X-Git-Url: http://git.plutz.net/?a=blobdiff_plain;f=markdown.awk;h=71c91fd47a825a984d6bad0026abe01962bb5a5c;hb=516bc32df2437bc45005b1ea1a5d452503bd1669;hp=af3d7224657418caa809b5fa6223ce3cc1411224;hpb=628929d8cbd7e2d59dd324f083fddd520c1c6c4d;p=cgilite diff --git a/markdown.awk b/markdown.awk index af3d722..71c91fd 100755 --- a/markdown.awk +++ b/markdown.awk @@ -38,6 +38,7 @@ # # Extensions - Block elements: # ---------------------------- +# - [x] Automatic
-wrapping (custom) # - ? Heading identifiers (php md, pandoc) # - [x] Automatic heading identifiers (custom) # - [x] Fenced code blocks (php md, pandoc) @@ -89,12 +90,13 @@ function HTML ( text ) { return text; } -function URL ( text ) { +function URL ( text, sharp ) { gsub( /&/, "%26", text ); gsub( /"/, "%22", text ); gsub( /'/, "%27", text ); + gsub( /`/, "%60", text ); gsub( /\?/, "%3F", text ); - gsub( /#/, "%23", text ); + if (sharp) gsub( /#/, "%23", text ); gsub( /\[/, "%5B", text ); gsub( /\]/, "%5D", text ); gsub( / /, "%20", text ); @@ -130,7 +132,7 @@ function inline( line, LOCAL, len, code, href, 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) + code = gensub( /^ | $/, "", "g" , code) # escape HTML within code span gsub( /&/, "\\&", code ); gsub( //, "\\>", code ); return "" code "" inline( substr( line, len + 1 ) ) @@ -194,16 +196,16 @@ function inline( line, LOCAL, len, code, href, guard ) { title = gensub(/^!\[([^]]*)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)(\{([a-zA-Z \t-]*)\})?/, "\\4", "g", substr(line, 1, len) ); attrib = gensub(/^!\[([^]]*)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)(\{([a-zA-Z \t-]*)\})?/, "\\6", "g", substr(line, 1, len) ); if ( title && attrib ) { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } else if ( title ) { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } else if ( attrib ) { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } else { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } @@ -214,10 +216,10 @@ function inline( line, LOCAL, len, code, href, guard ) { id = gensub(/^!\[([^\n]*)\] ?\[([^\n]*)\].*/, "\\2", 1, substr(line, 1, len) ); if ( ! id ) id = text; if ( rl_href[id] && rl_title[id] ) { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } else if ( rl_href[id] ) { - return "\""" \ + return "\""" \ inline( substr( line, len + 1) ); } else { return "" HTML(substr(line, 1, len)) inline( substr(line, len + 1) ); @@ -305,7 +307,39 @@ function inline( line, LOCAL, len, code, href, guard ) { } } -function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ) { +function headline( hlvl, htxt, LOCAL, sec, n, HL) { + split( gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\2" ,"1", hstack), HL); + + for ( n = hlvl; n <= 6; n++ ) { sec = sec (HL[n]?"
":""); } + HL[hlvl]++; for ( n = hlvl + 1; n <= 6; n++) { HL[n] = 0;} + hid = HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; } + # hid = hid ":" URL(htxt, 1); + + hstack = gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\1" ,"1", hstack) \ + HL[1] " " HL[2] " " HL[3] " " HL[4] " " HL[5] " " HL[6]; + + return sec "
" \ + "" inline( htxt ) \ + "" \ + "\n\n"; +} + +# Nested Block, resets heading counters +function _nblock( block, LOCAL, hlsav, sec ) { + hlsav = hstack; + hstack = hstack " 0 0 0 0 0 0"; + + # Block Level + blvl++; + + block = _block( block ); + split( gensub( /^(.* )?([0-9]+( [0-9]+){5})$/, "\\2" ,"1", hstack), HL); + sec = ""; for ( n = 1; n <= 6; n++ ) { sec = sec (HL[n]?"
":""); } + hstack = hlsav; blvl--; + return block sec; +} + +function _block( block, LOCAL, st, len, hlvl, htxt, sec, guard, code, indent, attrib ) { gsub( /^\n+|\n+$/, "", block ); if ( block == "" ) { @@ -343,7 +377,7 @@ 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 "
\n" _block( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "
\n\n" \ + return "
\n" _nblock( gensub( /(^|\n)> /, "\n", "g", substr(block, 1, st - 1) ) ) "
\n\n" \ _block( substr(block, st + len) ); # Pipe Tables (pandoc / php md / gfm ) @@ -446,7 +480,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ttext = "\n" for (cnt = 1; cnt <= cols; cnt++) - ttext = ttext "" _block(tarray[cnt]) "" + ttext = ttext "" _nblock(tarray[cnt]) "" ttext = ttext "\n" } @@ -469,10 +503,10 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib ttext = ttext "" for (cnt = 1; cnt <= cols; cnt++) - ttext = ttext "" _block(tarray[cnt]) "" + ttext = ttext "" _nblock(tarray[cnt]) "" ttext = ttext "\n" } - return "" ttext "
\n" _block(block); + return "" ttext "
\n" _nblock(block); # Line Blocks (pandoc) } else if ( match(block, /^\| [^\n]*(\n|$)(\| [^\n]*(\n|$)|[ \t]+[^\n[:space:]][^\n]*(\n|$))*/) ) { @@ -502,7 +536,7 @@ 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 "
" _block( substr(code, 1, st - 1) ) "
\n" \ + return "
" _nblock( substr(code, 1, st - 1) ) "
\n" \ _block( substr( code, st + len ) ); } else { match( block, /(^|\n)[[:space:]]*(\n|$)/ ) || match( block, /$/ ); @@ -553,35 +587,28 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib gsub("(^|\n) {0," indent "}", "\n", list); return "\n
    \n" _list( substr(list, 2) ) "
\n" _block( block ); - # First Order Heading + # First Order Heading H1 } else if ( match( block, /^[^\n]+\n===+(\n|$)/ ) ) { len = RLENGTH; - HL[1]++; HL[2] = 0; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0; - return "

" \ - inline( gensub( /\n.*$/, "", "g", block ) ) \ - "

\n\n" \ + + return headline(1, gensub( /\n.*$/, "", "g", block )) \ _block( substr( block, len + 1 ) ); - # Second Order Heading + # Second Order Heading H2 } else if ( match( block, /^[^\n]+\n---+(\n|$)/ ) ) { len = RLENGTH; - HL[2]++; HL[3] = 0; HL[4] = 0; HL[5] = 0; HL[6] = 0; - return "

" \ - inline( gensub( /\n.*$/, "", "g", block ) ) \ - "

\n\n" \ + + return headline(2, gensub( /\n.*$/, "", "g", block )) \ _block( substr( block, len + 1) ); - # Nth Order Heading + # Nth Order Heading H1 H2 H3 H4 H5 H6 } else if ( match( block, /^#{1,6}[ \t]*[^\n]+([ \t]*#*)(\n|$)/ ) ) { len = RLENGTH; - 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;} - hid = HL[1]; for ( n = 2; n <= hlvl; n++) { hid = hid "." HL[n] ; } - return "" inline( htxt ) \ - "\n\n" \ + + 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) \ + ) \ _block( substr( block, len + 1) ); # block images (wrapped in
) @@ -592,25 +619,25 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib title = gensub(/^!\[([^]]*)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)(\{([a-zA-Z \t-]*)\})?(\n.*)?$/, "\\4", "g", block); attrib = gensub(/^!\[([^]]*)\]\(([^"\)]+)([ \t]+"([^"]+)")?\)(\{([a-zA-Z \t-]*)\})?(\n.*)?$/, "\\6", "g", block); if ( title && attrib ) { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
" inline(title) "
" \ "
\n\n" \ _block( substr( block, len + 1) ); } else if ( title ) { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
" inline(title) "
" \ "
\n\n" \ _block( substr( block, len + 1) ); } else if ( attrib ) { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
\n\n" \ _block( substr( block, len + 1) ); } else { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
\n\n" \ _block( substr( block, len + 1) ); } @@ -622,14 +649,14 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib id = gensub(/^!\[([^\n]*)\] ?\[([^\n]*)\](\n.*)?$/, "\\2", 1, block); if ( ! id ) id = text; if ( rl_href[id] && rl_title[id] ) { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
" inline(rl_title[id]) "
" \ "
\n\n" \ _block( substr( block, len + 1) ); } else if ( rl_href[id] ) { - return "
" \ - "\""" \ + return "
" \ + "\""" \ "
\n\n" \ _block( substr( block, len + 1) ); } else { @@ -640,7 +667,7 @@ function _block( block, LOCAL, st, len, hlvl, htxt, guard, code, indent, attrib } else if ( AllowMacros && match( block, /^<<(([^>]|>[^>])+)>>(\n|$)/) ) { len = RLENGTH; text = gensub(/^<<(([^>]|>[^>])+)>>(\n.*)?$/, "\\1", 1, block); - return macro(text) _block(substr(block, len + 1)); + return macro(text) _block(substr(block, len + 1) ); # Split paragraphs } else if ( match( block, /(^|\n)[[:space:]]*(\n|$)/) ) { @@ -675,9 +702,9 @@ 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 ); sub( /<\/p>\n/, "", p ); } @@ -712,6 +739,7 @@ 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"; } @@ -735,5 +763,5 @@ 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 ); }