/[pcre]/code/trunk/doc/html/pcrepattern.html

Diff of /code/trunk/doc/html/pcrepattern.html

Parent Directory | Revision Log | View Patch Patch

-revision 454 by ph10,
Tue Sep 22 09:42:11 2009 UTC
+revision 461 by ph10,
Mon Oct  5 10:59:35 2009 UTC
 Line 61 
 description of PCRE's regular expression
  </P>
  <P>
  The original operation of PCRE was on strings of one-byte characters. However,
- there is now also support for UTF-8 character strings. To use this, you must
+ there is now also support for UTF-8 character strings. To use this,
- build PCRE to include UTF-8 support, and then call <b>pcre_compile()</b> with
+ PCRE must be built to include UTF-8 support, and you must call
- the PCRE_UTF8 option. There is also a special sequence that can be given at the
+ <b>pcre_compile()</b> or <b>pcre_compile2()</b> with the PCRE_UTF8 option. There
- start of a pattern:
+ is also a special sequence that can be given at the start of a pattern:
  <pre>
    (*UTF8)
  </pre>
 Line 111 
 string with one of the following five se
    (*ANYCRLF)   any of the three above
    (*ANY)       all Unicode newline sequences
  </pre>
- These override the default and the options given to <b>pcre_compile()</b>. For
+ These override the default and the options given to <b>pcre_compile()</b> or
- example, on a Unix system where LF is the default newline sequence, the pattern
+ <b>pcre_compile2()</b>. For example, on a Unix system where LF is the default
+ newline sequence, the pattern
  <pre>
    (*CR)a.b
  </pre>
-Line 228 
 Non-printing characters
+Line 229 
 Non-printing characters
  A second use of backslash provides a way of encoding non-printing characters
  in patterns in a visible manner. There is no restriction on the appearance of
  non-printing characters, apart from the binary zero that terminates a pattern,
- but when a pattern is being prepared by text editing, it is usually easier to
+ but when a pattern is being prepared by text editing, it is often easier to use
- use one of the following escape sequences than the binary character it
+ one of the following escape sequences than the binary character it represents:
- represents:
  <pre>
    \a        alarm, that is, the BEL character (hex 07)
    \cx       "control-x", where x is any character
 Line 334 
 a number enclosed either in angle bracke
  syntax for referencing a subpattern as a "subroutine". Details are discussed
  <a href="#onigurumasubroutines">later.</a>
  Note that \g{...} (Perl syntax) and \g&#60;...&#62; (Oniguruma syntax) are <i>not</i>
  synonymous. The former is a back reference; the latter is a
  <a href="#subpatternsassubroutines">subroutine</a>
  call.
  </P>
 Line 465 
 one of the following sequences:
    (*BSR_ANYCRLF)   CR, LF, or CRLF only
    (*BSR_UNICODE)   any Unicode newline sequence
  </pre>
- These override the default and the options given to <b>pcre_compile()</b>, but
+ These override the default and the options given to <b>pcre_compile()</b> or
- they can be overridden by options given to <b>pcre_exec()</b>. Note that these
+ <b>pcre_compile2()</b>, but they can be overridden by options given to
- special settings, which are not Perl-compatible, are recognized only at the
+ <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b>. Note that these special settings,
- very start of a pattern, and that they must be in upper case. If more than one
+ which are not Perl-compatible, are recognized only at the very start of a
- of them is present, the last one is used. They can be combined with a change of
+ pattern, and that they must be in upper case. If more than one of them is
- newline convention, for example, a pattern can start with:
+ present, the last one is used. They can be combined with a change of newline
+ convention, for example, a pattern can start with:
  <pre>
    (*ANY)(*BSR_ANYCRLF)
  </pre>
-Line 731 
 different meaning, namely the backspace
+Line 732 
 different meaning, namely the backspace
  A word boundary is a position in the subject string where the current character
  and the previous character do not both match \w or \W (i.e. one matches
  \w and the other matches \W), or the start or end of the string if the
- first or last character matches \w, respectively.
+ first or last character matches \w, respectively. Neither PCRE nor Perl has a
+ separte "start of word" or "end of word" metasequence. However, whatever
+ follows \b normally determines which it is. For example, the fragment
+ \ba matches "a" at the start of a word.
  </P>
  <P>
  The \A, \Z, and \z assertions differ from the traditional circumflex and
-Line 862 
 the lookbehind.
+Line 866 
 the lookbehind.
  <br><a name="SEC8" href="#TOC1">SQUARE BRACKETS AND CHARACTER CLASSES</a><br>
  <P>
  An opening square bracket introduces a character class, terminated by a closing
- square bracket. A closing square bracket on its own is not special. If a
+ square bracket. A closing square bracket on its own is not special by default.
- closing square bracket is required as a member of the class, it should be the
+ However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
- first data character in the class (after an initial circumflex, if present) or
+ bracket causes a compile-time error. If a closing square bracket is required as
- escaped with a backslash.
+ a member of the class, it should be the first data character in the class
+ (after an initial circumflex, if present) or escaped with a backslash.
  </P>
  <P>
  A character class matches a single character in the subject. In UTF-8 mode, the
- character may occupy more than one byte. A matched character must be in the set
+ character may be more than one byte long. A matched character must be in the
- of characters defined by the class, unless the first character in the class
+ set of characters defined by the class, unless the first character in the class
  definition is a circumflex, in which case the subject character must not be in
  the set defined by the class. If a circumflex is actually required as a member
  of the class, ensure it is not the first character, or escape it with a
-Line 881 
 For example, the character class [aeiou]
+Line 886 
 For example, the character class [aeiou]
  [^aeiou] matches any character that is not a lower case vowel. Note that a
  circumflex is just a convenient notation for specifying the characters that
  are in the class by enumerating those that are not. A class that starts with a
- circumflex is not an assertion: it still consumes a character from the subject
+ circumflex is not an assertion; it still consumes a character from the subject
  string, and therefore it fails if the current pointer is at the end of the
  string.
  </P>
-Line 897 
 caseful version would. In UTF-8 mode, PC
+Line 902 
 caseful version would. In UTF-8 mode, PC
  case for characters whose values are less than 128, so caseless matching is
  always possible. For characters with higher values, the concept of case is
  supported if PCRE is compiled with Unicode property support, but not otherwise.
- If you want to use caseless matching for characters 128 and above, you must
+ If you want to use caseless matching in UTF8-mode for characters 128 and above,
- ensure that PCRE is compiled with Unicode property support as well as with
+ you must ensure that PCRE is compiled with Unicode property support as well as
- UTF-8 support.
+ with UTF-8 support.
  </P>
  <P>
  Characters that might indicate line breaks are never treated in any special way
-Line 1127 
 match exactly the same set of strings. B
+Line 1132 
 match exactly the same set of strings. B
  from left to right, and options are not reset until the end of the subpattern
  is reached, an option setting in one branch does affect subsequent branches, so
  the above patterns match "SUNDAY" as well as "Saturday".
- </P>
+ <a name="dupsubpatternnumber"></a></P>
  <br><a name="SEC13" href="#TOC1">DUPLICATE SUBPATTERN NUMBERS</a><br>
  <P>
  Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
-Line 1152 
 stored.
+Line 1157 
 stored.
    / ( a )  (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
    # 1            2         2  3        2     3     4
  </pre>
- A backreference or a recursive call to a numbered subpattern always refers to
+ A backreference to a numbered subpattern uses the most recent value that is set
- the first one in the pattern with the given number.
+ for that number by any subpattern. The following pattern matches "abcabc" or
+ "defdef":
+ <pre>
+   /(?|(abc)|(def))\1/
+ </pre>
+ In contrast, a recursive or "subroutine" call to a numbered subpattern always
+ refers to the first one in the pattern with the given number. The following
+ pattern matches "abcabc" or "defabc":
+ <pre>
+   /(?|(abc)|(def))(?1)/
+ </pre>
+ If a
+ <a href="#conditions">condition test</a>
+ for a subpattern's having matched refers to a non-unique number, the test is
+ true if any of the subpatterns of that number have matched.
  </P>
  <P>
  An alternative approach to using this "branch reset" feature is to use
-Line 1167 
 if an expression is modified, the number
+Line 1186 
 if an expression is modified, the number
  difficulty, PCRE supports the naming of subpatterns. This feature was not
  added to Perl until release 5.10. Python had the feature earlier, and PCRE
  introduced it at release 4.0, using the Python syntax. PCRE now supports both
- the Perl and the Python syntax.
+ the Perl and the Python syntax. Perl allows identically numbered subpatterns to
+ have different names, but PCRE does not.
  </P>
  <P>
  In PCRE, a subpattern can be named in one of three ways: (?&#60;name&#62;...) or
-Line 1188 
 is also a convenience function for extra
+Line 1208 
 is also a convenience function for extra
  </P>
  <P>
  By default, a name must be unique within a pattern, but it is possible to relax
- this constraint by setting the PCRE_DUPNAMES option at compile time. This can
+ this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
- be useful for patterns where only one instance of the named parentheses can
+ names are also always permitted for subpatterns with the same number, set up as
- match. Suppose you want to match the name of a weekday, either as a 3-letter
+ described in the previous section.) Duplicate names can be useful for patterns
- abbreviation or as the full name, and in both cases you want to extract the
+ where only one instance of the named parentheses can match. Suppose you want to
- abbreviation. This pattern (ignoring the line breaks) does the job:
+ match the name of a weekday, either as a 3-letter abbreviation or as the full
+ name, and in both cases you want to extract the abbreviation. This pattern
+ (ignoring the line breaks) does the job:
  <pre>
    (?&#60;DN&#62;Mon|Fri|Sun)(?:day)?|
    (?&#60;DN&#62;Tue)(?:sday)?|
-Line 1207 
 subpattern, as described in the previous
+Line 1229 
 subpattern, as described in the previous
  <P>
  The convenience function for extracting the data by name returns the substring
  for the first (and in this example, the only) subpattern of that name that
- matched. This saves searching to find which numbered subpattern it was. If you
+ matched. This saves searching to find which numbered subpattern it was.
- make a reference to a non-unique named subpattern from elsewhere in the
+ </P>
- pattern, the one that corresponds to the lowest number is used. For further
+ <P>
- details of the interfaces for handling named subpatterns, see the
+ If you make a backreference to a non-unique named subpattern from elsewhere in
+ the pattern, the one that corresponds to the first occurrence of the name is
+ used. In the absence of duplicate numbers (see the previous section) this is
+ the one with the lowest number. If you use a named reference in a condition
+ test (see the
+ <a href="#conditions">section about conditions</a>
+ below), either to check whether a subpattern has matched, or to check for
+ recursion, all subpatterns with the same name are tested. If the condition is
+ true for any one of them, the overall condition is true. This is the same
+ behaviour as testing by number. For further details of the interfaces for
+ handling named subpatterns, see the
  <a href="pcreapi.html"><b>pcreapi</b></a>
  documentation.
  </P>
  <P>
  <b>Warning:</b> You cannot use different names to distinguish between two
- subpatterns with the same number (see the previous section) because PCRE uses
+ subpatterns with the same number because PCRE uses only the numbers when
- only the numbers when matching.
+ matching. For this reason, an error is given at compile time if different names
+ are given to subpatterns with the same number. However, you can give the same
+ name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
  </P>
  <br><a name="SEC15" href="#TOC1">REPETITION</a><br>
  <P>
-Line 1233 
 items:
+Line 1267 
 items:
    a character class
    a back reference (see next section)
    a parenthesized subpattern (unless it is an assertion)
+   a recursive or "subroutine" call to a subpattern
  </pre>
  The general repetition quantifier specifies a minimum and maximum number of
  permitted matches, by giving the two numbers in curly brackets (braces),
-Line 1564 
 after the reference.
+Line 1599 
 after the reference.
  <P>
  There may be more than one back reference to the same subpattern. If a
  subpattern has not actually been used in a particular match, any back
- references to it always fail. For example, the pattern
+ references to it always fail by default. For example, the pattern
  <pre>
    (a|(bc))\2
  </pre>
- always fails if it starts to match "a" rather than "bc". Because there may be
+ always fails if it starts to match "a" rather than "bc". However, if the
- many capturing parentheses in a pattern, all digits following the backslash are
+ PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
- taken as part of a potential back reference number. If the pattern continues
+ unset value matches an empty string.
- with a digit character, some delimiter must be used to terminate the back
+ </P>
- reference. If the PCRE_EXTENDED option is set, this can be whitespace.
+ <P>
- Otherwise an empty comment (see
+ Because there may be many capturing parentheses in a pattern, all digits
+ following a backslash are taken as part of a potential back reference number.
+ If the pattern continues with a digit character, some delimiter must be used to
+ terminate the back reference. If the PCRE_EXTENDED option is set, this can be
+ whitespace. Otherwise, the \g{ syntax or an empty comment (see
  <a href="#comments">"Comments"</a>
  below) can be used.
  </P>
-Line 1641 
 lookbehind assertion is needed to achiev
+Line 1680 
 lookbehind assertion is needed to achiev
  If you want to force a matching failure at some point in a pattern, the most
  convenient way to do it is with (?!) because an empty string always matches, so
  an assertion that requires there not to be an empty string must always fail.
+ The Perl 5.10 backtracking control verb (*FAIL) or (*F) is essentially a
+ synonym for (?!).
  <a name="lookbehind"></a></P>
  <br><b>
  Lookbehind assertions
-Line 1677 
 branches:
+Line 1718 
 branches:
  </pre>
  In some cases, the Perl 5.10 escape sequence \K
  <a href="#resetmatchstart">(see above)</a>
  can be used instead of a lookbehind assertion to get round the fixed-length
  restriction.
  </P>
  <P>
-Line 1695 
 different numbers of bytes, are also not
+Line 1736 
 different numbers of bytes, are also not
  <P>
  <a href="#subpatternsassubroutines">"Subroutine"</a>
  calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
  as the subpattern matches a fixed-length string.
  <a href="#recursion">Recursion,</a>
  however, is not supported.
  </P>
  <P>
  Possessive quantifiers can be used in conjunction with lookbehind assertions to
- specify efficient matching at the end of the subject string. Consider a simple
+ specify efficient matching of fixed-length strings at the end of subject
- pattern such as
+ strings. Consider a simple pattern such as
  <pre>
    abcd$
  </pre>
-Line 1764 
 characters that are not "999".
+Line 1805 
 characters that are not "999".
  <P>
  It is possible to cause the matching process to obey a subpattern
  conditionally or to choose between two alternative subpatterns, depending on
- the result of an assertion, or whether a previous capturing subpattern matched
+ the result of an assertion, or whether a specific capturing subpattern has
- or not. The two possible forms of conditional subpattern are
+ already been matched. The two possible forms of conditional subpattern are:
  <pre>
    (?(condition)yes-pattern)
    (?(condition)yes-pattern|no-pattern)
-Line 1783 
 Checking for a used subpattern by number
+Line 1824 
 Checking for a used subpattern by number
  </b><br>
  <P>
  If the text between the parentheses consists of a sequence of digits, the
- condition is true if the capturing subpattern of that number has previously
+ condition is true if a capturing subpattern of that number has previously
- matched. An alternative notation is to precede the digits with a plus or minus
+ matched. If there is more than one capturing subpattern with the same number
- sign. In this case, the subpattern number is relative rather than absolute.
+ (see the earlier
- The most recently opened parentheses can be referenced by (?(-1), the next most
+ <a href="#recursion">section about duplicate subpattern numbers),</a>
- recent by (?(-2), and so on. In looping constructs it can also make sense to
+ the condition is true if any of them have been set. An alternative notation is
- refer to subsequent groups with constructs such as (?(+2).
+ to precede the digits with a plus or minus sign. In this case, the subpattern
+ number is relative rather than absolute. The most recently opened parentheses
+ can be referenced by (?(-1), the next most recent by (?(-2), and so on. In
+ looping constructs it can also make sense to refer to subsequent groups with
+ constructs such as (?(+2).
  </P>
  <P>
  Consider the following pattern, which contains non-significant white space to
-Line 1832 
 names that consist entirely of digits is
+Line 1877 
 names that consist entirely of digits is
  Rewriting the above example to use a named subpattern gives this:
  <pre>
    (?&#60;OPEN&#62; \( )?    [^()]+    (?(&#60;OPEN&#62;) \) )
+ </pre>
- </PRE>
+ If the name used in a condition of this kind is a duplicate, the test is
+ applied to all subpatterns of the same name, and is true if any one of them has
+ matched.
  </P>
  <br><b>
  Checking for pattern recursion
-Line 1846 
 letter R, for example:
+Line 1893 
 letter R, for example:
  <pre>
    (?(R3)...) or (?(R&name)...)
  </pre>
- the condition is true if the most recent recursion is into the subpattern whose
+ the condition is true if the most recent recursion is into a subpattern whose
  number or name is given. This condition does not check the entire recursion
- stack.
+ stack. If the name used in a condition of this kind is a duplicate, the test is
+ applied to all subpatterns of the same name, and is true if any one of them is
+ the most recent recursion.
  </P>
  <P>
  At "top level", all these recursion test conditions are false.
- <a href="#recursion">Recursive patterns</a>
+ <a href="#recursion">The syntax for recursive patterns</a>
- are described below.
+ is described below.
  </P>
  <br><b>
  Defining subpatterns for use by reference only
-Line 1863 
 If the condition is the string (DEFINE),
+Line 1912 
 If the condition is the string (DEFINE),
  name DEFINE, the condition is always false. In this case, there may be only one
  alternative in the subpattern. It is always skipped if control reaches this
  point in the pattern; the idea of DEFINE is that it can be used to define
  "subroutines" that can be referenced from elsewhere. (The use of
  <a href="#subpatternsassubroutines">"subroutines"</a>
  is described below.) For example, a pattern to match an IPv4 address could be
  written like this (ignore whitespace and line breaks):
-Line 1874 
 written like this (ignore whitespace and
+Line 1923 
 written like this (ignore whitespace and
  The first part of the pattern is a DEFINE group inside which a another group
  named "byte" is defined. This matches an individual component of an IPv4
  address (a number less than 256). When matching takes place, this part of the
- pattern is skipped because DEFINE acts like a false condition.
+ pattern is skipped because DEFINE acts like a false condition. The rest of the
- </P>
+ pattern uses references to the named group to match the four dot-separated
- <P>
+ components of an IPv4 address, insisting on a word boundary at each end.
- The rest of the pattern uses references to the named group to match the four
- dot-separated components of an IPv4 address, insisting on a word boundary at
- each end.
  </P>
  <br><b>
  Assertion conditions
-Line 1939 
 this kind of recursion was subsequently
+Line 1985 
 this kind of recursion was subsequently
  <P>
  A special item that consists of (? followed by a number greater than zero and a
  closing parenthesis is a recursive call of the subpattern of the given number,
  provided that it occurs inside that subpattern. (If not, it is a
  <a href="#subpatternsassubroutines">"subroutine"</a>
  call, which is described in the next section.) The special item (?R) or (?0) is
  a recursive call of the entire regular expression.
-Line 1948 
 a recursive call of the entire regular e
+Line 1994 
 a recursive call of the entire regular e
  This PCRE pattern solves the nested parentheses problem (assume the
  PCRE_EXTENDED option is set so that white space is ignored):
  <pre>
-   \( ( (?&#62;[^()]+) | (?R) )* \)
+   \( ( [^()]++ | (?R) )* \)
  </pre>
  First it matches an opening parenthesis. Then it matches any number of
  substrings which can either be a sequence of non-parentheses, or a recursive
  match of the pattern itself (that is, a correctly parenthesized substring).
- Finally there is a closing parenthesis.
+ Finally there is a closing parenthesis. Note the use of a possessive quantifier
+ to avoid backtracking into sequences of non-parentheses.
  </P>
  <P>
  If this were part of a larger pattern, you would not want to recurse the entire
  pattern, so instead you could use this:
  <pre>
-   ( \( ( (?&#62;[^()]+) | (?1) )* \) )
+   ( \( ( [^()]++ | (?1) )* \) )
  </pre>
  We have put the pattern into parentheses, and caused the recursion to refer to
  them instead of the whole pattern.
  </P>
  <P>
  In a larger pattern, keeping track of parenthesis numbers can be tricky. This
- is made easier by the use of relative references. (A Perl 5.10 feature.)
+ is made easier by the use of relative references (a Perl 5.10 feature).
  Instead of (?1) in the pattern above you can write (?-2) to refer to the second
  most recently opened parentheses preceding the recursion. In other words, a
  negative number counts capturing parentheses leftwards from the point at which
-Line 1984 
 An alternative approach is to use named
+Line 2031 
 An alternative approach is to use named
  for this is (?&name); PCRE's earlier syntax (?P&#62;name) is also supported. We
  could rewrite the above example as follows:
  <pre>
-   (?&#60;pn&#62; \( ( (?&#62;[^()]+) | (?&pn) )* \) )
+   (?&#60;pn&#62; \( ( [^()]++ | (?&pn) )* \) )
  </pre>
  If there is more than one subpattern with the same name, the earliest one is
  used.
  </P>
  <P>
  This particular example pattern that we have been looking at contains nested
- unlimited repeats, and so the use of atomic grouping for matching strings of
+ unlimited repeats, and so the use of a possessive quantifier for matching
- non-parentheses is important when applying the pattern to strings that do not
+ strings of non-parentheses is important when applying the pattern to strings
- match. For example, when this pattern is applied to
+ that do not match. For example, when this pattern is applied to
  <pre>
    (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
  </pre>
- it yields "no match" quickly. However, if atomic grouping is not used,
+ it yields "no match" quickly. However, if a possessive quantifier is not used,
  the match runs for a very long time indeed because there are so many different
  ways the + and * repeats can carve up the subject, and all have to be tested
  before failure can be reported.
-Line 2015 
 documentation). If the pattern above is
+Line 2062 
 documentation). If the pattern above is
  the value for the capturing parentheses is "ef", which is the last value taken
  on at the top level. If additional parentheses are added, giving
  <pre>
-   \( ( ( (?&#62;[^()]+) | (?R) )* ) \)
+   \( ( ( [^()]++ | (?R) )* ) \)
       ^                        ^
       ^                        ^
  </pre>
-Line 2044 
 Recursion difference from Perl
+Line 2091 
 Recursion difference from Perl
  In PCRE (like Python, but unlike Perl), a recursive subpattern call is always
  treated as an atomic group. That is, once it has matched some of the subject
  string, it is never re-entered, even if it contains untried alternatives and
  there is a subsequent matching failure. This can be illustrated by the
  following pattern, which purports to match a palindromic string that contains
  an odd number of characters (for example, "a", "aba", "abcba", "abcdcba"):
  <pre>
    ^(.|(.)(?1)\2)$
  </pre>
  The idea is that it either matches a single character, or two identical
  characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
  it does not if the pattern is longer than three characters. Consider the
  subject string "abcba":
  </P>
  <P>
  At the top level, the first character is matched, but as it is not at the end
  of the string, the first alternative fails; the second alternative is taken
  and the recursion kicks in. The recursive call to subpattern 1 successfully
  matches the next character ("b"). (Note that the beginning and end of line
-Line 2064 
 tests are not part of the recursion).
+Line 2111 
 tests are not part of the recursion).
  </P>
  <P>
  Back at the top level, the next character ("c") is compared with what
  subpattern 2 matched, which was "a". This fails. Because the recursion is
  treated as an atomic group, there are now no backtracking points, and so the
  entire match fails. (Perl is able, at this point, to re-enter the recursion and
  try the second alternative.) However, if the pattern is written with the
-Line 2072 
 alternatives in the other order, things
+Line 2119 
 alternatives in the other order, things
  <pre>
    ^((.)(?1)\2|.)$
  </pre>
  This time, the recursing alternative is tried first, and continues to recurse
  until it runs out of characters, at which point the recursion fails. But this
  time we do have another alternative to try at the higher level. That is the big
  difference: in the previous case the remaining alternative is at a deeper
  recursion level, which PCRE cannot use.
  </P>
  <P>
  To change the pattern so that matches all palindromic strings, not just those
  with an odd number of characters, it is tempting to change the pattern to this:
  <pre>
    ^((.)(?1)\2|.?)$
  </pre>
  Again, this works in Perl, but not in PCRE, and for the same reason. When a
  deeper recursion has matched a single character, it cannot be entered again in
  order to match an empty string. The solution is to separate the two cases, and
  write out the odd and even cases as alternatives at the higher level:
  <pre>
    ^(?:((.)(?1)\2|)|((.)(?3)\4|.))
  </pre>
  If you want to match typical palindromic phrases, the pattern has to ignore all
  non-word characters, which can be done like this:
  <pre>
    ^\W*+(?:((.)\W*+(?1)\W*+\2|)|((.)\W*+(?3)\W*+\4|\W*+.\W*+))\W*+$
  </pre>
  If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
  man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
  the use of the possessive quantifier *+ to avoid backtracking into sequences of
  non-word characters. Without this, PCRE takes a great deal longer (ten times or
  more) to match typical phrases, and Perl takes so long that you think it has
  gone into a loop.
+ </P>
+ <P>
+ <b>WARNING</b>: The palindrome-matching patterns above work only if the subject
+ string does not start with a palindrome that is shorter than the entire string.
+ For example, although "abcba" is correctly matched, if the subject is "ababa",
+ PCRE finds the palindrome "aba" at the start, then fails at top level because
+ the end of the string does not follow. Once again, it cannot jump back into the
+ recursion to try other alternatives, so the entire match fails.
  <a name="subpatternsassubroutines"></a></P>
  <br><a name="SEC22" href="#TOC1">SUBPATTERNS AS SUBROUTINES</a><br>
  <P>
-Line 2212 
 failing negative assertion, they cause a
+Line 2267 
 failing negative assertion, they cause a
  <b>pcre_dfa_exec()</b>.
  </P>
  <P>
  If any of these verbs are used in an assertion subpattern, their effect is
  confined to that subpattern; it does not extend to the surrounding pattern.
  Note that assertion subpatterns are processed as anchored at the point where
  they are tested.
  </P>
  <P>
-Line 2234 
 The following verbs act as soon as they
+Line 2289 
 The following verbs act as soon as they
  </pre>
  This verb causes the match to end successfully, skipping the remainder of the
  pattern. When inside a recursion, only the innermost pattern is ended
- immediately. If the (*ACCEPT) is inside capturing parentheses, the data so far
+ immediately. If (*ACCEPT) is inside capturing parentheses, the data so far is
- is captured. (This feature was added to PCRE at release 8.00.) For example:
+ captured. (This feature was added to PCRE at release 8.00.) For example:
  <pre>
    A((?:A|B(*ACCEPT)|C)D)
  </pre>
  This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
  the outer parentheses.
  <pre>
    (*FAIL) or (*F)
-Line 2267 
 The verbs differ in exactly what kind of
+Line 2322 
 The verbs differ in exactly what kind of
  </pre>
  This verb causes the whole match to fail outright if the rest of the pattern
  does not match. Even if the pattern is unanchored, no further attempts to find
- a match by advancing the start point take place. Once (*COMMIT) has been
+ a match by advancing the starting point take place. Once (*COMMIT) has been
  passed, <b>pcre_exec()</b> is committed to finding a match at the current
  starting point, or not at all. For example:
  <pre>
-Line 2299 
 was matched leading up to it cannot be p
+Line 2354 
 was matched leading up to it cannot be p
  If the subject is "aaaac...", after the first match attempt fails (starting at
  the first character in the string), the starting point skips on to start the
  next attempt at "c". Note that a possessive quantifer does not have the same
- effect in this example; although it would suppress backtracking during the
+ effect as this example; although it would suppress backtracking during the
  first match attempt, the second attempt would start at the second character
  instead of skipping on to "c".
  <pre>
-Line 2319 
 is used outside of any alternation, it a
+Line 2374 
 is used outside of any alternation, it a
  </P>
  <br><a name="SEC26" href="#TOC1">SEE ALSO</a><br>
  <P>
- <b>pcreapi</b>(3), <b>pcrecallout</b>(3), <b>pcrematching</b>(3), <b>pcre</b>(3).
+ <b>pcreapi</b>(3), <b>pcrecallout</b>(3), <b>pcrematching</b>(3),
+ <b>pcresyntax</b>(3), <b>pcre</b>(3).
  </P>
  <br><a name="SEC27" href="#TOC1">AUTHOR</a><br>
  <P>
-Line 2332 
 Cambridge CB2 3QH, England.
+Line 2388 
 Cambridge CB2 3QH, England.
  </P>
  <br><a name="SEC28" href="#TOC1">REVISION</a><br>
  <P>
- Last updated: 22 September 2009
+ Last updated: 04 October 2009
  <br>
  Copyright &copy; 1997-2009 University of Cambridge.
  <br>

 Legend:



Removed from v.454
 


changed lines


 
Added in v.461
 Legend:



Removed from v.454
 


changed lines


 
Added in v.461
-Removed from v.454
+Added in v.461

	ViewVC Help
Powered by ViewVC 1.1.5