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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 505 - (show annotations) (download)
Tue Mar 9 16:50:47 2010 UTC (4 years, 4 months ago) by ph10
File size: 184737 byte(s)
Improve compile-time overrun checking.

1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.02 01-Mar-2010
5 ------------------------
6
7 1. The Unicode data tables have been updated to Unicode 5.2.0.
8
9 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
10 configured.
11
12 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
13 original author of that file, following a query about its status.
14
15 4. On systems that do not have stdint.h (e.g. Solaris), check for and include
16 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
17
18 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
19 quantifier applied to a forward-referencing subroutine call, could compile
20 incorrect code or give the error "internal error: previously-checked
21 referenced subpattern not found".
22
23 6. Both MS Visual Studio and Symbian OS have problems with initializing
24 variables to point to external functions. For these systems, therefore,
25 pcre_malloc etc. are now initialized to local functions that call the
26 relevant global functions.
27
28 7. There were two entries missing in the vectors called coptable and poptable
29 in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
30 I've fixed the data, and added a kludgy way of testing at compile time that
31 the lengths are correct (equal to the number of opcodes).
32
33 8. Following on from 7, I added a similar kludge to check the length of the
34 eint vector in pcreposix.c.
35
36 9. Error texts for pcre_compile() are held as one long string to avoid too
37 much relocation at load time. To find a text, the string is searched,
38 counting zeros. There was no check for running off the end of the string,
39 which could happen if a new error number was added without updating the
40 string.
41
42 10. \K gave a compile-time error if it appeared in a lookbehind assersion.
43
44 11. \K was not working if it appeared in an atomic group or in a group that
45 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
46 \K is "not well defined" if used in an assertion. PCRE now accepts it if
47 the assertion is positive, but not if it is negative.
48
49 12. Change 11 fortuitously reduced the size of the stack frame used in the
50 "match()" function of pcre_exec.c by one pointer. Forthcoming
51 implementation of support for (*MARK) will need an extra pointer on the
52 stack; I have reserved it now, so that the stack frame size does not
53 decrease.
54
55 13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
56 item in branch that calls a recursion is a subroutine call - as in the
57 second branch in the above example - was incorrectly given the compile-
58 time error "recursive call could loop indefinitely" because pcre_compile()
59 was not correctly checking the subroutine for matching a non-empty string.
60
61 14. The checks for overrunning compiling workspace could trigger after an
62 overrun had occurred. This is a "should never occur" error, but it can be
63 triggered by pathological patterns such as hundreds of nested parentheses.
64 The checks now trigger 100 bytes before the end of the workspace.
65
66
67 Version 8.01 19-Jan-2010
68 ------------------------
69
70 1. If a pattern contained a conditional subpattern with only one branch (in
71 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
72 computed the wrong minimum data length (which is of course zero for such
73 subpatterns). This could cause incorrect "no match" results.
74
75 2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
76 the pattern is reset in the first branch, pcre_compile() failed with
77 "internal error: code overflow at offset...". This happened only when
78 the reset was to the original external option setting. (An optimization
79 abstracts leading options settings into an external setting, which was the
80 cause of this.)
81
82 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
83 of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
84 assertion pattern did not match (meaning that the assertion was true), it
85 was incorrectly treated as false if the SKIP had been reached during the
86 matching. This also applied to assertions used as conditions.
87
88 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
89 assertion subpattern, including such a pattern used as a condition,
90 unpredictable results occurred, instead of the error return
91 PCRE_ERROR_DFA_UITEM.
92
93 5. The C++ GlobalReplace function was not working like Perl for the special
94 situation when an empty string is matched. It now does the fancy magic
95 stuff that is necessary.
96
97 6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
98 removed. (These were left over from very, very early versions of PCRE.)
99
100 7. Some cosmetic changes to the code to make life easier when compiling it
101 as part of something else:
102
103 (a) Change DEBUG to PCRE_DEBUG.
104
105 (b) In pcre_compile(), rename the member of the "branch_chain" structure
106 called "current" as "current_branch", to prevent a collision with the
107 Linux macro when compiled as a kernel module.
108
109 (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
110 prevent a collision with the Linux macro when compiled as a kernel
111 module.
112
113 8. In pcre_compile() there are some checks for integer overflows that used to
114 cast potentially large values to (double). This has been changed to that
115 when building, a check for int64_t is made, and if it is found, it is used
116 instead, thus avoiding the use of floating point arithmetic. (There is no
117 other use of FP in PCRE.) If int64_t is not found, the fallback is to
118 double.
119
120 9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
121 2005 (difference between two addresses compared to an unsigned value).
122
123 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
124 custom one, because of the following reported problem in Windows:
125
126 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
127 under Win32.
128 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
129 therefore missing the function definition.
130 - The compiler thus generates a "C" signature for the test function.
131 - The linker fails to find the "C" function.
132 - PCRE fails to configure if asked to do so against libbz2.
133
134 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
135 messages were output:
136
137 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
138 rerunning libtoolize, to keep the correct libtool macros in-tree.
139 Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
140
141 I have done both of these things.
142
143 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
144 most of the time, it *can* run out if it is given a pattern that contains a
145 runaway infinite recursion. I updated the discussion in the pcrestack man
146 page.
147
148 13. Now that we have gone to the x.xx style of version numbers, the minor
149 version may start with zero. Using 08 or 09 is a bad idea because users
150 might check the value of PCRE_MINOR in their code, and 08 or 09 may be
151 interpreted as invalid octal numbers. I've updated the previous comment in
152 configure.ac, and also added a check that gives an error if 08 or 09 are
153 used.
154
155 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
156 causing partial matching to fail when the end of the subject matched \W
157 in a UTF-8 pattern where \W was quantified with a minimum of 3.
158
159 15. There were some discrepancies between the declarations in pcre_internal.h
160 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
161 their definitions. The declarations used "const uschar *" and the
162 definitions used USPTR. Even though USPTR is normally defined as "const
163 unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
164 reported that: "This difference in casting confuses some C++ compilers, for
165 example, SunCC recognizes above declarations as different functions and
166 generates broken code for hbpcre." I have changed the declarations to use
167 USPTR.
168
169 16. GNU libtool is named differently on some systems. The autogen.sh script now
170 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
171 (FreeBSD).
172
173 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
174 (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
175 comment: "Figure out how to create a longlong from a string: strtoll and
176 equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
177 instance, but it only takes 2 args instead of 3!"
178
179 18. A subtle bug concerned with back references has been fixed by a change of
180 specification, with a corresponding code fix. A pattern such as
181 ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
182 refers, was giving matches when it shouldn't. For example, xa=xaaa would
183 match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
184 same bug. Such groups have to be quantified to be useful, or contained
185 inside another quantified group. (If there's no repetition, the reference
186 can never match.) The problem arises because, having left the group and
187 moved on to the rest of the pattern, a later failure that backtracks into
188 the group uses the captured value from the final iteration of the group
189 rather than the correct earlier one. I have fixed this in PCRE by forcing
190 any group that contains a reference to itself to be an atomic group; that
191 is, there cannot be any backtracking into it once it has completed. This is
192 similar to recursive and subroutine calls.
193
194
195 Version 8.00 19-Oct-09
196 ----------------------
197
198 1. The table for translating pcre_compile() error codes into POSIX error codes
199 was out-of-date, and there was no check on the pcre_compile() error code
200 being within the table. This could lead to an OK return being given in
201 error.
202
203 2. Changed the call to open a subject file in pcregrep from fopen(pathname,
204 "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
205 in a Windows environment.
206
207 3. The pcregrep --count option prints the count for each file even when it is
208 zero, as does GNU grep. However, pcregrep was also printing all files when
209 --files-with-matches was added. Now, when both options are given, it prints
210 counts only for those files that have at least one match. (GNU grep just
211 prints the file name in this circumstance, but including the count seems
212 more useful - otherwise, why use --count?) Also ensured that the
213 combination -clh just lists non-zero counts, with no names.
214
215 4. The long form of the pcregrep -F option was incorrectly implemented as
216 --fixed_strings instead of --fixed-strings. This is an incompatible change,
217 but it seems right to fix it, and I didn't think it was worth preserving
218 the old behaviour.
219
220 5. The command line items --regex=pattern and --regexp=pattern were not
221 recognized by pcregrep, which required --regex pattern or --regexp pattern
222 (with a space rather than an '='). The man page documented the '=' forms,
223 which are compatible with GNU grep; these now work.
224
225 6. No libpcreposix.pc file was created for pkg-config; there was just
226 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
227
228 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
229 when UCP support is not needed, by modifying the Python script that
230 generates it from Unicode data files. This should not matter if the module
231 is correctly used as a library, but I received one complaint about 50K of
232 unwanted data. My guess is that the person linked everything into his
233 program rather than using a library. Anyway, it does no harm.
234
235 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
236 was a minimum greater than 1 for a wide character in a possessive
237 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
238 which had an unlimited repeat of a nested, fixed maximum repeat of a wide
239 character. Chaos in the form of incorrect output or a compiling loop could
240 result.
241
242 9. The restrictions on what a pattern can contain when partial matching is
243 requested for pcre_exec() have been removed. All patterns can now be
244 partially matched by this function. In addition, if there are at least two
245 slots in the offset vector, the offset of the earliest inspected character
246 for the match and the offset of the end of the subject are set in them when
247 PCRE_ERROR_PARTIAL is returned.
248
249 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
250 synonymous with PCRE_PARTIAL, for backwards compatibility, and
251 PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
252 and may be more useful for multi-segment matching.
253
254 11. Partial matching with pcre_exec() is now more intuitive. A partial match
255 used to be given if ever the end of the subject was reached; now it is
256 given only if matching could not proceed because another character was
257 needed. This makes a difference in some odd cases such as Z(*FAIL) with the
258 string "Z", which now yields "no match" instead of "partial match". In the
259 case of pcre_dfa_exec(), "no match" is given if every matching path for the
260 final character ended with (*FAIL).
261
262 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
263 if the pattern had a "must contain" character that was already found in the
264 earlier partial match, unless partial matching was again requested. For
265 example, with the pattern /dog.(body)?/, the "must contain" character is
266 "g". If the first part-match was for the string "dog", restarting with
267 "sbody" failed. This bug has been fixed.
268
269 13. The string returned by pcre_dfa_exec() after a partial match has been
270 changed so that it starts at the first inspected character rather than the
271 first character of the match. This makes a difference only if the pattern
272 starts with a lookbehind assertion or \b or \B (\K is not supported by
273 pcre_dfa_exec()). It's an incompatible change, but it makes the two
274 matching functions compatible, and I think it's the right thing to do.
275
276 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
277 so that the demonstration program is easily available in environments where
278 PCRE has not been installed from source.
279
280 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
281 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
282 library.
283
284 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
285 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
286 is not the first non-POSIX option to be added. Clearly some people find
287 these options useful.
288
289 17. If a caller to the POSIX matching function regexec() passes a non-zero
290 value for nmatch with a NULL value for pmatch, the value of
291 nmatch is forced to zero.
292
293 18. RunGrepTest did not have a test for the availability of the -u option of
294 the diff command, as RunTest does. It now checks in the same way as
295 RunTest, and also checks for the -b option.
296
297 19. If an odd number of negated classes containing just a single character
298 interposed, within parentheses, between a forward reference to a named
299 subpattern and the definition of the subpattern, compilation crashed with
300 an internal error, complaining that it could not find the referenced
301 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
302 [The bug was that it was starting one character too far in when skipping
303 over the character class, thus treating the ] as data rather than
304 terminating the class. This meant it could skip too much.]
305
306 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
307 /g option in pcretest when the pattern contains \K, which makes it possible
308 to have an empty string match not at the start, even when the pattern is
309 anchored. Updated pcretest and pcredemo to use this option.
310
311 21. If the maximum number of capturing subpatterns in a recursion was greater
312 than the maximum at the outer level, the higher number was returned, but
313 with unset values at the outer level. The correct (outer level) value is
314 now given.
315
316 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
317 PCRE did not set those parentheses (unlike Perl). I have now found a way to
318 make it do so. The string so far is captured, making this feature
319 compatible with Perl.
320
321 23. The tests have been re-organized, adding tests 11 and 12, to make it
322 possible to check the Perl 5.10 features against Perl 5.10.
323
324 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
325 pattern matches a fixed length string. PCRE did not allow this; now it
326 does. Neither allows recursion.
327
328 25. I finally figured out how to implement a request to provide the minimum
329 length of subject string that was needed in order to match a given pattern.
330 (It was back references and recursion that I had previously got hung up
331 on.) This code has now been added to pcre_study(); it finds a lower bound
332 to the length of subject needed. It is not necessarily the greatest lower
333 bound, but using it to avoid searching strings that are too short does give
334 some useful speed-ups. The value is available to calling programs via
335 pcre_fullinfo().
336
337 26. While implementing 25, I discovered to my embarrassment that pcretest had
338 not been passing the result of pcre_study() to pcre_dfa_exec(), so the
339 study optimizations had never been tested with that matching function.
340 Oops. What is worse, even when it was passed study data, there was a bug in
341 pcre_dfa_exec() that meant it never actually used it. Double oops. There
342 were also very few tests of studied patterns with pcre_dfa_exec().
343
344 27. If (?| is used to create subpatterns with duplicate numbers, they are now
345 allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
346 on the other side of the coin, they are no longer allowed to have different
347 names, because these cannot be distinguished in PCRE, and this has caused
348 confusion. (This is a difference from Perl.)
349
350 28. When duplicate subpattern names are present (necessarily with different
351 numbers, as required by 27 above), and a test is made by name in a
352 conditional pattern, either for a subpattern having been matched, or for
353 recursion in such a pattern, all the associated numbered subpatterns are
354 tested, and the overall condition is true if the condition is true for any
355 one of them. This is the way Perl works, and is also more like the way
356 testing by number works.
357
358
359 Version 7.9 11-Apr-09
360 ---------------------
361
362 1. When building with support for bzlib/zlib (pcregrep) and/or readline
363 (pcretest), all targets were linked against these libraries. This included
364 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
365 libraries. This caused unwanted dependencies to be created. This problem
366 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
367 pcretest is linked with readline.
368
369 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
370 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
371 moved outside it again, because FALSE and TRUE are already defined in AIX,
372 but BOOL is not.
373
374 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
375 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
376
377 4. The pcregrep documentation said spaces were inserted as well as colons (or
378 hyphens) following file names and line numbers when outputting matching
379 lines. This is not true; no spaces are inserted. I have also clarified the
380 wording for the --colour (or --color) option.
381
382 5. In pcregrep, when --colour was used with -o, the list of matching strings
383 was not coloured; this is different to GNU grep, so I have changed it to be
384 the same.
385
386 6. When --colo(u)r was used in pcregrep, only the first matching substring in
387 each matching line was coloured. Now it goes on to look for further matches
388 of any of the test patterns, which is the same behaviour as GNU grep.
389
390 7. A pattern that could match an empty string could cause pcregrep to loop; it
391 doesn't make sense to accept an empty string match in pcregrep, so I have
392 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
393 seems to be how GNU grep behaves.
394
395 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
396 start or after a newline", because the conditional assertion was not being
397 correctly handled. The rule now is that both the assertion and what follows
398 in the first alternative must satisfy the test.
399
400 9. If auto-callout was enabled in a pattern with a conditional group whose
401 condition was an assertion, PCRE could crash during matching, both with
402 pcre_exec() and pcre_dfa_exec().
403
404 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
405 used for matching.
406
407 11. Unicode property support in character classes was not working for
408 characters (bytes) greater than 127 when not in UTF-8 mode.
409
410 12. Added the -M command line option to pcretest.
411
412 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
413
414 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
415
416 16. Added comments and documentation about mis-use of no_arg in the C++
417 wrapper.
418
419 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
420 from Martin Jerabek that uses macro names for all relevant character and
421 string constants.
422
423 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
424 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
425 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
426 these, but not everybody uses configure.
427
428 19. A conditional group that had only one branch was not being correctly
429 recognized as an item that could match an empty string. This meant that an
430 enclosing group might also not be so recognized, causing infinite looping
431 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
432 with the subject "ab", where knowledge that the repeated group can match
433 nothing is needed in order to break the loop.
434
435 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
436 exec(), but without supplying a callout function, matching went wrong.
437
438 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
439 leak if the size of the offset vector was greater than 30. When the vector
440 is smaller, the saved offsets during recursion go onto a local stack
441 vector, but for larger vectors malloc() is used. It was failing to free
442 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
443 error, in fact).
444
445 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
446 heapframe that is used only when UTF-8 support is enabled. This caused no
447 problem, but was untidy.
448
449 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
450 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
451 included within another project.
452
453 24. Steven Van Ingelgem's patches to add more options to the CMake support,
454 slightly modified by me:
455
456 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
457 not building pcregrep.
458
459 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
460 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
461
462 25. Forward references, both numeric and by name, in patterns that made use of
463 duplicate group numbers, could behave incorrectly or give incorrect errors,
464 because when scanning forward to find the reference group, PCRE was not
465 taking into account the duplicate group numbers. A pattern such as
466 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
467
468 26. Changed a few more instances of "const unsigned char *" to USPTR, making
469 the feature of a custom pointer more persuasive (as requested by a user).
470
471 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
472 pcretest.c, inside #ifndefs, because it seems they are sometimes already
473 pre-defined.
474
475 28. Added support for (*UTF8) at the start of a pattern.
476
477 29. Arrange for flags added by the "release type" setting in CMake to be shown
478 in the configuration summary.
479
480
481 Version 7.8 05-Sep-08
482 ---------------------
483
484 1. Replaced UCP searching code with optimized version as implemented for Ad
485 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
486 stage table and inline lookup instead of a function, giving speed ups of 2
487 to 5 times on some simple patterns that I tested. Permission was given to
488 distribute the MultiStage2.py script that generates the tables (it's not in
489 the tarball, but is in the Subversion repository).
490
491 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
492 scripts.
493
494 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
495 a group with a zero qualifier. The result of the study could be incorrect,
496 or the function might crash, depending on the pattern.
497
498 4. Caseless matching was not working for non-ASCII characters in back
499 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
500 It now works when Unicode Property Support is available.
501
502 5. In pcretest, an escape such as \x{de} in the data was always generating
503 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
504 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
505 truncation.
506
507 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
508
509 7. Added two (int) casts to pcregrep when printing the difference of two
510 pointers, in case they are 64-bit values.
511
512 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
513 test 2 if it fails.
514
515 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
516 and a #define of that name to empty if it is not externally set. This is to
517 allow users of MSVC to set it if necessary.
518
519 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
520 the convenience functions in the pcre_get.c source file.
521
522 11. An option change at the start of a pattern that had top-level alternatives
523 could cause overwriting and/or a crash. This command provoked a crash in
524 some environments:
525
526 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
527
528 This potential security problem was recorded as CVE-2008-2371.
529
530 12. For a pattern where the match had to start at the beginning or immediately
531 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
532 pcre_dfa_exec() could read past the end of the passed subject if there was
533 no match. To help with detecting such bugs (e.g. with valgrind), I modified
534 pcretest so that it places the subject at the end of its malloc-ed buffer.
535
536 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
537 exec() might read past the end of the data buffer in UTF-8 mode.
538
539 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
540 the data contained the byte 0x85 as part of a UTF-8 character within its
541 first line. This applied both to normal and DFA matching.
542
543 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
544 /^[^d]*?$/8 failed to match "abc".
545
546 16. Added a missing copyright notice to pcrecpp_internal.h.
547
548 17. Make it more clear in the documentation that values returned from
549 pcre_exec() in ovector are byte offsets, not character counts.
550
551 18. Tidied a few places to stop certain compilers from issuing warnings.
552
553 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
554 supplied by Stefan Weber. I made a further small update for 7.8 because
555 there is a change of source arrangements: the pcre_searchfuncs.c module is
556 replaced by pcre_ucd.c.
557
558
559 Version 7.7 07-May-08
560 ---------------------
561
562 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
563 a string to a long long, pretend we don't even have a long long." This is
564 done by checking for the strtoq, strtoll, and _strtoi64 functions.
565
566 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
567 pre-7.6 versions, which defined a global no_arg variable instead of putting
568 it in the RE class. (See also #8 below.)
569
570 3. Remove a line of dead code, identified by coverity and reported by Nuno
571 Lopes.
572
573 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
574
575 (1) The include/exclude patterns were being applied to the whole pathnames
576 of files, instead of just to the final components.
577
578 (2) If there was more than one level of directory, the subdirectories were
579 skipped unless they satisfied the include/exclude conditions. This is
580 inconsistent with GNU grep (and could even be seen as contrary to the
581 pcregrep specification - which I improved to make it absolutely clear).
582 The action now is always to scan all levels of directory, and just
583 apply the include/exclude patterns to regular files.
584
585 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
586 --exclude_dir in the tests to avoid scanning .svn directories.
587
588 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
589 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
590 doesn't support NULs in patterns.
591
592 7. Added some missing "const"s to declarations of static tables in
593 pcre_compile.c and pcre_dfa_exec.c.
594
595 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
596 caused by fix #2 above. (Subsequently also a second patch to fix the
597 first patch. And a third patch - this was a messy problem.)
598
599 9. Applied Craig's patch to remove the use of push_back().
600
601 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
602 matching function regexec().
603
604 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
605 which, however, unlike Perl's \g{...}, are subroutine calls, not back
606 references. PCRE supports relative numbers with this syntax (I don't think
607 Oniguruma does).
608
609 12. Previously, a group with a zero repeat such as (...){0} was completely
610 omitted from the compiled regex. However, this means that if the group
611 was called as a subroutine from elsewhere in the pattern, things went wrong
612 (an internal error was given). Such groups are now left in the compiled
613 pattern, with a new opcode that causes them to be skipped at execution
614 time.
615
616 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
617 to the way PCRE behaves:
618
619 (a) A lone ] character is dis-allowed (Perl treats it as data).
620
621 (b) A back reference to an unmatched subpattern matches an empty string
622 (Perl fails the current match path).
623
624 (c) A data ] in a character class must be notated as \] because if the
625 first data character in a class is ], it defines an empty class. (In
626 Perl it is not possible to have an empty class.) The empty class []
627 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
628 The negative empty class [^] matches any one character, independently
629 of the DOTALL setting.
630
631 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
632 non-existent subpattern following a character class starting with ']' and
633 containing () gave an internal compiling error instead of "reference to
634 non-existent subpattern". Fortunately, when the pattern did exist, the
635 compiled code was correct. (When scanning forwards to check for the
636 existencd of the subpattern, it was treating the data ']' as terminating
637 the class, so got the count wrong. When actually compiling, the reference
638 was subsequently set up correctly.)
639
640 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
641 it was being rejected as not supported by pcre_dfa_exec(), even though
642 other assertions are supported. I have made pcre_dfa_exec() support
643 (*FAIL).
644
645 16. The implementation of 13c above involved the invention of a new opcode,
646 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
647 cannot be changed at match time, I realized I could make a small
648 improvement to matching performance by compiling OP_ALLANY instead of
649 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
650 on the OP_ANY path.
651
652 17. Compiling pcretest on Windows with readline support failed without the
653 following two fixes: (1) Make the unistd.h include conditional on
654 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
655
656 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
657 ncurses library to be included for pcretest when ReadLine support is
658 requested, but also to allow for it to be overridden. This patch came from
659 Daniel Bergström.
660
661 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
662 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
663 any errors with the current Unicode tables. Thanks to Peter Kankowski for
664 spotting this.
665
666
667 Version 7.6 28-Jan-08
668 ---------------------
669
670 1. A character class containing a very large number of characters with
671 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
672 overflow.
673
674 2. Patch to cut out the "long long" test in pcrecpp_unittest when
675 HAVE_LONG_LONG is not defined.
676
677 3. Applied Christian Ehrlicher's patch to update the CMake build files to
678 bring them up to date and include new features. This patch includes:
679
680 - Fixed PH's badly added libz and libbz2 support.
681 - Fixed a problem with static linking.
682 - Added pcredemo. [But later removed - see 7 below.]
683 - Fixed dftables problem and added an option.
684 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
685 HAVE_LONG_LONG.
686 - Added readline support for pcretest.
687 - Added an listing of the option settings after cmake has run.
688
689 4. A user submitted a patch to Makefile that makes it easy to create
690 "pcre.dll" under mingw when using Configure/Make. I added stuff to
691 Makefile.am that cause it to include this special target, without
692 affecting anything else. Note that the same mingw target plus all
693 the other distribution libraries and programs are now supported
694 when configuring with CMake (see 6 below) instead of with
695 Configure/Make.
696
697 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
698 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
699 exported in the Windows port". It has not yet been confirmed that the patch
700 solves the problem, but it does no harm.
701
702 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
703 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
704 with CMake, and also correct the comment about stack recursion.
705
706 7. Remove the automatic building of pcredemo from the ./configure system and
707 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
708 of a program that users should build themselves after PCRE is installed, so
709 building it automatically is not really right. What is more, it gave
710 trouble in some build environments.
711
712 8. Further tidies to CMakeLists.txt from Sheri and Christian.
713
714
715 Version 7.5 10-Jan-08
716 ---------------------
717
718 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
719 values in parens when parsing an RE using the C++ wrapper."
720
721 2. Negative specials like \S did not work in character classes in UTF-8 mode.
722 Characters greater than 255 were excluded from the class instead of being
723 included.
724
725 3. The same bug as (2) above applied to negated POSIX classes such as
726 [:^space:].
727
728 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
729 defined or documented. It seems to have been a typo for PCRE_STATIC, so
730 I have changed it.
731
732 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
733 first named subpattern) and a construct such as (?&a) would reference the
734 first named subpattern whose name started with "a" (in other words, the
735 length check was missing). Both these problems are fixed. "Subpattern name
736 expected" is now given for (?&) (a zero-length name), and this patch also
737 makes it give the same error for \k'' (previously it complained that that
738 was a reference to a non-existent subpattern).
739
740 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
741 this is right because (?- can be followed by option settings as well as by
742 digits. I have, however, made the messages clearer.
743
744 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
745 than the number used in the conditional) now cause a compile-time error.
746 This is actually not compatible with Perl, which accepts such patterns, but
747 treats the conditional as always being FALSE (as PCRE used to), but it
748 seems to me that giving a diagnostic is better.
749
750 8. Change "alphameric" to the more common word "alphanumeric" in comments
751 and messages.
752
753 9. Fix two occurrences of "backslash" in comments that should have been
754 "backspace".
755
756 10. Remove two redundant lines of code that can never be obeyed (their function
757 was moved elsewhere).
758
759 11. The program that makes PCRE's Unicode character property table had a bug
760 which caused it to generate incorrect table entries for sequences of
761 characters that have the same character type, but are in different scripts.
762 It amalgamated them into a single range, with the script of the first of
763 them. In other words, some characters were in the wrong script. There were
764 thirteen such cases, affecting characters in the following ranges:
765
766 U+002b0 - U+002c1
767 U+0060c - U+0060d
768 U+0061e - U+00612
769 U+0064b - U+0065e
770 U+0074d - U+0076d
771 U+01800 - U+01805
772 U+01d00 - U+01d77
773 U+01d9b - U+01dbf
774 U+0200b - U+0200f
775 U+030fc - U+030fe
776 U+03260 - U+0327f
777 U+0fb46 - U+0fbb1
778 U+10450 - U+1049d
779
780 12. The -o option (show only the matching part of a line) for pcregrep was not
781 compatible with GNU grep in that, if there was more than one match in a
782 line, it showed only the first of them. It now behaves in the same way as
783 GNU grep.
784
785 13. If the -o and -v options were combined for pcregrep, it printed a blank
786 line for every non-matching line. GNU grep prints nothing, and pcregrep now
787 does the same. The return code can be used to tell if there were any
788 non-matching lines.
789
790 14. Added --file-offsets and --line-offsets to pcregrep.
791
792 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
793 infinitely looping recursion. The bug was that positive lookaheads were not
794 being skipped when checking for a possible empty match (negative lookaheads
795 and both kinds of lookbehind were skipped).
796
797 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
798 inclusion of <windows.h> to before rather than after the definition of
799 INVALID_FILE_ATTRIBUTES (patch from David Byron).
800
801 17. Specifying a possessive quantifier with a specific limit for a Unicode
802 character property caused pcre_compile() to compile bad code, which led at
803 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
804 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
805 caused the error; without that there was no problem.
806
807 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
808
809 19. Added --enable-pcretest-libreadline.
810
811 20. In pcrecpp.cc, the variable 'count' was incremented twice in
812 RE::GlobalReplace(). As a result, the number of replacements returned was
813 double what it should be. I removed one of the increments, but Craig sent a
814 later patch that removed the other one (the right fix) and added unit tests
815 that check the return values (which was not done before).
816
817 21. Several CMake things:
818
819 (1) Arranged that, when cmake is used on Unix, the libraries end up with
820 the names libpcre and libpcreposix, not just pcre and pcreposix.
821
822 (2) The above change means that pcretest and pcregrep are now correctly
823 linked with the newly-built libraries, not previously installed ones.
824
825 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
826
827 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
828 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
829 UTF-8 newline character). The key issue is that the pattern starts .*;
830 this means that the match must be either at the beginning, or after a
831 newline. The bug was in the code for advancing after a failed match and
832 checking that the new position followed a newline. It was not taking
833 account of UTF-8 characters correctly.
834
835 23. PCRE was behaving differently from Perl in the way it recognized POSIX
836 character classes. PCRE was not treating the sequence [:...:] as a
837 character class unless the ... were all letters. Perl, however, seems to
838 allow any characters between [: and :], though of course it rejects as
839 unknown any "names" that contain non-letters, because all the known class
840 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
841 for example, whereas PCRE did not - it did not recognize a POSIX character
842 class. This seemed a bit dangerous, so the code has been changed to be
843 closer to Perl. The behaviour is not identical to Perl, because PCRE will
844 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
845 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
846 Perl does, and where it didn't before.
847
848 24. Rewrite so as to remove the single use of %n from pcregrep because in some
849 Windows environments %n is disabled by default.
850
851
852 Version 7.4 21-Sep-07
853 ---------------------
854
855 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
856 means that a class such as [\s] counted as "explicit reference to CR or
857 LF". That isn't really right - the whole point of the change was to try to
858 help when there was an actual mention of one of the two characters. So now
859 the change happens only if \r or \n (or a literal CR or LF) character is
860 encountered.
861
862 2. The 32-bit options word was also used for 6 internal flags, but the numbers
863 of both had grown to the point where there were only 3 bits left.
864 Fortunately, there was spare space in the data structure, and so I have
865 moved the internal flags into a new 16-bit field to free up more option
866 bits.
867
868 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
869 but did not set the internal JCHANGED flag - either of these is enough to
870 control the way the "get" function works - but the PCRE_INFO_JCHANGED
871 facility is supposed to tell if (?J) was ever used, so now (?J) at the
872 start sets both bits.
873
874 4. Added options (at build time, compile time, exec time) to change \R from
875 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
876
877 5. doc/pcresyntax.html was missing from the distribution.
878
879 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
880 compatibility, even though it is no longer used.
881
882 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
883 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
884 windows.h file is present (where different names are used). [This was
885 reversed later after testing - see 16 below.]
886
887 8. Changed all #include <config.h> to #include "config.h". There were also
888 some further <pcre.h> cases that I changed to "pcre.h".
889
890 9. When pcregrep was used with the --colour option, it missed the line ending
891 sequence off the lines that it output.
892
893 10. It was pointed out to me that arrays of string pointers cause lots of
894 relocations when a shared library is dynamically loaded. A technique of
895 using a single long string with a table of offsets can drastically reduce
896 these. I have refactored PCRE in four places to do this. The result is
897 dramatic:
898
899 Originally: 290
900 After changing UCP table: 187
901 After changing error message table: 43
902 After changing table of "verbs" 36
903 After changing table of Posix names 22
904
905 Thanks to the folks working on Gregex for glib for this insight.
906
907 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
908 unicode-properties was also set.
909
910 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
911
912 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
913 checked only for CRLF.
914
915 14. Added casts to pcretest.c to avoid compiler warnings.
916
917 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
918
919 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
920 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
921 entirely. This removes changes made in 7 above.
922
923 17. The CMake files have been updated, and there is now more information about
924 building with CMake in the NON-UNIX-USE document.
925
926
927 Version 7.3 28-Aug-07
928 ---------------------
929
930 1. In the rejigging of the build system that eventually resulted in 7.1, the
931 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
932 brackets there is not right, since it causes compilers to look for an
933 installed pcre.h, not the version that is in the source that is being
934 compiled (which of course may be different). I have changed it back to:
935
936 #include "pcre.h"
937
938 I have a vague recollection that the change was concerned with compiling in
939 different directories, but in the new build system, that is taken care of
940 by the VPATH setting the Makefile.
941
942 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
943 when the subject happened to end in the byte 0x85 (e.g. if the last
944 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
945 characters but of course it shouldn't be taken as a newline when it is part
946 of another character. The bug was that, for an unlimited repeat of . in
947 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
948 characters when looking for a newline.
949
950 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
951
952 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
953 in debug output.
954
955 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
956 long printing in the pcrecpp unittest when running under MinGW.
957
958 6. ESC_K was left out of the EBCDIC table.
959
960 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
961 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
962 limit also applies to "virtual nesting" when a pattern is recursive, and in
963 this case 1000 isn't so big. I have been able to remove this limit at the
964 expense of backing off one optimization in certain circumstances. Normally,
965 when pcre_exec() would call its internal match() function recursively and
966 immediately return the result unconditionally, it uses a "tail recursion"
967 feature to save stack. However, when a subpattern that can match an empty
968 string has an unlimited repetition quantifier, it no longer makes this
969 optimization. That gives it a stack frame in which to save the data for
970 checking that an empty string has been matched. Previously this was taken
971 from the 1000-entry workspace that had been reserved. So now there is no
972 explicit limit, but more stack is used.
973
974 8. Applied Daniel's patches to solve problems with the import/export magic
975 syntax that is required for Windows, and which was going wrong for the
976 pcreposix and pcrecpp parts of the library. These were overlooked when this
977 problem was solved for the main library.
978
979 9. There were some crude static tests to avoid integer overflow when computing
980 the size of patterns that contain repeated groups with explicit upper
981 limits. As the maximum quantifier is 65535, the maximum group length was
982 set at 30,000 so that the product of these two numbers did not overflow a
983 32-bit integer. However, it turns out that people want to use groups that
984 are longer than 30,000 bytes (though not repeat them that many times).
985 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
986 made it possible to implement the integer overflow checks in a much more
987 dynamic way, which I have now done. The artificial limitation on group
988 length has been removed - we now have only the limit on the total length of
989 the compiled pattern, which depends on the LINK_SIZE setting.
990
991 10. Fixed a bug in the documentation for get/copy named substring when
992 duplicate names are permitted. If none of the named substrings are set, the
993 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
994 empty string.
995
996 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
997 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
998 because the ] is interpreted as the first data character and the
999 terminating ] is not found. PCRE has been made compatible with Perl in this
1000 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1001 cause memory overwriting.
1002
1003 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1004 string has been matched (to stop an infinite loop). It was not recognizing
1005 a conditional subpattern that could match an empty string if that
1006 subpattern was within another subpattern. For example, it looped when
1007 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1008 condition was not nested. This bug has been fixed.
1009
1010 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1011 past the start of the subject in the presence of bytes with the top bit
1012 set, for example "\x8aBCD".
1013
1014 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1015 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1016
1017 14. Optimized (?!) to (*FAIL).
1018
1019 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1020 This restricts code points to be within the range 0 to 0x10FFFF, excluding
1021 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1022 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1023 does: it's just the validity check that is more restrictive.
1024
1025 16. Inserted checks for integer overflows during escape sequence (backslash)
1026 processing, and also fixed erroneous offset values for syntax errors during
1027 backslash processing.
1028
1029 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1030 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1031
1032 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1033 caused an overrun.
1034
1035 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1036 something other than just ASCII characters) inside a group that had an
1037 unlimited repeat caused a loop at compile time (while checking to see
1038 whether the group could match an empty string).
1039
1040 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1041 [\P{Any}] did so. (Error in the code for printing property names.)
1042
1043 21. An orphan \E inside a character class could cause a crash.
1044
1045 22. A repeated capturing bracket such as (A)? could cause a wild memory
1046 reference during compilation.
1047
1048 23. There are several functions in pcre_compile() that scan along a compiled
1049 expression for various reasons (e.g. to see if it's fixed length for look
1050 behind). There were bugs in these functions when a repeated \p or \P was
1051 present in the pattern. These operators have additional parameters compared
1052 with \d, etc, and these were not being taken into account when moving along
1053 the compiled data. Specifically:
1054
1055 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1056 length.
1057
1058 (b) An item such as \pL+ within a repeated group could cause crashes or
1059 loops.
1060
1061 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1062 "reference to non-existent subpattern" error.
1063
1064 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1065
1066 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1067 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1068
1069 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1070
1071 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1072 character were causing crashes (broken optimization).
1073
1074 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1075 \p or \P) caused a compile-time loop.
1076
1077 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1078 break. For example, the unstudied pattern [\r\n]A does not match the string
1079 "\r\nA" because change 7.0/46 below moves the current point on by two
1080 characters after failing to match at the start. However, the pattern \nA
1081 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1082 the same is true. There doesn't seem any very clean way out of this, but
1083 what I have chosen to do makes the common cases work: PCRE now takes note
1084 of whether there can be an explicit match for \r or \n anywhere in the
1085 pattern, and if so, 7.0/46 no longer applies. As part of this change,
1086 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1087 pattern has explicit CR or LF references.
1088
1089 29. Added (*CR) etc for changing newline setting at start of pattern.
1090
1091
1092 Version 7.2 19-Jun-07
1093 ---------------------
1094
1095 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1096 which is apparently normally available under Windows.
1097
1098 2. Re-jig the pcregrep tests with different newline settings in an attempt
1099 to make them independent of the local environment's newline setting.
1100
1101 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1102
1103 4. Some of the "internals" tests were previously cut out when the link size
1104 was not 2, because the output contained actual offsets. The recent new
1105 "Z" feature of pcretest means that these can be cut out, making the tests
1106 usable with all link sizes.
1107
1108 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1109 stack recursion. This gives a massive performance boost under BSD, but just
1110 a small improvement under Linux. However, it saves one field in the frame
1111 in all cases.
1112
1113 6. Added more features from the forthcoming Perl 5.10:
1114
1115 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1116 recursion call. It refers to the nth most recently opened parentheses.
1117
1118 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1119 to be opened parentheses.
1120
1121 (c) Conditions that refer to capturing parentheses can be specified
1122 relatively, for example, (?(-2)... or (?(+3)...
1123
1124 (d) \K resets the start of the current match so that everything before
1125 is not part of it.
1126
1127 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1128
1129 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1130 reference syntax.
1131
1132 (g) (?| introduces a group in which the numbering of parentheses in each
1133 alternative starts with the same number.
1134
1135 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1136
1137 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1138 PCRE_INFO_JCHANGED.
1139
1140 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1141 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1142 for detecting groups that can match an empty string.
1143
1144 9. A pattern with a very large number of alternatives (more than several
1145 hundred) was running out of internal workspace during the pre-compile
1146 phase, where pcre_compile() figures out how much memory will be needed. A
1147 bit of new cunning has reduced the workspace needed for groups with
1148 alternatives. The 1000-alternative test pattern now uses 12 bytes of
1149 workspace instead of running out of the 4096 that are available.
1150
1151 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1152
1153 11. Applied patch from Google to remove an optimization that didn't quite work.
1154 The report of the bug said:
1155
1156 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1157 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1158 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1159
1160 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1161 it matched the wrong number of bytes.
1162
1163
1164 Version 7.1 24-Apr-07
1165 ---------------------
1166
1167 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1168 that is more "standard", making use of automake and other Autotools. There
1169 is some re-arrangement of the files and adjustment of comments consequent
1170 on this.
1171
1172 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1173 for recursive directory scanning broke on some systems because the files
1174 are not scanned in any specific order and on different systems the order
1175 was different. A call to "sort" has been inserted into RunGrepTest for the
1176 approprate test as a short-term fix. In the longer term there may be an
1177 alternative.
1178
1179 3. I had an email from Eric Raymond about problems translating some of PCRE's
1180 man pages to HTML (despite the fact that I distribute HTML pages, some
1181 people do their own conversions for various reasons). The problems
1182 concerned the use of low-level troff macros .br and .in. I have therefore
1183 removed all such uses from the man pages (some were redundant, some could
1184 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1185 HTML has been updated to handle .nf/.fi and to complain if it encounters
1186 .br or .in.
1187
1188 4. Updated comments in configure.ac that get placed in config.h.in and also
1189 arranged for config.h to be included in the distribution, with the name
1190 config.h.generic, for the benefit of those who have to compile without
1191 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1192
1193 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1194 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1195 makevp.bat for the current PCRE, using the additional files
1196 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1197
1198 6. A Windows user reported a minor discrepancy with test 2, which turned out
1199 to be caused by a trailing space on an input line that had got lost in his
1200 copy. The trailing space was an accident, so I've just removed it.
1201
1202 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1203 that is needed.
1204
1205 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1206 as "const" (a) because they are and (b) because it helps the PHP
1207 maintainers who have recently made a script to detect big data structures
1208 in the php code that should be moved to the .rodata section. I remembered
1209 to update Builducptable as well, so it won't revert if ucptable.h is ever
1210 re-created.
1211
1212 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1213 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1214 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1215 support is not required. This saves 1.5-2K of code, which is important in
1216 some applications.
1217
1218 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1219 so as not to refer to the tables, even though these functions will never be
1220 called when UTF-8 support is disabled. Otherwise there are problems with a
1221 shared library.
1222
1223 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1224
1225 (a) It was defining its arguments as char * instead of void *.
1226
1227 (b) It was assuming that all moves were upwards in memory; this was true
1228 a long time ago when I wrote it, but is no longer the case.
1229
1230 The emulated memove() is provided for those environments that have neither
1231 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1232 is clearly not the case, as these two bugs were recently reported.
1233
1234 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1235 and Detrail to create the HTML documentation, the .txt form of the man
1236 pages, and it removes trailing spaces from listed files. It also creates
1237 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1238 case, it wraps all the #defines with #ifndefs. This script should be run
1239 before "make dist".
1240
1241 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1242 with Unicode property support.
1243
1244 (a) For a maximizing quantifier, if the two different cases of the
1245 character were of different lengths in their UTF-8 codings (there are
1246 some cases like this - I found 11), and the matching function had to
1247 back up over a mixture of the two cases, it incorrectly assumed they
1248 were both the same length.
1249
1250 (b) When PCRE was configured to use the heap rather than the stack for
1251 recursion during matching, it was not correctly preserving the data for
1252 the other case of a UTF-8 character when checking ahead for a match
1253 while processing a minimizing repeat. If the check also involved
1254 matching a wide character, but failed, corruption could cause an
1255 erroneous result when trying to check for a repeat of the original
1256 character.
1257
1258 13. Some tidying changes to the testing mechanism:
1259
1260 (a) The RunTest script now detects the internal link size and whether there
1261 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1262 values substituted by "configure". (The RunGrepTest script already did
1263 this for UTF-8.) The configure.ac script no longer substitutes the
1264 relevant variables.
1265
1266 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1267 with length and offset values. This means that the output is different
1268 for different internal link sizes. Test 2 is skipped for link sizes
1269 other than 2 because of this, bypassing the problem. Unfortunately,
1270 there was also a test in test 3 (the locale tests) that used /B and
1271 failed for link sizes other than 2. Rather than cut the whole test out,
1272 I have added a new /Z option to pcretest that replaces the length and
1273 offset values with spaces. This is now used to make test 3 independent
1274 of link size. (Test 2 will be tidied up later.)
1275
1276 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1277 segmentation fault instead of returning the appropriate error message.
1278
1279 15. In multiline mode when the newline sequence was set to "any", the pattern
1280 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1281 This doesn't seem right; it now treats the CRLF combination as the line
1282 ending, and so does not match in that case. It's only a pattern such as ^$
1283 that would hit this one: something like ^ABC$ would have failed after \r
1284 and then tried again after \r\n.
1285
1286 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1287 in an attempt to make files that differ only in their line terminators
1288 compare equal. This works on Linux.
1289
1290 17. Under certain error circumstances pcregrep might try to free random memory
1291 as it exited. This is now fixed, thanks to valgrind.
1292
1293 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1294 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1295 was because its rules for how to advance for /g after matching an empty
1296 string at the end of a line did not allow for this case. They now check for
1297 it specially.
1298
1299 20. pcretest is supposed to handle patterns and data of any length, by
1300 extending its buffers when necessary. It was getting this wrong when the
1301 buffer for a data line had to be extended.
1302
1303 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1304 CRLF as a newline sequence.
1305
1306 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1307 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1308 I have nevertheless tidied it up.
1309
1310 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1311
1312 24. Added a man page for pcre-config.
1313
1314
1315 Version 7.0 19-Dec-06
1316 ---------------------
1317
1318 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1319 moving to gcc 4.1.1.
1320
1321 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1322 sys/time.h, which is documented as needed for this function. It doesn't
1323 seem to matter on Linux, but it showed up on some releases of OS X.
1324
1325 3. It seems that there are systems where bytes whose values are greater than
1326 127 match isprint() in the "C" locale. The "C" locale should be the
1327 default when a C program starts up. In most systems, only ASCII printing
1328 characters match isprint(). This difference caused the output from pcretest
1329 to vary, making some of the tests fail. I have changed pcretest so that:
1330
1331 (a) When it is outputting text in the compiled version of a pattern, bytes
1332 other than 32-126 are always shown as hex escapes.
1333
1334 (b) When it is outputting text that is a matched part of a subject string,
1335 it does the same, unless a different locale has been set for the match
1336 (using the /L modifier). In this case, it uses isprint() to decide.
1337
1338 4. Fixed a major bug that caused incorrect computation of the amount of memory
1339 required for a compiled pattern when options that changed within the
1340 pattern affected the logic of the preliminary scan that determines the
1341 length. The relevant options are -x, and -i in UTF-8 mode. The result was
1342 that the computed length was too small. The symptoms of this bug were
1343 either the PCRE error "internal error: code overflow" from pcre_compile(),
1344 or a glibc crash with a message such as "pcretest: free(): invalid next
1345 size (fast)". Examples of patterns that provoked this bug (shown in
1346 pcretest format) are:
1347
1348 /(?-x: )/x
1349 /(?x)(?-x: \s*#\s*)/
1350 /((?i)[\x{c0}])/8
1351 /(?i:[\x{c0}])/8
1352
1353 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1354 is now done differently.
1355
1356 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1357 wrapper classes; (b) implement a new function in the C++ scanner that is
1358 more efficient than the old way of doing things because it avoids levels of
1359 recursion in the regex matching; (c) add a paragraph to the documentation
1360 for the FullMatch() function.
1361
1362 6. The escape sequence \n was being treated as whatever was defined as
1363 "newline". Not only was this contrary to the documentation, which states
1364 that \n is character 10 (hex 0A), but it also went horribly wrong when
1365 "newline" was defined as CRLF. This has been fixed.
1366
1367 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1368 was being set to -1 for the "end of line" case (supposedly a value that no
1369 character can have). Though this value is never used (the check for end of
1370 line is "zero bytes in current character"), it caused compiler complaints.
1371 I've changed it to 0xffffffff.
1372
1373 8. In pcre_version.c, the version string was being built by a sequence of
1374 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1375 string (as it is for production releases) called a macro with an empty
1376 argument. The C standard says the result of this is undefined. The gcc
1377 compiler treats it as an empty string (which was what was wanted) but it is
1378 reported that Visual C gives an error. The source has been hacked around to
1379 avoid this problem.
1380
1381 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1382 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1383 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1384 of them did).
1385
1386 10. Originally, pcretest opened its input and output without "b"; then I was
1387 told that "b" was needed in some environments, so it was added for release
1388 5.0 to both the input and output. (It makes no difference on Unix-like
1389 systems.) Later I was told that it is wrong for the input on Windows. I've
1390 now abstracted the modes into two macros, to make it easier to fiddle with
1391 them, and removed "b" from the input mode under Windows.
1392
1393 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1394
1395 12. Added -help and --help to pcretest as an official way of being reminded
1396 of the options.
1397
1398 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1399 and pcrecpp.cc because they annoy compilers at high warning levels.
1400
1401 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1402
1403 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1404 scripts are not C programs :-) and which was not noticed because it works
1405 on Linux.
1406
1407 16. pcretest is supposed to handle any length of pattern and data line (as one
1408 line or as a continued sequence of lines) by extending its input buffer if
1409 necessary. This feature was broken for very long pattern lines, leading to
1410 a string of junk being passed to pcre_compile() if the pattern was longer
1411 than about 50K.
1412
1413 17. I have done a major re-factoring of the way pcre_compile() computes the
1414 amount of memory needed for a compiled pattern. Previously, there was code
1415 that made a preliminary scan of the pattern in order to do this. That was
1416 OK when PCRE was new, but as the facilities have expanded, it has become
1417 harder and harder to keep it in step with the real compile phase, and there
1418 have been a number of bugs (see for example, 4 above). I have now found a
1419 cunning way of running the real compile function in a "fake" mode that
1420 enables it to compute how much memory it would need, while actually only
1421 ever using a few hundred bytes of working memory and without too many
1422 tests of the mode. This should make future maintenance and development
1423 easier. A side effect of this work is that the limit of 200 on the nesting
1424 depth of parentheses has been removed (though this was never a serious
1425 limitation, I suspect). However, there is a downside: pcre_compile() now
1426 runs more slowly than before (30% or more, depending on the pattern). I
1427 hope this isn't a big issue. There is no effect on runtime performance.
1428
1429 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1430 newline (only possible for the last line of a file) and it was a
1431 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1432
1433 19. Added additional timing features to pcretest. (1) The -tm option now times
1434 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1435 separate command line item, by a number that specifies the number of
1436 repeats to use when timing. The default is 50000; this gives better
1437 precision, but takes uncomfortably long for very large patterns.
1438
1439 20. Extended pcre_study() to be more clever in cases where a branch of a
1440 subpattern has no definite first character. For example, (a*|b*)[cd] would
1441 previously give no result from pcre_study(). Now it recognizes that the
1442 first character must be a, b, c, or d.
1443
1444 21. There was an incorrect error "recursive call could loop indefinitely" if
1445 a subpattern (or the entire pattern) that was being tested for matching an
1446 empty string contained only one non-empty item after a nested subpattern.
1447 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1448 incorrectly, because the \d was being skipped in the check.
1449
1450 22. The pcretest program now has a new pattern option /B and a command line
1451 option -b, which is equivalent to adding /B to every pattern. This causes
1452 it to show the compiled bytecode, without the additional information that
1453 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1454 is the same as /B/I).
1455
1456 23. A new optimization is now able automatically to treat some sequences such
1457 as a*b as a*+b. More specifically, if something simple (such as a character
1458 or a simple class like \d) has an unlimited quantifier, and is followed by
1459 something that cannot possibly match the quantified thing, the quantifier
1460 is automatically "possessified".
1461
1462 24. A recursive reference to a subpattern whose number was greater than 39
1463 went wrong under certain circumstances in UTF-8 mode. This bug could also
1464 have affected the operation of pcre_study().
1465
1466 25. Realized that a little bit of performance could be had by replacing
1467 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1468
1469 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1470
1471 27. Possessive quantifiers such as a++ were previously implemented by turning
1472 them into atomic groups such as ($>a+). Now they have their own opcodes,
1473 which improves performance. This includes the automatically created ones
1474 from 23 above.
1475
1476 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1477 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1478 the first matched character to be a colon. This applied both to named and
1479 numbered groups.
1480
1481 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1482
1483 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1484 building PCRE on Windows easier, so I have included it in the distribution.
1485
1486 31. There is now a check in pcretest against a ridiculously large number being
1487 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1488 loop, the loop is abandoned.
1489
1490 32. Forward references to subpatterns in conditions such as (?(2)...) where
1491 subpattern 2 is defined later cause pcre_compile() to search forwards in
1492 the pattern for the relevant set of parentheses. This search went wrong
1493 when there were unescaped parentheses in a character class, parentheses
1494 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1495
1496 33. "Subroutine" calls and backreferences were previously restricted to
1497 referencing subpatterns earlier in the regex. This restriction has now
1498 been removed.
1499
1500 34. Added a number of extra features that are going to be in Perl 5.10. On the
1501 whole, these are just syntactic alternatives for features that PCRE had
1502 previously implemented using the Python syntax or my own invention. The
1503 other formats are all retained for compatibility.
1504
1505 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1506 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1507 also .NET compatible.
1508
1509 (b) A recursion or subroutine call to a named group can now be defined as
1510 (?&name) as well as (?P>name).
1511
1512 (c) A backreference to a named group can now be defined as \k<name> or
1513 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1514 5.10, are also .NET compatible.
1515
1516 (d) A conditional reference to a named group can now use the syntax
1517 (?(<name>) or (?('name') as well as (?(name).
1518
1519 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1520 groups (named and numbered) that are never evaluated inline, but can be
1521 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1522 is always false. There may be only one alternative in such a group.
1523
1524 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1525 as the simple (?(R). The condition is true only if the most recent
1526 recursion is that of the given number or name. It does not search out
1527 through the entire recursion stack.
1528
1529 (g) The escape \gN or \g{N} has been added, where N is a positive or
1530 negative number, specifying an absolute or relative reference.
1531
1532 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1533 some "unreachable code" warnings.
1534
1535 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1536 things, this adds five new scripts.
1537
1538 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1539 There were also incompatibilities regarding the handling of \Q..\E inside
1540 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1541 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1542
1543 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1544 matches an empty string, and forcibly breaks the loop. There were bugs in
1545 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1546 against aaaa the result was just "a" rather than "aaaa", for example. Two
1547 separate and independent bugs (that affected different cases) have been
1548 fixed.
1549
1550 39. Refactored the code to abolish the use of different opcodes for small
1551 capturing bracket numbers. This is a tidy that I avoided doing when I
1552 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1553 The new approach is not only tidier, it makes it possible to reduce the
1554 memory needed to fix the previous bug (38).
1555
1556 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1557 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1558 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1559 mode.
1560
1561 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1562 report.
1563
1564 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1565 copy construction and assignment in the C++ wrapper.
1566
1567 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1568 couple of bugs that could have given wrong results in the "--newline=crlf"
1569 case.
1570
1571 44. Added a number of casts and did some reorganization of signed/unsigned int
1572 variables following suggestions from Dair Grant. Also renamed the variable
1573 "this" as "item" because it is a C++ keyword.
1574
1575 45. Arranged for dftables to add
1576
1577 #include "pcre_internal.h"
1578
1579 to pcre_chartables.c because without it, gcc 4.x may remove the array
1580 definition from the final binary if PCRE is built into a static library and
1581 dead code stripping is activated.
1582
1583 46. For an unanchored pattern, if a match attempt fails at the start of a
1584 newline sequence, and the newline setting is CRLF or ANY, and the next two
1585 characters are CRLF, advance by two characters instead of one.
1586
1587
1588 Version 6.7 04-Jul-06
1589 ---------------------
1590
1591 1. In order to handle tests when input lines are enormously long, pcretest has
1592 been re-factored so that it automatically extends its buffers when
1593 necessary. The code is crude, but this _is_ just a test program. The
1594 default size has been increased from 32K to 50K.
1595
1596 2. The code in pcre_study() was using the value of the re argument before
1597 testing it for NULL. (Of course, in any sensible call of the function, it
1598 won't be NULL.)
1599
1600 3. The memmove() emulation function in pcre_internal.h, which is used on
1601 systems that lack both memmove() and bcopy() - that is, hardly ever -
1602 was missing a "static" storage class specifier.
1603
1604 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1605 containing an extended class (one that cannot be represented by a bitmap
1606 because it contains high-valued characters or Unicode property items, e.g.
1607 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1608 pattern, but PCRE should not loop if you do not (it no longer does).
1609 [Detail: two cases were found: (a) a repeated subpattern containing an
1610 extended class; (b) a recursive reference to a subpattern that followed a
1611 previous extended class. It wasn't skipping over the extended class
1612 correctly when UTF-8 mode was not set.]
1613
1614 5. A negated single-character class was not being recognized as fixed-length
1615 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1616 compile error "lookbehind assertion is not fixed length".
1617
1618 6. The RunPerlTest auxiliary script was showing an unexpected difference
1619 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1620 write a Perl script that can interpret lines of an input file either as
1621 byte characters or as UTF-8, which is what "perltest" was being required to
1622 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1623 can't do is switch easily at run time between having the "use utf8;" pragma
1624 or not. In the end, I fudged it by using the RunPerlTest script to insert
1625 "use utf8;" explicitly for the UTF-8 tests.
1626
1627 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1628 the end of the subject string, contrary to the documentation and to what
1629 Perl does. This was true of both matching functions. Now it matches only at
1630 the start of the subject and immediately after *internal* newlines.
1631
1632 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1633 a pointer to an int instead of a pointer to an unsigned long int. This
1634 caused problems on 64-bit systems.
1635
1636 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1637 instance of the 'standard' template library not being so standard".
1638
1639 10. There was no check on the number of named subpatterns nor the maximum
1640 length of a subpattern name. The product of these values is used to compute
1641 the size of the memory block for a compiled pattern. By supplying a very
1642 long subpattern name and a large number of named subpatterns, the size
1643 computation could be caused to overflow. This is now prevented by limiting
1644 the length of names to 32 characters, and the number of named subpatterns
1645 to 10,000.
1646
1647 11. Subpatterns that are repeated with specific counts have to be replicated in
1648 the compiled pattern. The size of memory for this was computed from the
1649 length of the subpattern and the repeat count. The latter is limited to
1650 65535, but there was no limit on the former, meaning that integer overflow
1651 could in principle occur. The compiled length of a repeated subpattern is
1652 now limited to 30,000 bytes in order to prevent this.
1653
1654 12. Added the optional facility to have named substrings with the same name.
1655
1656 13. Added the ability to use a named substring as a condition, using the
1657 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1658 are numbers (not recommended). Forward references are permitted.
1659
1660 14. Added forward references in named backreferences (if you see what I mean).
1661
1662 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1663 pattern could run off the end of the subject. For example, the pattern
1664 "(?s)(.{1,5})"8 did this with the subject "ab".
1665
1666 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1667 PCRE_CASELESS was set when matching characters that were quantified with ?
1668 or *.
1669
1670 17. A character class other than a single negated character that had a minimum
1671 but no maximum quantifier - for example [ab]{6,} - was not handled
1672 correctly by pce_dfa_exec(). It would match only one character.
1673
1674 18. A valid (though odd) pattern that looked like a POSIX character
1675 class but used an invalid character after [ (for example [[,abc,]]) caused
1676 pcre_compile() to give the error "Failed: internal error: code overflow" or
1677 in some cases to crash with a glibc free() error. This could even happen if
1678 the pattern terminated after [[ but there just happened to be a sequence of
1679 letters, a binary zero, and a closing ] in the memory that followed.
1680
1681 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1682 over the years. Originally (before any Unicode support), just the bottom 8
1683 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1684 output from "man perlunicode" includes this:
1685
1686 The regular expression compiler produces polymorphic opcodes. That
1687 is, the pattern adapts to the data and automatically switches to
1688 the Unicode character scheme when presented with Unicode data--or
1689 instead uses a traditional byte scheme when presented with byte
1690 data.
1691
1692 Sadly, a wide octal escape does not cause a switch, and in a string with
1693 no other multibyte characters, these octal escapes are treated as before.
1694 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1695 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1696 Unicode string.
1697
1698 I have not perpetrated such confusion in PCRE. Up till now, it took just
1699 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1700 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1701 translate to the appropriate multibyte character.
1702
1703 29. Applied some refactoring to reduce the number of warnings from Microsoft
1704 and Borland compilers. This has included removing the fudge introduced
1705 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1706 a warning about an unused variable.
1707
1708 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1709 characters since release 4.0, because Perl (from release 5.004) does not.
1710 [Or at least, is documented not to: some releases seem to be in conflict
1711 with the documentation.] However, when a pattern was studied with
1712 pcre_study() and all its branches started with \s, PCRE still included VT
1713 as a possible starting character. Of course, this did no harm; it just
1714 caused an unnecessary match attempt.
1715
1716 22. Removed a now-redundant internal flag bit that recorded the fact that case
1717 dependency changed within the pattern. This was once needed for "required
1718 byte" processing, but is no longer used. This recovers a now-scarce options
1719 bit. Also moved the least significant internal flag bit to the most-
1720 significant bit of the word, which was not previously used (hangover from
1721 the days when it was an int rather than a uint) to free up another bit for
1722 the future.
1723
1724 23. Added support for CRLF line endings as well as CR and LF. As well as the
1725 default being selectable at build time, it can now be changed at runtime
1726 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1727 specify that it is scanning data with non-default line endings.
1728
1729 24. Changed the definition of CXXLINK to make it agree with the definition of
1730 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1731
1732 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1733 recursions. This makes a big different to stack usage for some patterns.
1734
1735 26. If a subpattern containing a named recursion or subroutine reference such
1736 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1737 the space required for the compiled pattern went wrong and gave too small a
1738 value. Depending on the environment, this could lead to "Failed: internal
1739 error: code overflow at offset 49" or "glibc detected double free or
1740 corruption" errors.
1741
1742 27. Applied patches from Google (a) to support the new newline modes and (b) to
1743 advance over multibyte UTF-8 characters in GlobalReplace.
1744
1745 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1746 difference for some implementation of PCRE in some Windows version.
1747
1748 29. Added some extra testing facilities to pcretest:
1749
1750 \q<number> in a data line sets the "match limit" value
1751 \Q<number> in a data line sets the "match recursion limt" value
1752 -S <number> sets the stack size, where <number> is in megabytes
1753
1754 The -S option isn't available for Windows.
1755
1756
1757 Version 6.6 06-Feb-06
1758 ---------------------
1759
1760 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1761 in pcreposix.h. I have copied the definition from pcre.h.
1762
1763 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1764 because pcre.h is no longer a built file.
1765
1766 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1767 not normally included in the compiled code.
1768
1769
1770 Version 6.5 01-Feb-06
1771 ---------------------
1772
1773 1. When using the partial match feature with pcre_dfa_exec(), it was not
1774 anchoring the second and subsequent partial matches at the new starting
1775 point. This could lead to incorrect results. For example, with the pattern
1776 /1234/, partially matching against "123" and then "a4" gave a match.
1777
1778 2. Changes to pcregrep:
1779
1780 (a) All non-match returns from pcre_exec() were being treated as failures
1781 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1782 error message is output. Some extra information is given for the
1783 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1784 probably the only errors that are likely to be caused by users (by
1785 specifying a regex that has nested indefinite repeats, for instance).
1786 If there are more than 20 of these errors, pcregrep is abandoned.
1787
1788 (b) A binary zero was treated as data while matching, but terminated the
1789 output line if it was written out. This has been fixed: binary zeroes
1790 are now no different to any other data bytes.
1791
1792 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1793 used to set a locale for matching. The --locale=xxxx long option has
1794 been added (no short equivalent) to specify a locale explicitly on the
1795 pcregrep command, overriding the environment variables.
1796
1797 (d) When -B was used with -n, some line numbers in the output were one less
1798 than they should have been.
1799
1800 (e) Added the -o (--only-matching) option.
1801
1802 (f) If -A or -C was used with -c (count only), some lines of context were
1803 accidentally printed for the final match.
1804
1805 (g) Added the -H (--with-filename) option.
1806
1807 (h) The combination of options -rh failed to suppress file names for files
1808 that were found from directory arguments.
1809
1810 (i) Added the -D (--devices) and -d (--directories) options.
1811
1812 (j) Added the -F (--fixed-strings) option.
1813
1814 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1815
1816 (l) Added the --colo(u)r option.
1817
1818 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1819 is not present by default.
1820
1821 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1822 items such as (?R) or (?1), when the recursion could match a number of
1823 alternatives. If it matched one of the alternatives, but subsequently,
1824 outside the recursion, there was a failure, the code tried to back up into
1825 the recursion. However, because of the way PCRE is implemented, this is not
1826 possible, and the result was an incorrect result from the match.
1827
1828 In order to prevent this happening, the specification of recursion has
1829 been changed so that all such subpatterns are automatically treated as
1830 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1831
1832 4. I had overlooked the fact that, in some locales, there are characters for
1833 which isalpha() is true but neither isupper() nor islower() are true. In
1834 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1835 and ordfeminine) are like this. This affected the treatment of \w and \W
1836 when they appeared in character classes, but not when they appeared outside
1837 a character class. The bit map for "word" characters is now created
1838 separately from the results of isalnum() instead of just taking it from the
1839 upper, lower, and digit maps. (Plus the underscore character, of course.)
1840
1841 5. The above bug also affected the handling of POSIX character classes such as
1842 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1843 permanent tables. Instead, the bit maps for such a class were previously
1844 created as the appropriate unions of the upper, lower, and digit bitmaps.
1845 Now they are created by subtraction from the [[:word:]] class, which has
1846 its own bitmap.
1847
1848 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1849 It is created by subtracting the vertical space characters (\x09, \x0a,
1850 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1851 subtraction was done in the overall bitmap for a character class, meaning
1852 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1853 be recognized. This bug has been fixed.
1854
1855 7. Patches from the folks at Google:
1856
1857 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1858 real life, but is still worth protecting against".
1859
1860 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1861 regular expressions".
1862
1863 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1864 have it.
1865
1866 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1867 "configure" and the latter not, in order to fix a problem somebody had
1868 with compiling the Arg class on HP-UX.
1869
1870 (e) Improve the error-handling of the C++ wrapper a little bit.
1871
1872 (f) New tests for checking recursion limiting.
1873
1874 8. The pcre_memmove() function, which is used only if the environment does not
1875 have a standard memmove() function (and is therefore rarely compiled),
1876 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1877 returning a result (though PCRE never actually uses the result).
1878
1879 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1880 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1881 returned instead of calling malloc() with an overflowing number that would
1882 most likely cause subsequent chaos.
1883
1884 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1885
1886 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1887 with this option is matched, the nmatch and pmatch options of regexec() are
1888 ignored.
1889
1890 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1891 provided in case anyone wants to the the POSIX interface with UTF-8
1892 strings.
1893
1894 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1895 C++ linking (needed for some HP-UX environments).
1896
1897 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1898 (unused parameter) and in the pcre_printint() function (omitted "default"
1899 switch label when the default is to do nothing).
1900
1901 15. Added some code to make it possible, when PCRE is compiled as a C++
1902 library, to replace subject pointers for pcre_exec() with a smart pointer
1903 class, thus making it possible to process discontinuous strings.
1904
1905 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1906 much the same function. They were added by different people who were trying
1907 to make PCRE easy to compile on non-Unix systems. It has been suggested
1908 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1909 for compiling on Windows systems. I have therefore replaced it with
1910 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1911 defaults to "extern" for C or "extern C" for C++, which works fine on
1912 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1913 SCOPE with something explicit in config.h. In addition:
1914
1915 (a) pcreposix.h still had just "extern" instead of either of these macros;
1916 I have replaced it with PCRE_DATA_SCOPE.
1917
1918 (b) Functions such as _pcre_xclass(), which are internal to the library,
1919 but external in the C sense, all had PCRE_EXPORT in their definitions.
1920 This is apparently wrong for the Windows case, so I have removed it.
1921 (It makes no difference on Unix-like systems.)
1922
1923 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1924 of recursive calls to match(). This is different to MATCH_LIMIT because
1925 that limits the total number of calls to match(), not all of which increase
1926 the depth of recursion. Limiting the recursion depth limits the amount of
1927 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1928 when PCRE is compiled, and changed at run time. A patch from Google adds
1929 this functionality to the C++ interface.
1930
1931 18. Changes to the handling of Unicode character properties:
1932
1933 (a) Updated the table to Unicode 4.1.0.
1934
1935 (b) Recognize characters that are not in the table as "Cn" (undefined).
1936
1937 (c) I revised the way the table is implemented to a much improved format
1938 which includes recognition of ranges. It now supports the ranges that
1939 are defined in UnicodeData.txt, and it also amalgamates other
1940 characters into ranges. This has reduced the number of entries in the
1941 table from around 16,000 to around 3,000, thus reducing its size
1942 considerably. I realized I did not need to use a tree structure after
1943 all - a binary chop search is just as efficient. Having reduced the
1944 number of entries, I extended their size from 6 bytes to 8 bytes to
1945 allow for more data.
1946
1947 (d) Added support for Unicode script names via properties such as \p{Han}.
1948
1949 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1950 matching that character.
1951
1952 20. When matching a repeated Unicode property with a minimum greater than zero,
1953 (for example \pL{2,}), PCRE could look past the end of the subject if it
1954 reached it while seeking the minimum number of characters. This could
1955 happen only if some of the characters were more than one byte long, because
1956 there is a check for at least the minimum number of bytes.
1957
1958 21. Refactored the implementation of \p and \P so as to be more general, to
1959 allow for more different types of property in future. This has changed the
1960 compiled form incompatibly. Anybody with saved compiled patterns that use
1961 \p or \P will have to recompile them.
1962
1963 22. Added "Any" and "L&" to the supported property types.
1964
1965 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1966 but give a compile time error if the value is greater than 0xff.
1967
1968 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1969 accidentally not being installed or uninstalled.
1970
1971 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1972 made were to insert the current release number. This seemed silly, because
1973 it made things harder for people building PCRE on systems that don't run
1974 "configure". I have turned pcre.h into a distributed file, no longer built
1975 by "configure", with the version identification directly included. There is
1976 no longer a pcre.h.in file.
1977
1978 However, this change necessitated a change to the pcre-config script as
1979 well. It is built from pcre-config.in, and one of the substitutions was the
1980 release number. I have updated configure.ac so that ./configure now finds
1981 the release number by grepping pcre.h.
1982
1983 26. Added the ability to run the tests under valgrind.
1984
1985
1986 Version 6.4 05-Sep-05
1987 ---------------------
1988
1989 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1990 "--" to be printed when multiple files were scanned, even when none of the
1991 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1992 consider it to be a bug, and have restored the previous behaviour.
1993
1994 2. A couple of code tidies to get rid of compiler warnings.
1995
1996 3. The pcretest program used to cheat by referring to symbols in the library
1997 whose names begin with _pcre_. These are internal symbols that are not
1998 really supposed to be visible externally, and in some environments it is
1999 possible to suppress them. The cheating is now confined to including
2000 certain files from the library's source, which is a bit cleaner.
2001
2002 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2003 file's purpose clearer.
2004
2005 5. Reorganized pcre_ucp_findchar().
2006
2007
2008 Version 6.3 15-Aug-05
2009 ---------------------
2010
2011 1. The file libpcre.pc.in did not have general read permission in the tarball.
2012
2013 2. There were some problems when building without C++ support:
2014
2015 (a) If C++ support was not built, "make install" and "make test" still
2016 tried to test it.
2017
2018 (b) There were problems when the value of CXX was explicitly set. Some
2019 changes have been made to try to fix these, and ...
2020
2021 (c) --disable-cpp can now be used to explicitly disable C++ support.
2022
2023 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2024 backslash in a target when C++ was disabled. This confuses some
2025 versions of "make", apparently. Using an intermediate variable solves
2026 this. (Same for CPP_LOBJ.)
2027
2028 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2029 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2030 necessary on certain architectures.
2031
2032 4. Added a setting of -export-symbols-regex to the link command to remove
2033 those symbols that are exported in the C sense, but actually are local
2034 within the library, and not documented. Their names all begin with
2035 "_pcre_". This is not a perfect job, because (a) we have to except some
2036 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2037 available (and never for static libraries). I have made a note to try to
2038 find a way round (a) in the future.
2039
2040
2041 Version 6.2 01-Aug-05
2042 ---------------------
2043
2044 1. There was no test for integer overflow of quantifier values. A construction
2045 such as {1111111111111111} would give undefined results. What is worse, if
2046 a minimum quantifier for a parenthesized subpattern overflowed and became
2047 negative, the calculation of the memory size went wrong. This could have
2048 led to memory overwriting.
2049
2050 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2051
2052 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2053 operating environments where this matters.
2054
2055 4. Applied Giuseppe Maxia's patch to add additional features for controlling
2056 PCRE options from within the C++ wrapper.
2057
2058 5. Named capturing subpatterns were not being correctly counted when a pattern
2059 was compiled. This caused two problems: (a) If there were more than 100
2060 such subpatterns, the calculation of the memory needed for the whole
2061 compiled pattern went wrong, leading to an overflow error. (b) Numerical
2062 back references of the form \12, where the number was greater than 9, were
2063 not recognized as back references, even though there were sufficient
2064 previous subpatterns.
2065
2066 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2067 versions of gcc, e.g. 2.95.4.
2068
2069
2070 Version 6.1 21-Jun-05
2071 ---------------------
2072
2073 1. There was one reference to the variable "posix" in pcretest.c that was not
2074 surrounded by "#if !defined NOPOSIX".
2075
2076 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2077 the cross-check on the old pcre_info() function, for the benefit of the
2078 cut-down version of PCRE that is currently imported into Exim.
2079
2080 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2081 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2082 patterns that start (?i) but I don't think that matters. The use of (?i) is
2083 just an example; this all applies to the other options as well.
2084
2085 4. Since libtool seems to echo the compile commands it is issuing, the output
2086 from "make" can be reduced a bit by putting "@" in front of each libtool
2087 compile command.
2088
2089 5. Patch from the folks at Google for configure.in to be a bit more thorough
2090 in checking for a suitable C++ installation before trying to compile the
2091 C++ stuff. This should fix a reported problem when a compiler was present,
2092 but no suitable headers.
2093
2094 6. The man pages all had just "PCRE" as their title. I have changed them to
2095 be the relevant file name. I have also arranged that these names are
2096 retained in the file doc/pcre.txt, which is a concatenation in text format
2097 of all the man pages except the little individual ones for each function.
2098
2099 7. The NON-UNIX-USE file had not been updated for the different set of source
2100 files that come with release 6. I also added a few comments about the C++
2101 wrapper.
2102
2103
2104 Version 6.0 07-Jun-05
2105 ---------------------
2106
2107 1. Some minor internal re-organization to help with my DFA experiments.
2108
2109 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2110 didn't matter for the library itself when fully configured, but did matter
2111 when compiling without UCP support, or within Exim, where the ucp files are
2112 not imported.
2113
2114 3. Refactoring of the library code to split up the various functions into
2115 different source modules. The addition of the new DFA matching code (see
2116 below) to a single monolithic source would have made it really too
2117 unwieldy, quite apart from causing all the code to be include in a
2118 statically linked application, when only some functions are used. This is
2119 relevant even without the DFA addition now that patterns can be compiled in
2120 one application and matched in another.
2121
2122 The downside of splitting up is that there have to be some external
2123 functions and data tables that are used internally in different modules of
2124 the library but which are not part of the API. These have all had their
2125 names changed to start with "_pcre_" so that they are unlikely to clash
2126 with other external names.
2127
2128 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2129 a different (DFA) algorithm. Although it is slower than the original
2130 function, it does have some advantages for certain types of matching
2131 problem.
2132
2133 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2134 including restarting after a partial match.
2135
2136 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2137 defined when compiling for Windows was sent to me. I have put it into the
2138 code, though I have no means of testing or verifying it.
2139
2140 7. Added the pcre_refcount() auxiliary function.
2141
2142 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2143 match before or at the first newline in the subject string. In pcretest,
2144 the /f option on a pattern can be used to set this.
2145
2146 9. A repeated \w when used in UTF-8 mode with characters greater than 256
2147 would behave wrongly. This has been present in PCRE since release 4.0.
2148
2149 10. A number of changes to the pcregrep command:
2150
2151 (a) Refactored how -x works; insert ^(...)$ instead of setting
2152 PCRE_ANCHORED and checking the length, in preparation for adding
2153 something similar for -w.
2154
2155 (b) Added the -w (match as a word) option.
2156
2157 (c) Refactored the way lines are read and buffered so as to have more
2158 than one at a time available.
2159
2160 (d) Implemented a pcregrep test script.
2161
2162 (e) Added the -M (multiline match) option. This allows patterns to match
2163 over several lines of the subject. The buffering ensures that at least
2164 8K, or the rest of the document (whichever is the shorter) is available
2165 for matching (and similarly the previous 8K for lookbehind assertions).
2166
2167 (f) Changed the --help output so that it now says
2168
2169 -w, --word-regex(p)
2170
2171 instead of two lines, one with "regex" and the other with "regexp"
2172 because that confused at least one person since the short forms are the
2173 same. (This required a bit of code, as the output is generated
2174 automatically from a table. It wasn't just a text change.)
2175
2176 (g) -- can be used to terminate pcregrep options if the next thing isn't an
2177 option but starts with a hyphen. Could be a pattern or a path name
2178 starting with a hyphen, for instance.
2179
2180 (h) "-" can be given as a file name to represent stdin.
2181
2182 (i) When file names are being printed, "(standard input)" is used for
2183 the standard input, for compatibility with GNU grep. Previously
2184 "<stdin>" was used.
2185
2186 (j) The option --label=xxx can be used to supply a name to be used for
2187 stdin when file names are being printed. There is no short form.
2188
2189 (k) Re-factored the options decoding logic because we are going to add
2190 two more options that take data. Such options can now be given in four
2191 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2192
2193 (l) Added the -A, -B, and -C options for requesting that lines of context
2194 around matches be printed.
2195
2196 (m) Added the -L option to print the names of files that do not contain
2197 any matching lines, that is, the complement of -l.
2198
2199 (n) The return code is 2 if any file cannot be opened, but pcregrep does
2200 continue to scan other files.
2201
2202 (o) The -s option was incorrectly implemented. For compatibility with other
2203 greps, it now suppresses the error message for a non-existent or non-
2204 accessible file (but not the return code). There is a new option called
2205 -q that suppresses the output of matching lines, which was what -s was
2206 previously doing.
2207
2208 (p) Added --include and --exclude options to specify files for inclusion
2209 and exclusion when recursing.
2210
2211 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2212 Hopefully, it now does.
2213
2214 12. Missing cast in pcre_study().
2215
2216 13. Added an "uninstall" target to the makefile.
2217
2218 14. Replaced "extern" in the function prototypes in Makefile.in with
2219 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2220 world, but is set differently for Windows.
2221
2222 15. Added a second compiling function called pcre_compile2(). The only
2223 difference is that it has an extra argument, which is a pointer to an
2224 integer error code. When there is a compile-time failure, this is set
2225 non-zero, in addition to the error test pointer being set to point to an
2226 error message. The new argument may be NULL if no error number is required
2227 (but then you may as well call pcre_compile(), which is now just a
2228 wrapper). This facility is provided because some applications need a
2229 numeric error indication, but it has also enabled me to tidy up the way
2230 compile-time errors are handled in the POSIX wrapper.
2231
2232 16. Added VPATH=.libs to the makefile; this should help when building with one
2233 prefix path and installing with another. (Or so I'm told by someone who
2234 knows more about this stuff than I do.)
2235
2236 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2237 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2238 match everything, including newlines. This is not POSIX-compatible, but
2239 somebody wanted the feature. From pcretest it can be activated by using
2240 both the P and the s flags.
2241
2242 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2243
2244 19. libpcre.pc was being incorrectly installed as executable.
2245
2246 20. A couple of places in pcretest check for end-of-line by looking for '\n';
2247 it now also looks for '\r' so that it will work unmodified on Windows.
2248
2249 21. Added Google's contributed C++ wrapper to the distribution.
2250
2251 22. Added some untidy missing memory free() calls in pcretest, to keep
2252 Electric Fence happy when testing.
2253
2254
2255
2256 Version 5.0 13-Sep-04
2257 ---------------------
2258
2259 1. Internal change: literal characters are no longer packed up into items
2260 containing multiple characters in a single byte-string. Each character
2261 is now matched using a separate opcode. However, there may be more than one
2262 byte in the character in UTF-8 mode.
2263
2264 2. The pcre_callout_block structure has two new fields: pattern_position and
2265 next_item_length. These contain the offset in the pattern to the next match
2266 item, and its length, respectively.
2267
2268 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2269 insertion of callouts before each pattern item. Added the /C option to
2270 pcretest to make use of this.
2271
2272 4. On the advice of a Windows user, the lines
2273
2274 #if defined(_WIN32) || defined(WIN32)
2275 _setmode( _fileno( stdout ), 0x8000 );
2276 #endif /* defined(_WIN32) || defined(WIN32) */
2277
2278 have been added to the source of pcretest. This apparently does useful
2279 magic in relation to line terminators.
2280
2281 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2282 for the benefit of those environments where the "b" makes a difference.
2283
2284 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2285 to know about it. I have put a hack into configure.in that adds in code
2286 to set GCC=yes if CC=icc. This seems to end up at a point in the
2287 generated configure script that is early enough to affect the setting of
2288 compiler options, which is what is needed, but I have no means of testing
2289 whether it really works. (The user who reported this had patched the
2290 generated configure script, which of course I cannot do.)
2291
2292 LATER: After change 22 below (new libtool files), the configure script
2293 seems to know about icc (and also ecc). Therefore, I have commented out
2294 this hack in configure.in.
2295
2296 7. Added support for pkg-config (2 patches were sent in).
2297
2298 8. Negated POSIX character classes that used a combination of internal tables
2299 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2300 [[:^ascii]]. Typically, they would match almost any characters. The other
2301 POSIX classes were not broken in this way.
2302
2303 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2304 to find the match, as PCRE was deluded into thinking that the match had to
2305 start at the start point or following a newline. The same bug applied to
2306 patterns with negative forward assertions or any backward assertions
2307 preceding ".*" at the start, unless the pattern required a fixed first
2308 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2309
2310 10. In UTF-8 mode, when moving forwards in the subject after a failed match
2311 starting at the last subject character, bytes beyond the end of the subject
2312 string were read.
2313
2314 11. Renamed the variable "class" as "classbits" to make life easier for C++
2315 users. (Previously there was a macro definition, but it apparently wasn't
2316 enough.)
2317
2318 12. Added the new field "tables" to the extra data so that tables can be passed
2319 in at exec time, or the internal tables can be re-selected. This allows
2320 a compiled regex to be saved and re-used at a later time by a different
2321 program that might have everything at different addresses.
2322
2323 13. Modified the pcre-config script so that, when run on Solaris, it shows a
2324 -R library as well as a -L library.
2325
2326 14. The debugging options of pcretest (-d on the command line or D on a
2327 pattern) showed incorrect output for anything following an extended class
2328 that contained multibyte characters and which was followed by a quantifier.
2329
2330 15. Added optional support for general category Unicode character properties
2331 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2332 support. It adds about 90K to the size of the library. The meanings of the
2333 inbuilt class escapes such as \d and \s have NOT been changed.
2334
2335 16. Updated pcredemo.c to include calls to free() to release the memory for the
2336 compiled pattern.
2337
2338 17. The generated file chartables.c was being created in the source directory
2339 instead of in the building directory. This caused the build to fail if the
2340 source directory was different from the building directory, and was
2341 read-only.
2342
2343 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2344 file. No doubt somebody will tell me if they don't make sense... Also added
2345 Dan Mooney's comments about building on OpenVMS.
2346
2347 19. Added support for partial matching via the PCRE_PARTIAL option for
2348 pcre_exec() and the \P data escape in pcretest.
2349
2350 20. Extended pcretest with 3 new pattern features:
2351
2352 (i) A pattern option of the form ">rest-of-line" causes pcretest to
2353 write the compiled pattern to the file whose name is "rest-of-line".
2354 This is a straight binary dump of the data, with the saved pointer to
2355 the character tables forced to be NULL. The study data, if any, is
2356 written too. After writing, pcretest reads a new pattern.
2357
2358 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2359 compiled pattern from the given file. There must not be any
2360 occurrences of "<" in the file name (pretty unlikely); if there are,
2361 pcretest will instead treat the initial "<" as a pattern delimiter.
2362 After reading in the pattern, pcretest goes on to read data lines as
2363 usual.
2364
2365 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2366 and 16-bit fields in a compiled pattern, to simulate a pattern that
2367 was compiled on a host of opposite endianness.
2368
2369 21. The pcre-exec() function can now cope with patterns that were compiled on
2370 hosts of opposite endianness, with this restriction:
2371
2372 As for any compiled expression that is saved and used later, the tables
2373 pointer field cannot be preserved; the extra_data field in the arguments
2374 to pcre_exec() should be used to pass in a tables address if a value
2375 other than the default internal tables were used at compile time.
2376
2377 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2378 now diagnosed as an error. Previously, most of the time, a negative number
2379 would have been treated as zero, but if in addition "ovector" was passed as
2380 NULL, a crash could occur.
2381
2382 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2383 new versions from the libtool 1.5 distribution (the last one is a copy of
2384 a file called libtool.m4). This seems to have fixed the need to patch
2385 "configure" to support Darwin 1.3 (which I used to do). However, I still
2386 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2387 workstation).
2388
2389 24. Changed the PCRE licence to be the more standard "BSD" licence.
2390
2391
2392 Version 4.5 01-Dec-03
2393 ---------------------
2394
2395 1. There has been some re-arrangement of the code for the match() function so
2396 that it can be compiled in a version that does not call itself recursively.
2397 Instead, it keeps those local variables that need separate instances for
2398 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2399 needs to "recurse". Keeping track of where control must go is done by means
2400 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2401 hide most of the details from the main code, and operates only if
2402 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2403 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2404 operating.
2405
2406 To make it easier for callers to provide specially tailored get/free
2407 functions for this usage, two new functions, pcre_stack_malloc, and
2408 pcre_stack_free, are used. They are always called in strict stacking order,
2409 and the size of block requested is always the same.
2410
2411 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2412 PCRE has been compiled to use the stack or the heap for recursion. The
2413 -C option of pcretest uses this to show which version is compiled.
2414
2415 A new data escape \S, is added to pcretest; it causes the amounts of store
2416 obtained and freed by both kinds of malloc/free at match time to be added
2417 to the output.
2418
2419 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2420 what's available on my current Linux desktop machine.
2421
2422 3. When matching a UTF-8 string, the test for a valid string at the start has
2423 been extended. If start_offset is not zero, PCRE now checks that it points
2424 to a byte that is the start of a UTF-8 character. If not, it returns
2425 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2426 this is necessary because there may be backward assertions in the pattern.
2427 When matching the same subject several times, it may save resources to use
2428 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2429
2430 4. The code for checking the validity of UTF-8 strings has been tightened so
2431 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2432 containing "overlong sequences".
2433
2434 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2435 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2436 should have been "&", but it just so happened that all the cases this let
2437 through by mistake were picked up later in the function.
2438
2439 6. I had used a variable called "isblank" - this is a C99 function, causing
2440 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2441
2442 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2443 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2444 so the version is shown; (c) stop "make test" echoing "./RunTest".
2445
2446 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2447
2448 9. The prototype for memmove() for systems that don't have it was using
2449 size_t, but the inclusion of the header that defines size_t was later. I've
2450 moved the #includes for the C headers earlier to avoid this.
2451
2452 10. Added some adjustments to the code to make it easier to compiler on certain
2453 special systems:
2454
2455 (a) Some "const" qualifiers were missing.
2456 (b) Added the macro EXPORT before all exported functions; by default this
2457 is defined to be empty.
2458 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2459 that it reads its output file name as an argument instead of writing
2460 to the standard output and assuming this can be redirected.
2461
2462 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2463 class containing characters with values greater than 255, PCRE compilation
2464 went into a loop.
2465
2466 12. A recursive reference to a subpattern that was within another subpattern
2467 that had a minimum quantifier of zero caused PCRE to crash. For example,
2468 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2469 recursion. If the recursively-called subpattern itself had a zero repeat,
2470 that was OK.
2471
2472 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2473 buffer into which it was copied (for escape processing) was still set at
2474 1024, so long lines caused crashes.
2475
2476 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2477 "internal error: code overflow...". This applied to any character class
2478 that was followed by a possessive quantifier.
2479
2480 15. Modified the Makefile to add libpcre.la as a prerequisite for
2481 libpcreposix.la because I was told this is needed for a parallel build to
2482 work.
2483
2484 16. If a pattern that contained .* following optional items at the start was
2485 studied, the wrong optimizing data was generated, leading to matching
2486 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2487 matching string must start with a or b or c. The correct conclusion for
2488 this pattern is that a match can start with any character.
2489
2490
2491 Version 4.4 13-Aug-03
2492 ---------------------
2493
2494 1. In UTF-8 mode, a character class containing characters with values between
2495 127 and 255 was not handled correctly if the compiled pattern was studied.
2496 In fixing this, I have also improved the studying algorithm for such
2497 classes (slightly).
2498
2499 2. Three internal functions had redundant arguments passed to them. Removal
2500 might give a very teeny performance improvement.
2501
2502 3. Documentation bug: the value of the capture_top field in a callout is *one
2503 more than* the number of the hightest numbered captured substring.
2504
2505 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2506 in incorrectly linking with a previously installed version. They now link
2507 explicitly with libpcre.la.
2508
2509 5. configure.in no longer needs to recognize Cygwin specially.
2510
2511 6. A problem in pcre.in for Windows platforms is fixed.
2512
2513 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2514 pcretest, it used to include the size of the study block as part of its
2515 output. Unfortunately, the structure contains a field that has a different
2516 size on different hardware architectures. This meant that the tests that
2517 showed this size failed. As the block is currently always of a fixed size,
2518 this information isn't actually particularly useful in pcretest output, so
2519 I have just removed it.
2520
2521 8. Three pre-processor statements accidentally did not start in column 1.
2522 Sadly, there are *still* compilers around that complain, even though
2523 standard C has not required this for well over a decade. Sigh.
2524
2525 9. In pcretest, the code for checking callouts passed small integers in the
2526 callout_data field, which is a void * field. However, some picky compilers
2527 complained about the casts involved for this on 64-bit systems. Now
2528 pcretest passes the address of the small integer instead, which should get
2529 rid of the warnings.
2530
2531 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2532 both compile and run time, and gives an error if an invalid UTF-8 sequence
2533 is found. There is a option for disabling this check in cases where the
2534 string is known to be correct and/or the maximum performance is wanted.
2535
2536 11. In response to a bug report, I changed one line in Makefile.in from
2537
2538 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2539 to
2540 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2541
2542 to look similar to other lines, but I have no way of telling whether this
2543 is the right thing to do, as I do not use Windows. No doubt I'll get told
2544 if it's wrong...
2545
2546
2547 Version 4.3 21-May-03
2548 ---------------------
2549
2550 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2551 Makefile.
2552
2553 2. Some refactoring to improve the quality of the code:
2554
2555 (i) The utf8_table... variables are now declared "const".
2556
2557 (ii) The code for \cx, which used the "case flipping" table to upper case
2558 lower case letters, now just substracts 32. This is ASCII-specific,
2559 but the whole concept of \cx is ASCII-specific, so it seems
2560 reasonable.
2561
2562 (iii) PCRE was using its character types table to recognize decimal and
2563 hexadecimal digits in the pattern. This is silly, because it handles
2564 only 0-9, a-f, and A-F, but the character types table is locale-
2565 specific, which means strange things might happen. A private
2566 table is now used for this - though it costs 256 bytes, a table is
2567 much faster than multiple explicit tests. Of course, the standard
2568 character types table is still used for matching digits in subject
2569 strings against \d.
2570
2571 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2572 ending in _t are). So I've renamed it as ESC_tee.
2573
2574 3. The first argument for regexec() in the POSIX wrapper should have been
2575 defined as "const".
2576
2577 4. Changed pcretest to use malloc() for its buffers so that they can be
2578 Electric Fenced for debugging.
2579
2580 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2581 to read one or more bytes before the start of the subject string. Often this
2582 had no effect on PCRE's behaviour, but in some circumstances it could
2583 provoke a segmentation fault.
2584
2585 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2586 to try to read one or more bytes before the start of the subject string.
2587
2588 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2589 UTF-8 support could misbehave in various ways if the subject string
2590 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2591 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2592 back over UTF-8 characters.)
2593
2594
2595 Version 4.2 14-Apr-03
2596 ---------------------
2597
2598 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2599
2600 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2601 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2602 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2603 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2604 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2605 and BUILD_EXEEXT
2606 Note: automatic setting of the BUILD variables is not yet working
2607 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2608 compile-time but not at link-time
2609 [LINK]: use for linking executables only
2610 make different versions for Windows and non-Windows
2611 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2612 libraries
2613 [LINK_FOR_BUILD]: new variable
2614 [OBJEXT]: use throughout
2615 [EXEEXT]: use throughout
2616 <winshared>: new target
2617 <wininstall>: new target
2618 <dftables.o>: use native compiler
2619 <dftables>: use native linker
2620 <install>: handle Windows platform correctly
2621 <clean>: ditto
2622 <check>: ditto
2623 copy DLL to top builddir before testing
2624
2625 As part of these changes, -no-undefined was removed again. This was reported
2626 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2627 in any case.
2628
2629 3. Some tidies to get rid of compiler warnings:
2630
2631 . In the match_data structure, match_limit was an unsigned long int, whereas
2632 match_call_count was an int. I've made them both unsigned long ints.
2633
2634 . In pcretest the fact that a const uschar * doesn't automatically cast to
2635 a void * provoked a warning.
2636
2637 . Turning on some more compiler warnings threw up some "shadow" variables
2638 and a few more missing casts.
2639
2640 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2641 option, a class that contained a single character with a value between 128
2642 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2643
2644 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2645 option, a class that contained several characters, but with at least one
2646 whose value was between 128 and 255 caused PCRE to crash.
2647
2648
2649 Version 4.1 12-Mar-03
2650 ---------------------
2651
2652 1. Compiling with gcc -pedantic found a couple of places where casts were
2653 needed, and a string in dftables.c that was longer than standard compilers are
2654 required to support.
2655
2656 2. Compiling with Sun's compiler found a few more places where the code could
2657 be tidied up in order to avoid warnings.
2658
2659 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2660 first of these names is deprecated in the latest Autoconf in favour of the name
2661 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2662 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2663 analogy I have changed it to CFLAGS_FOR_BUILD.
2664
2665 4. Added -no-undefined to the linking command in the Makefile, because this is
2666 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2667 linking step for the pcreposix library.
2668
2669 5. PCRE was failing to diagnose the case of two named groups with the same
2670 name.
2671
2672 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2673 literal character that is needed in the subject for a match, and scans along to
2674 ensure that it is present before embarking on the full matching process. This
2675 saves time in cases of nested unlimited repeats that are never going to match.
2676 Problem: the scan can take a lot of time if the subject is very long (e.g.
2677 megabytes), thus penalizing straightforward matches. It is now done only if the
2678 amount of subject to be scanned is less than 1000 bytes.
2679
2680 7. A lesser problem with the same optimization is that it was recording the
2681 first character of an anchored pattern as "needed", thus provoking a search
2682 right along the subject, even when the first match of the pattern was going to
2683 fail. The "needed" character is now not set for anchored patterns, unless it
2684 follows something in the pattern that is of non-fixed length. Thus, it still
2685 fulfils its original purpose of finding quick non-matches in cases of nested
2686 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2687
2688
2689 Version 4.0 17-Feb-03
2690 ---------------------
2691
2692 1. If a comment in an extended regex that started immediately after a meta-item
2693 extended to the end of string, PCRE compiled incorrect data. This could lead to
2694 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2695
2696 2. Moved to autoconf 2.53 and libtool 1.4.2.
2697
2698 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2699 the special perltest8 script is no longer needed - all the tests can be run
2700 from a single perltest script.
2701
2702 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2703 by \s. It has now been removed in PCRE. This means it isn't recognized as
2704 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2705 class [:space:] *does* include VT, thereby creating a mess.
2706
2707 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2708 space and tab.
2709
2710 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2711 its new features into the main test script, reducing the number of scripts.
2712
2713 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2714 were backward compatible, and made the (?i) apply to the whole pattern, as if
2715 /i were given. Now it behaves more logically, and applies the option setting
2716 only to what follows. PCRE has been changed to follow suit. However, if it
2717 finds options settings right at the start of the pattern, it extracts them into
2718 the global options, as before. Thus, they show up in the info data.
2719
2720 8. Added support for the \Q...\E escape sequence. Characters in between are
2721 treated as literals. This is slightly different from Perl in that $ and @ are
2722 also handled as literals inside the quotes. In Perl, they will cause variable
2723 interpolation. Note the following examples:
2724
2725 Pattern PCRE matches Perl matches
2726
2727 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2728 \Qabc\$xyz\E abc\$xyz abc\$xyz
2729 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2730
2731 For compatibility with Perl, \Q...\E sequences are recognized inside character
2732 classes as well as outside them.
2733
2734 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2735 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2736 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2737 signed/unsigned warnings.
2738
2739 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2740 option for pcretest, so I've replaced it by a simple function that does just
2741 that job.
2742
2743 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2744 "pcregrep -".
2745
2746 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2747 Java package. This provides some syntactic sugar for simple cases of what my
2748 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2749 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2750 item, you can use this simplified notation. Note that only makes sense with
2751 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2752 greediness, whatever the setting of the PCRE_UNGREEDY option.
2753
2754 13. A change of greediness default within a pattern was not taking effect at
2755 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2756 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2757 was abstracted outside.
2758
2759 14. PCRE now supports the \G assertion. It is true when the current matching
2760 position is at the start point of the match. This differs from \A when the
2761 starting offset is non-zero. Used with the /g option of pcretest (or similar
2762 code), it works in the same way as it does for Perl's /g option. If all
2763 alternatives of a regex begin with \G, the expression is anchored to the start
2764 match position, and the "anchored" flag is set in the compiled expression.
2765
2766 15. Some bugs concerning the handling of certain option changes within patterns
2767 have been fixed. These applied to options other than (?ims). For example,
2768 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2769 been the other way round. Some of this was related to change 7 above.
2770
2771 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2772 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2773 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2774 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2775
2776 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2777 mode. Unlike ".", it always matches newline, whatever the setting of
2778 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2779 assertions. Perl allows it, but it doesn't (in general) work because it can't
2780 calculate the length of the lookbehind. At least, that's the case for Perl
2781 5.8.0 - I've been told they are going to document that it doesn't work in
2782 future.
2783
2784 18. Added an error diagnosis for escapes that PCRE does not support: these are
2785 \L, \l, \N, \P, \p, \U, \u, and \X.
2786
2787 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2788 reading past the end of the pattern in cases such as /[abcd/.
2789
2790 20. PCRE was getting more memory than necessary for patterns with classes that
2791 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2792
2793 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2794 compiling PCRE for use with Virtual Pascal.
2795
2796 22. Small fix to the Makefile to make it work properly if the build is done
2797 outside the source tree.
2798
2799 23. Added a new extension: a condition to go with recursion. If a conditional
2800 subpattern starts with (?(R) the "true" branch is used if recursion has
2801 happened, whereas the "false" branch is used only at the top level.
2802
2803 24. When there was a very long string of literal characters (over 255 bytes
2804 without UTF support, over 250 bytes with UTF support), the computation of how
2805 much memory was required could be incorrect, leading to segfaults or other
2806 strange effects.
2807
2808 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2809 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2810 there was a subsequent back reference to those brackets. This meant that, for
2811 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2812 possible to check for precisely this case. All we can do is abandon the
2813 optimization if .* occurs inside capturing brackets when there are any back
2814 references whatsoever. (See below for a better fix that came later.)
2815
2816 26. The handling of the optimization for finding the first character of a
2817 non-anchored pattern, and for finding a character that is required later in the
2818 match were failing in some cases. This didn't break the matching; it just
2819 failed to optimize when it could. The way this is done has been re-implemented.
2820
2821 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2822
2823 28. Added a new feature that provides some of the functionality that Perl
2824 provides with (?{...}). The facility is termed a "callout". The way it is done
2825 in PCRE is for the caller to provide an optional function, by setting
2826 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2827 global variable. By default it is unset, which disables all calling out. To get
2828 the function called, the regex must include (?C) at appropriate points. This
2829 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2830 This provides a means of identifying different callout points. When PCRE
2831 reaches such a point in the regex, if pcre_callout has been set, the external
2832 function is called. It is provided with data in a structure called
2833 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2834 matching continues; if it returns a non-zero value, the match at the current
2835 point fails. However, backtracking will occur if possible. [This was changed
2836 later and other features added - see item 49 below.]
2837
2838 29. pcretest is upgraded to test the callout functionality. It provides a
2839 callout function that displays information. By default, it shows the start of
2840 the match and the current position in the text. There are some new data escapes
2841 to vary what happens:
2842
2843 \C+ in addition, show current contents of captured substrings
2844 \C- do not supply a callout function
2845 \C!n return 1 when callout number n is reached
2846 \C!n!m return 1 when callout number n is reached for the mth time
2847
2848 30. If pcregrep was called with the -l option and just a single file name, it
2849 output "<stdin>" if a match was found, instead of the file name.
2850
2851 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2852 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2853 pcre_exec(). This saves a malloc/free per call. The default value of
2854 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2855 when configuring.
2856
2857 32. The default maximum size of a compiled pattern is 64K. There have been a
2858 few cases of people hitting this limit. The code now uses macros to handle the
2859 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2860 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2861 configuring. Tests 2 and 5 work only with 2-byte links because they output
2862 debugging information about compiled patterns.
2863
2864 33. Internal code re-arrangements:
2865
2866 (a) Moved the debugging function for printing out a compiled regex into
2867 its own source file (printint.c) and used #include to pull it into
2868 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2869 separate copies.
2870
2871 (b) Defined the list of op-code names for debugging as a macro in
2872 internal.h so that it is next to the definition of the opcodes.
2873
2874 (c) Defined a table of op-code lengths for simpler skipping along compiled
2875 code. This is again a macro in internal.h so that it is next to the
2876 definition of the opcodes.
2877
2878 34. Added support for recursive calls to individual subpatterns, along the
2879 lines of Robin Houston's patch (but implemented somewhat differently).
2880
2881 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2882 allow it to read and process whole directories in Win32. This code was
2883 contributed by Lionel Fourquaux; it has not been tested by me.
2884
2885 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2886 used to name a group. Names consist of alphanumerics and underscores, and must
2887 be unique. Back references use the syntax (?P=name) and recursive calls use
2888 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2889 numbers. The function pcre_fullinfo() can be used after compilation to extract
2890 a name/number map. There are three relevant calls:
2891
2892 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2893 PCRE_INFO_NAMECOUNT yields the number of entries
2894 PCRE_INFO_NAMETABLE yields a pointer to the map.
2895
2896 The map is a vector of fixed-size entries. The size of each entry depends on
2897 the length of the longest name used. The first two bytes of each entry are the
2898 group number, most significant byte first. There follows the corresponding
2899 name, zero terminated. The names are in alphabetical order.
2900
2901 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2902 case instead of 255. Making it the same both with and without UTF-8 support
2903 means that the same test output works with both.
2904
2905 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2906 calling malloc() with a zero argument.
2907
2908 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2909 optimization. I've improved things by keeping a bitmap of backreferences with
2910 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2911 fact referenced, the optimization can be applied. It is unlikely that a
2912 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2913 the match to follow \n) will appear inside brackets with a number greater than
2914 31, but if it does, any back reference > 31 suppresses the optimization.
2915
2916 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2917 of disabling numbered capturing parentheses. Any opening parenthesis that is
2918 not followed by ? behaves as if it were followed by ?: but named parentheses
2919 can still be used for capturing (and they will acquire numbers in the usual
2920 way).
2921
2922 41. Redesigned the return codes from the match() function into yes/no/error so
2923 that errors can be passed back from deep inside the nested calls. A malloc
2924 failure while inside a recursive subpattern call now causes the
2925 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2926
2927 42. It is now possible to set a limit on the number of times the match()
2928 function is called in a call to pcre_exec(). This facility makes it possible to
2929 limit the amount of recursion and backtracking, though not in a directly
2930 obvious way, because the match() function is used in a number of different
2931 circumstances. The count starts from zero for each position in the subject
2932 string (for non-anchored patterns). The default limit is, for compatibility, a
2933 large number, namely 10 000 000. You can change this in two ways:
2934
2935 (a) When configuring PCRE before making, you can use --with-match-limit=n
2936 to set a default value for the compiled library.
2937
2938 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2939 a different value is set. See 45 below.
2940
2941 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2942
2943 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2944 of things that can be changed at compile time. The first argument specifies
2945 what is wanted and the second points to where the information is to be placed.
2946 The current list of available information is:
2947
2948 PCRE_CONFIG_UTF8
2949
2950 The output is an integer that is set to one if UTF-8 support is available;
2951 otherwise it is set to zero.
2952
2953 PCRE_CONFIG_NEWLINE
2954
2955 The output is an integer that it set to the value of the code that is used for
2956 newline. It is either LF (10) or CR (13).
2957
2958 PCRE_CONFIG_LINK_SIZE
2959
2960 The output is an integer that contains the number of bytes used for internal
2961 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2962
2963 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2964
2965 The output is an integer that contains the threshold above which the POSIX
2966 interface uses malloc() for output vectors. See item 31 above.
2967
2968 PCRE_CONFIG_MATCH_LIMIT
2969
2970 The output is an unsigned integer that contains the default limit of the number
2971 of match() calls in a pcre_exec() execution. See 42 above.
2972
2973 44. pcretest has been upgraded by the addition of the -C option. This causes it
2974 to extract all the available output from the new pcre_config() function, and to
2975 output it. The program then exits immediately.
2976
2977 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2978 order to support additional features. One way would have been to define
2979 pcre_exec2() (for example) with extra arguments, but this would not have been
2980 extensible, and would also have required all calls to the original function to
2981 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2982 is used for passing in "extra" data from pcre_study().
2983
2984 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2985 contains the following fields:
2986
2987 flags a bitmap indicating which of the following fields are set
2988 study_data opaque data from pcre_study()
2989 match_limit a way of specifying a limit on match() calls for a specific
2990 call to pcre_exec()
2991 callout_data data for callouts (see 49 below)
2992
2993 The flag bits are also defined in pcre.h, and are
2994
2995 PCRE_EXTRA_STUDY_DATA
2996 PCRE_EXTRA_MATCH_LIMIT
2997 PCRE_EXTRA_CALLOUT_DATA
2998
2999 The pcre_study() function now returns one of these new pcre_extra blocks, with
3000 the actual study data pointed to by the study_data field, and the
3001 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3002 before. That is, this change is entirely upwards-compatible and requires no
3003 change to existing code.
3004
3005 If you want to pass in additional data to pcre_exec(), you can either place it
3006 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3007 block.
3008
3009 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3010 data string contains the escape sequence \M, pcretest calls pcre_exec() several
3011 times with different match limits, until it finds the minimum value needed for
3012 pcre_exec() to complete. The value is then output. This can be instructive; for
3013 most simple matches the number is quite small, but for pathological cases it
3014 gets very large very quickly.
3015
3016 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3017 returns the size of the data block pointed to by the study_data field in a
3018 pcre_extra block, that is, the value that was passed as the argument to
3019 pcre_malloc() when PCRE was getting memory in which to place the information
3020 created by pcre_study(). The fourth argument should point to a size_t variable.
3021 pcretest has been extended so that this information is shown after a successful
3022 pcre_study() call when information about the compiled regex is being displayed.
3023
3024 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3025 because what follows is always an absolute path. (Later: it turns out that this
3026 is more than cosmetic for MinGW, because it doesn't like empty path
3027 components.)
3028
3029 49. Some changes have been made to the callout feature (see 28 above):
3030
3031 (i) A callout function now has three choices for what it returns:
3032
3033 0 => success, carry on matching
3034 > 0 => failure at this point, but backtrack if possible
3035 < 0 => serious error, return this value from pcre_exec()
3036
3037 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3038 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3039 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3040 use by callout functions. It will never be used by PCRE itself.
3041
3042 (ii) The pcre_extra structure (see 45 above) has a void * field called
3043 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3044 pcre_callout_block structure has a field of the same name. The contents of
3045 the field passed in the pcre_extra structure are passed to the callout
3046 function in the corresponding field in the callout block. This makes it
3047 easier to use the same callout-containing regex from multiple threads. For
3048 testing, the pcretest program has a new data escape
3049
3050 \C*n pass the number n (may be negative) as callout_data
3051
3052 If the callout function in pcretest receives a non-zero value as
3053 callout_data, it returns that value.
3054
3055 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3056 there were some redundant $(CFLAGS) in commands that are now specified as
3057 $(LINK), which already includes $(CFLAGS).
3058
3059 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3060 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3061 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3062 one-byte characters throughout. Note that case-insensitive matching applies
3063 only to characters whose values are less than 256. PCRE doesn't support the
3064 notion of cases for higher-valued characters.
3065
3066 (i) A character class whose characters are all within 0-255 is handled as
3067 a bit map, and the map is inverted for negative classes. Previously, a
3068 character > 255 always failed to match such a class; however it should
3069 match if the class was a negative one (e.g. [^ab]). This has been fixed.
3070
3071 (ii) A negated character class with a single character < 255 is coded as
3072 "not this character" (OP_NOT). This wasn't working properly when the test
3073 character was multibyte, either singly or repeated.
3074
3075 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3076 mode, for example: \x{100}{2,3}.
3077
3078 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3079 singly or repeated) now correctly test multibyte characters. However,
3080 PCRE doesn't recognize any characters with values greater than 255 as
3081 digits, spaces, or word characters. Such characters always match \D, \S,
3082 and \W, and never match \d, \s, or \w.
3083
3084 (v) Classes may now contain characters and character ranges with values
3085 greater than 255. For example: [ab\x{100}-\x{400}].
3086
3087 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3088 PCRE in UTF-8 mode.
3089
3090 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3091 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3092 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3093 value.)
3094
3095 53. The single man page has become too large. I have therefore split it up into
3096 a number of separate man pages. These also give rise to individual HTML pages;
3097 these are now put in a separate directory, and there is an index.html page that
3098 lists them all. Some hyperlinking between the pages has been installed.
3099
3100 54. Added convenience functions for handling named capturing parentheses.
3101
3102 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3103 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3104 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3105 are faulted.
3106
3107 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3108 calling configure. These values are used when compiling the dftables.c program
3109 which is run to generate the source of the default character tables. They
3110 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3111 you will need to set these values.
3112
3113 57. Updated the building process for Windows DLL, as provided by Fred Cox.
3114
3115
3116 Version 3.9 02-Jan-02
3117 ---------------------
3118
3119 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3120
3121 2. If --disable-static was given, the building process failed when trying to
3122 build pcretest and pcregrep. (For some reason it was using libtool to compile
3123 them, which is not right, as they aren't part of the library.)
3124
3125
3126 Version 3.8 18-Dec-01
3127 ---------------------
3128
3129 1. The experimental UTF-8 code was completely screwed up. It was packing the
3130 bytes in the wrong order. How dumb can you get?
3131
3132
3133 Version 3.7 29-Oct-01
3134 ---------------------
3135
3136 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3137 This caused pcretest, when used on the test data, to segfault. Unfortunately,
3138 this didn't happen under Solaris 8, where I normally test things.
3139
3140 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3141 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3142 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3143 this fix an hour or so after the initial 3.7 release.)
3144
3145
3146 Version 3.6 23-Oct-01
3147 ---------------------
3148
3149 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3150 offsets passed as NULL with zero offset count.
3151
3152 2. The config.guess and config.sub files had not been updated when I moved to
3153 the latest autoconf.
3154
3155
3156 Version 3.5 15-Aug-01
3157 ---------------------
3158
3159 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3160 had been forgotten.
3161
3162 2. By using declared but undefined structures, we can avoid using "void"
3163 definitions in pcre.h while keeping the internal definitions of the structures
3164 private.
3165
3166 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3167 user point of view, this means that both static and shared libraries are built
3168 by default, but this can be individually controlled. More of the work of
3169 handling this static/shared cases is now inside libtool instead of PCRE's make
3170 file.
3171
3172 4. The pcretest utility is now installed along with pcregrep because it is
3173 useful for users (to test regexs) and by doing this, it automatically gets
3174 relinked by libtool. The documentation has been turned into a man page, so
3175 there are now .1, .txt, and .html versions in /doc.
3176
3177 5. Upgrades to pcregrep:
3178 (i) Added long-form option names like gnu grep.
3179 (ii) Added --help to list all options with an explanatory phrase.
3180 (iii) Added -r, --recursive to recurse into sub-directories.
3181 (iv) Added -f, --file to read patterns from a file.
3182
3183 6. pcre_exec() was referring to its "code" argument before testing that
3184 argument for NULL (and giving an error if it was NULL).
3185
3186 7. Upgraded Makefile.in to allow for compiling in a different directory from
3187 the source directory.
3188
3189 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3190 options bits, the pointer it was passed was to an int instead of to an unsigned
3191 long int. This mattered only on 64-bit systems.
3192
3193 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3194 generated) instead of pcre.in, which it its source. Also made the same change
3195 in several of the .c files.
3196
3197 10. A new release of gcc defines printf() as a macro, which broke pcretest
3198 because it had an ifdef in the middle of a string argument for printf(). Fixed
3199 by using separate calls to printf().
3200
3201 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3202 script, to force use of CR or LF instead of \n in the source. On non-Unix
3203 systems, the value can be set in config.h.
3204
3205 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3206 absolute limit. Changed the text of the error message to make this clear, and
3207 likewise updated the man page.
3208
3209 13. The limit of 99 on the number of capturing subpatterns has been removed.
3210 The new limit is 65535, which I hope will not be a "real" limit.
3211
3212
3213 Version 3.4 22-Aug-00
3214 ---------------------
3215
3216 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3217
3218 2. Diagnose condition (?(0) as an error instead of crashing on matching.
3219
3220
3221 Version 3.3 01-Aug-00
3222 ---------------------
3223
3224 1. If an octal character was given, but the value was greater than \377, it
3225 was not getting masked to the least significant bits, as documented. This could
3226 lead to crashes in some systems.
3227
3228 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3229 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3230
3231 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3232 These just pass their arguments on to (pcre_free)(), but they are provided
3233 because some uses of PCRE bind it to non-C systems that can call its functions,
3234 but cannot call free() or pcre_free() directly.
3235
3236 4. Add "make test" as a synonym for "make check". Corrected some comments in
3237 the Makefile.
3238
3239 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3240 Makefile.
3241
3242 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3243 command called pgrep for grepping around the active processes.
3244
3245 7. Added the beginnings of support for UTF-8 character strings.
3246
3247 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3248 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3249 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3250 out for the ar command.)
3251
3252
3253 Version 3.2 12-May-00
3254 ---------------------
3255
3256 This is purely a bug fixing release.
3257
3258 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3259 of ZA. This was just one example of several cases that could provoke this bug,
3260 which was introduced by change 9 of version 2.00. The code for breaking
3261 infinite loops after an iteration that matches an empty string was't working
3262 correctly.
3263
3264 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3265 when matched against abbab (for example). After matching an empty string, it
3266 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3267 caused it to match further down the string than it should.
3268
3269 3. The code contained an inclusion of sys/types.h. It isn't clear why this
3270 was there because it doesn't seem to be needed, and it causes trouble on some
3271 systems, as it is not a Standard C header. It has been removed.
3272
3273 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3274 were reported on the Macintosh. The changes were from
3275
3276 while ((c = *(++ptr)) != 0 && c != '\n');
3277 to
3278 while ((c = *(++ptr)) != 0 && c != '\n') ;
3279
3280 Totally extraordinary, but if that's what it takes...
3281
3282 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3283 available. Added HAVE_BCOPY and an autoconf test for it; if neither
3284 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3285 assumes the way PCRE uses memmove() (always moving upwards).
3286
3287 6. PCRE is being used in one environment where strchr() is not available. There
3288 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3289 faster code anyway.
3290
3291
3292 Version 3.1 09-Feb-00
3293 ---------------------
3294
3295 The only change in this release is the fixing of some bugs in Makefile.in for
3296 the "install" target:
3297
3298 (1) It was failing to install pcreposix.h.
3299
3300 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3301
3302
3303 Version 3.0 01-Feb-00
3304 ---------------------
3305
3306 1. Add support for the /+ modifier to perltest (to output $` like it does in
3307 pcretest).
3308
3309 2. Add support for the /g modifier to perltest.
3310
3311 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3312 matches null strings.
3313
3314 4. Fix perltest so that it doesn't do unwanted things when fed an empty
3315 pattern. Perl treats empty patterns specially - it reuses the most recent
3316 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3317 effect.
3318
3319 5. The POSIX interface was broken in that it was just handing over the POSIX
3320 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3321 required a bigger vector, with some working space on the end. This means that
3322 the POSIX wrapper now has to get and free some memory, and copy the results.
3323
3324 6. Added some simple autoconf support, placing the test data and the
3325 documentation in separate directories, re-organizing some of the
3326 information files, and making it build pcre-config (a GNU standard). Also added
3327 libtool support for building PCRE as a shared library, which is now the
3328 default.
3329
3330 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3331 09 are not valid octal constants. Single digits will be used for minor values
3332 less than 10.
3333
3334 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3335 existing programs that set these in the POSIX interface can use PCRE without
3336 modification.
3337
3338 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3339 return all that pcre_info() returns, plus additional data. The pcre_info()
3340 function is retained for compatibility, but is considered to be obsolete.
3341
3342 10. Added experimental recursion feature (?R) to handle one common case that
3343 Perl 5.6 will be able to do with (?p{...}).
3344
3345 11. Added support for POSIX character classes like [:alpha:], which Perl is
3346 adopting.
3347
3348
3349 Version 2.08 31-Aug-99
3350 ----------------------
3351
3352 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3353 trying to match at the startoffset position, but instead was moving forward to
3354 the next newline as if a previous match had failed.
3355
3356 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3357 and could get into a loop if a null string was matched other than at the start
3358 of the subject.
3359
3360 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3361 be distinguished at compile time, and for completeness also added PCRE_DATE.
3362
3363 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3364 in GnuWin32 environments.
3365
3366
3367 Version 2.07 29-Jul-99
3368 ----------------------
3369
3370 1. The documentation is now supplied in plain text form and HTML as well as in
3371 the form of man page sources.
3372
3373 2. C++ compilers don't like assigning (void *) values to other pointer types.
3374 In particular this affects malloc(). Although there is no problem in Standard
3375 C, I've put in casts to keep C++ compilers happy.
3376
3377 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3378 should be (const char *).
3379
3380 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3381 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3382 However, I haven't made this a standard facility. The documentation doesn't
3383 mention it, and the Makefile doesn't support it.
3384
3385 5. The Makefile now contains an "install" target, with editable destinations at
3386 the top of the file. The pcretest program is not installed.
3387
3388 6. pgrep -V now gives the PCRE version number and date.
3389
3390 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3391 causing the entire string to be ignored, instead of just the last character.
3392
3393 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3394 non-matching string, it can take a very, very long time, even for strings of
3395 quite modest length, because of the nested recursion. PCRE now does better in
3396 some of these cases. It does this by remembering the last required literal
3397 character in the pattern, and pre-searching the subject to ensure it is present
3398 before running the real match. In other words, it applies a heuristic to detect
3399 some types of certain failure quickly, and in the above example, if presented
3400 with a string that has no trailing " it gives "no match" very quickly.
3401
3402 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3403 other alternatives are tried instead.
3404
3405
3406 Version 2.06 09-Jun-99
3407 ----------------------
3408
3409 1. Change pcretest's output for amount of store used to show just the code
3410 space, because the remainder (the data block) varies in size between 32-bit and
3411 64-bit systems.
3412
3413 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3414 start matching at. This allows lookbehinds to work when searching for multiple
3415 occurrences in a string.
3416
3417 3. Added additional options to pcretest for testing multiple occurrences:
3418
3419 /+ outputs the rest of the string that follows a match
3420 /g loops for multiple occurrences, using the new startoffset argument
3421 /G loops for multiple occurrences by passing an incremented pointer
3422
3423 4. PCRE wasn't doing the "first character" optimization for patterns starting
3424 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3425 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3426 the letter 'x'. On long subject strings, this gives a significant speed-up.
3427
3428
3429 Version 2.05 21-Apr-99
3430 ----------------------
3431
3432 1. Changed the type of magic_number from int to long int so that it works
3433 properly on 16-bit systems.
3434
3435 2. Fixed a bug which caused patterns starting with .* not to work correctly
3436 when the subject string contained newline characters. PCRE was assuming
3437 anchoring for such patterns in all cases, which is not correct because .* will
3438 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3439 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3440 must be retried after every newline in the subject.
3441
3442
3443 Version 2.04 18-Feb-99
3444 ----------------------
3445
3446 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3447 computation of the store needed to hold the pattern was incorrect (too large).
3448 If such patterns were nested a few deep, this could multiply and become a real
3449 problem.
3450
3451 2. Added /M option to pcretest to show the memory requirement of a specific
3452 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3453
3454 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3455 compiled in such a way that the backtracking after subsequent failure was
3456 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3457 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3458
3459
3460 Version 2.03 02-Feb-99
3461 ----------------------
3462
3463 1. Fixed typo and small mistake in man page.
3464
3465 2. Added 4th condition (GPL supersedes if conflict) and created separate
3466 LICENCE file containing the conditions.
3467
3468 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3469 Perl, that is the internal \ allows the delimiter to be included in the
3470 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3471 the final delimiter, add \ to the end of the pattern (to test the error).
3472
3473 4. Added the convenience functions for extracting substrings after a successful
3474 match. Updated pcretest to make it able to test these functions.
3475
3476
3477 Version 2.02 14-Jan-99
3478 ----------------------
3479
3480 1. Initialized the working variables associated with each extraction so that
3481 their saving and restoring doesn't refer to uninitialized store.
3482
3483 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3484 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3485 fix the problem.
3486
3487 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3488 calls, and wasn't printing the correct value for compiling calls. Increased the
3489 default value of LOOPREPEAT, and the number of significant figures in the
3490 times.
3491
3492 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3493
3494 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3495 a building problem on Windows NT with a FAT file system.
3496
3497
3498 Version 2.01 21-Oct-98
3499 ----------------------
3500
3501 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3502 to character tables built by pcre_maketables() in the current locale. If NULL
3503 is passed, the default tables are used.
3504
3505
3506 Version 2.00 24-Sep-98
3507 ----------------------
3508
3509 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3510 it any more.
3511
3512 2. Allow quantification of (?>) groups, and make it work correctly.
3513
3514 3. The first character computation wasn't working for (?>) groups.
3515
3516 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3517 end of the subject) and add 5.005's \z, which really does match only at the
3518 very end of the subject.
3519
3520 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3521
3522 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3523 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3524 localized options. All options to pcre_study() were also removed.
3525
3526 7. Add other new features from 5.005:
3527
3528 $(?<= positive lookbehind
3529 $(?<! negative lookbehind
3530 (?imsx-imsx) added the unsetting capability
3531 such a setting is global if at outer level; local otherwise
3532 (?imsx-imsx:) non-capturing groups with option setting
3533 (?(cond)re|re) conditional pattern matching
3534
3535 A backreference to itself in a repeated group matches the previous
3536 captured string.
3537
3538 8. General tidying up of studying (both automatic and via "study")
3539 consequential on the addition of new assertions.
3540
3541 9. As in 5.005, unlimited repeated groups that could match an empty substring
3542 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3543 runtime if any iteration does actually match an empty substring.
3544
3545 10. Include the RunTest script in the distribution.
3546
3547 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3548 discrepancies, some of which were old and were also with respect to 5.004. They
3549 have now been fixed.
3550
3551
3552 Version 1.09 28-Apr-98
3553 ----------------------
3554
3555 1. A negated single character class followed by a quantifier with a minimum
3556 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3557 program crashes, or just wrong answers. This did not apply to negated classes
3558 containing more than one character, or to minima other than one.
3559
3560
3561 Version 1.08 27-Mar-98
3562 ----------------------
3563
3564 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3565
3566 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3567 latter must appear before anything that relies on it in the pattern.
3568
3569
3570 Version 1.07 16-Feb-98
3571 ----------------------
3572
3573 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3574 repeat of a potentially empty string).
3575
3576
3577 Version 1.06 23-Jan-98
3578 ----------------------
3579
3580 1. Added Markus Oberhumer's little patches for C++.
3581
3582 2. Literal strings longer than 255 characters were broken.
3583
3584
3585 Version 1.05 23-Dec-97
3586 ----------------------
3587
3588 1. Negated character classes containing more than one character were failing if
3589 PCRE_CASELESS was set at run time.
3590
3591
3592 Version 1.04 19-Dec-97
3593 ----------------------
3594
3595 1. Corrected the man page, where some "const" qualifiers had been omitted.
3596
3597 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3598 input syntax.
3599
3600 3. Fixed memory leak which occurred when a regex with back references was
3601 matched with an offsets vector that wasn't big enough. The temporary memory
3602 that is used in this case wasn't being freed if the match failed.
3603
3604 4. Tidied pcretest to ensure it frees memory that it gets.
3605
3606 5. Temporary memory was being obtained in the case where the passed offsets
3607 vector was exactly big enough.
3608
3609 6. Corrected definition of offsetof() from change 5 below.
3610
3611 7. I had screwed up change 6 below and broken the rules for the use of
3612 setjmp(). Now fixed.
3613
3614
3615 Version 1.03 18-Dec-97
3616 ----------------------
3617
3618 1. A erroneous regex with a missing opening parenthesis was correctly
3619 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3620 on some systems.
3621
3622 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3623 it was reported that one broken compiler failed on the former because "code" is
3624 also an independent variable.
3625
3626 3. The erroneous regex a[]b caused an array overrun reference.
3627
3628 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3629 fail on data ending with that character. (It was going on too far, and checking
3630 the next character, typically a binary zero.) This was specific to the
3631 optimized code for single-character negative classes.
3632
3633 5. Added a contributed patch from the TIN world which does the following:
3634
3635 + Add an undef for memmove, in case the the system defines a macro for it.
3636
3637 + Add a definition of offsetof(), in case there isn't one. (I don't know
3638 the reason behind this - offsetof() is part of the ANSI standard - but
3639 it does no harm).
3640
3641 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3642 most of the places where whitespace preceded '#'. I have given up and
3643 allowed the remaining 2 cases to be at the margin.
3644
3645 + Rename some variables in pcre to eliminate shadowing. This seems very
3646 pedantic, but does no harm, of course.
3647
3648 6. Moved the call to setjmp() into its own function, to get rid of warnings
3649 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3650
3651 7. Constructs such as \d{8,} were compiling into the equivalent of
3652 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3653 outcome, but in this particular case used more store than had been allocated,
3654 which caused the bug to be discovered because it threw up an internal error.
3655
3656 8. The debugging code in both pcre and pcretest for outputting the compiled
3657 form of a regex was going wrong in the case of back references followed by
3658 curly-bracketed repeats.
3659
3660
3661 Version 1.02 12-Dec-97
3662 ----------------------
3663
3664 1. Typos in pcre.3 and comments in the source fixed.
3665
3666 2. Applied a contributed patch to get rid of places where it used to remove
3667 'const' from variables, and fixed some signed/unsigned and uninitialized
3668 variable warnings.
3669
3670 3. Added the "runtest" target to Makefile.
3671
3672 4. Set default compiler flag to -O2 rather than just -O.
3673
3674
3675 Version 1.01 19-Nov-97
3676 ----------------------
3677
3678 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3679 like /([ab]*)*/, that is, for classes with more than one character in them.
3680
3681 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3682 as /((?>a*))*/ (a PCRE_EXTRA facility).
3683
3684
3685 Version 1.00 18-Nov-97
3686 ----------------------
3687
3688 1. Added compile-time macros to support systems such as SunOS4 which don't have
3689 memmove() or strerror() but have other things that can be used instead.
3690
3691 2. Arranged that "make clean" removes the executables.
3692
3693
3694 Version 0.99 27-Oct-97
3695 ----------------------
3696
3697 1. Fixed bug in code for optimizing classes with only one character. It was
3698 initializing a 32-byte map regardless, which could cause it to run off the end
3699 of the memory it had got.
3700
3701 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3702
3703
3704 Version 0.98 22-Oct-97
3705 ----------------------
3706
3707 1. Fixed bug in code for handling temporary memory usage when there are more
3708 back references than supplied space in the ovector. This could cause segfaults.
3709
3710
3711 Version 0.97 21-Oct-97
3712 ----------------------
3713
3714 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3715
3716 2. Optimized negated single characters not to use a bit map.
3717
3718 3. Brought error texts together as macro definitions; clarified some of them;
3719 fixed one that was wrong - it said "range out of order" when it meant "invalid
3720 escape sequence".
3721
3722 4. Changed some char * arguments to const char *.
3723
3724 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3725
3726 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3727 pcretest.
3728
3729
3730 Version 0.96 16-Oct-97
3731 ----------------------
3732
3733 1. Added a simple "pgrep" utility to the distribution.
3734
3735 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3736 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3737 where "ddd" means "one or more decimal digits".
3738
3739 3. Fixed serious bug. If a pattern had a back reference, but the call to
3740 pcre_exec() didn't supply a large enough ovector to record the related
3741 identifying subpattern, the match always failed. PCRE now remembers the number
3742 of the largest back reference, and gets some temporary memory in which to save
3743 the offsets during matching if necessary, in order to ensure that
3744 backreferences always work.
3745
3746 4. Increased the compatibility with Perl in a number of ways:
3747
3748 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3749 to request this handling. The option can be set at compile or exec time.
3750
3751 (b) $ matches before a terminating newline by default; an option
3752 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3753 mode). The option can be set at compile or exec time.
3754
3755 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3756 the same as Perl's. If the decimal number it represents is less than 10
3757 or there aren't that many previous left capturing parentheses, an octal
3758 escape is read. Inside a character class, it's always an octal escape,
3759 even if it is a single digit.
3760
3761 (d) An escaped but undefined alphabetic character is taken as a literal,
3762 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3763 escapes.
3764
3765 (e) {0} is now permitted. (The previous item is removed from the compiled
3766 pattern).
3767
3768 5. Changed all the names of code files so that the basic parts are no longer
3769 than 10 characters, and abolished the teeny "globals.c" file.
3770
3771 6. Changed the handling of character classes; they are now done with a 32-byte
3772 bit map always.
3773
3774 7. Added the -d and /D options to pcretest to make it possible to look at the
3775 internals of compilation without having to recompile pcre.
3776
3777
3778 Version 0.95 23-Sep-97
3779 ----------------------
3780
3781 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3782 \x20 at the start of a run of normal characters. These were being treated as
3783 real characters, instead of the source characters being re-checked.
3784
3785
3786 Version 0.94 18-Sep-97
3787 ----------------------
3788
3789 1. The functions are now thread-safe, with the caveat that the global variables
3790 containing pointers to malloc() and free() or alternative functions are the
3791 same for all threads.
3792
3793 2. Get pcre_study() to generate a bitmap of initial characters for non-
3794 anchored patterns when this is possible, and use it if passed to pcre_exec().
3795
3796
3797 Version 0.93 15-Sep-97
3798 ----------------------
3799
3800 1. /(b)|(:+)/ was computing an incorrect first character.
3801
3802 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3803 but not actually doing anything yet.
3804
3805 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3806 as Perl does (e.g. [-az] or [az-]).
3807
3808 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3809 all possible positions.
3810
3811 5. Split up into different modules to avoid including unneeded functions in a
3812 compiled binary. However, compile and exec are still in one module. The "study"
3813 function is split off.
3814
3815 6. The character tables are now in a separate module whose source is generated
3816 by an auxiliary program - but can then be edited by hand if required. There are
3817 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3818 toupper() in the code.
3819
3820 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3821 make them global. Abolish the function for setting them, as the caller can now
3822 set them directly.
3823
3824
3825 Version 0.92 11-Sep-97
3826 ----------------------
3827
3828 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3829 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3830
3831 2. Caseless matching was not working in character classes if the characters in
3832 the pattern were in upper case.
3833
3834 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3835
3836 4. Make PCRE_ANCHORED public and accept as a compile option.
3837
3838 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3839 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3840 pass them.
3841
3842 6. Give an error if bad option bits passed at compile or run time.
3843
3844 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3845 pcretest to cause it to pass that flag.
3846
3847 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3848 options, and the first character, if set.
3849
3850 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3851
3852
3853 Version 0.91 10-Sep-97
3854 ----------------------
3855
3856 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3857 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3858
3859 2. PCRE was looping on encountering an indefinitely repeated back reference to
3860 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3861 Perl does - treats the match as successful.
3862
3863 ****

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12