ViewVC logotype

Contents of /code/trunk/doc/html/pcrematching.html

Parent Directory Parent Directory | Revision Log Revision Log

Revision 836 - (show annotations)
Wed Dec 28 17:16:11 2011 UTC (7 years, 8 months ago) by ph10
File MIME type: text/html
File size: 10055 byte(s)
Merging all the changes from the pcre16 branch into the trunk.
1 <html>
2 <head>
3 <title>pcrematching specification</title>
4 </head>
5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 <h1>pcrematching 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">PCRE MATCHING ALGORITHMS</a>
17 <li><a name="TOC2" href="#SEC2">REGULAR EXPRESSIONS AS TREES</a>
18 <li><a name="TOC3" href="#SEC3">THE STANDARD MATCHING ALGORITHM</a>
22 <li><a name="TOC7" href="#SEC7">AUTHOR</a>
23 <li><a name="TOC8" href="#SEC8">REVISION</a>
24 </ul>
25 <br><a name="SEC1" href="#TOC1">PCRE MATCHING ALGORITHMS</a><br>
26 <P>
27 This document describes the two different algorithms that are available in PCRE
28 for matching a compiled regular expression against a given subject string. The
29 "standard" algorithm is the one provided by the <b>pcre_exec()</b> function.
30 This works in the same was as Perl's matching function, and provides a
31 Perl-compatible matching operation.
32 </P>
33 <P>
34 An alternative algorithm is provided by the <b>pcre_dfa_exec()</b> function;
35 this operates in a different way, and is not Perl-compatible. It has advantages
36 and disadvantages compared with the standard algorithm, and these are described
37 below.
38 </P>
39 <P>
40 When there is only one possible way in which a given subject string can match a
41 pattern, the two algorithms give the same answer. A difference arises, however,
42 when there are multiple possibilities. For example, if the pattern
43 <pre>
44 ^&#60;.*&#62;
45 </pre>
46 is matched against the string
47 <pre>
48 &#60;something&#62; &#60;something else&#62; &#60;something further&#62;
49 </pre>
50 there are three possible answers. The standard algorithm finds only one of
51 them, whereas the alternative algorithm finds all three.
52 </P>
53 <br><a name="SEC2" href="#TOC1">REGULAR EXPRESSIONS AS TREES</a><br>
54 <P>
55 The set of strings that are matched by a regular expression can be represented
56 as a tree structure. An unlimited repetition in the pattern makes the tree of
57 infinite size, but it is still a tree. Matching the pattern to a given subject
58 string (from a given starting point) can be thought of as a search of the tree.
59 There are two ways to search a tree: depth-first and breadth-first, and these
60 correspond to the two matching algorithms provided by PCRE.
61 </P>
62 <br><a name="SEC3" href="#TOC1">THE STANDARD MATCHING ALGORITHM</a><br>
63 <P>
64 In the terminology of Jeffrey Friedl's book "Mastering Regular
65 Expressions", the standard algorithm is an "NFA algorithm". It conducts a
66 depth-first search of the pattern tree. That is, it proceeds along a single
67 path through the tree, checking that the subject matches what is required. When
68 there is a mismatch, the algorithm tries any alternatives at the current point,
69 and if they all fail, it backs up to the previous branch point in the tree, and
70 tries the next alternative branch at that level. This often involves backing up
71 (moving to the left) in the subject string as well. The order in which
72 repetition branches are tried is controlled by the greedy or ungreedy nature of
73 the quantifier.
74 </P>
75 <P>
76 If a leaf node is reached, a matching string has been found, and at that point
77 the algorithm stops. Thus, if there is more than one possible match, this
78 algorithm returns the first one that it finds. Whether this is the shortest,
79 the longest, or some intermediate length depends on the way the greedy and
80 ungreedy repetition quantifiers are specified in the pattern.
81 </P>
82 <P>
83 Because it ends up with a single path through the tree, it is relatively
84 straightforward for this algorithm to keep track of the substrings that are
85 matched by portions of the pattern in parentheses. This provides support for
86 capturing parentheses and back references.
87 </P>
88 <br><a name="SEC4" href="#TOC1">THE ALTERNATIVE MATCHING ALGORITHM</a><br>
89 <P>
90 This algorithm conducts a breadth-first search of the tree. Starting from the
91 first matching point in the subject, it scans the subject string from left to
92 right, once, character by character, and as it does this, it remembers all the
93 paths through the tree that represent valid matches. In Friedl's terminology,
94 this is a kind of "DFA algorithm", though it is not implemented as a
95 traditional finite state machine (it keeps multiple states active
96 simultaneously).
97 </P>
98 <P>
99 Although the general principle of this matching algorithm is that it scans the
100 subject string only once, without backtracking, there is one exception: when a
101 lookaround assertion is encountered, the characters following or preceding the
102 current point have to be independently inspected.
103 </P>
104 <P>
105 The scan continues until either the end of the subject is reached, or there are
106 no more unterminated paths. At this point, terminated paths represent the
107 different matching possibilities (if there are none, the match has failed).
108 Thus, if there is more than one possible match, this algorithm finds all of
109 them, and in particular, it finds the longest. The matches are returned in
110 decreasing order of length. There is an option to stop the algorithm after the
111 first match (which is necessarily the shortest) is found.
112 </P>
113 <P>
114 Note that all the matches that are found start at the same point in the
115 subject. If the pattern
116 <pre>
117 cat(er(pillar)?)?
118 </pre>
119 is matched against the string "the caterpillar catchment", the result will be
120 the three strings "caterpillar", "cater", and "cat" that start at the fifth
121 character of the subject. The algorithm does not automatically move on to find
122 matches that start at later positions.
123 </P>
124 <P>
125 There are a number of features of PCRE regular expressions that are not
126 supported by the alternative matching algorithm. They are as follows:
127 </P>
128 <P>
129 1. Because the algorithm finds all possible matches, the greedy or ungreedy
130 nature of repetition quantifiers is not relevant. Greedy and ungreedy
131 quantifiers are treated in exactly the same way. However, possessive
132 quantifiers can make a difference when what follows could also match what is
133 quantified, for example in a pattern like this:
134 <pre>
135 ^a++\w!
136 </pre>
137 This pattern matches "aaab!" but not "aaa!", which would be matched by a
138 non-possessive quantifier. Similarly, if an atomic group is present, it is
139 matched as if it were a standalone pattern at the current point, and the
140 longest match is then "locked in" for the rest of the overall pattern.
141 </P>
142 <P>
143 2. When dealing with multiple paths through the tree simultaneously, it is not
144 straightforward to keep track of captured substrings for the different matching
145 possibilities, and PCRE's implementation of this algorithm does not attempt to
146 do this. This means that no captured substrings are available.
147 </P>
148 <P>
149 3. Because no substrings are captured, back references within the pattern are
150 not supported, and cause errors if encountered.
151 </P>
152 <P>
153 4. For the same reason, conditional expressions that use a backreference as the
154 condition or test for a specific group recursion are not supported.
155 </P>
156 <P>
157 5. Because many paths through the tree may be active, the \K escape sequence,
158 which resets the start of the match when encountered (but may be on some paths
159 and not on others), is not supported. It causes an error if encountered.
160 </P>
161 <P>
162 6. Callouts are supported, but the value of the <i>capture_top</i> field is
163 always 1, and the value of the <i>capture_last</i> field is always -1.
164 </P>
165 <P>
166 7. The \C escape sequence, which (in the standard algorithm) matches a single
167 byte, even in UTF-8 mode, is not supported in UTF-8 mode, because the
168 alternative algorithm moves through the subject string one character at a time,
169 for all active paths through the tree.
170 </P>
171 <P>
172 8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE) are not
173 supported. (*FAIL) is supported, and behaves like a failing negative assertion.
174 </P>
175 <br><a name="SEC5" href="#TOC1">ADVANTAGES OF THE ALTERNATIVE ALGORITHM</a><br>
176 <P>
177 Using the alternative matching algorithm provides the following advantages:
178 </P>
179 <P>
180 1. All possible matches (at a single point in the subject) are automatically
181 found, and in particular, the longest match is found. To find more than one
182 match using the standard algorithm, you have to do kludgy things with
183 callouts.
184 </P>
185 <P>
186 2. Because the alternative algorithm scans the subject string just once, and
187 never needs to backtrack, it is possible to pass very long subject strings to
188 the matching function in several pieces, checking for partial matching each
189 time. Although it is possible to do multi-segment matching using the standard
190 algorithm (<b>pcre_exec()</b>), by retaining partially matched substrings, it is
191 more complicated. The
192 <a href="pcrepartial.html"><b>pcrepartial</b></a>
193 documentation gives details of partial matching and discusses multi-segment
194 matching.
195 </P>
197 <P>
198 The alternative algorithm suffers from a number of disadvantages:
199 </P>
200 <P>
201 1. It is substantially slower than the standard algorithm. This is partly
202 because it has to search for all possible matches, but is also because it is
203 less susceptible to optimization.
204 </P>
205 <P>
206 2. Capturing parentheses and back references are not supported.
207 </P>
208 <P>
209 3. Although atomic groups are supported, their use does not provide the
210 performance advantage that it does for the standard algorithm.
211 </P>
212 <br><a name="SEC7" href="#TOC1">AUTHOR</a><br>
213 <P>
214 Philip Hazel
215 <br>
216 University Computing Service
217 <br>
218 Cambridge CB2 3QH, England.
219 <br>
220 </P>
221 <br><a name="SEC8" href="#TOC1">REVISION</a><br>
222 <P>
223 Last updated: 19 November 2011
224 <br>
225 Copyright &copy; 1997-2010 University of Cambridge.
226 <br>
227 <p>
228 Return to the <a href="index.html">PCRE index page</a>.
229 </p>


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

  ViewVC Help
Powered by ViewVC 1.1.5