EXPERIMENTAL: <section> wrapping for Headlines

diff --git a/markdown.awk b/markdown.awk
index 3fa248d25f2877fdc0e6f7eab9a5642635407d75..fbfc455355c14c6df2d34cd44a42daa8dee09794 100755 (executable)
--- a/markdown.awk
+++ b/markdown.awk
@@ -306,7 +306,18 @@ function inline( line, LOCAL, len, code, href, guard ) {
   }
 }
 
-function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ) {
+# Nested Block, resets heading counters
+function _nblock( block, LOCAL, st, len, hlvl, htxt, hlsav, sec, guard, code, indent, attrib ) {
+  hlsav = HL[1] " " HL[2] " " HL[3] " " HL[4] " " HL[5] " " HL[6];
+  HL[1] = 0; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
+
+  block = _block( block );
+  sec = ""; for ( n = 1; n <= 6; n++ ) { sec = sec (HL[n]?"</section>":""); }
+  split(hlsav, HL);
+  return block sec;
+}
+
+function _block( block, LOCAL, st, len, hlvl, htxt, sec, guard, code, indent, attrib ) {
   gsub( /^\n+|\n+$/, "", block );
 
   if ( block == "" ) {
@@ -344,7 +355,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib
   } 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" \
+    return "<blockquote>\n" _nblock( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "</blockquote>\n\n" \
            _block( substr(block, st + len) );
 
   # Pipe Tables (pandoc / php md / gfm )
@@ -447,7 +458,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib
 
       ttext = "<thead>\n<tr>"
       for (cnt = 1; cnt <= cols; cnt++)
-        ttext = ttext "<th align=\"" talign[cnt] "\">" _block(tarray[cnt]) "</th>"
+        ttext = ttext "<th align=\"" talign[cnt] "\">" _nblock(tarray[cnt]) "</th>"
       ttext = ttext "</tr>\n</thead>"
     }
 
@@ -470,10 +481,10 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib
 
       ttext = ttext "<tr>"
       for (cnt = 1; cnt <= cols; cnt++)
-        ttext = ttext "<td align=\"" talign[cnt] "\">" _block(tarray[cnt]) "</td>"
+        ttext = ttext "<td align=\"" talign[cnt] "\">" _nblock(tarray[cnt]) "</td>"
       ttext = ttext "</tr>\n"
     }
-    return "<table>" ttext "</tbody></table>\n" _block(block);
+    return "<table>" ttext "</tbody></table>\n" _nblock(block);
 
   # Line Blocks (pandoc)
   } else if ( match(block, /^\| [^\n]*(\n|$)(\| [^\n]*(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) {
@@ -503,7 +514,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, 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" \
+      return "<div class=\"" attrib "\">" _nblock( substr(code, 1, st - 1) ) "</div>\n" \
              _block( substr( code, st + len ) );
     } else {
       match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ );
@@ -554,34 +565,37 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib
   gsub("(^|\n) {0," indent "}", "\n", list);
   return "\n<ol>\n" _list( substr(list, 2) ) "</ol>\n" _block( block );
 
-  # First Order Heading
+  # First Order Heading H1
   } else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) {
-    len = RLENGTH;
+    len = RLENGTH; sec = "";
+    for ( n = 1; n <= 6; n++ ) { sec = sec (HL[n]?"</section>":""); }
     HL[1]++; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
-    return "<h1 id=\"" HL[1] ":" URL(gensub( /\n.*$/, "", "g", block )) "\">" \
+    return sec "<section id=\"" HL[1] ":" URL(gensub( /\n.*$/, "", "g", block )) "\"><h1>" \
            inline( gensub( /\n.*$/, "", "g", block ) ) \
            "<a class=\"anchor\" href=\"#" HL[1] ":" \
            URL(gensub( /\n.*$/, "", "g", block )) "\"></a></h1>\n\n" \
            _block( substr( block, len + 1 ) );
 
-  # Second Order Heading
+  # Second Order Heading H2
   } else if ( match( block, /^[^\n]+\n---+(\n|$)/ ) ) {
-    len = RLENGTH;
+    len = RLENGTH; sec = "";
+    for ( n = 2; n <= 6; n++ ) { sec = sec (HL[n]?"</section>":""); }
     HL[2]++; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0;
-    return "<h2 id=\"" HL[1] "." HL[2] ":" URL(gensub( /\n.*$/, "", "g", block )) "\">" \
-           inline( gensub( /\n.*$/, "", "g", block ) ) \
+    return sec "<section id=\"" HL[1] "." HL[2] ":" URL(gensub( /\n.*$/, "", "g", block )) "\">" \
+           "<h2>" inline( gensub( /\n.*$/, "", "g", block ) ) \
            "<a class=\"anchor\" href=\"#" HL[1] "." HL[2] ":" \
            URL(gensub( /\n.*$/, "", "g", block )) "\"></a></h2>\n\n" \
            _block( substr( block, len + 1) );
 
-  # Nth Order Heading
+  # Nth Order Heading H1 H2 H3 H4 H5 H6
   } else if ( match( block, /^#{1,6}[ \t]*[^\n]+([ \t]*#*)(\n|$)/ ) ) {
-    len = RLENGTH;
+    len = RLENGTH; sec = "";
     hlvl = length( gensub( /^(#{1,6}).*$/, "\\1", "g", block ) );
     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;}
+    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] ; }
-    return "<h" hlvl " id=\"" hid ":" URL(htxt) "\">" inline( htxt ) \
+    return sec "<section id=\"" hid ":" URL(htxt) "\"><h" hlvl ">" inline( htxt ) \
            "<a class=\"anchor\" href=\"#" hid "\"></a></h" hlvl ">\n\n" \
            _block( substr( block, len + 1) );
 
@@ -641,7 +655,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib
   } else if ( AllowMacros && match( block, /^<<(([^>]|>[^>])+)>>(\n|$)/) ) {
     len = RLENGTH;
     text = gensub(/^<<(([^>]|>[^>])+)>>(\n.*)?$/, "\\1", 1, block);
-    return macro(text) _block(substr(block, len + 1));
+    return macro(text) _block(substr(block, len + 1) );
 
   # Split paragraphs
   } else if ( match( block, /(^|\n)[[:space:]]*(\n|$)/) ) {
@@ -676,9 +690,9 @@ function _list( block, last, LOCAL, 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);
+    last = p; p = _nblock(p);
   } else {
-    last = p; p = _block(p);
+    last = p; p = _nblock(p);
     sub( /^<p>/, "", p );
     sub( /<\/p>\n/, "", p );
   }
@@ -713,6 +727,7 @@ BEGIN {
   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;
+  # hls = "0 0 0 0 0 0";
 
   # Buffering of full file ist necessary, e.g. to find reference links
   while (getline) { file = file $0 "\n"; }
@@ -736,5 +751,5 @@ BEGIN {
   # for (n in rl_href) { debug(n " | " rl_href[n] " | " rl_title[n] ); }
 
   # Run Block Processing -> The Actual Markdown!
-  printf "%s", _block( file );
+  printf "%s", _nblock( file );
 }