ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 172 - (show annotations)
Tue Jun 5 10:40:13 2007 UTC (14 years, 3 months ago) by ph10
File MIME type: text/html
File size: 9867 byte(s)
Error occurred while calculating annotation data.
Drastically reduce workspace used for alternatives in groups; also some 
trailing space removals for a test release.
1 <html>
2 <head>
3 <title>pcrepartial specification</title>
4 </head>
5 <body bgcolor="#FFFFFF" text="#00005A" link="#0066FF" alink="#3399FF" vlink="#2222BB">
6 <h1>pcrepartial 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">PARTIAL MATCHING IN PCRE</a>
19 <li><a name="TOC4" href="#SEC4">MULTI-SEGMENT MATCHING WITH pcre_dfa_exec()</a>
20 <li><a name="TOC5" href="#SEC5">AUTHOR</a>
21 <li><a name="TOC6" href="#SEC6">REVISION</a>
22 </ul>
23 <br><a name="SEC1" href="#TOC1">PARTIAL MATCHING IN PCRE</a><br>
24 <P>
25 In normal use of PCRE, if the subject string that is passed to
26 <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b> matches as far as it goes, but is
27 too short to match the entire pattern, PCRE_ERROR_NOMATCH is returned. There
28 are circumstances where it might be helpful to distinguish this case from other
29 cases in which there is no match.
30 </P>
31 <P>
32 Consider, for example, an application where a human is required to type in data
33 for a field with specific formatting requirements. An example might be a date
34 in the form <i>ddmmmyy</i>, defined by this pattern:
35 <pre>
36 ^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
37 </pre>
38 If the application sees the user's keystrokes one by one, and can check that
39 what has been typed so far is potentially valid, it is able to raise an error
40 as soon as a mistake is made, possibly beeping and not reflecting the
41 character that has been typed. This immediate feedback is likely to be a better
42 user interface than a check that is delayed until the entire string has been
43 entered.
44 </P>
45 <P>
46 PCRE supports the concept of partial matching by means of the PCRE_PARTIAL
47 option, which can be set when calling <b>pcre_exec()</b> or
48 <b>pcre_dfa_exec()</b>. When this flag is set for <b>pcre_exec()</b>, the return
49 code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if at any time
50 during the matching process the last part of the subject string matched part of
51 the pattern. Unfortunately, for non-anchored matching, it is not possible to
52 obtain the position of the start of the partial match. No captured data is set
53 when PCRE_ERROR_PARTIAL is returned.
54 </P>
55 <P>
56 When PCRE_PARTIAL is set for <b>pcre_dfa_exec()</b>, the return code
57 PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end of the
58 subject is reached, there have been no complete matches, but there is still at
59 least one matching possibility. The portion of the string that provided the
60 partial match is set as the first matching string.
61 </P>
62 <P>
63 Using PCRE_PARTIAL disables one of PCRE's optimizations. PCRE remembers the
64 last literal byte in a pattern, and abandons matching immediately if such a
65 byte is not present in the subject string. This optimization cannot be used
66 for a subject string that might match only partially.
67 </P>
68 <br><a name="SEC2" href="#TOC1">RESTRICTED PATTERNS FOR PCRE_PARTIAL</a><br>
69 <P>
70 Because of the way certain internal optimizations are implemented in the
71 <b>pcre_exec()</b> function, the PCRE_PARTIAL option cannot be used with all
72 patterns. These restrictions do not apply when <b>pcre_dfa_exec()</b> is used.
73 For <b>pcre_exec()</b>, repeated single characters such as
74 <pre>
75 a{2,4}
76 </pre>
77 and repeated single metasequences such as
78 <pre>
79 \d+
80 </pre>
81 are not permitted if the maximum number of occurrences is greater than one.
82 Optional items such as \d? (where the maximum is one) are permitted.
83 Quantifiers with any values are permitted after parentheses, so the invalid
84 examples above can be coded thus:
85 <pre>
86 (a){2,4}
87 (\d)+
88 </pre>
89 These constructions run more slowly, but for the kinds of application that are
90 envisaged for this facility, this is not felt to be a major restriction.
91 </P>
92 <P>
93 If PCRE_PARTIAL is set for a pattern that does not conform to the restrictions,
94 <b>pcre_exec()</b> returns the error code PCRE_ERROR_BADPARTIAL (-13).
95 You can use the PCRE_INFO_OKPARTIAL call to <b>pcre_fullinfo()</b> to find out
96 if a compiled pattern can be used for partial matching.
97 </P>
99 <P>
100 If the escape sequence \P is present in a <b>pcretest</b> data line, the
101 PCRE_PARTIAL flag is used for the match. Here is a run of <b>pcretest</b> that
102 uses the date example quoted above:
103 <pre>
104 re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
105 data&#62; 25jun04\P
106 0: 25jun04
107 1: jun
108 data&#62; 25dec3\P
109 Partial match
110 data&#62; 3ju\P
111 Partial match
112 data&#62; 3juj\P
113 No match
114 data&#62; j\P
115 No match
116 </pre>
117 The first data string is matched completely, so <b>pcretest</b> shows the
118 matched substrings. The remaining four strings do not match the complete
119 pattern, but the first two are partial matches. The same test, using
120 <b>pcre_dfa_exec()</b> matching (by means of the \D escape sequence), produces
121 the following output:
122 <pre>
123 re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
124 data&#62; 25jun04\P\D
125 0: 25jun04
126 data&#62; 23dec3\P\D
127 Partial match: 23dec3
128 data&#62; 3ju\P\D
129 Partial match: 3ju
130 data&#62; 3juj\P\D
131 No match
132 data&#62; j\P\D
133 No match
134 </pre>
135 Notice that in this case the portion of the string that was matched is made
136 available.
137 </P>
138 <br><a name="SEC4" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre_dfa_exec()</a><br>
139 <P>
140 When a partial match has been found using <b>pcre_dfa_exec()</b>, it is possible
141 to continue the match by providing additional subject data and calling
142 <b>pcre_dfa_exec()</b> again with the same compiled regular expression, this
143 time setting the PCRE_DFA_RESTART option. You must also pass the same working
144 space as before, because this is where details of the previous partial match
145 are stored. Here is an example using <b>pcretest</b>, using the \R escape
146 sequence to set the PCRE_DFA_RESTART option (\P and \D are as above):
147 <pre>
148 re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
149 data&#62; 23ja\P\D
150 Partial match: 23ja
151 data&#62; n05\R\D
152 0: n05
153 </pre>
154 The first call has "23ja" as the subject, and requests partial matching; the
155 second call has "n05" as the subject for the continued (restarted) match.
156 Notice that when the match is complete, only the last part is shown; PCRE does
157 not retain the previously partially-matched string. It is up to the calling
158 program to do that if it needs to.
159 </P>
160 <P>
161 You can set PCRE_PARTIAL with PCRE_DFA_RESTART to continue partial matching
162 over multiple segments. This facility can be used to pass very long subject
163 strings to <b>pcre_dfa_exec()</b>. However, some care is needed for certain
164 types of pattern.
165 </P>
166 <P>
167 1. If the pattern contains tests for the beginning or end of a line, you need
168 to pass the PCRE_NOTBOL or PCRE_NOTEOL options, as appropriate, when the
169 subject string for any call does not contain the beginning or end of a line.
170 </P>
171 <P>
172 2. If the pattern contains backward assertions (including \b or \B), you need
173 to arrange for some overlap in the subject strings to allow for this. For
174 example, you could pass the subject in chunks that are 500 bytes long, but in
175 a buffer of 700 bytes, with the starting offset set to 200 and the previous 200
176 bytes at the start of the buffer.
177 </P>
178 <P>
179 3. Matching a subject string that is split into multiple segments does not
180 always produce exactly the same result as matching over one single long string.
181 The difference arises when there are multiple matching possibilities, because a
182 partial match result is given only when there are no completed matches in a
183 call to <b>pcre_dfa_exec()</b>. This means that as soon as the shortest match has
184 been found, continuation to a new subject segment is no longer possible.
185 Consider this <b>pcretest</b> example:
186 <pre>
187 re&#62; /dog(sbody)?/
188 data&#62; do\P\D
189 Partial match: do
190 data&#62; gsb\R\P\D
191 0: g
192 data&#62; dogsbody\D
193 0: dogsbody
194 1: dog
195 </pre>
196 The pattern matches the words "dog" or "dogsbody". When the subject is
197 presented in several parts ("do" and "gsb" being the first two) the match stops
198 when "dog" has been found, and it is not possible to continue. On the other
199 hand, if "dogsbody" is presented as a single string, both matches are found.
200 </P>
201 <P>
202 Because of this phenomenon, it does not usually make sense to end a pattern
203 that is going to be matched in this way with a variable repeat.
204 </P>
205 <P>
206 4. Patterns that contain alternatives at the top level which do not all
207 start with the same pattern item may not work as expected. For example,
208 consider this pattern:
209 <pre>
210 1234|3789
211 </pre>
212 If the first part of the subject is "ABC123", a partial match of the first
213 alternative is found at offset 3. There is no partial match for the second
214 alternative, because such a match does not start at the same point in the
215 subject string. Attempting to continue with the string "789" does not yield a
216 match because only those alternatives that match at one point in the subject
217 are remembered. The problem arises because the start of the second alternative
218 matches within the first alternative. There is no problem with anchored
219 patterns or patterns such as:
220 <pre>
221 1234|ABCD
222 </pre>
223 where no string can be a partial match for both alternatives.
224 </P>
225 <br><a name="SEC5" href="#TOC1">AUTHOR</a><br>
226 <P>
227 Philip Hazel
228 <br>
229 University Computing Service
230 <br>
231 Cambridge CB2 3QH, England.
232 <br>
233 </P>
234 <br><a name="SEC6" href="#TOC1">REVISION</a><br>
235 <P>
236 Last updated: 04 June 2007
237 <br>
238 Copyright &copy; 1997-2007 University of Cambridge.
239 <br>
240 <p>
241 Return to the <a href="index.html">PCRE index page</a>.
242 </p>


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

  ViewVC Help
Powered by ViewVC 1.1.5