struct regexp;
struct reg_substr_datum {
- I32 min_offset;
- I32 max_offset;
+ SSize_t min_offset; /* min pos (in chars) that substr must appear */
+ SSize_t max_offset /* max pos (in chars) that substr must appear */;
SV *substr; /* non-utf8 variant */
SV *utf8_substr; /* utf8 variant */
- I32 end_shift;
+ SSize_t end_shift; /* how many fixed chars must end the string */
};
struct reg_substr_data {
+ U8 check_ix; /* index into data[] of check substr */
struct reg_substr_datum data[3]; /* Actual array */
};
/* offsets within a string of a particular /(.)/ capture */
typedef struct regexp_paren_pair {
- I32 start;
- I32 end;
+ SSize_t start;
+ SSize_t end;
/* 'start_tmp' records a new opening position before the matching end
* has been found, so that the old start and end values are still
* valid, e.g.
* "abc" =~ /(.(?{print "[$1]"}))+/
*outputs [][a][b]
* This field is not part of the API. */
- I32 start_tmp;
+ SSize_t start_tmp;
} regexp_paren_pair;
#if defined(PERL_IN_REGCOMP_C) || defined(PERL_IN_UTF8_C)
/* Information about the match that the perl core uses to */ \
/* manage things */ \
U32 extflags; /* Flags used both externally and internally */ \
- I32 minlen; /* mininum possible number of chars in string to match */\
- I32 minlenret; /* mininum possible number of chars in $& */ \
- U32 gofs; /* chars left of pos that we search from */ \
+ SSize_t minlen; /* mininum possible number of chars in string to match */\
+ SSize_t minlenret; /* mininum possible number of chars in $& */ \
+ STRLEN gofs; /* chars left of pos that we search from */ \
/* substring data about strings that must appear in the */ \
/* final match, used for optimisations */ \
struct reg_substr_data *substrs; \
/* saved or original string so \digit works forever. */ \
char *subbeg; \
SV_SAVED_COPY /* If non-NULL, SV which is COW from original */\
- I32 sublen; /* Length of string pointed by subbeg */ \
- I32 suboffset; /* byte offset of subbeg from logical start of str */ \
- I32 subcoffset; /* suboffset equiv, but in chars (for @-/@+) */ \
+ SSize_t sublen; /* Length of string pointed by subbeg */ \
+ SSize_t suboffset; /* byte offset of subbeg from logical start of str */ \
+ SSize_t subcoffset; /* suboffset equiv, but in chars (for @-/@+) */ \
/* Information about the match that isn't often used */ \
+ SSize_t maxlen; /* mininum possible number of chars in string to match */\
/* offset from wrapped to the start of precomp */ \
PERL_BITFIELD32 pre_prefix:4; \
/* original flags used to compile the pattern, may differ */ \
typedef struct re_scream_pos_data_s
{
char **scream_olds; /* match pos */
- I32 *scream_pos; /* Internal iterator of scream. */
+ SSize_t *scream_pos; /* Internal iterator of scream. */
} re_scream_pos_data;
/* regexp_engine structure. This is the dispatch table for regexes.
typedef struct regexp_engine {
REGEXP* (*comp) (pTHX_ SV * const pattern, U32 flags);
I32 (*exec) (pTHX_ REGEXP * const rx, char* stringarg, char* strend,
- char* strbeg, I32 minend, SV* sv,
+ char* strbeg, SSize_t minend, SV* sv,
void* data, U32 flags);
char* (*intuit) (pTHX_
REGEXP * const rx,
=for apidoc Am|REGEXP *|SvRX|SV *sv
-Convenience macro to get the REGEXP from a SV. This is approximately
+Convenience macro to get the REGEXP from a SV. This is approximately
equivalent to the following snippet:
if (SvMAGICAL(sv))
# error "RXf_SPLIT does not match RXf_PMf_SPLIT"
#endif
-/* Manually decorate this function with gcc-style attributes just to
- * avoid having to restructure the header files and their called order,
- * as proto.h would have to be included before this file, and isn't */
-
-PERL_STATIC_INLINE const char *
-get_regex_charset_name(const U32 flags, STRLEN* const lenp)
- __attribute__warn_unused_result__;
-
-#define MAX_CHARSET_NAME_LENGTH 2
-
-PERL_STATIC_INLINE const char *
-get_regex_charset_name(const U32 flags, STRLEN* const lenp)
-{
- /* Returns a string that corresponds to the name of the regex character set
- * given by 'flags', and *lenp is set the length of that string, which
- * cannot exceed MAX_CHARSET_NAME_LENGTH characters */
-
- *lenp = 1;
- switch (get_regex_charset(flags)) {
- case REGEX_DEPENDS_CHARSET: return DEPENDS_PAT_MODS;
- case REGEX_LOCALE_CHARSET: return LOCALE_PAT_MODS;
- case REGEX_UNICODE_CHARSET: return UNICODE_PAT_MODS;
- case REGEX_ASCII_RESTRICTED_CHARSET: return ASCII_RESTRICT_PAT_MODS;
- case REGEX_ASCII_MORE_RESTRICTED_CHARSET:
- *lenp = 2;
- return ASCII_MORE_RESTRICT_PAT_MODS;
- default:
- return "?"; /* Unknown */
- }
-}
-
-/* Anchor and GPOS related stuff */
-#define RXf_ANCH_BOL (1<<(RXf_BASE_SHIFT+0))
-#define RXf_ANCH_MBOL (1<<(RXf_BASE_SHIFT+1))
-#define RXf_ANCH_SBOL (1<<(RXf_BASE_SHIFT+2))
-#define RXf_ANCH_GPOS (1<<(RXf_BASE_SHIFT+3))
-#define RXf_GPOS_SEEN (1<<(RXf_BASE_SHIFT+4))
-#define RXf_GPOS_FLOAT (1<<(RXf_BASE_SHIFT+5))
-/* two bits here */
-#define RXf_ANCH (RXf_ANCH_BOL|RXf_ANCH_MBOL|RXf_ANCH_GPOS|RXf_ANCH_SBOL)
-#define RXf_GPOS_CHECK (RXf_GPOS_SEEN|RXf_ANCH_GPOS)
-#define RXf_ANCH_SINGLE (RXf_ANCH_SBOL|RXf_ANCH_GPOS)
+/* Unused +0 - +6 */
/* What we have seen */
-#define RXf_NO_INPLACE_SUBST (1<<(RXf_BASE_SHIFT+6))
-#define RXf_EVAL_SEEN (1<<(RXf_BASE_SHIFT+7))
-#define RXf_CANY_SEEN (1<<(RXf_BASE_SHIFT+8))
+#define RXf_NO_INPLACE_SUBST (1<<(RXf_BASE_SHIFT+7))
+#define RXf_EVAL_SEEN (1<<(RXf_BASE_SHIFT+8))
/* Special */
-#define RXf_NOSCAN (1<<(RXf_BASE_SHIFT+9))
+#define RXf_UNBOUNDED_QUANTIFIER_SEEN (1<<(RXf_BASE_SHIFT+9))
#define RXf_CHECK_ALL (1<<(RXf_BASE_SHIFT+10))
/* UTF8 related */
#define RXf_INTUIT_TAIL (1<<(RXf_BASE_SHIFT+14))
#define RXf_USE_INTUIT (RXf_USE_INTUIT_NOML|RXf_USE_INTUIT_ML)
+/* Do we have some sort of anchor? */
+#define RXf_IS_ANCHORED (1<<(RXf_BASE_SHIFT+15))
+
/* Copy and tainted info */
#define RXf_COPY_DONE (1<<(RXf_BASE_SHIFT+16))
-/* during execution: pattern temporarily tainted by executing locale ops;
- * post-execution: $1 et al are tainted */
+/* post-execution: $1 et al are tainted */
#define RXf_TAINTED_SEEN (1<<(RXf_BASE_SHIFT+17))
/* this pattern was tainted during compilation */
#define RXf_TAINTED (1<<(RXf_BASE_SHIFT+18))
*
*/
-#if NO_TAINT_SUPPORT
+#ifdef NO_TAINT_SUPPORT
# define RX_ISTAINTED(prog) 0
# define RX_TAINT_on(prog) NOOP
# define RXp_MATCH_TAINTED(prog) 0
#define RX_SAVED_COPY(prog) (ReANY(prog)->saved_copy)
/* last match was zero-length */
#define RX_ZERO_LEN(prog) \
- (RX_OFFS(prog)[0].start + RX_GOFS(prog) == (UV)RX_OFFS(prog)[0].end)
+ (RX_OFFS(prog)[0].start + (SSize_t)RX_GOFS(prog) \
+ == RX_OFFS(prog)[0].end)
#endif /* PLUGGABLE_RE_EXTENSION */
STRLEN suboffset; /* saved suboffset field from rex */
STRLEN subcoffset; /* saved subcoffset field from rex */
MAGIC *pos_magic; /* pos() magic attached to $_ */
- I32 pos; /* the original value of pos() in pos_magic */
+ SSize_t pos; /* the original value of pos() in pos_magic */
+ U8 pos_flags; /* flags to be restored; currently only MGf_BYTES*/
} regmatch_info_aux_eval;