ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 572 - (show annotations)
Wed Nov 17 17:55:57 2010 UTC (9 years, 2 months ago) by ph10
File MIME type: text/html
File size: 20210 byte(s)
Documentation updates and tidies.
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>
17 <li><a name="TOC2" href="#SEC2">PARTIAL MATCHING USING pcre_exec()</a>
18 <li><a name="TOC3" href="#SEC3">PARTIAL MATCHING USING pcre_dfa_exec()</a>
20 <li><a name="TOC5" href="#SEC5">FORMERLY RESTRICTED PATTERNS</a>
22 <li><a name="TOC7" href="#SEC7">MULTI-SEGMENT MATCHING WITH pcre_dfa_exec()</a>
23 <li><a name="TOC8" href="#SEC8">MULTI-SEGMENT MATCHING WITH pcre_exec()</a>
24 <li><a name="TOC9" href="#SEC9">ISSUES WITH MULTI-SEGMENT MATCHING</a>
25 <li><a name="TOC10" href="#SEC10">AUTHOR</a>
26 <li><a name="TOC11" href="#SEC11">REVISION</a>
27 </ul>
28 <br><a name="SEC1" href="#TOC1">PARTIAL MATCHING IN PCRE</a><br>
29 <P>
30 In normal use of PCRE, if the subject string that is passed to
31 <b>pcre_exec()</b> or <b>pcre_dfa_exec()</b> matches as far as it goes, but is
32 too short to match the entire pattern, PCRE_ERROR_NOMATCH is returned. There
33 are circumstances where it might be helpful to distinguish this case from other
34 cases in which there is no match.
35 </P>
36 <P>
37 Consider, for example, an application where a human is required to type in data
38 for a field with specific formatting requirements. An example might be a date
39 in the form <i>ddmmmyy</i>, defined by this pattern:
40 <pre>
41 ^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
42 </pre>
43 If the application sees the user's keystrokes one by one, and can check that
44 what has been typed so far is potentially valid, it is able to raise an error
45 as soon as a mistake is made, by beeping and not reflecting the character that
46 has been typed, for example. This immediate feedback is likely to be a better
47 user interface than a check that is delayed until the entire string has been
48 entered. Partial matching can also be useful when the subject string is very
49 long and is not all available at once.
50 </P>
51 <P>
52 PCRE supports partial matching by means of the PCRE_PARTIAL_SOFT and
53 PCRE_PARTIAL_HARD options, which can be set when calling <b>pcre_exec()</b> or
54 <b>pcre_dfa_exec()</b>. For backwards compatibility, PCRE_PARTIAL is a synonym
55 for PCRE_PARTIAL_SOFT. The essential difference between the two options is
56 whether or not a partial match is preferred to an alternative complete match,
57 though the details differ between the two matching functions. If both options
58 are set, PCRE_PARTIAL_HARD takes precedence.
59 </P>
60 <P>
61 Setting a partial matching option disables two of PCRE's optimizations. PCRE
62 remembers the last literal byte in a pattern, and abandons matching immediately
63 if such a byte is not present in the subject string. This optimization cannot
64 be used for a subject string that might match only partially. If the pattern
65 was studied, PCRE knows the minimum length of a matching string, and does not
66 bother to run the matching function on shorter strings. This optimization is
67 also disabled for partial matching.
68 </P>
69 <br><a name="SEC2" href="#TOC1">PARTIAL MATCHING USING pcre_exec()</a><br>
70 <P>
71 A partial match occurs during a call to <b>pcre_exec()</b> when the end of the
72 subject string is reached successfully, but matching cannot continue because
73 more characters are needed. However, at least one character in the subject must
74 have been inspected. This character need not form part of the final matched
75 string; lookbehind assertions and the \K escape sequence provide ways of
76 inspecting characters before the start of a matched substring. The requirement
77 for inspecting at least one character exists because an empty string can always
78 be matched; without such a restriction there would always be a partial match of
79 an empty string at the end of the subject.
80 </P>
81 <P>
82 If there are at least two slots in the offsets vector when <b>pcre_exec()</b>
83 returns with a partial match, the first slot is set to the offset of the
84 earliest character that was inspected when the partial match was found. For
85 convenience, the second offset points to the end of the subject so that a
86 substring can easily be identified.
87 </P>
88 <P>
89 For the majority of patterns, the first offset identifies the start of the
90 partially matched string. However, for patterns that contain lookbehind
91 assertions, or \K, or begin with \b or \B, earlier characters have been
92 inspected while carrying out the match. For example:
93 <pre>
94 /(?&#60;=abc)123/
95 </pre>
96 This pattern matches "123", but only if it is preceded by "abc". If the subject
97 string is "xyzabc12", the offsets after a partial match are for the substring
98 "abc12", because all these characters are needed if another match is tried
99 with extra characters added to the subject.
100 </P>
101 <P>
102 What happens when a partial match is identified depends on which of the two
103 partial matching options are set.
104 </P>
105 <br><b>
106 PCRE_PARTIAL_SOFT with pcre_exec()
107 </b><br>
108 <P>
109 If PCRE_PARTIAL_SOFT is set when <b>pcre_exec()</b> identifies a partial match,
110 the partial match is remembered, but matching continues as normal, and other
111 alternatives in the pattern are tried. If no complete match can be found,
112 <b>pcre_exec()</b> returns PCRE_ERROR_PARTIAL instead of PCRE_ERROR_NOMATCH.
113 </P>
114 <P>
115 This option is "soft" because it prefers a complete match over a partial match.
116 All the various matching items in a pattern behave as if the subject string is
117 potentially complete. For example, \z, \Z, and $ match at the end of the
118 subject, as normal, and for \b and \B the end of the subject is treated as a
119 non-alphanumeric.
120 </P>
121 <P>
122 If there is more than one partial match, the first one that was found provides
123 the data that is returned. Consider this pattern:
124 <pre>
125 /123\w+X|dogY/
126 </pre>
127 If this is matched against the subject string "abc123dog", both
128 alternatives fail to match, but the end of the subject is reached during
129 matching, so PCRE_ERROR_PARTIAL is returned. The offsets are set to 3 and 9,
130 identifying "123dog" as the first partial match that was found. (In this
131 example, there are two partial matches, because "dog" on its own partially
132 matches the second alternative.)
133 </P>
134 <br><b>
135 PCRE_PARTIAL_HARD with pcre_exec()
136 </b><br>
137 <P>
138 If PCRE_PARTIAL_HARD is set for <b>pcre_exec()</b>, it returns
139 PCRE_ERROR_PARTIAL as soon as a partial match is found, without continuing to
140 search for possible complete matches. This option is "hard" because it prefers
141 an earlier partial match over a later complete match. For this reason, the
142 assumption is made that the end of the supplied subject string may not be the
143 true end of the available data, and so, if \z, \Z, \b, \B, or $ are
144 encountered at the end of the subject, the result is PCRE_ERROR_PARTIAL.
145 </P>
146 <P>
147 Setting PCRE_PARTIAL_HARD also affects the way <b>pcre_exec()</b> checks UTF-8
148 subject strings for validity. Normally, an invalid UTF-8 sequence causes the
149 error PCRE_ERROR_BADUTF8. However, in the special case of a truncated UTF-8
150 character at the end of the subject, PCRE_ERROR_SHORTUTF8 is returned when
152 </P>
153 <br><b>
154 Comparing hard and soft partial matching
155 </b><br>
156 <P>
157 The difference between the two partial matching options can be illustrated by a
158 pattern such as:
159 <pre>
160 /dog(sbody)?/
161 </pre>
162 This matches either "dog" or "dogsbody", greedily (that is, it prefers the
163 longer string if possible). If it is matched against the string "dog" with
164 PCRE_PARTIAL_SOFT, it yields a complete match for "dog". However, if
165 PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL. On the other hand,
166 if the pattern is made ungreedy the result is different:
167 <pre>
168 /dog(sbody)??/
169 </pre>
170 In this case the result is always a complete match because <b>pcre_exec()</b>
171 finds that first, and it never continues after finding a match. It might be
172 easier to follow this explanation by thinking of the two patterns like this:
173 <pre>
174 /dog(sbody)?/ is the same as /dogsbody|dog/
175 /dog(sbody)??/ is the same as /dog|dogsbody/
176 </pre>
177 The second pattern will never match "dogsbody" when <b>pcre_exec()</b> is
178 used, because it will always find the shorter match first.
179 </P>
180 <br><a name="SEC3" href="#TOC1">PARTIAL MATCHING USING pcre_dfa_exec()</a><br>
181 <P>
182 The <b>pcre_dfa_exec()</b> function moves along the subject string character by
183 character, without backtracking, searching for all possible matches
184 simultaneously. If the end of the subject is reached before the end of the
185 pattern, there is the possibility of a partial match, again provided that at
186 least one character has been inspected.
187 </P>
188 <P>
189 When PCRE_PARTIAL_SOFT is set, PCRE_ERROR_PARTIAL is returned only if there
190 have been no complete matches. Otherwise, the complete matches are returned.
191 However, if PCRE_PARTIAL_HARD is set, a partial match takes precedence over any
192 complete matches. The portion of the string that was inspected when the longest
193 partial match was found is set as the first matching string, provided there are
194 at least two slots in the offsets vector.
195 </P>
196 <P>
197 Because <b>pcre_dfa_exec()</b> always searches for all possible matches, and
198 there is no difference between greedy and ungreedy repetition, its behaviour is
199 different from <b>pcre_exec</b> when PCRE_PARTIAL_HARD is set. Consider the
200 string "dog" matched against the ungreedy pattern shown above:
201 <pre>
202 /dog(sbody)??/
203 </pre>
204 Whereas <b>pcre_exec()</b> stops as soon as it finds the complete match for
205 "dog", <b>pcre_dfa_exec()</b> also finds the partial match for "dogsbody", and
206 so returns that when PCRE_PARTIAL_HARD is set.
207 </P>
208 <br><a name="SEC4" href="#TOC1">PARTIAL MATCHING AND WORD BOUNDARIES</a><br>
209 <P>
210 If a pattern ends with one of sequences \b or \B, which test for word
211 boundaries, partial matching with PCRE_PARTIAL_SOFT can give counter-intuitive
212 results. Consider this pattern:
213 <pre>
214 /\bcat\b/
215 </pre>
216 This matches "cat", provided there is a word boundary at either end. If the
217 subject string is "the cat", the comparison of the final "t" with a following
218 character cannot take place, so a partial match is found. However,
219 <b>pcre_exec()</b> carries on with normal matching, which matches \b at the end
220 of the subject when the last character is a letter, thus finding a complete
221 match. The result, therefore, is <i>not</i> PCRE_ERROR_PARTIAL. The same thing
222 happens with <b>pcre_dfa_exec()</b>, because it also finds the complete match.
223 </P>
224 <P>
225 Using PCRE_PARTIAL_HARD in this case does yield PCRE_ERROR_PARTIAL, because
226 then the partial match takes precedence.
227 </P>
228 <br><a name="SEC5" href="#TOC1">FORMERLY RESTRICTED PATTERNS</a><br>
229 <P>
230 For releases of PCRE prior to 8.00, because of the way certain internal
231 optimizations were implemented in the <b>pcre_exec()</b> function, the
232 PCRE_PARTIAL option (predecessor of PCRE_PARTIAL_SOFT) could not be used with
233 all patterns. From release 8.00 onwards, the restrictions no longer apply, and
234 partial matching with <b>pcre_exec()</b> can be requested for any pattern.
235 </P>
236 <P>
237 Items that were formerly restricted were repeated single characters and
238 repeated metasequences. If PCRE_PARTIAL was set for a pattern that did not
239 conform to the restrictions, <b>pcre_exec()</b> returned the error code
240 PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in use. The
241 PCRE_INFO_OKPARTIAL call to <b>pcre_fullinfo()</b> to find out if a compiled
242 pattern can be used for partial matching now always returns 1.
243 </P>
244 <br><a name="SEC6" href="#TOC1">EXAMPLE OF PARTIAL MATCHING USING PCRETEST</a><br>
245 <P>
246 If the escape sequence \P is present in a <b>pcretest</b> data line, the
247 PCRE_PARTIAL_SOFT option is used for the match. Here is a run of <b>pcretest</b>
248 that uses the date example quoted above:
249 <pre>
250 re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
251 data&#62; 25jun04\P
252 0: 25jun04
253 1: jun
254 data&#62; 25dec3\P
255 Partial match: 23dec3
256 data&#62; 3ju\P
257 Partial match: 3ju
258 data&#62; 3juj\P
259 No match
260 data&#62; j\P
261 No match
262 </pre>
263 The first data string is matched completely, so <b>pcretest</b> shows the
264 matched substrings. The remaining four strings do not match the complete
265 pattern, but the first two are partial matches. Similar output is obtained
266 when <b>pcre_dfa_exec()</b> is used.
267 </P>
268 <P>
269 If the escape sequence \P is present more than once in a <b>pcretest</b> data
270 line, the PCRE_PARTIAL_HARD option is set for the match.
271 </P>
272 <br><a name="SEC7" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre_dfa_exec()</a><br>
273 <P>
274 When a partial match has been found using <b>pcre_dfa_exec()</b>, it is possible
275 to continue the match by providing additional subject data and calling
276 <b>pcre_dfa_exec()</b> again with the same compiled regular expression, this
277 time setting the PCRE_DFA_RESTART option. You must pass the same working
278 space as before, because this is where details of the previous partial match
279 are stored. Here is an example using <b>pcretest</b>, using the \R escape
280 sequence to set the PCRE_DFA_RESTART option (\D specifies the use of
281 <b>pcre_dfa_exec()</b>):
282 <pre>
283 re&#62; /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
284 data&#62; 23ja\P\D
285 Partial match: 23ja
286 data&#62; n05\R\D
287 0: n05
288 </pre>
289 The first call has "23ja" as the subject, and requests partial matching; the
290 second call has "n05" as the subject for the continued (restarted) match.
291 Notice that when the match is complete, only the last part is shown; PCRE does
292 not retain the previously partially-matched string. It is up to the calling
293 program to do that if it needs to.
294 </P>
295 <P>
296 You can set the PCRE_PARTIAL_SOFT or PCRE_PARTIAL_HARD options with
297 PCRE_DFA_RESTART to continue partial matching over multiple segments. This
298 facility can be used to pass very long subject strings to
299 <b>pcre_dfa_exec()</b>.
300 </P>
301 <br><a name="SEC8" href="#TOC1">MULTI-SEGMENT MATCHING WITH pcre_exec()</a><br>
302 <P>
303 From release 8.00, <b>pcre_exec()</b> can also be used to do multi-segment
304 matching. Unlike <b>pcre_dfa_exec()</b>, it is not possible to restart the
305 previous match with a new segment of data. Instead, new data must be added to
306 the previous subject string, and the entire match re-run, starting from the
307 point where the partial match occurred. Earlier data can be discarded. It is
308 best to use PCRE_PARTIAL_HARD in this situation, because it does not treat the
309 end of a segment as the end of the subject when matching \z, \Z, \b, \B,
310 and $. Consider an unanchored pattern that matches dates:
311 <pre>
312 re&#62; /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
313 data&#62; The date is 23ja\P\P
314 Partial match: 23ja
315 </pre>
316 At this stage, an application could discard the text preceding "23ja", add on
317 text from the next segment, and call <b>pcre_exec()</b> again. Unlike
318 <b>pcre_dfa_exec()</b>, the entire matching string must always be available, and
319 the complete matching process occurs for each call, so more memory and more
320 processing time is needed.
321 </P>
322 <P>
323 <b>Note:</b> If the pattern contains lookbehind assertions, or \K, or starts
324 with \b or \B, the string that is returned for a partial match will include
325 characters that precede the partially matched string itself, because these must
326 be retained when adding on more characters for a subsequent matching attempt.
327 </P>
328 <br><a name="SEC9" href="#TOC1">ISSUES WITH MULTI-SEGMENT MATCHING</a><br>
329 <P>
330 Certain types of pattern may give problems with multi-segment matching,
331 whichever matching function is used.
332 </P>
333 <P>
334 1. If the pattern contains a test for the beginning of a line, you need to pass
335 the PCRE_NOTBOL option when the subject string for any call does start at the
336 beginning of a line. There is also a PCRE_NOTEOL option, but in practice when
337 doing multi-segment matching you should be using PCRE_PARTIAL_HARD, which
338 includes the effect of PCRE_NOTEOL.
339 </P>
340 <P>
341 2. Lookbehind assertions at the start of a pattern are catered for in the
342 offsets that are returned for a partial match. However, in theory, a lookbehind
343 assertion later in the pattern could require even earlier characters to be
344 inspected, and it might not have been reached when a partial match occurs. This
345 is probably an extremely unlikely case; you could guard against it to a certain
346 extent by always including extra characters at the start.
347 </P>
348 <P>
349 3. Matching a subject string that is split into multiple segments may not
350 always produce exactly the same result as matching over one single long string,
351 especially when PCRE_PARTIAL_SOFT is used. The section "Partial Matching and
352 Word Boundaries" above describes an issue that arises if the pattern ends with
353 \b or \B. Another kind of difference may occur when there are multiple
354 matching possibilities, because (for PCRE_PARTIAL_SOFT) a partial match result
355 is given only when there are no completed matches. This means that as soon as
356 the shortest match has been found, continuation to a new subject segment is no
357 longer possible. Consider again this <b>pcretest</b> example:
358 <pre>
359 re&#62; /dog(sbody)?/
360 data&#62; dogsb\P
361 0: dog
362 data&#62; do\P\D
363 Partial match: do
364 data&#62; gsb\R\P\D
365 0: g
366 data&#62; dogsbody\D
367 0: dogsbody
368 1: dog
369 </pre>
370 The first data line passes the string "dogsb" to <b>pcre_exec()</b>, setting the
371 PCRE_PARTIAL_SOFT option. Although the string is a partial match for
372 "dogsbody", the result is not PCRE_ERROR_PARTIAL, because the shorter string
373 "dog" is a complete match. Similarly, when the subject is presented to
374 <b>pcre_dfa_exec()</b> in several parts ("do" and "gsb" being the first two) the
375 match stops when "dog" has been found, and it is not possible to continue. On
376 the other hand, if "dogsbody" is presented as a single string,
377 <b>pcre_dfa_exec()</b> finds both matches.
378 </P>
379 <P>
380 Because of these problems, it is best to use PCRE_PARTIAL_HARD when matching
381 multi-segment data. The example above then behaves differently:
382 <pre>
383 re&#62; /dog(sbody)?/
384 data&#62; dogsb\P\P
385 Partial match: dogsb
386 data&#62; do\P\D
387 Partial match: do
388 data&#62; gsb\R\P\P\D
389 Partial match: gsb
390 </pre>
391 4. Patterns that contain alternatives at the top level which do not all
392 start with the same pattern item may not work as expected when
393 PCRE_DFA_RESTART is used with <b>pcre_dfa_exec()</b>. For example, consider this
394 pattern:
395 <pre>
396 1234|3789
397 </pre>
398 If the first part of the subject is "ABC123", a partial match of the first
399 alternative is found at offset 3. There is no partial match for the second
400 alternative, because such a match does not start at the same point in the
401 subject string. Attempting to continue with the string "7890" does not yield a
402 match because only those alternatives that match at one point in the subject
403 are remembered. The problem arises because the start of the second alternative
404 matches within the first alternative. There is no problem with anchored
405 patterns or patterns such as:
406 <pre>
407 1234|ABCD
408 </pre>
409 where no string can be a partial match for both alternatives. This is not a
410 problem if <b>pcre_exec()</b> is used, because the entire match has to be rerun
411 each time:
412 <pre>
413 re&#62; /1234|3789/
414 data&#62; ABC123\P\P
415 Partial match: 123
416 data&#62; 1237890
417 0: 3789
418 </pre>
419 Of course, instead of using PCRE_DFA_RESTART, the same technique of re-running
420 the entire match can also be used with <b>pcre_dfa_exec()</b>. Another
421 possibility is to work with two buffers. If a partial match at offset <i>n</i>
422 in the first buffer is followed by "no match" when PCRE_DFA_RESTART is used on
423 the second buffer, you can then try a new match starting at offset <i>n+1</i> in
424 the first buffer.
425 </P>
426 <br><a name="SEC10" href="#TOC1">AUTHOR</a><br>
427 <P>
428 Philip Hazel
429 <br>
430 University Computing Service
431 <br>
432 Cambridge CB2 3QH, England.
433 <br>
434 </P>
435 <br><a name="SEC11" href="#TOC1">REVISION</a><br>
436 <P>
437 Last updated: 07 November 2010
438 <br>
439 Copyright &copy; 1997-2010 University of Cambridge.
440 <br>
441 <p>
442 Return to the <a href="index.html">PCRE index page</a>.
443 </p>


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

  ViewVC Help
Powered by ViewVC 1.1.5