| 1 | # Test out Oil's regular expression syntax. |
| 2 | |
| 3 | #### /^.$/ |
| 4 | shopt -s oil:all |
| 5 | var pat = '' |
| 6 | |
| 7 | setvar pat = /^.$/ |
| 8 | echo pat=$pat |
| 9 | |
| 10 | setvar pat = /%start dot %end/ |
| 11 | echo pat=$pat |
| 12 | |
| 13 | if ('' ~ pat) { # ERE syntax |
| 14 | echo yes |
| 15 | } else { |
| 16 | echo no |
| 17 | } |
| 18 | # $pat is same as pat |
| 19 | if ('f' ~ pat) { # ERE syntax |
| 20 | echo yes |
| 21 | } else { |
| 22 | echo no |
| 23 | } |
| 24 | |
| 25 | ## STDOUT: |
| 26 | pat=^.$ |
| 27 | pat=^.$ |
| 28 | no |
| 29 | yes |
| 30 | ## END |
| 31 | |
| 32 | |
| 33 | #### /.+/ |
| 34 | shopt -s oil:all |
| 35 | |
| 36 | var pat = /.+/ |
| 37 | echo $pat |
| 38 | |
| 39 | var s = 'foo' |
| 40 | if (s ~ pat) { # ERE syntax |
| 41 | echo yes |
| 42 | } |
| 43 | var empty = '' |
| 44 | if (empty ~ pat) { echo yes } else { echo no } |
| 45 | ## STDOUT: |
| 46 | .+ |
| 47 | yes |
| 48 | no |
| 49 | ## END |
| 50 | |
| 51 | #### Positional captures with _match |
| 52 | shopt -s oil:all |
| 53 | |
| 54 | var x = 'zz 2020-08-20' |
| 55 | |
| 56 | if [[ $x =~ ([[:digit:]]+)-([[:digit:]]+) ]] { |
| 57 | argv.py "${BASH_REMATCH[@]}" |
| 58 | } |
| 59 | |
| 60 | # THIS IS A NO-OP. The variable is SHADOWED by the special name. |
| 61 | # I think that's OK. |
| 62 | setvar BASH_REMATCH = %(reset) |
| 63 | |
| 64 | if (x ~ /<d+> '-' <d+>/) { |
| 65 | argv.py "${BASH_REMATCH[@]}" |
| 66 | argv.py $[_match(0)] $[_match(1)] $[_match(2)] |
| 67 | |
| 68 | argv.py $[_match()] # synonym for _match(0) |
| 69 | |
| 70 | # TODO: Also test _start() and _end() |
| 71 | } |
| 72 | ## STDOUT: |
| 73 | ['2020-08', '2020', '08'] |
| 74 | ['2020-08', '2020', '08'] |
| 75 | ['2020-08', '2020', '08'] |
| 76 | ['2020-08'] |
| 77 | ## END |
| 78 | |
| 79 | #### s ~ regex and s !~ regex |
| 80 | shopt -s oil:upgrade |
| 81 | |
| 82 | var s = 'foo' |
| 83 | if (s ~ '.([[:alpha:]]+)') { # ERE syntax |
| 84 | echo matches |
| 85 | argv.py $[_match(0)] $[_match(1)] |
| 86 | } |
| 87 | if (s !~ '[[:digit:]]+') { |
| 88 | echo "does not match" |
| 89 | argv.py $[_match(0)] $[_match(1)] |
| 90 | } |
| 91 | |
| 92 | if (s ~ '[[:digit:]]+') { |
| 93 | echo "matches" |
| 94 | } |
| 95 | # Should be cleared now |
| 96 | # should this be Undef rather than ''? |
| 97 | var x = _match(0) |
| 98 | var y = _match(1) |
| 99 | if (x === null and y === null) { |
| 100 | echo 'cleared' |
| 101 | } |
| 102 | |
| 103 | ## STDOUT: |
| 104 | matches |
| 105 | ['foo', 'oo'] |
| 106 | does not match |
| 107 | ['foo', 'oo'] |
| 108 | cleared |
| 109 | ## END |
| 110 | |
| 111 | #### _start() and _end() |
| 112 | shopt -s oil:upgrade |
| 113 | |
| 114 | var s = 'foo123bar' |
| 115 | if (s ~ /digit+/) { |
| 116 | echo start=$[_start()] end=$[_end()] |
| 117 | } |
| 118 | |
| 119 | if (s ~ / word+ <digit+> /) { |
| 120 | echo start=$[_start(1)] end=$[_end(1)] |
| 121 | } |
| 122 | ## STDOUT: |
| 123 | start=3 end=6 |
| 124 | start=3 end=6 |
| 125 | ## END |
| 126 | |
| 127 | #### Repeat {1,3} etc. |
| 128 | var pat = null |
| 129 | |
| 130 | setvar pat = /d{2}/ |
| 131 | echo $pat |
| 132 | setvar pat = /d{1,3}/ |
| 133 | echo $pat |
| 134 | setvar pat = /d{1,}/ |
| 135 | echo $pat |
| 136 | setvar pat = /d{,3}/ |
| 137 | echo $pat |
| 138 | |
| 139 | |
| 140 | ## STDOUT: |
| 141 | [[:digit:]]{2} |
| 142 | [[:digit:]]{1,3} |
| 143 | [[:digit:]]{1,} |
| 144 | [[:digit:]]{,3} |
| 145 | ## END |
| 146 | |
| 147 | |
| 148 | #### d+ digit+ !d+ !digit+ |
| 149 | shopt -s oil:all |
| 150 | |
| 151 | var pat = '' |
| 152 | |
| 153 | setvar pat = /d+/ |
| 154 | echo $pat |
| 155 | if ('42' ~ pat) { echo yes } |
| 156 | |
| 157 | var empty = '' |
| 158 | if (empty ~ pat) { echo yes } else { echo no } |
| 159 | |
| 160 | setvar pat = /digit+/ |
| 161 | echo $pat |
| 162 | setvar pat = /!d+/ |
| 163 | echo $pat |
| 164 | setvar pat = /!digit+/ |
| 165 | echo $pat |
| 166 | |
| 167 | |
| 168 | ## STDOUT: |
| 169 | [[:digit:]]+ |
| 170 | yes |
| 171 | no |
| 172 | [[:digit:]]+ |
| 173 | [^[:digit:]]+ |
| 174 | [^[:digit:]]+ |
| 175 | ## END |
| 176 | |
| 177 | #### Alternation and sequence |
| 178 | var pat = '' |
| 179 | setvar pat = /s d+ | w*/ |
| 180 | echo $pat |
| 181 | setvar pat = /s d+ or w*/ |
| 182 | echo $pat |
| 183 | ## STDOUT: |
| 184 | [[:space:]][[:digit:]]+|[[:alpha:][:digit:]_]* |
| 185 | [[:space:]][[:digit:]]+|[[:alpha:][:digit:]_]* |
| 186 | ## END |
| 187 | |
| 188 | #### Char Class Ranges |
| 189 | shopt -s oil:all |
| 190 | |
| 191 | var pat = '' |
| 192 | setvar pat = /[0-9 a-f]+/ |
| 193 | echo $pat |
| 194 | # This is equivalent |
| 195 | setvar pat = /['0' - '9' 'a' - 'f']+/ |
| 196 | echo $pat |
| 197 | |
| 198 | if ('0123' ~ pat) { echo yes } else { echo no } |
| 199 | if ('zzz' ~ pat) { echo yes } else { echo no } |
| 200 | if ('' ~ pat) { echo yes } else { echo no } |
| 201 | ## STDOUT: |
| 202 | [0-9a-f]+ |
| 203 | [0-9a-f]+ |
| 204 | yes |
| 205 | no |
| 206 | no |
| 207 | ## END |
| 208 | |
| 209 | #### Char Class Set |
| 210 | shopt -s oil:all |
| 211 | var pat = '' |
| 212 | |
| 213 | # This is NOT allowed |
| 214 | # setvar pat = /[a b c]+/ |
| 215 | |
| 216 | setvar pat = /['abc']+/ |
| 217 | echo $pat |
| 218 | |
| 219 | if ('cbcb' ~ pat) { echo yes } else { echo no } |
| 220 | if ('0123' ~ pat) { echo yes } else { echo no } |
| 221 | if ('' ~ pat) { echo yes } else { echo no } |
| 222 | ## STDOUT: |
| 223 | [abc]+ |
| 224 | yes |
| 225 | no |
| 226 | no |
| 227 | ## END |
| 228 | |
| 229 | #### Range with escaped characters |
| 230 | shopt -s oil:all |
| 231 | |
| 232 | var pat = null |
| 233 | |
| 234 | setvar pat = / [ \x00 - \x0f ] / |
| 235 | echo $pat | od -A n -t x1 |
| 236 | |
| 237 | ## STDOUT: |
| 238 | 5b 00 2d 0f 5d 0a |
| 239 | ## END |
| 240 | |
| 241 | |
| 242 | #### Group () |
| 243 | shopt -s oil:all |
| 244 | var pat = '' |
| 245 | |
| 246 | setvar pat = /(%start s or d d)/ |
| 247 | echo $pat |
| 248 | |
| 249 | if (' foo' ~ pat) { echo yes } else { echo no } |
| 250 | if ('-00-' ~ pat) { echo yes } else { echo no } |
| 251 | if ('foo' ~ pat) { echo yes } else { echo no } |
| 252 | |
| 253 | ## STDOUT: |
| 254 | (^[[:space:]]|[[:digit:]][[:digit:]]) |
| 255 | yes |
| 256 | yes |
| 257 | no |
| 258 | ## END |
| 259 | |
| 260 | #### Capture is acceptable as a group |
| 261 | shopt -s oil:all |
| 262 | var pat = /<%start s | d d>/ |
| 263 | echo $pat |
| 264 | ## STDOUT: |
| 265 | (^[[:space:]]|[[:digit:]][[:digit:]]) |
| 266 | ## END |
| 267 | |
| 268 | #### Named captures with _match |
| 269 | shopt -s oil:all |
| 270 | |
| 271 | var x = 'zz 2020-08-20' |
| 272 | |
| 273 | if (x ~ /<d+ : year> '-' <d+ : month>/) { |
| 274 | argv.py $[_match('year')] $[_match('month')] |
| 275 | } |
| 276 | ## STDOUT: |
| 277 | ['2020', '08'] |
| 278 | ## END |
| 279 | |
| 280 | #### Named Capture Decays Without Name |
| 281 | shopt -s oil:all |
| 282 | var pat = /<d+ : month>/ |
| 283 | echo $pat |
| 284 | |
| 285 | if ('123' ~ pat) { |
| 286 | echo yes |
| 287 | } |
| 288 | |
| 289 | ## STDOUT: |
| 290 | ([[:digit:]]+) |
| 291 | yes |
| 292 | ## END |
| 293 | |
| 294 | #### Operator ~ assigns named variable |
| 295 | shopt -s oil:all |
| 296 | var pat = /<d+ : month>/ |
| 297 | echo $pat |
| 298 | |
| 299 | if ('123' ~ pat) { |
| 300 | echo yes |
| 301 | = month |
| 302 | } |
| 303 | ## STDOUT: |
| 304 | ([[:digit:]]+) |
| 305 | yes |
| 306 | TODO MONTH |
| 307 | ## END |
| 308 | |
| 309 | #### literal '' |
| 310 | shopt -s oil:all |
| 311 | var pat = '' |
| 312 | |
| 313 | setvar pat = /'abc' 'def'/ |
| 314 | echo $pat |
| 315 | |
| 316 | #setvar pat = /'abc' '^ + * ?'/ |
| 317 | #echo $pat |
| 318 | |
| 319 | if ('abcde' ~ pat) { echo yes } else { echo no } |
| 320 | if ('abcdef' ~ pat) { echo yes } else { echo no } |
| 321 | |
| 322 | ## STDOUT: |
| 323 | abcdef |
| 324 | no |
| 325 | yes |
| 326 | ## END |
| 327 | |
| 328 | #### Single quotes and splicing (do what "foo $x ${x}" used to) |
| 329 | shopt -s oil:all |
| 330 | var pat = '' |
| 331 | |
| 332 | var x = 'x' |
| 333 | var y = 'y' |
| 334 | setvar pat = / @x @x 'abc' @x @y / |
| 335 | echo $pat |
| 336 | |
| 337 | if ('xxabcx' ~ pat) { echo yes } else { echo no } |
| 338 | if ('xxabcxyf' ~ pat) { echo yes } else { echo no } |
| 339 | |
| 340 | ## STDOUT: |
| 341 | xxabcxy |
| 342 | no |
| 343 | yes |
| 344 | ## END |
| 345 | |
| 346 | #### @splice |
| 347 | shopt -s oil:all |
| 348 | var d = /d+/; |
| 349 | var ip = / @d '.' @d '.' @d '.' @d / |
| 350 | echo $ip |
| 351 | if ('0.0.0.0' ~ ip) { echo yes } else { echo no } |
| 352 | if ('0.0.0' ~ ip) { echo yes } else { echo no } |
| 353 | ## STDOUT: |
| 354 | [[:digit:]]+\.[[:digit:]]+\.[[:digit:]]+\.[[:digit:]]+ |
| 355 | yes |
| 356 | no |
| 357 | ## END |
| 358 | |
| 359 | #### splice with capital letters |
| 360 | shopt -s oil:all |
| 361 | var D = /d+/; |
| 362 | var ip = / D '.' D '.' D '.' D / |
| 363 | echo $ip |
| 364 | if ('0.0.0.0' ~ ip) { echo yes } else { echo no } |
| 365 | if ('0.0.0' ~ ip) { echo yes } else { echo no } |
| 366 | ## STDOUT: |
| 367 | [[:digit:]]+\.[[:digit:]]+\.[[:digit:]]+\.[[:digit:]]+ |
| 368 | yes |
| 369 | no |
| 370 | ## END |
| 371 | |
| 372 | #### Matching escaped tab character |
| 373 | shopt -s oil:all |
| 374 | |
| 375 | var lines=%($'aa\tbb' $'cc\tdd') |
| 376 | |
| 377 | var pat = / ('a' [\t] 'b') / |
| 378 | write pat=$pat |
| 379 | write @lines | egrep $pat |
| 380 | |
| 381 | ## stdout-json: "pat=(a[\t]b)\naa\tbb\n" |
| 382 | |
| 383 | #### Match unicode char |
| 384 | shopt -s oil:all |
| 385 | var pat = / 'a' dot 'b' / |
| 386 | |
| 387 | if ('axb' ~ pat ) { echo yes } else { echo no } |
| 388 | |
| 389 | # mu character |
| 390 | if ($'a\xce\xbcb' ~ pat ) { echo yes } else { echo no } |
| 391 | |
| 392 | if ('aZZb' ~ pat ) { echo yes } else { echo no } |
| 393 | ## STDOUT: |
| 394 | yes |
| 395 | yes |
| 396 | no |
| 397 | ## END |
| 398 | |
| 399 | #### Match non-ASCII byte denoted using $'\xff' (TODO: LANG=C) |
| 400 | |
| 401 | # NOTE: This pattern doesn't work with en_US.UTF-8. I think the user should |
| 402 | # set LANG=C or shopt --unset libc_utf8. |
| 403 | |
| 404 | shopt -s oil:all |
| 405 | var pat = /[ $'\xff' ]/; |
| 406 | |
| 407 | echo $pat | od -A n -t x1 |
| 408 | if ($'\xff' ~ pat) { echo yes } else { echo no } |
| 409 | if ($'\xfe' ~ pat) { echo yes } else { echo no } |
| 410 | |
| 411 | ## STDOUT: |
| 412 | 5b ff 5d 0a |
| 413 | yes |
| 414 | no |
| 415 | ## END |
| 416 | |
| 417 | #### Match non-ASCII byte denoted using \xff |
| 418 | shopt -s oil:all |
| 419 | var pat = /[ \xff ]/; |
| 420 | |
| 421 | # Show what it translates to |
| 422 | echo $pat | od -A n -t x1 |
| 423 | |
| 424 | # TODO: This might require LANG=C to work |
| 425 | #if ($'\xff' ~ pat) { echo yes } else { echo no } |
| 426 | #if ($'\xfe' ~ pat) { echo yes } else { echo no } |
| 427 | |
| 428 | ## STDOUT: |
| 429 | 5b ff 5d 0a |
| 430 | ## END |
| 431 | |
| 432 | #### ERE can express Unicode escapes that are in the ASCII range |
| 433 | shopt -s oil:all |
| 434 | var pat = /[ \u{7f} ]/; |
| 435 | |
| 436 | echo $pat | od -A n -t x1 |
| 437 | if ($'\x7f' ~ pat) { echo yes } else { echo no } |
| 438 | if ($'\x7e' ~ pat) { echo yes } else { echo no } |
| 439 | |
| 440 | var pat2 = /[ \u{7f} ]/; |
| 441 | var pat3 = /[ \u{0007f} ]/; |
| 442 | test "$pat2" = "$pat3" && echo 'equal' |
| 443 | |
| 444 | var range = / [ \u{70} - \u{7f} ] / |
| 445 | if ($'\x70' ~ range) { echo yes } else { echo no } |
| 446 | if ($'\x69' ~ range) { echo yes } else { echo no } |
| 447 | |
| 448 | ## STDOUT: |
| 449 | 5b 7f 5d 0a |
| 450 | yes |
| 451 | no |
| 452 | equal |
| 453 | yes |
| 454 | no |
| 455 | ## END |
| 456 | |
| 457 | #### ERE can't express higher Unicode escapes |
| 458 | shopt -s oil:all |
| 459 | var pat2 = /[ \u{00} - \u{ff} ]/; |
| 460 | |
| 461 | # This causes an error |
| 462 | echo $pat2 |
| 463 | |
| 464 | # This just prints it |
| 465 | = pat2 |
| 466 | |
| 467 | var pat1 = /[ \u{ff} ]/; |
| 468 | |
| 469 | echo $pat1 | od -A n -t x1 |
| 470 | if ($'\x7f' ~ pat) { echo yes } else { echo no } |
| 471 | if ($'\x7e' ~ pat) { echo yes } else { echo no } |
| 472 | |
| 473 | ## status: 1 |
| 474 | ## stdout-json: "" |
| 475 | |
| 476 | #### non-ASCII bytes must be singleton terms, e.g. '\x7f\xff' is disallowed |
| 477 | var bytes = $'\x7f\xff' |
| 478 | var pat = / [ @bytes ] / |
| 479 | echo $pat |
| 480 | ## status: 1 |
| 481 | ## stdout-json: "" |
| 482 | |
| 483 | #### Matching escaped tab character |
| 484 | shopt -s oil:all |
| 485 | |
| 486 | # BUG: need C strings in array literal |
| 487 | var lines=%($'aa\tbb' $'cc\tdd') |
| 488 | |
| 489 | var pat = / ('a' [\t] 'b') / |
| 490 | write pat=$pat |
| 491 | write @lines | egrep $pat |
| 492 | |
| 493 | ## stdout-json: "pat=(a[\t]b)\naa\tbb\n" |
| 494 | |
| 495 | #### Repeated String Literal With Single Char |
| 496 | shopt -s oil:all |
| 497 | |
| 498 | var literal = 'f' |
| 499 | var pat = null |
| 500 | |
| 501 | setvar pat = / %start @literal+ %end / |
| 502 | echo $pat |
| 503 | setvar pat = / %start (@literal)+ %end / |
| 504 | echo $pat |
| 505 | |
| 506 | if ('fff' ~ pat) { echo yes } |
| 507 | if ('foo' !~ pat) { echo no } |
| 508 | |
| 509 | ## STDOUT: |
| 510 | ^f+$ |
| 511 | ^(f)+$ |
| 512 | yes |
| 513 | no |
| 514 | ## END |
| 515 | |
| 516 | #### Error when unparenthesized string of more than one character is repeated |
| 517 | shopt -s oil:all |
| 518 | |
| 519 | var literal = 'foo' |
| 520 | var pat = null |
| 521 | |
| 522 | setvar pat = / %start @literal+ %end / |
| 523 | echo $pat |
| 524 | setvar pat = / %start (@literal)+ %end / |
| 525 | echo $pat |
| 526 | |
| 527 | if ('foofoo' ~ pat) { echo yes } |
| 528 | if ('foof' !~ pat) { echo no } |
| 529 | |
| 530 | ## status: 1 |
| 531 | ## stdout-json: "" |
| 532 | |
| 533 | #### Instead of $'foo\\bar' use 'foo' \\ 'bar' |
| 534 | shopt -s oil:all |
| 535 | var pat = /'foo' \\ 'bar'/ |
| 536 | echo $pat |
| 537 | |
| 538 | if (r'foo\bar' ~ pat) { echo yes } |
| 539 | if (r'foo.bar' !~ pat) { echo no } |
| 540 | ## STDOUT: |
| 541 | foo\\bar |
| 542 | yes |
| 543 | no |
| 544 | ## END |
| 545 | |
| 546 | #### Negation of Character Class ![a-z] |
| 547 | shopt -s oil:all |
| 548 | |
| 549 | var pat = / ![ a-z ] / |
| 550 | echo $pat |
| 551 | |
| 552 | if ('0' ~ pat) { echo yes } |
| 553 | if ('a' !~ pat) { echo no } |
| 554 | |
| 555 | ## STDOUT: |
| 556 | [^a-z] |
| 557 | yes |
| 558 | no |
| 559 | ## END |
| 560 | |
| 561 | #### Posix and Perl class in class literals |
| 562 | shopt -s oil:all |
| 563 | |
| 564 | var pat = null |
| 565 | |
| 566 | setvar pat = / [ space 'z' ] / |
| 567 | echo $pat |
| 568 | #setvar pat = / [ ~space 'z' ] / |
| 569 | #echo $pat |
| 570 | |
| 571 | # PROBLEM: can't negate individual POSIX classes. They would have to be a Perl |
| 572 | # class to be \D or \S. |
| 573 | # [[:space:]z] negates the whole thing! |
| 574 | # [^[:space:]] |
| 575 | |
| 576 | setvar pat = / [ digit 'z' ] / |
| 577 | echo $pat |
| 578 | #setvar pat = / [ ~digit 'z' ] / |
| 579 | #echo $pat |
| 580 | |
| 581 | ## STDOUT: |
| 582 | [[:space:]z] |
| 583 | [[:digit:]z] |
| 584 | ## END |
| 585 | |
| 586 | #### [!d] can't be negated because it's a literal character |
| 587 | setvar pat = / [ !d 'z' ] / |
| 588 | echo $pat |
| 589 | ## status: 2 |
| 590 | ## stdout-json: "" |
| 591 | |
| 592 | #### [!digit] can't be negated in POSIX ERE (but yes in Perl) |
| 593 | var pat = null |
| 594 | setvar pat = / [ !digit 'z' ] / |
| 595 | echo $pat |
| 596 | ## status: 1 |
| 597 | ## stdout-json: "" |
| 598 | |
| 599 | #### Long Python Example |
| 600 | |
| 601 | # https://docs.python.org/3/reference/lexical_analysis.html#integer-literals |
| 602 | |
| 603 | # integer ::= decinteger | bininteger | octinteger | hexinteger |
| 604 | # decinteger ::= nonzerodigit (["_"] digit)* | "0"+ (["_"] "0")* |
| 605 | # bininteger ::= "0" ("b" | "B") (["_"] bindigit)+ |
| 606 | # octinteger ::= "0" ("o" | "O") (["_"] octdigit)+ |
| 607 | # hexinteger ::= "0" ("x" | "X") (["_"] hexdigit)+ |
| 608 | # nonzerodigit ::= "1"..."9" |
| 609 | # digit ::= "0"..."9" |
| 610 | # bindigit ::= "0" | "1" |
| 611 | # octdigit ::= "0"..."7" |
| 612 | # hexdigit ::= digit | "a"..."f" | "A"..."F" |
| 613 | |
| 614 | shopt -s oil:all |
| 615 | |
| 616 | const DecDigit = / [0-9] / |
| 617 | const BinDigit = / [0-1] / |
| 618 | const OctDigit = / [0-7] / |
| 619 | const HexDigit = / [0-9 a-f A-F] / # note: not splicing Digit into character class |
| 620 | |
| 621 | const DecInt = / [1-9] ('_'? DecDigit)* | '0'+ ('_'? '0')* / |
| 622 | const BinInt = / '0' [b B] ('_'? BinDigit)+ / |
| 623 | const OctInt = / '0' [o O] ('_'? OctDigit)+ / |
| 624 | const HexInt = / '0' [x X] ('_'? HexDigit)+ / |
| 625 | |
| 626 | const Integer = / %start (DecInt | BinInt | OctInt | HexInt) %end / |
| 627 | |
| 628 | #echo $Integer |
| 629 | |
| 630 | if ( '123' ~ Integer) { echo 'Y' } |
| 631 | if ( 'zzz' !~ Integer) { echo 'N' } |
| 632 | |
| 633 | if ('123_000' ~ Integer) { echo 'Y decimal' } |
| 634 | if ('000_123' !~ Integer) { echo 'N decimal' } |
| 635 | |
| 636 | if ( '0b100' ~ Integer) { echo 'Y binary' } |
| 637 | if ( '0b102' !~ Integer) { echo 'N binary' } |
| 638 | |
| 639 | if ( '0o755' ~ Integer) { echo 'Y octal' } |
| 640 | if ( '0o778' !~ Integer) { echo 'N octal' } |
| 641 | |
| 642 | if ( '0xFF' ~ Integer) { echo 'Y hex' } |
| 643 | if ( '0xFG' !~ Integer) { echo 'N hex' } |
| 644 | |
| 645 | ## STDOUT: |
| 646 | Y |
| 647 | N |
| 648 | Y decimal |
| 649 | N decimal |
| 650 | Y binary |
| 651 | N binary |
| 652 | Y octal |
| 653 | N octal |
| 654 | Y hex |
| 655 | N hex |
| 656 | ## END |
| 657 | |
| 658 | #### Invalid sh operation on eggex |
| 659 | var pat = / d+ / |
| 660 | #pat[invalid]=1 |
| 661 | pat[invalid]+=1 |
| 662 | ## status: 1 |
| 663 | ## stdout-json: "" |
| 664 | |
| 665 | |
| 666 | #### Regex in a loop (bug regression) |
| 667 | |
| 668 | shopt --set oil:all |
| 669 | |
| 670 | var content = [ 1, 2 ] |
| 671 | var i = 0 |
| 672 | while (i < len(content)) { |
| 673 | var line = content[i] |
| 674 | write $[content[i]] |
| 675 | if (Str(line) ~ / s* 'imports' s* '=' s* .* /) { |
| 676 | exit |
| 677 | } |
| 678 | setvar i += 1 |
| 679 | } |
| 680 | |
| 681 | ## STDOUT: |
| 682 | 1 |
| 683 | 2 |
| 684 | ## END |
| 685 | |
| 686 | |
| 687 | #### Regex in a loop depending on var |
| 688 | |
| 689 | shopt --set oil:all |
| 690 | |
| 691 | var lines = ['foo', 'bar'] |
| 692 | for line in (lines) { |
| 693 | write "line $line" |
| 694 | |
| 695 | # = / $line / |
| 696 | |
| 697 | if ("x$line" ~ / dot @line /) { |
| 698 | #if (line ~ / $line /) { |
| 699 | write "matched $line" |
| 700 | } |
| 701 | } |
| 702 | |
| 703 | ## STDOUT: |
| 704 | line foo |
| 705 | matched foo |
| 706 | line bar |
| 707 | matched bar |
| 708 | ## END |
| 709 | |
| 710 | |
| 711 | #### Regex with [ (bug regression) |
| 712 | shopt --set oil:all |
| 713 | |
| 714 | if ('[' ~ / '[' /) { |
| 715 | echo 'sq' |
| 716 | } |
| 717 | |
| 718 | if ('[' ~ / [ '[' ] /) { |
| 719 | echo 'char class' |
| 720 | } |
| 721 | |
| 722 | # User-reported string |
| 723 | if ("a" ~ / s* 'imports' s* '=' s* '[' /) { |
| 724 | echo "yes" |
| 725 | } |
| 726 | |
| 727 | ## STDOUT: |
| 728 | sq |
| 729 | char class |
| 730 | ## END |
| 731 | |
| 732 | #### Operator chars in char classes (bash-like) |
| 733 | |
| 734 | pat='[-]' |
| 735 | [[ '-' =~ $pat ]] && echo hyphen |
| 736 | [[ '\' =~ $pat ]] && echo FAIL |
| 737 | |
| 738 | pat='[\]' |
| 739 | [[ '\' =~ $pat ]] && echo backslash |
| 740 | [[ '-' =~ $pat ]] && echo FAIL |
| 741 | |
| 742 | pat='[]]' |
| 743 | [[ ']' =~ $pat ]] && echo 'right bracket' |
| 744 | [[ '[' =~ $pat ]] && echo FAIL |
| 745 | |
| 746 | pat='[[]' |
| 747 | [[ '[' =~ $pat ]] && echo 'left bracket' |
| 748 | [[ ']' =~ $pat ]] && echo FAIL |
| 749 | |
| 750 | pat='[.]' |
| 751 | [[ '.' =~ $pat ]] && echo period |
| 752 | [[ '\' =~ $pat ]] && echo FAIL |
| 753 | |
| 754 | pat='[\^]' |
| 755 | [[ '^' =~ $pat ]] && echo caret |
| 756 | [[ '\' =~ $pat ]] && echo 'no way to have [^]' |
| 757 | |
| 758 | ## STDOUT: |
| 759 | hyphen |
| 760 | backslash |
| 761 | right bracket |
| 762 | left bracket |
| 763 | period |
| 764 | caret |
| 765 | no way to have [^] |
| 766 | ## END |
| 767 | |
| 768 | #### Operator chars in char classes (eggex) |
| 769 | shopt --set oil:upgrade |
| 770 | |
| 771 | var pat = / ['-'] / |
| 772 | #echo PAT=$pat |
| 773 | if ('-' ~ pat) { echo hyphen } |
| 774 | if ($'\\' ~ pat) { echo FAIL } |
| 775 | |
| 776 | var pat = / [ \\ ] / |
| 777 | [[ '\' =~ $pat ]] && echo backslash |
| 778 | [[ '-' =~ $pat ]] && echo FAIL |
| 779 | |
| 780 | var pat = / [ ']' ] / |
| 781 | [[ ']' =~ $pat ]] && echo 'right bracket' |
| 782 | [[ '[' =~ $pat ]] && echo FAIL |
| 783 | |
| 784 | var pat = / [ '[' ] / |
| 785 | [[ '[' =~ $pat ]] && echo 'left bracket' |
| 786 | [[ ']' =~ $pat ]] && echo FAIL |
| 787 | |
| 788 | var pat = / [ '.' ] / |
| 789 | [[ '.' =~ $pat ]] && echo period |
| 790 | [[ '\' =~ $pat ]] && echo FAIL |
| 791 | |
| 792 | var pat = / [ \\ '^' ] / |
| 793 | [[ '^' =~ $pat ]] && echo caret |
| 794 | [[ '\' =~ $pat ]] && echo 'no way to have [^]' |
| 795 | |
| 796 | |
| 797 | ## STDOUT: |
| 798 | hyphen |
| 799 | backslash |
| 800 | right bracket |
| 801 | left bracket |
| 802 | period |
| 803 | caret |
| 804 | no way to have [^] |
| 805 | ## END |
| 806 | |
| 807 | #### Matching ] and \ and ' and " in character classes |
| 808 | shopt -s oil:all |
| 809 | |
| 810 | # BUG: need C strings in array literal |
| 811 | var lines=%( |
| 812 | 'backslash \' |
| 813 | 'rbracket ]' |
| 814 | 'lbracket [' |
| 815 | "sq '" |
| 816 | 'dq "' |
| 817 | ) |
| 818 | |
| 819 | # Weird GNU quirk: ] has to come first! |
| 820 | # []abc] works. But [abc\]] does NOT work. Stupid rule! |
| 821 | |
| 822 | var pat = / [ ']' \\ \' \" ] / |
| 823 | write pat=$pat |
| 824 | write @lines | egrep $pat |
| 825 | |
| 826 | ## STDOUT: |
| 827 | pat=[]'"\\] |
| 828 | backslash \ |
| 829 | rbracket ] |
| 830 | sq ' |
| 831 | dq " |
| 832 | ## END |
| 833 | |
| 834 | #### Matching literal hyphen in character classes |
| 835 | shopt -s oil:all |
| 836 | |
| 837 | var literal = '-' |
| 838 | var pat = / [ 'a' 'b' @literal ] / |
| 839 | write pat=$pat |
| 840 | write 'c-d' 'ab' 'cd' | grep $pat |
| 841 | ## STDOUT: |
| 842 | pat=[ab-] |
| 843 | c-d |
| 844 | ab |
| 845 | ## END |
| 846 | |
| 847 | #### Char class special: ^ - ] \ |
| 848 | |
| 849 | # See demo/ere-char-class-literals.sh |
| 850 | # |
| 851 | # \ is special because of gawk |
| 852 | |
| 853 | shopt -s oil:upgrade |
| 854 | |
| 855 | |
| 856 | # Note: single caret disalowed |
| 857 | var caret = / ['^' 'x'] / |
| 858 | echo caret=$caret |
| 859 | |
| 860 | var caret2 = / [ \x5e 'x'] / |
| 861 | echo caret2=$caret2 |
| 862 | |
| 863 | var caret3 = / [ \u{5e} 'x'] / |
| 864 | echo caret3=$caret3 |
| 865 | |
| 866 | if ('x' ~ caret3) { |
| 867 | echo 'match x' |
| 868 | } |
| 869 | if ('^' ~ caret3) { |
| 870 | echo 'match ^' |
| 871 | } |
| 872 | |
| 873 | echo --- |
| 874 | |
| 875 | var hyphen = / ['a' '-' 'b'] / |
| 876 | echo hyphen=$hyphen |
| 877 | |
| 878 | var hyphen2 = / ['a' \x2d 'b' ] / |
| 879 | echo hyphen2=$hyphen2 |
| 880 | |
| 881 | if ('-' ~ hyphen2) { |
| 882 | echo 'match -' |
| 883 | } |
| 884 | |
| 885 | if ('a' ~ hyphen2) { |
| 886 | echo 'match a' |
| 887 | } |
| 888 | |
| 889 | if ('c' ~ hyphen2) { |
| 890 | echo 'match c' |
| 891 | } |
| 892 | |
| 893 | echo --- |
| 894 | |
| 895 | var rbracket = / [ '[' ']' ] / |
| 896 | echo rbracket=$rbracket |
| 897 | |
| 898 | var rbracket2 = / [ \x5b \x5d ] / |
| 899 | echo rbracket2=$rbracket2 |
| 900 | |
| 901 | if ('[' ~ rbracket2) { |
| 902 | echo 'match [' |
| 903 | } |
| 904 | |
| 905 | if (']' ~ rbracket2) { |
| 906 | echo 'match ]' |
| 907 | } |
| 908 | |
| 909 | echo --- |
| 910 | |
| 911 | var backslash = / [ 'x' \\ 'n' ] / |
| 912 | echo backslash=$backslash |
| 913 | |
| 914 | var backslash2 = / [ 'x' \x5c 'n' ] / |
| 915 | echo backslash2=$backslash2 |
| 916 | |
| 917 | var backslash3 = / [ 'x' $'\\' 'n' ] / |
| 918 | echo backslash3=$backslash3 |
| 919 | |
| 920 | if ('x' ~ backslash3) { |
| 921 | echo 'match x' |
| 922 | } |
| 923 | |
| 924 | if ('n' ~ backslash3) { |
| 925 | echo 'match n' |
| 926 | } |
| 927 | |
| 928 | if ($'\\' ~ backslash3) { |
| 929 | echo 'match backslash' |
| 930 | } |
| 931 | |
| 932 | if ($'\n' ~ backslash3) { |
| 933 | echo 'match nnewline' |
| 934 | } |
| 935 | |
| 936 | |
| 937 | ## STDOUT: |
| 938 | caret=[x^] |
| 939 | caret2=[x^] |
| 940 | caret3=[x^] |
| 941 | match x |
| 942 | match ^ |
| 943 | --- |
| 944 | hyphen=[ab-] |
| 945 | hyphen2=[ab-] |
| 946 | match - |
| 947 | match a |
| 948 | --- |
| 949 | rbracket=[][] |
| 950 | rbracket2=[][] |
| 951 | match [ |
| 952 | match ] |
| 953 | --- |
| 954 | backslash=[xn\\] |
| 955 | backslash2=[xn\\] |
| 956 | backslash3=[xn\\] |
| 957 | match x |
| 958 | match n |
| 959 | match backslash |
| 960 | ## END |
| 961 |