/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 89 - (show annotations) (download)
Sat Feb 24 21:41:27 2007 UTC (7 years, 5 months ago) by nigel
File size: 95469 byte(s)
Load pcre-6.6 into code/trunk.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 6.6 06-Feb-06
5 ---------------------
6
7 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
8 in pcreposix.h. I have copied the definition from pcre.h.
9
10 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
11 because pcre.h is no longer a built file.
12
13 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
14 not normally included in the compiled code.
15
16
17 Version 6.5 01-Feb-06
18 ---------------------
19
20 1. When using the partial match feature with pcre_dfa_exec(), it was not
21 anchoring the second and subsequent partial matches at the new starting
22 point. This could lead to incorrect results. For example, with the pattern
23 /1234/, partially matching against "123" and then "a4" gave a match.
24
25 2. Changes to pcregrep:
26
27 (a) All non-match returns from pcre_exec() were being treated as failures
28 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
29 error message is output. Some extra information is given for the
30 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
31 probably the only errors that are likely to be caused by users (by
32 specifying a regex that has nested indefinite repeats, for instance).
33 If there are more than 20 of these errors, pcregrep is abandoned.
34
35 (b) A binary zero was treated as data while matching, but terminated the
36 output line if it was written out. This has been fixed: binary zeroes
37 are now no different to any other data bytes.
38
39 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
40 used to set a locale for matching. The --locale=xxxx long option has
41 been added (no short equivalent) to specify a locale explicitly on the
42 pcregrep command, overriding the environment variables.
43
44 (d) When -B was used with -n, some line numbers in the output were one less
45 than they should have been.
46
47 (e) Added the -o (--only-matching) option.
48
49 (f) If -A or -C was used with -c (count only), some lines of context were
50 accidentally printed for the final match.
51
52 (g) Added the -H (--with-filename) option.
53
54 (h) The combination of options -rh failed to suppress file names for files
55 that were found from directory arguments.
56
57 (i) Added the -D (--devices) and -d (--directories) options.
58
59 (j) Added the -F (--fixed-strings) option.
60
61 (k) Allow "-" to be used as a file name for -f as well as for a data file.
62
63 (l) Added the --colo(u)r option.
64
65 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
66 is not present by default.
67
68 3. A nasty bug was discovered in the handling of recursive patterns, that is,
69 items such as (?R) or (?1), when the recursion could match a number of
70 alternatives. If it matched one of the alternatives, but subsequently,
71 outside the recursion, there was a failure, the code tried to back up into
72 the recursion. However, because of the way PCRE is implemented, this is not
73 possible, and the result was an incorrect result from the match.
74
75 In order to prevent this happening, the specification of recursion has
76 been changed so that all such subpatterns are automatically treated as
77 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
78
79 4. I had overlooked the fact that, in some locales, there are characters for
80 which isalpha() is true but neither isupper() nor islower() are true. In
81 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
82 and ordfeminine) are like this. This affected the treatment of \w and \W
83 when they appeared in character classes, but not when they appeared outside
84 a character class. The bit map for "word" characters is now created
85 separately from the results of isalnum() instead of just taking it from the
86 upper, lower, and digit maps. (Plus the underscore character, of course.)
87
88 5. The above bug also affected the handling of POSIX character classes such as
89 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
90 permanent tables. Instead, the bit maps for such a class were previously
91 created as the appropriate unions of the upper, lower, and digit bitmaps.
92 Now they are created by subtraction from the [[:word:]] class, which has
93 its own bitmap.
94
95 6. The [[:blank:]] character class matches horizontal, but not vertical space.
96 It is created by subtracting the vertical space characters (\x09, \x0a,
97 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
98 subtraction was done in the overall bitmap for a character class, meaning
99 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
100 be recognized. This bug has been fixed.
101
102 7. Patches from the folks at Google:
103
104 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
105 real life, but is still worth protecting against".
106
107 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
108 regular expressions".
109
110 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
111 have it.
112
113 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
114 "configure" and the latter not, in order to fix a problem somebody had
115 with compiling the Arg class on HP-UX.
116
117 (e) Improve the error-handling of the C++ wrapper a little bit.
118
119 (f) New tests for checking recursion limiting.
120
121 8. The pcre_memmove() function, which is used only if the environment does not
122 have a standard memmove() function (and is therefore rarely compiled),
123 contained two bugs: (a) use of int instead of size_t, and (b) it was not
124 returning a result (though PCRE never actually uses the result).
125
126 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
127 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
128 returned instead of calling malloc() with an overflowing number that would
129 most likely cause subsequent chaos.
130
131 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
132
133 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
134 with this option is matched, the nmatch and pmatch options of regexec() are
135 ignored.
136
137 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
138 provided in case anyone wants to the the POSIX interface with UTF-8
139 strings.
140
141 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
142 C++ linking (needed for some HP-UX environments).
143
144 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
145 (unused parameter) and in the pcre_printint() function (omitted "default"
146 switch label when the default is to do nothing).
147
148 15. Added some code to make it possible, when PCRE is compiled as a C++
149 library, to replace subject pointers for pcre_exec() with a smart pointer
150 class, thus making it possible to process discontinuous strings.
151
152 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
153 much the same function. They were added by different people who were trying
154 to make PCRE easy to compile on non-Unix systems. It has been suggested
155 that PCRE_EXPORT be abolished now that there is more automatic apparatus
156 for compiling on Windows systems. I have therefore replaced it with
157 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
158 defaults to "extern" for C or "extern C" for C++, which works fine on
159 Unix-like systems. It is now possible to override the value of PCRE_DATA_
160 SCOPE with something explicit in config.h. In addition:
161
162 (a) pcreposix.h still had just "extern" instead of either of these macros;
163 I have replaced it with PCRE_DATA_SCOPE.
164
165 (b) Functions such as _pcre_xclass(), which are internal to the library,
166 but external in the C sense, all had PCRE_EXPORT in their definitions.
167 This is apparently wrong for the Windows case, so I have removed it.
168 (It makes no difference on Unix-like systems.)
169
170 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
171 of recursive calls to match(). This is different to MATCH_LIMIT because
172 that limits the total number of calls to match(), not all of which increase
173 the depth of recursion. Limiting the recursion depth limits the amount of
174 stack (or heap if NO_RECURSE is set) that is used. The default can be set
175 when PCRE is compiled, and changed at run time. A patch from Google adds
176 this functionality to the C++ interface.
177
178 18. Changes to the handling of Unicode character properties:
179
180 (a) Updated the table to Unicode 4.1.0.
181
182 (b) Recognize characters that are not in the table as "Cn" (undefined).
183
184 (c) I revised the way the table is implemented to a much improved format
185 which includes recognition of ranges. It now supports the ranges that
186 are defined in UnicodeData.txt, and it also amalgamates other
187 characters into ranges. This has reduced the number of entries in the
188 table from around 16,000 to around 3,000, thus reducing its size
189 considerably. I realized I did not need to use a tree structure after
190 all - a binary chop search is just as efficient. Having reduced the
191 number of entries, I extended their size from 6 bytes to 8 bytes to
192 allow for more data.
193
194 (d) Added support for Unicode script names via properties such as \p{Han}.
195
196 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
197 matching that character.
198
199 20. When matching a repeated Unicode property with a minimum greater than zero,
200 (for example \pL{2,}), PCRE could look past the end of the subject if it
201 reached it while seeking the minimum number of characters. This could
202 happen only if some of the characters were more than one byte long, because
203 there is a check for at least the minimum number of bytes.
204
205 21. Refactored the implementation of \p and \P so as to be more general, to
206 allow for more different types of property in future. This has changed the
207 compiled form incompatibly. Anybody with saved compiled patterns that use
208 \p or \P will have to recompile them.
209
210 22. Added "Any" and "L&" to the supported property types.
211
212 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
213 but give a compile time error if the value is greater than 0xff.
214
215 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
216 accidentally not being installed or uninstalled.
217
218 25. The pcre.h file was built from pcre.h.in, but the only changes that were
219 made were to insert the current release number. This seemed silly, because
220 it made things harder for people building PCRE on systems that don't run
221 "configure". I have turned pcre.h into a distributed file, no longer built
222 by "configure", with the version identification directly included. There is
223 no longer a pcre.h.in file.
224
225 However, this change necessitated a change to the pcre-config script as
226 well. It is built from pcre-config.in, and one of the substitutions was the
227 release number. I have updated configure.ac so that ./configure now finds
228 the release number by grepping pcre.h.
229
230 26. Added the ability to run the tests under valgrind.
231
232
233 Version 6.4 05-Sep-05
234 ---------------------
235
236 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
237 "--" to be printed when multiple files were scanned, even when none of the
238 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
239 consider it to be a bug, and have restored the previous behaviour.
240
241 2. A couple of code tidies to get rid of compiler warnings.
242
243 3. The pcretest program used to cheat by referring to symbols in the library
244 whose names begin with _pcre_. These are internal symbols that are not
245 really supposed to be visible externally, and in some environments it is
246 possible to suppress them. The cheating is now confined to including
247 certain files from the library's source, which is a bit cleaner.
248
249 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
250 file's purpose clearer.
251
252 5. Reorganized pcre_ucp_findchar().
253
254
255 Version 6.3 15-Aug-05
256 ---------------------
257
258 1. The file libpcre.pc.in did not have general read permission in the tarball.
259
260 2. There were some problems when building without C++ support:
261
262 (a) If C++ support was not built, "make install" and "make test" still
263 tried to test it.
264
265 (b) There were problems when the value of CXX was explicitly set. Some
266 changes have been made to try to fix these, and ...
267
268 (c) --disable-cpp can now be used to explicitly disable C++ support.
269
270 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
271 backslash in a target when C++ was disabled. This confuses some
272 versions of "make", apparently. Using an intermediate variable solves
273 this. (Same for CPP_LOBJ.)
274
275 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
276 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
277 necessary on certain architectures.
278
279 4. Added a setting of -export-symbols-regex to the link command to remove
280 those symbols that are exported in the C sense, but actually are local
281 within the library, and not documented. Their names all begin with
282 "_pcre_". This is not a perfect job, because (a) we have to except some
283 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
284 available (and never for static libraries). I have made a note to try to
285 find a way round (a) in the future.
286
287
288 Version 6.2 01-Aug-05
289 ---------------------
290
291 1. There was no test for integer overflow of quantifier values. A construction
292 such as {1111111111111111} would give undefined results. What is worse, if
293 a minimum quantifier for a parenthesized subpattern overflowed and became
294 negative, the calculation of the memory size went wrong. This could have
295 led to memory overwriting.
296
297 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
298
299 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
300 operating environments where this matters.
301
302 4. Applied Giuseppe Maxia's patch to add additional features for controlling
303 PCRE options from within the C++ wrapper.
304
305 5. Named capturing subpatterns were not being correctly counted when a pattern
306 was compiled. This caused two problems: (a) If there were more than 100
307 such subpatterns, the calculation of the memory needed for the whole
308 compiled pattern went wrong, leading to an overflow error. (b) Numerical
309 back references of the form \12, where the number was greater than 9, were
310 not recognized as back references, even though there were sufficient
311 previous subpatterns.
312
313 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
314 versions of gcc, e.g. 2.95.4.
315
316
317 Version 6.1 21-Jun-05
318 ---------------------
319
320 1. There was one reference to the variable "posix" in pcretest.c that was not
321 surrounded by "#if !defined NOPOSIX".
322
323 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
324 the cross-check on the old pcre_info() function, for the benefit of the
325 cut-down version of PCRE that is currently imported into Exim.
326
327 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
328 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
329 patterns that start (?i) but I don't think that matters. The use of (?i) is
330 just an example; this all applies to the other options as well.
331
332 4. Since libtool seems to echo the compile commands it is issuing, the output
333 from "make" can be reduced a bit by putting "@" in front of each libtool
334 compile command.
335
336 5. Patch from the folks at Google for configure.in to be a bit more thorough
337 in checking for a suitable C++ installation before trying to compile the
338 C++ stuff. This should fix a reported problem when a compiler was present,
339 but no suitable headers.
340
341 6. The man pages all had just "PCRE" as their title. I have changed them to
342 be the relevant file name. I have also arranged that these names are
343 retained in the file doc/pcre.txt, which is a concatenation in text format
344 of all the man pages except the little individual ones for each function.
345
346 7. The NON-UNIX-USE file had not been updated for the different set of source
347 files that come with release 6. I also added a few comments about the C++
348 wrapper.
349
350
351 Version 6.0 07-Jun-05
352 ---------------------
353
354 1. Some minor internal re-organization to help with my DFA experiments.
355
356 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
357 didn't matter for the library itself when fully configured, but did matter
358 when compiling without UCP support, or within Exim, where the ucp files are
359 not imported.
360
361 3. Refactoring of the library code to split up the various functions into
362 different source modules. The addition of the new DFA matching code (see
363 below) to a single monolithic source would have made it really too
364 unwieldy, quite apart from causing all the code to be include in a
365 statically linked application, when only some functions are used. This is
366 relevant even without the DFA addition now that patterns can be compiled in
367 one application and matched in another.
368
369 The downside of splitting up is that there have to be some external
370 functions and data tables that are used internally in different modules of
371 the library but which are not part of the API. These have all had their
372 names changed to start with "_pcre_" so that they are unlikely to clash
373 with other external names.
374
375 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
376 a different (DFA) algorithm. Although it is slower than the original
377 function, it does have some advantages for certain types of matching
378 problem.
379
380 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
381 including restarting after a partial match.
382
383 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
384 defined when compiling for Windows was sent to me. I have put it into the
385 code, though I have no means of testing or verifying it.
386
387 7. Added the pcre_refcount() auxiliary function.
388
389 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
390 match before or at the first newline in the subject string. In pcretest,
391 the /f option on a pattern can be used to set this.
392
393 9. A repeated \w when used in UTF-8 mode with characters greater than 256
394 would behave wrongly. This has been present in PCRE since release 4.0.
395
396 10. A number of changes to the pcregrep command:
397
398 (a) Refactored how -x works; insert ^(...)$ instead of setting
399 PCRE_ANCHORED and checking the length, in preparation for adding
400 something similar for -w.
401
402 (b) Added the -w (match as a word) option.
403
404 (c) Refactored the way lines are read and buffered so as to have more
405 than one at a time available.
406
407 (d) Implemented a pcregrep test script.
408
409 (e) Added the -M (multiline match) option. This allows patterns to match
410 over several lines of the subject. The buffering ensures that at least
411 8K, or the rest of the document (whichever is the shorter) is available
412 for matching (and similarly the previous 8K for lookbehind assertions).
413
414 (f) Changed the --help output so that it now says
415
416 -w, --word-regex(p)
417
418 instead of two lines, one with "regex" and the other with "regexp"
419 because that confused at least one person since the short forms are the
420 same. (This required a bit of code, as the output is generated
421 automatically from a table. It wasn't just a text change.)
422
423 (g) -- can be used to terminate pcregrep options if the next thing isn't an
424 option but starts with a hyphen. Could be a pattern or a path name
425 starting with a hyphen, for instance.
426
427 (h) "-" can be given as a file name to represent stdin.
428
429 (i) When file names are being printed, "(standard input)" is used for
430 the standard input, for compatibility with GNU grep. Previously
431 "<stdin>" was used.
432
433 (j) The option --label=xxx can be used to supply a name to be used for
434 stdin when file names are being printed. There is no short form.
435
436 (k) Re-factored the options decoding logic because we are going to add
437 two more options that take data. Such options can now be given in four
438 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
439
440 (l) Added the -A, -B, and -C options for requesting that lines of context
441 around matches be printed.
442
443 (m) Added the -L option to print the names of files that do not contain
444 any matching lines, that is, the complement of -l.
445
446 (n) The return code is 2 if any file cannot be opened, but pcregrep does
447 continue to scan other files.
448
449 (o) The -s option was incorrectly implemented. For compatibility with other
450 greps, it now suppresses the error message for a non-existent or non-
451 accessible file (but not the return code). There is a new option called
452 -q that suppresses the output of matching lines, which was what -s was
453 previously doing.
454
455 (p) Added --include and --exclude options to specify files for inclusion
456 and exclusion when recursing.
457
458 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
459 Hopefully, it now does.
460
461 12. Missing cast in pcre_study().
462
463 13. Added an "uninstall" target to the makefile.
464
465 14. Replaced "extern" in the function prototypes in Makefile.in with
466 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
467 world, but is set differently for Windows.
468
469 15. Added a second compiling function called pcre_compile2(). The only
470 difference is that it has an extra argument, which is a pointer to an
471 integer error code. When there is a compile-time failure, this is set
472 non-zero, in addition to the error test pointer being set to point to an
473 error message. The new argument may be NULL if no error number is required
474 (but then you may as well call pcre_compile(), which is now just a
475 wrapper). This facility is provided because some applications need a
476 numeric error indication, but it has also enabled me to tidy up the way
477 compile-time errors are handled in the POSIX wrapper.
478
479 16. Added VPATH=.libs to the makefile; this should help when building with one
480 prefix path and installing with another. (Or so I'm told by someone who
481 knows more about this stuff than I do.)
482
483 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
484 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
485 match everything, including newlines. This is not POSIX-compatible, but
486 somebody wanted the feature. From pcretest it can be activated by using
487 both the P and the s flags.
488
489 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
490
491 19. libpcre.pc was being incorrectly installed as executable.
492
493 20. A couple of places in pcretest check for end-of-line by looking for '\n';
494 it now also looks for '\r' so that it will work unmodified on Windows.
495
496 21. Added Google's contributed C++ wrapper to the distribution.
497
498 22. Added some untidy missing memory free() calls in pcretest, to keep
499 Electric Fence happy when testing.
500
501
502
503 Version 5.0 13-Sep-04
504 ---------------------
505
506 1. Internal change: literal characters are no longer packed up into items
507 containing multiple characters in a single byte-string. Each character
508 is now matched using a separate opcode. However, there may be more than one
509 byte in the character in UTF-8 mode.
510
511 2. The pcre_callout_block structure has two new fields: pattern_position and
512 next_item_length. These contain the offset in the pattern to the next match
513 item, and its length, respectively.
514
515 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
516 insertion of callouts before each pattern item. Added the /C option to
517 pcretest to make use of this.
518
519 4. On the advice of a Windows user, the lines
520
521 #if defined(_WIN32) || defined(WIN32)
522 _setmode( _fileno( stdout ), 0x8000 );
523 #endif /* defined(_WIN32) || defined(WIN32) */
524
525 have been added to the source of pcretest. This apparently does useful
526 magic in relation to line terminators.
527
528 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
529 for the benefit of those environments where the "b" makes a difference.
530
531 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
532 to know about it. I have put a hack into configure.in that adds in code
533 to set GCC=yes if CC=icc. This seems to end up at a point in the
534 generated configure script that is early enough to affect the setting of
535 compiler options, which is what is needed, but I have no means of testing
536 whether it really works. (The user who reported this had patched the
537 generated configure script, which of course I cannot do.)
538
539 LATER: After change 22 below (new libtool files), the configure script
540 seems to know about icc (and also ecc). Therefore, I have commented out
541 this hack in configure.in.
542
543 7. Added support for pkg-config (2 patches were sent in).
544
545 8. Negated POSIX character classes that used a combination of internal tables
546 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
547 [[:^ascii]]. Typically, they would match almost any characters. The other
548 POSIX classes were not broken in this way.
549
550 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
551 to find the match, as PCRE was deluded into thinking that the match had to
552 start at the start point or following a newline. The same bug applied to
553 patterns with negative forward assertions or any backward assertions
554 preceding ".*" at the start, unless the pattern required a fixed first
555 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
556
557 10. In UTF-8 mode, when moving forwards in the subject after a failed match
558 starting at the last subject character, bytes beyond the end of the subject
559 string were read.
560
561 11. Renamed the variable "class" as "classbits" to make life easier for C++
562 users. (Previously there was a macro definition, but it apparently wasn't
563 enough.)
564
565 12. Added the new field "tables" to the extra data so that tables can be passed
566 in at exec time, or the internal tables can be re-selected. This allows
567 a compiled regex to be saved and re-used at a later time by a different
568 program that might have everything at different addresses.
569
570 13. Modified the pcre-config script so that, when run on Solaris, it shows a
571 -R library as well as a -L library.
572
573 14. The debugging options of pcretest (-d on the command line or D on a
574 pattern) showed incorrect output for anything following an extended class
575 that contained multibyte characters and which was followed by a quantifier.
576
577 15. Added optional support for general category Unicode character properties
578 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
579 support. It adds about 90K to the size of the library. The meanings of the
580 inbuilt class escapes such as \d and \s have NOT been changed.
581
582 16. Updated pcredemo.c to include calls to free() to release the memory for the
583 compiled pattern.
584
585 17. The generated file chartables.c was being created in the source directory
586 instead of in the building directory. This caused the build to fail if the
587 source directory was different from the building directory, and was
588 read-only.
589
590 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
591 file. No doubt somebody will tell me if they don't make sense... Also added
592 Dan Mooney's comments about building on OpenVMS.
593
594 19. Added support for partial matching via the PCRE_PARTIAL option for
595 pcre_exec() and the \P data escape in pcretest.
596
597 20. Extended pcretest with 3 new pattern features:
598
599 (i) A pattern option of the form ">rest-of-line" causes pcretest to
600 write the compiled pattern to the file whose name is "rest-of-line".
601 This is a straight binary dump of the data, with the saved pointer to
602 the character tables forced to be NULL. The study data, if any, is
603 written too. After writing, pcretest reads a new pattern.
604
605 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
606 compiled pattern from the given file. There must not be any
607 occurrences of "<" in the file name (pretty unlikely); if there are,
608 pcretest will instead treat the initial "<" as a pattern delimiter.
609 After reading in the pattern, pcretest goes on to read data lines as
610 usual.
611
612 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
613 and 16-bit fields in a compiled pattern, to simulate a pattern that
614 was compiled on a host of opposite endianness.
615
616 21. The pcre-exec() function can now cope with patterns that were compiled on
617 hosts of opposite endianness, with this restriction:
618
619 As for any compiled expression that is saved and used later, the tables
620 pointer field cannot be preserved; the extra_data field in the arguments
621 to pcre_exec() should be used to pass in a tables address if a value
622 other than the default internal tables were used at compile time.
623
624 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
625 now diagnosed as an error. Previously, most of the time, a negative number
626 would have been treated as zero, but if in addition "ovector" was passed as
627 NULL, a crash could occur.
628
629 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
630 new versions from the libtool 1.5 distribution (the last one is a copy of
631 a file called libtool.m4). This seems to have fixed the need to patch
632 "configure" to support Darwin 1.3 (which I used to do). However, I still
633 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
634 workstation).
635
636 24. Changed the PCRE licence to be the more standard "BSD" licence.
637
638
639 Version 4.5 01-Dec-03
640 ---------------------
641
642 1. There has been some re-arrangement of the code for the match() function so
643 that it can be compiled in a version that does not call itself recursively.
644 Instead, it keeps those local variables that need separate instances for
645 each "recursion" in a frame on the heap, and gets/frees frames whenever it
646 needs to "recurse". Keeping track of where control must go is done by means
647 of setjmp/longjmp. The whole thing is implemented by a set of macros that
648 hide most of the details from the main code, and operates only if
649 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
650 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
651 operating.
652
653 To make it easier for callers to provide specially tailored get/free
654 functions for this usage, two new functions, pcre_stack_malloc, and
655 pcre_stack_free, are used. They are always called in strict stacking order,
656 and the size of block requested is always the same.
657
658 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
659 PCRE has been compiled to use the stack or the heap for recursion. The
660 -C option of pcretest uses this to show which version is compiled.
661
662 A new data escape \S, is added to pcretest; it causes the amounts of store
663 obtained and freed by both kinds of malloc/free at match time to be added
664 to the output.
665
666 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
667 what's available on my current Linux desktop machine.
668
669 3. When matching a UTF-8 string, the test for a valid string at the start has
670 been extended. If start_offset is not zero, PCRE now checks that it points
671 to a byte that is the start of a UTF-8 character. If not, it returns
672 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
673 this is necessary because there may be backward assertions in the pattern.
674 When matching the same subject several times, it may save resources to use
675 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
676
677 4. The code for checking the validity of UTF-8 strings has been tightened so
678 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
679 containing "overlong sequences".
680
681 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
682 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
683 should have been "&", but it just so happened that all the cases this let
684 through by mistake were picked up later in the function.
685
686 6. I had used a variable called "isblank" - this is a C99 function, causing
687 some compilers to warn. To avoid this, I renamed it (as "blankclass").
688
689 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
690 prompting; (b) run "./pcretest /dev/null" at the start of the test script
691 so the version is shown; (c) stop "make test" echoing "./RunTest".
692
693 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
694
695 9. The prototype for memmove() for systems that don't have it was using
696 size_t, but the inclusion of the header that defines size_t was later. I've
697 moved the #includes for the C headers earlier to avoid this.
698
699 10. Added some adjustments to the code to make it easier to compiler on certain
700 special systems:
701
702 (a) Some "const" qualifiers were missing.
703 (b) Added the macro EXPORT before all exported functions; by default this
704 is defined to be empty.
705 (c) Changed the dftables auxiliary program (that builds chartables.c) so
706 that it reads its output file name as an argument instead of writing
707 to the standard output and assuming this can be redirected.
708
709 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
710 class containing characters with values greater than 255, PCRE compilation
711 went into a loop.
712
713 12. A recursive reference to a subpattern that was within another subpattern
714 that had a minimum quantifier of zero caused PCRE to crash. For example,
715 (x(y(?2))z)? provoked this bug with a subject that got as far as the
716 recursion. If the recursively-called subpattern itself had a zero repeat,
717 that was OK.
718
719 13. In pcretest, the buffer for reading a data line was set at 30K, but the
720 buffer into which it was copied (for escape processing) was still set at
721 1024, so long lines caused crashes.
722
723 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
724 "internal error: code overflow...". This applied to any character class
725 that was followed by a possessive quantifier.
726
727 15. Modified the Makefile to add libpcre.la as a prerequisite for
728 libpcreposix.la because I was told this is needed for a parallel build to
729 work.
730
731 16. If a pattern that contained .* following optional items at the start was
732 studied, the wrong optimizing data was generated, leading to matching
733 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
734 matching string must start with a or b or c. The correct conclusion for
735 this pattern is that a match can start with any character.
736
737
738 Version 4.4 13-Aug-03
739 ---------------------
740
741 1. In UTF-8 mode, a character class containing characters with values between
742 127 and 255 was not handled correctly if the compiled pattern was studied.
743 In fixing this, I have also improved the studying algorithm for such
744 classes (slightly).
745
746 2. Three internal functions had redundant arguments passed to them. Removal
747 might give a very teeny performance improvement.
748
749 3. Documentation bug: the value of the capture_top field in a callout is *one
750 more than* the number of the hightest numbered captured substring.
751
752 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
753 in incorrectly linking with a previously installed version. They now link
754 explicitly with libpcre.la.
755
756 5. configure.in no longer needs to recognize Cygwin specially.
757
758 6. A problem in pcre.in for Windows platforms is fixed.
759
760 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
761 pcretest, it used to include the size of the study block as part of its
762 output. Unfortunately, the structure contains a field that has a different
763 size on different hardware architectures. This meant that the tests that
764 showed this size failed. As the block is currently always of a fixed size,
765 this information isn't actually particularly useful in pcretest output, so
766 I have just removed it.
767
768 8. Three pre-processor statements accidentally did not start in column 1.
769 Sadly, there are *still* compilers around that complain, even though
770 standard C has not required this for well over a decade. Sigh.
771
772 9. In pcretest, the code for checking callouts passed small integers in the
773 callout_data field, which is a void * field. However, some picky compilers
774 complained about the casts involved for this on 64-bit systems. Now
775 pcretest passes the address of the small integer instead, which should get
776 rid of the warnings.
777
778 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
779 both compile and run time, and gives an error if an invalid UTF-8 sequence
780 is found. There is a option for disabling this check in cases where the
781 string is known to be correct and/or the maximum performance is wanted.
782
783 11. In response to a bug report, I changed one line in Makefile.in from
784
785 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
786 to
787 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
788
789 to look similar to other lines, but I have no way of telling whether this
790 is the right thing to do, as I do not use Windows. No doubt I'll get told
791 if it's wrong...
792
793
794 Version 4.3 21-May-03
795 ---------------------
796
797 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
798 Makefile.
799
800 2. Some refactoring to improve the quality of the code:
801
802 (i) The utf8_table... variables are now declared "const".
803
804 (ii) The code for \cx, which used the "case flipping" table to upper case
805 lower case letters, now just substracts 32. This is ASCII-specific,
806 but the whole concept of \cx is ASCII-specific, so it seems
807 reasonable.
808
809 (iii) PCRE was using its character types table to recognize decimal and
810 hexadecimal digits in the pattern. This is silly, because it handles
811 only 0-9, a-f, and A-F, but the character types table is locale-
812 specific, which means strange things might happen. A private
813 table is now used for this - though it costs 256 bytes, a table is
814 much faster than multiple explicit tests. Of course, the standard
815 character types table is still used for matching digits in subject
816 strings against \d.
817
818 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
819 ending in _t are). So I've renamed it as ESC_tee.
820
821 3. The first argument for regexec() in the POSIX wrapper should have been
822 defined as "const".
823
824 4. Changed pcretest to use malloc() for its buffers so that they can be
825 Electric Fenced for debugging.
826
827 5. There were several places in the code where, in UTF-8 mode, PCRE would try
828 to read one or more bytes before the start of the subject string. Often this
829 had no effect on PCRE's behaviour, but in some circumstances it could
830 provoke a segmentation fault.
831
832 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
833 to try to read one or more bytes before the start of the subject string.
834
835 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
836 UTF-8 support could misbehave in various ways if the subject string
837 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
838 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
839 back over UTF-8 characters.)
840
841
842 Version 4.2 14-Apr-03
843 ---------------------
844
845 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
846
847 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
848 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
849 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
850 [WIN_PREFIX]: new variable, "cyg" for Cygwin
851 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
852 and BUILD_EXEEXT
853 Note: automatic setting of the BUILD variables is not yet working
854 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
855 compile-time but not at link-time
856 [LINK]: use for linking executables only
857 make different versions for Windows and non-Windows
858 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
859 libraries
860 [LINK_FOR_BUILD]: new variable
861 [OBJEXT]: use throughout
862 [EXEEXT]: use throughout
863 <winshared>: new target
864 <wininstall>: new target
865 <dftables.o>: use native compiler
866 <dftables>: use native linker
867 <install>: handle Windows platform correctly
868 <clean>: ditto
869 <check>: ditto
870 copy DLL to top builddir before testing
871
872 As part of these changes, -no-undefined was removed again. This was reported
873 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
874 in any case.
875
876 3. Some tidies to get rid of compiler warnings:
877
878 . In the match_data structure, match_limit was an unsigned long int, whereas
879 match_call_count was an int. I've made them both unsigned long ints.
880
881 . In pcretest the fact that a const uschar * doesn't automatically cast to
882 a void * provoked a warning.
883
884 . Turning on some more compiler warnings threw up some "shadow" variables
885 and a few more missing casts.
886
887 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
888 option, a class that contained a single character with a value between 128
889 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
890
891 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
892 option, a class that contained several characters, but with at least one
893 whose value was between 128 and 255 caused PCRE to crash.
894
895
896 Version 4.1 12-Mar-03
897 ---------------------
898
899 1. Compiling with gcc -pedantic found a couple of places where casts were
900 needed, and a string in dftables.c that was longer than standard compilers are
901 required to support.
902
903 2. Compiling with Sun's compiler found a few more places where the code could
904 be tidied up in order to avoid warnings.
905
906 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
907 first of these names is deprecated in the latest Autoconf in favour of the name
908 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
909 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
910 analogy I have changed it to CFLAGS_FOR_BUILD.
911
912 4. Added -no-undefined to the linking command in the Makefile, because this is
913 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
914 linking step for the pcreposix library.
915
916 5. PCRE was failing to diagnose the case of two named groups with the same
917 name.
918
919 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
920 literal character that is needed in the subject for a match, and scans along to
921 ensure that it is present before embarking on the full matching process. This
922 saves time in cases of nested unlimited repeats that are never going to match.
923 Problem: the scan can take a lot of time if the subject is very long (e.g.
924 megabytes), thus penalizing straightforward matches. It is now done only if the
925 amount of subject to be scanned is less than 1000 bytes.
926
927 7. A lesser problem with the same optimization is that it was recording the
928 first character of an anchored pattern as "needed", thus provoking a search
929 right along the subject, even when the first match of the pattern was going to
930 fail. The "needed" character is now not set for anchored patterns, unless it
931 follows something in the pattern that is of non-fixed length. Thus, it still
932 fulfils its original purpose of finding quick non-matches in cases of nested
933 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
934
935
936 Version 4.0 17-Feb-03
937 ---------------------
938
939 1. If a comment in an extended regex that started immediately after a meta-item
940 extended to the end of string, PCRE compiled incorrect data. This could lead to
941 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
942
943 2. Moved to autoconf 2.53 and libtool 1.4.2.
944
945 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
946 the special perltest8 script is no longer needed - all the tests can be run
947 from a single perltest script.
948
949 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
950 by \s. It has now been removed in PCRE. This means it isn't recognized as
951 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
952 class [:space:] *does* include VT, thereby creating a mess.
953
954 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
955 space and tab.
956
957 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
958 its new features into the main test script, reducing the number of scripts.
959
960 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
961 were backward compatible, and made the (?i) apply to the whole pattern, as if
962 /i were given. Now it behaves more logically, and applies the option setting
963 only to what follows. PCRE has been changed to follow suit. However, if it
964 finds options settings right at the start of the pattern, it extracts them into
965 the global options, as before. Thus, they show up in the info data.
966
967 8. Added support for the \Q...\E escape sequence. Characters in between are
968 treated as literals. This is slightly different from Perl in that $ and @ are
969 also handled as literals inside the quotes. In Perl, they will cause variable
970 interpolation. Note the following examples:
971
972 Pattern PCRE matches Perl matches
973
974 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
975 \Qabc\$xyz\E abc\$xyz abc\$xyz
976 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
977
978 For compatibility with Perl, \Q...\E sequences are recognized inside character
979 classes as well as outside them.
980
981 9. Re-organized 3 code statements in pcretest to avoid "overflow in
982 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
983 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
984 signed/unsigned warnings.
985
986 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
987 option for pcretest, so I've replaced it by a simple function that does just
988 that job.
989
990 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
991 "pcregrep -".
992
993 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
994 Java package. This provides some syntactic sugar for simple cases of what my
995 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
996 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
997 item, you can use this simplified notation. Note that only makes sense with
998 greedy quantifiers. Consequently, the use of the possessive quantifier forces
999 greediness, whatever the setting of the PCRE_UNGREEDY option.
1000
1001 13. A change of greediness default within a pattern was not taking effect at
1002 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1003 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1004 was abstracted outside.
1005
1006 14. PCRE now supports the \G assertion. It is true when the current matching
1007 position is at the start point of the match. This differs from \A when the
1008 starting offset is non-zero. Used with the /g option of pcretest (or similar
1009 code), it works in the same way as it does for Perl's /g option. If all
1010 alternatives of a regex begin with \G, the expression is anchored to the start
1011 match position, and the "anchored" flag is set in the compiled expression.
1012
1013 15. Some bugs concerning the handling of certain option changes within patterns
1014 have been fixed. These applied to options other than (?ims). For example,
1015 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1016 been the other way round. Some of this was related to change 7 above.
1017
1018 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1019 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1020 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1021 POSIX classes only within a class (e.g. /[[:alpha:]]/).
1022
1023 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1024 mode. Unlike ".", it always matches newline, whatever the setting of
1025 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1026 assertions. Perl allows it, but it doesn't (in general) work because it can't
1027 calculate the length of the lookbehind. At least, that's the case for Perl
1028 5.8.0 - I've been told they are going to document that it doesn't work in
1029 future.
1030
1031 18. Added an error diagnosis for escapes that PCRE does not support: these are
1032 \L, \l, \N, \P, \p, \U, \u, and \X.
1033
1034 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1035 reading past the end of the pattern in cases such as /[abcd/.
1036
1037 20. PCRE was getting more memory than necessary for patterns with classes that
1038 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1039
1040 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1041 compiling PCRE for use with Virtual Pascal.
1042
1043 22. Small fix to the Makefile to make it work properly if the build is done
1044 outside the source tree.
1045
1046 23. Added a new extension: a condition to go with recursion. If a conditional
1047 subpattern starts with (?(R) the "true" branch is used if recursion has
1048 happened, whereas the "false" branch is used only at the top level.
1049
1050 24. When there was a very long string of literal characters (over 255 bytes
1051 without UTF support, over 250 bytes with UTF support), the computation of how
1052 much memory was required could be incorrect, leading to segfaults or other
1053 strange effects.
1054
1055 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1056 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1057 there was a subsequent back reference to those brackets. This meant that, for
1058 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1059 possible to check for precisely this case. All we can do is abandon the
1060 optimization if .* occurs inside capturing brackets when there are any back
1061 references whatsoever. (See below for a better fix that came later.)
1062
1063 26. The handling of the optimization for finding the first character of a
1064 non-anchored pattern, and for finding a character that is required later in the
1065 match were failing in some cases. This didn't break the matching; it just
1066 failed to optimize when it could. The way this is done has been re-implemented.
1067
1068 27. Fixed typo in error message for invalid (?R item (it said "(?p").
1069
1070 28. Added a new feature that provides some of the functionality that Perl
1071 provides with (?{...}). The facility is termed a "callout". The way it is done
1072 in PCRE is for the caller to provide an optional function, by setting
1073 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1074 global variable. By default it is unset, which disables all calling out. To get
1075 the function called, the regex must include (?C) at appropriate points. This
1076 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1077 This provides a means of identifying different callout points. When PCRE
1078 reaches such a point in the regex, if pcre_callout has been set, the external
1079 function is called. It is provided with data in a structure called
1080 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1081 matching continues; if it returns a non-zero value, the match at the current
1082 point fails. However, backtracking will occur if possible. [This was changed
1083 later and other features added - see item 49 below.]
1084
1085 29. pcretest is upgraded to test the callout functionality. It provides a
1086 callout function that displays information. By default, it shows the start of
1087 the match and the current position in the text. There are some new data escapes
1088 to vary what happens:
1089
1090 \C+ in addition, show current contents of captured substrings
1091 \C- do not supply a callout function
1092 \C!n return 1 when callout number n is reached
1093 \C!n!m return 1 when callout number n is reached for the mth time
1094
1095 30. If pcregrep was called with the -l option and just a single file name, it
1096 output "<stdin>" if a match was found, instead of the file name.
1097
1098 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1099 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1100 pcre_exec(). This saves a malloc/free per call. The default value of
1101 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1102 when configuring.
1103
1104 32. The default maximum size of a compiled pattern is 64K. There have been a
1105 few cases of people hitting this limit. The code now uses macros to handle the
1106 storing of links as offsets within the compiled pattern. It defaults to 2-byte
1107 links, but this can be changed to 3 or 4 bytes by --with-link-size when
1108 configuring. Tests 2 and 5 work only with 2-byte links because they output
1109 debugging information about compiled patterns.
1110
1111 33. Internal code re-arrangements:
1112
1113 (a) Moved the debugging function for printing out a compiled regex into
1114 its own source file (printint.c) and used #include to pull it into
1115 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1116 separate copies.
1117
1118 (b) Defined the list of op-code names for debugging as a macro in
1119 internal.h so that it is next to the definition of the opcodes.
1120
1121 (c) Defined a table of op-code lengths for simpler skipping along compiled
1122 code. This is again a macro in internal.h so that it is next to the
1123 definition of the opcodes.
1124
1125 34. Added support for recursive calls to individual subpatterns, along the
1126 lines of Robin Houston's patch (but implemented somewhat differently).
1127
1128 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1129 allow it to read and process whole directories in Win32. This code was
1130 contributed by Lionel Fourquaux; it has not been tested by me.
1131
1132 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1133 used to name a group. Names consist of alphanumerics and underscores, and must
1134 be unique. Back references use the syntax (?P=name) and recursive calls use
1135 (?P>name) which is a PCRE extension to the Python extension. Groups still have
1136 numbers. The function pcre_fullinfo() can be used after compilation to extract
1137 a name/number map. There are three relevant calls:
1138
1139 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
1140 PCRE_INFO_NAMECOUNT yields the number of entries
1141 PCRE_INFO_NAMETABLE yields a pointer to the map.
1142
1143 The map is a vector of fixed-size entries. The size of each entry depends on
1144 the length of the longest name used. The first two bytes of each entry are the
1145 group number, most significant byte first. There follows the corresponding
1146 name, zero terminated. The names are in alphabetical order.
1147
1148 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1149 case instead of 255. Making it the same both with and without UTF-8 support
1150 means that the same test output works with both.
1151
1152 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1153 calling malloc() with a zero argument.
1154
1155 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1156 optimization. I've improved things by keeping a bitmap of backreferences with
1157 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1158 fact referenced, the optimization can be applied. It is unlikely that a
1159 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1160 the match to follow \n) will appear inside brackets with a number greater than
1161 31, but if it does, any back reference > 31 suppresses the optimization.
1162
1163 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1164 of disabling numbered capturing parentheses. Any opening parenthesis that is
1165 not followed by ? behaves as if it were followed by ?: but named parentheses
1166 can still be used for capturing (and they will acquire numbers in the usual
1167 way).
1168
1169 41. Redesigned the return codes from the match() function into yes/no/error so
1170 that errors can be passed back from deep inside the nested calls. A malloc
1171 failure while inside a recursive subpattern call now causes the
1172 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1173
1174 42. It is now possible to set a limit on the number of times the match()
1175 function is called in a call to pcre_exec(). This facility makes it possible to
1176 limit the amount of recursion and backtracking, though not in a directly
1177 obvious way, because the match() function is used in a number of different
1178 circumstances. The count starts from zero for each position in the subject
1179 string (for non-anchored patterns). The default limit is, for compatibility, a
1180 large number, namely 10 000 000. You can change this in two ways:
1181
1182 (a) When configuring PCRE before making, you can use --with-match-limit=n
1183 to set a default value for the compiled library.
1184
1185 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1186 a different value is set. See 45 below.
1187
1188 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1189
1190 43. Added a new function pcre_config(int, void *) to enable run-time extraction
1191 of things that can be changed at compile time. The first argument specifies
1192 what is wanted and the second points to where the information is to be placed.
1193 The current list of available information is:
1194
1195 PCRE_CONFIG_UTF8
1196
1197 The output is an integer that is set to one if UTF-8 support is available;
1198 otherwise it is set to zero.
1199
1200 PCRE_CONFIG_NEWLINE
1201
1202 The output is an integer that it set to the value of the code that is used for
1203 newline. It is either LF (10) or CR (13).
1204
1205 PCRE_CONFIG_LINK_SIZE
1206
1207 The output is an integer that contains the number of bytes used for internal
1208 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1209
1210 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1211
1212 The output is an integer that contains the threshold above which the POSIX
1213 interface uses malloc() for output vectors. See item 31 above.
1214
1215 PCRE_CONFIG_MATCH_LIMIT
1216
1217 The output is an unsigned integer that contains the default limit of the number
1218 of match() calls in a pcre_exec() execution. See 42 above.
1219
1220 44. pcretest has been upgraded by the addition of the -C option. This causes it
1221 to extract all the available output from the new pcre_config() function, and to
1222 output it. The program then exits immediately.
1223
1224 45. A need has arisen to pass over additional data with calls to pcre_exec() in
1225 order to support additional features. One way would have been to define
1226 pcre_exec2() (for example) with extra arguments, but this would not have been
1227 extensible, and would also have required all calls to the original function to
1228 be mapped to the new one. Instead, I have chosen to extend the mechanism that
1229 is used for passing in "extra" data from pcre_study().
1230
1231 The pcre_extra structure is now exposed and defined in pcre.h. It currently
1232 contains the following fields:
1233
1234 flags a bitmap indicating which of the following fields are set
1235 study_data opaque data from pcre_study()
1236 match_limit a way of specifying a limit on match() calls for a specific
1237 call to pcre_exec()
1238 callout_data data for callouts (see 49 below)
1239
1240 The flag bits are also defined in pcre.h, and are
1241
1242 PCRE_EXTRA_STUDY_DATA
1243 PCRE_EXTRA_MATCH_LIMIT
1244 PCRE_EXTRA_CALLOUT_DATA
1245
1246 The pcre_study() function now returns one of these new pcre_extra blocks, with
1247 the actual study data pointed to by the study_data field, and the
1248 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1249 before. That is, this change is entirely upwards-compatible and requires no
1250 change to existing code.
1251
1252 If you want to pass in additional data to pcre_exec(), you can either place it
1253 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1254 block.
1255
1256 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1257 data string contains the escape sequence \M, pcretest calls pcre_exec() several
1258 times with different match limits, until it finds the minimum value needed for
1259 pcre_exec() to complete. The value is then output. This can be instructive; for
1260 most simple matches the number is quite small, but for pathological cases it
1261 gets very large very quickly.
1262
1263 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1264 returns the size of the data block pointed to by the study_data field in a
1265 pcre_extra block, that is, the value that was passed as the argument to
1266 pcre_malloc() when PCRE was getting memory in which to place the information
1267 created by pcre_study(). The fourth argument should point to a size_t variable.
1268 pcretest has been extended so that this information is shown after a successful
1269 pcre_study() call when information about the compiled regex is being displayed.
1270
1271 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1272 because what follows is always an absolute path. (Later: it turns out that this
1273 is more than cosmetic for MinGW, because it doesn't like empty path
1274 components.)
1275
1276 49. Some changes have been made to the callout feature (see 28 above):
1277
1278 (i) A callout function now has three choices for what it returns:
1279
1280 0 => success, carry on matching
1281 > 0 => failure at this point, but backtrack if possible
1282 < 0 => serious error, return this value from pcre_exec()
1283
1284 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1285 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1286 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1287 use by callout functions. It will never be used by PCRE itself.
1288
1289 (ii) The pcre_extra structure (see 45 above) has a void * field called
1290 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1291 pcre_callout_block structure has a field of the same name. The contents of
1292 the field passed in the pcre_extra structure are passed to the callout
1293 function in the corresponding field in the callout block. This makes it
1294 easier to use the same callout-containing regex from multiple threads. For
1295 testing, the pcretest program has a new data escape
1296
1297 \C*n pass the number n (may be negative) as callout_data
1298
1299 If the callout function in pcretest receives a non-zero value as
1300 callout_data, it returns that value.
1301
1302 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1303 there were some redundant $(CFLAGS) in commands that are now specified as
1304 $(LINK), which already includes $(CFLAGS).
1305
1306 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1307 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1308 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1309 one-byte characters throughout. Note that case-insensitive matching applies
1310 only to characters whose values are less than 256. PCRE doesn't support the
1311 notion of cases for higher-valued characters.
1312
1313 (i) A character class whose characters are all within 0-255 is handled as
1314 a bit map, and the map is inverted for negative classes. Previously, a
1315 character > 255 always failed to match such a class; however it should
1316 match if the class was a negative one (e.g. [^ab]). This has been fixed.
1317
1318 (ii) A negated character class with a single character < 255 is coded as
1319 "not this character" (OP_NOT). This wasn't working properly when the test
1320 character was multibyte, either singly or repeated.
1321
1322 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1323 mode, for example: \x{100}{2,3}.
1324
1325 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1326 singly or repeated) now correctly test multibyte characters. However,
1327 PCRE doesn't recognize any characters with values greater than 255 as
1328 digits, spaces, or word characters. Such characters always match \D, \S,
1329 and \W, and never match \d, \s, or \w.
1330
1331 (v) Classes may now contain characters and character ranges with values
1332 greater than 255. For example: [ab\x{100}-\x{400}].
1333
1334 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1335 PCRE in UTF-8 mode.
1336
1337 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1338 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1339 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1340 value.)
1341
1342 53. The single man page has become too large. I have therefore split it up into
1343 a number of separate man pages. These also give rise to individual HTML pages;
1344 these are now put in a separate directory, and there is an index.html page that
1345 lists them all. Some hyperlinking between the pages has been installed.
1346
1347 54. Added convenience functions for handling named capturing parentheses.
1348
1349 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1350 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1351 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1352 are faulted.
1353
1354 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1355 calling configure. These values are used when compiling the dftables.c program
1356 which is run to generate the source of the default character tables. They
1357 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1358 you will need to set these values.
1359
1360 57. Updated the building process for Windows DLL, as provided by Fred Cox.
1361
1362
1363 Version 3.9 02-Jan-02
1364 ---------------------
1365
1366 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1367
1368 2. If --disable-static was given, the building process failed when trying to
1369 build pcretest and pcregrep. (For some reason it was using libtool to compile
1370 them, which is not right, as they aren't part of the library.)
1371
1372
1373 Version 3.8 18-Dec-01
1374 ---------------------
1375
1376 1. The experimental UTF-8 code was completely screwed up. It was packing the
1377 bytes in the wrong order. How dumb can you get?
1378
1379
1380 Version 3.7 29-Oct-01
1381 ---------------------
1382
1383 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1384 This caused pcretest, when used on the test data, to segfault. Unfortunately,
1385 this didn't happen under Solaris 8, where I normally test things.
1386
1387 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1388 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1389 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1390 this fix an hour or so after the initial 3.7 release.)
1391
1392
1393 Version 3.6 23-Oct-01
1394 ---------------------
1395
1396 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1397 offsets passed as NULL with zero offset count.
1398
1399 2. The config.guess and config.sub files had not been updated when I moved to
1400 the latest autoconf.
1401
1402
1403 Version 3.5 15-Aug-01
1404 ---------------------
1405
1406 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1407 had been forgotten.
1408
1409 2. By using declared but undefined structures, we can avoid using "void"
1410 definitions in pcre.h while keeping the internal definitions of the structures
1411 private.
1412
1413 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1414 user point of view, this means that both static and shared libraries are built
1415 by default, but this can be individually controlled. More of the work of
1416 handling this static/shared cases is now inside libtool instead of PCRE's make
1417 file.
1418
1419 4. The pcretest utility is now installed along with pcregrep because it is
1420 useful for users (to test regexs) and by doing this, it automatically gets
1421 relinked by libtool. The documentation has been turned into a man page, so
1422 there are now .1, .txt, and .html versions in /doc.
1423
1424 5. Upgrades to pcregrep:
1425 (i) Added long-form option names like gnu grep.
1426 (ii) Added --help to list all options with an explanatory phrase.
1427 (iii) Added -r, --recursive to recurse into sub-directories.
1428 (iv) Added -f, --file to read patterns from a file.
1429
1430 6. pcre_exec() was referring to its "code" argument before testing that
1431 argument for NULL (and giving an error if it was NULL).
1432
1433 7. Upgraded Makefile.in to allow for compiling in a different directory from
1434 the source directory.
1435
1436 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
1437 options bits, the pointer it was passed was to an int instead of to an unsigned
1438 long int. This mattered only on 64-bit systems.
1439
1440 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
1441 generated) instead of pcre.in, which it its source. Also made the same change
1442 in several of the .c files.
1443
1444 10. A new release of gcc defines printf() as a macro, which broke pcretest
1445 because it had an ifdef in the middle of a string argument for printf(). Fixed
1446 by using separate calls to printf().
1447
1448 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
1449 script, to force use of CR or LF instead of \n in the source. On non-Unix
1450 systems, the value can be set in config.h.
1451
1452 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
1453 absolute limit. Changed the text of the error message to make this clear, and
1454 likewise updated the man page.
1455
1456 13. The limit of 99 on the number of capturing subpatterns has been removed.
1457 The new limit is 65535, which I hope will not be a "real" limit.
1458
1459
1460 Version 3.4 22-Aug-00
1461 ---------------------
1462
1463 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
1464
1465 2. Diagnose condition (?(0) as an error instead of crashing on matching.
1466
1467
1468 Version 3.3 01-Aug-00
1469 ---------------------
1470
1471 1. If an octal character was given, but the value was greater than \377, it
1472 was not getting masked to the least significant bits, as documented. This could
1473 lead to crashes in some systems.
1474
1475 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
1476 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
1477
1478 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
1479 These just pass their arguments on to (pcre_free)(), but they are provided
1480 because some uses of PCRE bind it to non-C systems that can call its functions,
1481 but cannot call free() or pcre_free() directly.
1482
1483 4. Add "make test" as a synonym for "make check". Corrected some comments in
1484 the Makefile.
1485
1486 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
1487 Makefile.
1488
1489 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
1490 command called pgrep for grepping around the active processes.
1491
1492 7. Added the beginnings of support for UTF-8 character strings.
1493
1494 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
1495 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
1496 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
1497 out for the ar command.)
1498
1499
1500 Version 3.2 12-May-00
1501 ---------------------
1502
1503 This is purely a bug fixing release.
1504
1505 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
1506 of ZA. This was just one example of several cases that could provoke this bug,
1507 which was introduced by change 9 of version 2.00. The code for breaking
1508 infinite loops after an iteration that matches an empty string was't working
1509 correctly.
1510
1511 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
1512 when matched against abbab (for example). After matching an empty string, it
1513 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
1514 caused it to match further down the string than it should.
1515
1516 3. The code contained an inclusion of sys/types.h. It isn't clear why this
1517 was there because it doesn't seem to be needed, and it causes trouble on some
1518 systems, as it is not a Standard C header. It has been removed.
1519
1520 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
1521 were reported on the Macintosh. The changes were from
1522
1523 while ((c = *(++ptr)) != 0 && c != '\n');
1524 to
1525 while ((c = *(++ptr)) != 0 && c != '\n') ;
1526
1527 Totally extraordinary, but if that's what it takes...
1528
1529 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
1530 available. Added HAVE_BCOPY and an autoconf test for it; if neither
1531 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
1532 assumes the way PCRE uses memmove() (always moving upwards).
1533
1534 6. PCRE is being used in one environment where strchr() is not available. There
1535 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
1536 faster code anyway.
1537
1538
1539 Version 3.1 09-Feb-00
1540 ---------------------
1541
1542 The only change in this release is the fixing of some bugs in Makefile.in for
1543 the "install" target:
1544
1545 (1) It was failing to install pcreposix.h.
1546
1547 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
1548
1549
1550 Version 3.0 01-Feb-00
1551 ---------------------
1552
1553 1. Add support for the /+ modifier to perltest (to output $` like it does in
1554 pcretest).
1555
1556 2. Add support for the /g modifier to perltest.
1557
1558 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
1559 matches null strings.
1560
1561 4. Fix perltest so that it doesn't do unwanted things when fed an empty
1562 pattern. Perl treats empty patterns specially - it reuses the most recent
1563 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
1564 effect.
1565
1566 5. The POSIX interface was broken in that it was just handing over the POSIX
1567 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
1568 required a bigger vector, with some working space on the end. This means that
1569 the POSIX wrapper now has to get and free some memory, and copy the results.
1570
1571 6. Added some simple autoconf support, placing the test data and the
1572 documentation in separate directories, re-organizing some of the
1573 information files, and making it build pcre-config (a GNU standard). Also added
1574 libtool support for building PCRE as a shared library, which is now the
1575 default.
1576
1577 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
1578 09 are not valid octal constants. Single digits will be used for minor values
1579 less than 10.
1580
1581 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
1582 existing programs that set these in the POSIX interface can use PCRE without
1583 modification.
1584
1585 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
1586 return all that pcre_info() returns, plus additional data. The pcre_info()
1587 function is retained for compatibility, but is considered to be obsolete.
1588
1589 10. Added experimental recursion feature (?R) to handle one common case that
1590 Perl 5.6 will be able to do with (?p{...}).
1591
1592 11. Added support for POSIX character classes like [:alpha:], which Perl is
1593 adopting.
1594
1595
1596 Version 2.08 31-Aug-99
1597 ----------------------
1598
1599 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
1600 trying to match at the startoffset position, but instead was moving forward to
1601 the next newline as if a previous match had failed.
1602
1603 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
1604 and could get into a loop if a null string was matched other than at the start
1605 of the subject.
1606
1607 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
1608 be distinguished at compile time, and for completeness also added PCRE_DATE.
1609
1610 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
1611 in GnuWin32 environments.
1612
1613
1614 Version 2.07 29-Jul-99
1615 ----------------------
1616
1617 1. The documentation is now supplied in plain text form and HTML as well as in
1618 the form of man page sources.
1619
1620 2. C++ compilers don't like assigning (void *) values to other pointer types.
1621 In particular this affects malloc(). Although there is no problem in Standard
1622 C, I've put in casts to keep C++ compilers happy.
1623
1624 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
1625 should be (const char *).
1626
1627 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
1628 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
1629 However, I haven't made this a standard facility. The documentation doesn't
1630 mention it, and the Makefile doesn't support it.
1631
1632 5. The Makefile now contains an "install" target, with editable destinations at
1633 the top of the file. The pcretest program is not installed.
1634
1635 6. pgrep -V now gives the PCRE version number and date.
1636
1637 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
1638 causing the entire string to be ignored, instead of just the last character.
1639
1640 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
1641 non-matching string, it can take a very, very long time, even for strings of
1642 quite modest length, because of the nested recursion. PCRE now does better in
1643 some of these cases. It does this by remembering the last required literal
1644 character in the pattern, and pre-searching the subject to ensure it is present
1645 before running the real match. In other words, it applies a heuristic to detect
1646 some types of certain failure quickly, and in the above example, if presented
1647 with a string that has no trailing " it gives "no match" very quickly.
1648
1649 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
1650 other alternatives are tried instead.
1651
1652
1653 Version 2.06 09-Jun-99
1654 ----------------------
1655
1656 1. Change pcretest's output for amount of store used to show just the code
1657 space, because the remainder (the data block) varies in size between 32-bit and
1658 64-bit systems.
1659
1660 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
1661 start matching at. This allows lookbehinds to work when searching for multiple
1662 occurrences in a string.
1663
1664 3. Added additional options to pcretest for testing multiple occurrences:
1665
1666 /+ outputs the rest of the string that follows a match
1667 /g loops for multiple occurrences, using the new startoffset argument
1668 /G loops for multiple occurrences by passing an incremented pointer
1669
1670 4. PCRE wasn't doing the "first character" optimization for patterns starting
1671 with \b or \B, though it was doing it for other lookbehind assertions. That is,
1672 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
1673 the letter 'x'. On long subject strings, this gives a significant speed-up.
1674
1675
1676 Version 2.05 21-Apr-99
1677 ----------------------
1678
1679 1. Changed the type of magic_number from int to long int so that it works
1680 properly on 16-bit systems.
1681
1682 2. Fixed a bug which caused patterns starting with .* not to work correctly
1683 when the subject string contained newline characters. PCRE was assuming
1684 anchoring for such patterns in all cases, which is not correct because .* will
1685 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
1686 DOTALL is set at top level; otherwise it knows that patterns starting with .*
1687 must be retried after every newline in the subject.
1688
1689
1690 Version 2.04 18-Feb-99
1691 ----------------------
1692
1693 1. For parenthesized subpatterns with repeats whose minimum was zero, the
1694 computation of the store needed to hold the pattern was incorrect (too large).
1695 If such patterns were nested a few deep, this could multiply and become a real
1696 problem.
1697
1698 2. Added /M option to pcretest to show the memory requirement of a specific
1699 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
1700
1701 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
1702 compiled in such a way that the backtracking after subsequent failure was
1703 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
1704 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
1705
1706
1707 Version 2.03 02-Feb-99
1708 ----------------------
1709
1710 1. Fixed typo and small mistake in man page.
1711
1712 2. Added 4th condition (GPL supersedes if conflict) and created separate
1713 LICENCE file containing the conditions.
1714
1715 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
1716 Perl, that is the internal \ allows the delimiter to be included in the
1717 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
1718 the final delimiter, add \ to the end of the pattern (to test the error).
1719
1720 4. Added the convenience functions for extracting substrings after a successful
1721 match. Updated pcretest to make it able to test these functions.
1722
1723
1724 Version 2.02 14-Jan-99
1725 ----------------------
1726
1727 1. Initialized the working variables associated with each extraction so that
1728 their saving and restoring doesn't refer to uninitialized store.
1729
1730 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
1731 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
1732 fix the problem.
1733
1734 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
1735 calls, and wasn't printing the correct value for compiling calls. Increased the
1736 default value of LOOPREPEAT, and the number of significant figures in the
1737 times.
1738
1739 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
1740
1741 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
1742 a building problem on Windows NT with a FAT file system.
1743
1744
1745 Version 2.01 21-Oct-98
1746 ----------------------
1747
1748 1. Changed the API for pcre_compile() to allow for the provision of a pointer
1749 to character tables built by pcre_maketables() in the current locale. If NULL
1750 is passed, the default tables are used.
1751
1752
1753 Version 2.00 24-Sep-98
1754 ----------------------
1755
1756 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
1757 it any more.
1758
1759 2. Allow quantification of (?>) groups, and make it work correctly.
1760
1761 3. The first character computation wasn't working for (?>) groups.
1762
1763 4. Correct the implementation of \Z (it is permitted to match on the \n at the
1764 end of the subject) and add 5.005's \z, which really does match only at the
1765 very end of the subject.
1766
1767 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
1768
1769 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
1770 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
1771 localized options. All options to pcre_study() were also removed.
1772
1773 7. Add other new features from 5.005:
1774
1775 $(?<= positive lookbehind
1776 $(?<! negative lookbehind
1777 (?imsx-imsx) added the unsetting capability
1778 such a setting is global if at outer level; local otherwise
1779 (?imsx-imsx:) non-capturing groups with option setting
1780 (?(cond)re|re) conditional pattern matching
1781
1782 A backreference to itself in a repeated group matches the previous
1783 captured string.
1784
1785 8. General tidying up of studying (both automatic and via "study")
1786 consequential on the addition of new assertions.
1787
1788 9. As in 5.005, unlimited repeated groups that could match an empty substring
1789 are no longer faulted at compile time. Instead, the loop is forcibly broken at
1790 runtime if any iteration does actually match an empty substring.
1791
1792 10. Include the RunTest script in the distribution.
1793
1794 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
1795 discrepancies, some of which were old and were also with respect to 5.004. They
1796 have now been fixed.
1797
1798
1799 Version 1.09 28-Apr-98
1800 ----------------------
1801
1802 1. A negated single character class followed by a quantifier with a minimum
1803 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
1804 program crashes, or just wrong answers. This did not apply to negated classes
1805 containing more than one character, or to minima other than one.
1806
1807
1808 Version 1.08 27-Mar-98
1809 ----------------------
1810
1811 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
1812
1813 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
1814 latter must appear before anything that relies on it in the pattern.
1815
1816
1817 Version 1.07 16-Feb-98
1818 ----------------------
1819
1820 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
1821 repeat of a potentially empty string).
1822
1823
1824 Version 1.06 23-Jan-98
1825 ----------------------
1826
1827 1. Added Markus Oberhumer's little patches for C++.
1828
1829 2. Literal strings longer than 255 characters were broken.
1830
1831
1832 Version 1.05 23-Dec-97
1833 ----------------------
1834
1835 1. Negated character classes containing more than one character were failing if
1836 PCRE_CASELESS was set at run time.
1837
1838
1839 Version 1.04 19-Dec-97
1840 ----------------------
1841
1842 1. Corrected the man page, where some "const" qualifiers had been omitted.
1843
1844 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
1845 input syntax.
1846
1847 3. Fixed memory leak which occurred when a regex with back references was
1848 matched with an offsets vector that wasn't big enough. The temporary memory
1849 that is used in this case wasn't being freed if the match failed.
1850
1851 4. Tidied pcretest to ensure it frees memory that it gets.
1852
1853 5. Temporary memory was being obtained in the case where the passed offsets
1854 vector was exactly big enough.
1855
1856 6. Corrected definition of offsetof() from change 5 below.
1857
1858 7. I had screwed up change 6 below and broken the rules for the use of
1859 setjmp(). Now fixed.
1860
1861
1862 Version 1.03 18-Dec-97
1863 ----------------------
1864
1865 1. A erroneous regex with a missing opening parenthesis was correctly
1866 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
1867 on some systems.
1868
1869 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
1870 it was reported that one broken compiler failed on the former because "code" is
1871 also an independent variable.
1872
1873 3. The erroneous regex a[]b caused an array overrun reference.
1874
1875 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
1876 fail on data ending with that character. (It was going on too far, and checking
1877 the next character, typically a binary zero.) This was specific to the
1878 optimized code for single-character negative classes.
1879
1880 5. Added a contributed patch from the TIN world which does the following:
1881
1882 + Add an undef for memmove, in case the the system defines a macro for it.
1883
1884 + Add a definition of offsetof(), in case there isn't one. (I don't know
1885 the reason behind this - offsetof() is part of the ANSI standard - but
1886 it does no harm).
1887
1888 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
1889 most of the places where whitespace preceded '#'. I have given up and
1890 allowed the remaining 2 cases to be at the margin.
1891
1892 + Rename some variables in pcre to eliminate shadowing. This seems very
1893 pedantic, but does no harm, of course.
1894
1895 6. Moved the call to setjmp() into its own function, to get rid of warnings
1896 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
1897
1898 7. Constructs such as \d{8,} were compiling into the equivalent of
1899 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
1900 outcome, but in this particular case used more store than had been allocated,
1901 which caused the bug to be discovered because it threw up an internal error.
1902
1903 8. The debugging code in both pcre and pcretest for outputting the compiled
1904 form of a regex was going wrong in the case of back references followed by
1905 curly-bracketed repeats.
1906
1907
1908 Version 1.02 12-Dec-97
1909 ----------------------
1910
1911 1. Typos in pcre.3 and comments in the source fixed.
1912
1913 2. Applied a contributed patch to get rid of places where it used to remove
1914 'const' from variables, and fixed some signed/unsigned and uninitialized
1915 variable warnings.
1916
1917 3. Added the "runtest" target to Makefile.
1918
1919 4. Set default compiler flag to -O2 rather than just -O.
1920
1921
1922 Version 1.01 19-Nov-97
1923 ----------------------
1924
1925 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
1926 like /([ab]*)*/, that is, for classes with more than one character in them.
1927
1928 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
1929 as /((?>a*))*/ (a PCRE_EXTRA facility).
1930
1931
1932 Version 1.00 18-Nov-97
1933 ----------------------
1934
1935 1. Added compile-time macros to support systems such as SunOS4 which don't have
1936 memmove() or strerror() but have other things that can be used instead.
1937
1938 2. Arranged that "make clean" removes the executables.
1939
1940
1941 Version 0.99 27-Oct-97
1942 ----------------------
1943
1944 1. Fixed bug in code for optimizing classes with only one character. It was
1945 initializing a 32-byte map regardless, which could cause it to run off the end
1946 of the memory it had got.
1947
1948 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
1949
1950
1951 Version 0.98 22-Oct-97
1952 ----------------------
1953
1954 1. Fixed bug in code for handling temporary memory usage when there are more
1955 back references than supplied space in the ovector. This could cause segfaults.
1956
1957
1958 Version 0.97 21-Oct-97
1959 ----------------------
1960
1961 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
1962
1963 2. Optimized negated single characters not to use a bit map.
1964
1965 3. Brought error texts together as macro definitions; clarified some of them;
1966 fixed one that was wrong - it said "range out of order" when it meant "invalid
1967 escape sequence".
1968
1969 4. Changed some char * arguments to const char *.
1970
1971 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
1972
1973 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
1974 pcretest.
1975
1976
1977 Version 0.96 16-Oct-97
1978 ----------------------
1979
1980 1. Added a simple "pgrep" utility to the distribution.
1981
1982 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
1983 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
1984 where "ddd" means "one or more decimal digits".
1985
1986 3. Fixed serious bug. If a pattern had a back reference, but the call to
1987 pcre_exec() didn't supply a large enough ovector to record the related
1988 identifying subpattern, the match always failed. PCRE now remembers the number
1989 of the largest back reference, and gets some temporary memory in which to save
1990 the offsets during matching if necessary, in order to ensure that
1991 backreferences always work.
1992
1993 4. Increased the compatibility with Perl in a number of ways:
1994
1995 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
1996 to request this handling. The option can be set at compile or exec time.
1997
1998 (b) $ matches before a terminating newline by default; an option
1999 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
2000 mode). The option can be set at compile or exec time.
2001
2002 (c) The handling of \ followed by a digit other than 0 is now supposed to be
2003 the same as Perl's. If the decimal number it represents is less than 10
2004 or there aren't that many previous left capturing parentheses, an octal
2005 escape is read. Inside a character class, it's always an octal escape,
2006 even if it is a single digit.
2007
2008 (d) An escaped but undefined alphabetic character is taken as a literal,
2009 unless PCRE_EXTRA is set. Currently this just reserves the remaining
2010 escapes.
2011
2012 (e) {0} is now permitted. (The previous item is removed from the compiled
2013 pattern).
2014
2015 5. Changed all the names of code files so that the basic parts are no longer
2016 than 10 characters, and abolished the teeny "globals.c" file.
2017
2018 6. Changed the handling of character classes; they are now done with a 32-byte
2019 bit map always.
2020
2021 7. Added the -d and /D options to pcretest to make it possible to look at the
2022 internals of compilation without having to recompile pcre.
2023
2024
2025 Version 0.95 23-Sep-97
2026 ----------------------
2027
2028 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
2029 \x20 at the start of a run of normal characters. These were being treated as
2030 real characters, instead of the source characters being re-checked.
2031
2032
2033 Version 0.94 18-Sep-97
2034 ----------------------
2035
2036 1. The functions are now thread-safe, with the caveat that the global variables
2037 containing pointers to malloc() and free() or alternative functions are the
2038 same for all threads.
2039
2040 2. Get pcre_study() to generate a bitmap of initial characters for non-
2041 anchored patterns when this is possible, and use it if passed to pcre_exec().
2042
2043
2044 Version 0.93 15-Sep-97
2045 ----------------------
2046
2047 1. /(b)|(:+)/ was computing an incorrect first character.
2048
2049 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
2050 but not actually doing anything yet.
2051
2052 3. Treat "-" characters in classes that cannot be part of ranges as literals,
2053 as Perl does (e.g. [-az] or [az-]).
2054
2055 4. Set the anchored flag if a branch starts with .* or .*? because that tests
2056 all possible positions.
2057
2058 5. Split up into different modules to avoid including unneeded functions in a
2059 compiled binary. However, compile and exec are still in one module. The "study"
2060 function is split off.
2061
2062 6. The character tables are now in a separate module whose source is generated
2063 by an auxiliary program - but can then be edited by hand if required. There are
2064 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
2065 toupper() in the code.
2066
2067 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
2068 make them global. Abolish the function for setting them, as the caller can now
2069 set them directly.
2070
2071
2072 Version 0.92 11-Sep-97
2073 ----------------------
2074
2075 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
2076 (e.g. /a{1,3}/) was broken (I mis-optimized it).
2077
2078 2. Caseless matching was not working in character classes if the characters in
2079 the pattern were in upper case.
2080
2081 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
2082
2083 4. Make PCRE_ANCHORED public and accept as a compile option.
2084
2085 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
2086 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
2087 pass them.
2088
2089 6. Give an error if bad option bits passed at compile or run time.
2090
2091 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
2092 pcretest to cause it to pass that flag.
2093
2094 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
2095 options, and the first character, if set.
2096
2097 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
2098
2099
2100 Version 0.91 10-Sep-97
2101 ----------------------
2102
2103 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
2104 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
2105
2106 2. PCRE was looping on encountering an indefinitely repeated back reference to
2107 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
2108 Perl does - treats the match as successful.
2109
2110 ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12