print header to STDERR
510 header = deck->header;
512 header->length > 0 &&
513 header->text->text[0] == '%') {
515 // skip descriptor word (e.g. %title:)
516 offset = next_blank(header->text, 0) + 1;
518 fprintf(stderr, "header: %s\n", &header->text->text[offset]);
519 header = header->next;
524 slide_t *slide = deck->slide;
527 // print slide/line count to STDERR
532 fprintf(stderr, " slide %i: %i lines\n", sc, slide->lines);
534 } else if(debug > 1) {
536 // also print bits and line length
537 fprintf(stderr, " slide %i:\n", sc);
542 fprintf(stderr, " line %i: bits = %i, length = %i\n", lc, line->bits, line->length);
551 void adjust_line_length(line_t *line) {
553 const static char *special = "\\*_`"; // list of interpreted chars
554 const char *c = &line->text->text[line->offset];
555 cstack_t *stack = cstack_init();
557 // for each char in line
559 // if char is in special char list
560 if(strchr(special, *c)) {
562 // closing special char (or second backslash)
563 if((stack->top)(stack, *c)) {
567 // treat special as regular char
568 } else if((stack->top)(stack, '\\')) {
572 // opening special char
574 (stack->push)(stack, *c);
578 // remove backslash from stack
579 if((stack->top)(stack, '\\'))
585 if(CHECK_BIT(line->bits, IS_H1_ATX))
587 if(CHECK_BIT(line->bits, IS_H2_ATX))
592 (stack->delete)(stack);
595 bool is_utf8(char ch) {
596 return (ch & 0x80) != 0x00;
599 int length_utf8(char ch) {
601 int i = 0; // increment
611 int next_nonblank(cstring_t *text, int i) {
612 while ((i < text->size) && isspace((unsigned char) (text->text)[i]))
618 int prev_blank(cstring_t *text, int i) {
619 while ((i > 0) && !isspace((unsigned char) (text->text)[i]))
625 int next_blank(cstring_t *text, int i) {
626 while ((i < text->size) && !isspace((unsigned char) (text->text)[i]))
632 int next_word(cstring_t *text, int i) {
633 return next_nonblank(text, next_blank(text, i));