| 1 |
nigel |
77 |
/************************************************* |
| 2 |
|
|
* Perl-Compatible Regular Expressions * |
| 3 |
|
|
*************************************************/ |
| 4 |
|
|
|
| 5 |
|
|
/* PCRE is a library of functions to support regular expressions whose syntax |
| 6 |
|
|
and semantics are as close as possible to those of the Perl 5 language. |
| 7 |
|
|
|
| 8 |
|
|
Written by Philip Hazel |
| 9 |
nigel |
87 |
Copyright (c) 1997-2006 University of Cambridge |
| 10 |
nigel |
77 |
|
| 11 |
|
|
----------------------------------------------------------------------------- |
| 12 |
|
|
Redistribution and use in source and binary forms, with or without |
| 13 |
|
|
modification, are permitted provided that the following conditions are met: |
| 14 |
|
|
|
| 15 |
|
|
* Redistributions of source code must retain the above copyright notice, |
| 16 |
|
|
this list of conditions and the following disclaimer. |
| 17 |
|
|
|
| 18 |
|
|
* Redistributions in binary form must reproduce the above copyright |
| 19 |
|
|
notice, this list of conditions and the following disclaimer in the |
| 20 |
|
|
documentation and/or other materials provided with the distribution. |
| 21 |
|
|
|
| 22 |
|
|
* Neither the name of the University of Cambridge nor the names of its |
| 23 |
|
|
contributors may be used to endorse or promote products derived from |
| 24 |
|
|
this software without specific prior written permission. |
| 25 |
|
|
|
| 26 |
|
|
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" |
| 27 |
|
|
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
| 28 |
|
|
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE |
| 29 |
|
|
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE |
| 30 |
|
|
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR |
| 31 |
|
|
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF |
| 32 |
|
|
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS |
| 33 |
|
|
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN |
| 34 |
|
|
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) |
| 35 |
|
|
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE |
| 36 |
|
|
POSSIBILITY OF SUCH DAMAGE. |
| 37 |
|
|
----------------------------------------------------------------------------- |
| 38 |
|
|
*/ |
| 39 |
|
|
|
| 40 |
|
|
|
| 41 |
|
|
/* This module contains the external function pcre_maketables(), which builds |
| 42 |
|
|
character tables for PCRE in the current locale. The file is compiled on its |
| 43 |
|
|
own as part of the PCRE library. However, it is also included in the |
| 44 |
|
|
compilation of dftables.c, in which case the macro DFTABLES is defined. */ |
| 45 |
|
|
|
| 46 |
|
|
|
| 47 |
|
|
#ifndef DFTABLES |
| 48 |
|
|
#include "pcre_internal.h" |
| 49 |
|
|
#endif |
| 50 |
|
|
|
| 51 |
|
|
|
| 52 |
|
|
/************************************************* |
| 53 |
|
|
* Create PCRE character tables * |
| 54 |
|
|
*************************************************/ |
| 55 |
|
|
|
| 56 |
|
|
/* This function builds a set of character tables for use by PCRE and returns |
| 57 |
|
|
a pointer to them. They are build using the ctype functions, and consequently |
| 58 |
|
|
their contents will depend upon the current locale setting. When compiled as |
| 59 |
|
|
part of the library, the store is obtained via pcre_malloc(), but when compiled |
| 60 |
|
|
inside dftables, use malloc(). |
| 61 |
|
|
|
| 62 |
|
|
Arguments: none |
| 63 |
|
|
Returns: pointer to the contiguous block of data |
| 64 |
|
|
*/ |
| 65 |
|
|
|
| 66 |
|
|
const unsigned char * |
| 67 |
|
|
pcre_maketables(void) |
| 68 |
|
|
{ |
| 69 |
|
|
unsigned char *yield, *p; |
| 70 |
|
|
int i; |
| 71 |
|
|
|
| 72 |
|
|
#ifndef DFTABLES |
| 73 |
|
|
yield = (unsigned char*)(pcre_malloc)(tables_length); |
| 74 |
|
|
#else |
| 75 |
|
|
yield = (unsigned char*)malloc(tables_length); |
| 76 |
|
|
#endif |
| 77 |
|
|
|
| 78 |
|
|
if (yield == NULL) return NULL; |
| 79 |
|
|
p = yield; |
| 80 |
|
|
|
| 81 |
|
|
/* First comes the lower casing table */ |
| 82 |
|
|
|
| 83 |
|
|
for (i = 0; i < 256; i++) *p++ = tolower(i); |
| 84 |
|
|
|
| 85 |
|
|
/* Next the case-flipping table */ |
| 86 |
|
|
|
| 87 |
|
|
for (i = 0; i < 256; i++) *p++ = islower(i)? toupper(i) : tolower(i); |
| 88 |
|
|
|
| 89 |
nigel |
87 |
/* Then the character class tables. Don't try to be clever and save effort on |
| 90 |
|
|
exclusive ones - in some locales things may be different. Note that the table |
| 91 |
|
|
for "space" includes everything "isspace" gives, including VT in the default |
| 92 |
|
|
locale. This makes it work for the POSIX class [:space:]. Note also that it is |
| 93 |
|
|
possible for a character to be alnum or alpha without being lower or upper, |
| 94 |
|
|
such as "male and female ordinals" (\xAA and \xBA) in the fr_FR locale (at |
| 95 |
|
|
least under Debian Linux's locales as of 12/2005). So we must test for alnum |
| 96 |
|
|
specially. */ |
| 97 |
nigel |
77 |
|
| 98 |
|
|
memset(p, 0, cbit_length); |
| 99 |
|
|
for (i = 0; i < 256; i++) |
| 100 |
|
|
{ |
| 101 |
nigel |
87 |
if (isdigit(i)) p[cbit_digit + i/8] |= 1 << (i&7); |
| 102 |
|
|
if (isupper(i)) p[cbit_upper + i/8] |= 1 << (i&7); |
| 103 |
|
|
if (islower(i)) p[cbit_lower + i/8] |= 1 << (i&7); |
| 104 |
|
|
if (isalnum(i)) p[cbit_word + i/8] |= 1 << (i&7); |
| 105 |
nigel |
77 |
if (i == '_') p[cbit_word + i/8] |= 1 << (i&7); |
| 106 |
|
|
if (isspace(i)) p[cbit_space + i/8] |= 1 << (i&7); |
| 107 |
|
|
if (isxdigit(i))p[cbit_xdigit + i/8] |= 1 << (i&7); |
| 108 |
|
|
if (isgraph(i)) p[cbit_graph + i/8] |= 1 << (i&7); |
| 109 |
|
|
if (isprint(i)) p[cbit_print + i/8] |= 1 << (i&7); |
| 110 |
|
|
if (ispunct(i)) p[cbit_punct + i/8] |= 1 << (i&7); |
| 111 |
|
|
if (iscntrl(i)) p[cbit_cntrl + i/8] |= 1 << (i&7); |
| 112 |
|
|
} |
| 113 |
|
|
p += cbit_length; |
| 114 |
|
|
|
| 115 |
|
|
/* Finally, the character type table. In this, we exclude VT from the white |
| 116 |
|
|
space chars, because Perl doesn't recognize it as such for \s and for comments |
| 117 |
|
|
within regexes. */ |
| 118 |
|
|
|
| 119 |
|
|
for (i = 0; i < 256; i++) |
| 120 |
|
|
{ |
| 121 |
|
|
int x = 0; |
| 122 |
|
|
if (i != 0x0b && isspace(i)) x += ctype_space; |
| 123 |
|
|
if (isalpha(i)) x += ctype_letter; |
| 124 |
|
|
if (isdigit(i)) x += ctype_digit; |
| 125 |
|
|
if (isxdigit(i)) x += ctype_xdigit; |
| 126 |
|
|
if (isalnum(i) || i == '_') x += ctype_word; |
| 127 |
|
|
|
| 128 |
|
|
/* Note: strchr includes the terminating zero in the characters it considers. |
| 129 |
|
|
In this instance, that is ok because we want binary zero to be flagged as a |
| 130 |
|
|
meta-character, which in this sense is any character that terminates a run |
| 131 |
|
|
of data characters. */ |
| 132 |
|
|
|
| 133 |
|
|
if (strchr("*+?{^.$|()[", i) != 0) x += ctype_meta; *p++ = x; } |
| 134 |
|
|
|
| 135 |
|
|
return yield; |
| 136 |
|
|
} |
| 137 |
|
|
|
| 138 |
|
|
/* End of pcre_maketables.c */ |