460b9539 |
1 | /* |
2 | * This file is part of DisOrder |
3 | * Copyright (C) 2004, 2005 Richard Kettlewell |
4 | * |
5 | * This program is free software; you can redistribute it and/or modify |
6 | * it under the terms of the GNU General Public License as published by |
7 | * the Free Software Foundation; either version 2 of the License, or |
8 | * (at your option) any later version. |
9 | * |
10 | * This program is distributed in the hope that it will be useful, but |
11 | * WITHOUT ANY WARRANTY; without even the implied warranty of |
12 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
13 | * General Public License for more details. |
14 | * |
15 | * You should have received a copy of the GNU General Public License |
16 | * along with this program; if not, write to the Free Software |
17 | * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 |
18 | * USA |
19 | */ |
20 | |
21 | #include <config.h> |
22 | #include "types.h" |
23 | |
24 | #include <string.h> |
25 | #include <pcre.h> |
26 | |
27 | #include "regsub.h" |
28 | #include "mem.h" |
29 | #include "vector.h" |
30 | #include "log.h" |
31 | |
32 | #define PREMATCH (-1) /* fictitious pre-match substring */ |
33 | #define POSTMATCH (-2) /* fictitious post-match substring */ |
34 | |
35 | static inline int substring_start(const char attribute((unused)) *subject, |
36 | const int *ovector, |
37 | int n) { |
38 | switch(n) { |
39 | case PREMATCH: return 0; |
40 | case POSTMATCH: return ovector[1]; |
41 | default: return ovector[2 * n]; |
42 | } |
43 | } |
44 | |
45 | static inline int substring_end(const char *subject, |
46 | const int *ovector, |
47 | int n) { |
48 | switch(n) { |
49 | case PREMATCH: return ovector[0]; |
50 | case POSTMATCH: return strlen(subject); |
51 | default: return ovector[2 * n + 1]; |
52 | } |
53 | } |
54 | |
55 | static void transform_append(struct dynstr *d, |
56 | const char *subject, |
57 | const int *ovector, |
58 | int n) { |
59 | int start = substring_start(subject, ovector, n); |
60 | int end = substring_end(subject, ovector, n); |
61 | |
62 | if(start != -1) |
63 | dynstr_append_bytes(d, subject + start, end - start); |
64 | } |
65 | |
66 | static void replace_core(struct dynstr *d, |
67 | const char *subject, |
68 | const char *replace, |
69 | int rc, |
70 | const int *ovector) { |
71 | int substr; |
72 | |
73 | while(*replace) { |
74 | if(*replace == '$') |
75 | switch(replace[1]) { |
76 | case '&': |
77 | transform_append(d, subject, ovector, 0); |
78 | replace += 2; |
79 | break; |
80 | case '1': case '2': case '3': |
81 | case '4': case '5': case '6': |
82 | case '7': case '8': case '9': |
83 | substr = replace[1] - '0'; |
84 | if(substr < rc) |
85 | transform_append(d, subject, ovector, substr); |
86 | replace += 2; |
87 | break; |
88 | case '$': |
89 | dynstr_append(d, '$'); |
90 | replace += 2; |
91 | break; |
92 | default: |
93 | dynstr_append(d, *replace++); |
94 | break; |
95 | } |
96 | else |
97 | dynstr_append(d, *replace++); |
98 | } |
99 | } |
100 | |
101 | unsigned regsub_flags(const char *flags) { |
102 | unsigned f = 0; |
103 | |
104 | while(*flags) { |
105 | switch(*flags++) { |
106 | case 'g': f |= REGSUB_GLOBAL; break; |
107 | case 'i': f |= REGSUB_CASE_INDEPENDENT; break; |
108 | default: break; |
109 | } |
110 | } |
111 | return f; |
112 | } |
113 | |
114 | int regsub_compile_options(unsigned flags) { |
115 | int options = 0; |
116 | |
117 | if(flags & REGSUB_CASE_INDEPENDENT) |
118 | options |= PCRE_CASELESS; |
119 | return options; |
120 | } |
121 | |
122 | const char *regsub(const pcre *re, const char *subject, const char *replace, |
123 | unsigned flags) { |
124 | int rc, ovector[99], matches; |
125 | struct dynstr d; |
126 | |
127 | dynstr_init(&d); |
128 | matches = 0; |
129 | /* find the next match */ |
130 | while((rc = pcre_exec(re, 0, subject, strlen(subject), 0, |
131 | 0, ovector, sizeof ovector / sizeof (int))) > 0) { |
132 | /* text just before the match */ |
133 | if(!(flags & REGSUB_REPLACE)) |
134 | transform_append(&d, subject, ovector, PREMATCH); |
135 | /* the replacement text */ |
136 | replace_core(&d, subject, replace, rc, ovector); |
137 | ++matches; |
138 | if(!*subject) /* end of subject */ |
139 | break; |
140 | if(flags & REGSUB_REPLACE) /* replace subject entirely */ |
141 | break; |
142 | /* step over the matched substring */ |
143 | subject += substring_start(subject, ovector, POSTMATCH); |
144 | if(!(flags & REGSUB_GLOBAL)) |
145 | break; |
146 | } |
147 | if(rc <= 0 && rc != PCRE_ERROR_NOMATCH) { |
148 | error(0, "pcre_exec returned %d, subject '%s'", rc, subject); |
149 | return 0; |
150 | } |
151 | if((flags & REGSUB_MUST_MATCH) && matches == 0) |
152 | return 0; |
153 | /* append the remainder of the subject */ |
154 | if(!(flags & REGSUB_REPLACE)) |
155 | dynstr_append_string(&d, subject); |
156 | dynstr_terminate(&d); |
157 | return d.vec; |
158 | } |
159 | |
160 | /* |
161 | Local Variables: |
162 | c-basic-offset:2 |
163 | comment-column:40 |
164 | End: |
165 | */ |