From: Paul Hänsch <paul@plutz.net>
Date: Tue, 29 Jun 2021 16:58:24 +0000 (+0200)
Subject: include markdown processor
X-Git-Url: https://git.plutz.net/?a=commitdiff_plain;h=3d2264ce0ed3d74b084531e364e27e5427ac1cf4;p=cgilite

include markdown processor
---

diff --git a/markdown.awk b/markdown.awk
new file mode 100755
index 0000000..134002d
--- /dev/null
+++ b/markdown.awk
@@ -0,0 +1,336 @@
+#!/bin/awk -f
+#!/opt/busybox/awk -f
+
+# 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:
+# ==========================
+# [x] done    [ ] todo    [-] not planned    [?] unsure
+#
+# Basic Markdown - Block elements:
+# -------------------------------
+# - [x] Paragraphs
+#   - [ ] Double space line breaks
+# - [x] Proper block element nesting
+# - [x] Headings
+# - [x] ATX-Style Headings
+# - [x] Blockquotes
+# - [x] Lists (ordered, unordered)
+# - [x] Code blocks (using indention)
+# - [ ] Horizontal rules
+#
+# Basic Markdown - Inline elements:
+# ---------------------------------
+# - [x] Links
+# - [x] Reference style links
+# - [x] Emphasis *em*/**strong** (*Asterisk*, _Underscore_)
+# - [x] `code`, also ``code containing `backticks` ``
+# - [ ] Images / reference style images
+# - [x] <automatic links>
+# - [x] backslash escapes
+#
+# Extensions - Block elements:
+# ----------------------------
+# - [?] Heading identifiers (phpmarkdown, pandoc)
+# - [x] Fenced code blocks (phpmarkdown, pandoc)
+#   - [-] Fenced code attributes
+# - [ ] Tables
+#   - [?] Simple table (pandoc)
+#   - [?] Multiline table (pandoc)
+#   - [?] Grid table (pandoc)
+#   - [?] Pipe table (phpmarkdown, pandoc)
+# - [x] Line blocks (pandoc)
+# - [ ] Task lists (pandoc)
+# - [ ] Definition lists (phpmarkdown, pandoc)
+# - [-] Numbered example lists (pandoc)
+# - [-] Metadata blocks (pandoc)
+# - [-] Fenced Divs (pandoc)
+#
+# Extensions - Inline elements:
+# ----------------------------
+# - [ ] Ignore embedded_underscores (phpmarkdown, pandoc)
+# - [x] ~~strikeout~~ (pandoc)
+# - [x] ^Superscript^ ~Subscript~ (pandoc)
+# - [-] Bracketed spans (pandoc)
+#   - [-] Inline attributes (pandoc)
+# - [-] TEX-Math (pandoc)
+# - [?] Footnotes (phpmarkdown)
+# - [?] Abbreviations (phpmarkdown)
+# - [?] "Curly quotes" (smartypants)
+# - [ ] em-dashes (--) (smartypants old)
+# - [?] ... three-dot ellipsis (smartypants)
+# - [-] en-dash (smartypants)
+# - [ ] Automatic em-dash / en-dash
+
+function debug(text) { printf "\n---\n%s\n---\n", text > "/dev/stderr"; }
+
+function HTML ( text ) {
+  gsub( /&/,  "\\&amp;",  text );
+  gsub( /</,  "\\&lt;",   text );
+  gsub( />/,  "\\&gt;",   text );
+  gsub( /"/,  "\\&quot;", text );
+  gsub( /'/,  "\\&#x27;", text );
+  gsub( /\\/, "\\&#x5C;", text );
+  return text;
+}
+
+function inline( line, LOCAL, len, code, href, guard ) {
+  nu = "(\\\\\\\\|\\\\[^\\\\]|[^\\\\_])*"    # not underline (except when escaped)
+  na = "(\\\\\\\\|\\\\[^\\\\]|[^\\\\\\*])*"  # not asterisk (except when escaped)
+  ieu =  "_([^_[:space:]]|[^_[:space:]]" nu "[^_[:space:]])_"                 # inner <em> (underline)
+  isu = "__([^_[:space:]]|[^_[:space:]]" nu "[^_[:space:]])__"                # inner <strong> (underline)
+  iea =    "\\*([^\\*[:space:]]|[^\\*[:space:]]" na "[^\\*[:space:]])\\*"     # inner <em> (asterisk)
+  isa = "\\*\\*([^\\*[:space:]]|[^\\*[:space:]]" na "[^\\*[:space:]])\\*\\*"  # inner <strong> (asterisk)
+
+  if ( line ~ /^$/ ) {  # Recursion End
+    return "";
+
+  #  omit processing of escaped characters
+  } else if ( line ~ /^\\[]\\`\*_\{\}\(\)#\+-\.![]/) {
+    return substr(line, 2, 1) inline( substr(line, 3) );
+
+  #  ``code spans``
+  } else if ( match( line, /^`+/) ) {
+    len = RLENGTH
+    guard = substr( line, 1, len )
+    if ( match(line, guard ".*" guard) ) {
+      code = substr( line, len + 1, match( substr(line, len + 1), guard ) - 1)
+      len = 2 * length(guard) + length(code)
+      #  strip single surrounding white spaces
+      code = gensub( / (.*) /, "\\1", "1" , code)
+      #  escape HTML within code span
+      gsub( /&/, "\\&amp;", code ); gsub( /</, "\\&lt;", code ); gsub( />/, "\\&gt;", code );
+      return "<code>" code "</code>" inline( substr( line, len + 1 ) )
+    }
+
+  #  quick links ("automatic links" in md doc)
+  } else if ( match( line, /^<[a-zA-Z]+:\/\/([-\.[:alnum:]]+)(:[0-9]*)?(\/[^>]*)?>/ ) ) {
+    len = RLENGTH;
+    href = HTML( substr( line, 2, len - 2) );
+    return "<a href=\"" href "\">" href "</a>" inline( substr( line, len + 1) );
+
+  # inline links
+  } 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);
+    if ( title ) {
+      return "<a href=\"" HTML(href) "\" title=\"" HTML(title) "\">" inline( text ) "</a>" inline( substr( line, len + 1) );
+    } else {
+      return "<a href=\"" HTML(href) "\">" inline( text ) "</a>" inline( substr( line, len + 1) );
+    }
+
+  # reference style links
+  } else if ( match(line, /^\[([^]]+)\] ?\[([^]]*)\]/ ) ) {
+    len = RLENGTH;
+    text = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\1", 1, line);
+      id = gensub(/^\[([^\n]+)\] ?\[([^\n]*)\].*/, "\\2", 1, line);
+    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) );
+    } else if ( rl_href[id] ) {
+      return "<a href=\"" rl_href[id] "\">" inline(text) "</a>" inline( substr( line, len + 1) );
+    } else {
+      return "" substr(line, 1, len) inline( substr(line, len + 1) );
+    }
+
+  #  ~~strikeout~~ (pandoc)
+  } else if ( match(line, /^~~([[:graph:]]|[[:graph:]]([^~]|~[^~])*[[:graph:]])~~/) ) {
+    len = RLENGTH;
+    return "<del>" inline( substr( line, 3, len - 4 ) ) "</del>" inline( substr( line, len + 1 ) );
+
+  #  ^superscript^ (pandoc)
+  } else if ( match(line, /^\^([^[:space:]^]|\\[ ^])+\^/) ) {
+    len = RLENGTH;
+    return "<sup>" inline( substr( line, 2, len - 2 ) ) "</sup>" inline( substr( line, len + 1 ) );
+
+  #  ~subscript~ (pandoc)
+  } else if ( match(line, /^~([^[:space:]~]|\\[ ~])+~/) ) {
+    len = RLENGTH;
+    return "<sub>" inline( substr( line, 2, len - 2 ) ) "</sub>" inline( substr( line, len + 1 ) );
+
+  #  __strong__
+  } else if ( match(line, "^__(([^_[:space:]]|" ieu ")|([^_[:space:]]|" ieu ")(" nu "|" ieu ")*([^_[:space:]]|" ieu "))__") ) {
+    len = RLENGTH;
+    return "<strong>" inline( substr( line, 3, len - 4 ) ) "</strong>" inline( substr( line, len + 1 ) );
+
+  #  **strong**
+  } else if ( match(line, "^\\*\\*(([^\\*[:space:]]|" iea ")|([^\\*[:space:]]|" iea ")(" na "|" iea ")*([^\\*[:space:]]|" iea "))\\*\\*") ) {
+    len = RLENGTH;
+    return "<strong>" inline( substr( line, 3, len - 4 ) ) "</strong>" inline( substr( line, len + 1 ) );
+
+  #  _em_
+  } else if ( match(line, "^_(([^_[:space:]]|" isu ")|([^_[:space:]]|" isu ")(" nu "|" isu ")*([^_[:space:]]|" isu "))_") ) {
+    len = RLENGTH;
+    return "<em>" inline( substr( line, 2, len - 2 ) ) "</em>" inline( substr( line, len + 1 ) );
+
+  #  *em*
+  } else if ( match(line, "^\\*(([^\\*[:space:]]|" isa ")|([^\\*[:space:]]|" isa ")(" na "|" isa ")*([^\\*[:space:]]|" isa "))\\*") ) {
+    len = RLENGTH;
+    return "<em>" inline( substr( line, 2, len - 2 ) ) "</em>" inline( substr( line, len + 1 ) );
+
+  #  continue walk over string
+  } else {
+    return substr(line, 1, 1) inline( substr(line, 2) );
+  }
+}
+
+function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent ) {
+  gsub( /^\n+|\n+$/, "", block );
+
+  if ( block == "" ) {
+    return "";
+ 
+  # Blockquote (leading >)
+  } else if ( match( block, /^> /) ) {
+    match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match(block, /$/);
+    len = RLENGTH; st = RSTART;
+    return "<blockquote>\n" _block( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "</blockquote>\n\n" \
+           _block( substr(block, st + len) );
+
+  # Line Blocks (pandoc)
+  } else if ( match(block, /^\| [^\n]*(\n|$)(\| [^\n]*(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) {
+    len = RLENGTH; st = RSTART;
+    code = substr(block, 1, len);
+    gsub(/\n[[:space:]]+/, " ", code);
+    gsub(/\n\| /, "\n", code);
+    gsub(/^\| |\n$/, "", code);
+    return "<div class=\"line-block\">" gensub(/\n/, "<br />\n", "g", inline( code )) "</div>\n" \
+           _block( substr( block, len + 1) );
+
+  # Indented Code Block
+  } else if ( match(block, /^(    |\t)[^\n]+(\n|$)((    |\t)[^\n]+(\n|$)|[ \t]*(\n|$))*/) ) {
+    len = RLENGTH; st = RSTART;
+    code = substr(block, 1, len);
+    gsub(/(^|\n)(    |\t)/, "\n", code);
+    gsub(/^\n|\n+$/, "", code);
+    return "<pre><code>" HTML( code ) "</code></pre>\n" \
+           _block( substr( block, len + 1 ) );
+
+  # Fenced Code Block (pandoc)
+  } else if ( match( block, /^(~~~+|```+)/ ) ) {
+    guard = substr( block, 1, RLENGTH );
+    code = gensub(/^[^\n]+\n/, "", 1, block);
+    if ( match(code, "(^|\n)" guard "+(\n|$)" ) ) {
+      len = RLENGTH; st = RSTART;
+      return "<pre><code>" HTML( substr(code, 1, st - 1) ) "</code></pre>\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) );
+    }
+
+  # Unordered list
+  } else if ( match( block, "^ ? ? ?[-+*][ \t]+[^\n]+(\n|$)" \
+                            "(([ \t]*\n)* ? ? ?[-+*][ \t]+[^\n]+(\n|$)" \
+                            "|([ \t]*\n)*( ? ? ?\t|  +)[^\n]+(\n|$)" \
+                            "|[^\n]+(\n|$))*" ) ) {
+  list = substr( block, 1, RLENGTH);
+  block = substr( block, RLENGTH + 1);
+  indent = length( gensub(/[-+*][ \t]+[^\n]+.*$/, "", 1, list) );
+
+  gsub("(^|\n) {0," indent "}", "\n", list);
+  return "\n<ul>\n" _list( substr(list, 2) ) "</ul>\n" _block( block );
+
+  # Ordered list
+  } else if ( match( block, "^ ? ? ?([0-9]+|#)\\.[ \t]+[^\n]+(\n|$)" \
+                            "(([ \t]*\n)* ? ? ?([0-9]+|#)\\.[ \t]+[^\n]+(\n|$)" \
+                            "|([ \t]*\n)*( ? ? ?\t|  +)[^\n]+(\n|$)" \
+                            "|[^\n]+(\n|$))*" ) ) {
+  list = substr( block, 1, RLENGTH);
+  block = substr( block, RLENGTH + 1);
+  indent = length( gensub(/([0-9]+|#)\.[ \t]+[^\n]+.*$/, "", 1, list) );
+
+  gsub("(^|\n) {0," indent "}", "\n", list);
+  return "\n<ol>\n" _list( substr(list, 2) ) "</ol>\n" _block( block );
+
+  # First Order Heading
+  } else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) {
+    len = RLENGTH;
+    return "<h1>" inline( gensub( /\n.*$/, "", "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" \
+           _block( substr( block, len + 1) );
+
+  # Nth Order Heading
+  } else if ( match( block, /^#{1,6}[[:space:]]*[^\n]+([[:space:]]*#*)(\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" \
+           _block( substr( block, len + 1) );
+
+  # Plain paragraph
+  } 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) );
+  }
+}
+
+function _list( block, last, LOCAL, p) {
+  if ( ! length(block) ) return "";
+  gsub(/^([-+*]|[0-9]+\.|#\.)(  ? ? ?|\t)/, "", block)
+
+  # slice next list item from input
+  if ( match( block, /\n([-+*]|[0-9]+\.|#\.)[ \t]+[^\n]+/) ) {
+    p = substr( block, 1, RSTART);
+    block = substr( block, RSTART + 1);
+  } else {
+    p = block; block = "";
+  }
+  sub( /\n +([-+*]|[0-9]+\.|#\.)/, "\n&", p );
+
+  # if this should be a paragraph item
+  # either previous item (last) or current item (p) contains blank lines
+  if (match(last, /\n[[:space:]]*\n/) || match(p, /\n[[:space:]]*\n/) ) {
+    last = p; p = _block(p);
+  } else {
+    last = p; p = _block(p);
+    sub( /^<p>/, "", p );
+    sub( /<\/p>\n/, "", p );
+  }
+  sub( /\n$/, "", p );
+  return "<li>" p "</li>\n" _list( block, last );
+}
+
+BEGIN {
+  # Global Vars
+  file = ""; rl_href[""] = ""; rl_title[""] = "";
+
+  # Buffering of full file ist necessary, e.g. to find reference links
+  while (getline) { file = file $0 "\n"; }
+
+  # Fill array of reference links
+  f = file; rl_id;
+  re_reflink = "(^|\n) ? ? ?\\[([^]\n]+)\\]: ([^ \t\n]+)(\n?[ \t]+(\"([^\"]+)\"|'([^']+)'|\\(([^)]+)\\)))?(\n|$)";
+  # /(^|\n) ? ? ?\[([^]\n]+)\]: ([^ \t\n]+)(\n?[ \t]+("([^"]+)"|'([^']+)'|\(([^)]+)\)))?(\n|$)/
+  while ( match(f, re_reflink ) ) {
+    rl_id           = gensub( re_reflink, "\\2", 1, substr(f, RSTART, RLENGTH) );
+    rl_href[rl_id]  = gensub( re_reflink, "\\3", 1, substr(f, RSTART, RLENGTH) );
+    rl_title[rl_id] = gensub( re_reflink, "\\5", 1, substr(f, RSTART, RLENGTH) );
+    f = substr(f, RSTART + RLENGTH);
+    rl_title[rl_id] = substr( rl_title[rl_id], 2, length(rl_title[rl_id]) - 2 );
+    if ( rl_href[rl_id] ~ /<.*>/ ) rl_href[rl_id] = substr( rl_href[rl_id], 2, length(rl_href[rl_id]) - 2 );
+  }
+  # Clear reflinks from File
+  while( gsub(re_reflink, "\n", file ) );
+  # for (n in rl_href) { debug(n " | " rl_href[n] " | " rl_title[n] ); }
+
+  # Run Block Processing -> The Actual Markdown!
+  printf "%s", _block( file );
+}