123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902 |
- #ifndef _REGEX_INTERNAL_H
- #define _REGEX_INTERNAL_H 1
- #include <assert.h>
- #include <ctype.h>
- #include <stdio.h>
- #include <stdlib.h>
- #include <string.h>
- #include <langinfo.h>
- #include <locale.h>
- #include <wchar.h>
- #include <wctype.h>
- #include <stdbool.h>
- #include <stdint.h>
- #include "intprops.h"
- #ifdef _LIBC
- # include <libc-lock.h>
- # define lock_define(name) __libc_lock_define (, name)
- # define lock_init(lock) (__libc_lock_init (lock), 0)
- # define lock_fini(lock) 0
- # define lock_lock(lock) __libc_lock_lock (lock)
- # define lock_unlock(lock) __libc_lock_unlock (lock)
- #elif defined GNULIB_LOCK && !defined USE_UNLOCKED_IO
- # include "glthread/lock.h"
-
- # if ((defined __GNUC__ && !defined __STRICT_ANSI__) \
- || (defined __STDC_VERSION__ && 199901L <= __STDC_VERSION__))
- # define lock_define(name) gl_lock_define (, name)
- # elif USE_POSIX_THREADS
- # define lock_define(name) pthread_mutex_t name;
- # elif USE_PTH_THREADS
- # define lock_define(name) pth_mutex_t name;
- # elif USE_SOLARIS_THREADS
- # define lock_define(name) mutex_t name;
- # elif USE_WINDOWS_THREADS
- # define lock_define(name) gl_lock_t name;
- # else
- # define lock_define(name)
- # endif
- # define lock_init(lock) glthread_lock_init (&(lock))
- # define lock_fini(lock) glthread_lock_destroy (&(lock))
- # define lock_lock(lock) glthread_lock_lock (&(lock))
- # define lock_unlock(lock) glthread_lock_unlock (&(lock))
- #elif defined GNULIB_PTHREAD && !defined USE_UNLOCKED_IO
- # include <pthread.h>
- # define lock_define(name) pthread_mutex_t name;
- # define lock_init(lock) pthread_mutex_init (&(lock), 0)
- # define lock_fini(lock) pthread_mutex_destroy (&(lock))
- # define lock_lock(lock) pthread_mutex_lock (&(lock))
- # define lock_unlock(lock) pthread_mutex_unlock (&(lock))
- #else
- # define lock_define(name)
- # define lock_init(lock) 0
- # define lock_fini(lock) ((void) 0)
-
- # define lock_lock(lock) ((void) dfa)
- # define lock_unlock(lock) ((void) 0)
- #endif
- #if !defined _LIBC && ! (defined isblank || (HAVE_ISBLANK && HAVE_DECL_ISBLANK))
- # define isblank(ch) ((ch) == ' ' || (ch) == '\t')
- #endif
- #ifdef _LIBC
- # ifndef _RE_DEFINE_LOCALE_FUNCTIONS
- # define _RE_DEFINE_LOCALE_FUNCTIONS 1
- # include <locale/localeinfo.h>
- # include <locale/coll-lookup.h>
- # endif
- #endif
- #if (HAVE_LIBINTL_H && ENABLE_NLS) || defined _LIBC
- # include <libintl.h>
- # ifdef _LIBC
- # undef gettext
- # define gettext(msgid) \
- __dcgettext (_libc_intl_domainname, msgid, LC_MESSAGES)
- # endif
- #else
- # define gettext(msgid) (msgid)
- #endif
- #ifndef gettext_noop
- # define gettext_noop(String) String
- #endif
- #if (defined MB_CUR_MAX && HAVE_WCTYPE_H && HAVE_ISWCTYPE) || _LIBC
- # define RE_ENABLE_I18N
- #endif
- #define BE(expr, val) __builtin_expect (expr, val)
- #define ASCII_CHARS 0x80
- #define SBC_MAX (UCHAR_MAX + 1)
- #define COLL_ELEM_LEN_MAX 8
- #define NEWLINE_CHAR '\n'
- #define WIDE_NEWLINE_CHAR L'\n'
- #ifndef _LIBC
- # undef __wctype
- # undef __iswctype
- # define __wctype wctype
- # define __iswalnum iswalnum
- # define __iswctype iswctype
- # define __towlower towlower
- # define __towupper towupper
- # define __btowc btowc
- # define __mbrtowc mbrtowc
- # define __wcrtomb wcrtomb
- # define __regfree regfree
- # define attribute_hidden
- #endif /* not _LIBC */
- #if __GNUC__ < 3 + (__GNUC_MINOR__ < 1)
- # define __attribute__(arg)
- #endif
- #ifndef SSIZE_MAX
- # define SSIZE_MAX ((ssize_t) (SIZE_MAX / 2))
- #endif
- typedef regoff_t Idx;
- #ifdef _REGEX_LARGE_OFFSETS
- # define IDX_MAX SSIZE_MAX
- #else
- # define IDX_MAX INT_MAX
- #endif
- typedef __re_size_t re_hashval_t;
- typedef unsigned long int bitset_word_t;
- #define BITSET_WORD_MAX ULONG_MAX
- #if BITSET_WORD_MAX == 0xffffffffUL
- # define BITSET_WORD_BITS 32
- #elif BITSET_WORD_MAX >> 31 >> 4 == 1
- # define BITSET_WORD_BITS 36
- #elif BITSET_WORD_MAX >> 31 >> 16 == 1
- # define BITSET_WORD_BITS 48
- #elif BITSET_WORD_MAX >> 31 >> 28 == 1
- # define BITSET_WORD_BITS 60
- #elif BITSET_WORD_MAX >> 31 >> 31 >> 1 == 1
- # define BITSET_WORD_BITS 64
- #elif BITSET_WORD_MAX >> 31 >> 31 >> 9 == 1
- # define BITSET_WORD_BITS 72
- #elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 3 == 1
- # define BITSET_WORD_BITS 128
- #elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 == 1
- # define BITSET_WORD_BITS 256
- #elif BITSET_WORD_MAX >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 31 >> 7 > 1
- # define BITSET_WORD_BITS 257 /* any value > SBC_MAX will do here */
- # if BITSET_WORD_BITS <= SBC_MAX
- # error "Invalid SBC_MAX"
- # endif
- #else
- # error "Add case for new bitset_word_t size"
- #endif
- #define BITSET_WORDS ((SBC_MAX + BITSET_WORD_BITS - 1) / BITSET_WORD_BITS)
- typedef bitset_word_t bitset_t[BITSET_WORDS];
- typedef bitset_word_t *re_bitset_ptr_t;
- typedef const bitset_word_t *re_const_bitset_ptr_t;
- #define PREV_WORD_CONSTRAINT 0x0001
- #define PREV_NOTWORD_CONSTRAINT 0x0002
- #define NEXT_WORD_CONSTRAINT 0x0004
- #define NEXT_NOTWORD_CONSTRAINT 0x0008
- #define PREV_NEWLINE_CONSTRAINT 0x0010
- #define NEXT_NEWLINE_CONSTRAINT 0x0020
- #define PREV_BEGBUF_CONSTRAINT 0x0040
- #define NEXT_ENDBUF_CONSTRAINT 0x0080
- #define WORD_DELIM_CONSTRAINT 0x0100
- #define NOT_WORD_DELIM_CONSTRAINT 0x0200
- typedef enum
- {
- INSIDE_WORD = PREV_WORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
- WORD_FIRST = PREV_NOTWORD_CONSTRAINT | NEXT_WORD_CONSTRAINT,
- WORD_LAST = PREV_WORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
- INSIDE_NOTWORD = PREV_NOTWORD_CONSTRAINT | NEXT_NOTWORD_CONSTRAINT,
- LINE_FIRST = PREV_NEWLINE_CONSTRAINT,
- LINE_LAST = NEXT_NEWLINE_CONSTRAINT,
- BUF_FIRST = PREV_BEGBUF_CONSTRAINT,
- BUF_LAST = NEXT_ENDBUF_CONSTRAINT,
- WORD_DELIM = WORD_DELIM_CONSTRAINT,
- NOT_WORD_DELIM = NOT_WORD_DELIM_CONSTRAINT
- } re_context_type;
- typedef struct
- {
- Idx alloc;
- Idx nelem;
- Idx *elems;
- } re_node_set;
- typedef enum
- {
- NON_TYPE = 0,
-
- CHARACTER = 1,
- END_OF_RE = 2,
- SIMPLE_BRACKET = 3,
- OP_BACK_REF = 4,
- OP_PERIOD = 5,
- #ifdef RE_ENABLE_I18N
- COMPLEX_BRACKET = 6,
- OP_UTF8_PERIOD = 7,
- #endif /* RE_ENABLE_I18N */
-
- #define EPSILON_BIT 8
- OP_OPEN_SUBEXP = EPSILON_BIT | 0,
- OP_CLOSE_SUBEXP = EPSILON_BIT | 1,
- OP_ALT = EPSILON_BIT | 2,
- OP_DUP_ASTERISK = EPSILON_BIT | 3,
- ANCHOR = EPSILON_BIT | 4,
-
- CONCAT = 16,
- SUBEXP = 17,
-
- OP_DUP_PLUS = 18,
- OP_DUP_QUESTION,
- OP_OPEN_BRACKET,
- OP_CLOSE_BRACKET,
- OP_CHARSET_RANGE,
- OP_OPEN_DUP_NUM,
- OP_CLOSE_DUP_NUM,
- OP_NON_MATCH_LIST,
- OP_OPEN_COLL_ELEM,
- OP_CLOSE_COLL_ELEM,
- OP_OPEN_EQUIV_CLASS,
- OP_CLOSE_EQUIV_CLASS,
- OP_OPEN_CHAR_CLASS,
- OP_CLOSE_CHAR_CLASS,
- OP_WORD,
- OP_NOTWORD,
- OP_SPACE,
- OP_NOTSPACE,
- BACK_SLASH
- } re_token_type_t;
- #ifdef RE_ENABLE_I18N
- typedef struct
- {
-
- wchar_t *mbchars;
-
- # ifdef _LIBC
- int32_t *coll_syms;
- # endif
-
- # ifdef _LIBC
- int32_t *equiv_classes;
- # endif
-
- # ifdef _LIBC
- uint32_t *range_starts;
- uint32_t *range_ends;
- # else /* not _LIBC */
- wchar_t *range_starts;
- wchar_t *range_ends;
- # endif /* not _LIBC */
-
- wctype_t *char_classes;
-
- unsigned int non_match : 1;
-
- Idx nmbchars;
-
- Idx ncoll_syms;
-
- Idx nequiv_classes;
-
- Idx nranges;
-
- Idx nchar_classes;
- } re_charset_t;
- #endif /* RE_ENABLE_I18N */
- typedef struct
- {
- union
- {
- unsigned char c;
- re_bitset_ptr_t sbcset;
- #ifdef RE_ENABLE_I18N
- re_charset_t *mbcset;
- #endif /* RE_ENABLE_I18N */
- Idx idx;
- re_context_type ctx_type;
- } opr;
- #if __GNUC__ >= 2 && !defined __STRICT_ANSI__
- re_token_type_t type : 8;
- #else
- re_token_type_t type;
- #endif
- unsigned int constraint : 10;
- unsigned int duplicated : 1;
- unsigned int opt_subexp : 1;
- #ifdef RE_ENABLE_I18N
- unsigned int accept_mb : 1;
-
- unsigned int mb_partial : 1;
- #endif
- unsigned int word_char : 1;
- } re_token_t;
- #define IS_EPSILON_NODE(type) ((type) & EPSILON_BIT)
- struct re_string_t
- {
-
- const unsigned char *raw_mbs;
-
- unsigned char *mbs;
- #ifdef RE_ENABLE_I18N
-
- wint_t *wcs;
- Idx *offsets;
- mbstate_t cur_state;
- #endif
-
- Idx raw_mbs_idx;
-
- Idx valid_len;
-
- Idx valid_raw_len;
-
- Idx bufs_len;
-
- Idx cur_idx;
-
- Idx raw_len;
-
- Idx len;
-
- Idx raw_stop;
-
- Idx stop;
-
- unsigned int tip_context;
-
- RE_TRANSLATE_TYPE trans;
-
- re_const_bitset_ptr_t word_char;
-
- unsigned char icase;
- unsigned char is_utf8;
- unsigned char map_notascii;
- unsigned char mbs_allocated;
- unsigned char offsets_needed;
- unsigned char newline_anchor;
- unsigned char word_ops_used;
- int mb_cur_max;
- };
- typedef struct re_string_t re_string_t;
- struct re_dfa_t;
- typedef struct re_dfa_t re_dfa_t;
- #ifndef _LIBC
- # define internal_function
- # define IS_IN(libc) false
- #endif
- static reg_errcode_t re_string_realloc_buffers (re_string_t *pstr,
- Idx new_buf_len)
- internal_function;
- #ifdef RE_ENABLE_I18N
- static void build_wcs_buffer (re_string_t *pstr) internal_function;
- static reg_errcode_t build_wcs_upper_buffer (re_string_t *pstr)
- internal_function;
- #endif /* RE_ENABLE_I18N */
- static void build_upper_buffer (re_string_t *pstr) internal_function;
- static void re_string_translate_buffer (re_string_t *pstr) internal_function;
- static unsigned int re_string_context_at (const re_string_t *input, Idx idx,
- int eflags)
- internal_function __attribute__ ((pure));
- #define re_string_peek_byte(pstr, offset) \
- ((pstr)->mbs[(pstr)->cur_idx + offset])
- #define re_string_fetch_byte(pstr) \
- ((pstr)->mbs[(pstr)->cur_idx++])
- #define re_string_first_byte(pstr, idx) \
- ((idx) == (pstr)->valid_len || (pstr)->wcs[idx] != WEOF)
- #define re_string_is_single_byte_char(pstr, idx) \
- ((pstr)->wcs[idx] != WEOF && ((pstr)->valid_len == (idx) + 1 \
- || (pstr)->wcs[(idx) + 1] != WEOF))
- #define re_string_eoi(pstr) ((pstr)->stop <= (pstr)->cur_idx)
- #define re_string_cur_idx(pstr) ((pstr)->cur_idx)
- #define re_string_get_buffer(pstr) ((pstr)->mbs)
- #define re_string_length(pstr) ((pstr)->len)
- #define re_string_byte_at(pstr,idx) ((pstr)->mbs[idx])
- #define re_string_skip_bytes(pstr,idx) ((pstr)->cur_idx += (idx))
- #define re_string_set_index(pstr,idx) ((pstr)->cur_idx = (idx))
- #if defined _LIBC || HAVE_ALLOCA
- # include <alloca.h>
- #endif
- #ifndef _LIBC
- # if HAVE_ALLOCA
- # define __libc_use_alloca(n) ((n) < 4032)
- # else
- # define __libc_use_alloca(n) 0
- # undef alloca
- # define alloca(n) malloc (n)
- # endif
- #endif
- #ifdef _LIBC
- # define MALLOC_0_IS_NONNULL 1
- #elif !defined MALLOC_0_IS_NONNULL
- # define MALLOC_0_IS_NONNULL 0
- #endif
- #ifndef MAX
- # define MAX(a,b) ((a) < (b) ? (b) : (a))
- #endif
- #ifndef MIN
- # define MIN(a,b) ((a) < (b) ? (a) : (b))
- #endif
- #define re_malloc(t,n) ((t *) malloc ((n) * sizeof (t)))
- #define re_realloc(p,t,n) ((t *) realloc (p, (n) * sizeof (t)))
- #define re_free(p) free (p)
- struct bin_tree_t
- {
- struct bin_tree_t *parent;
- struct bin_tree_t *left;
- struct bin_tree_t *right;
- struct bin_tree_t *first;
- struct bin_tree_t *next;
- re_token_t token;
-
- Idx node_idx;
- };
- typedef struct bin_tree_t bin_tree_t;
- #define BIN_TREE_STORAGE_SIZE \
- ((1024 - sizeof (void *)) / sizeof (bin_tree_t))
- struct bin_tree_storage_t
- {
- struct bin_tree_storage_t *next;
- bin_tree_t data[BIN_TREE_STORAGE_SIZE];
- };
- typedef struct bin_tree_storage_t bin_tree_storage_t;
- #define CONTEXT_WORD 1
- #define CONTEXT_NEWLINE (CONTEXT_WORD << 1)
- #define CONTEXT_BEGBUF (CONTEXT_NEWLINE << 1)
- #define CONTEXT_ENDBUF (CONTEXT_BEGBUF << 1)
- #define IS_WORD_CONTEXT(c) ((c) & CONTEXT_WORD)
- #define IS_NEWLINE_CONTEXT(c) ((c) & CONTEXT_NEWLINE)
- #define IS_BEGBUF_CONTEXT(c) ((c) & CONTEXT_BEGBUF)
- #define IS_ENDBUF_CONTEXT(c) ((c) & CONTEXT_ENDBUF)
- #define IS_ORDINARY_CONTEXT(c) ((c) == 0)
- #define IS_WORD_CHAR(ch) (isalnum (ch) || (ch) == '_')
- #define IS_NEWLINE(ch) ((ch) == NEWLINE_CHAR)
- #define IS_WIDE_WORD_CHAR(ch) (__iswalnum (ch) || (ch) == L'_')
- #define IS_WIDE_NEWLINE(ch) ((ch) == WIDE_NEWLINE_CHAR)
- #define NOT_SATISFY_PREV_CONSTRAINT(constraint,context) \
- ((((constraint) & PREV_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
- || ((constraint & PREV_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
- || ((constraint & PREV_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context))\
- || ((constraint & PREV_BEGBUF_CONSTRAINT) && !IS_BEGBUF_CONTEXT (context)))
- #define NOT_SATISFY_NEXT_CONSTRAINT(constraint,context) \
- ((((constraint) & NEXT_WORD_CONSTRAINT) && !IS_WORD_CONTEXT (context)) \
- || (((constraint) & NEXT_NOTWORD_CONSTRAINT) && IS_WORD_CONTEXT (context)) \
- || (((constraint) & NEXT_NEWLINE_CONSTRAINT) && !IS_NEWLINE_CONTEXT (context)) \
- || (((constraint) & NEXT_ENDBUF_CONSTRAINT) && !IS_ENDBUF_CONTEXT (context)))
- struct re_dfastate_t
- {
- re_hashval_t hash;
- re_node_set nodes;
- re_node_set non_eps_nodes;
- re_node_set inveclosure;
- re_node_set *entrance_nodes;
- struct re_dfastate_t **trtable, **word_trtable;
- unsigned int context : 4;
- unsigned int halt : 1;
-
- unsigned int accept_mb : 1;
-
- unsigned int has_backref : 1;
- unsigned int has_constraint : 1;
- };
- typedef struct re_dfastate_t re_dfastate_t;
- struct re_state_table_entry
- {
- Idx num;
- Idx alloc;
- re_dfastate_t **array;
- };
- typedef struct
- {
- Idx next_idx;
- Idx alloc;
- re_dfastate_t **array;
- } state_array_t;
- typedef struct
- {
- Idx node;
- Idx str_idx;
- state_array_t path;
- } re_sub_match_last_t;
- typedef struct
- {
- Idx str_idx;
- Idx node;
- state_array_t *path;
- Idx alasts;
- Idx nlasts;
- re_sub_match_last_t **lasts;
- } re_sub_match_top_t;
- struct re_backref_cache_entry
- {
- Idx node;
- Idx str_idx;
- Idx subexp_from;
- Idx subexp_to;
- char more;
- char unused;
- unsigned short int eps_reachable_subexps_map;
- };
- typedef struct
- {
-
- re_string_t input;
- #if defined _LIBC || (defined __STDC_VERSION__ && __STDC_VERSION__ >= 199901L)
- const re_dfa_t *const dfa;
- #else
- const re_dfa_t *dfa;
- #endif
-
- int eflags;
-
- Idx match_last;
- Idx last_node;
-
- re_dfastate_t **state_log;
- Idx state_log_top;
-
- Idx nbkref_ents;
- Idx abkref_ents;
- struct re_backref_cache_entry *bkref_ents;
- int max_mb_elem_len;
- Idx nsub_tops;
- Idx asub_tops;
- re_sub_match_top_t **sub_tops;
- } re_match_context_t;
- typedef struct
- {
- re_dfastate_t **sifted_states;
- re_dfastate_t **limited_states;
- Idx last_node;
- Idx last_str_idx;
- re_node_set limits;
- } re_sift_context_t;
- struct re_fail_stack_ent_t
- {
- Idx idx;
- Idx node;
- regmatch_t *regs;
- re_node_set eps_via_nodes;
- };
- struct re_fail_stack_t
- {
- Idx num;
- Idx alloc;
- struct re_fail_stack_ent_t *stack;
- };
- struct re_dfa_t
- {
- re_token_t *nodes;
- size_t nodes_alloc;
- size_t nodes_len;
- Idx *nexts;
- Idx *org_indices;
- re_node_set *edests;
- re_node_set *eclosures;
- re_node_set *inveclosures;
- struct re_state_table_entry *state_table;
- re_dfastate_t *init_state;
- re_dfastate_t *init_state_word;
- re_dfastate_t *init_state_nl;
- re_dfastate_t *init_state_begbuf;
- bin_tree_t *str_tree;
- bin_tree_storage_t *str_tree_storage;
- re_bitset_ptr_t sb_char;
- int str_tree_storage_idx;
-
- re_hashval_t state_hash_mask;
- Idx init_node;
- Idx nbackref;
-
- bitset_word_t used_bkref_map;
- bitset_word_t completed_bkref_map;
- unsigned int has_plural_match : 1;
-
- unsigned int has_mb_node : 1;
- unsigned int is_utf8 : 1;
- unsigned int map_notascii : 1;
- unsigned int word_ops_used : 1;
- int mb_cur_max;
- bitset_t word_char;
- reg_syntax_t syntax;
- Idx *subexp_map;
- #ifdef DEBUG
- char* re_str;
- #endif
- lock_define (lock)
- };
- #define re_node_set_init_empty(set) memset (set, '\0', sizeof (re_node_set))
- #define re_node_set_remove(set,id) \
- (re_node_set_remove_at (set, re_node_set_contains (set, id) - 1))
- #define re_node_set_empty(p) ((p)->nelem = 0)
- #define re_node_set_free(set) re_free ((set)->elems)
- typedef enum
- {
- SB_CHAR,
- MB_CHAR,
- EQUIV_CLASS,
- COLL_SYM,
- CHAR_CLASS
- } bracket_elem_type;
- typedef struct
- {
- bracket_elem_type type;
- union
- {
- unsigned char ch;
- unsigned char *name;
- wchar_t wch;
- } opr;
- } bracket_elem_t;
- static void
- bitset_set (bitset_t set, Idx i)
- {
- set[i / BITSET_WORD_BITS] |= (bitset_word_t) 1 << i % BITSET_WORD_BITS;
- }
- static void
- bitset_clear (bitset_t set, Idx i)
- {
- set[i / BITSET_WORD_BITS] &= ~ ((bitset_word_t) 1 << i % BITSET_WORD_BITS);
- }
- static bool
- bitset_contain (const bitset_t set, Idx i)
- {
- return (set[i / BITSET_WORD_BITS] >> i % BITSET_WORD_BITS) & 1;
- }
- static void
- bitset_empty (bitset_t set)
- {
- memset (set, '\0', sizeof (bitset_t));
- }
- static void
- bitset_set_all (bitset_t set)
- {
- memset (set, -1, sizeof (bitset_word_t) * (SBC_MAX / BITSET_WORD_BITS));
- if (SBC_MAX % BITSET_WORD_BITS != 0)
- set[BITSET_WORDS - 1] =
- ((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1;
- }
- static void
- bitset_copy (bitset_t dest, const bitset_t src)
- {
- memcpy (dest, src, sizeof (bitset_t));
- }
- static void __attribute__ ((unused))
- bitset_not (bitset_t set)
- {
- int bitset_i;
- for (bitset_i = 0; bitset_i < SBC_MAX / BITSET_WORD_BITS; ++bitset_i)
- set[bitset_i] = ~set[bitset_i];
- if (SBC_MAX % BITSET_WORD_BITS != 0)
- set[BITSET_WORDS - 1] =
- ((((bitset_word_t) 1 << SBC_MAX % BITSET_WORD_BITS) - 1)
- & ~set[BITSET_WORDS - 1]);
- }
- static void __attribute__ ((unused))
- bitset_merge (bitset_t dest, const bitset_t src)
- {
- int bitset_i;
- for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
- dest[bitset_i] |= src[bitset_i];
- }
- static void __attribute__ ((unused))
- bitset_mask (bitset_t dest, const bitset_t src)
- {
- int bitset_i;
- for (bitset_i = 0; bitset_i < BITSET_WORDS; ++bitset_i)
- dest[bitset_i] &= src[bitset_i];
- }
- #ifdef RE_ENABLE_I18N
- static int
- internal_function __attribute__ ((pure, unused))
- re_string_char_size_at (const re_string_t *pstr, Idx idx)
- {
- int byte_idx;
- if (pstr->mb_cur_max == 1)
- return 1;
- for (byte_idx = 1; idx + byte_idx < pstr->valid_len; ++byte_idx)
- if (pstr->wcs[idx + byte_idx] != WEOF)
- break;
- return byte_idx;
- }
- static wint_t
- internal_function __attribute__ ((pure, unused))
- re_string_wchar_at (const re_string_t *pstr, Idx idx)
- {
- if (pstr->mb_cur_max == 1)
- return (wint_t) pstr->mbs[idx];
- return (wint_t) pstr->wcs[idx];
- }
- # ifdef _LIBC
- # include <locale/weight.h>
- # endif
- static int
- internal_function __attribute__ ((pure, unused))
- re_string_elem_size_at (const re_string_t *pstr, Idx idx)
- {
- # ifdef _LIBC
- const unsigned char *p, *extra;
- const int32_t *table, *indirect;
- uint_fast32_t nrules = _NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES);
- if (nrules != 0)
- {
- table = (const int32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_TABLEMB);
- extra = (const unsigned char *)
- _NL_CURRENT (LC_COLLATE, _NL_COLLATE_EXTRAMB);
- indirect = (const int32_t *) _NL_CURRENT (LC_COLLATE,
- _NL_COLLATE_INDIRECTMB);
- p = pstr->mbs + idx;
- findidx (table, indirect, extra, &p, pstr->len - idx);
- return p - pstr->mbs - idx;
- }
- else
- # endif /* _LIBC */
- return 1;
- }
- #endif /* RE_ENABLE_I18N */
- #ifndef __GNUC_PREREQ
- # if defined __GNUC__ && defined __GNUC_MINOR__
- # define __GNUC_PREREQ(maj, min) \
- ((__GNUC__ << 16) + __GNUC_MINOR__ >= ((maj) << 16) + (min))
- # else
- # define __GNUC_PREREQ(maj, min) 0
- # endif
- #endif
- #if __GNUC_PREREQ (3,4)
- # undef __attribute_warn_unused_result__
- # define __attribute_warn_unused_result__ \
- __attribute__ ((__warn_unused_result__))
- #else
- # define __attribute_warn_unused_result__ /* empty */
- #endif
- #endif /* _REGEX_INTERNAL_H */
|