[352] | 1 | /*
|
---|
| 2 | tre-internal.h - TRE internal definitions
|
---|
| 3 |
|
---|
| 4 | Copyright (c) 2001-2009 Ville Laurikari <vl@iki.fi>
|
---|
| 5 | All rights reserved.
|
---|
| 6 |
|
---|
| 7 | Redistribution and use in source and binary forms, with or without
|
---|
| 8 | modification, are permitted provided that the following conditions
|
---|
| 9 | are met:
|
---|
| 10 |
|
---|
| 11 | 1. Redistributions of source code must retain the above copyright
|
---|
| 12 | notice, this list of conditions and the following disclaimer.
|
---|
| 13 |
|
---|
| 14 | 2. Redistributions in binary form must reproduce the above copyright
|
---|
| 15 | notice, this list of conditions and the following disclaimer in the
|
---|
| 16 | documentation and/or other materials provided with the distribution.
|
---|
| 17 |
|
---|
| 18 | THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDER AND CONTRIBUTORS
|
---|
| 19 | ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
---|
| 20 | LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
---|
| 21 | A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
---|
| 22 | HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
---|
| 23 | SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
---|
| 24 | LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
---|
| 25 | DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
---|
| 26 | THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
---|
| 27 | (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
---|
| 28 | OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
---|
| 29 |
|
---|
| 30 | */
|
---|
| 31 |
|
---|
| 32 | #include <regex.h>
|
---|
| 33 | #include <wchar.h>
|
---|
| 34 | #include <wctype.h>
|
---|
| 35 |
|
---|
| 36 | #undef TRE_MBSTATE
|
---|
| 37 |
|
---|
| 38 | #define NDEBUG
|
---|
| 39 |
|
---|
| 40 | #define TRE_REGEX_T_FIELD __opaque
|
---|
| 41 | typedef int reg_errcode_t;
|
---|
| 42 |
|
---|
| 43 | typedef wchar_t tre_char_t;
|
---|
| 44 |
|
---|
| 45 | #define DPRINT(msg) do { } while(0)
|
---|
| 46 |
|
---|
| 47 | #define elementsof(x) ( sizeof(x) / sizeof(x[0]) )
|
---|
| 48 |
|
---|
| 49 | #define tre_mbrtowc(pwc, s, n, ps) (mbtowc((pwc), (s), (n)))
|
---|
| 50 |
|
---|
| 51 | /* Wide characters. */
|
---|
| 52 | typedef wint_t tre_cint_t;
|
---|
| 53 | #define TRE_CHAR_MAX 0x10ffff
|
---|
| 54 |
|
---|
| 55 | #define tre_isalnum iswalnum
|
---|
| 56 | #define tre_isalpha iswalpha
|
---|
| 57 | #define tre_isblank iswblank
|
---|
| 58 | #define tre_iscntrl iswcntrl
|
---|
| 59 | #define tre_isdigit iswdigit
|
---|
| 60 | #define tre_isgraph iswgraph
|
---|
| 61 | #define tre_islower iswlower
|
---|
| 62 | #define tre_isprint iswprint
|
---|
| 63 | #define tre_ispunct iswpunct
|
---|
| 64 | #define tre_isspace iswspace
|
---|
| 65 | #define tre_isupper iswupper
|
---|
| 66 | #define tre_isxdigit iswxdigit
|
---|
| 67 |
|
---|
| 68 | #define tre_tolower towlower
|
---|
| 69 | #define tre_toupper towupper
|
---|
| 70 | #define tre_strlen wcslen
|
---|
| 71 |
|
---|
| 72 | /* Use system provided iswctype() and wctype(). */
|
---|
| 73 | typedef wctype_t tre_ctype_t;
|
---|
| 74 | #define tre_isctype iswctype
|
---|
| 75 | #define tre_ctype wctype
|
---|
| 76 |
|
---|
| 77 | /* Returns number of bytes to add to (char *)ptr to make it
|
---|
| 78 | properly aligned for the type. */
|
---|
| 79 | #define ALIGN(ptr, type) \
|
---|
| 80 | ((((long)ptr) % sizeof(type)) \
|
---|
| 81 | ? (sizeof(type) - (((long)ptr) % sizeof(type))) \
|
---|
| 82 | : 0)
|
---|
| 83 |
|
---|
| 84 | #undef MAX
|
---|
| 85 | #undef MIN
|
---|
| 86 | #define MAX(a, b) (((a) >= (b)) ? (a) : (b))
|
---|
| 87 | #define MIN(a, b) (((a) <= (b)) ? (a) : (b))
|
---|
| 88 |
|
---|
| 89 | /* TNFA transition type. A TNFA state is an array of transitions,
|
---|
| 90 | the terminator is a transition with NULL `state'. */
|
---|
| 91 | typedef struct tnfa_transition tre_tnfa_transition_t;
|
---|
| 92 |
|
---|
| 93 | struct tnfa_transition {
|
---|
| 94 | /* Range of accepted characters. */
|
---|
| 95 | tre_cint_t code_min;
|
---|
| 96 | tre_cint_t code_max;
|
---|
| 97 | /* Pointer to the destination state. */
|
---|
| 98 | tre_tnfa_transition_t *state;
|
---|
| 99 | /* ID number of the destination state. */
|
---|
| 100 | int state_id;
|
---|
| 101 | /* -1 terminated array of tags (or NULL). */
|
---|
| 102 | int *tags;
|
---|
| 103 | /* Assertion bitmap. */
|
---|
| 104 | int assertions;
|
---|
| 105 | /* Assertion parameters. */
|
---|
| 106 | union {
|
---|
| 107 | /* Character class assertion. */
|
---|
| 108 | tre_ctype_t class;
|
---|
| 109 | /* Back reference assertion. */
|
---|
| 110 | int backref;
|
---|
| 111 | } u;
|
---|
| 112 | /* Negative character class assertions. */
|
---|
| 113 | tre_ctype_t *neg_classes;
|
---|
| 114 | };
|
---|
| 115 |
|
---|
| 116 |
|
---|
| 117 | /* Assertions. */
|
---|
| 118 | #define ASSERT_AT_BOL 1 /* Beginning of line. */
|
---|
| 119 | #define ASSERT_AT_EOL 2 /* End of line. */
|
---|
| 120 | #define ASSERT_CHAR_CLASS 4 /* Character class in `class'. */
|
---|
| 121 | #define ASSERT_CHAR_CLASS_NEG 8 /* Character classes in `neg_classes'. */
|
---|
| 122 | #define ASSERT_AT_BOW 16 /* Beginning of word. */
|
---|
| 123 | #define ASSERT_AT_EOW 32 /* End of word. */
|
---|
| 124 | #define ASSERT_AT_WB 64 /* Word boundary. */
|
---|
| 125 | #define ASSERT_AT_WB_NEG 128 /* Not a word boundary. */
|
---|
| 126 | #define ASSERT_BACKREF 256 /* A back reference in `backref'. */
|
---|
| 127 | #define ASSERT_LAST 256
|
---|
| 128 |
|
---|
| 129 | /* Tag directions. */
|
---|
| 130 | typedef enum {
|
---|
| 131 | TRE_TAG_MINIMIZE = 0,
|
---|
| 132 | TRE_TAG_MAXIMIZE = 1
|
---|
| 133 | } tre_tag_direction_t;
|
---|
| 134 |
|
---|
| 135 | /* Instructions to compute submatch register values from tag values
|
---|
| 136 | after a successful match. */
|
---|
| 137 | struct tre_submatch_data {
|
---|
| 138 | /* Tag that gives the value for rm_so (submatch start offset). */
|
---|
| 139 | int so_tag;
|
---|
| 140 | /* Tag that gives the value for rm_eo (submatch end offset). */
|
---|
| 141 | int eo_tag;
|
---|
| 142 | /* List of submatches this submatch is contained in. */
|
---|
| 143 | int *parents;
|
---|
| 144 | };
|
---|
| 145 |
|
---|
| 146 | typedef struct tre_submatch_data tre_submatch_data_t;
|
---|
| 147 |
|
---|
| 148 |
|
---|
| 149 | /* TNFA definition. */
|
---|
| 150 | typedef struct tnfa tre_tnfa_t;
|
---|
| 151 |
|
---|
| 152 | struct tnfa {
|
---|
| 153 | tre_tnfa_transition_t *transitions;
|
---|
| 154 | unsigned int num_transitions;
|
---|
| 155 | tre_tnfa_transition_t *initial;
|
---|
| 156 | tre_tnfa_transition_t *final;
|
---|
| 157 | tre_submatch_data_t *submatch_data;
|
---|
| 158 | char *firstpos_chars;
|
---|
| 159 | int first_char;
|
---|
| 160 | unsigned int num_submatches;
|
---|
| 161 | tre_tag_direction_t *tag_directions;
|
---|
| 162 | int *minimal_tags;
|
---|
| 163 | int num_tags;
|
---|
| 164 | int num_minimals;
|
---|
| 165 | int end_tag;
|
---|
| 166 | int num_states;
|
---|
| 167 | int cflags;
|
---|
| 168 | int have_backrefs;
|
---|
| 169 | int have_approx;
|
---|
| 170 | };
|
---|
| 171 |
|
---|
| 172 | /* from tre-mem.h: */
|
---|
| 173 |
|
---|
| 174 | #define TRE_MEM_BLOCK_SIZE 1024
|
---|
| 175 |
|
---|
| 176 | typedef struct tre_list {
|
---|
| 177 | void *data;
|
---|
| 178 | struct tre_list *next;
|
---|
| 179 | } tre_list_t;
|
---|
| 180 |
|
---|
| 181 | typedef struct tre_mem_struct {
|
---|
| 182 | tre_list_t *blocks;
|
---|
| 183 | tre_list_t *current;
|
---|
| 184 | char *ptr;
|
---|
| 185 | size_t n;
|
---|
| 186 | int failed;
|
---|
| 187 | void **provided;
|
---|
| 188 | } *tre_mem_t;
|
---|
| 189 |
|
---|
| 190 | #define tre_mem_new_impl __tre_mem_new_impl
|
---|
| 191 | #define tre_mem_alloc_impl __tre_mem_alloc_impl
|
---|
| 192 | #define tre_mem_destroy __tre_mem_destroy
|
---|
| 193 |
|
---|
| 194 | tre_mem_t tre_mem_new_impl(int provided, void *provided_block);
|
---|
| 195 | void *tre_mem_alloc_impl(tre_mem_t mem, int provided, void *provided_block,
|
---|
| 196 | int zero, size_t size);
|
---|
| 197 |
|
---|
| 198 | /* Returns a new memory allocator or NULL if out of memory. */
|
---|
| 199 | #define tre_mem_new() tre_mem_new_impl(0, NULL)
|
---|
| 200 |
|
---|
| 201 | /* Allocates a block of `size' bytes from `mem'. Returns a pointer to the
|
---|
| 202 | allocated block or NULL if an underlying malloc() failed. */
|
---|
| 203 | #define tre_mem_alloc(mem, size) tre_mem_alloc_impl(mem, 0, NULL, 0, size)
|
---|
| 204 |
|
---|
| 205 | /* Allocates a block of `size' bytes from `mem'. Returns a pointer to the
|
---|
| 206 | allocated block or NULL if an underlying malloc() failed. The memory
|
---|
| 207 | is set to zero. */
|
---|
| 208 | #define tre_mem_calloc(mem, size) tre_mem_alloc_impl(mem, 0, NULL, 1, size)
|
---|
| 209 |
|
---|
| 210 | #ifdef TRE_USE_ALLOCA
|
---|
| 211 | /* alloca() versions. Like above, but memory is allocated with alloca()
|
---|
| 212 | instead of malloc(). */
|
---|
| 213 |
|
---|
| 214 | #define tre_mem_newa() \
|
---|
| 215 | tre_mem_new_impl(1, alloca(sizeof(struct tre_mem_struct)))
|
---|
| 216 |
|
---|
| 217 | #define tre_mem_alloca(mem, size) \
|
---|
| 218 | ((mem)->n >= (size) \
|
---|
| 219 | ? tre_mem_alloc_impl((mem), 1, NULL, 0, (size)) \
|
---|
| 220 | : tre_mem_alloc_impl((mem), 1, alloca(TRE_MEM_BLOCK_SIZE), 0, (size)))
|
---|
| 221 | #endif /* TRE_USE_ALLOCA */
|
---|
| 222 |
|
---|
| 223 |
|
---|
| 224 | /* Frees the memory allocator and all memory allocated with it. */
|
---|
| 225 | void tre_mem_destroy(tre_mem_t mem);
|
---|
| 226 |
|
---|
| 227 | #define xmalloc malloc
|
---|
| 228 | #define xcalloc calloc
|
---|
| 229 | #define xfree free
|
---|
| 230 | #define xrealloc realloc
|
---|
| 231 |
|
---|