Home | History | Annotate | Download | only in html
      1 <html>
      2 <head>
      3 <title>pcrecpp specification</title>
      4 </head>
      5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
      6 <h1>pcrecpp man page</h1>
      7 <p>
      8 Return to the <a href="index.html">PCRE index page</a>.
      9 </p>
     10 <p>
     11 This page is part of the PCRE HTML documentation. It was generated automatically
     12 from the original man page. If there is any nonsense in it, please consult the
     13 man page, in case the conversion went wrong.
     14 <br>
     15 <ul>
     16 <li><a name="TOC1" href="#SEC1">SYNOPSIS OF C++ WRAPPER</a>
     17 <li><a name="TOC2" href="#SEC2">DESCRIPTION</a>
     18 <li><a name="TOC3" href="#SEC3">MATCHING INTERFACE</a>
     19 <li><a name="TOC4" href="#SEC4">QUOTING METACHARACTERS</a>
     20 <li><a name="TOC5" href="#SEC5">PARTIAL MATCHES</a>
     21 <li><a name="TOC6" href="#SEC6">UTF-8 AND THE MATCHING INTERFACE</a>
     22 <li><a name="TOC7" href="#SEC7">PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE</a>
     23 <li><a name="TOC8" href="#SEC8">SCANNING TEXT INCREMENTALLY</a>
     24 <li><a name="TOC9" href="#SEC9">PARSING HEX/OCTAL/C-RADIX NUMBERS</a>
     25 <li><a name="TOC10" href="#SEC10">REPLACING PARTS OF STRINGS</a>
     26 <li><a name="TOC11" href="#SEC11">AUTHOR</a>
     27 <li><a name="TOC12" href="#SEC12">REVISION</a>
     28 </ul>
     29 <br><a name="SEC1" href="#TOC1">SYNOPSIS OF C++ WRAPPER</a><br>
     30 <P>
     31 <b>#include &#60;pcrecpp.h&#62;</b>
     32 </P>
     33 <br><a name="SEC2" href="#TOC1">DESCRIPTION</a><br>
     34 <P>
     35 The C++ wrapper for PCRE was provided by Google Inc. Some additional
     36 functionality was added by Giuseppe Maxia. This brief man page was constructed
     37 from the notes in the <i>pcrecpp.h</i> file, which should be consulted for
     38 further details. Note that the C++ wrapper supports only the original 8-bit
     39 PCRE library. There is no 16-bit or 32-bit support at present.
     40 </P>
     41 <br><a name="SEC3" href="#TOC1">MATCHING INTERFACE</a><br>
     42 <P>
     43 The "FullMatch" operation checks that supplied text matches a supplied pattern
     44 exactly. If pointer arguments are supplied, it copies matched sub-strings that
     45 match sub-patterns into them.
     46 <pre>
     47   Example: successful match
     48      pcrecpp::RE re("h.*o");
     49      re.FullMatch("hello");
     50 
     51   Example: unsuccessful match (requires full match):
     52      pcrecpp::RE re("e");
     53      !re.FullMatch("hello");
     54 
     55   Example: creating a temporary RE object:
     56      pcrecpp::RE("h.*o").FullMatch("hello");
     57 </pre>
     58 You can pass in a "const char*" or a "string" for "text". The examples below
     59 tend to use a const char*. You can, as in the different examples above, store
     60 the RE object explicitly in a variable or use a temporary RE object. The
     61 examples below use one mode or the other arbitrarily. Either could correctly be
     62 used for any of these examples.
     63 </P>
     64 <P>
     65 You must supply extra pointer arguments to extract matched subpieces.
     66 <pre>
     67   Example: extracts "ruby" into "s" and 1234 into "i"
     68      int i;
     69      string s;
     70      pcrecpp::RE re("(\\w+):(\\d+)");
     71      re.FullMatch("ruby:1234", &s, &i);
     72 
     73   Example: does not try to extract any extra sub-patterns
     74      re.FullMatch("ruby:1234", &s);
     75 
     76   Example: does not try to extract into NULL
     77      re.FullMatch("ruby:1234", NULL, &i);
     78 
     79   Example: integer overflow causes failure
     80      !re.FullMatch("ruby:1234567891234", NULL, &i);
     81 
     82   Example: fails because there aren't enough sub-patterns:
     83      !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s);
     84 
     85   Example: fails because string cannot be stored in integer
     86      !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
     87 </pre>
     88 The provided pointer arguments can be pointers to any scalar numeric
     89 type, or one of:
     90 <pre>
     91    string        (matched piece is copied to string)
     92    StringPiece   (StringPiece is mutated to point to matched piece)
     93    T             (where "bool T::ParseFrom(const char*, int)" exists)
     94    NULL          (the corresponding matched sub-pattern is not copied)
     95 </pre>
     96 The function returns true iff all of the following conditions are satisfied:
     97 <pre>
     98   a. "text" matches "pattern" exactly;
     99 
    100   b. The number of matched sub-patterns is &#62;= number of supplied
    101      pointers;
    102 
    103   c. The "i"th argument has a suitable type for holding the
    104      string captured as the "i"th sub-pattern. If you pass in
    105      void * NULL for the "i"th argument, or a non-void * NULL
    106      of the correct type, or pass fewer arguments than the
    107      number of sub-patterns, "i"th captured sub-pattern is
    108      ignored.
    109 </pre>
    110 CAVEAT: An optional sub-pattern that does not exist in the matched
    111 string is assigned the empty string. Therefore, the following will
    112 return false (because the empty string is not a valid number):
    113 <pre>
    114    int number;
    115    pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number);
    116 </pre>
    117 The matching interface supports at most 16 arguments per call.
    118 If you need more, consider using the more general interface
    119 <b>pcrecpp::RE::DoMatch</b>. See <b>pcrecpp.h</b> for the signature for
    120 <b>DoMatch</b>.
    121 </P>
    122 <P>
    123 NOTE: Do not use <b>no_arg</b>, which is used internally to mark the end of a
    124 list of optional arguments, as a placeholder for missing arguments, as this can
    125 lead to segfaults.
    126 </P>
    127 <br><a name="SEC4" href="#TOC1">QUOTING METACHARACTERS</a><br>
    128 <P>
    129 You can use the "QuoteMeta" operation to insert backslashes before all
    130 potentially meaningful characters in a string. The returned string, used as a
    131 regular expression, will exactly match the original string.
    132 <pre>
    133   Example:
    134      string quoted = RE::QuoteMeta(unquoted);
    135 </pre>
    136 Note that it's legal to escape a character even if it has no special meaning in
    137 a regular expression -- so this function does that. (This also makes it
    138 identical to the perl function of the same name; see "perldoc -f quotemeta".)
    139 For example, "1.5-2.0?" becomes "1\.5\-2\.0\?".
    140 </P>
    141 <br><a name="SEC5" href="#TOC1">PARTIAL MATCHES</a><br>
    142 <P>
    143 You can use the "PartialMatch" operation when you want the pattern
    144 to match any substring of the text.
    145 <pre>
    146   Example: simple search for a string:
    147      pcrecpp::RE("ell").PartialMatch("hello");
    148 
    149   Example: find first number in a string:
    150      int number;
    151      pcrecpp::RE re("(\\d+)");
    152      re.PartialMatch("x*100 + 20", &number);
    153      assert(number == 100);
    154 </PRE>
    155 </P>
    156 <br><a name="SEC6" href="#TOC1">UTF-8 AND THE MATCHING INTERFACE</a><br>
    157 <P>
    158 By default, pattern and text are plain text, one byte per character. The UTF8
    159 flag, passed to the constructor, causes both pattern and string to be treated
    160 as UTF-8 text, still a byte stream but potentially multiple bytes per
    161 character. In practice, the text is likelier to be UTF-8 than the pattern, but
    162 the match returned may depend on the UTF8 flag, so always use it when matching
    163 UTF8 text. For example, "." will match one byte normally but with UTF8 set may
    164 match up to three bytes of a multi-byte character.
    165 <pre>
    166   Example:
    167      pcrecpp::RE_Options options;
    168      options.set_utf8();
    169      pcrecpp::RE re(utf8_pattern, options);
    170      re.FullMatch(utf8_string);
    171 
    172   Example: using the convenience function UTF8():
    173      pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
    174      re.FullMatch(utf8_string);
    175 </pre>
    176 NOTE: The UTF8 flag is ignored if pcre was not configured with the
    177 <pre>
    178       --enable-utf8 flag.
    179 </PRE>
    180 </P>
    181 <br><a name="SEC7" href="#TOC1">PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE</a><br>
    182 <P>
    183 PCRE defines some modifiers to change the behavior of the regular expression
    184 engine. The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle to
    185 pass such modifiers to a RE class. Currently, the following modifiers are
    186 supported:
    187 <pre>
    188    modifier              description               Perl corresponding
    189 
    190    PCRE_CASELESS         case insensitive match      /i
    191    PCRE_MULTILINE        multiple lines match        /m
    192    PCRE_DOTALL           dot matches newlines        /s
    193    PCRE_DOLLAR_ENDONLY   $ matches only at end       N/A
    194    PCRE_EXTRA            strict escape parsing       N/A
    195    PCRE_EXTENDED         ignore white spaces         /x
    196    PCRE_UTF8             handles UTF8 chars          built-in
    197    PCRE_UNGREEDY         reverses * and *?           N/A
    198    PCRE_NO_AUTO_CAPTURE  disables capturing parens   N/A (*)
    199 </pre>
    200 (*) Both Perl and PCRE allow non capturing parentheses by means of the
    201 "?:" modifier within the pattern itself. e.g. (?:ab|cd) does not
    202 capture, while (ab|cd) does.
    203 </P>
    204 <P>
    205 For a full account on how each modifier works, please check the
    206 PCRE API reference page.
    207 </P>
    208 <P>
    209 For each modifier, there are two member functions whose name is made
    210 out of the modifier in lowercase, without the "PCRE_" prefix. For
    211 instance, PCRE_CASELESS is handled by
    212 <pre>
    213   bool caseless()
    214 </pre>
    215 which returns true if the modifier is set, and
    216 <pre>
    217   RE_Options & set_caseless(bool)
    218 </pre>
    219 which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can be
    220 accessed through the <b>set_match_limit()</b> and <b>match_limit()</b> member
    221 functions. Setting <i>match_limit</i> to a non-zero value will limit the
    222 execution of pcre to keep it from doing bad things like blowing the stack or
    223 taking an eternity to return a result. A value of 5000 is good enough to stop
    224 stack blowup in a 2MB thread stack. Setting <i>match_limit</i> to zero disables
    225 match limiting. Alternatively, you can call <b>match_limit_recursion()</b>
    226 which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much PCRE
    227 recurses. <b>match_limit()</b> limits the number of matches PCRE does;
    228 <b>match_limit_recursion()</b> limits the depth of internal recursion, and
    229 therefore the amount of stack that is used.
    230 </P>
    231 <P>
    232 Normally, to pass one or more modifiers to a RE class, you declare
    233 a <i>RE_Options</i> object, set the appropriate options, and pass this
    234 object to a RE constructor. Example:
    235 <pre>
    236    RE_Options opt;
    237    opt.set_caseless(true);
    238    if (RE("HELLO", opt).PartialMatch("hello world")) ...
    239 </pre>
    240 RE_options has two constructors. The default constructor takes no arguments and
    241 creates a set of flags that are off by default. The optional parameter
    242 <i>option_flags</i> is to facilitate transfer of legacy code from C programs.
    243 This lets you do
    244 <pre>
    245    RE(pattern,
    246      RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
    247 </pre>
    248 However, new code is better off doing
    249 <pre>
    250    RE(pattern,
    251      RE_Options().set_caseless(true).set_multiline(true))
    252        .PartialMatch(str);
    253 </pre>
    254 If you are going to pass one of the most used modifiers, there are some
    255 convenience functions that return a RE_Options class with the
    256 appropriate modifier already set: <b>CASELESS()</b>, <b>UTF8()</b>,
    257 <b>MULTILINE()</b>, <b>DOTALL</b>(), and <b>EXTENDED()</b>.
    258 </P>
    259 <P>
    260 If you need to set several options at once, and you don't want to go through
    261 the pains of declaring a RE_Options object and setting several options, there
    262 is a parallel method that give you such ability on the fly. You can concatenate
    263 several <b>set_xxxxx()</b> member functions, since each of them returns a
    264 reference to its class object. For example, to pass PCRE_CASELESS,
    265 PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one statement, you may write:
    266 <pre>
    267    RE(" ^ xyz \\s+ .* blah$",
    268      RE_Options()
    269        .set_caseless(true)
    270        .set_extended(true)
    271        .set_multiline(true)).PartialMatch(sometext);
    272 
    273 </PRE>
    274 </P>
    275 <br><a name="SEC8" href="#TOC1">SCANNING TEXT INCREMENTALLY</a><br>
    276 <P>
    277 The "Consume" operation may be useful if you want to repeatedly
    278 match regular expressions at the front of a string and skip over
    279 them as they match. This requires use of the "StringPiece" type,
    280 which represents a sub-range of a real string. Like RE, StringPiece
    281 is defined in the pcrecpp namespace.
    282 <pre>
    283   Example: read lines of the form "var = value" from a string.
    284      string contents = ...;                 // Fill string somehow
    285      pcrecpp::StringPiece input(contents);  // Wrap in a StringPiece
    286 
    287      string var;
    288      int value;
    289      pcrecpp::RE re("(\\w+) = (\\d+)\n");
    290      while (re.Consume(&input, &var, &value)) {
    291        ...;
    292      }
    293 </pre>
    294 Each successful call to "Consume" will set "var/value", and also
    295 advance "input" so it points past the matched text.
    296 </P>
    297 <P>
    298 The "FindAndConsume" operation is similar to "Consume" but does not
    299 anchor your match at the beginning of the string. For example, you
    300 could extract all words from a string by repeatedly calling
    301 <pre>
    302   pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word)
    303 </PRE>
    304 </P>
    305 <br><a name="SEC9" href="#TOC1">PARSING HEX/OCTAL/C-RADIX NUMBERS</a><br>
    306 <P>
    307 By default, if you pass a pointer to a numeric value, the
    308 corresponding text is interpreted as a base-10 number. You can
    309 instead wrap the pointer with a call to one of the operators Hex(),
    310 Octal(), or CRadix() to interpret the text in another base. The
    311 CRadix operator interprets C-style "0" (base-8) and "0x" (base-16)
    312 prefixes, but defaults to base-10.
    313 <pre>
    314   Example:
    315     int a, b, c, d;
    316     pcrecpp::RE re("(.*) (.*) (.*) (.*)");
    317     re.FullMatch("100 40 0100 0x40",
    318                  pcrecpp::Octal(&a), pcrecpp::Hex(&b),
    319                  pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
    320 </pre>
    321 will leave 64 in a, b, c, and d.
    322 </P>
    323 <br><a name="SEC10" href="#TOC1">REPLACING PARTS OF STRINGS</a><br>
    324 <P>
    325 You can replace the first match of "pattern" in "str" with "rewrite".
    326 Within "rewrite", backslash-escaped digits (\1 to \9) can be
    327 used to insert text matching corresponding parenthesized group
    328 from the pattern. \0 in "rewrite" refers to the entire matching
    329 text. For example:
    330 <pre>
    331   string s = "yabba dabba doo";
    332   pcrecpp::RE("b+").Replace("d", &s);
    333 </pre>
    334 will leave "s" containing "yada dabba doo". The result is true if the pattern
    335 matches and a replacement occurs, false otherwise.
    336 </P>
    337 <P>
    338 <b>GlobalReplace</b> is like <b>Replace</b> except that it replaces all
    339 occurrences of the pattern in the string with the rewrite. Replacements are
    340 not subject to re-matching. For example:
    341 <pre>
    342   string s = "yabba dabba doo";
    343   pcrecpp::RE("b+").GlobalReplace("d", &s);
    344 </pre>
    345 will leave "s" containing "yada dada doo". It returns the number of
    346 replacements made.
    347 </P>
    348 <P>
    349 <b>Extract</b> is like <b>Replace</b>, except that if the pattern matches,
    350 "rewrite" is copied into "out" (an additional argument) with substitutions.
    351 The non-matching portions of "text" are ignored. Returns true iff a match
    352 occurred and the extraction happened successfully;  if no match occurs, the
    353 string is left unaffected.
    354 </P>
    355 <br><a name="SEC11" href="#TOC1">AUTHOR</a><br>
    356 <P>
    357 The C++ wrapper was contributed by Google Inc.
    358 <br>
    359 Copyright &copy; 2007 Google Inc.
    360 <br>
    361 </P>
    362 <br><a name="SEC12" href="#TOC1">REVISION</a><br>
    363 <P>
    364 Last updated: 08 January 2012
    365 <br>
    366 <p>
    367 Return to the <a href="index.html">PCRE index page</a>.
    368 </p>
    369