4 # EXPERIMENTAL Markdown processor with minimal dependencies.
5 # Meant to support all features of John Grubers basic Markdown
6 # + a number of common extensions, mostly inspired by Pandoc Markdown
8 # Supported Features / TODO:
9 # ==========================
10 # [x] done [ ] todo [-] not planned ? unsure
12 # Basic Markdown - Block elements:
13 # -------------------------------
15 # - [x] Double space line breaks
16 # - [x] Proper block element nesting
18 # - [x] ATX-Style Headings
20 # - [x] Lists (ordered, unordered)
21 # - [x] Code blocks (using indention)
22 # - [x] Horizontal rules
23 # - [x] Verbatim HTML block (disabled by default)
25 # Basic Markdown - Inline elements:
26 # ---------------------------------
28 # - [x] Reference style links
29 # - [x] Emphasis *em*/**strong** (*Asterisk*, _Underscore_)
30 # - [x] `code`, also ``code containing `backticks` ``
31 # - [x] Images / reference style images
32 # - [x] <automatic links>
33 # - [x] backslash escapes
34 # - [x] Verbatim HTML inline (disabled by default)
37 # NOTE: Set the environment variable MD_HTML=true to enable verbatim HTML
39 # Extensions - Block elements:
40 # ----------------------------
41 # - ? Heading identifiers (php md, pandoc)
42 # - [x] Automatic heading identifiers (custom)
43 # - [x] Fenced code blocks (php md, pandoc)
44 # - [x] Fenced code attributes
46 # - ? Simple table (pandoc)
47 # - ? Multiline table (pandoc)
48 # - [x] Grid table (pandoc)
49 # - [x] Pipe table (php md pandoc)
50 # - [x] Line blocks (pandoc)
51 # - [x] Task lists (pandoc, custom)
52 # - [ ] Definition lists (php md, pandoc)
53 # - [-] Numbered example lists (pandoc)
54 # - [-] Metadata blocks (pandoc)
55 # - [x] Metadata blocks (custom)
56 # - [x] Fenced Divs (pandoc)
58 # Extensions - Inline elements:
59 # ----------------------------
60 # - [x] Ignore embedded_underscores (php md, pandoc)
61 # - [x] ~~strikeout~~ (pandoc)
62 # - [x] ^Superscript^ ~Subscript~ (pandoc)
63 # - [-] Bracketed spans (pandoc)
64 # - [-] Inline attributes (pandoc)
65 # - [-] TEX-Math (pandoc)
66 # - ? Footnotes (php md)
67 # - ? Abbreviations (php md)
68 # - ? "Curly quotes" (smartypants)
69 # - [ ] em-dashes (--) (smartypants old)
70 # - ? ... three-dot ellipsis (smartypants)
71 # - [-] en-dash (smartypants)
72 # - [ ] Automatic em-dash / en-dash
73 # - [ ] Automatic -> Arrows <-
75 function debug(text) { printf "\n---\n%s\n---\n", text > "/dev/stderr"; }
77 function HTML ( text ) {
78 gsub( /&/, "\\&", text );
79 gsub( /</, "\\<", text );
80 gsub( />/, "\\>", text );
81 gsub( /"/, "\\"", text );
82 gsub( /'/, "\\'", text );
83 gsub( /\\/, "\\\", text );
87 function inline( line, LOCAL, len, code, href, guard ) {
88 nu = "(\\\\\\\\|\\\\[^\\\\]|[^\\\\_]|_[[:alnum:]])*" # not underline (except when escaped)
89 na = "(\\\\\\\\|\\\\[^\\\\]|[^\\\\\\*])*" # not asterisk (except when escaped)
90 ieu = "_([^_[:space:]]|[^_[:space:]]" nu "[^_[:space:]])_" # inner <em> (underline)
91 isu = "__([^_[:space:]]|[^_[:space:]]" nu "[^_[:space:]])__" # inner <strong> (underline)
92 iea = "\\*([^\\*[:space:]]|[^\\*[:space:]]" na "[^\\*[:space:]])\\*" # inner <em> (asterisk)
93 isa = "\\*\\*([^\\*[:space:]]|[^\\*[:space:]]" na "[^\\*[:space:]])\\*\\*" # inner <strong> (asterisk)
95 if ( line ~ /^$/ ) { # Recursion End
98 # omit processing of escaped characters
99 } else if ( line ~ /^\\[]\\`\*_\{\}\(\)#\+-\.![]/) {
100 return substr(line, 2, 1) inline( substr(line, 3) );
103 } else if ( match(line, /^ \n/) ) {
104 return "<br />\n" inline( substr(line, RLENGTH + 1) );
107 } else if ( match( line, /^`+/) ) {
109 guard = substr( line, 1, len )
110 if ( match(line, guard ".*" guard) ) {
111 code = substr( line, len + 1, match( substr(line, len + 1), guard ) - 1)
112 len = 2 * length(guard) + length(code)
113 # strip single surrounding white spaces
114 code = gensub( / (.*) /, "\\1", "1" , code)
115 # escape HTML within code span
116 gsub( /&/, "\\&", code ); gsub( /</, "\\<", code ); gsub( />/, "\\>", code );
117 return "<code>" code "</code>" inline( substr( line, len + 1 ) )
120 # quick links ("automatic links" in md doc)
121 } else if ( match( line, /^<[a-zA-Z]+:\/\/([-\.[:alnum:]]+)(:[0-9]*)?(\/[^>]*)?>/ ) ) {
123 href = HTML( substr( line, 2, len - 2) );
124 return "<a href=\"" href "\">" href "</a>" inline( substr( line, len + 1) );
127 } else if ( match( line, /^<[a-zA-Z0-9.!#$%&'\''*+\/=?^_`{|}~-]+@[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(\.[a-zA-Z0-9]([a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*>/ ) ) {
129 href = HTML( substr( line, 2, len - 2) );
130 return "<a href=\"mailto:" href "\">" href "</a>" inline( substr( line, len + 1) );
133 } else if ( match(line, /^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/) ) {
135 text = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\1", 1, substr(line, 1, len) );
136 href = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\2", 1, substr(line, 1, len) );
137 title = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\4", 1, substr(line, 1, len) );
139 return "<a href=\"" HTML(href) "\" title=\"" HTML(title) "\">" inline( text ) "</a>" inline( substr( line, len + 1) );
141 return "<a href=\"" HTML(href) "\">" inline( text ) "</a>" inline( substr( line, len + 1) );
144 # reference style links
145 } else if ( match(line, /^\[([^]]+)\] ?\[([^]]*)\]/ ) ) {
147 text = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, substr(line, 1, len) );
148 id = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, substr(line, 1, len) );
149 if ( ! id ) id = text;
150 if ( rl_href[id] && rl_title[id] ) {
151 return "<a href=\"" HTML(rl_href[id]) "\" title=\"" HTML(rl_title[id]) "\">" inline(text) "</a>" inline( substr( line, len + 1) );
152 } else if ( rl_href[id] ) {
153 return "<a href=\"" HTML(rl_href[id]) "\">" inline(text) "</a>" inline( substr( line, len + 1) );
155 return "" HTML(substr(line, 1, len)) inline( substr(line, len + 1) );
159 } else if ( match(line, /^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/) ) {
161 text = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\1", "g", substr(line, 1, len) );
162 href = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\2", "g", substr(line, 1, len) );
163 title = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\4", "g", substr(line, 1, len) );
165 return "<img src=\"" HTML(href) "\" alt=\"" HTML(text) "\" title=\"" HTML(title) "\" />" inline( substr( line, len + 1) );
167 return "<img src=\"" HTML(href) "\" alt=\"" HTML(text) "\" />" inline( substr( line, len + 1) );
170 # reference style images
171 } else if ( match(line, /^!\[([^]]+)\] ?\[([^]]*)\]/ ) ) {
173 text = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, substr(line, 1, len) );
174 id = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, substr(line, 1, len) );
175 if ( ! id ) id = text;
176 if ( rl_href[id] && rl_title[id] ) {
177 return "<img src=\"" HTML(rl_href[id]) "\" alt=\"" HTML(text) "\" title=\"" HTML(rl_title[id]) "\" />" inline( substr( line, len + 1) );
178 } else if ( rl_href[id] ) {
179 return "<img src=\"" HTML(rl_href[id]) "\" alt=\"" HTML(text) "\" />" inline( substr( line, len + 1) );
181 return "" HTML(substr(line, 1, len)) inline( substr(line, len + 1) );
184 # ~~strikeout~~ (pandoc)
185 } else if ( match(line, /^~~([[:graph:]]|[[:graph:]]([^~]|~[^~])*[[:graph:]])~~/) ) {
187 return "<del>" inline( substr( line, 3, len - 4 ) ) "</del>" inline( substr( line, len + 1 ) );
189 # ^superscript^ (pandoc)
190 } else if ( match(line, /^\^([^[:space:]^]|\\[ ^])+\^/) ) {
192 return "<sup>" inline( substr( line, 2, len - 2 ) ) "</sup>" inline( substr( line, len + 1 ) );
194 # ~subscript~ (pandoc)
195 } else if ( match(line, /^~([^[:space:]~]|\\[ ~])+~/) ) {
197 return "<sub>" inline( substr( line, 2, len - 2 ) ) "</sub>" inline( substr( line, len + 1 ) );
199 # ignore embedded underscores (pandoc, php md)
200 } else if ( match(line, "^[[:alnum:]](__|_)") ) {
201 return HTML(substr( line, 1, RLENGTH)) inline( substr(line, RLENGTH + 1) );
204 } else if ( match(line, "^__(([^_[:space:]]|" ieu ")|([^_[:space:]]|" ieu ")(" nu "|" ieu ")*([^_[:space:]]|" ieu "))__$") ) {
206 return "<strong>" inline( substr( line, 3, len - 4 ) ) "</strong>" inline( substr( line, len + 1 ) );
209 } else if ( match(line, "^__(([^_[:space:]]|" ieu ")|([^_[:space:]]|" ieu ")(" nu "|" ieu ")*([^_[:space:]]|" ieu "))__[[:space:][:punct:]]") ) {
211 return "<strong>" inline( substr( line, 3, len - 5 ) ) "</strong>" inline( substr( line, len) );
214 } else if ( match(line, "^\\*\\*(([^\\*[:space:]]|" iea ")|([^\\*[:space:]]|" iea ")(" na "|" iea ")*([^\\*[:space:]]|" iea "))\\*\\*") ) {
216 return "<strong>" inline( substr( line, 3, len - 4 ) ) "</strong>" inline( substr( line, len + 1 ) );
219 } else if ( match(line, "^_(([^_[:space:]]|" isu ")|([^_[:space:]]|" isu ")(" nu "|" isu ")*([^_[:space:]]|" isu "))_$") ) {
221 return "<em>" inline( substr( line, 2, len - 2 ) ) "</em>" inline( substr( line, len + 1 ) );
224 } else if ( match(line, "^_(([^_[:space:]]|" isu ")|([^_[:space:]]|" isu ")(" nu "|" isu ")*([^_[:space:]]|" isu "))_[[:space:][:punct:]]") ) {
226 return "<em>" inline( substr( line, 2, len - 3 ) ) "</em>" inline( substr( line, len ) );
229 } else if ( match(line, "^\\*(([^\\*[:space:]]|" isa ")|([^\\*[:space:]]|" isa ")(" na "|" isa ")*([^\\*[:space:]]|" isa "))\\*") ) {
231 return "<em>" inline( substr( line, 2, len - 2 ) ) "</em>" inline( substr( line, len + 1 ) );
234 } else if ( AllowMacros && match( line, /^<<([^>]|>[^>])+>>/) ) {
236 return macro( substr( line, 3, len - 4 ) ) inline(substr(line, len + 1));
238 # Verbatim inline HTML
239 } else if ( AllowHTML && match( line, /^(<!--([^-]|-[^-]|--[^>])*-->|<\?([^\?]|\?[^>])*\?>|<![A-Z][^>]*>|<!\[CDATA\[([^\]]|\][^\]]|\]\][^>])*\]\]>|<\/[A-Za-z][A-Za-z0-9-]*[[:space:]]*>|<[A-Za-z][A-Za-z0-9-]*([[:space:]]+[A-Za-z_:][A-Za-z0-9_\.:-]*([[:space:]]*=[[:space:]]*([[:space:]"'=<>`]+|"[^"]*"|'[^']*'))?)*[[:space:]]*\/?>)/) ) {
241 return substr( line, 1, len) inline(substr(line, len + 1));
243 # Literal HTML entities
244 } else if ( match( line, /^&([a-zA-Z]{2,32}|#[0-9]{1,7}|#[xX][0-9a-fA-F]{1,6});/) ) {
246 return substr( line, 1, len ) inline(substr(line, len + 1));
248 # Escape lone HTML character
249 } else if ( match( line, /^[&<>"']/) ) {
250 return HTML(substr(line, 1, 1)) inline(substr(line, 2));
252 # continue walk over string
254 return substr(line, 1, 1) inline( substr(line, 2) );
258 function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ) {
259 gsub( /^\n+|\n+$/, "", block );
265 } else if ( AllowHTML && match( block, /(^|\n) ? ? ?(<!--([^-]|-[^-]|--[^>])*(-->|$)|<\?([^\?]|\?[^>])*(\?>|$)|<![A-Z][^>]*(>|$)|<!\[CDATA\[([^\]]|\][^\]]|\]\][^>])*(\]\]>|$))/) ) {
266 len = RLENGTH; st = RSTART;
267 return _block(substr(block, 1, st - 1)) substr(block, st, len) _block(substr(block, st + len));
270 } else if ( AllowHTML && match( tolower(block), /(^|\n) ? ? ?<\/?(address|article|aside|base|basefont|blockquote|body|caption|center|col|colgroup|dd|details|dialog|dir|div|dl|dt|fieldset|figcaption|figure|footer|form|frame|frameset|h[123456]|head|header|hr|html|iframe|legend|li|link|main|menu|menuitem|nav|noframes|ol|optgroup|option|p|param|section|source|summary|table|tbody|td|tfoot|th|thead|title|tr|track|ul)([[:space:]\n>]|\/>)([^\n]|\n[ \t]*[^\n])*(\n[[:space:]]*\n|$)/) ) {
271 len = RLENGTH; st = RSTART;
272 return _block(substr(block, 1, st - 1)) substr(block, st, len) _block(substr(block, st + len));
275 } else if ( AllowHTML && match( tolower(block), /(^|\n) ? ? ?<(script|pre|style)([[:space:]\n>]).*(<\/script>|<\/pre>|<\/style>|$)/) ) {
276 len = RLENGTH; st = RSTART;
277 match( tolower(substr(block, st, len)), /(<\/script>|<\/pre>|<\/style>)/);
278 len = RSTART + RLENGTH;
279 return _block(substr(block, 1, st - 1)) substr(block, st, len) _block(substr(block, st + len));
282 } else if ( AllowHTML && match( block, /^ ? ? ?(<\/[A-Za-z][A-Za-z0-9-]*[[:space:]]*>|<[A-Za-z][A-Za-z0-9-]*([[:space:]]+[A-Za-z_:][A-Za-z0-9_\.:-]*([[:space:]]*=[[:space:]]*([[:space:]"'=<>`]+|"[^"]*"|'[^']*'))?)*[[:space:]]*\/?>)([[:space:]]*\n)([^\n]|\n[ \t]*[^\n])*(\n[[:space:]]*\n|$)/) ) {
283 len = RLENGTH; st = RSTART;
284 return substr(block, st, len) _block(substr(block, st + len));
286 # Metadata (custom, block starting with %something)
287 # Metadata is ignored but can be interpreted externally
288 } else if ( match(block, /^%[a-zA-Z]+([[:space:]][^\n]*)?(\n|$)(%[a-zA-Z]+([[:space:]][^\n]*)?(\n|$)|%([[:space:]][^\n]*)?(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) {
289 len = RLENGTH; st = RSTART;
290 return _block( substr( block, len + 1) );
292 # Blockquote (leading >)
293 } else if ( match( block, /^> /) ) {
294 match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match(block, /$/);
295 len = RLENGTH; st = RSTART;
296 return "<blockquote>\n" _block( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "</blockquote>\n\n" \
297 _block( substr(block, st + len) );
299 # Pipe Tables (pandoc / php md / gfm )
300 } else if ( match(block, "^((\\|)?([^\n]+\\|)+[^\n]+(\\|)?)\n" \
301 "((\\|)?:?(-+:?[\\|+])+:?-+:?(\\|)?)\n" \
302 "((\\|)?([^\n]+\\|)+[^\n]+(\\|)?(\n|$))+" ) ) {
303 len = RLENGTH; st = RSTART;
304 #initialize empty arrays
305 split("", talign); split("", tarray);
306 cols = 0; cnt=0; ttext = "";
308 # table header and alignment
309 split( gensub( /(^\||\|$)/, "", "g", \
310 gensub( /(^|[^\\])\\\|/, "\\1\\|", "g", \
311 substr(block, 1, match(block, /(\n|$)/)) \
313 block = substr(block, match(block, /(\n|$)/) + 1 );
315 gensub( /(^\||\|$)/, "", "g", \
316 substr(block, 1, match(block, /(\n|$)/)) \
318 block = substr(block, match(block, /(\n|$)/) + 1 );
320 for( cnt = 1; cnt < cols; cnt++ ) {
321 if (match(talign[cnt], /:-+:/)) talign[cnt]="center";
322 else if (match(talign[cnt], /-+:/)) talign[cnt]="right";
323 else if (match(talign[cnt], /:-+/)) talign[cnt]="left";
327 ttext = "<thead>\n<tr>"
328 for (cnt = 1; cnt < cols; cnt++)
329 ttext = ttext "<th align=\"" talign[cnt] "\">" inline(tarray[cnt]) "</th>"
330 ttext = ttext "</tr>\n</thead><tbody>\n"
332 while ( match(block, "^((\\|)?([^\n]+\\|)+[^\n]+(\\|)?(\n|$))+" ) ){
333 split( gensub( /(^\||\|$)/, "", "g", \
334 gensub( /(^|[^\\])\\\|/, "\\1\\|", "g", \
335 substr(block, 1, match(block, /(\n|$)/)) \
337 block = substr(block, match(block, /(\n|$)/) + 1 );
340 for (cnt = 1; cnt < cols; cnt++)
341 ttext = ttext "<td align=\"" talign[cnt] "\">" inline(tarray[cnt]) "</td>"
342 ttext = ttext "</tr>\n"
344 return "<table>" ttext "</tbody></table>\n" _block(block);
346 # Grid Tables (pandoc)
347 } else if ( match(block, "^\\+(-+\\+)+\n" \
348 "(\\|([^\n]+\\|)+\n)+" \
349 "\\+(:?=+:?\\+)+\n" \
350 "((\\|([^\n]+\\|)+\n)+" \
351 "\\+(-+\\+)+(\n|$))+" \
353 len = RLENGTH; st = RSTART;
354 #initialize empty arrays
355 split("", talign); split("", tarray); split("", tread);
356 cols = 0; cnt=0; ttext = "";
358 # table header and alignment
359 block = substr(block, match(block, /(\n|$)/) + 1 );
360 while ( match(block, "^\\|([^\n]+\\|)+\n") ) {
361 cols = split( gensub( /(^\||\|$)/, "", "g", \
362 gensub( /(^|[^\\])\\\|/, "\\1\\|", "g", \
363 substr(block, 1, match(block, /(\n|$)/)) \
365 block = substr(block, match(block, /(\n|$)/) + 1 );
366 for (cnt = 1; cnt < cols; cnt++)
367 tarray[cnt] = tarray[cnt] "\n" tread[cnt];
371 gensub( /(^\+|\+$)/, "", "g", \
372 substr(block, 1, match(block, /(\n|$)/)) \
374 block = substr(block, match(block, /(\n|$)/) + 1 );
376 for (cnt = 1; cnt < cols; cnt++) {
377 if (match(talign[cnt], /:=+:/)) talign[cnt]="center";
378 else if (match(talign[cnt], /=+:/)) talign[cnt]="right";
379 else if (match(talign[cnt], /:=+/ )) talign[cnt]="left";
383 ttext = "<thead>\n<tr>"
384 for (cnt = 1; cnt < cols; cnt++)
385 ttext = ttext "<th align=\"" talign[cnt] "\">" _block(tarray[cnt]) "</th>"
386 ttext = ttext "</tr>\n</thead><tbody>\n"
388 while ( match(block, /^((\|([^\n]+\|)+\n)+\+(-+\+)+(\n|$))+/ ) ){
390 while ( match(block, /^\|([^\n]+\|)+\n/) ) {
391 split( gensub( /(^\||\|$)/, "", "g", \
392 gensub( /(^|[^\\])\\\|/, "\\1\\|", "g", \
393 substr(block, 1, match(block, /(\n|$)/)) \
395 block = substr(block, match(block, /(\n|$)/) + 1 );
396 for (cnt = 1; cnt < cols; cnt++)
397 tarray[cnt] = tarray[cnt] "\n" tread[cnt];
399 block = substr(block, match(block, /(\n|$)/) + 1 );
402 for (cnt = 1; cnt < cols; cnt++)
403 ttext = ttext "<td align=\"" talign[cnt] "\">" _block(tarray[cnt]) "</td>"
404 ttext = ttext "</tr>\n"
406 return "<table>" ttext "</tbody></table>\n" _block(block);
408 # Line Blocks (pandoc)
409 } else if ( match(block, /^\| [^\n]*(\n|$)(\| [^\n]*(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) {
410 len = RLENGTH; st = RSTART;
411 code = substr(block, 1, len);
412 gsub(/\n[[:space:]]+/, " ", code);
413 gsub(/\n\| /, "\n", code);
414 gsub(/^\| |\n$/, "", code);
415 return "<div class=\"line-block\">" gensub(/\n/, "<br />\n", "g", inline( code )) "</div>\n" \
416 _block( substr( block, len + 1) );
418 # Indented Code Block
419 } else if ( match(block, /^( |\t)( *\t*[^ \t\n])+(\n|$)(( |\t)[^\n]+(\n|$)|[ \t]*(\n|$))*/) ) {
420 len = RLENGTH; st = RSTART;
421 code = substr(block, 1, len);
422 gsub(/(^|\n)( |\t)/, "\n", code);
423 gsub(/^\n|\n+$/, "", code);
424 return "<pre><code>" HTML( code ) "</code></pre>\n" \
425 _block( substr( block, len + 1 ) );
427 # Fenced Divs (pandoc, custom)
428 } else if ( match( block, /^(:::+)/ ) ) {
429 guard = substr( block, 1, RLENGTH );
430 code = gensub(/^[^\n]+\n/, "", 1, block);
431 attrib = gensub(/^:::+[ \t]*\{?[ \t]*([^\}\n]*)\}?[ \t]*\n.*$/, "\\1", 1, block);
432 gsub(/[^a-zA-Z0-9_-]+/, " ", attrib);
433 gsub(/(^ | $)/, "", attrib);
434 if ( match(code, "(^|\n)" guard "+(\n|$)" ) ) {
435 len = RLENGTH; st = RSTART;
436 return "<div class=\"" attrib "\">" _block( substr(code, 1, st - 1) ) "</div>\n" \
437 _block( substr( code, st + len ) );
439 match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
440 len = RLENGTH; st = RSTART;
441 return "<p>" inline( substr(block, 1, st - 1) ) "</p>\n" \
442 _block( substr(block, st + len) );
445 # Fenced Code Block (pandoc)
446 } else if ( match( block, /^(~~~+|```+)/ ) ) {
447 guard = substr( block, 1, RLENGTH );
448 code = gensub(/^[^\n]+\n/, "", 1, block);
449 attrib = gensub(/^(~~~+|```+)[ \t]*\{?[ \t]*([^\}\n]*)\}?[ \t]*\n.*$/, "\\2", 1, block);
450 gsub(/[^a-zA-Z0-9_-]+/, " ", attrib);
451 gsub(/(^ | $)/, "", attrib);
452 if ( match(code, "(^|\n)" guard "+(\n|$)" ) ) {
453 len = RLENGTH; st = RSTART;
454 return "<pre><code class=\"" attrib "\">" HTML( substr(code, 1, st - 1) ) "</code></pre>\n" \
455 _block( substr( code, st + len ) );
457 match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
458 len = RLENGTH; st = RSTART;
459 return "<p>" inline( substr(block, 1, st - 1) ) "</p>\n" \
460 _block( substr(block, st + len) );
464 } else if ( match( block, "^ ? ? ?[-+*][ \t]+[^\n]+(\n|$)" \
465 "(([ \t]*\n)* ? ? ?[-+*][ \t]+[^\n]+(\n|$)" \
466 "|([ \t]*\n)*( ? ? ?\t| +)[^\n]+(\n|$)" \
467 "|[^\n]+(\n|$))*" ) ) {
468 list = substr( block, 1, RLENGTH);
469 block = substr( block, RLENGTH + 1);
470 indent = length( gensub(/[-+*][ \t]+[^\n]+.*$/, "", 1, list) );
472 gsub("(^|\n) {0," indent "}", "\n", list);
473 return "\n<ul>\n" _list( substr(list, 2) ) "</ul>\n" _block( block );
476 } else if ( match( block, "^ ? ? ?([0-9]+|#)\\.[ \t]+[^\n]+(\n|$)" \
477 "(([ \t]*\n)* ? ? ?([0-9]+|#)\\.[ \t]+[^\n]+(\n|$)" \
478 "|([ \t]*\n)*( ? ? ?\t| +)[^\n]+(\n|$)" \
479 "|[^\n]+(\n|$))*" ) ) {
480 list = substr( block, 1, RLENGTH);
481 block = substr( block, RLENGTH + 1);
482 indent = length( gensub(/([0-9]+|#)\.[ \t]+[^\n]+.*$/, "", 1, list) );
484 gsub("(^|\n) {0," indent "}", "\n", list);
485 return "\n<ol>\n" _list( substr(list, 2) ) "</ol>\n" _block( block );
487 # First Order Heading
488 } else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) {
490 HL[1]++; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
491 return "<h1 id=\"" HL[1] " - " HTML(gensub( /\n.*$/, "", "g", block )) "\">" \
492 inline( gensub( /\n.*$/, "", "g", block ) ) \
493 "<a class=\"anchor\" href=\"" HL[1] " - " \
494 HTML(gensub( /\n.*$/, "", "g", block )) "\"></a></h1>\n\n" \
495 _block( substr( block, len + 1 ) );
497 # Second Order Heading
498 } else if ( match( block, /^[^\n]+\n---+(\n|$)/ ) ) {
500 HL[2]++; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
501 return "<h2 id=\"" HL[1] "." HL[2] " - " HTML(gensub( /\n.*$/, "", "g", block )) "\">" \
502 inline( gensub( /\n.*$/, "", "g", block ) ) \
503 "<a class=\"anchor\" href=\"" HL[1] "." HL[2] " - " \
504 HTML(gensub( /\n.*$/, "", "g", block )) "\"></a></h2>\n\n" \
505 _block( substr( block, len + 1) );
508 } else if ( match( block, /^#{1,6}[ \t]*[^\n]+([ \t]*#*)(\n|$)/ ) ) {
510 hlvl = length( gensub( /^(#{1,6}).*$/, "\\1", "g", block ) );
511 htxt = gensub(/^#{1,6}[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[^\n#])+)([ \t]*#*)(\n.*)?$/, "\\1", 1, block);
512 HL[hlvl]++; for ( n = hlvl + 1; n < 7; n++) { HL[n] = 0;}
513 hid = HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; }
514 return "<h" hlvl " id=\"" hid " - " HTML(htxt) "\">" inline( htxt ) \
515 "<a class=\"anchor\" href=\"" hid "\"></a></h" hlvl ">\n\n" \
516 _block( substr( block, len + 1) );
519 } else if ( match( block, /(^|\n)[[:space:]]*(\n|$)/) ) {
520 len = RLENGTH; st = RSTART;
521 return _block( substr(block, 1, st - 1) ) "\n" \
522 _block( substr(block, st + len) );
525 } else if ( match( block, /(^|\n) ? ? ?((\* *){3,}|(- *){3,}|(_ *){3,})($|\n)/) ) {
526 len = RLENGTH; st = RSTART;
527 return _block(substr(block, 1, st - 1)) "<hr />\n" _block(substr(block, st + len));
531 return "<p>" inline(block) "</p>\n";
535 function _list( block, last, LOCAL, p) {
536 if ( ! length(block) ) return "";
537 gsub(/^([-+*]|[0-9]+\.|#\.)( ? ? ?|\t)/, "", block)
539 # slice next list item from input
540 if ( match( block, /\n([-+*]|[0-9]+\.|#\.)[ \t]+[^\n]+/) ) {
541 p = substr( block, 1, RSTART);
542 block = substr( block, RSTART + 1);
544 p = block; block = "";
546 sub( /\n +([-+*]|[0-9]+\.|#\.)/, "\n&", p );
548 # if this should be a paragraph item
549 # either previous item (last) or current item (p) contains blank lines
550 if (match(last, /\n[[:space:]]*\n/) || match(p, /\n[[:space:]]*\n/) ) {
551 last = p; p = _block(p);
553 last = p; p = _block(p);
554 sub( /^<p>/, "", p );
555 sub( /<\/p>\n/, "", p );
559 # Task List (pandoc, custom)
560 if ( p ~ /^\[ \].*/ ) { return "<li class=\"task pending\"><input type=checkbox disabled />" \
561 substr(p, 4) "</li>\n" _list( block, last );
562 } else if ( p ~ /^\[-\].*/ ) { return "<li class=\"task negative\"><input type=checkbox disabled />" \
563 substr(p, 4) "</li>\n" _list( block, last );
564 } else if ( p ~ /^\[\?\].*/ ) { return "<li class=\"task unsure\"><input type=checkbox disabled />" \
565 substr(p, 4) "</li>\n" _list( block, last );
566 } else if ( p ~ /^\[\/\].*/ ) { return "<li class=\"task partial\"><input type=checkbox disabled />" \
567 substr(p, 4) "</li>\n" _list( block, last );
568 } else if ( p ~ /^\[[xX]\].*/ ) { return "<li class=\"task done\"><input type=checkbox disabled checked />" \
569 substr(p, 4) "</li>\n" _list( block, last );
570 } else if ( p ~ /^<p>\[ \].*/ ) { return "<li class=\"task pending\"><p><input type=checkbox disabled />" \
571 substr(p, 7) "</li>\n" _list( block, last );
572 } else if ( p ~ /^<p>\[-\].*/ ) { return "<li class=\"task negative\"><p><input type=checkbox disabled />" \
573 substr(p, 7) "</li>\n" _list( block, last );
574 } else if ( p ~ /^<p>\[\?\].*/ ) { return "<li class=\"task unsure\"><p><input type=checkbox disabled />" \
575 substr(p, 7) "</li>\n" _list( block, last );
576 } else if ( p ~ /^<p>\[\/\].*/ ) { return "<li class=\"task partial\"><p><input type=checkbox disabled />" \
577 substr(p, 7) "</li>\n" _list( block, last );
578 } else if ( p ~ /^<p>\[[xX]\].*/ ) { return "<li class=\"task done\"><p><input type=checkbox disabled checked />" \
579 substr(p, 7) "</li>\n" _list( block, last );
580 } else { return "<li>" p "</li>\n" _list( block, last ); }
585 file = ""; rl_href[""] = ""; rl_title[""] = "";
586 if (ENVIRON["MD_HTML"] == "true") { AllowHTML = "true"; }
587 HL[1] = 0; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
589 # Buffering of full file ist necessary, e.g. to find reference links
590 while (getline) { file = file $0 "\n"; }
591 # Clean up MS-DOS line breaks
592 gsub(/\r\n/, "\n", file);
594 # Fill array of reference links
596 re_reflink = "(^|\n) ? ? ?\\[([^]\n]+)\\]: ([^ \t\n]+)(\n?[ \t]+(\"([^\"]+)\"|'([^']+)'|\\(([^)]+)\\)))?(\n|$)";
597 # /(^|\n) ? ? ?\[([^]\n]+)\]: ([^ \t\n]+)(\n?[ \t]+("([^"]+)"|'([^']+)'|\(([^)]+)\)))?(\n|$)/
598 while ( match(f, re_reflink ) ) {
599 rl_id = gensub( re_reflink, "\\2", 1, substr(f, RSTART, RLENGTH) );
600 rl_href[rl_id] = gensub( re_reflink, "\\3", 1, substr(f, RSTART, RLENGTH) );
601 rl_title[rl_id] = gensub( re_reflink, "\\5", 1, substr(f, RSTART, RLENGTH) );
602 f = substr(f, RSTART + RLENGTH);
603 rl_title[rl_id] = substr( rl_title[rl_id], 2, length(rl_title[rl_id]) - 2 );
604 if ( rl_href[rl_id] ~ /<.*>/ ) rl_href[rl_id] = substr( rl_href[rl_id], 2, length(rl_href[rl_id]) - 2 );
606 # Clear reflinks from File
607 while( gsub(re_reflink, "\n", file ) );
608 # for (n in rl_href) { debug(n " | " rl_href[n] " | " rl_title[n] ); }
610 # Run Block Processing -> The Actual Markdown!
611 printf "%s", _block( file );