# EXPERIMENTAL Markdown processor with minimal dependencies.
# Meant to support all features of John Grubers basic Markdown
# + a number of common extensions, mostly inspired by Pandoc Markdown
-#
-# ToDo:
-# - HTML processing / escaping (according to environment flag)
-# - em-dashes and arrows
# Supported Features / TODO:
# ==========================
# Extensions - Block elements:
# ----------------------------
# - ? Heading identifiers (php md, pandoc)
+# - [x] Automatic heading identifiers (custom)
# - [x] Fenced code blocks (php md, pandoc)
-# - [-] Fenced code attributes
+# - [x] Fenced code attributes
# - [ ] Tables
# - ? Simple table (pandoc)
# - ? Multiline table (pandoc)
# - ? Grid table (pandoc)
# - ? Pipe table (php md pandoc)
# - [x] Line blocks (pandoc)
-# - [ ] Task lists (pandoc)
+# - [x] Task lists (pandoc)
# - [ ] Definition lists (php md, pandoc)
# - [-] Numbered example lists (pandoc)
# - [-] Metadata blocks (pandoc)
-# - [-] Fenced Divs (pandoc)
+# - [x] Metadata blocks (custom)
+# - [x] Fenced Divs (pandoc)
#
# Extensions - Inline elements:
# ----------------------------
href = HTML( substr( line, 2, len - 2) );
return "<a href=\"" href "\">" href "</a>" inline( substr( line, len + 1) );
+ # quick link email
+ } 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])?)*>/ ) ) {
+ len = RLENGTH;
+ href = HTML( substr( line, 2, len - 2) );
+ return "<a href=\"mailto:" href "\">" href "</a>" inline( substr( line, len + 1) );
+
# inline links
- } else if ( match(line, /^\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/) ) {
+ } else if ( match(line, /^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/) ) {
len = RLENGTH;
- text = gensub(/^\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\1", "g", line);
- href = gensub(/^\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\2", "g", line);
- title = gensub(/^\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\4", "g", line);
+ text = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\1", 1, substr(line, 1, len) );
+ href = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\2", 1, substr(line, 1, len) );
+ title = gensub(/^\[([^]]+)\]\(([^"\)]+)([[:space:]]+"([^"]+)")?\)/, "\\4", 1, substr(line, 1, len) );
if ( title ) {
return "<a href=\"" HTML(href) "\" title=\"" HTML(title) "\">" inline( text ) "</a>" inline( substr( line, len + 1) );
} else {
# reference style links
} else if ( match(line, /^\[([^]]+)\] ?\[([^]]*)\]/ ) ) {
len = RLENGTH;
- text = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, line);
- id = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, line);
+ text = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, substr(line, 1, len) );
+ id = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, substr(line, 1, len) );
if ( ! id ) id = text;
if ( rl_href[id] && rl_title[id] ) {
- return "<a href=\"" rl_href[id] "\" title=\"" rl_title[id] "\">" inline(text) "</a>" inline( substr( line, len + 1) );
+ return "<a href=\"" HTML(rl_href[id]) "\" title=\"" HTML(rl_title[id]) "\">" inline(text) "</a>" inline( substr( line, len + 1) );
} else if ( rl_href[id] ) {
- return "<a href=\"" rl_href[id] "\">" inline(text) "</a>" inline( substr( line, len + 1) );
+ return "<a href=\"" HTML(rl_href[id]) "\">" inline(text) "</a>" inline( substr( line, len + 1) );
} else {
- return "" substr(line, 1, len) inline( substr(line, len + 1) );
+ return "" HTML(substr(line, 1, len)) inline( substr(line, len + 1) );
}
# inline images
} else if ( match(line, /^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/) ) {
len = RLENGTH;
- text = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\1", "g", line);
- href = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\2", "g", line);
- title = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\4", "g", line);
+ text = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\1", "g", substr(line, 1, len) );
+ href = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\2", "g", substr(line, 1, len) );
+ title = gensub(/^!\[([^]]+)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)/, "\\4", "g", substr(line, 1, len) );
if ( title ) {
return "<img src=\"" HTML(href) "\" alt=\"" HTML(text) "\" title=\"" HTML(title) "\" />" inline( substr( line, len + 1) );
} else {
# reference style images
} else if ( match(line, /^!\[([^]]+)\] ?\[([^]]*)\]/ ) ) {
len = RLENGTH;
- text = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, line);
- id = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, line);
+ text = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, substr(line, 1, len) );
+ id = gensub(/^!\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, substr(line, 1, len) );
if ( ! id ) id = text;
if ( rl_href[id] && rl_title[id] ) {
- return "<img src=\"" rl_href[id] "\" alt=\"" HTML(text) "\" title=\"" rl_title[id] "\" />" inline( substr( line, len + 1) );
+ return "<img src=\"" HTML(rl_href[id]) "\" alt=\"" HTML(text) "\" title=\"" HTML(rl_title[id]) "\" />" inline( substr( line, len + 1) );
} else if ( rl_href[id] ) {
- return "<img src=\"" rl_href[id] "\" alt=\"" HTML(text) "\" />" inline( substr( line, len + 1) );
+ return "<img src=\"" HTML(rl_href[id]) "\" alt=\"" HTML(text) "\" />" inline( substr( line, len + 1) );
} else {
- return "" substr(line, 1, len) inline( substr(line, len + 1) );
+ return "" HTML(substr(line, 1, len)) inline( substr(line, len + 1) );
}
# ~~strikeout~~ (pandoc)
# ignore embedded underscores (pandoc, php md)
} else if ( match(line, "^[[:alnum:]](__|_)") ) {
- return substr( line, 1, RLENGTH) inline( substr(line, RLENGTH + 1) );
+ return HTML(substr( line, 1, RLENGTH)) inline( substr(line, RLENGTH + 1) );
# __strong__$
} else if ( match(line, "^__(([^_[:space:]]|" ieu ")|([^_[:space:]]|" ieu ")(" nu "|" ieu ")*([^_[:space:]]|" ieu "))__$") ) {
len = RLENGTH;
return "<em>" inline( substr( line, 2, len - 2 ) ) "</em>" inline( substr( line, len + 1 ) );
+ # Macros
+ } else if ( AllowMacros && match( line, /^<<([^>]|>[^>])+>>/) ) {
+ len = RLENGTH;
+ return macro( substr( line, 3, len - 4 ) ) inline(substr(line, len + 1));
+
# Verbatim inline HTML
} 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:]]*\/?>)/) ) {
len = RLENGTH;
}
}
-function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent ) {
+function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ) {
gsub( /^\n+|\n+$/, "", block );
if ( block == "" ) {
len = RLENGTH; st = RSTART;
return substr(block, st, len) _block(substr(block, st + len));
- # Horizontal rule
- } else if ( match( block, /(^|\n) ? ? ?((\* *){3,}|(- *){3,}|(_ *){3,})($|\n)/) ) {
+ # Metadata (custom, block starting with %something)
+ # Metadata is ignored but can be interpreted externally
+ } else if ( match(block, /^%[a-zA-Z]+([[:space:]][^\n]*)?(\n|$)(%[a-zA-Z]+([[:space:]][^\n]*)?(\n|$)|%([[:space:]][^\n]*)?(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) {
len = RLENGTH; st = RSTART;
- return _block(substr(block, 1, st - 1)) "<hr />\n" _block(substr(block, st + len));
+ return _block( substr( block, len + 1) );
# Blockquote (leading >)
} else if ( match( block, /^> /) ) {
return "<pre><code>" HTML( code ) "</code></pre>\n" \
_block( substr( block, len + 1 ) );
+ # Fenced Divs (pandoc, custom)
+ } else if ( match( block, /^(:::+)/ ) ) {
+ guard = substr( block, 1, RLENGTH );
+ code = gensub(/^[^\n]+\n/, "", 1, block);
+ attrib = gensub(/^:::+[ \t]*\{?[ \t]*([^\}\n]*)\}?[ \t]*\n.*$/, "\\1", 1, block);
+ gsub(/[^a-zA-Z0-9_-]+/, " ", attrib);
+ gsub(/(^ | $)/, "", attrib);
+ if ( match(code, "(^|\n)" guard "+(\n|$)" ) ) {
+ len = RLENGTH; st = RSTART;
+ return "<div class=\"" attrib "\">" _block( substr(code, 1, st - 1) ) "</div>\n" \
+ _block( substr( code, st + len ) );
+ } else {
+ match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
+ len = RLENGTH; st = RSTART;
+ return "<p>" inline( substr(block, 1, st - 1) ) "</p>\n" \
+ _block( substr(block, st + len) );
+ }
+
# Fenced Code Block (pandoc)
} else if ( match( block, /^(~~~+|```+)/ ) ) {
guard = substr( block, 1, RLENGTH );
code = gensub(/^[^\n]+\n/, "", 1, block);
+ attrib = gensub(/^:::+[ \t]*\{?[ \t]*([^\}\n]*)\}?[ \t]*\n.*$/, "\\1", 1, block);
+ gsub(/[^a-zA-Z0-9_-]+/, " ", attrib);
+ gsub(/(^ | $)/, "", attrib);
if ( match(code, "(^|\n)" guard "+(\n|$)" ) ) {
len = RLENGTH; st = RSTART;
- return "<pre><code>" HTML( substr(code, 1, st - 1) ) "</code></pre>\n" \
+ return "<pre><code class=\"" attrib "\">" HTML( substr(code, 1, st - 1) ) "</code></pre>\n" \
_block( substr( code, st + len ) );
} else {
match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
# First Order Heading
} else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) {
len = RLENGTH;
- return "<h1>" inline( gensub( /\n.*$/, "", "g", block ) ) "</h1>\n\n" \
+ HL[1]++; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
+ return "<h1 id=\"" HL[1] " - " HTML(gensub( /\n.*$/, "", "g", block )) "\">" inline( gensub( /\n.*$/, "", "g", block ) ) "</h1>\n\n" \
_block( substr( block, len + 1 ) );
# Second Order Heading
} else if ( match( block, /^[^\n]+\n---+(\n|$)/ ) ) {
len = RLENGTH;
- return "<h2>" inline( gensub( /\n.*$/, "", "g", block ) ) "</h2>\n\n" \
+ HL[2]++; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
+ return "<h2 id=\"" HL[1] "." HL[2] " - " HTML(gensub( /\n.*$/, "", "g", block )) "\">" inline( gensub( /\n.*$/, "", "g", block ) ) "</h2>\n\n" \
_block( substr( block, len + 1) );
# Nth Order Heading
- } else if ( match( block, /^#{1,6}[[:space:]]*[^\n]+([[:space:]]*#*)(\n|$)/ ) ) {
+ } else if ( match( block, /^#{1,6}[ \t]*[^\n]+([ \t]*#*)(\n|$)/ ) ) {
len = RLENGTH;
hlvl = length( gensub( /^(#{1,6}).*$/, "\\1", "g", block ) );
- htxt = gensub( /[[:space:]]*#*$/, "", "1", gensub( /^#{1,6}[[:space:]]*([^\n]+)([[:space:]]*#*)\n.*$/, "\\1", "g", block ) )
- return "<h" hlvl ">" inline( htxt ) "</h" hlvl ">\n\n" \
+ htxt = gensub(/^#{1,6}[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[^\n#])+)([ \t]*#*)(\n.*)?$/, "\\1", 1, block);
+ HL[hlvl]++; for ( n = hlvl + 1; n < 7; n++) { HL[n] = 0;}
+ hid = HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; }
+ return "<h" hlvl " id=\"" hid " - " HTML(htxt) "\">" inline( htxt ) "</h" hlvl ">\n\n" \
_block( substr( block, len + 1) );
- # Plain paragraph
- } else {
- match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
+ # Split paragraphs
+ } else if ( match( block, /(^|\n)[[:space:]]*(\n|$)/) ) {
len = RLENGTH; st = RSTART;
- return "<p>" inline( substr(block, 1, st - 1) ) "</p>\n" \
+ return _block( substr(block, 1, st - 1) ) "\n" \
_block( substr(block, st + len) );
+
+ # Horizontal rule
+ } else if ( match( block, /(^|\n) ? ? ?((\* *){3,}|(- *){3,}|(_ *){3,})($|\n)/) ) {
+ len = RLENGTH; st = RSTART;
+ return _block(substr(block, 1, st - 1)) "<hr />\n" _block(substr(block, st + len));
+
+ # Plain paragraph
+ } else {
+ return "<p>" inline(block) "</p>\n";
}
}
sub( /<\/p>\n/, "", p );
}
sub( /\n$/, "", p );
+
+ # Task List (pandoc)
+ if ( p ~ /^\[ \].*/ ) { p = "<input type=checkbox disabled />" substr(p, 4); }
+ else if ( p ~ /^\[[xX]\].*/ ) { p = "<input type=checkbox disabled checked />" substr(p, 4); }
+ else if ( p ~ /^<p>\[ \].*/ ) { p = "<p><input type=checkbox disabled />" substr(p, 7); }
+ else if ( p ~ /^<p>\[[xX]\].*/ ) { p = "<p><input type=checkbox disabled checked />" substr(p, 7); }
return "<li>" p "</li>\n" _list( block, last );
}
# Global Vars
file = ""; rl_href[""] = ""; rl_title[""] = "";
if (ENVIRON["MD_HTML"] == "true") { AllowHTML = "true"; }
+ HL[1] = 0; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
# Buffering of full file ist necessary, e.g. to find reference links
while (getline) { file = file $0 "\n"; }
+ # Clean up MS-DOS line breaks
+ gsub(/\r\n/, "\n", file);
# Fill array of reference links
f = file; rl_id;