- regpiece - something followed by possible [*+?] * * Note that the branching code sequences used for ? and the general cases * of * and + are somewhat optimized: they use the same NOTHING node as * both the endmarker for their branch list and the body of the last branch. * It might seem that this node could be dispensed with entirely, but the * endmarker role is not redundant. */
| 415 | * endmarker role is not redundant. |
| 416 | */ |
| 417 | static char * |
| 418 | regpiece( int32_t *flagp ) |
| 419 | { |
| 420 | char *ret; |
| 421 | char op; |
| 422 | char *next; |
| 423 | int32_t flags; |
| 424 | |
| 425 | ret = regatom(&flags); |
| 426 | if (ret == NULL) |
| 427 | return(NULL); |
| 428 | |
| 429 | op = *regparse; |
| 430 | if (!ISMULT(op)) { |
| 431 | *flagp = flags; |
| 432 | return(ret); |
| 433 | } |
| 434 | |
| 435 | if (!(flags&HASWIDTH) && op != '?') |
| 436 | FAIL("*+ operand could be empty"); |
| 437 | *flagp = (op != '+') ? (WORST|SPSTART) : (WORST|HASWIDTH); |
| 438 | |
| 439 | if (op == '*' && (flags&SIMPLE)) |
| 440 | reginsert(STAR, ret); |
| 441 | else if (op == '*') { |
| 442 | /* Emit x* as (x&|), where & means "self". */ |
| 443 | reginsert(BRANCH, ret); /* Either x */ |
| 444 | regoptail(ret, regnode(BACK)); /* and loop */ |
| 445 | regoptail(ret, ret); /* back */ |
| 446 | regtail(ret, regnode(BRANCH)); /* or */ |
| 447 | regtail(ret, regnode(NOTHING)); /* null. */ |
| 448 | } else if (op == '+' && (flags&SIMPLE)) |
| 449 | reginsert(PLUS, ret); |
| 450 | else if (op == '+') { |
| 451 | /* Emit x+ as x(&|), where & means "self". */ |
| 452 | next = regnode(BRANCH); /* Either */ |
| 453 | regtail(ret, next); |
| 454 | regtail(regnode(BACK), ret); /* loop back */ |
| 455 | regtail(next, regnode(BRANCH)); /* or */ |
| 456 | regtail(ret, regnode(NOTHING)); /* null. */ |
| 457 | } else if (op == '?') { |
| 458 | /* Emit x? as (x|) */ |
| 459 | reginsert(BRANCH, ret); /* Either x */ |
| 460 | regtail(ret, regnode(BRANCH)); /* or */ |
| 461 | next = regnode(NOTHING); /* null. */ |
| 462 | regtail(ret, next); |
| 463 | regoptail(ret, next); |
| 464 | } |
| 465 | regparse++; |
| 466 | if (ISMULT(*regparse)) |
| 467 | FAIL("nested *?+"); |
| 468 | |
| 469 | return(ret); |
| 470 | } |
| 471 | |
| 472 | /* |
| 473 | - regatom - the lowest level |