mp_syntax.mpsl

   1 /*
   2
   3     Minimum Profit 5.x
   4     A Programmer's Text Editor
   5
   6     Syntax highlight definitions.
   7
   8     Copyright (C) 1991-2011 Angel Ortega <angel@triptico.com>
   9
  10     This program is free software; you can redistribute it and/or
  11     modify it under the terms of the GNU General Public License
  12     as published by the Free Software Foundation; either version 2
  13     of the License, or (at your option) any later version.
  14
  15     This program is distributed in the hope that it will be useful,
  16     but WITHOUT ANY WARRANTY; without even the implied warranty of
  17     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  18     GNU General Public License for more details.
  19
  20     You should have received a copy of the GNU General Public License
  21     along with this program; if not, write to the Free Software
  22     Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  23
  24     http://www.triptico.com
  25
  26 */
  27
  28 /** editor actions **/
  29
  30 sub mp.actions.help(d) {
  31     local w, h, r;
  32
  33     /* get the special word regex separator for help (if any) */
  34     r = d.syntax.help_word_regex;
  35
  36     /* if there is a word at the cursor position,
  37         try searching for help */
  38     if ((w = mp.get_word(d, r)) != NULL) {
  39         mp.busy(1);
  40         h = mp.help(d, w);
  41         mp.busy(0);
  42     }
  43
  44     if (h == NULL)
  45         mp.alert(sprintf(L("No help for '%s'."), w));
  46     else {
  47         local hd;
  48
  49         hd = mp.open(sprintf(L("<help on '%s'>"), w));
  50
  51         hd.txt.lines    = h;
  52         hd.txt.y        = 0;
  53         hd.read_only    = 1;
  54
  55         hd.syntax = d.syntax;
  56     }
  57 };
  58
  59 sub mp.actions.section_list(d) {
  60     local l = mp.section_list(d);
  61
  62     if (l == NULL)
  63         mp.alert(L("No detection for sections for this mode."));
  64     else {
  65         if (size(l) == 0)
  66             mp.alert(L("No sections were found in this document."));
  67         else {
  68             local pos = -1;
  69
  70             /* set pos to the section nearest the cursor */
  71             foreach (e, l) {
  72                 if (e[1] > d.txt.y)
  73                     break;
  74
  75                 pos++;
  76             }
  77
  78             if (pos < 0 || pos > size(l))
  79                 pos = 0;
  80
  81             local r = mp.form(
  82                 [
  83                     {
  84                         label:  L("Section list"),
  85                         type:   'list',
  86                         list:   map(l, sub (e) { e[0]; }),
  87                         value:  pos
  88                     }
  89                 ]
  90             );
  91
  92             if (r != NULL) {
  93                 mp.search_set_y(d, l[r[0]][1]);
  94                 mp.set_x(d, 0);
  95             }
  96         }
  97     }
  98 };
  99
 100 /** default key bindings **/
 101
 102 mp.keycodes['f1']       = "help";
 103 mp.keycodes['ctrl-d']   = "section_list";
 104
 105 /** action descriptions **/
 106
 107 mp.actdesc['help']          = LL("Help on word over cursor");
 108 mp.actdesc['section_list']  = LL("Section list...");
 109
 110 /** data **/
 111
 112 mp.syntax = {};
 113
 114 /** syntax definitions **/
 115
 116 sub mp.syn_token_list(l) { '/\b(' ~ join(l, "|") ~ ')\b/'; }
 117
 118 mp.syntax.c = {
 119         'id'            =>      'c',
 120         'name'          =>      'C / C++',
 121         'filenames'     =>      [ '/\.c$/i', '/\.h$/i', '/\.l$/i', '/\.y$/i', '/\.d$/i',
 122                                 '/\.cpp$/i', '/\.hpp$/i', '/\.c++$/i', '/\.cxx$/i', '/\.xpm$/i' ],
 123         'help'          =>      [ "man 2 %s", "man 3 %s" ],
 124     'help_func' =>  sub (w) { mp.c_gather_help(w); },
 125         'defs'          =>      [
 126                 'word1',        [
 127                                 mp.syn_token_list( [
 128                                 "auto", "break", "case", "catch", "class", "const", "const_cast",
 129                                 "continue", "default", "delete", "do", "dynamic_cast", "else", "enum",
 130                                 "explicit", "extern", "for", "friend", "goto", "if", "inline", "mutable",
 131                                 "namespace", "new", "operator", "private", "protected", "public",
 132                                 "register", "reinterpret_cast", "restrict", "return", "sizeof", "static",
 133                                 "static_cast", "struct", "switch", "template", "this", "throw", "try",
 134                                 "typedef", "typeid", "typename", "union", "using", "virtual", "volatile",
 135                                 "while", "not", "not_eq", "xor", "xor_eq", "or",  "or_eq", "and", "and_eq",
 136                                 "bitor", "bitand", "compl"
 137                                 ])
 138                 ],
 139                 'word2',        [
 140                                 mp.syn_token_list( [
 141                                 "bool", "char", "double", "float", "int", "long", "short", "signed",
 142                                 "unsigned", "wchar_t", "size_t", "ssize_t", "off_t", "wchar_t",
 143                                 "ptrdiff_t", "void", "sig_atomic_t", "fpos_t", "clock_t", "time_t",
 144                                 "va_list", "jmp_buf", "FILE", "DIR", "div_t", "ldiv_t", "mbstate_t",
 145                                 "wctrans_t", "wint_t", "wctype_t", "complex", "int8_t", "int16_t",
 146                                 "int32_t", "int64_t", "uint8_t", "uint16_t", "uint32_t", "uint64_t",
 147                                 "int_least8_t", "int_least16_t", "int_least32_t", "int_least64_t",
 148                                 "uint_least8_t", "uint_least16_t", "uint_least32_t", "uint_least64_t",
 149                                 "int_fast8_t", "int_fast16_t", "int_fast32_t", "int_fast64_t",
 150                                 "uint_fast8_t", "uint_fast16_t", "uint_fast32_t", "uint_fast64_t",
 151                                 "intptr_t", "uintptr_t", "intmax_t", "uintmax_t"
 152                                  ]),
 153                                 '/[-=<>:\?\+\*\/\!\%&\|~\^\.]+/'
 154                 ],
 155                 'quotes',       [
 156                         /* from http://ad.hominem.org/log/2005/05/quoted_strings.php */
 157                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 158                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 159                         "/\b-?[0-9]+\b/",                       /* numbers */
 160                         "/\b[0-9]+e[0-9]+\b/",                  /* numbers in exp format */
 161                         "/\b0[0-7]+\b/",                        /* octal numbers */
 162                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 163                         "/\b__(DATE|FILE|LINE|STDC|TIME|STDC_HOSTED|STDC_VERSION|func)__\b/", /* ISO macros */
 164                         "/\b(true|false|NULL)\b/"           /* symbolic constant names */
 165                 ],
 166                 'comments',     [
 167                         [ '|/\*|', '|\*/|' ],                   /* C-like */
 168                         [ '|#if 0|', '#endif' ],                /* CPP "comments" */
 169                         '|//.*$|m',                             /* C++ */
 170                         '/^\s*#\s*[a-z]+/m'                     /* CPP directives */
 171                 ],
 172                 'documentation',        [
 173                         [ "|/\*\*\n|", '|\*/|' ],               /* mp_doccer */
 174                         '/^/\*\*.*\*\*/$/m'                     /* section mark */
 175                 ]
 176         ],
 177         'section'       =>      [ '/(^\/\*\*.*\*\*\/$|^#pragma mark|^[A-Za-z_])/' ]
 178 };
 179
 180 mp.syntax.rc = {
 181         'id'            =>      'rc',
 182         'name'          =>      'Resource file',
 183         'filenames'     =>      [ '/\.rc$/i' ],
 184         'defs'          =>      [
 185                 'word1',        [ ],
 186                 'word2',        [ ],
 187                 'quotes',       [
 188                         /* from http://ad.hominem.org/log/2005/05/quoted_strings.php */
 189                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 190                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 191                         "/\b-?[0-9]+\b/",                       /* numbers */
 192                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 193                         "/\b[0-9[:upper:]_]+\b/"                /* all-caps words */
 194                 ],
 195                 'comments',     [
 196                         [ '|/\*|', '|\*/|' ],                   /* C-like */
 197                         '|//.*$|m',                             /* C++ */
 198                         '/^\s*#[a-z]+/m'                        /* CPP directives */
 199                 ]
 200         ]
 201 };
 202
 203 mp.syntax.perl = {
 204         'id'            =>      'perl',
 205         'name'          =>      'Perl',
 206         'filenames'     =>      [ '/\.pl$/i', '/\.pm$/i' ],
 207         'help'          =>      [ 'perldoc -f %s', 'perldoc %s' ],
 208         'help_word_regex' =>    '/[A-Z_][A-Z0-9_:]*/i',
 209         'defs'  =>      [
 210                 'word1',        [
 211                                 mp.syn_token_list( [
 212                                 "for", "if", "next", "last", "else", "elsif",
 213                                 "unless", "while", "shift", "unshift", "push",
 214                                 "pop", "delete", "new", "bless", "return",
 215                                 "foreach", "keys", "values", "sort", "grep",
 216                                 "tr", "length", "system", "exec", "fork", "map",
 217                                 "print", "write", "open", "close", "chop",
 218                                 "chomp", "exit", "sleep", "split", "join",
 219                                 "sub", "printf", "sprintf", "s", "glob",
 220                                 "scalar", "my", "local", "undef", "defined",
 221                                 "use", "package", "require", "ref", "can", "isa",
 222                                 "qw", "qq", "eq", "ne", "or", "exists",
 223                                 "and", "not", "import", "our", "caller" ]),
 224                                 '/->/'
 225                 ],
 226                 'word2',        [
 227                                 '/[:\?\+\*\/\!\$@\%&\|~\.]+/',
 228                                 '/[\$@%]\w+/'
 229                 ],
 230                 'quotes',       [
 231                         /* from http://ad.hominem.org/log/2005/05/quoted_strings.php */
 232                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 233                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 234                         "/\b-?[0-9]+\b/",                       /* numbers */
 235                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 236                         "/\b[0-9[:upper:]_]+\b/",               /* all-caps words */
 237                         [ "/q\(/", "/\)/" ],                    /* quote */
 238                         [ "/qw\(/", "/\)/" ],                   /* quote word */
 239                         '/\w+\s*=>/',                           /* barewords as hash keys 1 */
 240                         '/\{\s*-?\w+\s*\}/',                    /* barewords as hash keys 2 */
 241                         [ "/`/", "/`/" ],                       /* backticks */
 242                         [ "/<<[\"']?EOF.*$/m", "/^EOF$/m" ]     /* 'document here' */
 243                 ],
 244                 /* color all => as word2 */
 245                 'word2',        [ '/=>/' ],
 246                 /* color curly brackets as word */
 247                 'word',         [ '/[{}]/' ],
 248                 'comments',     [
 249                         "/#.*$/m"                               /* Comments */
 250                 ],
 251                 'documentation',        [
 252                         "/__END__\n.*$/",                       /* __END__ */
 253                         '/^## .*$/m',                           /* section mark */
 254                         [ "/^=(head[1-4]|over|item|back|pod|begin|end|for)/m",
 255                                 "/^=cut$/m" ]                   /* POD */
 256                 ]
 257         ],
 258         'detect'        =>      sub (d) {
 259                 /* take the first line */
 260                 local f = d.txt.lines[0];
 261
 262                 /* is it a 'she-bang' for Perl? */
 263                 return regex(f, '/^#!\s*/usr/bin/(env )?perl/');
 264         },
 265         'section'       =>      [ '/(^sub \w+|^package|^## )/' ]
 266 };
 267
 268
 269 mp.syntax.mpsl = {
 270         'id'            =>      'mpsl',
 271         'name'          =>      'MPSL',
 272         'filenames'     =>      [ '/\.mpsl?$/i' ],
 273         'defs'          =>      [
 274                 'word1',        [
 275                                 mp.syn_token_list( [
 276                                         'if', 'else', 'while', 'foreach',
 277                                         'sub', 'break', 'return', 'eq', 'ne' ]),
 278                                 mp.syn_token_list( keys(MPSL.CORE) ),
 279                                 '/[\{\}]/'
 280                 ],
 281                 'word2',        [
 282                                 "/(NULL|local)/",
 283                                 '/[-=<>\?\+\*\/\!\%&\|]+/'
 284                 ],
 285                 'quotes',       [
 286             /* from http://ad.hominem.org/log/2005/05/quoted_strings.php */
 287             "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/", /* double-quoted strings */
 288             "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",     /* single-quoted strings */
 289             "/\b-?[0-9]+\b/",                           /* numbers */
 290             "/\b0x[0-9a-f]+\b/i",                       /* hex numbers */
 291             '/\w+\s*:/',                                /* JS-type hashes */
 292             "/\b[0-9[:upper:]_]+\b/"                    /* all-caps words */
 293         ],
 294         'word2',    [
 295             '/:/'
 296         ],
 297                 'comments',     [
 298                         [ '|/\*|', '|\*/|' ]                    /* C-like */
 299                 ],
 300                 'documentation',        [
 301                         [ "|/\*\*\n|", '|\*/|' ],               /* mp_doccer */
 302                         '/^/\*\*.*\*\*/$/m'                     /* section mark */
 303                 ]
 304         ],
 305         'detect'        =>      sub (d) {
 306                 /* take the first line */
 307                 local f = d.txt.lines[0];
 308
 309                 return regex(f, '|^#!\s*/usr/bin/(env )?mpsl|');
 310         },
 311         'section'       =>      [ '/(^[A-Za-z_]|^\/\*\*.*\*\*\/$)/' ]
 312 };
 313
 314 mp.syntax.sh = {
 315         'id'            =>      'sh',
 316         'name'          =>      'Shell script',
 317         'filenames'     =>      [ '/\.sh$/i', '/makefile/i' ],
 318         'defs'          =>      [
 319                 'word1',        [
 320                                 mp.syn_token_list( [
 321                                 "if", "then", "else", "elif",
 322                                 "fi", "case", "do", "done", "esac",
 323                                 "for", "until", "while", "break",
 324                                 "in", "source", "alias", "cd",
 325                                 "continue", "echo", "eval", "exec",
 326                                 "exit", "export", "kill", "logout",
 327                                 "printf", "pwd", "read", "return",
 328                                 "shift", "test", "trap", "ulimit",
 329                                 "umask", "unset", "wait", "cp", "rm" ]),
 330                                 '/[\{\}]/'
 331
 332                 ],
 333                 'word2',        [
 334                                 '/\b(local|let|set)\b/',
 335                                 '/[-=<>:\?\+\*\!\%&\|]+/',
 336                                 '/\$\w+/',
 337                                 '/\$\{\w+\}/',
 338                                 "/\b[0-9[:upper:]_]+\b/"        /* all-caps words */
 339                 ],
 340                 'quotes',       [
 341                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 342                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 343                         "/\b-?[0-9]+\b/",                       /* numbers */
 344                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 345                         "/\([A-Za-z0-9_]+\)/",                  /* parens */
 346                         [ "/`/", "/`/" ],                       /* backticks */
 347                         [ "/<<[\"']?EOF[\"']?;$/m", "/^EOF$/m" ]        /* 'document here' */
 348                 ],
 349                 'comments',     [ "/#.*$/m" ]
 350         ],
 351         'detect'        =>      sub (d) {
 352                 /* take the first line */
 353                 local f = d.txt.lines[0];
 354
 355                 /* is it a 'she-bang' for usual shells? */
 356                 return regex(f, '/^#!\s*/bin/(sh|bash|csh|dash|ksh)/') ||
 357                        regex(f, '/^#!\s*/usr/bin/make -f/');
 358         },
 359         'section'       =>      [ '/(^\w+\(\))/', '/^[A-Za-z0-9_\.-]+:/' ]
 360 };
 361
 362
 363 mp.syntax.html = {
 364         'id'            =>      'html',
 365         'name'          =>      'HTML',
 366         'filenames'     =>      [ '/\.html$/i', '/\.htm$/i' ],
 367         'defs'          =>      [
 368                 'word1',        [
 369                                 "/<[\/]?[ \t]*(" ~
 370                                 join([
 371                                 "a", "abbr", "acronym", "address",
 372                                 "area", "b", "base", "bdo", "big",
 373                                 "blockquote", "body", "br", "button",
 374                                 "caption", "center", "cite", "code", "col",
 375                                 "colgroup", "dd", "del", "dfn", "div",
 376                                 "dl", "dt", "em", "fieldset", "form",
 377                                 "h1", "h2", "h3", "h4", "h5", "h6",
 378                                 "head", "hr", "html", "i", "img",
 379                                 "input", "ins", "kbd", "label", "legend",
 380                                 "li", "link", "map", "meta", "noscript",
 381                                 "object", "ol", "optgroup", "option",
 382                                 "p", "param", "pre", "q", "samp",
 383                                 "script", "select", "small", "span",
 384                                 "strong", "style", "sub", "sup", "table",
 385                                 "tbody", "td", "textarea", "tfoot", "th",
 386                                 "thead", "title", "tr", "tt", "ul",
 387                                 "var" ], "|") ~
 388                                 ")[^<>]*>/i"
 389                 ],
 390                 'word2',        [
 391                                 mp.syn_token_list( [
 392                                  "!DOCTYPE", "class", "type",
 393                                 "cellspacing", "cellpadding",
 394                                 "href", "align", "valign", "name", "lang",
 395                                 "value", "action", "width", "height",
 396                                 "content", "http-equiv", "src", "alt",
 397                                 "bgcolor", "text", "link", "vlink", "alink",
 398                                 "media" ])
 399                 ],
 400                 'quotes',       [
 401                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 402                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/"          /* single-quoted strings */
 403                 ],
 404                 'comments',     [
 405                                 [ '/<!--/', '/-->/' ]
 406                 ]
 407         ],
 408         'section'       =>      [ '/<\s*h[0-9]\s*>/' ]
 409 };
 410
 411
 412 mp.syntax.conf = {
 413         'id'            =>      'conf',
 414         'name'          =>      'Config file',
 415         'filenames'     =>      [ '/\.conf$/i', '/\.cfg$/i', '/^.*\/?\.[0-9a-z_-]+rc$/i' ],
 416         'defs'          =>      [
 417                 'word1',        [ '/^\[.+\]$/m' ],
 418                 'word2',        [ "/^[^:=\n]+[:=]/m" ],
 419                 'quotes',       [
 420                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 421                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 422                         "/\b-?[0-9]+\b/",                       /* numbers */
 423                         "/\b0x[0-9a-f]+\b/i"                    /* hex numbers */
 424                 ],
 425                 'comments',     [ "/#.*$/m" ]
 426         ]
 427 };
 428
 429
 430 mp.syntax.php = {
 431         'id'            =>      'php',
 432         'name'          =>      'PHP',
 433         'filenames'     =>      [ '/\.php[345]?$/i', '/\.inc$/i' ],
 434         'defs'          =>      [
 435                 'word1',        [
 436                                 mp.syn_token_list( [
 437                                 "and", "array", "as",
 438                                 "bool", "boolean", "break", "case", "class",
 439                                 "const", "continue", "declare", "default",
 440                                 "die", "do", "double", "echo", "else", "elseif",
 441                                 "empty", "enddeclare", "endfor", "endforeach",
 442                                 "endif", "endswitch", "endwhile", "eval",
 443                                 "exit", "extends", "__FILE__", "float", "for",
 444                                 "foreach", "function", "cfunction", "global",
 445                                 "if", "include", "include_once", "int",
 446                                 "integer", "isset", "__LINE__", "list", "new",
 447                                 "object", "old_function", "or", "print", "real",
 448                                 "require", "require_once", "return",
 449                                 "static", "string", "switch", "unset", "use",
 450                                 "var", "while", "xor", 'true', 'false' ]
 451                                 )
 452                 ],
 453                 'word2',        [ '/\$\w+/', '/[-=<>:\?\+\*\!\%&\|]+/' ],
 454                 'quotes',       [
 455                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 456                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 457                         "/\b-?[0-9]+\b/",                       /* numbers */
 458                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 459                         "/\b[0-9[:upper:]_]+\b/",               /* all-caps words */
 460                         [ "/`/", "/`/" ]                        /* backticks */
 461                 ],
 462                 'comments',     [
 463                         [ '|/\*|', '|\*/|' ],                   /* C-like */
 464                         '|//.*$|m'                              /* C++ */
 465                 ]
 466         ]
 467 };
 468
 469 mp.syntax.python = {
 470         'id'            =>      'python',
 471         'name'          =>      'Python',
 472         'filenames'     =>      [ '/\.py$/i' ],
 473         'defs'          =>      [
 474                 'word1',        [
 475                         mp.syn_token_list( [
 476                                 "and", "assert", "break", "class", "continue",
 477                                 "def", "del", "elif", "else", "except", "exec",
 478                                 "finally", "for", "from", "if", "import", "in",
 479                                 "is", "lambda", "not", "or", "pass", "print",
 480                                 "raise", "return", "try", "while", "yield"
 481                                 ]
 482                         )
 483                 ],
 484                 'word2',        [ '/global/', '/\$\w+/', '/[-=<>:\?\+\*\!\%&\|{}]+/' ],
 485                 'quotes',       [
 486                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 487                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 488                         "/\b-?[0-9]+\b/",                       /* numbers */
 489                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 490                         "/\b[0-9[:upper:]_]+\b/",               /* all-caps words */
 491                         [ "/`/", "/`/" ]                        /* backticks */
 492                 ],
 493                 'comments',             [ "/#.*$/m" ],
 494                 'documentation',        [ [ '/""".+[^"]$/m', '/"""$/m' ] ]
 495         ],
 496         'detect'        =>      sub (d) {
 497                 /* take the first line */
 498                 local f = d.txt.lines[0];
 499
 500                 /* is it a 'she-bang'? */
 501                 return regex(f, '/^#!\s*/usr/bin/(env )?python/');
 502         },
 503         'section'       =>      [ '/^[ \t]*def/' ]
 504 };
 505
 506 mp.syntax.ruby = {
 507         'id'            =>      'ruby',
 508         'name'          =>      'Ruby',
 509         'filenames'     =>      [ '/\.rb$/i' ],
 510         'defs'          =>      [
 511                 'word1',        [
 512                         mp.syn_token_list( [
 513                                 "BEGIN", "END", "alias", "and", "begin",
 514                                 "break", "case", "class", "def", "defined",
 515                                 "do", "else", "elsif", "end", "ensure",
 516                                 "false", "for", "if", "in", "module", "next",
 517                                 "nil", "not", "or", "redo", "rescue", "retry",
 518                                 "return", "self", "super", "then", "true",
 519                                 "undef", "unless", "until", "when", "while",
 520                                 "yield", "require", "include" ]
 521                         )
 522                 ],
 523                 'word2',        [ '/[-=<>:\?\+\*\!\%&\|{}]+/', '/=(begin|end)/' ],
 524                 'quotes',       [
 525                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 526                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",         /* single-quoted strings */
 527                         "/\b-?[0-9]+\b/",                       /* numbers */
 528                         "/\b0x[0-9a-f]+\b/i",                   /* hex numbers */
 529                         "/\b[0-9[:upper:]_]+\b/",               /* all-caps words */
 530                         [ "/`/", "/`/" ]                        /* backticks */
 531                 ],
 532                 'comments',     [ "/#.*$/m" ]
 533         ],
 534         'detect'        =>      sub (d) {
 535                 /* take the first line */
 536                 local f = d.txt.lines[0];
 537
 538                 /* is it a 'she-bang'? */
 539                 return regex(f, '/^#!\s*/usr/bin/(env )?ruby/');
 540         }
 541 };
 542
 543 mp.syntax.diff = {
 544     id:         'diff',
 545     name:       'diff',
 546     filenames:  [
 547         '/\.diff$/i',
 548         '/\.patch$/i'
 549     ],
 550     defs: [
 551         'word1',    [ '/^\+.+$/m' ],
 552         'word2',    [ '/^\-.+$/m' ],
 553         'quotes',   [ '/^@@.+@@.*$/m' ]
 554     ],
 555     section:        [ '/^--- ', '/^@@/' ],
 556     detect:         sub (d) {
 557         local r, n;
 558
 559         n = 0;
 560         while (n < 100) {
 561             if (regex(d.txt.lines[n++], '/^--- /'))
 562                 break;
 563         }
 564
 565         if (regex(d.txt.lines[n++], '/^\+\+\+ /') &&
 566             regex(d.txt.lines[n++], '/^@@ /'))
 567             r = 1;
 568
 569         return r;
 570     }
 571 };
 572
 573 mp.syntax.commit_msg = {
 574         'id'            =>      'commit_msg',
 575         'name'          =>      'VCS commit message',
 576         'filenames'     =>      [ '/sv[kn]-commit.*\./' ],
 577         'defs'          =>      [
 578                 'word1',        [ '/^AM? .+$/m' ],
 579                 'word2',        [ '/^D .+$/m' ],
 580                 'quotes',       [ '/^M .+$/m' ],
 581                 'comments',     [ '/^=== .+$/m', '/^--.+$/m' ]
 582         ]
 583 };
 584
 585 mp.syntax.po = {
 586         'id'            =>      'po',
 587         'name'          =>      'Gettext file',
 588         'filenames'     =>      [ '/\.po$/i' ],
 589         'defs'          =>      [
 590                 'word1',        [ '/^msgid/m' ],
 591                 'word2',        [ '/^msgstr/m' ],
 592                 'comments',     [ "/#.*$/m" ],
 593                 'quotes',       [
 594                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/"      /* double-quoted strings */
 595                 ]
 596         ]
 597 };
 598
 599 mp.syntax.xml = {
 600         'id'            =>      'xml',
 601         'name'          =>      'XML / XGML',
 602         'filenames'     =>      [ '/\.xml$/i', '/\.sgml$/i' ],
 603         'defs'          =>      [
 604                 'word1',        [ '/<[^>]+>/' ],
 605                 'word2',        [ '/<\?[^\?]+\?>/' ],
 606                 'quotes',       [
 607                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",     /* double-quoted strings */
 608                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/"          /* single-quoted strings */
 609                 ],
 610                 'comments',     [
 611                                 [ '/<!--/', '/-->/' ]
 612                 ]
 613         ],
 614         'detect'        =>      sub (d) {
 615                 /* take the first line */
 616                 local f = d.txt.lines[0];
 617
 618                 return regex(f, '/<\?xml/');
 619         }
 620 };
 621
 622 mp.syntax.make_output = {
 623         'id'            =>      'make_output',
 624         'name'          =>      'Make output',
 625         'defs'          =>      [
 626                 'word1',        [ "/^.*warning:.*$/m" ],
 627                 'word2',        [ "/^.*error:.*$/m" ]
 628         ]
 629 };
 630
 631 mp.syntax.euphoria = {
 632         'id'            =>      'euphoria',
 633         'name'          =>      'euphoria',
 634         'filenames'     =>      [ '/\.e$/', '/\.eu$/', '/\.ew$/', '/\.ed$/',
 635                                 '/\.ex$/', '/\.exw$/', '/\.exu$/' ],
 636         'defs'          =>      [
 637                 'word1',        [
 638                         mp.syn_token_list( [
 639                                 "as", "and", "break", "by", "case", "constant", "continue", "do", "end",
 640                                 "else", "elsif", "elsifdef",
 641                                 "exit", "entry", "enum", "export", "for", "function", "global", "include",
 642                                 "if", "ifdef", "label",
 643                                 "not", "or", "procedure", "return", "retry", "switch", "then", "type",
 644                                 "to", "while", "with", "without", "xor"
 645                         ] )
 646                 ],
 647                 'word2',        [
 648                         mp.syn_token_list( [
 649                                 "atom", "integer", "sequence", "object"
 650                         ] )
 651                 ],
 652                 'quotes',       [
 653                         "/\"([^\"\\\\\n]*(\\\\.[^\"\\\\\n]*)*)\"/",    /* double-quoted strings */
 654                         "/'([^'\\\\\n]*(\\\\.[^'\\\\\n]*)*)'/",        /* single-quoted strings */
 655                         "/\b-?#[0-9A-F]+\b/",                /* hex numbers */
 656                         "/\b-?[0-9]+\b/",                /* numbers */
 657                         "/\b[0-9[:upper:]_]+\b/"            /* all-caps words */
 658                 ],
 659                 'comments',             [ "/--.*$/m" ]
 660         ],
 661         'section'       =>      [ '/^[ \t]*(global|export)*[ \t]*(function|procedure)/' ]
 662 };
 663
 664
 665 mp.syntax.mp_templates = {
 666         'id'            => 'mp_templates',
 667         'name'          => 'Minimum Profit template file',
 668         'filenames'     => [ '/\.mp_templates$/' ],
 669         'defs'          => [
 670                 'documentation',        [ '/^%%.*$/m' ]
 671         ],
 672         'section'       => [ '/^%%/' ]
 673 };
 674
 675
 676 mp.syntax.hex_view = {
 677         'id'            => 'hex_view',
 678         'name'          => 'Hexadecimal view',
 679         'defs'          => [
 680                 'word1',                [ "/^\| [0-9A-F]+ \|/m" ],
 681                 'word2',                [ "/\|/" ]
 682         ]
 683 };
 684
 685
 686 mp.syntax.grutatxt = {
 687         'id'            => 'grutatxt',
 688         'name'          => 'Grutatxt',
 689         'defs'          => [
 690                 'word1',        [ "/[^\n]+\n[-=~]+\n/", "/ \* [^.\n:]+:/" ],
 691                 'comments',     [ "/\b_[^ \t\n][^_\n]*[^ \t\n]_\b/" ],
 692                 'word2',        [ "/\*[^ \t\n][^\*\n]*[^ \t\n]\*/" ],
 693                 'quotes',       [ "/`[^ \t\n][^'\n]*[^ \t\n]'/" ]
 694         ],
 695         'section'       => [ "/^[-=~]+$/" ],
 696         'detect'        =>      sub (d) {
 697                 local n = 0;
 698
 699                 /* search the first lines for a heading */
 700                 while (n < 40) {
 701                         local l = d.txt.lines[n++];
 702
 703                         if (l == NULL)
 704                                 break;
 705
 706                         if (regex(l, "/^[-=~]+$/"))
 707                                 return 1;
 708                 }
 709
 710                 return 0;
 711         }
 712 };
 713
 714
 715 /** code **/
 716
 717 sub mp.detect_syntax(doc)
 718 /* tries to detect the syntax of a document */
 719 {
 720         doc.syntax = NULL;
 721
 722         /* loops the syntax highlight definitions */
 723         foreach (n, keys(mp.syntax)) {
 724                 local s = mp.syntax[n];
 725
 726                 /* test the extensions */
 727                 foreach (ext, s.filenames) {
 728                         if (regex(doc.name, ext)) {
 729                                 doc.syntax = s;
 730                                 return;
 731                         }
 732                 }
 733         }
 734
 735         /* not by extension? try the 'detect' subroutine */
 736         foreach (n, keys(mp.syntax)) {
 737                 local s = mp.syntax[n];
 738
 739                 if (is_exec(s.detect) && s.detect(doc)) {
 740                         doc.syntax = s;
 741                         return;
 742                 }
 743         }
 744 }
 745
 746
 747 sub mp.help(doc, word)
 748 {
 749     local h;
 750
 751     foreach (c, doc.syntax.help) {
 752         local f;
 753
 754         /* format the command */
 755         c = sprintf(c, word);
 756
 757         /* pipe from it */
 758         if ((f = popen(c, "r")) != NULL) {
 759             local l;
 760             h = [];
 761
 762             while ((l = read(f)) != NULL) {
 763                 l = sregex(l, "/\x{0008}./g");
 764                 push(h, mp.chomp(l));
 765             }
 766
 767             /* fails? */
 768             if (pclose(f) != 0)
 769                 h = NULL;
 770         }
 771
 772         /* is there already help? don't look for more */
 773         if (h != NULL)
 774             break;
 775     }
 776
 777     if (h == NULL && doc.syntax.help_func)
 778         h = doc.syntax.help_func(word);
 779
 780     return h;
 781 }
 782
 783
 784 /**
 785  * mp.section_list - Returns the list of sections of a document.
 786  * @doc: the document
 787  *
 788  * Applies the `section' array of regular expressions of the
 789  * document's syntax definition and returns it as an array of
 790  * line and line number pairs.
 791  *
 792  * If the document has no syntax highlight definition, or it has
 793  * one without a `section' definition, NULL is returned. Otherwise,
 794  * an array of line, line number pairs is returned (it can be
 795  * an empty list if the document has no recognizable sections).
 796  */
 797 sub mp.section_list(doc)
 798 {
 799     local r;
 800
 801     if (doc.syntax.section) {
 802         local n, l;
 803
 804         r = [];
 805
 806         while ((l = doc.txt.lines[n]) != NULL) {
 807             foreach (ex, doc.syntax.section) {
 808                 if (regex(l, ex)) {
 809                     push(r, [ l, n ]);
 810                     break;
 811                 }
 812             }
 813
 814             n++;
 815         }
 816     }
 817
 818     return r;
 819 }
 820
 821
 822 /**
 823  * mp.c_gather_help - Gathers help in C-style files
 824  * @word: the help term
 825  *
 826  * Searches in all applicable files for code snippets that may
 827  * conform help for @word (mp_doccer style help, struct or
 828  * function definitions, etc).
 829  *
 830  * Returns an array of text with the gathered information.
 831  */
 832 sub mp.c_gather_help(word)
 833 {
 834     local h = [];
 835
 836     foreach (fn, glob()) {
 837         local f;
 838         local ok = 0;
 839
 840         /* test if this file is relevant to C */
 841                 foreach (ext, mp.syntax.c.filenames) {
 842                         if (regex(fn, ext)) {
 843                 ok = 1;
 844                 break;
 845                         }
 846         }
 847
 848         if (ok && (f = open(fn, "r")) != NULL) {
 849             local l;
 850             local n = 0;
 851
 852             while ((l = read(f)) != NULL) {
 853                 n++;
 854                 local where = fn ~ ':' ~ n ~ ':';
 855
 856                 l = mp.chomp(l);
 857
 858                 if (regex(l, "/^[ \t]*#define[ \t]+" ~ word ~ "/")) {
 859                     /* it's a CPP define */
 860                     push(h, where, l);
 861                 }
 862                 else
 863                 if (regex(l, "/^[A-Za-z_]+.*[ \t]+" ~ word ~ "[ \t]*\(.*/")) {
 864                     /* it's a function definition or prototype */
 865                     push(h, where, l);
 866                 }
 867                 else
 868                 if (regex(l, "/^(typedef[ \]*|static[ \]*)?struct[ \t]+.*" ~ word ~ ".*\{/")) {
 869                     /* it's a structure definition */
 870                     push(h, where, l);
 871
 872                     /* add up to the end of the struct */
 873                     while ((l = read(f)) != NULL) {
 874                         n++;
 875                         push(h, l);
 876
 877                         if (regex(l, "/^\};/"))
 878                             break;
 879                     }
 880                 }
 881                 else
 882                 if (regex(l, "/^\/\*\*$/")) {
 883                     /* mp_doccer help: is it for this word? */
 884                     l = read(f);
 885                     n++;
 886
 887                     if (regex(l, "/ \* " ~ word ~ " - /")) {
 888                         /* it is; dump up to a ; */
 889                         push(h, where, '/' ~ '**', l);
 890
 891                         while ((l = read(f)) != NULL) {
 892                             n++;
 893                             if (regex(l, "/^\{/"))
 894                                 break;
 895
 896                             push(h, l);
 897                         }
 898                     }
 899                 }
 900             }
 901
 902             close(f);
 903         }
 904     }
 905
 906     if (size(h)) {
 907         ins(h, sprintf(L("Gathered information on %s:"), word), 0);
 908         ins(h, '', 1);
 909     }
 910     else
 911         h = NULL;
 912
 913     return h;
 914 }