ViewVC logotype

Contents of /code/trunk/doc/pcrepartial.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 468 - (show annotations)
Mon Oct 19 11:46:26 2009 UTC (11 years, 6 months ago) by ph10
File size: 16290 byte(s)
Error occurred while calculating annotation data.
Fix typos and other things.
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 In normal use of PCRE, if the subject string that is passed to
8 \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP matches as far as it goes, but is
9 too short to match the entire pattern, PCRE_ERROR_NOMATCH is returned. There
10 are circumstances where it might be helpful to distinguish this case from other
11 cases in which there is no match.
12 .P
13 Consider, for example, an application where a human is required to type in data
14 for a field with specific formatting requirements. An example might be a date
15 in the form \fIddmmmyy\fP, defined by this pattern:
16 .sp
17 ^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$
18 .sp
19 If the application sees the user's keystrokes one by one, and can check that
20 what has been typed so far is potentially valid, it is able to raise an error
21 as soon as a mistake is made, by beeping and not reflecting the character that
22 has been typed, for example. This immediate feedback is likely to be a better
23 user interface than a check that is delayed until the entire string has been
24 entered. Partial matching can also sometimes be useful when the subject string
25 is very long and is not all available at once.
26 .P
27 PCRE supports partial matching by means of the PCRE_PARTIAL_SOFT and
28 PCRE_PARTIAL_HARD options, which can be set when calling \fBpcre_exec()\fP or
29 \fBpcre_dfa_exec()\fP. For backwards compatibility, PCRE_PARTIAL is a synonym
30 for PCRE_PARTIAL_SOFT. The essential difference between the two options is
31 whether or not a partial match is preferred to an alternative complete match,
32 though the details differ between the two matching functions. If both options
33 are set, PCRE_PARTIAL_HARD takes precedence.
34 .P
35 Setting a partial matching option disables two of PCRE's optimizations. PCRE
36 remembers the last literal byte in a pattern, and abandons matching immediately
37 if such a byte is not present in the subject string. This optimization cannot
38 be used for a subject string that might match only partially. If the pattern
39 was studied, PCRE knows the minimum length of a matching string, and does not
40 bother to run the matching function on shorter strings. This optimization is
41 also disabled for partial matching.
42 .
43 .
45 .rs
46 .sp
47 A partial match occurs during a call to \fBpcre_exec()\fP whenever the end of
48 the subject string is reached successfully, but matching cannot continue
49 because more characters are needed. However, at least one character must have
50 been matched. (In other words, a partial match can never be an empty string.)
51 .P
52 If PCRE_PARTIAL_SOFT is set, the partial match is remembered, but matching
53 continues as normal, and other alternatives in the pattern are tried. If no
54 complete match can be found, \fBpcre_exec()\fP returns PCRE_ERROR_PARTIAL
55 instead of PCRE_ERROR_NOMATCH. If there are at least two slots in the offsets
56 vector, the first of them is set to the offset of the earliest character that
57 was inspected when the partial match was found. For convenience, the second
58 offset points to the end of the string so that a substring can easily be
59 identified.
60 .P
61 For the majority of patterns, the first offset identifies the start of the
62 partially matched string. However, for patterns that contain lookbehind
63 assertions, or \eK, or begin with \eb or \eB, earlier characters have been
64 inspected while carrying out the match. For example:
65 .sp
66 /(?<=abc)123/
67 .sp
68 This pattern matches "123", but only if it is preceded by "abc". If the subject
69 string is "xyzabc12", the offsets after a partial match are for the substring
70 "abc12", because all these characters are needed if another match is tried
71 with extra characters added.
72 .P
73 If there is more than one partial match, the first one that was found provides
74 the data that is returned. Consider this pattern:
75 .sp
76 /123\ew+X|dogY/
77 .sp
78 If this is matched against the subject string "abc123dog", both
79 alternatives fail to match, but the end of the subject is reached during
80 matching, so PCRE_ERROR_PARTIAL is returned instead of PCRE_ERROR_NOMATCH. The
81 offsets are set to 3 and 9, identifying "123dog" as the first partial match
82 that was found. (In this example, there are two partial matches, because "dog"
83 on its own partially matches the second alternative.)
84 .P
85 If PCRE_PARTIAL_HARD is set for \fBpcre_exec()\fP, it returns
86 PCRE_ERROR_PARTIAL as soon as a partial match is found, without continuing to
87 search for possible complete matches. The difference between the two options
88 can be illustrated by a pattern such as:
89 .sp
90 /dog(sbody)?/
91 .sp
92 This matches either "dog" or "dogsbody", greedily (that is, it prefers the
93 longer string if possible). If it is matched against the string "dog" with
94 PCRE_PARTIAL_SOFT, it yields a complete match for "dog". However, if
95 PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL. On the other hand,
96 if the pattern is made ungreedy the result is different:
97 .sp
98 /dog(sbody)??/
99 .sp
100 In this case the result is always a complete match because \fBpcre_exec()\fP
101 finds that first, and it never continues after finding a match. It might be
102 easier to follow this explanation by thinking of the two patterns like this:
103 .sp
104 /dog(sbody)?/ is the same as /dogsbody|dog/
105 /dog(sbody)??/ is the same as /dog|dogsbody/
106 .sp
107 The second pattern will never match "dogsbody" when \fBpcre_exec()\fP is
108 used, because it will always find the shorter match first.
109 .
110 .
111 .SH "PARTIAL MATCHING USING pcre_dfa_exec()"
112 .rs
113 .sp
114 The \fBpcre_dfa_exec()\fP function moves along the subject string character by
115 character, without backtracking, searching for all possible matches
116 simultaneously. If the end of the subject is reached before the end of the
117 pattern, there is the possibility of a partial match, again provided that at
118 least one character has matched.
119 .P
120 When PCRE_PARTIAL_SOFT is set, PCRE_ERROR_PARTIAL is returned only if there
121 have been no complete matches. Otherwise, the complete matches are returned.
122 However, if PCRE_PARTIAL_HARD is set, a partial match takes precedence over any
123 complete matches. The portion of the string that was inspected when the longest
124 partial match was found is set as the first matching string, provided there are
125 at least two slots in the offsets vector.
126 .P
127 Because \fBpcre_dfa_exec()\fP always searches for all possible matches, and
128 there is no difference between greedy and ungreedy repetition, its behaviour is
129 different from \fBpcre_exec\fP when PCRE_PARTIAL_HARD is set. Consider the
130 string "dog" matched against the ungreedy pattern shown above:
131 .sp
132 /dog(sbody)??/
133 .sp
134 Whereas \fBpcre_exec()\fP stops as soon as it finds the complete match for
135 "dog", \fBpcre_dfa_exec()\fP also finds the partial match for "dogsbody", and
136 so returns that when PCRE_PARTIAL_HARD is set.
137 .
138 .
140 .rs
141 .sp
142 If a pattern ends with one of sequences \eb or \eB, which test for word
143 boundaries, partial matching with PCRE_PARTIAL_SOFT can give counter-intuitive
144 results. Consider this pattern:
145 .sp
146 /\ebcat\eb/
147 .sp
148 This matches "cat", provided there is a word boundary at either end. If the
149 subject string is "the cat", the comparison of the final "t" with a following
150 character cannot take place, so a partial match is found. However,
151 \fBpcre_exec()\fP carries on with normal matching, which matches \eb at the end
152 of the subject when the last character is a letter, thus finding a complete
153 match. The result, therefore, is \fInot\fP PCRE_ERROR_PARTIAL. The same thing
154 happens with \fBpcre_dfa_exec()\fP, because it also finds the complete match.
155 .P
156 Using PCRE_PARTIAL_HARD in this case does yield PCRE_ERROR_PARTIAL, because
157 then the partial match takes precedence.
158 .
159 .
161 .rs
162 .sp
163 For releases of PCRE prior to 8.00, because of the way certain internal
164 optimizations were implemented in the \fBpcre_exec()\fP function, the
165 PCRE_PARTIAL option (predecessor of PCRE_PARTIAL_SOFT) could not be used with
166 all patterns. From release 8.00 onwards, the restrictions no longer apply, and
167 partial matching with \fBpcre_exec()\fP can be requested for any pattern.
168 .P
169 Items that were formerly restricted were repeated single characters and
170 repeated metasequences. If PCRE_PARTIAL was set for a pattern that did not
171 conform to the restrictions, \fBpcre_exec()\fP returned the error code
172 PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in use. The
173 PCRE_INFO_OKPARTIAL call to \fBpcre_fullinfo()\fP to find out if a compiled
174 pattern can be used for partial matching now always returns 1.
175 .
176 .
178 .rs
179 .sp
180 If the escape sequence \eP is present in a \fBpcretest\fP data line, the
181 PCRE_PARTIAL_SOFT option is used for the match. Here is a run of \fBpcretest\fP
182 that uses the date example quoted above:
183 .sp
184 re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
185 data> 25jun04\eP
186 0: 25jun04
187 1: jun
188 data> 25dec3\eP
189 Partial match: 23dec3
190 data> 3ju\eP
191 Partial match: 3ju
192 data> 3juj\eP
193 No match
194 data> j\eP
195 No match
196 .sp
197 The first data string is matched completely, so \fBpcretest\fP shows the
198 matched substrings. The remaining four strings do not match the complete
199 pattern, but the first two are partial matches. Similar output is obtained
200 when \fBpcre_dfa_exec()\fP is used.
201 .P
202 If the escape sequence \eP is present more than once in a \fBpcretest\fP data
203 line, the PCRE_PARTIAL_HARD option is set for the match.
204 .
205 .
206 .SH "MULTI-SEGMENT MATCHING WITH pcre_dfa_exec()"
207 .rs
208 .sp
209 When a partial match has been found using \fBpcre_dfa_exec()\fP, it is possible
210 to continue the match by providing additional subject data and calling
211 \fBpcre_dfa_exec()\fP again with the same compiled regular expression, this
212 time setting the PCRE_DFA_RESTART option. You must pass the same working
213 space as before, because this is where details of the previous partial match
214 are stored. Here is an example using \fBpcretest\fP, using the \eR escape
215 sequence to set the PCRE_DFA_RESTART option (\eD specifies the use of
216 \fBpcre_dfa_exec()\fP):
217 .sp
218 re> /^\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed$/
219 data> 23ja\eP\eD
220 Partial match: 23ja
221 data> n05\eR\eD
222 0: n05
223 .sp
224 The first call has "23ja" as the subject, and requests partial matching; the
225 second call has "n05" as the subject for the continued (restarted) match.
226 Notice that when the match is complete, only the last part is shown; PCRE does
227 not retain the previously partially-matched string. It is up to the calling
228 program to do that if it needs to.
229 .P
230 You can set the PCRE_PARTIAL_SOFT or PCRE_PARTIAL_HARD options with
231 PCRE_DFA_RESTART to continue partial matching over multiple segments. This
232 facility can be used to pass very long subject strings to
233 \fBpcre_dfa_exec()\fP.
234 .
235 .
237 .rs
238 .sp
239 From release 8.00, \fBpcre_exec()\fP can also be used to do multi-segment
240 matching. Unlike \fBpcre_dfa_exec()\fP, it is not possible to restart the
241 previous match with a new segment of data. Instead, new data must be added to
242 the previous subject string, and the entire match re-run, starting from the
243 point where the partial match occurred. Earlier data can be discarded.
244 Consider an unanchored pattern that matches dates:
245 .sp
246 re> /\ed?\ed(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\ed\ed/
247 data> The date is 23ja\eP
248 Partial match: 23ja
249 .sp
250 At this stage, an application could discard the text preceding "23ja", add on
251 text from the next segment, and call \fBpcre_exec()\fP again. Unlike
252 \fBpcre_dfa_exec()\fP, the entire matching string must always be available, and
253 the complete matching process occurs for each call, so more memory and more
254 processing time is needed.
255 .P
256 \fBNote:\fP If the pattern contains lookbehind assertions, or \eK, or starts
257 with \eb or \eB, the string that is returned for a partial match will include
258 characters that precede the partially matched string itself, because these must
259 be retained when adding on more characters for a subsequent matching attempt.
260 .
261 .
263 .rs
264 .sp
265 Certain types of pattern may give problems with multi-segment matching,
266 whichever matching function is used.
267 .P
268 1. If the pattern contains tests for the beginning or end of a line, you need
269 to pass the PCRE_NOTBOL or PCRE_NOTEOL options, as appropriate, when the
270 subject string for any call does not contain the beginning or end of a line.
271 .P
272 2. Lookbehind assertions at the start of a pattern are catered for in the
273 offsets that are returned for a partial match. However, in theory, a lookbehind
274 assertion later in the pattern could require even earlier characters to be
275 inspected, and it might not have been reached when a partial match occurs. This
276 is probably an extremely unlikely case; you could guard against it to a certain
277 extent by always including extra characters at the start.
278 .P
279 3. Matching a subject string that is split into multiple segments may not
280 always produce exactly the same result as matching over one single long string,
281 especially when PCRE_PARTIAL_SOFT is used. The section "Partial Matching and
282 Word Boundaries" above describes an issue that arises if the pattern ends with
283 \eb or \eB. Another kind of difference may occur when there are multiple
284 matching possibilities, because a partial match result is given only when there
285 are no completed matches. This means that as soon as the shortest match has
286 been found, continuation to a new subject segment is no longer possible.
287 Consider again this \fBpcretest\fP example:
288 .sp
289 re> /dog(sbody)?/
290 data> dogsb\eP
291 0: dog
292 data> do\eP\eD
293 Partial match: do
294 data> gsb\eR\eP\eD
295 0: g
296 data> dogsbody\eD
297 0: dogsbody
298 1: dog
299 .sp
300 The first data line passes the string "dogsb" to \fBpcre_exec()\fP, setting the
301 PCRE_PARTIAL_SOFT option. Although the string is a partial match for
302 "dogsbody", the result is not PCRE_ERROR_PARTIAL, because the shorter string
303 "dog" is a complete match. Similarly, when the subject is presented to
304 \fBpcre_dfa_exec()\fP in several parts ("do" and "gsb" being the first two) the
305 match stops when "dog" has been found, and it is not possible to continue. On
306 the other hand, if "dogsbody" is presented as a single string,
307 \fBpcre_dfa_exec()\fP finds both matches.
308 .P
309 Because of these problems, it is probably best to use PCRE_PARTIAL_HARD when
310 matching multi-segment data. The example above then behaves differently:
311 .sp
312 re> /dog(sbody)?/
313 data> dogsb\eP\eP
314 Partial match: dogsb
315 data> do\eP\eD
316 Partial match: do
317 data> gsb\eR\eP\eP\eD
318 Partial match: gsb
319 .sp
320 .P
321 4. Patterns that contain alternatives at the top level which do not all
322 start with the same pattern item may not work as expected when
323 PCRE_DFA_RESTART is used with \fBpcre_dfa_exec()\fP. For example, consider this
324 pattern:
325 .sp
326 1234|3789
327 .sp
328 If the first part of the subject is "ABC123", a partial match of the first
329 alternative is found at offset 3. There is no partial match for the second
330 alternative, because such a match does not start at the same point in the
331 subject string. Attempting to continue with the string "7890" does not yield a
332 match because only those alternatives that match at one point in the subject
333 are remembered. The problem arises because the start of the second alternative
334 matches within the first alternative. There is no problem with anchored
335 patterns or patterns such as:
336 .sp
337 1234|ABCD
338 .sp
339 where no string can be a partial match for both alternatives. This is not a
340 problem if \fBpcre_exec()\fP is used, because the entire match has to be rerun
341 each time:
342 .sp
343 re> /1234|3789/
344 data> ABC123\eP
345 Partial match: 123
346 data> 1237890
347 0: 3789
348 .sp
349 Of course, instead of using PCRE_DFA_PARTIAL, the same technique of re-running
350 the entire match can also be used with \fBpcre_dfa_exec()\fP. Another
351 possibility is to work with two buffers. If a partial match at offset \fIn\fP
352 in the first buffer is followed by "no match" when PCRE_DFA_RESTART is used on
353 the second buffer, you can then try a new match starting at offset \fIn+1\fP in
354 the first buffer.
355 .
356 .
358 .rs
359 .sp
360 .nf
361 Philip Hazel
362 University Computing Service
363 Cambridge CB2 3QH, England.
364 .fi
365 .
366 .
368 .rs
369 .sp
370 .nf
371 Last updated: 19 October 2009
372 Copyright (c) 1997-2009 University of Cambridge.
373 .fi


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

  ViewVC Help
Powered by ViewVC 1.1.5