markdown: avoid adding newlines to blockquote, css: style for blockquote

[cgilite] / markdown.awk

diff --git a/markdown.awk b/markdown.awk
index a2bb989bd3d6e0af3752578695656d37941ef67d..ffa225b477f6e8ff3dd1b157bc0f6e715df0a494 100755 (executable)
--- a/markdown.awk
+++ b/markdown.awk
@@ -40,6 +40,7 @@
 # ----------------------------
 # - [x] Automatic <section>-wrapping (custom)
 # -  ?  Heading identifiers (php md, pandoc)
+#   - [x] Heading attributes (custom)
 # - [x] Automatic heading identifiers (custom)
 # - [x] Fenced code blocks (php md, pandoc)
 #   - [x] Fenced code attributes
@@ -307,32 +308,35 @@ function inline( line, LOCAL, len, code, href, guard ) {
   }
 }
 
-function headline( hlvl, htxt, LOCAL, sec, n, HL) {
-  split( gensub( /^(.* )?(([0-9]+ ){5}[0-9]+)$/, "\\2" ,"1", hstack),  HL);
+function headline( hlvl, htxt, attrib, LOCAL, sec, n, HL) {
+  split( gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\2" ,"1", hstack),  HL);
 
   for ( n = hlvl; n <= 6; n++ ) { sec = sec (HL[n]?"</section>":""); }
   HL[hlvl]++; for ( n = hlvl + 1; n <= 6; n++) { HL[n] = 0;}
-  hid = HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; }
-  # hid = hid ":" URL(htxt, 1);
+  hid = ""; for ( n = 2; n <= blvl; n++) { hid = hid BL[n] "/"; }
+  hid = hid HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; }
+  hid = hid ":" URL(htxt, 1);
 
-  hstack = HL[1] " " HL[2] " " HL[3] " " HL[4] " " HL[5] " " HL[6];
+  hstack = gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\1" ,"1", hstack) \
+           HL[1] " " HL[2] " " HL[3] " " HL[4] " " HL[5] " " HL[6];
 
-  return sec "<section class=\"h1\"" ((blvl > 10)?"":" id=\"" hid "\"") ">" \
-         "<h1>" inline( htxt ) \
-         ((blvl > 0)?"":"<a class=\"anchor\" href=\"#" hid "\"></a>") \
-         "</h1>\n\n";
+  return sec "<section class=\"" (attrib ? "h" hlvl " " attrib : "h" hlvl)  "\" id=\"" hid "\">" \
+         "<h" hlvl (attrib ? " class=\"" attrib "\"" : "") ">" inline( htxt ) \
+         "<a class=\"anchor\" href=\"#" hid "\"></a>" \
+         "</h" hlvl ">\n\n";
 }
 
 # Nested Block, resets heading counters
-function _nblock( block, LOCAL, hlsav, sec ) {
+function _nblock( block, LOCAL, hlsav, sec, n ) {
   hlsav = hstack;
-  hstack = "0 0 0 0 0 0";
+  hstack = hstack " 0 0 0 0 0 0";
 
   # Block Level
-  blvl++;
+  blvl++; BL[blvl]++;
+  for ( n = blvl + 1; n in BL; n++) { delete BL[n]; }
 
   block = _block( block );
-  split( gensub( /^(.* )?(([0-9]+ ){5}[0-9]+)$/, "\2" ,"1", hstack),  HL);
+  split( gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\2" ,"1", hstack),  HL);
   sec = ""; for ( n = 1; n <= 6; n++ ) { sec = sec (HL[n]?"</section>":""); }
   hstack = hlsav; blvl--;
   return block sec;
@@ -376,7 +380,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, sec, guard, code, indent, at
   } else if ( match( block, /^> /) ) {
     match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match(block, /$/);
     len = RLENGTH; st = RSTART;
-    return "<blockquote>\n" _nblock( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "</blockquote>\n\n" \
+    return "<blockquote>" gensub( /^\n|\n$/, "", "g",  _nblock( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) ) "</blockquote>\n\n" \
            _block( substr(block, st + len) );
 
   # Pipe Tables (pandoc / php md / gfm )
@@ -586,28 +590,60 @@ function _block( block, LOCAL, st, len, hlvl, htxt, sec, guard, code, indent, at
   gsub("(^|\n) {0," indent "}", "\n", list);
   return "\n<ol>\n" _list( substr(list, 2) ) "</ol>\n" _block( block );
 
+  # First Order Heading H1 + Attrib
+  } else if ( match( block, /^([^\n]+)([ \t]*\{([^\}\n]+)\})\n===+(\n|$)/ ) ) {
+    len = RLENGTH;
+    text   = gensub(/^([^\n]+)([ \t]*\{([^\}\n]+)\})\n===+(\n.*)?$/, "\\1", 1, block)
+    attrib = gensub(/^([^\n]+)([ \t]*\{([^\}\n]+)\})\n===+(\n.*)?$/, "\\3", 1, block)
+    gsub(/[^a-zA-Z0-9_-]+/, " ", attrib); gsub(/(^ | $)/, "", attrib);
+
+    return headline(1, text, attrib) \
+           _block( substr( block, len + 1 ) );
+
   # First Order Heading H1
-  } else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) {
+  } else if ( match( block, /^([^\n]+)\n===+(\n|$)/ ) ) {
     len = RLENGTH;
+    text   = gensub(/^([^\n]+)\n===+(\n.*)?$/, "\\1", 1, block)
 
-    return headline(1, gensub( /\n.*$/, "", "g", block )) \
+    return headline(1, text, 0) \
            _block( substr( block, len + 1 ) );
 
+  # Second Order Heading H2 + Attrib
+  } else if ( match( block, /^([^\n]+)([ \t]*\{([^\}\n]+)\})\n---+(\n|$)/ ) ) {
+    len = RLENGTH;
+    text   = gensub(/^([^\n]+)([ \t]*\{([^\}\n]+)\})\n---+(\n.*)?$/, "\\1", 1, block)
+    attrib = gensub(/^([^\n]+)([ \t]*\{([^\}\n]+)\})\n---+(\n.*)?$/, "\\3", 1, block)
+    gsub(/[^a-zA-Z0-9_-]+/, " ", attrib); gsub(/(^ | $)/, "", attrib);
+
+    return headline(2, text, attrib) \
+           _block( substr( block, len + 1) );
+
   # Second Order Heading H2
-  } else if ( match( block, /^[^\n]+\n---+(\n|$)/ ) ) {
+  } else if ( match( block, /^([^\n]+)\n---+(\n|$)/ ) ) {
+    len = RLENGTH;
+    text   = gensub(/^([^\n]+)\n---+(\n.*)?$/, "\\1", 1, block)
+
+    return headline(2, text, 0) \
+           _block( substr( block, len + 1) );
+
+  # Nth Order Heading H1 H2 H3 H4 H5 H6 + Attrib
+  } else if ( match( block, /^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*([ \t]*\{([a-zA-Z \t-]*)\})(\n|$)/ ) ) {
     len = RLENGTH;
+    n      = gensub(/^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*([ \t]*\{([a-zA-Z \t-]*)\})(\n.*)?$/, "\\1", 1, block);
+    text   = gensub(/^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*([ \t]*\{([a-zA-Z \t-]*)\})(\n.*)?$/, "\\2", 1, block);
+    attrib = gensub(/^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*([ \t]*\{([a-zA-Z \t-]*)\})(\n.*)?$/, "\\5", 1, block);
+    gsub(/[^a-zA-Z0-9_-]+/, " ", attrib); gsub(/(^ | $)/, "", attrib);
 
-    return headline(2, gensub( /\n.*$/, "", "g", block )) \
+    return headline( length(n), text, attrib ) \
            _block( substr( block, len + 1) );
 
   # Nth Order Heading H1 H2 H3 H4 H5 H6
-  } else if ( match( block, /^#{1,6}[ \t]*[^\n]+([ \t]*#*)(\n|$)/ ) ) {
+  } else if ( match( block, /^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*(\n|$)/ ) ) {
     len = RLENGTH;
+    n      = gensub(/^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*(\n.*)?$/, "\\1", 1, block);
+    text   = gensub(/^(#{1,6})[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[ \t]*[^ \t\n#])+)[ \t]*#*(\n.*)?$/, "\\2", 1, block);
 
-    return headline( \
-             length( gensub( /^(#{1,6}).*$/, "\\1", "g", block ) ), \
-             gensub(/^#{1,6}[ \t]*(([^ \t\n]+|[ \t]+[^ \t\n#]|[ \t]+#+[^\n#])+)([ \t]*#*)(\n.*)?$/, "\\1", 1, block) \
-           ) \
+    return headline( length(n), text, 0 ) \
            _block( substr( block, len + 1) );
 
   # block images (wrapped in <figure>)