1 | /* |
2 | * Copyright (c) 1998, 1999 Henry Spencer. All rights reserved. |
3 | * |
4 | * Development of this software was funded, in part, by Cray Research Inc., |
5 | * UUNET Communications Services Inc., Sun Microsystems Inc., and Scriptics |
6 | * Corporation, none of whom are responsible for the results. The author |
7 | * thanks all of them. |
8 | * |
9 | * Redistribution and use in source and binary forms -- with or without |
10 | * modification -- are permitted for any purpose, provided that |
11 | * redistributions in source form retain this entire copyright notice and |
12 | * indicate the origin and nature of any modifications. |
13 | * |
14 | * I'd appreciate being given credit for this package in the documentation |
15 | * of software which uses it, but that is not a requirement. |
16 | * |
17 | * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, |
18 | * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY |
19 | * AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL |
20 | * HENRY SPENCER BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, |
21 | * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, |
22 | * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; |
23 | * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, |
24 | * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR |
25 | * OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF |
26 | * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
27 | * |
28 | * src/include/regex/regcustom.h |
29 | */ |
30 | |
31 | /* headers if any */ |
32 | |
33 | /* |
34 | * It's against Postgres coding conventions to include postgres.h in a |
35 | * header file, but we allow the violation here because the regexp library |
36 | * files specifically intend this file to supply application-dependent |
37 | * headers, and are careful to include this file before anything else. |
38 | */ |
39 | #include "postgres.h" |
40 | |
41 | #include <ctype.h> |
42 | #include <limits.h> |
43 | |
44 | /* |
45 | * towlower() and friends should be in <wctype.h>, but some pre-C99 systems |
46 | * declare them in <wchar.h>. |
47 | */ |
48 | #ifdef HAVE_WCHAR_H |
49 | #include <wchar.h> |
50 | #endif |
51 | #ifdef HAVE_WCTYPE_H |
52 | #include <wctype.h> |
53 | #endif |
54 | |
55 | #include "mb/pg_wchar.h" |
56 | |
57 | #include "miscadmin.h" /* needed by rcancelrequested/rstacktoodeep */ |
58 | |
59 | |
60 | /* overrides for regguts.h definitions, if any */ |
61 | #define FUNCPTR(name, args) (*name) args |
62 | #define MALLOC(n) malloc(n) |
63 | #define FREE(p) free(VS(p)) |
64 | #define REALLOC(p,n) realloc(VS(p),n) |
65 | #define assert(x) Assert(x) |
66 | |
67 | /* internal character type and related */ |
68 | typedef pg_wchar chr; /* the type itself */ |
69 | typedef unsigned uchr; /* unsigned type that will hold a chr */ |
70 | |
71 | #define CHR(c) ((unsigned char) (c)) /* turn char literal into chr literal */ |
72 | #define DIGITVAL(c) ((c)-'0') /* turn chr digit into its value */ |
73 | #define CHRBITS 32 /* bits in a chr; must not use sizeof */ |
74 | #define CHR_MIN 0x00000000 /* smallest and largest chr; the value */ |
75 | #define CHR_MAX 0x7ffffffe /* CHR_MAX-CHR_MIN+1 must fit in an int, and |
76 | * CHR_MAX+1 must fit in a chr variable */ |
77 | |
78 | /* |
79 | * Check if a chr value is in range. Ideally we'd just write this as |
80 | * ((c) >= CHR_MIN && (c) <= CHR_MAX) |
81 | * However, if chr is unsigned and CHR_MIN is zero, the first part of that |
82 | * is a no-op, and certain overly-nannyish compilers give warnings about it. |
83 | * So we leave that out here. If you want to make chr signed and/or CHR_MIN |
84 | * not zero, redefine this macro as above. Callers should assume that the |
85 | * macro may multiply evaluate its argument, even though it does not today. |
86 | */ |
87 | #define CHR_IS_IN_RANGE(c) ((c) <= CHR_MAX) |
88 | |
89 | /* |
90 | * MAX_SIMPLE_CHR is the cutoff between "simple" and "complicated" processing |
91 | * in the color map logic. It should usually be chosen high enough to ensure |
92 | * that all common characters are <= MAX_SIMPLE_CHR. However, very large |
93 | * values will be counterproductive since they cause more regex setup time. |
94 | * Also, small values can be helpful for testing the high-color-map logic |
95 | * with plain old ASCII input. |
96 | */ |
97 | #define MAX_SIMPLE_CHR 0x7FF /* suitable value for Unicode */ |
98 | |
99 | /* functions operating on chr */ |
100 | #define iscalnum(x) pg_wc_isalnum(x) |
101 | #define iscalpha(x) pg_wc_isalpha(x) |
102 | #define iscdigit(x) pg_wc_isdigit(x) |
103 | #define iscspace(x) pg_wc_isspace(x) |
104 | |
105 | /* and pick up the standard header */ |
106 | #include "regex.h" |
107 | |