/* regcomp.h
*
* Copyright (C) 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999,
- * 2000, 2001, 2002, 2003, 2005, 2006 by Larry Wall and others
+ * 2000, 2001, 2002, 2003, 2005, 2006, 2007, 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.
*/
#include "regcharclass.h"
-typedef OP OP_4tree; /* Will be redefined later. */
-
-
/* Convert branch sequences to more efficient trie ops? */
#define PERL_ENABLE_TRIE_OPTIMISATION 1
-/* Be really agressive about optimising patterns with trie sequences? */
+/* Be really aggressive about optimising patterns with trie sequences? */
#define PERL_ENABLE_EXTENDED_TRIE_OPTIMISATION 1
/* Should the optimiser take positive assertions into account? */
Used to make it easier to clone and free arbitrary
data that the regops need. Often the ARG field of
a regop is an index into this structure */
+ struct reg_code_block *code_blocks;/* positions of literal (?{}) */
+ int num_code_blocks; /* size of code_blocks[] */
regnode program[1]; /* Unwarranted chumminess with compiler. */
} regexp_internal;
#define PREGf_NAUGHTY 0x00000004 /* how exponential is this pattern? */
#define PREGf_VERBARG_SEEN 0x00000008
#define PREGf_CUTGROUP_SEEN 0x00000010
+#define PREGf_USE_RE_EVAL 0x00000020 /* compiled with "use re 'eval'" */
/* this is where the old regcomp.h started */
#define ANYOF_BITMAP_SIZE 32 /* 256 b/(8 b/B) */
-#define ANYOF_CLASSBITMAP_SIZE 4 /* up to 40 (8*5) named classes */
+#define ANYOF_CLASSBITMAP_SIZE 4 /* up to 32 (8*4) named classes */
/* also used by trie */
struct regnode_charclass {
U8 flags;
U8 type;
U16 next_off;
- U32 arg1;
+ U32 arg1; /* used as ptr in S_regclass */
char bitmap[ANYOF_BITMAP_SIZE]; /* only compile-time */
};
-struct regnode_charclass_class { /* has [[:blah:]] classes */
- U8 flags; /* should have ANYOF_CLASS here */
+/* has runtime (locale) \d, \w, ..., [:posix:] classes */
+struct regnode_charclass_class {
+ U8 flags; /* ANYOF_CLASS bit must go here */
U8 type;
U16 next_off;
- U32 arg1;
+ U32 arg1; /* used as ptr in S_regclass */
char bitmap[ANYOF_BITMAP_SIZE]; /* both compile-time */
char classflags[ANYOF_CLASSBITMAP_SIZE]; /* and run-time */
};
#undef STRING
#define OP(p) ((p)->type)
+#define FLAGS(p) ((p)->flags) /* Caution: Doesn't apply to all \
+ regnode types. For some, it's the \
+ character set of the regnode */
#define OPERAND(p) (((struct regnode_string *)p)->string)
#define MASK(p) ((char*)OPERAND(p))
#define STR_LEN(p) (((struct regnode_string *)p)->str_len)
#define SIZE_ONLY (RExC_emit == &PL_regdummy)
-/* Flags for node->flags of ANYOF */
-
-#define ANYOF_CLASS 0x08 /* has [[:blah:]] classes */
-#define ANYOF_INVERT 0x04
-#define ANYOF_FOLD 0x02
-#define ANYOF_LOCALE 0x01
-
-/* Used for regstclass only */
-#define ANYOF_EOS 0x10 /* Can match an empty string too */
-
-/* There is a character or a range past 0xff */
-#define ANYOF_UNICODE 0x20
-#define ANYOF_UNICODE_ALL 0x40 /* Can match any char past 0xff */
-
-/* size of node is large (includes class pointer) */
-#define ANYOF_LARGE 0x80
-
-/* Are there any runtime flags on in this node? */
-#define ANYOF_RUNTIME(s) (ANYOF_FLAGS(s) & 0x0f)
+/* If the bitmap doesn't fully represent what this ANYOF node can match, the
+ * ARG is set to this special value (since 0, 1, ... are legal, but will never
+ * reach this high). */
+#define ANYOF_NONBITMAP_EMPTY ((U32) -1)
+
+/* The information used to be stored as as combination of the ANYOF_UTF8 and
+ * ANYOF_NONBITMAP_NON_UTF8 bits in the flags field, but was moved out of there
+ * to free up a bit for other uses. This tries to hide the change from
+ * existing code as much as possible. Now, the data structure that goes in ARG
+ * is not allocated unless it is needed, and that is what is used to determine
+ * if there is something outside the bitmap. The code now assumes that if
+ * that structure exists, that any UTF-8 encoded string should be tried against
+ * it, but a non-UTF8-encoded string will be tried only if the
+ * ANYOF_NONBITMAP_NON_UTF8 bit is also set. */
+#define ANYOF_NONBITMAP(node) (ARG(node) != ANYOF_NONBITMAP_EMPTY)
+
+/* Flags for node->flags of ANYOF. These are in short supply, so some games
+ * are done to share them, as described below. Already, the ANYOF_LOCALE and
+ * ANYOF_CLASS bits are shared, making a space penalty for all locale nodes.
+ * An option would be to push them into new nodes. E.g. there could be an
+ * ANYOF_LOCALE node that would be in place of the flag of the same name. But
+ * there are better options. The UNICODE_ALL bit could be freed up by
+ * resorting to creating a swash containing everything above 255. This
+ * introduces a performance penalty. Better would be to split it off into a
+ * separate node, which actually would improve performance by allowing adding a
+ * case statement to regexec.c use the bit map for code points under 256, and
+ * to match everything above. If flags need to be added that are applicable to
+ * the synthetic start class only, with some work, they could be put in the
+ * next-node field, or in an unused bit of the classflags field. This could be
+ * done with the current EOS flag, and a new node type created that is just for
+ * the scc, freeing up that bit */
+
+#define ANYOF_LOCALE 0x01 /* /l modifier */
+
+/* The fold is calculated and stored in the bitmap where possible at compile
+ * time. However under locale, the actual folding varies depending on
+ * what the locale is at the time of execution, so it has to be deferred until
+ * then */
+#define ANYOF_LOC_FOLD 0x02
+
+#define ANYOF_INVERT 0x04
+
+/* Set if this is a struct regnode_charclass_class vs a regnode_charclass. This
+ * is used for runtime \d, \w, [:posix:], ..., which are used only in locale
+ * and the optimizer's synthetic start class. Non-locale \d, etc are resolved
+ * at compile-time. Now shared with ANYOF_LOCALE, forcing all locale nodes to
+ * be large */
+#define ANYOF_CLASS ANYOF_LOCALE
+#define ANYOF_LARGE ANYOF_CLASS /* Same; name retained for back compat */
+
+/* Should this character class warn if matched against a character above
+ * Unicode */
+#define ANYOF_WARN_SUPER 0x08
+
+/* EOS, meaning that it can match an empty string too, is used for the
+ * synthetic start class only. */
+#define ANYOF_EOS 0x10
+
+/* ? Is this node the synthetic start class (ssc). This bit is shared with
+ * ANYOF_EOS, as the latter is used only for the ssc, and then not used by
+ * regexec.c. And, the code is structured so that if it is set, the ssc is
+ * not used, so it is guaranteed to be 0 for the ssc by the time regexec.c
+ * gets executed, and 0 for a non-ssc ANYOF node, as it only ever gets set for
+ * a potential ssc candidate. Thus setting it to 1 after it has been
+ * determined that the ssc will be used is not ambiguous */
+#define ANYOF_IS_SYNTHETIC ANYOF_EOS
+
+/* Can match something outside the bitmap that isn't in utf8 */
+#define ANYOF_NONBITMAP_NON_UTF8 0x20
+
+/* Matches every code point 0x100 and above*/
+#define ANYOF_UNICODE_ALL 0x40
+
+/* Match all Latin1 characters that aren't ASCII when the target string is not
+ * in utf8. */
+#define ANYOF_NON_UTF8_LATIN1_ALL 0x80
#define ANYOF_FLAGS_ALL 0xff
+/* These are the flags that ANYOF_INVERT being set or not doesn't affect
+ * whether they are operative or not. e.g., the node still has LOCALE
+ * regardless of being inverted; whereas ANYOF_UNICODE_ALL means something
+ * different if inverted */
+#define INVERSION_UNAFFECTED_FLAGS (ANYOF_LOCALE \
+ |ANYOF_LOC_FOLD \
+ |ANYOF_CLASS \
+ |ANYOF_EOS \
+ |ANYOF_NONBITMAP_NON_UTF8)
+
/* Character classes for node->classflags of ANYOF */
/* Should be synchronized with a table in regprop() */
-/* 2n should pair with 2n+1 */
-
-#define ANYOF_ALNUM 0 /* \w, PL_utf8_alnum, utf8::IsWord, ALNUM */
-#define ANYOF_NALNUM 1
-#define ANYOF_SPACE 2 /* \s */
-#define ANYOF_NSPACE 3
-#define ANYOF_DIGIT 4
-#define ANYOF_NDIGIT 5
-#define ANYOF_ALNUMC 6 /* isalnum(3), utf8::IsAlnum, ALNUMC */
-#define ANYOF_NALNUMC 7
-#define ANYOF_ALPHA 8
-#define ANYOF_NALPHA 9
-#define ANYOF_ASCII 10
-#define ANYOF_NASCII 11
-#define ANYOF_CNTRL 12
-#define ANYOF_NCNTRL 13
-#define ANYOF_GRAPH 14
-#define ANYOF_NGRAPH 15
-#define ANYOF_LOWER 16
-#define ANYOF_NLOWER 17
-#define ANYOF_PRINT 18
-#define ANYOF_NPRINT 19
-#define ANYOF_PUNCT 20
-#define ANYOF_NPUNCT 21
-#define ANYOF_UPPER 22
-#define ANYOF_NUPPER 23
-#define ANYOF_XDIGIT 24
-#define ANYOF_NXDIGIT 25
-#define ANYOF_PSXSPC 26 /* POSIX space: \s plus the vertical tab */
-#define ANYOF_NPSXSPC 27
-#define ANYOF_BLANK 28 /* GNU extension: space and tab: non-vertical space */
-#define ANYOF_NBLANK 29
+/* 2n should be the normal one, paired with its complement at 2n+1 */
+
+#define ANYOF_WORDCHAR ((_CC_WORDCHAR) * 2) /* \w, PL_utf8_alnum, utf8::IsWord, ALNUM */
+#define ANYOF_NWORDCHAR ((ANYOF_WORDCHAR) + 1)
+#define ANYOF_SPACE ((_CC_SPACE) * 2) /* \s */
+#define ANYOF_NSPACE ((ANYOF_SPACE) + 1)
+#define ANYOF_DIGIT ((_CC_DIGIT) * 2) /* \d */
+#define ANYOF_NDIGIT ((ANYOF_DIGIT) + 1)
+#define ANYOF_ALNUMC ((_CC_ALNUMC) * 2) /* [[:alnum:]] isalnum(3), utf8::IsAlnum, ALNUMC */
+#define ANYOF_NALNUMC ((ANYOF_ALNUMC) + 1)
+#define ANYOF_ALPHA ((_CC_ALPHA) * 2)
+#define ANYOF_NALPHA ((ANYOF_ALPHA) + 1)
+#define ANYOF_ASCII ((_CC_ASCII) * 2)
+#define ANYOF_NASCII ((ANYOF_ASCII) + 1)
+#define ANYOF_CNTRL ((_CC_CNTRL) * 2)
+#define ANYOF_NCNTRL ((ANYOF_CNTRL) + 1)
+#define ANYOF_GRAPH ((_CC_GRAPH) * 2)
+#define ANYOF_NGRAPH ((ANYOF_GRAPH) + 1)
+#define ANYOF_LOWER ((_CC_LOWER) * 2)
+#define ANYOF_NLOWER ((ANYOF_LOWER) + 1)
+#define ANYOF_PRINT ((_CC_PRINT) * 2)
+#define ANYOF_NPRINT ((ANYOF_PRINT) + 1)
+#define ANYOF_PUNCT ((_CC_PUNCT) * 2)
+#define ANYOF_NPUNCT ((ANYOF_PUNCT) + 1)
+#define ANYOF_UPPER ((_CC_UPPER) * 2)
+#define ANYOF_NUPPER ((ANYOF_UPPER) + 1)
+#define ANYOF_XDIGIT ((_CC_XDIGIT) * 2)
+#define ANYOF_NXDIGIT ((ANYOF_XDIGIT) + 1)
+#define ANYOF_PSXSPC ((_CC_PSXSPC) * 2) /* POSIX space: \s plus the vertical tab */
+#define ANYOF_NPSXSPC ((ANYOF_PSXSPC) + 1)
+#define ANYOF_BLANK ((_CC_BLANK) * 2) /* GNU extension: space and tab: non-vertical space */
+#define ANYOF_NBLANK ((ANYOF_BLANK) + 1)
#define ANYOF_MAX 32
+#if (ANYOF_MAX <= _HIGHEST_REGCOMP_DOT_H_SYNC * 2 + 1)
+# error Problem with handy.h _CC_foo #defines
+#endif
/* pseudo classes, not stored in the class bitmap, but used as flags
during compilation of char classes */
#define ANYOF_NALNUML ANYOF_NALNUM
#define ANYOF_SPACEL ANYOF_SPACE
#define ANYOF_NSPACEL ANYOF_NSPACE
+#define ANYOF_ALNUM ANYOF_WORDCHAR
+#define ANYOF_NALNUM ANYOF_NWORDCHAR
/* Utility macros for the bitmap and classes of ANYOF */
#define ANYOF_CLASS_TEST(p, c) (ANYOF_CLASS_BYTE(p, c) & ANYOF_BIT(c))
#define ANYOF_CLASS_ZERO(ret) Zero(((struct regnode_charclass_class*)(ret))->classflags, ANYOF_CLASSBITMAP_SIZE, char)
+#define ANYOF_CLASS_SETALL(ret) \
+ memset (((struct regnode_charclass_class*)(ret))->classflags, 255, ANYOF_CLASSBITMAP_SIZE)
#define ANYOF_BITMAP_ZERO(ret) Zero(((struct regnode_charclass*)(ret))->bitmap, ANYOF_BITMAP_SIZE, char)
#define ANYOF_BITMAP(p) (((struct regnode_charclass*)(p))->bitmap)
#define ANYOF_BITMAP_CLEARALL(p) \
Zero (ANYOF_BITMAP(p), ANYOF_BITMAP_SIZE)
/* Check that all 256 bits are all set. Used in S_cl_is_anything() */
-#define ANYOF_BITMAP_TESTALLSET(p) \
+#define ANYOF_BITMAP_TESTALLSET(p) /* Assumes sizeof(p) == 32 */ \
memEQ (ANYOF_BITMAP(p), "\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377\377", ANYOF_BITMAP_SIZE)
#define ANYOF_SKIP ((ANYOF_SIZE - 1)/sizeof(regnode))
#define ANYOF_CLASS_SKIP ((ANYOF_CLASS_SIZE - 1)/sizeof(regnode))
-#define ANYOF_CLASS_ADD_SKIP (ANYOF_CLASS_SKIP - ANYOF_SKIP)
+
+#if ANYOF_CLASSBITMAP_SIZE != 4
+# error ANYOF_CLASSBITMAP_SIZE is expected to be 4
+#endif
+#define ANYOF_CLASS_TEST_ANY_SET(p) ((ANYOF_FLAGS(p) & ANYOF_CLASS) \
+ && memNE (((struct regnode_charclass_class*)(p))->classflags, \
+ "\0\0\0\0", ANYOF_CLASSBITMAP_SIZE))
+/*#define ANYOF_CLASS_ADD_SKIP (ANYOF_CLASS_SKIP - ANYOF_SKIP)
+ * */
/*
#define REG_SEEN_ZERO_LEN 0x00000001
#define REG_SEEN_LOOKBEHIND 0x00000002
#define REG_SEEN_GPOS 0x00000004
-#define REG_SEEN_EVAL 0x00000008
+/* spare */
#define REG_SEEN_CANY 0x00000010
#define REG_SEEN_SANY REG_SEEN_CANY /* src bckwrd cmpt */
#define REG_SEEN_RECURSE 0x00000020
#define REG_SEEN_VERBARG 0x00000080
#define REG_SEEN_CUTGROUP 0x00000100
#define REG_SEEN_RUN_ON_COMMENT 0x00000200
+#define REG_SEEN_EXACTF_SHARP_S 0x00000400
START_EXTERN_C
#include "regnodes.h"
#endif
-/* The following have no fixed length. U8 so we can do strchr() on it. */
-#ifndef DOINIT
-EXTCONST U8 PL_varies[];
-#else
-EXTCONST U8 PL_varies[] = {
- BRANCH, BACK, STAR, PLUS, CURLY, CURLYX, REF, REFF, REFFL,
- WHILEM, CURLYM, CURLYN, BRANCHJ, IFTHEN, SUSPEND, CLUMP,
- NREF, NREFF, NREFFL,
- 0
-};
-#endif
-
-/* The following always have a length of 1. U8 we can do strchr() on it. */
-/* (Note that length 1 means "one character" under UTF8, not "one octet".) */
-#ifndef DOINIT
-EXTCONST U8 PL_simple[];
-#else
-EXTCONST U8 PL_simple[] = {
- REG_ANY, SANY, CANY,
- ANYOF,
- ALNUM, ALNUML,
- NALNUM, NALNUML,
- SPACE, SPACEL,
- NSPACE, NSPACEL,
- DIGIT, NDIGIT,
- VERTWS, NVERTWS,
- HORIZWS, NHORIZWS,
- 0
-};
-#endif
-
#ifndef PLUGGABLE_RE_EXTENSION
#ifndef DOINIT
EXTCONST regexp_engine PL_core_reg_engine;
Perl_reg_named_buff_iter,
Perl_reg_qr_package,
#if defined(USE_ITHREADS)
- Perl_regdupe_internal
+ Perl_regdupe_internal,
#endif
+ Perl_re_op_compile
};
#endif /* DOINIT */
#endif /* PLUGGABLE_RE_EXTENSION */
/* .what is a character array with one character for each member of .data
* The character describes the function of the corresponding .data item:
+ * a - AV for paren_name_list under DEBUGGING
* f - start-class data for regstclass optimization
- * n - Root of op tree for (?{EVAL}) item
- * o - Start op for (?{EVAL}) item
- * p - Pad for (?{EVAL}) item
+ * l - start op for literal (?{EVAL}) item
+ * L - start op for literal (?{EVAL}) item, with separate CV (qr//)
+ * r - pointer to an embedded code-containing qr, e.g. /ab$qr/
* s - swash for Unicode-style character class, and the multicharacter
* strings resulting from casefolding the single-character entries
* in the character class
#define check_offset_max substrs->data[2].max_offset
#define check_end_shift substrs->data[2].end_shift
-
+#define RX_ANCHORED_SUBSTR(rx) (ReANY(rx)->anchored_substr)
+#define RX_ANCHORED_UTF8(rx) (ReANY(rx)->anchored_utf8)
+#define RX_FLOAT_SUBSTR(rx) (ReANY(rx)->float_substr)
+#define RX_FLOAT_UTF8(rx) (ReANY(rx)->float_utf8)
/* trie related stuff */
} trans;
};
+/* info per word; indexed by wordnum */
+typedef struct {
+ U16 prev; /* previous word in acceptance chain; eg in
+ * zzz|abc|ab/ after matching the chars abc, the
+ * accepted word is #2, and the previous accepted
+ * word is #3 */
+ U32 len; /* how many chars long is this word? */
+ U32 accept; /* accept state for this word */
+} reg_trie_wordinfo;
typedef struct _reg_trie_state reg_trie_state;
reg_trie_state *states; /* state data */
reg_trie_trans *trans; /* array of transition elements */
char *bitmap; /* stclass bitmap */
- U32 *wordlen; /* array of lengths of words */
U16 *jump; /* optional 1 indexed array of offsets before tail
for the node following a given word. */
- U16 *nextword; /* optional 1 indexed array to support linked list
- of duplicate wordnums */
+ reg_trie_wordinfo *wordinfo; /* array of info per word */
U16 uniquecharcount; /* unique chars in trie (width of trans table) */
U32 startstate; /* initial state - used for common prefix optimisation */
STRLEN minlen; /* minimum length of words in trie - build/opt only? */
STRLEN maxlen; /* maximum length of words in trie - build/opt only? */
+ U32 prefixlen; /* #chars in common prefix */
U32 statecount; /* Build only - number of states in the states array
(including the unused zero state) */
U32 wordcount; /* Build only */
};
typedef struct _reg_ac_data reg_ac_data;
-/* ANY_BIT doesnt use the structure, so we can borrow it here.
+/* ANY_BIT doesn't use the structure, so we can borrow it here.
This is simpler than refactoring all of it as wed end up with
three different sets... */
#define RE_DEBUG_EXTRA_STATE 0x080000
#define RE_DEBUG_EXTRA_OPTIMISE 0x100000
#define RE_DEBUG_EXTRA_BUFFERS 0x400000
+#define RE_DEBUG_EXTRA_GPOS 0x800000
/* combined */
#define RE_DEBUG_EXTRA_STACK 0x280000
#define DEBUG_TRIE_r(x) DEBUG_r( \
if (re_debug_flags & (RE_DEBUG_COMPILE_TRIE \
| RE_DEBUG_EXECUTE_TRIE )) x )
+#define DEBUG_GPOS_r(x) DEBUG_r( \
+ if (re_debug_flags & RE_DEBUG_EXTRA_GPOS) x )
/* initialization */
-/* get_sv() can return NULL during global destruction. */
+/* get_sv() can return NULL during global destruction. re_debug_flags can get
+ * clobbered by a longjmp, so must be initialized */
#define GET_RE_DEBUG_FLAGS DEBUG_r({ \
SV * re_debug_flags_sv = NULL; \
+ re_debug_flags = 0; \
re_debug_flags_sv = get_sv(RE_DEBUG_FLAGS, 1); \
if (re_debug_flags_sv) { \
if (!SvIOK(re_debug_flags_sv)) \
#ifdef DEBUGGING
-#define GET_RE_DEBUG_FLAGS_DECL IV re_debug_flags = 0; GET_RE_DEBUG_FLAGS;
+#define GET_RE_DEBUG_FLAGS_DECL VOL IV re_debug_flags \
+ PERL_UNUSED_DECL = 0; GET_RE_DEBUG_FLAGS;
#define RE_PV_COLOR_DECL(rpv,rlen,isuni,dsv,pv,l,m,c1,c2) \
const char * const rpv = \
pv_pretty((dsv), (pv), (l), (m), \
PL_colors[(c1)],PL_colors[(c2)], \
- PERL_PV_ESCAPE_RE |((isuni) ? PERL_PV_ESCAPE_UNI : 0) ); \
+ PERL_PV_ESCAPE_RE|PERL_PV_ESCAPE_NONASCII |((isuni) ? PERL_PV_ESCAPE_UNI : 0) ); \
const int rlen = SvCUR(dsv)
#define RE_SV_ESCAPE(rpv,isuni,dsv,sv,m) \
const char * const rpv = \
pv_pretty((dsv), (SvPV_nolen_const(sv)), (SvCUR(sv)), (m), \
PL_colors[(c1)],PL_colors[(c2)], \
- PERL_PV_ESCAPE_RE |((isuni) ? PERL_PV_ESCAPE_UNI : 0) )
+ PERL_PV_ESCAPE_RE|PERL_PV_ESCAPE_NONASCII |((isuni) ? PERL_PV_ESCAPE_UNI : 0) )
#define RE_PV_QUOTED_DECL(rpv,isuni,dsv,pv,l,m) \
const char * const rpv = \
pv_pretty((dsv), (pv), (l), (m), \
PL_colors[0], PL_colors[1], \
- ( PERL_PV_PRETTY_QUOTE | PERL_PV_ESCAPE_RE | PERL_PV_PRETTY_ELLIPSES | \
+ ( PERL_PV_PRETTY_QUOTE | PERL_PV_ESCAPE_RE | PERL_PV_ESCAPE_NONASCII | PERL_PV_PRETTY_ELLIPSES | \
((isuni) ? PERL_PV_ESCAPE_UNI : 0)) \
)
#endif /* DEBUG RELATED DEFINES */
-
-
+/*
+ * Local variables:
+ * c-indentation-style: bsd
+ * c-basic-offset: 4
+ * indent-tabs-mode: nil
+ * End:
+ *
+ * ex: set ts=8 sts=4 sw=4 et:
+ */
https://perl5.git.perl.org / perl5.git / blobdiff