+/* regexp.h
+ *
+ * Copyright (C) 1993, 1994, 1996, 1997, 1999, 2000, 2001, 2003,
+ * by Larry Wall and others
+ *
+ * You may distribute under the terms of either the GNU General Public
+ * License or the Artistic License, as specified in the README file.
+ *
+ */
+
/*
* Definitions etc. for regexp(3) routines.
*
* not the System V one.
*/
-/* $RCSfile: regexp.h,v $$Revision: 4.0.1.2 $$Date: 91/11/05 18:24:31 $
- *
- * $Log: regexp.h,v $
- * Revision 4.0.1.2 91/11/05 18:24:31 lwall
- * patch11: minimum match length calculation in regexp is now cumulative
- * patch11: initial .* in pattern had dependency on value of $*
- *
- * Revision 4.0.1.1 91/06/07 11:51:18 lwall
- * patch4: new copyright notice
- * patch4: // wouldn't use previous pattern if it started with a null character
- * patch4: $` was busted inside s///
- *
- * Revision 4.0 91/03/20 01:39:23 lwall
- * 4.0 baseline.
- *
- */
+
+struct regnode {
+ U8 flags;
+ U8 type;
+ U16 next_off;
+};
+
+typedef struct regnode regnode;
+
+struct reg_substr_data;
+
+struct reg_data;
typedef struct regexp {
- char **startp;
- char **endp;
- STR *regstart; /* Internal use only. */
- char *regstclass;
- STR *regmust; /* Internal use only. */
- int regback; /* Can regmust locate first try? */
- int minlen; /* mininum possible length of $& */
- int prelen; /* length of precomp */
+ I32 *startp;
+ I32 *endp;
+ regnode *regstclass;
+ struct reg_substr_data *substrs;
char *precomp; /* pre-compilation regular expression */
- char *subbase; /* saved string so \digit works forever */
- char *subbeg; /* same, but not responsible for allocation */
- char *subend; /* end of subbase */
- char reganch; /* Internal use only. */
- char do_folding; /* do case-insensitive match? */
- char lastparen; /* last paren matched */
- char nparens; /* number of parentheses */
- char program[1]; /* Unwarranted chumminess with compiler. */
+ struct reg_data *data; /* Additional data. */
+ char *subbeg; /* saved or original string
+ so \digit works forever. */
+#ifdef PERL_COPY_ON_WRITE
+ SV *saved_copy; /* If non-NULL, SV which is COW from original */
+#endif
+ U32 *offsets; /* offset annotations 20001228 MJD */
+ I32 sublen; /* Length of string pointed by subbeg */
+ I32 refcnt;
+ I32 minlen; /* mininum possible length of $& */
+ I32 prelen; /* length of precomp */
+ U32 nparens; /* number of parentheses */
+ U32 lastparen; /* last paren matched */
+ U32 lastcloseparen; /* last paren matched */
+ U32 reganch; /* Internal use only +
+ Tainted information used by regexec? */
+ regnode program[1]; /* Unwarranted chumminess with compiler. */
} regexp;
-#define ROPT_ANCH 1
-#define ROPT_SKIP 2
-#define ROPT_IMPLICIT 4
+#define ROPT_ANCH (ROPT_ANCH_BOL|ROPT_ANCH_MBOL|ROPT_ANCH_GPOS|ROPT_ANCH_SBOL)
+#define ROPT_ANCH_SINGLE (ROPT_ANCH_SBOL|ROPT_ANCH_GPOS)
+#define ROPT_ANCH_BOL 0x00001
+#define ROPT_ANCH_MBOL 0x00002
+#define ROPT_ANCH_SBOL 0x00004
+#define ROPT_ANCH_GPOS 0x00008
+#define ROPT_SKIP 0x00010
+#define ROPT_IMPLICIT 0x00020 /* Converted .* to ^.* */
+#define ROPT_NOSCAN 0x00040 /* Check-string always at start. */
+#define ROPT_GPOS_SEEN 0x00080
+#define ROPT_CHECK_ALL 0x00100
+#define ROPT_LOOKBEHIND_SEEN 0x00200
+#define ROPT_EVAL_SEEN 0x00400
+#define ROPT_CANY_SEEN 0x00800
+#define ROPT_SANY_SEEN ROPT_CANY_SEEN /* src bckwrd cmpt */
+
+/* 0xf800 of reganch is used by PMf_COMPILETIME */
+
+#define ROPT_UTF8 0x10000
+#define ROPT_NAUGHTY 0x20000 /* how exponential is this pattern? */
+#define ROPT_COPY_DONE 0x40000 /* subbeg is a copy of the string */
+#define ROPT_TAINTED_SEEN 0x80000
+#define ROPT_MATCH_UTF8 0x10000000 /* subbeg is utf-8 */
+
+#define RE_USE_INTUIT_NOML 0x0100000 /* Best to intuit before matching */
+#define RE_USE_INTUIT_ML 0x0200000
+#define REINT_AUTORITATIVE_NOML 0x0400000 /* Can trust a positive answer */
+#define REINT_AUTORITATIVE_ML 0x0800000
+#define REINT_ONCE_NOML 0x1000000 /* Intuit can succed once only. */
+#define REINT_ONCE_ML 0x2000000
+#define RE_INTUIT_ONECHAR 0x4000000
+#define RE_INTUIT_TAIL 0x8000000
+
+#define RE_USE_INTUIT (RE_USE_INTUIT_NOML|RE_USE_INTUIT_ML)
+#define REINT_AUTORITATIVE (REINT_AUTORITATIVE_NOML|REINT_AUTORITATIVE_ML)
+#define REINT_ONCE (REINT_ONCE_NOML|REINT_ONCE_ML)
+
+#define RX_MATCH_TAINTED(prog) ((prog)->reganch & ROPT_TAINTED_SEEN)
+#define RX_MATCH_TAINTED_on(prog) ((prog)->reganch |= ROPT_TAINTED_SEEN)
+#define RX_MATCH_TAINTED_off(prog) ((prog)->reganch &= ~ROPT_TAINTED_SEEN)
+#define RX_MATCH_TAINTED_set(prog, t) ((t) \
+ ? RX_MATCH_TAINTED_on(prog) \
+ : RX_MATCH_TAINTED_off(prog))
+
+#define RX_MATCH_COPIED(prog) ((prog)->reganch & ROPT_COPY_DONE)
+#define RX_MATCH_COPIED_on(prog) ((prog)->reganch |= ROPT_COPY_DONE)
+#define RX_MATCH_COPIED_off(prog) ((prog)->reganch &= ~ROPT_COPY_DONE)
+#define RX_MATCH_COPIED_set(prog,t) ((t) \
+ ? RX_MATCH_COPIED_on(prog) \
+ : RX_MATCH_COPIED_off(prog))
+
+#ifdef PERL_COPY_ON_WRITE
+#define RX_MATCH_COPY_FREE(rx) \
+ STMT_START {if (rx->saved_copy) { \
+ SV_CHECK_THINKFIRST_COW_DROP(rx->saved_copy); \
+ } \
+ if (RX_MATCH_COPIED(rx)) { \
+ Safefree(rx->subbeg); \
+ RX_MATCH_COPIED_off(rx); \
+ }} STMT_END
+#else
+#define RX_MATCH_COPY_FREE(rx) \
+ STMT_START {if (RX_MATCH_COPIED(rx)) { \
+ Safefree(rx->subbeg); \
+ RX_MATCH_COPIED_off(rx); \
+ }} STMT_END
+#endif
+
+#define RX_MATCH_UTF8(prog) ((prog)->reganch & ROPT_MATCH_UTF8)
+#define RX_MATCH_UTF8_on(prog) ((prog)->reganch |= ROPT_MATCH_UTF8)
+#define RX_MATCH_UTF8_off(prog) ((prog)->reganch &= ~ROPT_MATCH_UTF8)
+#define RX_MATCH_UTF8_set(prog, t) ((t) \
+ ? (RX_MATCH_UTF8_on(prog), (PL_reg_match_utf8 = 1)) \
+ : (RX_MATCH_UTF8_off(prog), (PL_reg_match_utf8 = 0)))
+
+#define REXEC_COPY_STR 0x01 /* Need to copy the string. */
+#define REXEC_CHECKED 0x02 /* check_substr already checked. */
+#define REXEC_SCREAM 0x04 /* use scream table. */
+#define REXEC_IGNOREPOS 0x08 /* \G matches at start. */
+#define REXEC_NOT_FIRST 0x10 /* This is another iteration of //g. */
+#define REXEC_ML 0x20 /* $* was set. */
+
+#define ReREFCNT_inc(re) ((void)(re && re->refcnt++), re)
+#define ReREFCNT_dec(re) CALLREGFREE(aTHX_ re)
+
+#define FBMcf_TAIL_DOLLAR 1
+#define FBMcf_TAIL_DOLLARM 2
+#define FBMcf_TAIL_Z 4
+#define FBMcf_TAIL_z 8
+#define FBMcf_TAIL (FBMcf_TAIL_DOLLAR|FBMcf_TAIL_DOLLARM|FBMcf_TAIL_Z|FBMcf_TAIL_z)
+
+#define FBMrf_MULTILINE 1
-regexp *regcomp();
-int regexec();
+struct re_scream_pos_data_s;
https://perl5.git.perl.org / perl5.git / blobdiff