| 1 | # EXTRACT VARIOUSLY DELIMITED TEXT SEQUENCES FROM STRINGS. |
| 2 | # FOR FULL DOCUMENTATION SEE Balanced.pod |
| 3 | |
| 4 | use 5.005; |
| 5 | use strict; |
| 6 | |
| 7 | package Text::Balanced; |
| 8 | |
| 9 | use Exporter; |
| 10 | use SelfLoader; |
| 11 | use vars qw { $VERSION @ISA %EXPORT_TAGS }; |
| 12 | |
| 13 | $VERSION = '1.77'; |
| 14 | @ISA = qw ( Exporter ); |
| 15 | |
| 16 | %EXPORT_TAGS = ( ALL => [ qw( |
| 17 | &extract_delimited |
| 18 | &extract_bracketed |
| 19 | &extract_quotelike |
| 20 | &extract_codeblock |
| 21 | &extract_variable |
| 22 | &extract_tagged |
| 23 | &extract_multiple |
| 24 | |
| 25 | &gen_delimited_pat |
| 26 | &gen_extract_tagged |
| 27 | |
| 28 | &delimited_pat |
| 29 | ) ] ); |
| 30 | |
| 31 | Exporter::export_ok_tags('ALL'); |
| 32 | |
| 33 | # PROTOTYPES |
| 34 | |
| 35 | sub _match_bracketed($$$$$$); |
| 36 | sub _match_variable($$); |
| 37 | sub _match_codeblock($$$$$$$); |
| 38 | sub _match_quotelike($$$$); |
| 39 | |
| 40 | # HANDLE RETURN VALUES IN VARIOUS CONTEXTS |
| 41 | |
| 42 | sub _fail |
| 43 | { |
| 44 | my ($wantarray,$textref) = @_; |
| 45 | return ("",$$textref,"") if $wantarray; |
| 46 | return undef; |
| 47 | } |
| 48 | |
| 49 | sub _succeed |
| 50 | { |
| 51 | $@ = undef; |
| 52 | my ($wantarray,$textref) = splice @_, 0, 2; |
| 53 | if ($wantarray) |
| 54 | { |
| 55 | # print join ("|", @_), "\n"; |
| 56 | my @res; |
| 57 | pos($$textref) = $_[2]; # RESET \G |
| 58 | while (my ($from, $len) = splice @_, 0, 2) |
| 59 | { |
| 60 | push @res, substr($$textref,$from,$len); |
| 61 | } |
| 62 | return @res; |
| 63 | } |
| 64 | else |
| 65 | { |
| 66 | my $match = substr($$textref,$_[0],$_[1]); |
| 67 | eval {substr($$textref,$_[4],$_[1]+$_[5])=""} ; #CHOP OUT PREFIX & MATCH, IF POSSIBLE |
| 68 | pos($$textref) = $_[4]; # RESET \G |
| 69 | return $match; |
| 70 | } |
| 71 | } |
| 72 | |
| 73 | # BUILD A PATTERN MATCHING A SIMPLE DELIMITED STRING |
| 74 | |
| 75 | sub gen_delimited_pat($;$) # ($delimiters;$escapes) |
| 76 | { |
| 77 | my ($dels, $escs) = @_; |
| 78 | return "" unless $dels =~ /\S/; |
| 79 | $escs = '\\' unless $escs; |
| 80 | $escs .= substr($escs,-1) x (length($dels)-length($escs)); |
| 81 | my @pat = (); |
| 82 | my $i; |
| 83 | for ($i=0; $i<length $dels; $i++) |
| 84 | { |
| 85 | my $del = quotemeta substr($dels,$i,1); |
| 86 | my $esc = quotemeta substr($escs,$i,1); |
| 87 | if ($del eq $esc) |
| 88 | { |
| 89 | push @pat, "$del(?:[^$del]*(?:(?:$del$del)[^$del]*)*)$del"; |
| 90 | } |
| 91 | else |
| 92 | { |
| 93 | push @pat, "$del(?:[^$esc$del]*(?:$esc.[^$esc$del]*)*)$del"; |
| 94 | } |
| 95 | } |
| 96 | my $pat = join '|', @pat; |
| 97 | return "(?:$pat)"; |
| 98 | } |
| 99 | |
| 100 | *delimited_pat = \&gen_delimited_pat; |
| 101 | |
| 102 | |
| 103 | # THE EXTRACTION FUNCTIONS |
| 104 | |
| 105 | sub extract_delimited (;$$$$) |
| 106 | { |
| 107 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 108 | my $wantarray = wantarray; |
| 109 | my $del = defined $_[1] ? $_[1] : qq{\'\"\`}; |
| 110 | my $pre = defined $_[2] ? $_[2] : '\s*'; |
| 111 | my $esc = defined $_[3] ? $_[3] : qq{\\}; |
| 112 | my $pat = gen_delimited_pat($del, $esc); |
| 113 | my $startpos = pos $$textref || 0; |
| 114 | return _fail($wantarray, $textref) |
| 115 | unless $$textref =~ m/\G($pre)($pat)/gc; |
| 116 | my $prelen = length($1); |
| 117 | my $matchpos = $startpos+$prelen; |
| 118 | my $endpos = pos $$textref; |
| 119 | return _succeed $wantarray, $textref, |
| 120 | $matchpos, $endpos-$matchpos, # MATCH |
| 121 | $endpos, length($$textref)-$endpos, # REMAINDER |
| 122 | $startpos, $prelen; # PREFIX |
| 123 | } |
| 124 | |
| 125 | sub extract_bracketed (;$$$) |
| 126 | { |
| 127 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 128 | my $ldel = defined $_[1] ? $_[1] : '{([<'; |
| 129 | my $pre = defined $_[2] ? $_[2] : '\s*'; |
| 130 | my $wantarray = wantarray; |
| 131 | my $qdel = ""; |
| 132 | my $quotelike; |
| 133 | $ldel =~ s/'//g and $qdel .= q{'}; |
| 134 | $ldel =~ s/"//g and $qdel .= q{"}; |
| 135 | $ldel =~ s/`//g and $qdel .= q{`}; |
| 136 | $ldel =~ s/q//g and $quotelike = 1; |
| 137 | $ldel =~ tr/[](){}<>\0-\377/[[(({{<</ds; |
| 138 | my $rdel = $ldel; |
| 139 | unless ($rdel =~ tr/[({</])}>/) |
| 140 | { |
| 141 | $@ = "Did not find a suitable bracket in delimiter: \"$_[1]\""; |
| 142 | return _fail $wantarray, $textref; |
| 143 | } |
| 144 | my $posbug = pos; |
| 145 | $ldel = join('|', map { quotemeta $_ } split('', $ldel)); |
| 146 | $rdel = join('|', map { quotemeta $_ } split('', $rdel)); |
| 147 | pos = $posbug; |
| 148 | |
| 149 | my $startpos = pos $$textref || 0; |
| 150 | my @match = _match_bracketed($textref,$pre, $ldel, $qdel, $quotelike, $rdel); |
| 151 | |
| 152 | return _fail ($wantarray, $textref) unless @match; |
| 153 | |
| 154 | return _succeed ( $wantarray, $textref, |
| 155 | $match[2], $match[5]+2, # MATCH |
| 156 | @match[8,9], # REMAINDER |
| 157 | @match[0,1], # PREFIX |
| 158 | ); |
| 159 | } |
| 160 | |
| 161 | sub _match_bracketed($$$$$$) # $textref, $pre, $ldel, $qdel, $quotelike, $rdel |
| 162 | { |
| 163 | my ($textref, $pre, $ldel, $qdel, $quotelike, $rdel) = @_; |
| 164 | my ($startpos, $ldelpos, $endpos) = (pos $$textref = pos $$textref||0); |
| 165 | unless ($$textref =~ m/\G$pre/gc) |
| 166 | { |
| 167 | $@ = "Did not find prefix: /$pre/"; |
| 168 | return; |
| 169 | } |
| 170 | |
| 171 | $ldelpos = pos $$textref; |
| 172 | |
| 173 | unless ($$textref =~ m/\G($ldel)/gc) |
| 174 | { |
| 175 | $@ = "Did not find opening bracket after prefix: \"$pre\""; |
| 176 | pos $$textref = $startpos; |
| 177 | return; |
| 178 | } |
| 179 | |
| 180 | my @nesting = ( $1 ); |
| 181 | my $textlen = length $$textref; |
| 182 | while (pos $$textref < $textlen) |
| 183 | { |
| 184 | next if $$textref =~ m/\G\\./gcs; |
| 185 | |
| 186 | if ($$textref =~ m/\G($ldel)/gc) |
| 187 | { |
| 188 | push @nesting, $1; |
| 189 | } |
| 190 | elsif ($$textref =~ m/\G($rdel)/gc) |
| 191 | { |
| 192 | my ($found, $brackettype) = ($1, $1); |
| 193 | if ($#nesting < 0) |
| 194 | { |
| 195 | $@ = "Unmatched closing bracket: \"$found\""; |
| 196 | pos $$textref = $startpos; |
| 197 | return; |
| 198 | } |
| 199 | my $expected = pop(@nesting); |
| 200 | $expected =~ tr/({[</)}]>/; |
| 201 | if ($expected ne $brackettype) |
| 202 | { |
| 203 | $@ = qq{Mismatched closing bracket: expected "$expected" but found "$found"}; |
| 204 | pos $$textref = $startpos; |
| 205 | return; |
| 206 | } |
| 207 | last if $#nesting < 0; |
| 208 | } |
| 209 | elsif ($qdel && $$textref =~ m/\G([$qdel])/gc) |
| 210 | { |
| 211 | $$textref =~ m/\G[^\\$1]*(?:\\.[^\\$1]*)*(\Q$1\E)/gc and next; |
| 212 | $@ = "Unmatched embedded quote ($1)"; |
| 213 | pos $$textref = $startpos; |
| 214 | return; |
| 215 | } |
| 216 | elsif ($quotelike && _match_quotelike($textref,"",1,0)) |
| 217 | { |
| 218 | next; |
| 219 | } |
| 220 | |
| 221 | else { $$textref =~ m/\G(?:[a-zA-Z0-9]+|.)/gcs } |
| 222 | } |
| 223 | if ($#nesting>=0) |
| 224 | { |
| 225 | $@ = "Unmatched opening bracket(s): " |
| 226 | . join("..",@nesting).".."; |
| 227 | pos $$textref = $startpos; |
| 228 | return; |
| 229 | } |
| 230 | |
| 231 | $endpos = pos $$textref; |
| 232 | |
| 233 | return ( |
| 234 | $startpos, $ldelpos-$startpos, # PREFIX |
| 235 | $ldelpos, 1, # OPENING BRACKET |
| 236 | $ldelpos+1, $endpos-$ldelpos-2, # CONTENTS |
| 237 | $endpos-1, 1, # CLOSING BRACKET |
| 238 | $endpos, length($$textref)-$endpos, # REMAINDER |
| 239 | ); |
| 240 | } |
| 241 | |
| 242 | sub revbracket($) |
| 243 | { |
| 244 | my $brack = reverse $_[0]; |
| 245 | $brack =~ tr/[({</])}>/; |
| 246 | return $brack; |
| 247 | } |
| 248 | |
| 249 | my $XMLNAME = q{[a-zA-Z_:][a-zA-Z0-9_:.-]*}; |
| 250 | |
| 251 | sub extract_tagged (;$$$$$) # ($text, $opentag, $closetag, $pre, \%options) |
| 252 | { |
| 253 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 254 | my $ldel = $_[1]; |
| 255 | my $rdel = $_[2]; |
| 256 | my $pre = defined $_[3] ? $_[3] : '\s*'; |
| 257 | my %options = defined $_[4] ? %{$_[4]} : (); |
| 258 | my $omode = defined $options{fail} ? $options{fail} : ''; |
| 259 | my $bad = ref($options{reject}) eq 'ARRAY' ? join('|', @{$options{reject}}) |
| 260 | : defined($options{reject}) ? $options{reject} |
| 261 | : '' |
| 262 | ; |
| 263 | my $ignore = ref($options{ignore}) eq 'ARRAY' ? join('|', @{$options{ignore}}) |
| 264 | : defined($options{ignore}) ? $options{ignore} |
| 265 | : '' |
| 266 | ; |
| 267 | |
| 268 | if (!defined $ldel) { $ldel = '<\w+(?:' . gen_delimited_pat(q{'"}) . '|[^>])*>'; } |
| 269 | $@ = undef; |
| 270 | |
| 271 | my @match = _match_tagged($textref, $pre, $ldel, $rdel, $omode, $bad, $ignore); |
| 272 | |
| 273 | return _fail(wantarray, $textref) unless @match; |
| 274 | return _succeed wantarray, $textref, |
| 275 | $match[2], $match[3]+$match[5]+$match[7], # MATCH |
| 276 | @match[8..9,0..1,2..7]; # REM, PRE, BITS |
| 277 | } |
| 278 | |
| 279 | sub _match_tagged # ($$$$$$$) |
| 280 | { |
| 281 | my ($textref, $pre, $ldel, $rdel, $omode, $bad, $ignore) = @_; |
| 282 | my $rdelspec; |
| 283 | |
| 284 | my ($startpos, $opentagpos, $textpos, $parapos, $closetagpos, $endpos) = ( pos($$textref) = pos($$textref)||0 ); |
| 285 | |
| 286 | unless ($$textref =~ m/\G($pre)/gc) |
| 287 | { |
| 288 | $@ = "Did not find prefix: /$pre/"; |
| 289 | goto failed; |
| 290 | } |
| 291 | |
| 292 | $opentagpos = pos($$textref); |
| 293 | |
| 294 | unless ($$textref =~ m/\G$ldel/gc) |
| 295 | { |
| 296 | $@ = "Did not find opening tag: /$ldel/"; |
| 297 | goto failed; |
| 298 | } |
| 299 | |
| 300 | $textpos = pos($$textref); |
| 301 | |
| 302 | if (!defined $rdel) |
| 303 | { |
| 304 | $rdelspec = $&; |
| 305 | unless ($rdelspec =~ s/\A([[(<{]+)($XMLNAME).*/ "$1\/$2". revbracket($1) /oes) |
| 306 | { |
| 307 | $@ = "Unable to construct closing tag to match: $rdel"; |
| 308 | goto failed; |
| 309 | } |
| 310 | } |
| 311 | else |
| 312 | { |
| 313 | $rdelspec = eval "qq{$rdel}"; |
| 314 | } |
| 315 | |
| 316 | while (pos($$textref) < length($$textref)) |
| 317 | { |
| 318 | next if $$textref =~ m/\G\\./gc; |
| 319 | |
| 320 | if ($$textref =~ m/\G(\n[ \t]*\n)/gc ) |
| 321 | { |
| 322 | $parapos = pos($$textref) - length($1) |
| 323 | unless defined $parapos; |
| 324 | } |
| 325 | elsif ($$textref =~ m/\G($rdelspec)/gc ) |
| 326 | { |
| 327 | $closetagpos = pos($$textref)-length($1); |
| 328 | goto matched; |
| 329 | } |
| 330 | elsif ($ignore && $$textref =~ m/\G(?:$ignore)/gc) |
| 331 | { |
| 332 | next; |
| 333 | } |
| 334 | elsif ($bad && $$textref =~ m/\G($bad)/gcs) |
| 335 | { |
| 336 | pos($$textref) -= length($1); # CUT OFF WHATEVER CAUSED THE SHORTNESS |
| 337 | goto short if ($omode eq 'PARA' || $omode eq 'MAX'); |
| 338 | $@ = "Found invalid nested tag: $1"; |
| 339 | goto failed; |
| 340 | } |
| 341 | elsif ($$textref =~ m/\G($ldel)/gc) |
| 342 | { |
| 343 | my $tag = $1; |
| 344 | pos($$textref) -= length($tag); # REWIND TO NESTED TAG |
| 345 | unless (_match_tagged(@_)) # MATCH NESTED TAG |
| 346 | { |
| 347 | goto short if $omode eq 'PARA' || $omode eq 'MAX'; |
| 348 | $@ = "Found unbalanced nested tag: $tag"; |
| 349 | goto failed; |
| 350 | } |
| 351 | } |
| 352 | else { $$textref =~ m/./gcs } |
| 353 | } |
| 354 | |
| 355 | short: |
| 356 | $closetagpos = pos($$textref); |
| 357 | goto matched if $omode eq 'MAX'; |
| 358 | goto failed unless $omode eq 'PARA'; |
| 359 | |
| 360 | if (defined $parapos) { pos($$textref) = $parapos } |
| 361 | else { $parapos = pos($$textref) } |
| 362 | |
| 363 | return ( |
| 364 | $startpos, $opentagpos-$startpos, # PREFIX |
| 365 | $opentagpos, $textpos-$opentagpos, # OPENING TAG |
| 366 | $textpos, $parapos-$textpos, # TEXT |
| 367 | $parapos, 0, # NO CLOSING TAG |
| 368 | $parapos, length($$textref)-$parapos, # REMAINDER |
| 369 | ); |
| 370 | |
| 371 | matched: |
| 372 | $endpos = pos($$textref); |
| 373 | return ( |
| 374 | $startpos, $opentagpos-$startpos, # PREFIX |
| 375 | $opentagpos, $textpos-$opentagpos, # OPENING TAG |
| 376 | $textpos, $closetagpos-$textpos, # TEXT |
| 377 | $closetagpos, $endpos-$closetagpos, # CLOSING TAG |
| 378 | $endpos, length($$textref)-$endpos, # REMAINDER |
| 379 | ); |
| 380 | |
| 381 | failed: |
| 382 | $@ = "Did not find closing tag" unless $@; |
| 383 | pos($$textref) = $startpos; |
| 384 | return; |
| 385 | } |
| 386 | |
| 387 | sub extract_variable (;$$) |
| 388 | { |
| 389 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 390 | return ("","","") unless defined $$textref; |
| 391 | my $pre = defined $_[1] ? $_[1] : '\s*'; |
| 392 | |
| 393 | my @match = _match_variable($textref,$pre); |
| 394 | |
| 395 | return _fail wantarray, $textref unless @match; |
| 396 | |
| 397 | return _succeed wantarray, $textref, |
| 398 | @match[2..3,4..5,0..1]; # MATCH, REMAINDER, PREFIX |
| 399 | } |
| 400 | |
| 401 | sub _match_variable($$) |
| 402 | { |
| 403 | my ($textref, $pre) = @_; |
| 404 | my $startpos = pos($$textref) = pos($$textref)||0; |
| 405 | unless ($$textref =~ m/\G($pre)/gc) |
| 406 | { |
| 407 | $@ = "Did not find prefix: /$pre/"; |
| 408 | return; |
| 409 | } |
| 410 | my $varpos = pos($$textref); |
| 411 | unless ($$textref =~ m/\G(\$#?|[*\@\%]|\\&)+/gc) |
| 412 | { |
| 413 | $@ = "Did not find leading dereferencer"; |
| 414 | pos $$textref = $startpos; |
| 415 | return; |
| 416 | } |
| 417 | |
| 418 | unless ($$textref =~ m/\G\s*(?:::)?(?:[_a-z]\w*::)*[_a-z]\w*/gci |
| 419 | or _match_codeblock($textref, "", '\{', '\}', '\{', '\}', 0)) |
| 420 | { |
| 421 | $@ = "Bad identifier after dereferencer"; |
| 422 | pos $$textref = $startpos; |
| 423 | return; |
| 424 | } |
| 425 | |
| 426 | while (1) |
| 427 | { |
| 428 | next if _match_codeblock($textref, |
| 429 | qr/\s*->\s*(?:[a-zA-Z]\w+\s*)?/, |
| 430 | qr/[({[]/, qr/[)}\]]/, |
| 431 | qr/[({[]/, qr/[)}\]]/, 0); |
| 432 | next if _match_codeblock($textref, |
| 433 | qr/\s*/, qr/[{[]/, qr/[}\]]/, |
| 434 | qr/[{[]/, qr/[}\]]/, 0); |
| 435 | next if _match_variable($textref,'\s*->\s*'); |
| 436 | next if $$textref =~ m/\G\s*->\s*\w+(?![{([])/gc; |
| 437 | last; |
| 438 | } |
| 439 | |
| 440 | my $endpos = pos($$textref); |
| 441 | return ($startpos, $varpos-$startpos, |
| 442 | $varpos, $endpos-$varpos, |
| 443 | $endpos, length($$textref)-$endpos |
| 444 | ); |
| 445 | } |
| 446 | |
| 447 | sub extract_codeblock (;$$$$$) |
| 448 | { |
| 449 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 450 | my $wantarray = wantarray; |
| 451 | my $ldel_inner = defined $_[1] ? $_[1] : '{'; |
| 452 | my $pre = defined $_[2] ? $_[2] : '\s*'; |
| 453 | my $ldel_outer = defined $_[3] ? $_[3] : $ldel_inner; |
| 454 | my $rd = $_[4]; |
| 455 | my $rdel_inner = $ldel_inner; |
| 456 | my $rdel_outer = $ldel_outer; |
| 457 | my $posbug = pos; |
| 458 | for ($ldel_inner, $ldel_outer) { tr/[]()<>{}\0-\377/[[((<<{{/ds } |
| 459 | for ($rdel_inner, $rdel_outer) { tr/[]()<>{}\0-\377/]]))>>}}/ds } |
| 460 | for ($ldel_inner, $ldel_outer, $rdel_inner, $rdel_outer) |
| 461 | { |
| 462 | $_ = '('.join('|',map { quotemeta $_ } split('',$_)).')' |
| 463 | } |
| 464 | pos = $posbug; |
| 465 | |
| 466 | my @match = _match_codeblock($textref, $pre, |
| 467 | $ldel_outer, $rdel_outer, |
| 468 | $ldel_inner, $rdel_inner, |
| 469 | $rd); |
| 470 | return _fail($wantarray, $textref) unless @match; |
| 471 | return _succeed($wantarray, $textref, |
| 472 | @match[2..3,4..5,0..1] # MATCH, REMAINDER, PREFIX |
| 473 | ); |
| 474 | |
| 475 | } |
| 476 | |
| 477 | sub _match_codeblock($$$$$$$) |
| 478 | { |
| 479 | my ($textref, $pre, $ldel_outer, $rdel_outer, $ldel_inner, $rdel_inner, $rd) = @_; |
| 480 | my $startpos = pos($$textref) = pos($$textref) || 0; |
| 481 | unless ($$textref =~ m/\G($pre)/gc) |
| 482 | { |
| 483 | $@ = qq{Did not match prefix /$pre/ at"} . |
| 484 | substr($$textref,pos($$textref),20) . |
| 485 | q{..."}; |
| 486 | return; |
| 487 | } |
| 488 | my $codepos = pos($$textref); |
| 489 | unless ($$textref =~ m/\G($ldel_outer)/gc) # OUTERMOST DELIMITER |
| 490 | { |
| 491 | $@ = qq{Did not find expected opening bracket at "} . |
| 492 | substr($$textref,pos($$textref),20) . |
| 493 | q{..."}; |
| 494 | pos $$textref = $startpos; |
| 495 | return; |
| 496 | } |
| 497 | my $closing = $1; |
| 498 | $closing =~ tr/([<{/)]>}/; |
| 499 | my $matched; |
| 500 | my $patvalid = 1; |
| 501 | while (pos($$textref) < length($$textref)) |
| 502 | { |
| 503 | $matched = ''; |
| 504 | if ($rd && $$textref =~ m#\G(\Q(?)\E|\Q(s?)\E|\Q(s)\E)#gc) |
| 505 | { |
| 506 | $patvalid = 0; |
| 507 | next; |
| 508 | } |
| 509 | |
| 510 | if ($$textref =~ m/\G\s*#.*/gc) |
| 511 | { |
| 512 | next; |
| 513 | } |
| 514 | |
| 515 | if ($$textref =~ m/\G\s*($rdel_outer)/gc) |
| 516 | { |
| 517 | unless ($matched = ($closing && $1 eq $closing) ) |
| 518 | { |
| 519 | next if $1 eq '>'; # MIGHT BE A "LESS THAN" |
| 520 | $@ = q{Mismatched closing bracket at "} . |
| 521 | substr($$textref,pos($$textref),20) . |
| 522 | qq{...". Expected '$closing'}; |
| 523 | } |
| 524 | last; |
| 525 | } |
| 526 | |
| 527 | if (_match_variable($textref,'\s*') || |
| 528 | _match_quotelike($textref,'\s*',$patvalid,$patvalid) ) |
| 529 | { |
| 530 | $patvalid = 0; |
| 531 | next; |
| 532 | } |
| 533 | |
| 534 | |
| 535 | # NEED TO COVER MANY MORE CASES HERE!!! |
| 536 | if ($$textref =~ m#\G\s*( [-+*x/%^&|.]=? |
| 537 | | =(?!>) |
| 538 | | (\*\*|&&|\|\||<<|>>)=? |
| 539 | | [!=][~=] |
| 540 | | split|grep|map|return |
| 541 | )#gcx) |
| 542 | { |
| 543 | $patvalid = 1; |
| 544 | next; |
| 545 | } |
| 546 | |
| 547 | if ( _match_codeblock($textref, '\s*', $ldel_inner, $rdel_inner, $ldel_inner, $rdel_inner, $rd) ) |
| 548 | { |
| 549 | $patvalid = 1; |
| 550 | next; |
| 551 | } |
| 552 | |
| 553 | if ($$textref =~ m/\G\s*$ldel_outer/gc) |
| 554 | { |
| 555 | $@ = q{Improperly nested codeblock at "} . |
| 556 | substr($$textref,pos($$textref),20) . |
| 557 | q{..."}; |
| 558 | last; |
| 559 | } |
| 560 | |
| 561 | $patvalid = 0; |
| 562 | $$textref =~ m/\G\s*(\w+|[-=>]>|.|\Z)/gc; |
| 563 | } |
| 564 | |
| 565 | unless ($matched) |
| 566 | { |
| 567 | $@ = 'No match found for opening bracket' unless $@; |
| 568 | return; |
| 569 | } |
| 570 | |
| 571 | my $endpos = pos($$textref); |
| 572 | return ( $startpos, $codepos-$startpos, |
| 573 | $codepos, $endpos-$codepos, |
| 574 | $endpos, length($$textref)-$endpos, |
| 575 | ); |
| 576 | } |
| 577 | |
| 578 | |
| 579 | my %mods = ( |
| 580 | 'none' => '[cgimsox]*', |
| 581 | 'm' => '[cgimsox]*', |
| 582 | 's' => '[cegimsox]*', |
| 583 | 'tr' => '[cds]*', |
| 584 | 'y' => '[cds]*', |
| 585 | 'qq' => '', |
| 586 | 'qx' => '', |
| 587 | 'qw' => '', |
| 588 | 'qr' => '[imsx]*', |
| 589 | 'q' => '', |
| 590 | ); |
| 591 | |
| 592 | sub extract_quotelike (;$$) |
| 593 | { |
| 594 | my $textref = $_[0] ? \$_[0] : \$_; |
| 595 | my $wantarray = wantarray; |
| 596 | my $pre = defined $_[1] ? $_[1] : '\s*'; |
| 597 | |
| 598 | my @match = _match_quotelike($textref,$pre,1,0); |
| 599 | return _fail($wantarray, $textref) unless @match; |
| 600 | return _succeed($wantarray, $textref, |
| 601 | $match[2], $match[18]-$match[2], # MATCH |
| 602 | @match[18,19], # REMAINDER |
| 603 | @match[0,1], # PREFIX |
| 604 | @match[2..17], # THE BITS |
| 605 | ); |
| 606 | }; |
| 607 | |
| 608 | sub _match_quotelike($$$$) # ($textref, $prepat, $allow_raw_match) |
| 609 | { |
| 610 | my ($textref, $pre, $rawmatch, $qmark) = @_; |
| 611 | |
| 612 | my ($textlen,$startpos, |
| 613 | $oppos, |
| 614 | $preld1pos,$ld1pos,$str1pos,$rd1pos, |
| 615 | $preld2pos,$ld2pos,$str2pos,$rd2pos, |
| 616 | $modpos) = ( length($$textref), pos($$textref) = pos($$textref) || 0 ); |
| 617 | |
| 618 | unless ($$textref =~ m/\G($pre)/gc) |
| 619 | { |
| 620 | $@ = qq{Did not find prefix /$pre/ at "} . |
| 621 | substr($$textref, pos($$textref), 20) . |
| 622 | q{..."}; |
| 623 | return; |
| 624 | } |
| 625 | $oppos = pos($$textref); |
| 626 | |
| 627 | my $initial = substr($$textref,$oppos,1); |
| 628 | |
| 629 | if ($initial && $initial =~ m|^[\"\'\`]| |
| 630 | || $rawmatch && $initial =~ m|^/| |
| 631 | || $qmark && $initial =~ m|^\?|) |
| 632 | { |
| 633 | unless ($$textref =~ m/ \Q$initial\E [^\\$initial]* (\\.[^\\$initial]*)* \Q$initial\E /gcx) |
| 634 | { |
| 635 | $@ = qq{Did not find closing delimiter to match '$initial' at "} . |
| 636 | substr($$textref, $oppos, 20) . |
| 637 | q{..."}; |
| 638 | pos $$textref = $startpos; |
| 639 | return; |
| 640 | } |
| 641 | $modpos= pos($$textref); |
| 642 | $rd1pos = $modpos-1; |
| 643 | |
| 644 | if ($initial eq '/' || $initial eq '?') |
| 645 | { |
| 646 | $$textref =~ m/\G$mods{none}/gc |
| 647 | } |
| 648 | |
| 649 | my $endpos = pos($$textref); |
| 650 | return ( |
| 651 | $startpos, $oppos-$startpos, # PREFIX |
| 652 | $oppos, 0, # NO OPERATOR |
| 653 | $oppos, 1, # LEFT DEL |
| 654 | $oppos+1, $rd1pos-$oppos-1, # STR/PAT |
| 655 | $rd1pos, 1, # RIGHT DEL |
| 656 | $modpos, 0, # NO 2ND LDEL |
| 657 | $modpos, 0, # NO 2ND STR |
| 658 | $modpos, 0, # NO 2ND RDEL |
| 659 | $modpos, $endpos-$modpos, # MODIFIERS |
| 660 | $endpos, $textlen-$endpos, # REMAINDER |
| 661 | ); |
| 662 | } |
| 663 | |
| 664 | unless ($$textref =~ m!\G(m|s|qq|qx|qw|q|qr|tr|y)\b(?=\s*\S)!gc) |
| 665 | { |
| 666 | $@ = q{No quotelike operator found after prefix at "} . |
| 667 | substr($$textref, pos($$textref), 20) . |
| 668 | q{..."}; |
| 669 | pos $$textref = $startpos; |
| 670 | return; |
| 671 | } |
| 672 | |
| 673 | my $op = $1; |
| 674 | $preld1pos = pos($$textref); |
| 675 | $$textref =~ m/\G\s*/gc; |
| 676 | $ld1pos = pos($$textref); |
| 677 | $str1pos = $ld1pos+1; |
| 678 | |
| 679 | unless ($$textref =~ m/\G(\S)/gc) # SHOULD USE LOOKAHEAD |
| 680 | { |
| 681 | $@ = "No block delimiter found after quotelike $op"; |
| 682 | pos $$textref = $startpos; |
| 683 | return; |
| 684 | } |
| 685 | pos($$textref) = $ld1pos; # HAVE TO DO THIS BECAUSE LOOKAHEAD BROKEN |
| 686 | my ($ldel1, $rdel1) = ("\Q$1","\Q$1"); |
| 687 | if ($ldel1 =~ /[[(<{]/) |
| 688 | { |
| 689 | $rdel1 =~ tr/[({</])}>/; |
| 690 | _match_bracketed($textref,"",$ldel1,"","",$rdel1) |
| 691 | || do { pos $$textref = $startpos; return }; |
| 692 | } |
| 693 | else |
| 694 | { |
| 695 | $$textref =~ /$ldel1[^\\$ldel1]*(\\.[^\\$ldel1]*)*$ldel1/gc |
| 696 | || do { pos $$textref = $startpos; return }; |
| 697 | } |
| 698 | $ld2pos = $rd1pos = pos($$textref)-1; |
| 699 | |
| 700 | my $second_arg = $op =~ /s|tr|y/ ? 1 : 0; |
| 701 | if ($second_arg) |
| 702 | { |
| 703 | my ($ldel2, $rdel2); |
| 704 | if ($ldel1 =~ /[[(<{]/) |
| 705 | { |
| 706 | unless ($$textref =~ /\G\s*(\S)/gc) # SHOULD USE LOOKAHEAD |
| 707 | { |
| 708 | $@ = "Missing second block for quotelike $op"; |
| 709 | pos $$textref = $startpos; |
| 710 | return; |
| 711 | } |
| 712 | $ldel2 = $rdel2 = "\Q$1"; |
| 713 | $rdel2 =~ tr/[({</])}>/; |
| 714 | } |
| 715 | else |
| 716 | { |
| 717 | $ldel2 = $rdel2 = $ldel1; |
| 718 | } |
| 719 | $str2pos = $ld2pos+1; |
| 720 | |
| 721 | if ($ldel2 =~ /[[(<{]/) |
| 722 | { |
| 723 | pos($$textref)--; # OVERCOME BROKEN LOOKAHEAD |
| 724 | _match_bracketed($textref,"",$ldel2,"","",$rdel2) |
| 725 | || do { pos $$textref = $startpos; return }; |
| 726 | } |
| 727 | else |
| 728 | { |
| 729 | $$textref =~ /[^\\$ldel2]*(\\.[^\\$ldel2]*)*$ldel2/gc |
| 730 | || do { pos $$textref = $startpos; return }; |
| 731 | } |
| 732 | $rd2pos = pos($$textref)-1; |
| 733 | } |
| 734 | else |
| 735 | { |
| 736 | $ld2pos = $str2pos = $rd2pos = $rd1pos; |
| 737 | } |
| 738 | |
| 739 | $modpos = pos $$textref; |
| 740 | |
| 741 | $$textref =~ m/\G($mods{$op})/gc; |
| 742 | my $endpos = pos $$textref; |
| 743 | |
| 744 | return ( |
| 745 | $startpos, $startpos-$oppos, # PREFIX |
| 746 | $oppos, length($op), # OPERATOR |
| 747 | $ld1pos, 1, # LEFT DEL |
| 748 | $str1pos, $rd1pos-$str1pos, # STR/PAT |
| 749 | $rd1pos, 1, # RIGHT DEL |
| 750 | $ld2pos, $second_arg, # 2ND LDEL (MAYBE) |
| 751 | $str2pos, $rd2pos-$str2pos, # 2ND STR (MAYBE) |
| 752 | $rd2pos, $second_arg, # 2ND RDEL (MAYBE) |
| 753 | $modpos, $endpos-$modpos, # MODIFIERS |
| 754 | $endpos, $textlen-$endpos, # REMAINDER |
| 755 | ); |
| 756 | } |
| 757 | |
| 758 | my $def_func = |
| 759 | [ |
| 760 | sub { extract_variable($_[0], '') }, |
| 761 | sub { extract_quotelike($_[0],'') }, |
| 762 | sub { extract_codeblock($_[0],'{}','') }, |
| 763 | ]; |
| 764 | |
| 765 | sub extract_multiple (;$$$$) # ($text, $functions_ref, $max_fields, $ignoreunknown) |
| 766 | { |
| 767 | my $textref = defined($_[0]) ? \$_[0] : \$_; |
| 768 | my $posbug = pos; |
| 769 | my ($lastpos, $firstpos); |
| 770 | my @fields = (); |
| 771 | |
| 772 | for ($$textref) |
| 773 | { |
| 774 | my @func = defined $_[1] ? @{$_[1]} : @{$def_func}; |
| 775 | my $max = defined $_[2] && $_[2]>0 ? $_[2] : 1_000_000_000; |
| 776 | my $igunk = $_[3]; |
| 777 | |
| 778 | pos ||= 0; |
| 779 | |
| 780 | unless (wantarray) |
| 781 | { |
| 782 | use Carp; |
| 783 | carp "extract_multiple reset maximal count to 1 in scalar context" |
| 784 | if $^W && defined($_[2]) && $max > 1; |
| 785 | $max = 1 |
| 786 | } |
| 787 | |
| 788 | my $unkpos; |
| 789 | my $func; |
| 790 | my $class; |
| 791 | |
| 792 | my @class; |
| 793 | foreach $func ( @func ) |
| 794 | { |
| 795 | if (ref($func) eq 'HASH') |
| 796 | { |
| 797 | push @class, (keys %$func)[0]; |
| 798 | $func = (values %$func)[0]; |
| 799 | } |
| 800 | else |
| 801 | { |
| 802 | push @class, undef; |
| 803 | } |
| 804 | } |
| 805 | |
| 806 | FIELD: while (pos() < length()) |
| 807 | { |
| 808 | my $field; |
| 809 | foreach my $i ( 0..$#func ) |
| 810 | { |
| 811 | $func = $func[$i]; |
| 812 | $class = $class[$i]; |
| 813 | $lastpos = pos; |
| 814 | if (ref($func) eq 'CODE') |
| 815 | { ($field) = $func->($_) } |
| 816 | elsif( m/\G$func/gc ) |
| 817 | { $field = defined($1) ? $1 : $& } |
| 818 | |
| 819 | if (defined($field) && length($field)) |
| 820 | { |
| 821 | if (defined($unkpos) && !$igunk) |
| 822 | { |
| 823 | push @fields, substr($_, $unkpos, $lastpos-$unkpos); |
| 824 | $firstpos = $unkpos unless defined $firstpos; |
| 825 | undef $unkpos; |
| 826 | last FIELD if @fields == $max; |
| 827 | } |
| 828 | push @fields, $class |
| 829 | ? bless(\$field, $class) |
| 830 | : $field; |
| 831 | $firstpos = $lastpos unless defined $firstpos; |
| 832 | $lastpos = pos; |
| 833 | last FIELD if @fields == $max; |
| 834 | next FIELD; |
| 835 | } |
| 836 | } |
| 837 | if (/\G(.)/gcs) |
| 838 | { |
| 839 | $unkpos = pos()-1 |
| 840 | unless $igunk || defined $unkpos; |
| 841 | } |
| 842 | } |
| 843 | |
| 844 | if (defined $unkpos) |
| 845 | { |
| 846 | push @fields, substr($_, $unkpos); |
| 847 | $firstpos = $unkpos unless defined $firstpos; |
| 848 | $lastpos = length; |
| 849 | } |
| 850 | last; |
| 851 | } |
| 852 | |
| 853 | pos $$textref = $lastpos; |
| 854 | return @fields if wantarray; |
| 855 | |
| 856 | $firstpos ||= 0; |
| 857 | eval { substr($$textref,$firstpos,$lastpos-$firstpos)=""; |
| 858 | pos $$textref = $firstpos }; |
| 859 | return $fields[0]; |
| 860 | } |
| 861 | |
| 862 | |
| 863 | sub gen_extract_tagged # ($opentag, $closetag, $pre, \%options) |
| 864 | { |
| 865 | my $ldel = $_[0]; |
| 866 | my $rdel = $_[1]; |
| 867 | my $pre = defined $_[2] ? $_[2] : '\s*'; |
| 868 | my %options = defined $_[3] ? %{$_[3]} : (); |
| 869 | my $omode = defined $options{fail} ? $options{fail} : ''; |
| 870 | my $bad = ref($options{reject}) eq 'ARRAY' ? join('|', @{$options{reject}}) |
| 871 | : defined($options{reject}) ? $options{reject} |
| 872 | : '' |
| 873 | ; |
| 874 | my $ignore = ref($options{ignore}) eq 'ARRAY' ? join('|', @{$options{ignore}}) |
| 875 | : defined($options{ignore}) ? $options{ignore} |
| 876 | : '' |
| 877 | ; |
| 878 | |
| 879 | if (!defined $ldel) { $ldel = '<\w+(?:' . gen_delimited_pat(q{'"}) . '|[^>])*>'; } |
| 880 | |
| 881 | my $posbug = pos; |
| 882 | for ($ldel, $pre, $bad, $ignore) { $_ = qr/$_/ if $_ } |
| 883 | pos = $posbug; |
| 884 | |
| 885 | my $closure = sub |
| 886 | { |
| 887 | my $textref = defined $_[0] ? \$_[0] : \$_; |
| 888 | my @match = Text::Balanced::_match_tagged($textref, $pre, $ldel, $rdel, $omode, $bad, $ignore); |
| 889 | |
| 890 | return _fail(wantarray, $textref) unless @match; |
| 891 | return _succeed wantarray, $textref, |
| 892 | $match[2], $match[3]+$match[5]+$match[7], # MATCH |
| 893 | @match[8..9,0..1,2..7]; # REM, PRE, BITS |
| 894 | }; |
| 895 | |
| 896 | bless $closure, 'Text::Balanced::Extractor'; |
| 897 | } |
| 898 | |
| 899 | package Text::Balanced::Extractor; |
| 900 | |
| 901 | sub extract($$) # ($self, $text) |
| 902 | { |
| 903 | &{$_[0]}($_[1]); |
| 904 | } |
| 905 | |
| 906 | 1; |