libraries/spcre/libpcre/pcre/doc/html/pcrematching.html
author Tom Sutcliffe <thomas.sutcliffe@accenture.com>
Tue, 07 Dec 2010 17:29:09 +0000
changeset 114 ceac7084e2e5
parent 0 7f656887cf89
permissions -rw-r--r--
Implemented RObjectIx-based memoryaccess APIs. Upshot is that objinfo now works again on platforms that define FSHELL_NO_DOBJECTIX_SUPPORT.
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
0
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     1
<html>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     2
<head>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     3
<title>pcrematching specification</title>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     4
</head>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     5
<body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     6
<h1>pcrematching man page</h1>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     7
<p>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     8
Return to the <a href="index.html">PCRE index page</a>.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
     9
</p>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    10
<p>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    11
This page is part of the PCRE HTML documentation. It was generated automatically
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    12
from the original man page. If there is any nonsense in it, please consult the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    13
man page, in case the conversion went wrong.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    14
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    15
<ul>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    16
<li><a name="TOC1" href="#SEC1">PCRE MATCHING ALGORITHMS</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    17
<li><a name="TOC2" href="#SEC2">REGULAR EXPRESSIONS AS TREES</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    18
<li><a name="TOC3" href="#SEC3">THE STANDARD MATCHING ALGORITHM</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    19
<li><a name="TOC4" href="#SEC4">THE ALTERNATIVE MATCHING ALGORITHM</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    20
<li><a name="TOC5" href="#SEC5">ADVANTAGES OF THE ALTERNATIVE ALGORITHM</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    21
<li><a name="TOC6" href="#SEC6">DISADVANTAGES OF THE ALTERNATIVE ALGORITHM</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    22
<li><a name="TOC7" href="#SEC7">AUTHOR</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    23
<li><a name="TOC8" href="#SEC8">REVISION</a>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    24
</ul>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    25
<br><a name="SEC1" href="#TOC1">PCRE MATCHING ALGORITHMS</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    26
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    27
This document describes the two different algorithms that are available in PCRE
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    28
for matching a compiled regular expression against a given subject string. The
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    29
"standard" algorithm is the one provided by the <b>pcre_exec()</b> function.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    30
This works in the same was as Perl's matching function, and provides a
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    31
Perl-compatible matching operation.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    32
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    33
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    34
An alternative algorithm is provided by the <b>pcre_dfa_exec()</b> function;
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    35
this operates in a different way, and is not Perl-compatible. It has advantages
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    36
and disadvantages compared with the standard algorithm, and these are described
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    37
below.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    38
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    39
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    40
When there is only one possible way in which a given subject string can match a
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    41
pattern, the two algorithms give the same answer. A difference arises, however,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    42
when there are multiple possibilities. For example, if the pattern
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    43
<pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    44
  ^&#60;.*&#62;
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    45
</pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    46
is matched against the string
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    47
<pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    48
  &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    49
</pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    50
there are three possible answers. The standard algorithm finds only one of
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    51
them, whereas the alternative algorithm finds all three.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    52
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    53
<br><a name="SEC2" href="#TOC1">REGULAR EXPRESSIONS AS TREES</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    54
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    55
The set of strings that are matched by a regular expression can be represented
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    56
as a tree structure. An unlimited repetition in the pattern makes the tree of
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    57
infinite size, but it is still a tree. Matching the pattern to a given subject
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    58
string (from a given starting point) can be thought of as a search of the tree.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    59
There are two ways to search a tree: depth-first and breadth-first, and these
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    60
correspond to the two matching algorithms provided by PCRE.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    61
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    62
<br><a name="SEC3" href="#TOC1">THE STANDARD MATCHING ALGORITHM</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    63
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    64
In the terminology of Jeffrey Friedl's book "Mastering Regular
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    65
Expressions", the standard algorithm is an "NFA algorithm". It conducts a
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    66
depth-first search of the pattern tree. That is, it proceeds along a single
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    67
path through the tree, checking that the subject matches what is required. When
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    68
there is a mismatch, the algorithm tries any alternatives at the current point,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    69
and if they all fail, it backs up to the previous branch point in the tree, and
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    70
tries the next alternative branch at that level. This often involves backing up
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    71
(moving to the left) in the subject string as well. The order in which
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    72
repetition branches are tried is controlled by the greedy or ungreedy nature of
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    73
the quantifier.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    74
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    75
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    76
If a leaf node is reached, a matching string has been found, and at that point
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    77
the algorithm stops. Thus, if there is more than one possible match, this
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    78
algorithm returns the first one that it finds. Whether this is the shortest,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    79
the longest, or some intermediate length depends on the way the greedy and
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    80
ungreedy repetition quantifiers are specified in the pattern.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    81
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    82
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    83
Because it ends up with a single path through the tree, it is relatively
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    84
straightforward for this algorithm to keep track of the substrings that are
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    85
matched by portions of the pattern in parentheses. This provides support for
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    86
capturing parentheses and back references.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    87
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    88
<br><a name="SEC4" href="#TOC1">THE ALTERNATIVE MATCHING ALGORITHM</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    89
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    90
This algorithm conducts a breadth-first search of the tree. Starting from the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    91
first matching point in the subject, it scans the subject string from left to
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    92
right, once, character by character, and as it does this, it remembers all the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    93
paths through the tree that represent valid matches. In Friedl's terminology,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    94
this is a kind of "DFA algorithm", though it is not implemented as a
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    95
traditional finite state machine (it keeps multiple states active
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    96
simultaneously).
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    97
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    98
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
    99
The scan continues until either the end of the subject is reached, or there are
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   100
no more unterminated paths. At this point, terminated paths represent the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   101
different matching possibilities (if there are none, the match has failed).
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   102
Thus, if there is more than one possible match, this algorithm finds all of
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   103
them, and in particular, it finds the longest. In PCRE, there is an option to
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   104
stop the algorithm after the first match (which is necessarily the shortest)
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   105
has been found.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   106
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   107
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   108
Note that all the matches that are found start at the same point in the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   109
subject. If the pattern
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   110
<pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   111
  cat(er(pillar)?)
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   112
</pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   113
is matched against the string "the caterpillar catchment", the result will be
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   114
the three strings "cat", "cater", and "caterpillar" that start at the fourth
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   115
character of the subject. The algorithm does not automatically move on to find
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   116
matches that start at later positions.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   117
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   118
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   119
There are a number of features of PCRE regular expressions that are not
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   120
supported by the alternative matching algorithm. They are as follows:
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   121
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   122
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   123
1. Because the algorithm finds all possible matches, the greedy or ungreedy
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   124
nature of repetition quantifiers is not relevant. Greedy and ungreedy
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   125
quantifiers are treated in exactly the same way. However, possessive
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   126
quantifiers can make a difference when what follows could also match what is
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   127
quantified, for example in a pattern like this:
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   128
<pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   129
  ^a++\w!
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   130
</pre>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   131
This pattern matches "aaab!" but not "aaa!", which would be matched by a
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   132
non-possessive quantifier. Similarly, if an atomic group is present, it is
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   133
matched as if it were a standalone pattern at the current point, and the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   134
longest match is then "locked in" for the rest of the overall pattern.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   135
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   136
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   137
2. When dealing with multiple paths through the tree simultaneously, it is not
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   138
straightforward to keep track of captured substrings for the different matching
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   139
possibilities, and PCRE's implementation of this algorithm does not attempt to
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   140
do this. This means that no captured substrings are available.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   141
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   142
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   143
3. Because no substrings are captured, back references within the pattern are
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   144
not supported, and cause errors if encountered.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   145
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   146
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   147
4. For the same reason, conditional expressions that use a backreference as the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   148
condition or test for a specific group recursion are not supported.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   149
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   150
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   151
5. Because many paths through the tree may be active, the \K escape sequence,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   152
which resets the start of the match when encountered (but may be on some paths
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   153
and not on others), is not supported. It causes an error if encountered.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   154
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   155
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   156
6. Callouts are supported, but the value of the <i>capture_top</i> field is
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   157
always 1, and the value of the <i>capture_last</i> field is always -1.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   158
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   159
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   160
7. The \C escape sequence, which (in the standard algorithm) matches a single
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   161
byte, even in UTF-8 mode, is not supported because the alternative algorithm
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   162
moves through the subject string one character at a time, for all active paths
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   163
through the tree.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   164
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   165
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   166
8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE) are not
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   167
supported. (*FAIL) is supported, and behaves like a failing negative assertion.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   168
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   169
<br><a name="SEC5" href="#TOC1">ADVANTAGES OF THE ALTERNATIVE ALGORITHM</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   170
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   171
Using the alternative matching algorithm provides the following advantages:
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   172
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   173
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   174
1. All possible matches (at a single point in the subject) are automatically
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   175
found, and in particular, the longest match is found. To find more than one
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   176
match using the standard algorithm, you have to do kludgy things with
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   177
callouts.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   178
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   179
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   180
2. There is much better support for partial matching. The restrictions on the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   181
content of the pattern that apply when using the standard algorithm for partial
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   182
matching do not apply to the alternative algorithm. For non-anchored patterns,
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   183
the starting position of a partial match is available.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   184
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   185
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   186
3. Because the alternative algorithm scans the subject string just once, and
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   187
never needs to backtrack, it is possible to pass very long subject strings to
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   188
the matching function in several pieces, checking for partial matching each
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   189
time.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   190
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   191
<br><a name="SEC6" href="#TOC1">DISADVANTAGES OF THE ALTERNATIVE ALGORITHM</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   192
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   193
The alternative algorithm suffers from a number of disadvantages:
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   194
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   195
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   196
1. It is substantially slower than the standard algorithm. This is partly
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   197
because it has to search for all possible matches, but is also because it is
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   198
less susceptible to optimization.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   199
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   200
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   201
2. Capturing parentheses and back references are not supported.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   202
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   203
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   204
3. Although atomic groups are supported, their use does not provide the
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   205
performance advantage that it does for the standard algorithm.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   206
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   207
<br><a name="SEC7" href="#TOC1">AUTHOR</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   208
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   209
Philip Hazel
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   210
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   211
University Computing Service
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   212
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   213
Cambridge CB2 3QH, England.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   214
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   215
</P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   216
<br><a name="SEC8" href="#TOC1">REVISION</a><br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   217
<P>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   218
Last updated: 19 April 2008
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   219
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   220
Copyright &copy; 1997-2008 University of Cambridge.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   221
<br>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   222
<p>
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   223
Return to the <a href="index.html">PCRE index page</a>.
7f656887cf89 First submission to Symbian Foundation staging server.
Tom Sutcliffe <thomas.sutcliffe@accenture.com>
parents:
diff changeset
   224
</p>