Tue, 30 Jun 2026 11:55:49 +0200
fix memory leak when program exits due to incorrect regex pattern
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
1 | /* |
|
34
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
2 | * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS HEADER. |
|
57
68018eac46c3
adds simple tiny test suite and updates license headers
Mike Becker <universe@uap-core.de>
parents:
54
diff
changeset
|
3 | * Copyright 2018 Mike Becker. All rights reserved. |
|
34
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
4 | * |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
5 | * Redistribution and use in source and binary forms, with or without |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
6 | * modification, are permitted provided that the following conditions are met: |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
7 | * |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
8 | * 1. Redistributions of source code must retain the above copyright |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
9 | * notice, this list of conditions and the following disclaimer. |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
10 | * |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
11 | * 2. Redistributions in binary form must reproduce the above copyright |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
12 | * notice, this list of conditions and the following disclaimer in the |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
13 | * documentation and/or other materials provided with the distribution. |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
14 | * |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
15 | * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
16 | * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
17 | * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
18 | * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
19 | * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
20 | * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
21 | * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
22 | * CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, |
|
fa9bda32de17
moved src files to src subdirectory and added licence text
Mike Becker <universe@uap-core.de>
parents:
29
diff
changeset
|
23 | * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
|
57
68018eac46c3
adds simple tiny test suite and updates license headers
Mike Becker <universe@uap-core.de>
parents:
54
diff
changeset
|
24 | * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
25 | */ |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
26 | |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
27 | #include "regex_parser.h" |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
28 | #include <ctype.h> |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
29 | |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
30 | regex_parser_t* new_regex_parser_t() { |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
31 | regex_parser_t* ret = malloc(sizeof(regex_parser_t)); |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
32 | if (ret != NULL) { |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
33 | ret->pattern_list = new_string_list_t(); |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
34 | ret->matched_counted = 0; |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
35 | ret->pattern_match = 0; |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
36 | ret->compiled_patterns = NULL; |
| 28 | 37 | ret->compiled_pattern_count = 0; |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
38 | ret->count_chars = false; |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
39 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
40 | return ret; |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
41 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
42 | |
|
54
76d46533b9a9
regex parser was not properly reset before each file, sometimes resulting in wrong line counts, when the previous scanned file ended with a match
Mike Becker <universe@uap-core.de>
parents:
48
diff
changeset
|
43 | void regex_parser_reset(regex_parser_t* parser) { |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
44 | parser->pattern_match = parser->matched_counted = 0; |
|
54
76d46533b9a9
regex parser was not properly reset before each file, sometimes resulting in wrong line counts, when the previous scanned file ended with a match
Mike Becker <universe@uap-core.de>
parents:
48
diff
changeset
|
45 | } |
|
76d46533b9a9
regex parser was not properly reset before each file, sometimes resulting in wrong line counts, when the previous scanned file ended with a match
Mike Becker <universe@uap-core.de>
parents:
48
diff
changeset
|
46 | |
| 28 | 47 | void regex_destcomppats(regex_parser_t* parser) { |
| 48 | if (parser->compiled_patterns != NULL) { | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
49 | for (unsigned i = 0 ; i < parser->compiled_pattern_count ; i++) { |
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
50 | regfree(parser->compiled_patterns + i); |
| 28 | 51 | } |
| 52 | free(parser->compiled_patterns); | |
| 53 | parser->compiled_patterns = NULL; | |
| 54 | parser->compiled_pattern_count = 0; | |
| 55 | } | |
| 56 | } | |
| 57 | ||
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
58 | void destroy_regex_parser_t(regex_parser_t* parser) { |
| 28 | 59 | regex_destcomppats(parser); |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
60 | destroy_string_list_t(parser->pattern_list); |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
61 | free(parser); |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
62 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
63 | |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
64 | bool regex_parser_matching(regex_parser_t* parser) { |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
65 | return parser->pattern_match > 0; |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
66 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
67 | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
68 | static unsigned regex_parser_count_chars(const char* input, |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
69 | unsigned start, unsigned end) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
70 | unsigned ret = 0; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
71 | for (unsigned i = start ; i < end ; i++) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
72 | ret += isspace(input[i]) ? 0 : 1; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
73 | } |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
74 | return ret; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
75 | } |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
76 | |
| 28 | 77 | int regex_parser_do(regex_parser_t* parser, char* input) { |
| 78 | int err = REG_NOMATCH; | |
| 79 | if (parser->compiled_pattern_count > 0) { | |
| 80 | regmatch_t match; | |
| 81 | ||
| 82 | if (regex_parser_matching(parser)) { | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
83 | if (parser->count_chars) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
84 | parser->matched_counted += |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
85 | regex_parser_count_chars(input, 0, strlen(input)); |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
86 | } else { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
87 | parser->matched_counted++; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
88 | } |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
89 | |
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
90 | err = regexec(parser->compiled_patterns + parser->pattern_match, |
| 28 | 91 | input, 1, &match, 0); |
| 92 | if (err > 0 && err != REG_NOMATCH) { | |
| 93 | fprintf(stderr, "Regex-Error: 0x%08x", err); | |
| 94 | } | |
| 95 | if (err == 0) { | |
| 96 | parser->pattern_match = 0; | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
97 | size_t input_len = strlen(input); |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
98 | if (match.rm_eo < input_len) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
99 | if (parser->count_chars) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
100 | /* do not exclude chars that occur after pattern end */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
101 | parser->matched_counted -= |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
102 | regex_parser_count_chars(input, match.rm_eo, input_len); |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
103 | } else { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
104 | /* do not exclude line, if it does not end with the pattern */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
105 | parser->matched_counted--; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
106 | } |
| 28 | 107 | } |
| 108 | } | |
| 109 | } else { | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
110 | for (unsigned i = 0 ; i < parser->compiled_pattern_count - 1 ; i += 2) { |
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
111 | err = regexec(parser->compiled_patterns + i, input, 1, &match, 0); |
| 28 | 112 | if (err > 0 && err != REG_NOMATCH) { |
| 113 | fprintf(stderr, "Regex-Error: 0x%08x", err); | |
| 114 | } | |
| 115 | if (err == 0) { | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
116 | /* a start pattern matches, start counting */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
117 | parser->matched_counted = 0; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
118 | /* Check, if end pattern is also in this line */ |
| 28 | 119 | parser->pattern_match = i+1; |
| 120 | regex_parser_do(parser, input); | |
|
66
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
121 | /* If something was found, determine what exactly to exclude */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
122 | if (parser->matched_counted > 0) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
123 | if (parser->count_chars) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
124 | /* do not exclude the characters before the pattern */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
125 | parser->matched_counted -= |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
126 | regex_parser_count_chars(input, 0, match.rm_so); |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
127 | } else { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
128 | /* do not match line, if it does not start with the pattern */ |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
129 | if (match.rm_so > 0) { |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
130 | parser->matched_counted--; |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
131 | } |
|
be2084398c37
new feature: count non-whitespace characters
Mike Becker <universe@uap-core.de>
parents:
57
diff
changeset
|
132 | } |
| 28 | 133 | } |
| 134 | break; | |
| 135 | } | |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
136 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
137 | } |
|
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
138 | } |
| 28 | 139 | return err; |
|
27
95a958e3de88
added regexp_parser struct and compile function
Mike Becker <universe@uap-core.de>
parents:
diff
changeset
|
140 | } |
| 28 | 141 | |
| 142 | bool regex_compile_all(regex_parser_t* parser) { | |
| 143 | size_t pcount = parser->pattern_list->count; | |
| 144 | if (pcount > 0) { | |
| 145 | regex_destcomppats(parser); | |
| 146 | parser->compiled_patterns = calloc(pcount, sizeof(regex_t)); | |
| 147 | ||
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
148 | for (size_t i = 0 ; i < pcount ; i++) { |
|
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
149 | if (regcomp(parser->compiled_patterns + i, |
|
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
150 | parser->pattern_list->items[i], REG_EXTENDED) != 0) { |
| 28 | 151 | fprintf(stderr, "Cannot compile pattern: %s\n", |
| 152 | (parser->pattern_list->items[i])); | |
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
153 | parser->compiled_pattern_count = i; |
|
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
154 | return false; |
| 28 | 155 | } |
| 156 | } | |
| 157 | } | |
|
92
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
158 | parser->compiled_pattern_count = pcount; |
|
51270d470cda
fix memory leak in regex parser (wrong free for compiled patterns)
Mike Becker <universe@uap-core.de>
parents:
66
diff
changeset
|
159 | return true; |
| 28 | 160 | } |