ViewVC logotype

Contents of /code/trunk/doc/pcrecpp.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1055 - (show annotations)
Tue Oct 16 15:53:30 2012 UTC (8 years, 6 months ago) by chpe
File size: 12709 byte(s)
pcre32: Add 32-bit library

Create libpcre32 that operates on 32-bit characters (UTF-32).

This turned out to be surprisingly simple after the UTF-16 support
was introduced; mostly just extra ifdefs and adjusting and adding
some tests.
1 .TH PCRECPP 3 "08 January 2012" "PCRE 8.30"
3 PCRE - Perl-compatible regular expressions.
5 .rs
6 .sp
7 .B #include <pcrecpp.h>
8 .
10 .rs
11 .sp
12 The C++ wrapper for PCRE was provided by Google Inc. Some additional
13 functionality was added by Giuseppe Maxia. This brief man page was constructed
14 from the notes in the \fIpcrecpp.h\fP file, which should be consulted for
15 further details. Note that the C++ wrapper supports only the original 8-bit
16 PCRE library. There is no 16-bit or 32-bit support at present.
17 .
18 .
20 .rs
21 .sp
22 The "FullMatch" operation checks that supplied text matches a supplied pattern
23 exactly. If pointer arguments are supplied, it copies matched sub-strings that
24 match sub-patterns into them.
25 .sp
26 Example: successful match
27 pcrecpp::RE re("h.*o");
28 re.FullMatch("hello");
29 .sp
30 Example: unsuccessful match (requires full match):
31 pcrecpp::RE re("e");
32 !re.FullMatch("hello");
33 .sp
34 Example: creating a temporary RE object:
35 pcrecpp::RE("h.*o").FullMatch("hello");
36 .sp
37 You can pass in a "const char*" or a "string" for "text". The examples below
38 tend to use a const char*. You can, as in the different examples above, store
39 the RE object explicitly in a variable or use a temporary RE object. The
40 examples below use one mode or the other arbitrarily. Either could correctly be
41 used for any of these examples.
42 .P
43 You must supply extra pointer arguments to extract matched subpieces.
44 .sp
45 Example: extracts "ruby" into "s" and 1234 into "i"
46 int i;
47 string s;
48 pcrecpp::RE re("(\e\ew+):(\e\ed+)");
49 re.FullMatch("ruby:1234", &s, &i);
50 .sp
51 Example: does not try to extract any extra sub-patterns
52 re.FullMatch("ruby:1234", &s);
53 .sp
54 Example: does not try to extract into NULL
55 re.FullMatch("ruby:1234", NULL, &i);
56 .sp
57 Example: integer overflow causes failure
58 !re.FullMatch("ruby:1234567891234", NULL, &i);
59 .sp
60 Example: fails because there aren't enough sub-patterns:
61 !pcrecpp::RE("\e\ew+:\e\ed+").FullMatch("ruby:1234", &s);
62 .sp
63 Example: fails because string cannot be stored in integer
64 !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
65 .sp
66 The provided pointer arguments can be pointers to any scalar numeric
67 type, or one of:
68 .sp
69 string (matched piece is copied to string)
70 StringPiece (StringPiece is mutated to point to matched piece)
71 T (where "bool T::ParseFrom(const char*, int)" exists)
72 NULL (the corresponding matched sub-pattern is not copied)
73 .sp
74 The function returns true iff all of the following conditions are satisfied:
75 .sp
76 a. "text" matches "pattern" exactly;
77 .sp
78 b. The number of matched sub-patterns is >= number of supplied
79 pointers;
80 .sp
81 c. The "i"th argument has a suitable type for holding the
82 string captured as the "i"th sub-pattern. If you pass in
83 void * NULL for the "i"th argument, or a non-void * NULL
84 of the correct type, or pass fewer arguments than the
85 number of sub-patterns, "i"th captured sub-pattern is
86 ignored.
87 .sp
88 CAVEAT: An optional sub-pattern that does not exist in the matched
89 string is assigned the empty string. Therefore, the following will
90 return false (because the empty string is not a valid number):
91 .sp
92 int number;
93 pcrecpp::RE::FullMatch("abc", "[a-z]+(\e\ed+)?", &number);
94 .sp
95 The matching interface supports at most 16 arguments per call.
96 If you need more, consider using the more general interface
97 \fBpcrecpp::RE::DoMatch\fP. See \fBpcrecpp.h\fP for the signature for
98 \fBDoMatch\fP.
99 .P
100 NOTE: Do not use \fBno_arg\fP, which is used internally to mark the end of a
101 list of optional arguments, as a placeholder for missing arguments, as this can
102 lead to segfaults.
103 .
104 .
106 .rs
107 .sp
108 You can use the "QuoteMeta" operation to insert backslashes before all
109 potentially meaningful characters in a string. The returned string, used as a
110 regular expression, will exactly match the original string.
111 .sp
112 Example:
113 string quoted = RE::QuoteMeta(unquoted);
114 .sp
115 Note that it's legal to escape a character even if it has no special meaning in
116 a regular expression -- so this function does that. (This also makes it
117 identical to the perl function of the same name; see "perldoc -f quotemeta".)
118 For example, "1.5-2.0?" becomes "1\e.5\e-2\e.0\e?".
119 .
121 .rs
122 .sp
123 You can use the "PartialMatch" operation when you want the pattern
124 to match any substring of the text.
125 .sp
126 Example: simple search for a string:
127 pcrecpp::RE("ell").PartialMatch("hello");
128 .sp
129 Example: find first number in a string:
130 int number;
131 pcrecpp::RE re("(\e\ed+)");
132 re.PartialMatch("x*100 + 20", &number);
133 assert(number == 100);
134 .
135 .
137 .rs
138 .sp
139 By default, pattern and text are plain text, one byte per character. The UTF8
140 flag, passed to the constructor, causes both pattern and string to be treated
141 as UTF-8 text, still a byte stream but potentially multiple bytes per
142 character. In practice, the text is likelier to be UTF-8 than the pattern, but
143 the match returned may depend on the UTF8 flag, so always use it when matching
144 UTF8 text. For example, "." will match one byte normally but with UTF8 set may
145 match up to three bytes of a multi-byte character.
146 .sp
147 Example:
148 pcrecpp::RE_Options options;
149 options.set_utf8();
150 pcrecpp::RE re(utf8_pattern, options);
151 re.FullMatch(utf8_string);
152 .sp
153 Example: using the convenience function UTF8():
154 pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
155 re.FullMatch(utf8_string);
156 .sp
157 NOTE: The UTF8 flag is ignored if pcre was not configured with the
158 --enable-utf8 flag.
159 .
160 .
162 .rs
163 .sp
164 PCRE defines some modifiers to change the behavior of the regular expression
165 engine. The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle to
166 pass such modifiers to a RE class. Currently, the following modifiers are
167 supported:
168 .sp
169 modifier description Perl corresponding
170 .sp
171 PCRE_CASELESS case insensitive match /i
172 PCRE_MULTILINE multiple lines match /m
173 PCRE_DOTALL dot matches newlines /s
174 PCRE_DOLLAR_ENDONLY $ matches only at end N/A
175 PCRE_EXTRA strict escape parsing N/A
176 PCRE_EXTENDED ignore white spaces /x
177 PCRE_UTF8 handles UTF8 chars built-in
178 PCRE_UNGREEDY reverses * and *? N/A
179 PCRE_NO_AUTO_CAPTURE disables capturing parens N/A (*)
180 .sp
181 (*) Both Perl and PCRE allow non capturing parentheses by means of the
182 "?:" modifier within the pattern itself. e.g. (?:ab|cd) does not
183 capture, while (ab|cd) does.
184 .P
185 For a full account on how each modifier works, please check the
186 PCRE API reference page.
187 .P
188 For each modifier, there are two member functions whose name is made
189 out of the modifier in lowercase, without the "PCRE_" prefix. For
190 instance, PCRE_CASELESS is handled by
191 .sp
192 bool caseless()
193 .sp
194 which returns true if the modifier is set, and
195 .sp
196 RE_Options & set_caseless(bool)
197 .sp
198 which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can be
199 accessed through the \fBset_match_limit()\fP and \fBmatch_limit()\fP member
200 functions. Setting \fImatch_limit\fP to a non-zero value will limit the
201 execution of pcre to keep it from doing bad things like blowing the stack or
202 taking an eternity to return a result. A value of 5000 is good enough to stop
203 stack blowup in a 2MB thread stack. Setting \fImatch_limit\fP to zero disables
204 match limiting. Alternatively, you can call \fBmatch_limit_recursion()\fP
205 which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much PCRE
206 recurses. \fBmatch_limit()\fP limits the number of matches PCRE does;
207 \fBmatch_limit_recursion()\fP limits the depth of internal recursion, and
208 therefore the amount of stack that is used.
209 .P
210 Normally, to pass one or more modifiers to a RE class, you declare
211 a \fIRE_Options\fP object, set the appropriate options, and pass this
212 object to a RE constructor. Example:
213 .sp
214 RE_Options opt;
215 opt.set_caseless(true);
216 if (RE("HELLO", opt).PartialMatch("hello world")) ...
217 .sp
218 RE_options has two constructors. The default constructor takes no arguments and
219 creates a set of flags that are off by default. The optional parameter
220 \fIoption_flags\fP is to facilitate transfer of legacy code from C programs.
221 This lets you do
222 .sp
223 RE(pattern,
224 RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
225 .sp
226 However, new code is better off doing
227 .sp
228 RE(pattern,
229 RE_Options().set_caseless(true).set_multiline(true))
230 .PartialMatch(str);
231 .sp
232 If you are going to pass one of the most used modifiers, there are some
233 convenience functions that return a RE_Options class with the
234 appropriate modifier already set: \fBCASELESS()\fP, \fBUTF8()\fP,
235 \fBMULTILINE()\fP, \fBDOTALL\fP(), and \fBEXTENDED()\fP.
236 .P
237 If you need to set several options at once, and you don't want to go through
238 the pains of declaring a RE_Options object and setting several options, there
239 is a parallel method that give you such ability on the fly. You can concatenate
240 several \fBset_xxxxx()\fP member functions, since each of them returns a
241 reference to its class object. For example, to pass PCRE_CASELESS,
242 PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one statement, you may write:
243 .sp
244 RE(" ^ xyz \e\es+ .* blah$",
245 RE_Options()
246 .set_caseless(true)
247 .set_extended(true)
248 .set_multiline(true)).PartialMatch(sometext);
249 .sp
250 .
251 .
253 .rs
254 .sp
255 The "Consume" operation may be useful if you want to repeatedly
256 match regular expressions at the front of a string and skip over
257 them as they match. This requires use of the "StringPiece" type,
258 which represents a sub-range of a real string. Like RE, StringPiece
259 is defined in the pcrecpp namespace.
260 .sp
261 Example: read lines of the form "var = value" from a string.
262 string contents = ...; // Fill string somehow
263 pcrecpp::StringPiece input(contents); // Wrap in a StringPiece
264 .sp
265 string var;
266 int value;
267 pcrecpp::RE re("(\e\ew+) = (\e\ed+)\en");
268 while (re.Consume(&input, &var, &value)) {
269 ...;
270 }
271 .sp
272 Each successful call to "Consume" will set "var/value", and also
273 advance "input" so it points past the matched text.
274 .P
275 The "FindAndConsume" operation is similar to "Consume" but does not
276 anchor your match at the beginning of the string. For example, you
277 could extract all words from a string by repeatedly calling
278 .sp
279 pcrecpp::RE("(\e\ew+)").FindAndConsume(&input, &word)
280 .
281 .
283 .rs
284 .sp
285 By default, if you pass a pointer to a numeric value, the
286 corresponding text is interpreted as a base-10 number. You can
287 instead wrap the pointer with a call to one of the operators Hex(),
288 Octal(), or CRadix() to interpret the text in another base. The
289 CRadix operator interprets C-style "0" (base-8) and "0x" (base-16)
290 prefixes, but defaults to base-10.
291 .sp
292 Example:
293 int a, b, c, d;
294 pcrecpp::RE re("(.*) (.*) (.*) (.*)");
295 re.FullMatch("100 40 0100 0x40",
296 pcrecpp::Octal(&a), pcrecpp::Hex(&b),
297 pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
298 .sp
299 will leave 64 in a, b, c, and d.
300 .
301 .
303 .rs
304 .sp
305 You can replace the first match of "pattern" in "str" with "rewrite".
306 Within "rewrite", backslash-escaped digits (\e1 to \e9) can be
307 used to insert text matching corresponding parenthesized group
308 from the pattern. \e0 in "rewrite" refers to the entire matching
309 text. For example:
310 .sp
311 string s = "yabba dabba doo";
312 pcrecpp::RE("b+").Replace("d", &s);
313 .sp
314 will leave "s" containing "yada dabba doo". The result is true if the pattern
315 matches and a replacement occurs, false otherwise.
316 .P
317 \fBGlobalReplace\fP is like \fBReplace\fP except that it replaces all
318 occurrences of the pattern in the string with the rewrite. Replacements are
319 not subject to re-matching. For example:
320 .sp
321 string s = "yabba dabba doo";
322 pcrecpp::RE("b+").GlobalReplace("d", &s);
323 .sp
324 will leave "s" containing "yada dada doo". It returns the number of
325 replacements made.
326 .P
327 \fBExtract\fP is like \fBReplace\fP, except that if the pattern matches,
328 "rewrite" is copied into "out" (an additional argument) with substitutions.
329 The non-matching portions of "text" are ignored. Returns true iff a match
330 occurred and the extraction happened successfully; if no match occurs, the
331 string is left unaffected.
332 .
333 .
335 .rs
336 .sp
337 .nf
338 The C++ wrapper was contributed by Google Inc.
339 Copyright (c) 2007 Google Inc.
340 .fi
341 .
342 .
344 .rs
345 .sp
346 .nf
347 Last updated: 08 January 2012
348 .fi


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5