newfangle

   1 #! /usr/bin/awk -f
   2 # newfangle - fully featured notangle replacement in awk
   3 #
   4 #  Copyright (C) Sam Liddicott 2009
   5 #
   6 #  This program is free software: you can redistribute it and/or modify
   7 #  it under the terms of the GNU General Public License as published by
   8 #  the Free Software Foundation, either version 3 of the License, or
   9 #  (at your option) any later version.
  10 #
  11 #  This program is distributed in the hope that it will be useful,
  12 #  but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 #  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 #  GNU General Public License for more details.
  15 #
  16 #  You should have received a copy of the GNU General Public License
  17 #  along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18 #
  19
  20
  21 # NOTE: Arnold Robbins public domain getopt for awk is also used:
  22 # getopt.awk --- do C library getopt(3) function in awk
  23 #
  24 # Arnold Robbins, arnold@skeeve.com, Public Domain
  25 #
  26 # Initial version: March, 1991
  27 # Revised: May, 1993
  28
  29
  30 function getopt(argc, argv, options,    thisopt, i)
  31 {
  32     if (length(options) == 0)    # no options given
  33         return -1
  34     if (argv[Optind] == "--") {  # all done
  35         Optind++
  36         _opti = 0
  37         return -1
  38     } else if (argv[Optind] !~ /^-[^: \t\n\f\r\v\b]/) {
  39         _opti = 0
  40         return -1
  41     }
  42     if (_opti == 0)
  43         _opti = 2
  44     thisopt = substr(argv[Optind], _opti, 1)
  45     Optopt = thisopt
  46     i = index(options, thisopt)
  47     if (i == 0) {
  48         if (Opterr)
  49             printf("%c -- invalid option\n",
  50                                   thisopt) > "/dev/stderr"
  51         if (_opti >= length(argv[Optind])) {
  52             Optind++
  53             _opti = 0
  54         } else
  55             _opti++
  56         return "?"
  57     }
  58     if (substr(options, i + 1, 1) == ":") {
  59         # get option argument
  60         if (length(substr(argv[Optind], _opti + 1)) > 0)
  61             Optarg = substr(argv[Optind], _opti + 1)
  62         else
  63             Optarg = argv[++Optind]
  64         _opti = 0
  65     } else
  66         Optarg = ""
  67     if (_opti == 0 || _opti >= length(argv[Optind])) {
  68         Optind++
  69         _opti = 0
  70     } else
  71         _opti++
  72     return thisopt
  73 }
  74
  75 function error(message)
  76 {
  77   print "ERROR: " FILENAME ":" FNR " " message > "/dev/stderr";
  78   exit 1;
  79 }
  80 function warning(message)
  81 {
  82   print "WARNING: " FILENAME ":" FNR " " message > "/dev/stderr";
  83   warnings++;
  84 }
  85 function new_mode_tracker(context, language, mode) {
  86   context[""] = 0;
  87   context[0, "language"] = language;
  88   context[0, "mode"] = mode;
  89 }
  90 function push_mode_tracker(context, language, mode,
  91   # local vars
  92   top)
  93 {
  94   if (! ("" in context)) {
  95     split("", context);
  96     new_mode_tracker(context, language, mode);
  97   } else {
  98     top = context[""];
  99     if (context[top, "language"] == language && mode=="") mode = context[top, "mode"];
 100     top++;
 101     context[top, "language"] = language;
 102     context[top, "mode"] = mode;
 103     context[""] = top;
 104   }
 105 }
 106 function dump_mode_tracker(context,
 107   c, d)
 108 {
 109   for(c=0; c <= context[""]; c++) {
 110     printf(" %2d   %s:%s\n", c, context[c, "language"], context[c, "mode"]) > "/dev/stderr";
 111     for(d=1; ( (c, "values", d) in context); d++) {
 112       printf("   %2d %s\n", d, context[c, "values", d]) > "/dev/stderr";
 113     }
 114   }
 115 }
 116 function finalize_mode_tracker(context)
 117 {
 118   if ( ("" in context) && context[""] != 0) return 0;
 119   return 1;
 120 }
 121 function mode_tracker(context, text, values,
 122   # optional parameters
 123   # local vars
 124   mode, submodes, language,
 125   cindex, c, a, part, item, name, result, new_values, new_mode,
 126   delimiters, terminators)
 127 {
 128   cindex = context[""] + 0;
 129   mode = context[cindex, "mode"];
 130   language = context[cindex, "language" ];
 131   submodes=modes[language, mode, "submodes"];
 132
 133   if ((language, mode, "delimiters") in modes) {
 134     delimiters = modes[language, mode, "delimiters"];
 135     if (length(submodes)>0) submodes = submodes "|";
 136     submodes=submodes delimiters;
 137   } else delimiters="";
 138   if ((language, mode, "terminators") in modes) {
 139     terminators = modes[language, mode, "terminators"];
 140     if (length(submodes)>0) submodes = submodes "|";
 141     submodes=submodes terminators;
 142   } else terminators="";
 143   if (! length(submodes)) return text;
 144   while((cindex >= 0) && length(text)) {
 145     if (match(text, "(" submodes ")", a)) {
 146       if (RLENGTH<1) {
 147         error(sprintf("Internal error, matched zero length submode, should be impossible - likely regex computation error\n" \
 148                 "Language=%s\nmode=%s\nmatch=%s\n", language, mode, submodes));
 149       }
 150       part = substr(text, 1, RSTART -1);
 151       item = item part;
 152       if (match(a[1], "^" terminators "$")) {
 153 #printf("%2d EXIT  MODE [%s] by [%s] [%s]\n", cindex, mode, a[1], text) > "/dev/stderr"
 154         context[cindex, "values", ++context[cindex, "values"]] = item;
 155         delete context[cindex];
 156         context[""] = --cindex;
 157         if (cindex>=0) {
 158           mode = context[cindex, "mode"];
 159           language = context[cindex, "language"];
 160             submodes=modes[language, mode, "submodes"];
 161
 162             if ((language, mode, "delimiters") in modes) {
 163               delimiters = modes[language, mode, "delimiters"];
 164               if (length(submodes)>0) submodes = submodes "|";
 165               submodes=submodes delimiters;
 166             } else delimiters="";
 167             if ((language, mode, "terminators") in modes) {
 168               terminators = modes[language, mode, "terminators"];
 169               if (length(submodes)>0) submodes = submodes "|";
 170               submodes=submodes terminators;
 171             } else terminators="";
 172             if (! length(submodes)) return text;
 173         }
 174         item = item a[1];
 175         text = substr(text, 1 + length(part) + length(a[1]));
 176       }
 177       else if (match(a[1], "^" delimiters "$")) {
 178         if (cindex==0) {
 179           context[cindex, "values", ++context[cindex, "values"]] = item;
 180           item = "";
 181         } else {
 182           item = item a[1];
 183         }
 184         text = substr(text, 1 + length(part) + length(a[1]));
 185       }
 186  else if ((language, a[1], "terminators") in modes) {
 187         #check if new_mode is defined
 188         item = item a[1];
 189 #printf("%2d ENTER MODE [%s] in [%s]\n", cindex, a[1], text) > "/dev/stderr"
 190         text = substr(text, 1 + length(part) + length(a[1]));
 191         context[""] = ++cindex;
 192         context[cindex, "mode"] = a[1];
 193         context[cindex, "language"] = language;
 194         mode = a[1];
 195           submodes=modes[language, mode, "submodes"];
 196
 197           if ((language, mode, "delimiters") in modes) {
 198             delimiters = modes[language, mode, "delimiters"];
 199             if (length(submodes)>0) submodes = submodes "|";
 200             submodes=submodes delimiters;
 201           } else delimiters="";
 202           if ((language, mode, "terminators") in modes) {
 203             terminators = modes[language, mode, "terminators"];
 204             if (length(submodes)>0) submodes = submodes "|";
 205             submodes=submodes terminators;
 206           } else terminators="";
 207           if (! length(submodes)) return text;
 208       } else {
 209         error(sprintf("Submode '%s' set unknown mode in text: %s\nLanguage %s Mode %s\n", a[1], text, language, mode));
 210         text = substr(text, 1 + length(part) + length(a[1]));
 211       }
 212     }
 213 else {
 214       context[cindex, "values", ++context[cindex, "values"]] = item text;
 215       text = "";
 216       item = "";
 217     }
 218   }
 219
 220   context["item"] = item;
 221
 222   if (length(item)) context[cindex, "values", ++context[cindex, "values"]] = item;
 223   return text;
 224 }
 225 function transform_escape(s, r, text,
 226     # optional
 227     max,
 228         # local vars
 229         c)
 230 {
 231   for(c=1; c <= max && (c in s); c++) {
 232     gsub(s[c], r[c], text);
 233   }
 234   return text;
 235 }
 236 function mode_escaper(context, s, r, src,
 237   c, cp, cpl)
 238 {
 239         for(c = context[""]; c >= 0; c--) {
 240                 if ( (context[c, "language"], context[c, "mode"]) in escapes) {
 241                         cpl = escapes[context[c, "language"], context[c, "mode"]];
 242                         for (cp = 1; cp <= cpl; cp ++) {
 243                                 ++src;
 244                                 s[src] = escapes[context[c, "language"], context[c, "mode"], cp, "s"];
 245                                 r[src] = escapes[context[c, "language"], context[c, "mode"], cp, "r"];
 246                         }
 247                 }
 248         }
 249         return src;
 250 }
 251 function dump_escaper(c, s, r, cc) {
 252   for(cc=1; cc<=c; cc++) {
 253     printf("%2d s[%s] r[%s]\n", cc, s[cc], r[cc]) > "/dev/stderr"
 254   }
 255 }
 256 function parse_chunk_args(language, text, values, mode,
 257   # local vars
 258   c, context, rest)
 259 {
 260   split("", context);
 261   new_mode_tracker(context, language, mode);
 262   rest = mode_tracker(context, text, values);
 263   # extract values
 264   for(c=1; c <= context[0, "values"]; c++) {
 265     values[c] = context[0, "values", c];
 266   }
 267   return rest;
 268 }
 269 function new_chunk(chunk_name, params,
 270   # local vars
 271   p, append )
 272 {
 273   # HACK WHILE WE CHANGE TO ( ) for PARAM CHUNKS
 274   gsub("\\(\\)$", "", chunk_name);
 275   if (! (chunk_name in chunk_names)) {
 276     if (debug) print "New chunk " chunk_name;
 277     chunk_names[chunk_name];
 278     for (p in params) {
 279       chunks[chunk_name, p] = params[p];
 280     }
 281     if ("append" in params) {
 282       append=params["append"];
 283       if (! (append in chunk_names)) {
 284         warning("Chunk " chunk_name " is appended to chunk " append " which is not defined yet");
 285         new_chunk(append);
 286       }
 287       chunk_include(append, chunk_name);
 288       chunk_line(append, ORS);
 289     }
 290   }
 291   active_chunk = chunk_name;
 292   prime_chunk(chunk_name);
 293 }
 294
 295 function prime_chunk(chunk_name)
 296 {
 297   chunks[chunk_name, "part", ++chunks[chunk_name, "part"] ] = \
 298          chunk_name SUBSEP "chunklet" SUBSEP "" ++chunks[chunk_name, "chunklet"];
 299   chunks[chunk_name, "part", chunks[chunk_name, "part"], "FILENAME"] = FILENAME;
 300   chunks[chunk_name, "part", chunks[chunk_name, "part"], "LINENO"] = FNR + 1;
 301 }
 302
 303 function chunk_line(chunk_name, line){
 304   chunks[chunk_name, "chunklet", chunks[chunk_name, "chunklet"],
 305          ++chunks[chunk_name, "chunklet", chunks[chunk_name, "chunklet"], "line"]  ] = line;
 306 }
 307
 308 function chunk_include(chunk_name, chunk_ref, indent, tail)
 309 {
 310   chunks[chunk_name, "part", ++chunks[chunk_name, "part"] ] = chunk_ref;
 311   chunks[chunk_name, "part", chunks[chunk_name, "part"], "type" ] = part_type_chunk;
 312   chunks[chunk_name, "part", chunks[chunk_name, "part"], "indent" ] = indent_string(indent);
 313   chunks[chunk_name, "part", chunks[chunk_name, "part"], "tail" ] = tail;
 314   prime_chunk(chunk_name);
 315 }
 316
 317 function indent_string(indent) {
 318   return sprintf("%" indent "s", "");
 319 }
 320 function output_chunk_names(   c, prefix, suffix)
 321 {
 322   if (notangle_mode) {
 323     prefix="<<";
 324     suffix=">>";
 325   }
 326   for (c in chunk_names) {
 327     print prefix c suffix "\n";
 328   }
 329 }
 330 function output_chunks(  a)
 331 {
 332   for (a in chunk_names) {
 333     output_chunk(chunk_names[a]);
 334   }
 335 }
 336
 337 function output_chunk(chunk) {
 338   newline = 1;
 339   lineno_needed = linenos;
 340
 341   write_chunk(chunk);
 342 }
 343
 344 function write_chunk(chunk_name) {
 345   split("", context);
 346   return write_chunk_r(chunk_name, context);
 347 }
 348 function write_chunk_r(chunk_name, context, indent, tail,
 349   # optional vars
 350   chunk_path, chunk_args,
 351   s, r, src, new_src,
 352   # local vars
 353   chunk_params, part, max_part, part_line, frag, max_frag, text,
 354   chunklet, only_part, call_chunk_args, new_context)
 355 {
 356   if (match(chunk_name, "^(.*)\\[([0-9]*)\\]$", chunk_name_parts)) {
 357     chunk_name = chunk_name_parts[1];
 358     only_part = chunk_name_parts[2];
 359   }
 360   split("", context);
 361   new_mode_tracker(context, chunks[chunk_name, "language"], "");
 362   split(chunks[chunk_name, "params"], chunk_params, " *; *");
 363   if (! (chunk_name in chunk_names)) {
 364     error(sprintf(_"The root module <<%s>> was not defined.\nUsed by: %s",\
 365                   chunk_name, chunk_path));
 366   }
 367
 368   max_part = chunks[chunk_name, "part"];
 369   for(part = 1; part <= max_part; part++) {
 370     if (! only_part || part == only_part) {
 371       if (linenos && (chunk_name SUBSEP "part" SUBSEP part SUBSEP "FILENAME" in chunks)) {
 372         a_filename = chunks[chunk_name, "part", part, "FILENAME"];
 373         a_lineno = chunks[chunk_name, "part", part, "LINENO"];
 374         if (a_filename != filename || a_lineno != lineno) {
 375           lineno_needed++;
 376         }
 377       }
 378
 379       chunklet = chunks[chunk_name, "part", part];
 380       if (chunks[chunk_name, "part", part, "type"] == part_type_chunk) {
 381         if (match(chunklet, "^([^\\[\\(]*)\\((.*)\\)$", chunklet_parts)) {
 382           chunklet = chunklet_parts[1];
 383           parse_chunk_args("c-like", chunklet_parts[2], call_chunk_args, "(");
 384           for (c in call_chunk_args) {
 385             call_chunk_args[c] = expand_chunk_args(call_chunk_args[c], chunk_params, chunk_args);
 386           }
 387         } else {
 388           split("", call_chunk_args);
 389         }
 390         # update the transforms arrays
 391         new_src = mode_escaper(context, s, r, src);
 392         split("", new_context);
 393         write_chunk_r(chunklet, new_context,
 394                     chunks[chunk_name, "part", part, "indent"] indent,
 395                     chunks[chunk_name, "part", part, "tail"],
 396                     chunk_path "\n         " chunk_name,
 397                     call_chunk_args,
 398                                 s, r, new_src);
 399       } else if (chunklet SUBSEP "line" in chunks) {
 400         max_frag = chunks[chunklet, "line"];
 401         for(frag = 1; frag <= max_frag; frag++) {
 402           if (newline && lineno_needed && ! lineno_suppressed) {
 403             filename = a_filename;
 404             lineno = a_lineno;
 405             print "#line " lineno " \"" filename "\"\n"
 406             lineno_needed = 0;
 407           }
 408
 409           text = chunks[chunklet, frag];
 410
 411           /* check params */
 412           text = expand_chunk_args(text, chunk_params, chunk_args);
 413
 414          if (text == "\n") {
 415             lineno++;
 416             if (part == max_part && frag == max_frag && length(chunk_path)) {
 417               text = "";
 418               break;
 419             } else {
 420               newline = 1;
 421             }
 422           } else if (length(text) || length(tail)) {
 423             if (newline) text = indent text;
 424             newline = 0;
 425           }
 426
 427           text = text tail;
 428           mode_tracker(context, text);
 429           print transform_escape(s, r, text, src);
 430           if (linenos) {
 431             lineno_suppressed = substr(lastline, length(lastline)) == "\\";
 432           }
 433         }
 434       } else {
 435         # empty last chunklet
 436       }
 437     }
 438   }
 439   if (! finalize_mode_tracker(context)) {
 440     dump_mode_tracker(context);
 441     error(sprintf(_"Module %s did not close context properly.\nUsed by: %s\n", chunk_name, chunk_path));
 442   }
 443 }
 444 function expand_chunk_args(text, params, args,
 445   p, text_array, next_text, v, t, l)
 446 {
 447   if (split(text, text_array, "\\${")) {
 448     for(p in params) {
 449       v[params[p]]=args[p];
 450     }
 451     text=text_array[1];
 452     for(t=2; t in text_array; t++) {
 453       if (match(text_array[t], "^([a-zA-Z_][a-zA-Z0-9_]*)}", l) &&
 454           l[1] in v)
 455       {
 456         text = text v[l[1]] substr(text_array[t], length(l[1])+2);
 457       } else {
 458         text = text "${" text_array[t];
 459       }
 460     }
 461   }
 462
 463   return text;
 464 }
 465 BEGIN {
 466   part_type_chunk=1;
 467   SUBSEP=",";
 468   modes["c-like", "",  "submodes" ]="\\\\|\"|'|{|\\(|\\[";
 469   modes["c-like", "",  "delimiters"]=" *, *";
 470   modes["c-like", "\\", "terminators"]=".";
 471   modes["c-like", "\"", "submodes" ]="\\\\";
 472   modes["c-like", "\"", "terminators"]="\"";
 473   escapes["c-like", "\"", ++escapes["c-like", "\""], "s"]="\\\\";
 474   escapes["c-like", "\"",   escapes["c-like", "\""], "r"]="\\\\";
 475   escapes["c-like", "\"", ++escapes["c-like", "\""], "s"]="\"";
 476   escapes["c-like", "\"",   escapes["c-like", "\""], "r"]="\\" "\"";
 477   escapes["c-like", "\"", ++escapes["c-like", "\""], "s"]="\n";
 478   escapes["c-like", "\"",   escapes["c-like", "\""], "r"]="\\n";
 479   modes["c-like", "'", "submodes" ]="\\\\";
 480   modes["c-like", "'", "terminators"]="'";
 481   escapes["c-like", "'", ++escapes["c-like", "'"], "s"]="\\\\";
 482   escapes["c-like", "'",   escapes["c-like", "'"], "r"]="\\\\";
 483   escapes["c-like", "'", ++escapes["c-like", "'"], "s"]="'";
 484   escapes["c-like", "'",   escapes["c-like", "'"], "r"]="\\" "'";
 485   escapes["c-like", "'", ++escapes["c-like", "'"], "s"]="\n";
 486   escapes["c-like", "'",   escapes["c-like", "'"], "r"]="\\n";
 487   modes["c-like", "{",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 488   modes["c-like", "{",  "delimiters"]=" *, *";
 489   modes["c-like", "{",  "terminators"]="}";
 490   modes["c-like", "[",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 491   modes["c-like", "[",  "delimiters"]=" *, *";
 492   modes["c-like", "[",  "terminators"]="\\]";
 493   modes["c-like", "(",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 494   modes["c-like", "(",  "delimiters"]=" *, *";
 495   modes["c-like", "(",  "terminators"]="\\)";
 496
 497   modes["c", "",  "submodes" ]="\\\\|\"|'|{|\\(|\\[";
 498   modes["c", "",  "delimiters"]=" *, *";
 499   modes["c", "\\", "terminators"]=".";
 500   modes["c", "\"", "submodes" ]="\\\\";
 501   modes["c", "\"", "terminators"]="\"";
 502   escapes["c", "\"", ++escapes["c", "\""], "s"]="\\\\";
 503   escapes["c", "\"",   escapes["c", "\""], "r"]="\\\\";
 504   escapes["c", "\"", ++escapes["c", "\""], "s"]="\"";
 505   escapes["c", "\"",   escapes["c", "\""], "r"]="\\" "\"";
 506   escapes["c", "\"", ++escapes["c", "\""], "s"]="\n";
 507   escapes["c", "\"",   escapes["c", "\""], "r"]="\\n";
 508   modes["c", "'", "submodes" ]="\\\\";
 509   modes["c", "'", "terminators"]="'";
 510   escapes["c", "'", ++escapes["c", "'"], "s"]="\\\\";
 511   escapes["c", "'",   escapes["c", "'"], "r"]="\\\\";
 512   escapes["c", "'", ++escapes["c", "'"], "s"]="'";
 513   escapes["c", "'",   escapes["c", "'"], "r"]="\\" "'";
 514   escapes["c", "'", ++escapes["c", "'"], "s"]="\n";
 515   escapes["c", "'",   escapes["c", "'"], "r"]="\\n";
 516   modes["c", "{",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 517   modes["c", "{",  "delimiters"]=" *, *";
 518   modes["c", "{",  "terminators"]="}";
 519   modes["c", "[",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 520   modes["c", "[",  "delimiters"]=" *, *";
 521   modes["c", "[",  "terminators"]="\\]";
 522   modes["c", "(",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 523   modes["c", "(",  "delimiters"]=" *, *";
 524   modes["c", "(",  "terminators"]="\\)";
 525   modes["c", "", "submodes"] = modes["c", "", "submodes"] "|" "/\\*";
 526   modes["c", "/*", "terminators"]="\\*/";
 527   modes["c", "", "submodes"] = modes["c", "", "submodes"] "|" "//";
 528   modes["c", "//", "terminators"]="\n";
 529   escapes["c", "//", ++escapes["c", "//"], "s"]="\n";
 530   escapes["c", "//",   escapes["c", "//"], "r"]="\n//";
 531   modes["c", "", "submodes"] = modes["c", "", "submodes"] "|" "#";
 532   modes["c", "#", "submodes" ]="\\\\";
 533   modes["c", "#", "terminators"]="\n";
 534   escapes["c", "#", ++escapes["c", "#"], "s"]="\n";
 535   escapes["c", "#",   escapes["c", "#"], "r"]="\\\\\n";
 536
 537   modes["awk", "",  "submodes" ]="\\\\|\"|'|{|\\(|\\[";
 538   modes["awk", "",  "delimiters"]=" *, *";
 539   modes["awk", "\\", "terminators"]=".";
 540   modes["awk", "\"", "submodes" ]="\\\\";
 541   modes["awk", "\"", "terminators"]="\"";
 542   escapes["awk", "\"", ++escapes["awk", "\""], "s"]="\\\\";
 543   escapes["awk", "\"",   escapes["awk", "\""], "r"]="\\\\";
 544   escapes["awk", "\"", ++escapes["awk", "\""], "s"]="\"";
 545   escapes["awk", "\"",   escapes["awk", "\""], "r"]="\\" "\"";
 546   escapes["awk", "\"", ++escapes["awk", "\""], "s"]="\n";
 547   escapes["awk", "\"",   escapes["awk", "\""], "r"]="\\n";
 548   modes["awk", "'", "submodes" ]="\\\\";
 549   modes["awk", "'", "terminators"]="'";
 550   escapes["awk", "'", ++escapes["awk", "'"], "s"]="\\\\";
 551   escapes["awk", "'",   escapes["awk", "'"], "r"]="\\\\";
 552   escapes["awk", "'", ++escapes["awk", "'"], "s"]="'";
 553   escapes["awk", "'",   escapes["awk", "'"], "r"]="\\" "'";
 554   escapes["awk", "'", ++escapes["awk", "'"], "s"]="\n";
 555   escapes["awk", "'",   escapes["awk", "'"], "r"]="\\n";
 556   modes["awk", "{",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 557   modes["awk", "{",  "delimiters"]=" *, *";
 558   modes["awk", "{",  "terminators"]="}";
 559   modes["awk", "[",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 560   modes["awk", "[",  "delimiters"]=" *, *";
 561   modes["awk", "[",  "terminators"]="\\]";
 562   modes["awk", "(",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 563   modes["awk", "(",  "delimiters"]=" *, *";
 564   modes["awk", "(",  "terminators"]="\\)";
 565   modes["awk", "", "submodes"] = modes["awk", "", "submodes"] "|" "#";
 566   modes["awk", "#", "terminators"]="\n";
 567   escapes["awk", "#", ++escapes["awk", "#"], "s"]="\n";
 568   escapes["awk", "#",   escapes["awk", "#"], "r"]="\n#";
 569   modes["awk", "", "submodes"] = modes["awk", "", "submodes"] "|" "/\\^";
 570   modes["awk", "/^", "terminators"]="/";
 571   modes["perl", "",  "submodes" ]="\\\\|\"|'|{|\\(|\\[";
 572   modes["perl", "",  "delimiters"]=" *, *";
 573   modes["perl", "\\", "terminators"]=".";
 574   modes["perl", "\"", "submodes" ]="\\\\";
 575   modes["perl", "\"", "terminators"]="\"";
 576   escapes["perl", "\"", ++escapes["perl", "\""], "s"]="\\\\";
 577   escapes["perl", "\"",   escapes["perl", "\""], "r"]="\\\\";
 578   escapes["perl", "\"", ++escapes["perl", "\""], "s"]="\"";
 579   escapes["perl", "\"",   escapes["perl", "\""], "r"]="\\" "\"";
 580   escapes["perl", "\"", ++escapes["perl", "\""], "s"]="\n";
 581   escapes["perl", "\"",   escapes["perl", "\""], "r"]="\\n";
 582   modes["perl", "'", "submodes" ]="\\\\";
 583   modes["perl", "'", "terminators"]="'";
 584   escapes["perl", "'", ++escapes["perl", "'"], "s"]="\\\\";
 585   escapes["perl", "'",   escapes["perl", "'"], "r"]="\\\\";
 586   escapes["perl", "'", ++escapes["perl", "'"], "s"]="'";
 587   escapes["perl", "'",   escapes["perl", "'"], "r"]="\\" "'";
 588   escapes["perl", "'", ++escapes["perl", "'"], "s"]="\n";
 589   escapes["perl", "'",   escapes["perl", "'"], "r"]="\\n";
 590   modes["perl", "{",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 591   modes["perl", "{",  "delimiters"]=" *, *";
 592   modes["perl", "{",  "terminators"]="}";
 593   modes["perl", "[",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 594   modes["perl", "[",  "delimiters"]=" *, *";
 595   modes["perl", "[",  "terminators"]="\\]";
 596   modes["perl", "(",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 597   modes["perl", "(",  "delimiters"]=" *, *";
 598   modes["perl", "(",  "terminators"]="\\)";
 599   modes["perl", "", "submodes"] = modes["perl", "", "submodes"] "|" "/\\*";
 600   modes["perl", "/*", "terminators"]="\\*/";
 601   modes["perl", "", "submodes"] = modes["perl", "", "submodes"] "|" "#";
 602   modes["perl", "#", "terminators"]="\n";
 603   escapes["perl", "#", ++escapes["perl", "#"], "s"]="\n";
 604   escapes["perl", "#",   escapes["perl", "#"], "r"]="\n#";
 605   modes["sh", "",  "submodes" ]="\\\\|\"|'|{|\\(|\\[";
 606   modes["sh", "",  "delimiters"]=" *, *";
 607   modes["sh", "\\", "terminators"]=".";
 608   modes["sh", "\"", "submodes" ]="\\\\";
 609   modes["sh", "\"", "terminators"]="\"";
 610   escapes["sh", "\"", ++escapes["sh", "\""], "s"]="\\\\";
 611   escapes["sh", "\"",   escapes["sh", "\""], "r"]="\\\\";
 612   escapes["sh", "\"", ++escapes["sh", "\""], "s"]="\"";
 613   escapes["sh", "\"",   escapes["sh", "\""], "r"]="\\" "\"";
 614   escapes["sh", "\"", ++escapes["sh", "\""], "s"]="\n";
 615   escapes["sh", "\"",   escapes["sh", "\""], "r"]="\\n";
 616   modes["sh", "'", "submodes" ]="\\\\";
 617   modes["sh", "'", "terminators"]="'";
 618   escapes["sh", "'", ++escapes["sh", "'"], "s"]="\\\\";
 619   escapes["sh", "'",   escapes["sh", "'"], "r"]="\\\\";
 620   escapes["sh", "'", ++escapes["sh", "'"], "s"]="'";
 621   escapes["sh", "'",   escapes["sh", "'"], "r"]="\\" "'";
 622   escapes["sh", "'", ++escapes["sh", "'"], "s"]="\n";
 623   escapes["sh", "'",   escapes["sh", "'"], "r"]="\\n";
 624   modes["sh", "{",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 625   modes["sh", "{",  "delimiters"]=" *, *";
 626   modes["sh", "{",  "terminators"]="}";
 627   modes["sh", "[",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 628   modes["sh", "[",  "delimiters"]=" *, *";
 629   modes["sh", "[",  "terminators"]="\\]";
 630   modes["sh", "(",  "submodes" ]="\\\\|\"|{|\\(|\\[|'|/\\*";
 631   modes["sh", "(",  "delimiters"]=" *, *";
 632   modes["sh", "(",  "terminators"]="\\)";
 633   #<\chunkref{mode:common-string}("sh", "\textbackslash{}"")>
 634   #<\chunkref{mode:common-string}("sh", "'")>
 635   modes["sh", "", "submodes"] = modes["sh", "", "submodes"] "|" "#";
 636   modes["sh", "#", "terminators"]="\n";
 637   escapes["sh", "#", ++escapes["sh", "#"], "s"]="\n";
 638   escapes["sh", "#",   escapes["sh", "#"], "r"]="\n#";
 639   escapes["sh", "\"", ++escapes["sh", "\""], "s"]="\\$";
 640   escapes["sh", "\"",   escapes["sh", "\""], "r"]="\\$";
 641   debug=0;
 642   linenos=0;
 643   notangle_mode=0;
 644   root="*";
 645   tabs = "";
 646
 647   Optind = 1    # skip ARGV[0]
 648   while(getopt(ARGC, ARGV, "R:LdT:hr")!=-1) {
 649     if (Optopt == "R") root = Optarg;
 650     else if (Optopt == "r") root="";
 651     else if (Optopt == "L") linenos = 1;
 652     else if (Optopt == "d") debug = 1;
 653     else if (Optopt == "T") tabs = indent_string(Optarg+0);
 654     else if (Optopt == "h") help();
 655     else if (Optopt == "?") help();
 656   }
 657   for (i=1; i<Optind; i++) { ARGV[i]=""; }
 658 }
 659 /^\\Chunk{/ {
 660   if (match($0, "^\\\\Chunk{ *([^ ,}]*),?(.*)}", line)) {
 661     next_chunk_name = line[1];
 662     get_chunk_args(line[2], next_chunk_args);
 663   }
 664   next;
 665 }
 666 /^\\begin{lstlisting}|^\\begin{Chunk}/ {
 667   if (match($0, "}.*[[,] *name= *{? *([^], }]*)", line)) {
 668     new_chunk(line[1]);
 669   } else {
 670     new_chunk(next_chunk_name, next_chunk_args);
 671   }
 672   chunking=1;
 673   next;
 674 }
 675 /^[<]<.*[>]>=/ {
 676   if (match($0, "^[<]<(.*)[>]>= *$", line)) {
 677     chunking=1;
 678     notangle_mode=1;
 679     new_chunk(line[1]);
 680     next;
 681   }
 682 }
 683 /^\\[e]nd{lstlisting}|^\\[e]nd{Chunk}/ {
 684   chunking=0;
 685   active_chunk="";
 686   next;
 687 }
 688 /^@ *$/ {
 689   chunking=0;
 690   active_chunk="";
 691 }
 692 ! chunking { next; }
 693 length(active_chunk) {
 694   if (length(tabs)) {
 695     gsub("\t", tabs);
 696   }
 697   chunk = $0;
 698   indent = 0;
 699   while(match(chunk,
 700               "([=]<\\\\chunkref{([^}>]*)}(\\(.*\\)|)>|<<([a-zA-Z_][-a-zA-Z0-9_]*)>>)",
 701               line)\
 702   ) {
 703     chunklet = substr(chunk, 1, RSTART - 1);
 704     indent += length(chunklet);
 705     chunk_line(active_chunk, chunklet);
 706     chunk = substr(chunk, RSTART + RLENGTH);
 707     if (substr(line[1], 1, 1) == "=") {
 708       # chunk name up to }
 709         # FILTHY HACK
 710    gsub("\\\\#", "#", line[3]);
 711    gsub("\\\\textbackslash{}", "\\", line[3]);
 712    gsub("\\\\\\^", "^", line[3]);
 713       chunk_include(active_chunk, line[2] line[3], indent);
 714     } else if (substr(line[1], 1, 1) == "<") {
 715       chunk_include(active_chunk, line[4], indent);
 716     } else {
 717       error("Unknown chunk fragment: " line[1]);
 718     }
 719   }
 720   chunk_line(active_chunk, chunk);
 721   chunk_line(active_chunk, "\n");
 722 }
 723 END {
 724   if (debug) {
 725     print "------ chunk names "
 726     output_chunk_names();
 727     print "====== chunks"
 728     output_chunks();
 729     print "++++++ debug"
 730     for (a in chunks) {
 731       print a "=" chunks[a];
 732     }
 733   }
 734   ORS="";
 735   if (length(root)) output_chunk(root);
 736   else output_chunk_names();
 737 }
 738 function get_chunk_args(text, values,
 739   # optional parameters
 740   path, # hierarchical precursors
 741   # local vars
 742   a, name)
 743 {
 744   split("", next_chunk_args);
 745   while(length(text)) {
 746     if (match(text, "^ *}(.*)", a)) {
 747       return a[1];
 748     }
 749     if (! match(text, " *([^,=]*[^,= ]) *(([,=]) *(([^,}]*) *,* *(.*))|)$", a)) {
 750       return text;
 751     }
 752     name=a[1];
 753     if (a[3] == "=") {
 754       if (substr(a[4],1,1) == "{") {
 755         text = get_chunk_args(substr(a[4],2), values, path name SUBSEP);
 756       } else {
 757         values[path name]=a[5];
 758         text = a[6];
 759       }
 760     } else {
 761       values[path name]="";
 762       text = a[2];
 763     }
 764   }
 765   return text;
 766 }