summaryrefslogtreecommitdiffstats
path: root/xbmc/utils/RegExp.h
diff options
context:
space:
mode:
Diffstat (limited to 'xbmc/utils/RegExp.h')
-rw-r--r--xbmc/utils/RegExp.h165
1 files changed, 165 insertions, 0 deletions
diff --git a/xbmc/utils/RegExp.h b/xbmc/utils/RegExp.h
new file mode 100644
index 0000000..53f6019
--- /dev/null
+++ b/xbmc/utils/RegExp.h
@@ -0,0 +1,165 @@
1/*
2 * Copyright (C) 2005-2018 Team Kodi
3 * This file is part of Kodi - https://kodi.tv
4 *
5 * SPDX-License-Identifier: GPL-2.0-or-later
6 * See LICENSES/README.md for more information.
7 */
8
9#pragma once
10
11//! @todo - move to std::regex (after switching to gcc 4.9 or higher) and get rid of CRegExp
12
13#include <string>
14#include <vector>
15
16/* make sure stdlib.h is included before including pcre.h inside the
17 namespace; this works around stdlib.h definitions also living in
18 the PCRE namespace */
19#include <stdlib.h>
20
21namespace PCRE {
22struct real_pcre_jit_stack; // forward declaration for PCRE without JIT
23typedef struct real_pcre_jit_stack pcre_jit_stack;
24#include <pcre.h>
25}
26
27class CRegExp
28{
29public:
30 enum studyMode
31 {
32 NoStudy = 0, // do not study expression
33 StudyRegExp = 1, // study expression (slower compilation, faster find)
34 StudyWithJitComp // study expression and JIT-compile it, if possible (heavyweight optimization)
35 };
36 enum utf8Mode
37 {
38 autoUtf8 = -1, // analyze regexp for UTF-8 multi-byte chars, for Unicode codes > 0xFF
39 // or explicit Unicode properties (\p, \P and \X), enable UTF-8 mode if any of them are found
40 asciiOnly = 0, // process regexp and strings as single-byte encoded strings
41 forceUtf8 = 1 // enable UTF-8 mode (with Unicode properties)
42 };
43
44 static const int m_MaxNumOfBackrefrences = 20;
45 /**
46 * @param caseless (optional) Matching will be case insensitive if set to true
47 * or case sensitive if set to false
48 * @param utf8 (optional) Control UTF-8 processing
49 */
50 CRegExp(bool caseless = false, utf8Mode utf8 = asciiOnly);
51 /**
52 * Create new CRegExp object and compile regexp expression in one step
53 * @warning Use only with hardcoded regexp when you're sure that regexp is compiled without errors
54 * @param caseless Matching will be case insensitive if set to true
55 * or case sensitive if set to false
56 * @param utf8 Control UTF-8 processing
57 * @param re The regular expression
58 * @param study (optional) Controls study of expression, useful if expression will be used
59 * several times
60 */
61 CRegExp(bool caseless, utf8Mode utf8, const char *re, studyMode study = NoStudy);
62
63 CRegExp(const CRegExp& re);
64 ~CRegExp();
65
66 /**
67 * Compile (prepare) regular expression
68 * @param re The regular expression
69 * @param study (optional) Controls study of expression, useful if expression will be used
70 * several times
71 * @return true on success, false on any error
72 */
73 bool RegComp(const char *re, studyMode study = NoStudy);
74
75 /**
76 * Compile (prepare) regular expression
77 * @param re The regular expression
78 * @param study (optional) Controls study of expression, useful if expression will be used
79 * several times
80 * @return true on success, false on any error
81 */
82 bool RegComp(const std::string& re, studyMode study = NoStudy)
83 { return RegComp(re.c_str(), study); }
84
85 /**
86 * Find first match of regular expression in given string
87 * @param str The string to match against regular expression
88 * @param startoffset (optional) The string offset to start matching
89 * @param maxNumberOfCharsToTest (optional) The maximum number of characters to test (match) in
90 * string. If set to -1 string checked up to the end.
91 * @return staring position of match in string, negative value in case of error or no match
92 */
93 int RegFind(const char* str, unsigned int startoffset = 0, int maxNumberOfCharsToTest = -1);
94 /**
95 * Find first match of regular expression in given string
96 * @param str The string to match against regular expression
97 * @param startoffset (optional) The string offset to start matching
98 * @param maxNumberOfCharsToTest (optional) The maximum number of characters to test (match) in
99 * string. If set to -1 string checked up to the end.
100 * @return staring position of match in string, negative value in case of error or no match
101 */
102 int RegFind(const std::string& str, unsigned int startoffset = 0, int maxNumberOfCharsToTest = -1)
103 { return PrivateRegFind(str.length(), str.c_str(), startoffset, maxNumberOfCharsToTest); }
104 std::string GetReplaceString(const std::string& sReplaceExp) const;
105 int GetFindLen() const
106 {
107 if (!m_re || !m_bMatched)
108 return 0;
109
110 return (m_iOvector[1] - m_iOvector[0]);
111 };
112 int GetSubCount() const { return m_iMatchCount - 1; } // PCRE returns the number of sub-patterns + 1
113 int GetSubStart(int iSub) const;
114 int GetSubStart(const std::string& subName) const;
115 int GetSubLength(int iSub) const;
116 int GetSubLength(const std::string& subName) const;
117 int GetCaptureTotal() const;
118 std::string GetMatch(int iSub = 0) const;
119 std::string GetMatch(const std::string& subName) const;
120 const std::string& GetPattern() const { return m_pattern; }
121 bool GetNamedSubPattern(const char* strName, std::string& strMatch) const;
122 int GetNamedSubPatternNumber(const char* strName) const;
123 void DumpOvector(int iLog);
124 /**
125 * Check is RegExp object is ready for matching
126 * @return true if RegExp object is ready for matching, false otherwise
127 */
128 inline bool IsCompiled(void) const
129 { return !m_pattern.empty(); }
130 CRegExp& operator= (const CRegExp& re);
131 static bool IsUtf8Supported(void);
132 static bool AreUnicodePropertiesSupported(void);
133 static bool LogCheckUtf8Support(void);
134 static bool IsJitSupported(void);
135
136private:
137 int PrivateRegFind(size_t bufferLen, const char *str, unsigned int startoffset = 0, int maxNumberOfCharsToTest = -1);
138 void InitValues(bool caseless = false, CRegExp::utf8Mode utf8 = asciiOnly);
139 static bool requireUtf8(const std::string& regexp);
140 static int readCharXCode(const std::string& regexp, size_t& pos);
141 static bool isCharClassWithUnicode(const std::string& regexp, size_t& pos);
142
143 void Cleanup();
144 inline bool IsValidSubNumber(int iSub) const;
145
146 PCRE::pcre* m_re;
147 PCRE::pcre_extra* m_sd;
148 static const int OVECCOUNT=(m_MaxNumOfBackrefrences + 1) * 3;
149 unsigned int m_offset;
150 int m_iOvector[OVECCOUNT];
151 utf8Mode m_utf8Mode;
152 int m_iMatchCount;
153 int m_iOptions;
154 bool m_jitCompiled;
155 bool m_bMatched;
156 PCRE::pcre_jit_stack* m_jitStack;
157 std::string m_subject;
158 std::string m_pattern;
159 static int m_Utf8Supported;
160 static int m_UcpSupported;
161 static int m_JitSupported;
162};
163
164typedef std::vector<CRegExp> VECCREGEXP;
165
generated by cgit v1.2.3 (git 2.25.1) at 2026-01-12 10:18:39 +0000