struct reg_data;
+struct regexp_engine;
+
typedef struct regexp {
I32 *startp;
I32 *endp;
U32 lastcloseparen; /* last paren matched */
U32 reganch; /* Internal use only +
Tainted information used by regexec? */
+ HV *paren_names; /* Paren names */
+ const struct regexp_engine* engine;
regnode program[1]; /* Unwarranted chumminess with compiler. */
} regexp;
+
+typedef struct re_scream_pos_data_s
+{
+ char **scream_olds; /* match pos */
+ I32 *scream_pos; /* Internal iterator of scream. */
+} re_scream_pos_data;
+
+typedef struct regexp_engine {
+ regexp* (*comp) (pTHX_ char* exp, char* xend, PMOP* pm);
+ I32 (*exec) (pTHX_ regexp* prog, char* stringarg, char* strend,
+ char* strbeg, I32 minend, SV* screamer,
+ void* data, U32 flags);
+ char* (*intuit) (pTHX_ regexp *prog, SV *sv, char *strpos,
+ char *strend, U32 flags,
+ struct re_scream_pos_data_s *data);
+ SV* (*checkstr) (pTHX_ regexp *prog);
+ void (*free) (pTHX_ struct regexp* r);
+#ifdef USE_ITHREADS
+ regexp* (*dupe) (pTHX_ const regexp *r, CLONE_PARAMS *param);
+#endif
+} regexp_engine;
+
#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 0x00000001
#define ROPT_EVAL_SEEN 0x00000400
#define ROPT_CANY_SEEN 0x00000800
#define ROPT_SANY_SEEN ROPT_CANY_SEEN /* src bckwrd cmpt */
+#define ROPT_GPOS_CHECK (ROPT_GPOS_SEEN|ROPT_ANCH_GPOS)
/* 0xf800 of reganch is used by PMf_COMPILETIME */
#define ROPT_COPY_DONE 0x00040000 /* subbeg is a copy of the string */
#define ROPT_TAINTED_SEEN 0x00080000
#define ROPT_MATCH_UTF8 0x10000000 /* subbeg is utf-8 */
+#define ROPT_RECURSE_SEEN 0x20000000
+#define ROPT_VERBARG_SEEN 0x40000000
#define RE_USE_INTUIT_NOML 0x00100000 /* Best to intuit before matching */
#define RE_USE_INTUIT_ML 0x00200000
#define RX_MATCH_COPIED_set(prog,t) ((t) \
? RX_MATCH_COPIED_on(prog) \
: RX_MATCH_COPIED_off(prog))
+
#endif /* PLUGGABLE_RE_EXTENSION */
/* Stuff that needs to be included in the plugable extension goes below here */
#define REXEC_NOT_FIRST 0x10 /* This is another iteration of //g. */
#define ReREFCNT_inc(re) ((void)(re && re->refcnt++), re)
-#define ReREFCNT_dec(re) CALLREGFREE(aTHX_ re)
+#define ReREFCNT_dec(re) CALLREGFREE(re)
#define FBMcf_TAIL_DOLLAR 1
#define FBMcf_TAIL_DOLLARM 2
char *till;
SV *sv;
char *ganch;
+ char *cutpoint;
} regmatch_info;
/* structures for holding and saving the state maintained by regmatch() */
+#define MAX_RECURSE_EVAL_NOCHANGE_DEPTH 50
+
typedef I32 CHECKPOINT;
typedef struct regmatch_state {
-
- /* these vars contain state that needs to be maintained
- * across the main while loop ... */
-
int resume_state; /* where to jump to on return */
- regnode *scan; /* Current node. */
- regnode *next; /* Next node. */
- bool minmod; /* the next "{n,m}" is a "{n,m}?" */
- bool sw; /* the condition value in (?(cond)a|b) */
- int logical;
- I32 unwind; /* savestack index of current unwind block */
- char *locinput;
-
- /* ... while the rest of these are local to an individual branch */
-
- I32 n; /* no or next */
- I32 ln; /* len or last */
+ char *locinput; /* where to backtrack in string on failure */
union {
struct {
reg_trie_accepted *accept_buff;
U32 accepted; /* how many accepting states we have seen */
- U16 *jump; /* negative offsets from B */
+ U16 *jump; /* positive offsets from me */
regnode *B; /* node following the trie */
- regnode *me; /* only needed for debugging */
+ regnode *me; /* Which node am I - needed for jump tries*/
} trie;
struct {
/* this first element must match u.yes */
struct regmatch_state *prev_yes_state;
+ struct regmatch_state *prev_eval;
+ struct regmatch_state *prev_curlyx;
regexp *prev_rex;
- int toggleutf;
+ U32 toggle_reg_flags; /* what bits in PL_reg_flags to
+ flip when transitioning between
+ inner and outer rexen */
CHECKPOINT cp; /* remember current savestack indexes */
CHECKPOINT lastcp;
regnode *B; /* the node following us */
+ U32 close_paren; /* which close bracket is our end */
} eval;
struct {
- CHECKPOINT cp; /* remember current savestack indexes */
- struct regmatch_state *outercc; /* outer CURLYX state if any */
-
- /* these contain the current curly state, and are accessed
- * by subsequent WHILEMs */
+ /* this first element must match u.yes */
+ struct regmatch_state *prev_yes_state;
+ struct regmatch_state *prev_curlyx; /* previous cur_curlyx */
+ CHECKPOINT cp; /* remember current savestack index */
+ bool minmod;
int parenfloor;/* how far back to strip paren data */
- int cur; /* how many instances of scan we've matched */
- int min; /* the minimal number of scans to match */
- int max; /* the maximal number of scans to match */
- regnode * scan; /* the thing to match */
- char * lastloc;/* where we started matching this scan */
+ int min; /* the minimal number of A's to match */
+ int max; /* the maximal number of A's to match */
+ regnode *A, *B; /* the nodes corresponding to /A*B/ */
+
+ /* these two are modified by WHILEM */
+ int count; /* how many instances of A we've matched */
+ char *lastloc;/* where previous A matched (0-len detect) */
} curlyx;
struct {
- CHECKPOINT cp; /* remember current savestack indexes */
- CHECKPOINT lastcp;
- struct regmatch_state *savecc;
- char *lastloc; /* Detection of 0-len. */
- I32 cache_offset;
- I32 cache_bit;
+ /* this first element must match u.yes */
+ struct regmatch_state *prev_yes_state;
+ struct regmatch_state *save_curlyx;
+ CHECKPOINT cp; /* remember current savestack indexes */
+ CHECKPOINT lastcp;
+ char *save_lastloc; /* previous curlyx.lastloc */
+ I32 cache_offset;
+ I32 cache_mask;
} whilem;
struct {
/* this first element must match u.yes */
struct regmatch_state *prev_yes_state;
I32 wanted;
+ I32 logical; /* saved copy of 'logical' var */
regnode *me; /* the IFMATCH/SUSPEND/UNLESSM node */
} ifmatch; /* and SUSPEND/UNLESSM */
+
+ struct {
+ /* this first element must match u.yes */
+ struct regmatch_state *prev_yes_state;
+ struct regmatch_state *prev_mark;
+ SV* mark_name;
+ char *mark_loc;
+ } mark;
} u;
} regmatch_state;
#define PL_reg_start_tmp PL_reg_state.re_state_reg_start_tmp
#define PL_reg_start_tmpl PL_reg_state.re_state_reg_start_tmpl
#define PL_reg_eval_set PL_reg_state.re_state_reg_eval_set
-#define PL_regindent PL_reg_state.re_state_regindent
#define PL_reg_match_utf8 PL_reg_state.re_state_reg_match_utf8
#define PL_reg_magic PL_reg_state.re_state_reg_magic
#define PL_reg_oldpos PL_reg_state.re_state_reg_oldpos
char **re_state_reg_start_tmp; /* from regexec.c */
U32 re_state_reg_start_tmpl; /* from regexec.c */
I32 re_state_reg_eval_set; /* from regexec.c */
- int re_state_regindent; /* from regexec.c */
bool re_state_reg_match_utf8; /* from regexec.c */
MAGIC *re_state_reg_magic; /* from regexec.c */
I32 re_state_reg_oldpos; /* from regexec.c */