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) 2015 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/>.
34 deck_t *markdown_load(FILE *input) {
36 wchar_t c = L'\0'; // char
37 int i = 0; // increment
38 int hc = 0; // header count
39 int lc = 0; // line count
40 int sc = 1; // slide count
41 int bits = 0; // markdown bits
42 int prev = 0; // markdown bits of previous line
44 deck_t *deck = new_deck();
45 slide_t *slide = deck->slide;
48 cstring_t *text = cstring_init();
50 // initialize bits as empty line
51 SET_BIT(bits, IS_EMPTY);
53 while ((c = fgetwc(input)) != WEOF) {
55 fprintf(stderr, "markdown_load() failed to read input: %s\n", strerror(errno));
63 bits = markdown_analyse(text, prev);
65 // if first line in file is markdown hr
66 if(!line && CHECK_BIT(bits, IS_HR)) {
71 } else if(line && CHECK_BIT(bits, IS_STOP)) {
73 // set stop bit on last line
74 SET_BIT(line->bits, IS_STOP);
79 // if text is markdown hr
80 } else if(CHECK_BIT(bits, IS_HR) &&
81 CHECK_BIT(line->bits, IS_EMPTY)) {
89 slide = next_slide(slide);
94 // if slide ! has line
95 if(!slide->line || !line) {
105 line = next_line(line);
117 line->offset = next_nonblank(text, 0);
119 // adjust line length dynamicaly - excluding markup
120 if(line->text->value)
121 adjust_line_length(line);
124 text = cstring_init();
127 } else if(c == L'\t') {
129 // expand tab to spaces
130 for (i = 0; i < EXPAND_TABS; i++) {
131 (text->expand)(text, L' ');
134 } else if(c == L'\\') {
137 (text->expand)(text, c);
139 // if !IS_CODE add next char to line
140 // and do not increase line count
141 if(next_nonblank(text, 0) < CODE_INDENT) {
144 (text->expand)(text, c);
147 } else if(iswprint(c) || iswspace(c)) {
150 (text->expand)(text, c);
153 (text->delete)(text);
159 line = deck->slide->line;
160 if(line && line->text->size > 0 && line->text->value[0] == L'%') {
162 // assign header to deck
165 // find first non-header line
166 while(line && line->text->size > 0 && line->text->value[0] == L'%') {
171 // only split header if any non-header line is found
175 line->prev->next = NULL;
178 // remove header lines from slide
179 deck->slide->line = line;
183 deck->slide->lines -= hc;
186 // remove header from deck
195 // ignore mdpress format attributes
198 !CHECK_BIT(line->bits, IS_EMPTY) &&
199 line->text->value[line->offset] == L'=' &&
200 line->text->value[line->offset + 1] == L' ') {
202 // remove line from linked list
203 slide->line = line->next;
204 line->next->prev = NULL;
206 // maintain loop condition
214 (tmp->text->delete)(tmp->text);
219 // combine underlined H1/H2 in single line
220 if((CHECK_BIT(line->bits, IS_H1) ||
221 CHECK_BIT(line->bits, IS_H2)) &&
222 CHECK_BIT(line->bits, IS_EMPTY) &&
224 !CHECK_BIT(line->prev->bits, IS_EMPTY)) {
227 // remove line from linked list
228 line->prev->next = line->next;
230 line->next->prev = line->prev;
232 // set bits on previous line
233 if(CHECK_BIT(line->bits, IS_H1)) {
234 SET_BIT(line->prev->bits, IS_H1);
236 SET_BIT(line->prev->bits, IS_H2);
242 // maintain loop condition
247 (tmp->text->delete)(tmp->text);
250 // pass enclosing flag IS_UNORDERED_LIST_3
251 // to nested levels for unordered lists
252 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_3)) {
254 line_t *list_last_level_3 = line;
257 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
258 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3)) {
259 list_last_level_3 = tmp;
264 for(tmp = line; tmp != list_last_level_3; tmp = tmp->next) {
265 SET_BIT(tmp->bits, IS_UNORDERED_LIST_3);
268 // pass enclosing flag IS_UNORDERED_LIST_2
269 // to nested levels for unordered lists
270 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_2)) {
272 line_t *list_last_level_2 = line;
275 (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
276 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
277 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2)) {
278 list_last_level_2 = tmp;
283 for(tmp = line; tmp != list_last_level_2; tmp = tmp->next) {
284 SET_BIT(tmp->bits, IS_UNORDERED_LIST_2);
287 // pass enclosing flag IS_UNORDERED_LIST_1
288 // to nested levels for unordered lists
289 } else if(CHECK_BIT(line->bits, IS_UNORDERED_LIST_1)) {
291 line_t *list_last_level_1 = line;
294 (CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1) ||
295 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_2) ||
296 CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_3))) {
297 if(CHECK_BIT(tmp->bits, IS_UNORDERED_LIST_1)) {
298 list_last_level_1 = tmp;
303 for(tmp = line; tmp != list_last_level_1; tmp = tmp->next) {
304 SET_BIT(tmp->bits, IS_UNORDERED_LIST_1);
316 int markdown_analyse(cstring_t *text, int prev) {
318 // static variables can not be redeclaired, but changed outside of a declaration
319 // the program remembers their value on every function calls
320 static int unordered_list_level = 0;
321 static int unordered_list_level_offset[] = {-1, -1, -1, -1};
323 int i = 0; // increment
324 int bits = 0; // markdown bits
325 int offset = 0; // text offset
326 int eol = 0; // end of line
328 int equals = 0, hashes = 0,
329 stars = 0, minus = 0,
330 spaces = 0, other = 0; // special character counts
332 const int unordered_list_offset = unordered_list_level_offset[unordered_list_level];
334 // return IS_EMPTY on null pointers
335 if(!text || !text->value) {
336 SET_BIT(bits, IS_EMPTY);
340 // count leading spaces
341 offset = next_nonblank(text, 0);
344 if((offset < CODE_INDENT || !CHECK_BIT(prev, IS_CODE)) &&
345 (!wcsncmp(&text->value[offset], L"<br>", 4) ||
346 !wcsncmp(&text->value[offset], L"<BR>", 4) ||
347 !wcsncmp(&text->value[offset], L"^", 1))) {
348 SET_BIT(bits, IS_STOP);
352 // strip trailing spaces
353 for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);
355 // IS_UNORDERED_LIST_#
356 if(text->size >= offset + 2 &&
357 (text->value[offset] == L'*' || text->value[offset] == L'-') &&
358 iswspace(text->value[offset + 1])) {
360 // if different from last lines offset
361 if(offset != unordered_list_offset) {
363 // test if offset matches a lower indent level
364 for(i = unordered_list_level; i >= 0; i--) {
365 if(unordered_list_level_offset[i] == offset) {
366 unordered_list_level = i;
370 // if offset doesn't match any previously stored indent level
371 if(i != unordered_list_level) {
372 unordered_list_level = MIN(unordered_list_level + 1, UNORDERED_LIST_MAX_LEVEL);
373 // memorize the offset as next bigger indent level
374 unordered_list_level_offset[unordered_list_level] = offset;
378 // if no previous indent level matches, this must be the first line of the list
379 if(unordered_list_level == 0) {
380 unordered_list_level = 1;
381 unordered_list_level_offset[1] = offset;
384 switch(unordered_list_level) {
385 case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
386 case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
387 case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
392 if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
393 !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
394 !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
396 // continue list if indent level is still the same as in previous line
397 if ((CHECK_BIT(prev, IS_UNORDERED_LIST_1) ||
398 CHECK_BIT(prev, IS_UNORDERED_LIST_2) ||
399 CHECK_BIT(prev, IS_UNORDERED_LIST_3)) &&
400 offset >= unordered_list_offset) {
402 switch(unordered_list_level) {
403 case 1: SET_BIT(bits, IS_UNORDERED_LIST_1); break;
404 case 2: SET_BIT(bits, IS_UNORDERED_LIST_2); break;
405 case 3: SET_BIT(bits, IS_UNORDERED_LIST_3); break;
409 // this line extends the previous list item
410 SET_BIT(bits, IS_UNORDERED_LIST_EXT);
412 // or reset indent level
414 unordered_list_level = 0;
418 if(!CHECK_BIT(bits, IS_UNORDERED_LIST_1) &&
419 !CHECK_BIT(bits, IS_UNORDERED_LIST_2) &&
420 !CHECK_BIT(bits, IS_UNORDERED_LIST_3)) {
423 if(offset >= CODE_INDENT &&
424 (CHECK_BIT(prev, IS_EMPTY) ||
425 CHECK_BIT(prev, IS_CODE) ||
426 CHECK_BIT(prev, IS_STOP))) {
427 SET_BIT(bits, IS_CODE);
432 if(text->value[offset] == L'>') {
433 SET_BIT(bits, IS_QUOTE);
437 if(text->size >= offset + 3 &&
438 text->value[offset] == L'-' &&
439 text->value[offset + 1] == L'>' &&
440 iswspace(text->value[offset + 2])) {
441 SET_BIT(bits, IS_CENTER);
444 (text->strip)(text, offset, 3);
447 if(text->size >= offset + 3 &&
448 text->value[eol - 1] == L'-' &&
449 text->value[eol - 2] == L'<' &&
450 iswspace(text->value[eol - 3])) {
453 (text->strip)(text, eol - 3, 3);
455 // adjust end of line
456 for(eol = text->size; eol > offset && iswspace(text->value[eol - 1]); eol--);
461 for(i = offset; i < eol; i++) {
463 if(iswspace(text->value[i])) {
467 switch(text->value[i]) {
468 case L'=': equals++; break;
469 case L'#': hashes++; break;
470 case L'*': stars++; break;
471 case L'-': minus++; break;
472 case L'\\': other++; i++; break;
473 default: other++; break;
480 hashes + stars + minus + spaces + other == 0) {
481 SET_BIT(bits, IS_H1);
483 if(text->value[offset] == L'#' &&
484 iswspace(text->value[offset+1])) {
485 SET_BIT(bits, IS_H1);
486 SET_BIT(bits, IS_H1_ATX);
491 equals + hashes + stars + spaces + other == 0) {
492 SET_BIT(bits, IS_H2);
494 if(text->value[offset] == L'#' &&
495 text->value[offset+1] == L'#' &&
496 iswspace(text->value[offset+2])) {
497 SET_BIT(bits, IS_H2);
498 SET_BIT(bits, IS_H2_ATX);
502 if((minus >= 3 && equals + hashes + stars + other == 0) ||
503 (stars >= 3 && equals + hashes + minus + other == 0)) {
505 SET_BIT(bits, IS_HR);
510 SET_BIT(bits, IS_EMPTY);
518 void markdown_debug(deck_t *deck, int debug) {
520 int sc = 0; // slide count
521 int lc = 0; // line count
527 fwprintf(stderr, L"headers: %i\nslides: %i\n", deck->headers, deck->slides);
529 } else if(debug > 1) {
531 // print header to STDERR
533 header = deck->header;
535 header->length > 0 &&
536 header->text->value[0] == L'%') {
538 // skip descriptor word (e.g. %title:)
539 offset = next_blank(header->text, 0) + 1;
541 fwprintf(stderr, L"header: %S\n", &header->text->value[offset]);
542 header = header->next;
547 slide_t *slide = deck->slide;
550 // print slide/line count to STDERR
555 fwprintf(stderr, L" slide %i: %i lines\n", sc, slide->lines);
557 } else if(debug > 1) {
559 // also print bits and line length
560 fwprintf(stderr, L" slide %i:\n", sc);
565 fwprintf(stderr, L" line %i: bits = %i, length = %i\n", lc, line->bits, line->length);
574 void adjust_line_length(line_t *line) {
576 const static wchar_t *special = L"\\*_`"; // list of interpreted chars
577 const wchar_t *c = &line->text->value[line->offset];
578 cstack_t *stack = cstack_init();
580 // for each char in line
582 // if char is in special char list
583 if(wcschr(special, *c)) {
585 // closing special char (or second backslash)
586 if((stack->top)(stack, *c)) {
590 // treat special as regular char
591 } else if((stack->top)(stack, L'\\')) {
595 // opening special char
597 (stack->push)(stack, *c);
601 // remove backslash from stack
602 if((stack->top)(stack, L'\\'))
608 if(CHECK_BIT(line->bits, IS_H1_ATX))
610 if(CHECK_BIT(line->bits, IS_H2_ATX))
615 (stack->delete)(stack);
618 int next_nonblank(cstring_t *text, int i) {
619 while ((i < text->size) && iswspace((text->value)[i]))
625 int prev_blank(cstring_t *text, int i) {
626 while ((i > 0) && !iswspace((text->value)[i]))
632 int next_blank(cstring_t *text, int i) {
633 while ((i < text->size) && !iswspace((text->value)[i]))
639 int next_word(cstring_t *text, int i) {
640 return next_nonblank(text, next_blank(text, i));