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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 515 - (hide annotations) (download)
Tue May 4 09:12:25 2010 UTC (4 years, 2 months ago) by ph10
File size: 185861 byte(s)
Avoid warnings about fwrite in pcregrep.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 ph10 513 Version 8.10 03 May-2010
5 ph10 510 ------------------------
6    
7 ph10 512 1. Added support for (*MARK:ARG) and for ARG additions to PRUNE, SKIP, and
8 ph10 510 THEN.
9    
10 ph10 512 2. (*ACCEPT) was not working when inside an atomic group.
11 ph10 510
12 ph10 513 3. Inside a character class, \B is treated as a literal by default, but
13     faulted if PCRE_EXTRA is set. This mimics Perl's behaviour (the -w option
14     causes the error). The code is unchanged, but I tidied the documentation.
15    
16     4. Inside a character class, PCRE always treated \R and \X as literals,
17     whereas Perl faults them if its -w option is set. I have changed PCRE so
18     that it faults them when PCRE_EXTRA is set.
19 ph10 514
20     5. Added support for \N, which always matches any character other than
21     newline. (It is the same as "." when PCRE_DOTALL is not set.)
22 ph10 515
23     6. When compiling pcregrep with newer versions of gcc which may have
24     FORTIFY_SOURCE set, several warnings "ignoring return value of 'fwrite',
25     declared with attribute warn_unused_result" were given. Just casting the
26     result to (void) does not stop the warnings; a more elaborate fudge is
27     needed. I've used a macro to implement this.
28 ph10 512
29 ph10 513
30 ph10 508 Version 8.02 19-Mar-2010
31 ph10 491 ------------------------
32    
33     1. The Unicode data tables have been updated to Unicode 5.2.0.
34    
35 ph10 492 2. Added the option --libs-cpp to pcre-config, but only when C++ support is
36     configured.
37 ph10 507
38 ph10 493 3. Updated the licensing terms in the pcregexp.pas file, as agreed with the
39 ph10 507 original author of that file, following a query about its status.
40    
41     4. On systems that do not have stdint.h (e.g. Solaris), check for and include
42 ph10 494 inttypes.h instead. This fixes a bug that was introduced by change 8.01/8.
43 ph10 507
44 ph10 496 5. A pattern such as (?&t)*+(?(DEFINE)(?<t>.)) which has a possessive
45     quantifier applied to a forward-referencing subroutine call, could compile
46     incorrect code or give the error "internal error: previously-checked
47     referenced subpattern not found".
48 ph10 507
49     6. Both MS Visual Studio and Symbian OS have problems with initializing
50 ph10 497 variables to point to external functions. For these systems, therefore,
51     pcre_malloc etc. are now initialized to local functions that call the
52     relevant global functions.
53 ph10 507
54 ph10 498 7. There were two entries missing in the vectors called coptable and poptable
55     in pcre_dfa_exec.c. This could lead to memory accesses outsize the vectors.
56 ph10 507 I've fixed the data, and added a kludgy way of testing at compile time that
57     the lengths are correct (equal to the number of opcodes).
58    
59     8. Following on from 7, I added a similar kludge to check the length of the
60     eint vector in pcreposix.c.
61    
62     9. Error texts for pcre_compile() are held as one long string to avoid too
63     much relocation at load time. To find a text, the string is searched,
64 ph10 499 counting zeros. There was no check for running off the end of the string,
65     which could happen if a new error number was added without updating the
66 ph10 507 string.
67    
68     10. \K gave a compile-time error if it appeared in a lookbehind assersion.
69    
70 ph10 500 11. \K was not working if it appeared in an atomic group or in a group that
71 ph10 507 was called as a "subroutine", or in an assertion. Perl 5.11 documents that
72     \K is "not well defined" if used in an assertion. PCRE now accepts it if
73     the assertion is positive, but not if it is negative.
74    
75 ph10 501 12. Change 11 fortuitously reduced the size of the stack frame used in the
76 ph10 507 "match()" function of pcre_exec.c by one pointer. Forthcoming
77     implementation of support for (*MARK) will need an extra pointer on the
78 ph10 501 stack; I have reserved it now, so that the stack frame size does not
79     decrease.
80 ph10 507
81     13. A pattern such as (?P<L1>(?P<L2>0)|(?P>L2)(?P>L1)) in which the only other
82     item in branch that calls a recursion is a subroutine call - as in the
83 ph10 503 second branch in the above example - was incorrectly given the compile-
84     time error "recursive call could loop indefinitely" because pcre_compile()
85 ph10 507 was not correctly checking the subroutine for matching a non-empty string.
86    
87 ph10 505 14. The checks for overrunning compiling workspace could trigger after an
88 ph10 507 overrun had occurred. This is a "should never occur" error, but it can be
89 ph10 505 triggered by pathological patterns such as hundreds of nested parentheses.
90 ph10 507 The checks now trigger 100 bytes before the end of the workspace.
91 ph10 491
92 ph10 507 15. Fix typo in configure.ac: "srtoq" should be "strtoq".
93 ph10 492
94 ph10 507
95 ph10 489 Version 8.01 19-Jan-2010
96     ------------------------
97 ph10 471
98     1. If a pattern contained a conditional subpattern with only one branch (in
99 ph10 489 particular, this includes all (*DEFINE) patterns), a call to pcre_study()
100 ph10 473 computed the wrong minimum data length (which is of course zero for such
101 ph10 489 subpatterns). This could cause incorrect "no match" results.
102 ph10 473
103     2. For patterns such as (?i)a(?-i)b|c where an option setting at the start of
104 ph10 472 the pattern is reset in the first branch, pcre_compile() failed with
105     "internal error: code overflow at offset...". This happened only when
106 ph10 473 the reset was to the original external option setting. (An optimization
107     abstracts leading options settings into an external setting, which was the
108 ph10 472 cause of this.)
109 ph10 471
110 ph10 473 3. A pattern such as ^(?!a(*SKIP)b) where a negative assertion contained one
111     of the verbs SKIP, PRUNE, or COMMIT, did not work correctly. When the
112     assertion pattern did not match (meaning that the assertion was true), it
113     was incorrectly treated as false if the SKIP had been reached during the
114     matching. This also applied to assertions used as conditions.
115 ph10 471
116 ph10 473 4. If an item that is not supported by pcre_dfa_exec() was encountered in an
117     assertion subpattern, including such a pattern used as a condition,
118     unpredictable results occurred, instead of the error return
119     PCRE_ERROR_DFA_UITEM.
120 ph10 478
121 ph10 474 5. The C++ GlobalReplace function was not working like Perl for the special
122     situation when an empty string is matched. It now does the fancy magic
123 ph10 478 stuff that is necessary.
124    
125     6. In pcre_internal.h, obsolete includes to setjmp.h and stdarg.h have been
126 ph10 475 removed. (These were left over from very, very early versions of PCRE.)
127 ph10 478
128 ph10 475 7. Some cosmetic changes to the code to make life easier when compiling it
129     as part of something else:
130 ph10 478
131     (a) Change DEBUG to PCRE_DEBUG.
132    
133     (b) In pcre_compile(), rename the member of the "branch_chain" structure
134     called "current" as "current_branch", to prevent a collision with the
135 ph10 475 Linux macro when compiled as a kernel module.
136 ph10 478
137     (c) In pcre_study(), rename the function set_bit() as set_table_bit(), to
138     prevent a collision with the Linux macro when compiled as a kernel
139 ph10 475 module.
140 ph10 478
141 ph10 475 8. In pcre_compile() there are some checks for integer overflows that used to
142     cast potentially large values to (double). This has been changed to that
143 ph10 478 when building, a check for int64_t is made, and if it is found, it is used
144     instead, thus avoiding the use of floating point arithmetic. (There is no
145     other use of FP in PCRE.) If int64_t is not found, the fallback is to
146     double.
147    
148     9. Added two casts to avoid signed/unsigned warnings from VS Studio Express
149 ph10 476 2005 (difference between two addresses compared to an unsigned value).
150 ph10 473
151 ph10 478 10. Change the standard AC_CHECK_LIB test for libbz2 in configure.ac to a
152     custom one, because of the following reported problem in Windows:
153 ph10 473
154 ph10 478 - libbz2 uses the Pascal calling convention (WINAPI) for the functions
155 ph10 489 under Win32.
156 ph10 478 - The standard autoconf AC_CHECK_LIB fails to include "bzlib.h",
157 ph10 489 therefore missing the function definition.
158 ph10 478 - The compiler thus generates a "C" signature for the test function.
159     - The linker fails to find the "C" function.
160     - PCRE fails to configure if asked to do so against libbz2.
161 ph10 486
162 ph10 479 11. When running libtoolize from libtool-2.2.6b as part of autogen.sh, these
163     messages were output:
164 ph10 478
165 ph10 479 Consider adding `AC_CONFIG_MACRO_DIR([m4])' to configure.ac and
166     rerunning libtoolize, to keep the correct libtool macros in-tree.
167     Consider adding `-I m4' to ACLOCAL_AMFLAGS in Makefile.am.
168 ph10 486
169 ph10 479 I have done both of these things.
170 ph10 486
171 ph10 480 12. Although pcre_dfa_exec() does not use nearly as much stack as pcre_exec()
172 ph10 486 most of the time, it *can* run out if it is given a pattern that contains a
173     runaway infinite recursion. I updated the discussion in the pcrestack man
174     page.
175    
176     13. Now that we have gone to the x.xx style of version numbers, the minor
177     version may start with zero. Using 08 or 09 is a bad idea because users
178     might check the value of PCRE_MINOR in their code, and 08 or 09 may be
179     interpreted as invalid octal numbers. I've updated the previous comment in
180     configure.ac, and also added a check that gives an error if 08 or 09 are
181 ph10 481 used.
182 ph10 486
183 ph10 482 14. Change 8.00/11 was not quite complete: code had been accidentally omitted,
184 ph10 483 causing partial matching to fail when the end of the subject matched \W
185     in a UTF-8 pattern where \W was quantified with a minimum of 3.
186 ph10 486
187 ph10 483 15. There were some discrepancies between the declarations in pcre_internal.h
188 ph10 486 of _pcre_is_newline(), _pcre_was_newline(), and _pcre_valid_utf8() and
189 ph10 483 their definitions. The declarations used "const uschar *" and the
190     definitions used USPTR. Even though USPTR is normally defined as "const
191     unsigned char *" (and uschar is typedeffed as "unsigned char"), it was
192     reported that: "This difference in casting confuses some C++ compilers, for
193     example, SunCC recognizes above declarations as different functions and
194 ph10 486 generates broken code for hbpcre." I have changed the declarations to use
195 ph10 483 USPTR.
196 ph10 486
197     16. GNU libtool is named differently on some systems. The autogen.sh script now
198 ph10 484 tries several variants such as glibtoolize (MacOSX) and libtoolize1x
199     (FreeBSD).
200 ph10 478
201 ph10 486 17. Applied Craig's patch that fixes an HP aCC compile error in pcre 8.00
202     (strtoXX undefined when compiling pcrecpp.cc). The patch contains this
203     comment: "Figure out how to create a longlong from a string: strtoll and
204     equivalent. It's not enough to call AC_CHECK_FUNCS: hpux has a strtoll, for
205     instance, but it only takes 2 args instead of 3!"
206 ph10 479
207 ph10 488 18. A subtle bug concerned with back references has been fixed by a change of
208     specification, with a corresponding code fix. A pattern such as
209     ^(xa|=?\1a)+$ which contains a back reference inside the group to which it
210     refers, was giving matches when it shouldn't. For example, xa=xaaa would
211     match that pattern. Interestingly, Perl (at least up to 5.11.3) has the
212     same bug. Such groups have to be quantified to be useful, or contained
213     inside another quantified group. (If there's no repetition, the reference
214     can never match.) The problem arises because, having left the group and
215     moved on to the rest of the pattern, a later failure that backtracks into
216     the group uses the captured value from the final iteration of the group
217     rather than the correct earlier one. I have fixed this in PCRE by forcing
218     any group that contains a reference to itself to be an atomic group; that
219     is, there cannot be any backtracking into it once it has completed. This is
220     similar to recursive and subroutine calls.
221 ph10 486
222 ph10 488
223 ph10 469 Version 8.00 19-Oct-09
224 ph10 418 ----------------------
225    
226     1. The table for translating pcre_compile() error codes into POSIX error codes
227 ph10 461 was out-of-date, and there was no check on the pcre_compile() error code
228     being within the table. This could lead to an OK return being given in
229 ph10 418 error.
230 ph10 461
231     2. Changed the call to open a subject file in pcregrep from fopen(pathname,
232     "r") to fopen(pathname, "rb"), which fixed a problem with some of the tests
233     in a Windows environment.
234    
235 ph10 420 3. The pcregrep --count option prints the count for each file even when it is
236     zero, as does GNU grep. However, pcregrep was also printing all files when
237     --files-with-matches was added. Now, when both options are given, it prints
238     counts only for those files that have at least one match. (GNU grep just
239 ph10 461 prints the file name in this circumstance, but including the count seems
240     more useful - otherwise, why use --count?) Also ensured that the
241 ph10 420 combination -clh just lists non-zero counts, with no names.
242 ph10 461
243     4. The long form of the pcregrep -F option was incorrectly implemented as
244     --fixed_strings instead of --fixed-strings. This is an incompatible change,
245     but it seems right to fix it, and I didn't think it was worth preserving
246     the old behaviour.
247    
248     5. The command line items --regex=pattern and --regexp=pattern were not
249 ph10 422 recognized by pcregrep, which required --regex pattern or --regexp pattern
250 ph10 461 (with a space rather than an '='). The man page documented the '=' forms,
251 ph10 422 which are compatible with GNU grep; these now work.
252 ph10 461
253     6. No libpcreposix.pc file was created for pkg-config; there was just
254 ph10 423 libpcre.pc and libpcrecpp.pc. The omission has been rectified.
255 ph10 461
256 ph10 425 7. Added #ifndef SUPPORT_UCP into the pcre_ucd.c module, to reduce its size
257 ph10 461 when UCP support is not needed, by modifying the Python script that
258 ph10 425 generates it from Unicode data files. This should not matter if the module
259     is correctly used as a library, but I received one complaint about 50K of
260     unwanted data. My guess is that the person linked everything into his
261     program rather than using a library. Anyway, it does no harm.
262 ph10 461
263 ph10 426 8. A pattern such as /\x{123}{2,2}+/8 was incorrectly compiled; the trigger
264 ph10 461 was a minimum greater than 1 for a wide character in a possessive
265 ph10 431 repetition. The same bug could also affect patterns like /(\x{ff}{0,2})*/8
266     which had an unlimited repeat of a nested, fixed maximum repeat of a wide
267     character. Chaos in the form of incorrect output or a compiling loop could
268     result.
269 ph10 461
270 ph10 426 9. The restrictions on what a pattern can contain when partial matching is
271 ph10 461 requested for pcre_exec() have been removed. All patterns can now be
272 ph10 426 partially matched by this function. In addition, if there are at least two
273 ph10 435 slots in the offset vector, the offset of the earliest inspected character
274     for the match and the offset of the end of the subject are set in them when
275 ph10 461 PCRE_ERROR_PARTIAL is returned.
276    
277 ph10 428 10. Partial matching has been split into two forms: PCRE_PARTIAL_SOFT, which is
278     synonymous with PCRE_PARTIAL, for backwards compatibility, and
279     PCRE_PARTIAL_HARD, which causes a partial match to supersede a full match,
280 ph10 462 and may be more useful for multi-segment matching.
281 ph10 461
282     11. Partial matching with pcre_exec() is now more intuitive. A partial match
283     used to be given if ever the end of the subject was reached; now it is
284     given only if matching could not proceed because another character was
285     needed. This makes a difference in some odd cases such as Z(*FAIL) with the
286     string "Z", which now yields "no match" instead of "partial match". In the
287     case of pcre_dfa_exec(), "no match" is given if every matching path for the
288     final character ended with (*FAIL).
289    
290 ph10 428 12. Restarting a match using pcre_dfa_exec() after a partial match did not work
291 ph10 461 if the pattern had a "must contain" character that was already found in the
292 ph10 428 earlier partial match, unless partial matching was again requested. For
293     example, with the pattern /dog.(body)?/, the "must contain" character is
294     "g". If the first part-match was for the string "dog", restarting with
295 ph10 435 "sbody" failed. This bug has been fixed.
296 ph10 461
297     13. The string returned by pcre_dfa_exec() after a partial match has been
298     changed so that it starts at the first inspected character rather than the
299     first character of the match. This makes a difference only if the pattern
300     starts with a lookbehind assertion or \b or \B (\K is not supported by
301     pcre_dfa_exec()). It's an incompatible change, but it makes the two
302 ph10 435 matching functions compatible, and I think it's the right thing to do.
303 ph10 461
304 ph10 435 14. Added a pcredemo man page, created automatically from the pcredemo.c file,
305 ph10 461 so that the demonstration program is easily available in environments where
306     PCRE has not been installed from source.
307    
308 ph10 435 15. Arranged to add -DPCRE_STATIC to cflags in libpcre.pc, libpcreposix.cp,
309 ph10 430 libpcrecpp.pc and pcre-config when PCRE is not compiled as a shared
310     library.
311 ph10 461
312 ph10 435 16. Added REG_UNGREEDY to the pcreposix interface, at the request of a user.
313 ph10 432 It maps to PCRE_UNGREEDY. It is not, of course, POSIX-compatible, but it
314 ph10 461 is not the first non-POSIX option to be added. Clearly some people find
315 ph10 433 these options useful.
316 ph10 461
317     17. If a caller to the POSIX matching function regexec() passes a non-zero
318 ph10 438 value for nmatch with a NULL value for pmatch, the value of
319 ph10 461 nmatch is forced to zero.
320    
321 ph10 437 18. RunGrepTest did not have a test for the availability of the -u option of
322 ph10 461 the diff command, as RunTest does. It now checks in the same way as
323 ph10 437 RunTest, and also checks for the -b option.
324 ph10 461
325 ph10 438 19. If an odd number of negated classes containing just a single character
326     interposed, within parentheses, between a forward reference to a named
327 ph10 461 subpattern and the definition of the subpattern, compilation crashed with
328     an internal error, complaining that it could not find the referenced
329 ph10 438 subpattern. An example of a crashing pattern is /(?&A)(([^m])(?<A>))/.
330 ph10 461 [The bug was that it was starting one character too far in when skipping
331     over the character class, thus treating the ] as data rather than
332     terminating the class. This meant it could skip too much.]
333    
334 ph10 442 20. Added PCRE_NOTEMPTY_ATSTART in order to be able to correctly implement the
335 ph10 461 /g option in pcretest when the pattern contains \K, which makes it possible
336 ph10 442 to have an empty string match not at the start, even when the pattern is
337 ph10 461 anchored. Updated pcretest and pcredemo to use this option.
338    
339 ph10 446 21. If the maximum number of capturing subpatterns in a recursion was greater
340 ph10 461 than the maximum at the outer level, the higher number was returned, but
341     with unset values at the outer level. The correct (outer level) value is
342 ph10 446 now given.
343 ph10 461
344 ph10 447 22. If (*ACCEPT) appeared inside capturing parentheses, previous releases of
345     PCRE did not set those parentheses (unlike Perl). I have now found a way to
346     make it do so. The string so far is captured, making this feature
347     compatible with Perl.
348 ph10 461
349     23. The tests have been re-organized, adding tests 11 and 12, to make it
350 ph10 448 possible to check the Perl 5.10 features against Perl 5.10.
351 ph10 461
352 ph10 454 24. Perl 5.10 allows subroutine calls in lookbehinds, as long as the subroutine
353 ph10 461 pattern matches a fixed length string. PCRE did not allow this; now it
354     does. Neither allows recursion.
355    
356     25. I finally figured out how to implement a request to provide the minimum
357     length of subject string that was needed in order to match a given pattern.
358     (It was back references and recursion that I had previously got hung up
359     on.) This code has now been added to pcre_study(); it finds a lower bound
360 ph10 455 to the length of subject needed. It is not necessarily the greatest lower
361     bound, but using it to avoid searching strings that are too short does give
362     some useful speed-ups. The value is available to calling programs via
363     pcre_fullinfo().
364 ph10 461
365 ph10 455 26. While implementing 25, I discovered to my embarrassment that pcretest had
366     not been passing the result of pcre_study() to pcre_dfa_exec(), so the
367     study optimizations had never been tested with that matching function.
368     Oops. What is worse, even when it was passed study data, there was a bug in
369     pcre_dfa_exec() that meant it never actually used it. Double oops. There
370     were also very few tests of studied patterns with pcre_dfa_exec().
371 ph10 461
372 ph10 457 27. If (?| is used to create subpatterns with duplicate numbers, they are now
373     allowed to have the same name, even if PCRE_DUPNAMES is not set. However,
374     on the other side of the coin, they are no longer allowed to have different
375     names, because these cannot be distinguished in PCRE, and this has caused
376     confusion. (This is a difference from Perl.)
377 ph10 461
378     28. When duplicate subpattern names are present (necessarily with different
379     numbers, as required by 27 above), and a test is made by name in a
380     conditional pattern, either for a subpattern having been matched, or for
381     recursion in such a pattern, all the associated numbered subpatterns are
382 ph10 459 tested, and the overall condition is true if the condition is true for any
383     one of them. This is the way Perl works, and is also more like the way
384     testing by number works.
385 ph10 418
386 ph10 461
387 ph10 412 Version 7.9 11-Apr-09
388 ph10 376 ---------------------
389    
390 ph10 392 1. When building with support for bzlib/zlib (pcregrep) and/or readline
391 ph10 376 (pcretest), all targets were linked against these libraries. This included
392     libpcre, libpcreposix, and libpcrecpp, even though they do not use these
393     libraries. This caused unwanted dependencies to be created. This problem
394 ph10 392 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
395 ph10 376 pcretest is linked with readline.
396 ph10 392
397 ph10 376 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
398     "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
399     moved outside it again, because FALSE and TRUE are already defined in AIX,
400     but BOOL is not.
401 ph10 392
402     3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
403 ph10 409 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
404 ph10 392
405 ph10 376 4. The pcregrep documentation said spaces were inserted as well as colons (or
406     hyphens) following file names and line numbers when outputting matching
407 ph10 392 lines. This is not true; no spaces are inserted. I have also clarified the
408 ph10 376 wording for the --colour (or --color) option.
409 ph10 392
410 ph10 378 5. In pcregrep, when --colour was used with -o, the list of matching strings
411     was not coloured; this is different to GNU grep, so I have changed it to be
412     the same.
413 ph10 392
414 ph10 378 6. When --colo(u)r was used in pcregrep, only the first matching substring in
415 ph10 392 each matching line was coloured. Now it goes on to look for further matches
416     of any of the test patterns, which is the same behaviour as GNU grep.
417    
418     7. A pattern that could match an empty string could cause pcregrep to loop; it
419     doesn't make sense to accept an empty string match in pcregrep, so I have
420 ph10 379 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
421     seems to be how GNU grep behaves.
422 ph10 392
423 ph10 380 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
424 ph10 392 start or after a newline", because the conditional assertion was not being
425     correctly handled. The rule now is that both the assertion and what follows
426     in the first alternative must satisfy the test.
427    
428 ph10 405 9. If auto-callout was enabled in a pattern with a conditional group whose
429 ph10 399 condition was an assertion, PCRE could crash during matching, both with
430     pcre_exec() and pcre_dfa_exec().
431 ph10 392
432     10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
433     used for matching.
434    
435     11. Unicode property support in character classes was not working for
436 ph10 385 characters (bytes) greater than 127 when not in UTF-8 mode.
437 ph10 388
438 ph10 392 12. Added the -M command line option to pcretest.
439    
440 ph10 388 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
441 ph10 389
442     15. Added the PCRE_NO_START_OPTIMIZE match-time option.
443 ph10 390
444 ph10 392 16. Added comments and documentation about mis-use of no_arg in the C++
445 ph10 390 wrapper.
446 ph10 392
447 ph10 391 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
448 ph10 392 from Martin Jerabek that uses macro names for all relevant character and
449 ph10 391 string constants.
450 ph10 405
451 ph10 393 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
452     SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
453 ph10 405 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
454     these, but not everybody uses configure.
455    
456     19. A conditional group that had only one branch was not being correctly
457     recognized as an item that could match an empty string. This meant that an
458     enclosing group might also not be so recognized, causing infinite looping
459     (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
460     with the subject "ab", where knowledge that the repeated group can match
461 ph10 396 nothing is needed in order to break the loop.
462 ph10 405
463 ph10 397 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
464 ph10 405 exec(), but without supplying a callout function, matching went wrong.
465    
466     21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
467     leak if the size of the offset vector was greater than 30. When the vector
468     is smaller, the saved offsets during recursion go onto a local stack
469     vector, but for larger vectors malloc() is used. It was failing to free
470     when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
471 ph10 400 error, in fact).
472 ph10 405
473 ph10 402 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
474 ph10 405 heapframe that is used only when UTF-8 support is enabled. This caused no
475     problem, but was untidy.
476 ph10 376
477 ph10 405 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
478     CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
479     included within another project.
480 ph10 411
481 ph10 407 24. Steven Van Ingelgem's patches to add more options to the CMake support,
482     slightly modified by me:
483 ph10 411
484     (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
485     not building pcregrep.
486    
487 ph10 407 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
488 ph10 411 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
489    
490 ph10 408 25. Forward references, both numeric and by name, in patterns that made use of
491     duplicate group numbers, could behave incorrectly or give incorrect errors,
492     because when scanning forward to find the reference group, PCRE was not
493     taking into account the duplicate group numbers. A pattern such as
494     ^X(?3)(a)(?|(b)|(q))(Y) is an example.
495 ph10 411
496     26. Changed a few more instances of "const unsigned char *" to USPTR, making
497     the feature of a custom pointer more persuasive (as requested by a user).
498 ph10 416
499 ph10 411 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
500 ph10 416 pcretest.c, inside #ifndefs, because it seems they are sometimes already
501     pre-defined.
502 ph10 392
503 ph10 416 28. Added support for (*UTF8) at the start of a pattern.
504    
505 ph10 413 29. Arrange for flags added by the "release type" setting in CMake to be shown
506     in the configuration summary.
507 ph10 405
508 ph10 411
509 ph10 374 Version 7.8 05-Sep-08
510 ph10 349 ---------------------
511    
512     1. Replaced UCP searching code with optimized version as implemented for Ad
513     Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
514 ph10 351 stage table and inline lookup instead of a function, giving speed ups of 2
515     to 5 times on some simple patterns that I tested. Permission was given to
516     distribute the MultiStage2.py script that generates the tables (it's not in
517     the tarball, but is in the Subversion repository).
518 ph10 349
519 ph10 351 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
520     scripts.
521 ph10 358
522 ph10 353 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
523     a group with a zero qualifier. The result of the study could be incorrect,
524 ph10 358 or the function might crash, depending on the pattern.
525    
526     4. Caseless matching was not working for non-ASCII characters in back
527 ph10 354 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
528 ph10 358 It now works when Unicode Property Support is available.
529    
530 ph10 355 5. In pcretest, an escape such as \x{de} in the data was always generating
531     a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
532     non-UTF-8 mode. If the value is greater than 255, it gives a warning about
533 ph10 358 truncation.
534 ph10 349
535 ph10 358 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
536    
537 ph10 357 7. Added two (int) casts to pcregrep when printing the difference of two
538     pointers, in case they are 64-bit values.
539 ph10 371
540     8. Added comments about Mac OS X stack usage to the pcrestack man page and to
541     test 2 if it fails.
542    
543 ph10 359 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
544 ph10 371 and a #define of that name to empty if it is not externally set. This is to
545     allow users of MSVC to set it if necessary.
546    
547     10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
548 ph10 359 the convenience functions in the pcre_get.c source file.
549 ph10 371
550 ph10 360 11. An option change at the start of a pattern that had top-level alternatives
551     could cause overwriting and/or a crash. This command provoked a crash in
552 ph10 371 some environments:
553    
554     printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
555    
556 ph10 360 This potential security problem was recorded as CVE-2008-2371.
557 ph10 371
558 ph10 361 12. For a pattern where the match had to start at the beginning or immediately
559     after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
560 ph10 371 pcre_dfa_exec() could read past the end of the passed subject if there was
561 ph10 361 no match. To help with detecting such bugs (e.g. with valgrind), I modified
562     pcretest so that it places the subject at the end of its malloc-ed buffer.
563 ph10 371
564 ph10 364 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
565 ph10 371 exec() might read past the end of the data buffer in UTF-8 mode.
566    
567 ph10 364 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
568 ph10 371 the data contained the byte 0x85 as part of a UTF-8 character within its
569     first line. This applied both to normal and DFA matching.
570    
571 ph10 366 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
572 ph10 371 /^[^d]*?$/8 failed to match "abc".
573 ph10 351
574 ph10 371 16. Added a missing copyright notice to pcrecpp_internal.h.
575    
576     17. Make it more clear in the documentation that values returned from
577 ph10 368 pcre_exec() in ovector are byte offsets, not character counts.
578 ph10 357
579 ph10 371 18. Tidied a few places to stop certain compilers from issuing warnings.
580 ph10 368
581 ph10 373 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
582     supplied by Stefan Weber. I made a further small update for 7.8 because
583 ph10 374 there is a change of source arrangements: the pcre_searchfuncs.c module is
584     replaced by pcre_ucd.c.
585 ph10 371
586 ph10 373
587 ph10 347 Version 7.7 07-May-08
588 ph10 321 ---------------------
589    
590     1. Applied Craig's patch to sort out a long long problem: "If we can't convert
591 ph10 345 a string to a long long, pretend we don't even have a long long." This is
592 ph10 321 done by checking for the strtoq, strtoll, and _strtoi64 functions.
593 ph10 345
594 ph10 322 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
595     pre-7.6 versions, which defined a global no_arg variable instead of putting
596 ph10 332 it in the RE class. (See also #8 below.)
597 ph10 345
598     3. Remove a line of dead code, identified by coverity and reported by Nuno
599     Lopes.
600    
601 ph10 324 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
602 ph10 321
603 ph10 324 (1) The include/exclude patterns were being applied to the whole pathnames
604 ph10 345 of files, instead of just to the final components.
605    
606 ph10 324 (2) If there was more than one level of directory, the subdirectories were
607     skipped unless they satisfied the include/exclude conditions. This is
608     inconsistent with GNU grep (and could even be seen as contrary to the
609     pcregrep specification - which I improved to make it absolutely clear).
610     The action now is always to scan all levels of directory, and just
611     apply the include/exclude patterns to regular files.
612 ph10 345
613 ph10 325 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
614 ph10 345 --exclude_dir in the tests to avoid scanning .svn directories.
615    
616 ph10 326 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
617 ph10 345 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
618 ph10 326 doesn't support NULs in patterns.
619 ph10 345
620     7. Added some missing "const"s to declarations of static tables in
621     pcre_compile.c and pcre_dfa_exec.c.
622    
623 ph10 328 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
624 ph10 329 caused by fix #2 above. (Subsequently also a second patch to fix the
625 ph10 345 first patch. And a third patch - this was a messy problem.)
626 ph10 321
627 ph10 345 9. Applied Craig's patch to remove the use of push_back().
628    
629 ph10 332 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
630     matching function regexec().
631 ph10 345
632     11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
633     which, however, unlike Perl's \g{...}, are subroutine calls, not back
634 ph10 333 references. PCRE supports relative numbers with this syntax (I don't think
635     Oniguruma does).
636 ph10 345
637     12. Previously, a group with a zero repeat such as (...){0} was completely
638 ph10 335 omitted from the compiled regex. However, this means that if the group
639     was called as a subroutine from elsewhere in the pattern, things went wrong
640 ph10 345 (an internal error was given). Such groups are now left in the compiled
641     pattern, with a new opcode that causes them to be skipped at execution
642 ph10 335 time.
643 ph10 345
644 ph10 341 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
645     to the way PCRE behaves:
646 ph10 345
647 ph10 336 (a) A lone ] character is dis-allowed (Perl treats it as data).
648 ph10 345
649     (b) A back reference to an unmatched subpattern matches an empty string
650 ph10 336 (Perl fails the current match path).
651 ph10 345
652 ph10 341 (c) A data ] in a character class must be notated as \] because if the
653 ph10 345 first data character in a class is ], it defines an empty class. (In
654 ph10 341 Perl it is not possible to have an empty class.) The empty class []
655     never matches; it forces failure and is equivalent to (*FAIL) or (?!).
656     The negative empty class [^] matches any one character, independently
657     of the DOTALL setting.
658 ph10 345
659     14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
660 ph10 340 non-existent subpattern following a character class starting with ']' and
661     containing () gave an internal compiling error instead of "reference to
662     non-existent subpattern". Fortunately, when the pattern did exist, the
663 ph10 345 compiled code was correct. (When scanning forwards to check for the
664 ph10 340 existencd of the subpattern, it was treating the data ']' as terminating
665 ph10 345 the class, so got the count wrong. When actually compiling, the reference
666 ph10 340 was subsequently set up correctly.)
667 ph10 345
668 ph10 341 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
669 ph10 345 it was being rejected as not supported by pcre_dfa_exec(), even though
670     other assertions are supported. I have made pcre_dfa_exec() support
671     (*FAIL).
672    
673 ph10 342 16. The implementation of 13c above involved the invention of a new opcode,
674 ph10 345 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
675     cannot be changed at match time, I realized I could make a small
676     improvement to matching performance by compiling OP_ALLANY instead of
677     OP_ANY for "." when DOTALL was set, and then removing the runtime tests
678     on the OP_ANY path.
679    
680     17. Compiling pcretest on Windows with readline support failed without the
681 ph10 343 following two fixes: (1) Make the unistd.h include conditional on
682 ph10 345 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
683    
684 ph10 344 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
685 ph10 345 ncurses library to be included for pcretest when ReadLine support is
686     requested, but also to allow for it to be overridden. This patch came from
687     Daniel Bergström.
688 ph10 347
689 ph10 346 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
690 ph10 347 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
691     any errors with the current Unicode tables. Thanks to Peter Kankowski for
692     spotting this.
693 ph10 324
694 ph10 332
695 ph10 319 Version 7.6 28-Jan-08
696 ph10 300 ---------------------
697    
698 ph10 302 1. A character class containing a very large number of characters with
699     codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
700     overflow.
701 ph10 309
702     2. Patch to cut out the "long long" test in pcrecpp_unittest when
703     HAVE_LONG_LONG is not defined.
704    
705 ph10 303 3. Applied Christian Ehrlicher's patch to update the CMake build files to
706 ph10 304 bring them up to date and include new features. This patch includes:
707 ph10 309
708 ph10 304 - Fixed PH's badly added libz and libbz2 support.
709     - Fixed a problem with static linking.
710 ph10 312 - Added pcredemo. [But later removed - see 7 below.]
711 ph10 304 - Fixed dftables problem and added an option.
712     - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
713     HAVE_LONG_LONG.
714     - Added readline support for pcretest.
715 ph10 309 - Added an listing of the option settings after cmake has run.
716    
717 ph10 314 4. A user submitted a patch to Makefile that makes it easy to create
718     "pcre.dll" under mingw when using Configure/Make. I added stuff to
719     Makefile.am that cause it to include this special target, without
720     affecting anything else. Note that the same mingw target plus all
721     the other distribution libraries and programs are now supported
722     when configuring with CMake (see 6 below) instead of with
723     Configure/Make.
724 ph10 309
725 ph10 308 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
726     This is an attempt to solve the reported problem "pcrecpp::no_arg is not
727 ph10 309 exported in the Windows port". It has not yet been confirmed that the patch
728     solves the problem, but it does no harm.
729 ph10 313
730 ph10 311 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
731 ph10 319 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
732 ph10 314 with CMake, and also correct the comment about stack recursion.
733 ph10 313
734 ph10 312 7. Remove the automatic building of pcredemo from the ./configure system and
735 ph10 313 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
736     of a program that users should build themselves after PCRE is installed, so
737     building it automatically is not really right. What is more, it gave
738 ph10 312 trouble in some build environments.
739 ph10 300
740 ph10 319 8. Further tidies to CMakeLists.txt from Sheri and Christian.
741 ph10 308
742 ph10 319
743 ph10 298 Version 7.5 10-Jan-08
744 ph10 263 ---------------------
745    
746     1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
747     values in parens when parsing an RE using the C++ wrapper."
748 ph10 286
749 ph10 264 2. Negative specials like \S did not work in character classes in UTF-8 mode.
750     Characters greater than 255 were excluded from the class instead of being
751     included.
752 ph10 286
753     3. The same bug as (2) above applied to negated POSIX classes such as
754 ph10 265 [:^space:].
755 ph10 286
756 ph10 267 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
757     defined or documented. It seems to have been a typo for PCRE_STATIC, so
758 ph10 286 I have changed it.
759    
760 ph10 268 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
761 ph10 286 first named subpattern) and a construct such as (?&a) would reference the
762     first named subpattern whose name started with "a" (in other words, the
763 ph10 272 length check was missing). Both these problems are fixed. "Subpattern name
764     expected" is now given for (?&) (a zero-length name), and this patch also
765     makes it give the same error for \k'' (previously it complained that that
766     was a reference to a non-existent subpattern).
767 ph10 286
768 ph10 269 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
769 ph10 286 this is right because (?- can be followed by option settings as well as by
770 ph10 269 digits. I have, however, made the messages clearer.
771 ph10 286
772 ph10 270 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
773     than the number used in the conditional) now cause a compile-time error.
774     This is actually not compatible with Perl, which accepts such patterns, but
775     treats the conditional as always being FALSE (as PCRE used to), but it
776     seems to me that giving a diagnostic is better.
777 ph10 286
778 ph10 274 8. Change "alphameric" to the more common word "alphanumeric" in comments
779 ph10 275 and messages.
780 ph10 286
781     9. Fix two occurrences of "backslash" in comments that should have been
782     "backspace".
783    
784     10. Remove two redundant lines of code that can never be obeyed (their function
785     was moved elsewhere).
786    
787     11. The program that makes PCRE's Unicode character property table had a bug
788     which caused it to generate incorrect table entries for sequences of
789 ph10 277 characters that have the same character type, but are in different scripts.
790 ph10 286 It amalgamated them into a single range, with the script of the first of
791 ph10 277 them. In other words, some characters were in the wrong script. There were
792     thirteen such cases, affecting characters in the following ranges:
793 ph10 286
794 ph10 277 U+002b0 - U+002c1
795     U+0060c - U+0060d
796 ph10 286 U+0061e - U+00612
797 ph10 277 U+0064b - U+0065e
798     U+0074d - U+0076d
799     U+01800 - U+01805
800     U+01d00 - U+01d77
801     U+01d9b - U+01dbf
802     U+0200b - U+0200f
803     U+030fc - U+030fe
804     U+03260 - U+0327f
805     U+0fb46 - U+0fbb1
806     U+10450 - U+1049d
807 ph10 286
808 ph10 279 12. The -o option (show only the matching part of a line) for pcregrep was not
809     compatible with GNU grep in that, if there was more than one match in a
810     line, it showed only the first of them. It now behaves in the same way as
811     GNU grep.
812 ph10 286
813     13. If the -o and -v options were combined for pcregrep, it printed a blank
814     line for every non-matching line. GNU grep prints nothing, and pcregrep now
815     does the same. The return code can be used to tell if there were any
816     non-matching lines.
817    
818 ph10 289 14. Added --file-offsets and --line-offsets to pcregrep.
819    
820     15. The pattern (?=something)(?R) was not being diagnosed as a potentially
821 ph10 282 infinitely looping recursion. The bug was that positive lookaheads were not
822 ph10 286 being skipped when checking for a possible empty match (negative lookaheads
823     and both kinds of lookbehind were skipped).
824    
825 ph10 289 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
826 ph10 284 inclusion of <windows.h> to before rather than after the definition of
827     INVALID_FILE_ATTRIBUTES (patch from David Byron).
828 ph10 263
829 ph10 289 17. Specifying a possessive quantifier with a specific limit for a Unicode
830 ph10 286 character property caused pcre_compile() to compile bad code, which led at
831     runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
832     are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
833     caused the error; without that there was no problem.
834 ph10 263
835 ph10 289 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
836 ph10 286
837 ph10 289 19. Added --enable-pcretest-libreadline.
838 ph10 286
839 ph10 289 20. In pcrecpp.cc, the variable 'count' was incremented twice in
840 ph10 288 RE::GlobalReplace(). As a result, the number of replacements returned was
841 ph10 298 double what it should be. I removed one of the increments, but Craig sent a
842     later patch that removed the other one (the right fix) and added unit tests
843     that check the return values (which was not done before).
844 ph10 292
845 ph10 291 21. Several CMake things:
846 ph10 287
847 ph10 291 (1) Arranged that, when cmake is used on Unix, the libraries end up with
848 ph10 292 the names libpcre and libpcreposix, not just pcre and pcreposix.
849 ph10 288
850 ph10 292 (2) The above change means that pcretest and pcregrep are now correctly
851     linked with the newly-built libraries, not previously installed ones.
852 ph10 291
853 ph10 292 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
854 ph10 298
855     22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
856     crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
857     UTF-8 newline character). The key issue is that the pattern starts .*;
858     this means that the match must be either at the beginning, or after a
859     newline. The bug was in the code for advancing after a failed match and
860     checking that the new position followed a newline. It was not taking
861 ph10 294 account of UTF-8 characters correctly.
862 ph10 298
863     23. PCRE was behaving differently from Perl in the way it recognized POSIX
864     character classes. PCRE was not treating the sequence [:...:] as a
865     character class unless the ... were all letters. Perl, however, seems to
866     allow any characters between [: and :], though of course it rejects as
867     unknown any "names" that contain non-letters, because all the known class
868     names consist only of letters. Thus, Perl gives an error for [[:1234:]],
869     for example, whereas PCRE did not - it did not recognize a POSIX character
870     class. This seemed a bit dangerous, so the code has been changed to be
871     closer to Perl. The behaviour is not identical to Perl, because PCRE will
872     diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
873     treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
874     Perl does, and where it didn't before.
875    
876 ph10 296 24. Rewrite so as to remove the single use of %n from pcregrep because in some
877     Windows environments %n is disabled by default.
878 ph10 292
879    
880 ph10 260 Version 7.4 21-Sep-07
881 ph10 230 ---------------------
882    
883     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
884 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
885     LF". That isn't really right - the whole point of the change was to try to
886     help when there was an actual mention of one of the two characters. So now
887     the change happens only if \r or \n (or a literal CR or LF) character is
888 ph10 230 encountered.
889 ph10 231
890     2. The 32-bit options word was also used for 6 internal flags, but the numbers
891     of both had grown to the point where there were only 3 bits left.
892     Fortunately, there was spare space in the data structure, and so I have
893     moved the internal flags into a new 16-bit field to free up more option
894 ph10 230 bits.
895 ph10 231
896     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
897     but did not set the internal JCHANGED flag - either of these is enough to
898     control the way the "get" function works - but the PCRE_INFO_JCHANGED
899     facility is supposed to tell if (?J) was ever used, so now (?J) at the
900 ph10 230 start sets both bits.
901    
902 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
903     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
904 ph10 230
905 ph10 243 5. doc/pcresyntax.html was missing from the distribution.
906    
907     6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
908 ph10 233 compatibility, even though it is no longer used.
909 ph10 243
910     7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
911 ph10 254 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
912 ph10 259 windows.h file is present (where different names are used). [This was
913 ph10 257 reversed later after testing - see 16 below.]
914 ph10 243
915     8. Changed all #include <config.h> to #include "config.h". There were also
916 ph10 236 some further <pcre.h> cases that I changed to "pcre.h".
917 ph10 243
918 ph10 239 9. When pcregrep was used with the --colour option, it missed the line ending
919 ph10 243 sequence off the lines that it output.
920    
921     10. It was pointed out to me that arrays of string pointers cause lots of
922     relocations when a shared library is dynamically loaded. A technique of
923     using a single long string with a table of offsets can drastically reduce
924     these. I have refactored PCRE in four places to do this. The result is
925 ph10 240 dramatic:
926 ph10 243
927 ph10 240 Originally: 290
928     After changing UCP table: 187
929 ph10 243 After changing error message table: 43
930 ph10 240 After changing table of "verbs" 36
931     After changing table of Posix names 22
932 ph10 243
933 ph10 240 Thanks to the folks working on Gregex for glib for this insight.
934 ph10 247
935 ph10 244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
936 ph10 247 unicode-properties was also set.
937    
938 ph10 248 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
939 ph10 231
940 ph10 253 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
941     checked only for CRLF.
942 ph10 233
943 ph10 259 14. Added casts to pcretest.c to avoid compiler warnings.
944    
945 ph10 256 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
946 ph10 253
947 ph10 257 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
948 ph10 259 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
949     entirely. This removes changes made in 7 above.
950 ph10 256
951 ph10 261 17. The CMake files have been updated, and there is now more information about
952     building with CMake in the NON-UNIX-USE document.
953 ph10 257
954 ph10 261
955 ph10 228 Version 7.3 28-Aug-07
956 ph10 157 ---------------------
957    
958 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
959     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
960     brackets there is not right, since it causes compilers to look for an
961     installed pcre.h, not the version that is in the source that is being
962     compiled (which of course may be different). I have changed it back to:
963 ph10 197
964 ph10 189 #include "pcre.h"
965 ph10 197
966     I have a vague recollection that the change was concerned with compiling in
967     different directories, but in the new build system, that is taken care of
968     by the VPATH setting the Makefile.
969    
970 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
971     when the subject happened to end in the byte 0x85 (e.g. if the last
972     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
973     characters but of course it shouldn't be taken as a newline when it is part
974     of another character. The bug was that, for an unlimited repeat of . in
975     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
976     characters when looking for a newline.
977 ph10 197
978     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
979    
980     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
981     in debug output.
982    
983 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
984     long printing in the pcrecpp unittest when running under MinGW.
985 ph10 189
986 ph10 197 6. ESC_K was left out of the EBCDIC table.
987 ph10 189
988 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
989     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
990     limit also applies to "virtual nesting" when a pattern is recursive, and in
991     this case 1000 isn't so big. I have been able to remove this limit at the
992     expense of backing off one optimization in certain circumstances. Normally,
993     when pcre_exec() would call its internal match() function recursively and
994     immediately return the result unconditionally, it uses a "tail recursion"
995     feature to save stack. However, when a subpattern that can match an empty
996     string has an unlimited repetition quantifier, it no longer makes this
997     optimization. That gives it a stack frame in which to save the data for
998     checking that an empty string has been matched. Previously this was taken
999     from the 1000-entry workspace that had been reserved. So now there is no
1000     explicit limit, but more stack is used.
1001 ph10 189
1002 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
1003     syntax that is required for Windows, and which was going wrong for the
1004     pcreposix and pcrecpp parts of the library. These were overlooked when this
1005     problem was solved for the main library.
1006 ph10 197
1007 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
1008     the size of patterns that contain repeated groups with explicit upper
1009     limits. As the maximum quantifier is 65535, the maximum group length was
1010     set at 30,000 so that the product of these two numbers did not overflow a
1011     32-bit integer. However, it turns out that people want to use groups that
1012     are longer than 30,000 bytes (though not repeat them that many times).
1013     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
1014     made it possible to implement the integer overflow checks in a much more
1015     dynamic way, which I have now done. The artificial limitation on group
1016     length has been removed - we now have only the limit on the total length of
1017     the compiled pattern, which depends on the LINK_SIZE setting.
1018 ph10 208
1019     10. Fixed a bug in the documentation for get/copy named substring when
1020     duplicate names are permitted. If none of the named substrings are set, the
1021     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
1022     empty string.
1023    
1024     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
1025     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
1026     because the ] is interpreted as the first data character and the
1027     terminating ] is not found. PCRE has been made compatible with Perl in this
1028     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
1029     cause memory overwriting.
1030    
1031 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
1032     string has been matched (to stop an infinite loop). It was not recognizing
1033 ph10 208 a conditional subpattern that could match an empty string if that
1034 ph10 206 subpattern was within another subpattern. For example, it looped when
1035 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
1036 ph10 206 condition was not nested. This bug has been fixed.
1037 ph10 208
1038 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
1039     past the start of the subject in the presence of bytes with the top bit
1040     set, for example "\x8aBCD".
1041 ph10 211
1042 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
1043     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
1044 ph10 200
1045 ph10 211 14. Optimized (?!) to (*FAIL).
1046 ph10 202
1047 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
1048     This restricts code points to be within the range 0 to 0x10FFFF, excluding
1049     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
1050     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
1051     does: it's just the validity check that is more restrictive.
1052 ph10 220
1053     16. Inserted checks for integer overflows during escape sequence (backslash)
1054     processing, and also fixed erroneous offset values for syntax errors during
1055     backslash processing.
1056    
1057 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
1058 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
1059    
1060 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
1061     caused an overrun.
1062 ph10 220
1063     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
1064     something other than just ASCII characters) inside a group that had an
1065     unlimited repeat caused a loop at compile time (while checking to see
1066     whether the group could match an empty string).
1067    
1068 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
1069     [\P{Any}] did so. (Error in the code for printing property names.)
1070 ph10 210
1071 ph10 220 21. An orphan \E inside a character class could cause a crash.
1072    
1073     22. A repeated capturing bracket such as (A)? could cause a wild memory
1074 ph10 218 reference during compilation.
1075 ph10 220
1076     23. There are several functions in pcre_compile() that scan along a compiled
1077     expression for various reasons (e.g. to see if it's fixed length for look
1078 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
1079 ph10 220 present in the pattern. These operators have additional parameters compared
1080     with \d, etc, and these were not being taken into account when moving along
1081 ph10 218 the compiled data. Specifically:
1082 ph10 220
1083     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
1084     length.
1085    
1086     (b) An item such as \pL+ within a repeated group could cause crashes or
1087 ph10 218 loops.
1088 ph10 220
1089 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
1090 ph10 220 "reference to non-existent subpattern" error.
1091    
1092 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
1093    
1094 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
1095 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
1096 ph10 211
1097 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
1098 ph10 218
1099 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
1100 ph10 223 character were causing crashes (broken optimization).
1101 ph10 227
1102 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
1103 ph10 227 \p or \P) caused a compile-time loop.
1104    
1105 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
1106     break. For example, the unstudied pattern [\r\n]A does not match the string
1107     "\r\nA" because change 7.0/46 below moves the current point on by two
1108     characters after failing to match at the start. However, the pattern \nA
1109 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
1110     the same is true. There doesn't seem any very clean way out of this, but
1111     what I have chosen to do makes the common cases work: PCRE now takes note
1112     of whether there can be an explicit match for \r or \n anywhere in the
1113     pattern, and if so, 7.0/46 no longer applies. As part of this change,
1114     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
1115     pattern has explicit CR or LF references.
1116 ph10 222
1117 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
1118 ph10 223
1119 ph10 227
1120 ph10 189 Version 7.2 19-Jun-07
1121     ---------------------
1122    
1123 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
1124     which is apparently normally available under Windows.
1125 ph10 159
1126     2. Re-jig the pcregrep tests with different newline settings in an attempt
1127     to make them independent of the local environment's newline setting.
1128 ph10 157
1129 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
1130 ph10 165
1131 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
1132     was not 2, because the output contained actual offsets. The recent new
1133 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
1134     usable with all link sizes.
1135    
1136 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
1137     stack recursion. This gives a massive performance boost under BSD, but just
1138 ph10 165 a small improvement under Linux. However, it saves one field in the frame
1139 ph10 164 in all cases.
1140 ph10 172
1141 ph10 166 6. Added more features from the forthcoming Perl 5.10:
1142 ph10 172
1143 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
1144     recursion call. It refers to the nth most recently opened parentheses.
1145 ph10 172
1146 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
1147 ph10 172 to be opened parentheses.
1148    
1149     (c) Conditions that refer to capturing parentheses can be specified
1150 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
1151 ph10 172
1152 ph10 168 (d) \K resets the start of the current match so that everything before
1153 ph10 172 is not part of it.
1154    
1155 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
1156 ph10 172
1157 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
1158 ph10 172 reference syntax.
1159 ph10 182
1160 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
1161 ph10 182 alternative starts with the same number.
1162 ph10 172
1163 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
1164    
1165 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
1166     PCRE_INFO_JCHANGED.
1167    
1168     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
1169     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
1170 ph10 170 for detecting groups that can match an empty string.
1171 ph10 159
1172 ph10 172 9. A pattern with a very large number of alternatives (more than several
1173     hundred) was running out of internal workspace during the pre-compile
1174     phase, where pcre_compile() figures out how much memory will be needed. A
1175     bit of new cunning has reduced the workspace needed for groups with
1176     alternatives. The 1000-alternative test pattern now uses 12 bytes of
1177     workspace instead of running out of the 4096 that are available.
1178 ph10 182
1179 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
1180 ph10 172
1181 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
1182     The report of the bug said:
1183 ph10 182
1184 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
1185     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
1186 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
1187 ph10 185
1188 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
1189 ph10 185 it matched the wrong number of bytes.
1190 ph10 172
1191 ph10 179
1192 ph10 155 Version 7.1 24-Apr-07
1193 ph10 98 ---------------------
1194    
1195 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
1196 ph10 122 that is more "standard", making use of automake and other Autotools. There
1197 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
1198     on this.
1199 ph10 111
1200     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
1201     for recursive directory scanning broke on some systems because the files
1202     are not scanned in any specific order and on different systems the order
1203     was different. A call to "sort" has been inserted into RunGrepTest for the
1204     approprate test as a short-term fix. In the longer term there may be an
1205 ph10 100 alternative.
1206 ph10 111
1207 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
1208 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
1209     people do their own conversions for various reasons). The problems
1210     concerned the use of low-level troff macros .br and .in. I have therefore
1211     removed all such uses from the man pages (some were redundant, some could
1212 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
1213     HTML has been updated to handle .nf/.fi and to complain if it encounters
1214     .br or .in.
1215 ph10 111
1216 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
1217 ph10 123 arranged for config.h to be included in the distribution, with the name
1218 ph10 111 config.h.generic, for the benefit of those who have to compile without
1219     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
1220    
1221     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
1222     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
1223 ph10 127 makevp.bat for the current PCRE, using the additional files
1224 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
1225 ph10 111
1226     6. A Windows user reported a minor discrepancy with test 2, which turned out
1227     to be caused by a trailing space on an input line that had got lost in his
1228 ph10 102 copy. The trailing space was an accident, so I've just removed it.
1229 ph10 111
1230 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
1231 ph10 111 that is needed.
1232    
1233 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
1234     as "const" (a) because they are and (b) because it helps the PHP
1235     maintainers who have recently made a script to detect big data structures
1236 ph10 111 in the php code that should be moved to the .rodata section. I remembered
1237     to update Builducptable as well, so it won't revert if ucptable.h is ever
1238 ph10 105 re-created.
1239 ph10 111
1240     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
1241 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
1242 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
1243     support is not required. This saves 1.5-2K of code, which is important in
1244 ph10 107 some applications.
1245 ph10 111
1246 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
1247 ph10 111 so as not to refer to the tables, even though these functions will never be
1248     called when UTF-8 support is disabled. Otherwise there are problems with a
1249     shared library.
1250    
1251 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
1252    
1253     (a) It was defining its arguments as char * instead of void *.
1254    
1255 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
1256     a long time ago when I wrote it, but is no longer the case.
1257    
1258 ph10 118 The emulated memove() is provided for those environments that have neither
1259 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
1260 ph10 118 is clearly not the case, as these two bugs were recently reported.
1261 ph10 123
1262 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
1263 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
1264     pages, and it removes trailing spaces from listed files. It also creates
1265     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
1266     case, it wraps all the #defines with #ifndefs. This script should be run
1267 ph10 111 before "make dist".
1268 ph10 123
1269 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
1270     with Unicode property support.
1271 ph10 123
1272     (a) For a maximizing quantifier, if the two different cases of the
1273     character were of different lengths in their UTF-8 codings (there are
1274     some cases like this - I found 11), and the matching function had to
1275 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
1276     were both the same length.
1277 ph10 123
1278     (b) When PCRE was configured to use the heap rather than the stack for
1279     recursion during matching, it was not correctly preserving the data for
1280     the other case of a UTF-8 character when checking ahead for a match
1281     while processing a minimizing repeat. If the check also involved
1282 ph10 115 matching a wide character, but failed, corruption could cause an
1283     erroneous result when trying to check for a repeat of the original
1284     character.
1285 ph10 123
1286 ph10 116 13. Some tidying changes to the testing mechanism:
1287 ph10 98
1288 ph10 116 (a) The RunTest script now detects the internal link size and whether there
1289     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
1290 ph10 123 values substituted by "configure". (The RunGrepTest script already did
1291     this for UTF-8.) The configure.ac script no longer substitutes the
1292     relevant variables.
1293    
1294 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
1295     with length and offset values. This means that the output is different
1296     for different internal link sizes. Test 2 is skipped for link sizes
1297     other than 2 because of this, bypassing the problem. Unfortunately,
1298     there was also a test in test 3 (the locale tests) that used /B and
1299 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
1300     I have added a new /Z option to pcretest that replaces the length and
1301     offset values with spaces. This is now used to make test 3 independent
1302 ph10 122 of link size. (Test 2 will be tidied up later.)
1303 ph10 123
1304     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
1305 ph10 122 segmentation fault instead of returning the appropriate error message.
1306 ph10 134
1307 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
1308 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
1309     This doesn't seem right; it now treats the CRLF combination as the line
1310     ending, and so does not match in that case. It's only a pattern such as ^$
1311     that would hit this one: something like ^ABC$ would have failed after \r
1312     and then tried again after \r\n.
1313    
1314 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
1315 ph10 134 in an attempt to make files that differ only in their line terminators
1316     compare equal. This works on Linux.
1317 ph10 142
1318 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
1319     as it exited. This is now fixed, thanks to valgrind.
1320 ph10 142
1321 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
1322 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
1323     was because its rules for how to advance for /g after matching an empty
1324 ph10 143 string at the end of a line did not allow for this case. They now check for
1325     it specially.
1326 ph10 150
1327     20. pcretest is supposed to handle patterns and data of any length, by
1328     extending its buffers when necessary. It was getting this wrong when the
1329 ph10 147 buffer for a data line had to be extended.
1330 ph10 150
1331 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
1332 ph10 150 CRLF as a newline sequence.
1333 ph10 152
1334 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
1335 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
1336     I have nevertheless tidied it up.
1337 ph10 111
1338 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
1339 ph10 123
1340 ph10 153 24. Added a man page for pcre-config.
1341 ph10 152
1342 ph10 153
1343 nigel 93 Version 7.0 19-Dec-06
1344     ---------------------
1345    
1346     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
1347     moving to gcc 4.1.1.
1348    
1349     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
1350     sys/time.h, which is documented as needed for this function. It doesn't
1351     seem to matter on Linux, but it showed up on some releases of OS X.
1352    
1353     3. It seems that there are systems where bytes whose values are greater than
1354     127 match isprint() in the "C" locale. The "C" locale should be the
1355     default when a C program starts up. In most systems, only ASCII printing
1356     characters match isprint(). This difference caused the output from pcretest
1357     to vary, making some of the tests fail. I have changed pcretest so that:
1358    
1359     (a) When it is outputting text in the compiled version of a pattern, bytes
1360     other than 32-126 are always shown as hex escapes.
1361    
1362     (b) When it is outputting text that is a matched part of a subject string,
1363     it does the same, unless a different locale has been set for the match
1364     (using the /L modifier). In this case, it uses isprint() to decide.
1365    
1366     4. Fixed a major bug that caused incorrect computation of the amount of memory
1367     required for a compiled pattern when options that changed within the
1368     pattern affected the logic of the preliminary scan that determines the
1369     length. The relevant options are -x, and -i in UTF-8 mode. The result was
1370     that the computed length was too small. The symptoms of this bug were
1371     either the PCRE error "internal error: code overflow" from pcre_compile(),
1372     or a glibc crash with a message such as "pcretest: free(): invalid next
1373     size (fast)". Examples of patterns that provoked this bug (shown in
1374     pcretest format) are:
1375    
1376     /(?-x: )/x
1377     /(?x)(?-x: \s*#\s*)/
1378     /((?i)[\x{c0}])/8
1379     /(?i:[\x{c0}])/8
1380    
1381     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
1382     is now done differently.
1383    
1384     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
1385     wrapper classes; (b) implement a new function in the C++ scanner that is
1386     more efficient than the old way of doing things because it avoids levels of
1387     recursion in the regex matching; (c) add a paragraph to the documentation
1388     for the FullMatch() function.
1389    
1390     6. The escape sequence \n was being treated as whatever was defined as
1391     "newline". Not only was this contrary to the documentation, which states
1392     that \n is character 10 (hex 0A), but it also went horribly wrong when
1393     "newline" was defined as CRLF. This has been fixed.
1394    
1395     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1396     was being set to -1 for the "end of line" case (supposedly a value that no
1397     character can have). Though this value is never used (the check for end of
1398     line is "zero bytes in current character"), it caused compiler complaints.
1399     I've changed it to 0xffffffff.
1400    
1401     8. In pcre_version.c, the version string was being built by a sequence of
1402     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1403     string (as it is for production releases) called a macro with an empty
1404     argument. The C standard says the result of this is undefined. The gcc
1405     compiler treats it as an empty string (which was what was wanted) but it is
1406     reported that Visual C gives an error. The source has been hacked around to
1407     avoid this problem.
1408    
1409     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1410     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1411     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1412     of them did).
1413    
1414     10. Originally, pcretest opened its input and output without "b"; then I was
1415     told that "b" was needed in some environments, so it was added for release
1416     5.0 to both the input and output. (It makes no difference on Unix-like
1417     systems.) Later I was told that it is wrong for the input on Windows. I've
1418     now abstracted the modes into two macros, to make it easier to fiddle with
1419     them, and removed "b" from the input mode under Windows.
1420    
1421     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1422    
1423     12. Added -help and --help to pcretest as an official way of being reminded
1424     of the options.
1425    
1426     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1427     and pcrecpp.cc because they annoy compilers at high warning levels.
1428    
1429     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1430    
1431     15. Fixed an occurrence of == in configure.ac that should have been = (shell
1432     scripts are not C programs :-) and which was not noticed because it works
1433     on Linux.
1434    
1435     16. pcretest is supposed to handle any length of pattern and data line (as one
1436     line or as a continued sequence of lines) by extending its input buffer if
1437     necessary. This feature was broken for very long pattern lines, leading to
1438     a string of junk being passed to pcre_compile() if the pattern was longer
1439     than about 50K.
1440    
1441     17. I have done a major re-factoring of the way pcre_compile() computes the
1442     amount of memory needed for a compiled pattern. Previously, there was code
1443     that made a preliminary scan of the pattern in order to do this. That was
1444     OK when PCRE was new, but as the facilities have expanded, it has become
1445     harder and harder to keep it in step with the real compile phase, and there
1446     have been a number of bugs (see for example, 4 above). I have now found a
1447     cunning way of running the real compile function in a "fake" mode that
1448     enables it to compute how much memory it would need, while actually only
1449     ever using a few hundred bytes of working memory and without too many
1450     tests of the mode. This should make future maintenance and development
1451     easier. A side effect of this work is that the limit of 200 on the nesting
1452     depth of parentheses has been removed (though this was never a serious
1453     limitation, I suspect). However, there is a downside: pcre_compile() now
1454     runs more slowly than before (30% or more, depending on the pattern). I
1455     hope this isn't a big issue. There is no effect on runtime performance.
1456    
1457     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1458     newline (only possible for the last line of a file) and it was a
1459     pattern that set a locale (followed by /Lsomething), pcretest crashed.
1460    
1461     19. Added additional timing features to pcretest. (1) The -tm option now times
1462     matching only, not compiling. (2) Both -t and -tm can be followed, as a
1463     separate command line item, by a number that specifies the number of
1464     repeats to use when timing. The default is 50000; this gives better
1465     precision, but takes uncomfortably long for very large patterns.
1466    
1467     20. Extended pcre_study() to be more clever in cases where a branch of a
1468     subpattern has no definite first character. For example, (a*|b*)[cd] would
1469     previously give no result from pcre_study(). Now it recognizes that the
1470     first character must be a, b, c, or d.
1471    
1472     21. There was an incorrect error "recursive call could loop indefinitely" if
1473     a subpattern (or the entire pattern) that was being tested for matching an
1474     empty string contained only one non-empty item after a nested subpattern.
1475     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1476     incorrectly, because the \d was being skipped in the check.
1477    
1478     22. The pcretest program now has a new pattern option /B and a command line
1479     option -b, which is equivalent to adding /B to every pattern. This causes
1480     it to show the compiled bytecode, without the additional information that
1481     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1482     is the same as /B/I).
1483    
1484     23. A new optimization is now able automatically to treat some sequences such
1485     as a*b as a*+b. More specifically, if something simple (such as a character
1486     or a simple class like \d) has an unlimited quantifier, and is followed by
1487     something that cannot possibly match the quantified thing, the quantifier
1488     is automatically "possessified".
1489    
1490     24. A recursive reference to a subpattern whose number was greater than 39
1491     went wrong under certain circumstances in UTF-8 mode. This bug could also
1492     have affected the operation of pcre_study().
1493    
1494     25. Realized that a little bit of performance could be had by replacing
1495     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1496    
1497     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1498    
1499     27. Possessive quantifiers such as a++ were previously implemented by turning
1500     them into atomic groups such as ($>a+). Now they have their own opcodes,
1501     which improves performance. This includes the automatically created ones
1502     from 23 above.
1503    
1504     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1505     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1506     the first matched character to be a colon. This applied both to named and
1507     numbered groups.
1508    
1509     29. The ucpinternal.h header file was missing its idempotency #ifdef.
1510    
1511     30. I was sent a "project" file called libpcre.a.dev which I understand makes
1512     building PCRE on Windows easier, so I have included it in the distribution.
1513    
1514     31. There is now a check in pcretest against a ridiculously large number being
1515     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1516     loop, the loop is abandoned.
1517    
1518     32. Forward references to subpatterns in conditions such as (?(2)...) where
1519     subpattern 2 is defined later cause pcre_compile() to search forwards in
1520     the pattern for the relevant set of parentheses. This search went wrong
1521     when there were unescaped parentheses in a character class, parentheses
1522     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1523    
1524     33. "Subroutine" calls and backreferences were previously restricted to
1525     referencing subpatterns earlier in the regex. This restriction has now
1526     been removed.
1527    
1528     34. Added a number of extra features that are going to be in Perl 5.10. On the
1529     whole, these are just syntactic alternatives for features that PCRE had
1530     previously implemented using the Python syntax or my own invention. The
1531     other formats are all retained for compatibility.
1532    
1533     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1534     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1535     also .NET compatible.
1536    
1537     (b) A recursion or subroutine call to a named group can now be defined as
1538     (?&name) as well as (?P>name).
1539    
1540     (c) A backreference to a named group can now be defined as \k<name> or
1541     \k'name' as well as (?P=name). The new forms, as well as being in Perl
1542     5.10, are also .NET compatible.
1543    
1544     (d) A conditional reference to a named group can now use the syntax
1545     (?(<name>) or (?('name') as well as (?(name).
1546    
1547     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1548     groups (named and numbered) that are never evaluated inline, but can be
1549     called as "subroutines" from elsewhere. In effect, the DEFINE condition
1550     is always false. There may be only one alternative in such a group.
1551    
1552     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1553     as the simple (?(R). The condition is true only if the most recent
1554     recursion is that of the given number or name. It does not search out
1555     through the entire recursion stack.
1556    
1557     (g) The escape \gN or \g{N} has been added, where N is a positive or
1558     negative number, specifying an absolute or relative reference.
1559    
1560     35. Tidied to get rid of some further signed/unsigned compiler warnings and
1561     some "unreachable code" warnings.
1562    
1563     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1564     things, this adds five new scripts.
1565    
1566     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1567     There were also incompatibilities regarding the handling of \Q..\E inside
1568     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1569     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1570    
1571     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1572     matches an empty string, and forcibly breaks the loop. There were bugs in
1573     this code in non-simple cases. For a pattern such as ^(a()*)* matched
1574     against aaaa the result was just "a" rather than "aaaa", for example. Two
1575     separate and independent bugs (that affected different cases) have been
1576     fixed.
1577    
1578     39. Refactored the code to abolish the use of different opcodes for small
1579     capturing bracket numbers. This is a tidy that I avoided doing when I
1580     removed the limit on the number of capturing brackets for 3.5 back in 2001.
1581     The new approach is not only tidier, it makes it possible to reduce the
1582     memory needed to fix the previous bug (38).
1583    
1584     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1585     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1586     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1587     mode.
1588    
1589     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1590     report.
1591    
1592     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1593     copy construction and assignment in the C++ wrapper.
1594    
1595     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1596     couple of bugs that could have given wrong results in the "--newline=crlf"
1597     case.
1598    
1599     44. Added a number of casts and did some reorganization of signed/unsigned int
1600     variables following suggestions from Dair Grant. Also renamed the variable
1601     "this" as "item" because it is a C++ keyword.
1602    
1603     45. Arranged for dftables to add
1604    
1605     #include "pcre_internal.h"
1606    
1607     to pcre_chartables.c because without it, gcc 4.x may remove the array
1608     definition from the final binary if PCRE is built into a static library and
1609     dead code stripping is activated.
1610    
1611     46. For an unanchored pattern, if a match attempt fails at the start of a
1612     newline sequence, and the newline setting is CRLF or ANY, and the next two
1613     characters are CRLF, advance by two characters instead of one.
1614    
1615    
1616 nigel 91 Version 6.7 04-Jul-06
1617     ---------------------
1618    
1619     1. In order to handle tests when input lines are enormously long, pcretest has
1620     been re-factored so that it automatically extends its buffers when
1621     necessary. The code is crude, but this _is_ just a test program. The
1622     default size has been increased from 32K to 50K.
1623    
1624     2. The code in pcre_study() was using the value of the re argument before
1625     testing it for NULL. (Of course, in any sensible call of the function, it
1626     won't be NULL.)
1627    
1628     3. The memmove() emulation function in pcre_internal.h, which is used on
1629     systems that lack both memmove() and bcopy() - that is, hardly ever -
1630     was missing a "static" storage class specifier.
1631    
1632     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1633     containing an extended class (one that cannot be represented by a bitmap
1634     because it contains high-valued characters or Unicode property items, e.g.
1635     [\pZ]). Almost always one would set UTF-8 mode when processing such a
1636     pattern, but PCRE should not loop if you do not (it no longer does).
1637     [Detail: two cases were found: (a) a repeated subpattern containing an
1638     extended class; (b) a recursive reference to a subpattern that followed a
1639     previous extended class. It wasn't skipping over the extended class
1640     correctly when UTF-8 mode was not set.]
1641    
1642     5. A negated single-character class was not being recognized as fixed-length
1643     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1644     compile error "lookbehind assertion is not fixed length".
1645    
1646     6. The RunPerlTest auxiliary script was showing an unexpected difference
1647     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1648     write a Perl script that can interpret lines of an input file either as
1649     byte characters or as UTF-8, which is what "perltest" was being required to
1650     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1651     can't do is switch easily at run time between having the "use utf8;" pragma
1652     or not. In the end, I fudged it by using the RunPerlTest script to insert
1653     "use utf8;" explicitly for the UTF-8 tests.
1654    
1655     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1656     the end of the subject string, contrary to the documentation and to what
1657     Perl does. This was true of both matching functions. Now it matches only at
1658     the start of the subject and immediately after *internal* newlines.
1659    
1660     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1661     a pointer to an int instead of a pointer to an unsigned long int. This
1662     caused problems on 64-bit systems.
1663    
1664     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1665     instance of the 'standard' template library not being so standard".
1666    
1667     10. There was no check on the number of named subpatterns nor the maximum
1668     length of a subpattern name. The product of these values is used to compute
1669     the size of the memory block for a compiled pattern. By supplying a very
1670     long subpattern name and a large number of named subpatterns, the size
1671     computation could be caused to overflow. This is now prevented by limiting
1672     the length of names to 32 characters, and the number of named subpatterns
1673     to 10,000.
1674    
1675     11. Subpatterns that are repeated with specific counts have to be replicated in
1676     the compiled pattern. The size of memory for this was computed from the
1677     length of the subpattern and the repeat count. The latter is limited to
1678     65535, but there was no limit on the former, meaning that integer overflow
1679     could in principle occur. The compiled length of a repeated subpattern is
1680     now limited to 30,000 bytes in order to prevent this.
1681    
1682     12. Added the optional facility to have named substrings with the same name.
1683    
1684     13. Added the ability to use a named substring as a condition, using the
1685     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1686     are numbers (not recommended). Forward references are permitted.
1687    
1688     14. Added forward references in named backreferences (if you see what I mean).
1689    
1690     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1691     pattern could run off the end of the subject. For example, the pattern
1692     "(?s)(.{1,5})"8 did this with the subject "ab".
1693    
1694     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1695     PCRE_CASELESS was set when matching characters that were quantified with ?
1696     or *.
1697    
1698     17. A character class other than a single negated character that had a minimum
1699     but no maximum quantifier - for example [ab]{6,} - was not handled
1700     correctly by pce_dfa_exec(). It would match only one character.
1701    
1702     18. A valid (though odd) pattern that looked like a POSIX character
1703     class but used an invalid character after [ (for example [[,abc,]]) caused
1704     pcre_compile() to give the error "Failed: internal error: code overflow" or
1705     in some cases to crash with a glibc free() error. This could even happen if
1706     the pattern terminated after [[ but there just happened to be a sequence of
1707     letters, a binary zero, and a closing ] in the memory that followed.
1708    
1709     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1710     over the years. Originally (before any Unicode support), just the bottom 8
1711     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1712     output from "man perlunicode" includes this:
1713    
1714     The regular expression compiler produces polymorphic opcodes. That
1715     is, the pattern adapts to the data and automatically switches to
1716     the Unicode character scheme when presented with Unicode data--or
1717     instead uses a traditional byte scheme when presented with byte
1718     data.
1719    
1720     Sadly, a wide octal escape does not cause a switch, and in a string with
1721     no other multibyte characters, these octal escapes are treated as before.
1722     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1723     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1724     Unicode string.
1725    
1726     I have not perpetrated such confusion in PCRE. Up till now, it took just
1727     the bottom 8 bits, as in old Perl. I have now made octal escapes with
1728     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1729     translate to the appropriate multibyte character.
1730    
1731     29. Applied some refactoring to reduce the number of warnings from Microsoft
1732     and Borland compilers. This has included removing the fudge introduced
1733     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1734     a warning about an unused variable.
1735    
1736     21. PCRE has not included VT (character 0x0b) in the set of whitespace
1737     characters since release 4.0, because Perl (from release 5.004) does not.
1738     [Or at least, is documented not to: some releases seem to be in conflict
1739     with the documentation.] However, when a pattern was studied with
1740     pcre_study() and all its branches started with \s, PCRE still included VT
1741     as a possible starting character. Of course, this did no harm; it just
1742     caused an unnecessary match attempt.
1743    
1744     22. Removed a now-redundant internal flag bit that recorded the fact that case
1745     dependency changed within the pattern. This was once needed for "required
1746     byte" processing, but is no longer used. This recovers a now-scarce options
1747     bit. Also moved the least significant internal flag bit to the most-
1748     significant bit of the word, which was not previously used (hangover from
1749     the days when it was an int rather than a uint) to free up another bit for
1750     the future.
1751    
1752     23. Added support for CRLF line endings as well as CR and LF. As well as the
1753     default being selectable at build time, it can now be changed at runtime
1754     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1755     specify that it is scanning data with non-default line endings.
1756    
1757     24. Changed the definition of CXXLINK to make it agree with the definition of
1758     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1759    
1760     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1761     recursions. This makes a big different to stack usage for some patterns.
1762    
1763     26. If a subpattern containing a named recursion or subroutine reference such
1764     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1765     the space required for the compiled pattern went wrong and gave too small a
1766     value. Depending on the environment, this could lead to "Failed: internal
1767     error: code overflow at offset 49" or "glibc detected double free or
1768     corruption" errors.
1769    
1770     27. Applied patches from Google (a) to support the new newline modes and (b) to
1771     advance over multibyte UTF-8 characters in GlobalReplace.
1772    
1773     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1774     difference for some implementation of PCRE in some Windows version.
1775    
1776     29. Added some extra testing facilities to pcretest:
1777    
1778     \q<number> in a data line sets the "match limit" value
1779     \Q<number> in a data line sets the "match recursion limt" value
1780     -S <number> sets the stack size, where <number> is in megabytes
1781    
1782     The -S option isn't available for Windows.
1783    
1784    
1785 nigel 89 Version 6.6 06-Feb-06
1786     ---------------------
1787    
1788     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1789     in pcreposix.h. I have copied the definition from pcre.h.
1790    
1791     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1792     because pcre.h is no longer a built file.
1793    
1794     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1795     not normally included in the compiled code.
1796    
1797    
1798 nigel 87 Version 6.5 01-Feb-06
1799     ---------------------
1800    
1801     1. When using the partial match feature with pcre_dfa_exec(), it was not
1802     anchoring the second and subsequent partial matches at the new starting
1803     point. This could lead to incorrect results. For example, with the pattern
1804     /1234/, partially matching against "123" and then "a4" gave a match.
1805    
1806     2. Changes to pcregrep:
1807    
1808     (a) All non-match returns from pcre_exec() were being treated as failures
1809     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1810     error message is output. Some extra information is given for the
1811     PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1812     probably the only errors that are likely to be caused by users (by
1813     specifying a regex that has nested indefinite repeats, for instance).
1814     If there are more than 20 of these errors, pcregrep is abandoned.
1815    
1816     (b) A binary zero was treated as data while matching, but terminated the
1817     output line if it was written out. This has been fixed: binary zeroes
1818     are now no different to any other data bytes.
1819    
1820     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1821     used to set a locale for matching. The --locale=xxxx long option has
1822     been added (no short equivalent) to specify a locale explicitly on the
1823     pcregrep command, overriding the environment variables.
1824    
1825     (d) When -B was used with -n, some line numbers in the output were one less
1826     than they should have been.
1827    
1828     (e) Added the -o (--only-matching) option.
1829    
1830     (f) If -A or -C was used with -c (count only), some lines of context were
1831     accidentally printed for the final match.
1832    
1833     (g) Added the -H (--with-filename) option.
1834    
1835     (h) The combination of options -rh failed to suppress file names for files
1836     that were found from directory arguments.
1837    
1838     (i) Added the -D (--devices) and -d (--directories) options.
1839    
1840     (j) Added the -F (--fixed-strings) option.
1841    
1842     (k) Allow "-" to be used as a file name for -f as well as for a data file.
1843    
1844     (l) Added the --colo(u)r option.
1845    
1846     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1847     is not present by default.
1848    
1849     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1850     items such as (?R) or (?1), when the recursion could match a number of
1851     alternatives. If it matched one of the alternatives, but subsequently,
1852     outside the recursion, there was a failure, the code tried to back up into
1853     the recursion. However, because of the way PCRE is implemented, this is not
1854     possible, and the result was an incorrect result from the match.
1855    
1856     In order to prevent this happening, the specification of recursion has
1857     been changed so that all such subpatterns are automatically treated as
1858     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1859    
1860     4. I had overlooked the fact that, in some locales, there are characters for
1861     which isalpha() is true but neither isupper() nor islower() are true. In
1862     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1863     and ordfeminine) are like this. This affected the treatment of \w and \W
1864     when they appeared in character classes, but not when they appeared outside
1865     a character class. The bit map for "word" characters is now created
1866     separately from the results of isalnum() instead of just taking it from the
1867     upper, lower, and digit maps. (Plus the underscore character, of course.)
1868    
1869     5. The above bug also affected the handling of POSIX character classes such as
1870     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1871     permanent tables. Instead, the bit maps for such a class were previously
1872     created as the appropriate unions of the upper, lower, and digit bitmaps.
1873     Now they are created by subtraction from the [[:word:]] class, which has
1874     its own bitmap.
1875    
1876     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1877     It is created by subtracting the vertical space characters (\x09, \x0a,
1878     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1879     subtraction was done in the overall bitmap for a character class, meaning
1880     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1881     be recognized. This bug has been fixed.
1882    
1883     7. Patches from the folks at Google:
1884    
1885     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1886     real life, but is still worth protecting against".
1887    
1888     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1889     regular expressions".
1890    
1891     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1892     have it.
1893    
1894     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1895     "configure" and the latter not, in order to fix a problem somebody had
1896     with compiling the Arg class on HP-UX.
1897    
1898     (e) Improve the error-handling of the C++ wrapper a little bit.
1899    
1900     (f) New tests for checking recursion limiting.
1901    
1902     8. The pcre_memmove() function, which is used only if the environment does not
1903     have a standard memmove() function (and is therefore rarely compiled),
1904     contained two bugs: (a) use of int instead of size_t, and (b) it was not
1905     returning a result (though PCRE never actually uses the result).
1906    
1907     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1908     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1909     returned instead of calling malloc() with an overflowing number that would
1910     most likely cause subsequent chaos.
1911    
1912     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1913    
1914     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1915     with this option is matched, the nmatch and pmatch options of regexec() are
1916     ignored.
1917    
1918     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1919     provided in case anyone wants to the the POSIX interface with UTF-8
1920     strings.
1921    
1922     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1923     C++ linking (needed for some HP-UX environments).
1924    
1925     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1926     (unused parameter) and in the pcre_printint() function (omitted "default"
1927     switch label when the default is to do nothing).
1928    
1929     15. Added some code to make it possible, when PCRE is compiled as a C++
1930     library, to replace subject pointers for pcre_exec() with a smart pointer
1931     class, thus making it possible to process discontinuous strings.
1932    
1933     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1934     much the same function. They were added by different people who were trying
1935     to make PCRE easy to compile on non-Unix systems. It has been suggested
1936     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1937     for compiling on Windows systems. I have therefore replaced it with
1938     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1939     defaults to "extern" for C or "extern C" for C++, which works fine on
1940     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1941     SCOPE with something explicit in config.h. In addition:
1942    
1943     (a) pcreposix.h still had just "extern" instead of either of these macros;
1944     I have replaced it with PCRE_DATA_SCOPE.
1945    
1946     (b) Functions such as _pcre_xclass(), which are internal to the library,
1947     but external in the C sense, all had PCRE_EXPORT in their definitions.
1948     This is apparently wrong for the Windows case, so I have removed it.
1949     (It makes no difference on Unix-like systems.)
1950    
1951     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1952     of recursive calls to match(). This is different to MATCH_LIMIT because
1953     that limits the total number of calls to match(), not all of which increase
1954     the depth of recursion. Limiting the recursion depth limits the amount of
1955     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1956     when PCRE is compiled, and changed at run time. A patch from Google adds
1957     this functionality to the C++ interface.
1958    
1959     18. Changes to the handling of Unicode character properties:
1960    
1961     (a) Updated the table to Unicode 4.1.0.
1962    
1963     (b) Recognize characters that are not in the table as "Cn" (undefined).
1964    
1965     (c) I revised the way the table is implemented to a much improved format
1966     which includes recognition of ranges. It now supports the ranges that
1967     are defined in UnicodeData.txt, and it also amalgamates other
1968     characters into ranges. This has reduced the number of entries in the
1969     table from around 16,000 to around 3,000, thus reducing its size
1970     considerably. I realized I did not need to use a tree structure after
1971     all - a binary chop search is just as efficient. Having reduced the
1972     number of entries, I extended their size from 6 bytes to 8 bytes to
1973     allow for more data.
1974    
1975     (d) Added support for Unicode script names via properties such as \p{Han}.
1976    
1977     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1978     matching that character.
1979    
1980     20. When matching a repeated Unicode property with a minimum greater than zero,
1981     (for example \pL{2,}), PCRE could look past the end of the subject if it
1982     reached it while seeking the minimum number of characters. This could
1983     happen only if some of the characters were more than one byte long, because
1984     there is a check for at least the minimum number of bytes.
1985    
1986     21. Refactored the implementation of \p and \P so as to be more general, to
1987     allow for more different types of property in future. This has changed the
1988     compiled form incompatibly. Anybody with saved compiled patterns that use
1989     \p or \P will have to recompile them.
1990    
1991     22. Added "Any" and "L&" to the supported property types.
1992    
1993     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1994     but give a compile time error if the value is greater than 0xff.
1995    
1996     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1997     accidentally not being installed or uninstalled.
1998    
1999     25. The pcre.h file was built from pcre.h.in, but the only changes that were
2000     made were to insert the current release number. This seemed silly, because
2001     it made things harder for people building PCRE on systems that don't run
2002     "configure". I have turned pcre.h into a distributed file, no longer built
2003     by "configure", with the version identification directly included. There is
2004     no longer a pcre.h.in file.
2005    
2006     However, this change necessitated a change to the pcre-config script as
2007     well. It is built from pcre-config.in, and one of the substitutions was the
2008     release number. I have updated configure.ac so that ./configure now finds
2009     the release number by grepping pcre.h.
2010    
2011     26. Added the ability to run the tests under valgrind.
2012    
2013    
2014 nigel 85 Version 6.4 05-Sep-05
2015     ---------------------
2016    
2017     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
2018     "--" to be printed when multiple files were scanned, even when none of the
2019     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
2020     consider it to be a bug, and have restored the previous behaviour.
2021    
2022     2. A couple of code tidies to get rid of compiler warnings.
2023    
2024     3. The pcretest program used to cheat by referring to symbols in the library
2025     whose names begin with _pcre_. These are internal symbols that are not
2026     really supposed to be visible externally, and in some environments it is
2027     possible to suppress them. The cheating is now confined to including
2028     certain files from the library's source, which is a bit cleaner.
2029    
2030     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
2031     file's purpose clearer.
2032    
2033     5. Reorganized pcre_ucp_findchar().
2034    
2035    
2036 nigel 83 Version 6.3 15-Aug-05
2037     ---------------------
2038    
2039     1. The file libpcre.pc.in did not have general read permission in the tarball.
2040    
2041     2. There were some problems when building without C++ support:
2042    
2043     (a) If C++ support was not built, "make install" and "make test" still
2044     tried to test it.
2045    
2046     (b) There were problems when the value of CXX was explicitly set. Some
2047     changes have been made to try to fix these, and ...
2048    
2049     (c) --disable-cpp can now be used to explicitly disable C++ support.
2050    
2051     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
2052     backslash in a target when C++ was disabled. This confuses some
2053     versions of "make", apparently. Using an intermediate variable solves
2054     this. (Same for CPP_LOBJ.)
2055    
2056     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
2057     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
2058     necessary on certain architectures.
2059    
2060     4. Added a setting of -export-symbols-regex to the link command to remove
2061     those symbols that are exported in the C sense, but actually are local
2062     within the library, and not documented. Their names all begin with
2063     "_pcre_". This is not a perfect job, because (a) we have to except some
2064     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
2065     available (and never for static libraries). I have made a note to try to
2066     find a way round (a) in the future.
2067    
2068    
2069 nigel 81 Version 6.2 01-Aug-05
2070     ---------------------
2071    
2072     1. There was no test for integer overflow of quantifier values. A construction
2073     such as {1111111111111111} would give undefined results. What is worse, if
2074     a minimum quantifier for a parenthesized subpattern overflowed and became
2075     negative, the calculation of the memory size went wrong. This could have
2076     led to memory overwriting.
2077    
2078     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
2079    
2080     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
2081     operating environments where this matters.
2082    
2083     4. Applied Giuseppe Maxia's patch to add additional features for controlling
2084     PCRE options from within the C++ wrapper.
2085    
2086     5. Named capturing subpatterns were not being correctly counted when a pattern
2087     was compiled. This caused two problems: (a) If there were more than 100
2088     such subpatterns, the calculation of the memory needed for the whole
2089     compiled pattern went wrong, leading to an overflow error. (b) Numerical
2090     back references of the form \12, where the number was greater than 9, were
2091     not recognized as back references, even though there were sufficient
2092     previous subpatterns.
2093    
2094     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
2095     versions of gcc, e.g. 2.95.4.
2096    
2097    
2098 nigel 79 Version 6.1 21-Jun-05
2099     ---------------------
2100    
2101     1. There was one reference to the variable "posix" in pcretest.c that was not
2102     surrounded by "#if !defined NOPOSIX".
2103    
2104     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
2105     the cross-check on the old pcre_info() function, for the benefit of the
2106     cut-down version of PCRE that is currently imported into Exim.
2107    
2108     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
2109     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
2110     patterns that start (?i) but I don't think that matters. The use of (?i) is
2111     just an example; this all applies to the other options as well.
2112    
2113     4. Since libtool seems to echo the compile commands it is issuing, the output
2114     from "make" can be reduced a bit by putting "@" in front of each libtool
2115     compile command.
2116    
2117     5. Patch from the folks at Google for configure.in to be a bit more thorough
2118     in checking for a suitable C++ installation before trying to compile the
2119     C++ stuff. This should fix a reported problem when a compiler was present,
2120     but no suitable headers.
2121    
2122     6. The man pages all had just "PCRE" as their title. I have changed them to
2123     be the relevant file name. I have also arranged that these names are
2124     retained in the file doc/pcre.txt, which is a concatenation in text format
2125     of all the man pages except the little individual ones for each function.
2126    
2127     7. The NON-UNIX-USE file had not been updated for the different set of source
2128     files that come with release 6. I also added a few comments about the C++
2129     wrapper.
2130    
2131    
2132 nigel 77 Version 6.0 07-Jun-05
2133     ---------------------
2134    
2135     1. Some minor internal re-organization to help with my DFA experiments.
2136    
2137     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
2138     didn't matter for the library itself when fully configured, but did matter
2139     when compiling without UCP support, or within Exim, where the ucp files are
2140     not imported.
2141    
2142     3. Refactoring of the library code to split up the various functions into
2143     different source modules. The addition of the new DFA matching code (see
2144     below) to a single monolithic source would have made it really too
2145     unwieldy, quite apart from causing all the code to be include in a
2146     statically linked application, when only some functions are used. This is
2147     relevant even without the DFA addition now that patterns can be compiled in
2148     one application and matched in another.
2149    
2150     The downside of splitting up is that there have to be some external
2151     functions and data tables that are used internally in different modules of
2152     the library but which are not part of the API. These have all had their
2153     names changed to start with "_pcre_" so that they are unlikely to clash
2154     with other external names.
2155    
2156     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
2157     a different (DFA) algorithm. Although it is slower than the original
2158     function, it does have some advantages for certain types of matching
2159     problem.
2160    
2161     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
2162     including restarting after a partial match.
2163    
2164     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
2165     defined when compiling for Windows was sent to me. I have put it into the
2166     code, though I have no means of testing or verifying it.
2167    
2168     7. Added the pcre_refcount() auxiliary function.
2169    
2170     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
2171     match before or at the first newline in the subject string. In pcretest,
2172     the /f option on a pattern can be used to set this.
2173    
2174     9. A repeated \w when used in UTF-8 mode with characters greater than 256
2175     would behave wrongly. This has been present in PCRE since release 4.0.
2176    
2177     10. A number of changes to the pcregrep command:
2178    
2179     (a) Refactored how -x works; insert ^(...)$ instead of setting
2180     PCRE_ANCHORED and checking the length, in preparation for adding
2181     something similar for -w.
2182    
2183     (b) Added the -w (match as a word) option.
2184    
2185     (c) Refactored the way lines are read and buffered so as to have more
2186     than one at a time available.
2187    
2188     (d) Implemented a pcregrep test script.
2189    
2190     (e) Added the -M (multiline match) option. This allows patterns to match
2191     over several lines of the subject. The buffering ensures that at least
2192     8K, or the rest of the document (whichever is the shorter) is available
2193     for matching (and similarly the previous 8K for lookbehind assertions).
2194    
2195     (f) Changed the --help output so that it now says
2196    
2197     -w, --word-regex(p)
2198    
2199     instead of two lines, one with "regex" and the other with "regexp"
2200     because that confused at least one person since the short forms are the
2201     same. (This required a bit of code, as the output is generated
2202     automatically from a table. It wasn't just a text change.)
2203    
2204     (g) -- can be used to terminate pcregrep options if the next thing isn't an
2205     option but starts with a hyphen. Could be a pattern or a path name
2206     starting with a hyphen, for instance.
2207    
2208     (h) "-" can be given as a file name to represent stdin.
2209    
2210     (i) When file names are being printed, "(standard input)" is used for
2211     the standard input, for compatibility with GNU grep. Previously
2212     "<stdin>" was used.
2213    
2214     (j) The option --label=xxx can be used to supply a name to be used for
2215     stdin when file names are being printed. There is no short form.
2216    
2217     (k) Re-factored the options decoding logic because we are going to add
2218     two more options that take data. Such options can now be given in four
2219     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
2220    
2221     (l) Added the -A, -B, and -C options for requesting that lines of context
2222     around matches be printed.
2223    
2224     (m) Added the -L option to print the names of files that do not contain
2225     any matching lines, that is, the complement of -l.
2226    
2227     (n) The return code is 2 if any file cannot be opened, but pcregrep does
2228     continue to scan other files.
2229    
2230     (o) The -s option was incorrectly implemented. For compatibility with other
2231     greps, it now suppresses the error message for a non-existent or non-
2232     accessible file (but not the return code). There is a new option called
2233     -q that suppresses the output of matching lines, which was what -s was
2234     previously doing.
2235    
2236     (p) Added --include and --exclude options to specify files for inclusion
2237     and exclusion when recursing.
2238    
2239     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
2240     Hopefully, it now does.
2241    
2242     12. Missing cast in pcre_study().
2243    
2244     13. Added an "uninstall" target to the makefile.
2245    
2246     14. Replaced "extern" in the function prototypes in Makefile.in with
2247     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
2248     world, but is set differently for Windows.
2249    
2250     15. Added a second compiling function called pcre_compile2(). The only
2251     difference is that it has an extra argument, which is a pointer to an
2252     integer error code. When there is a compile-time failure, this is set
2253     non-zero, in addition to the error test pointer being set to point to an
2254     error message. The new argument may be NULL if no error number is required
2255     (but then you may as well call pcre_compile(), which is now just a
2256     wrapper). This facility is provided because some applications need a
2257     numeric error indication, but it has also enabled me to tidy up the way
2258     compile-time errors are handled in the POSIX wrapper.
2259    
2260     16. Added VPATH=.libs to the makefile; this should help when building with one
2261     prefix path and installing with another. (Or so I'm told by someone who
2262     knows more about this stuff than I do.)
2263    
2264     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
2265     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
2266     match everything, including newlines. This is not POSIX-compatible, but
2267     somebody wanted the feature. From pcretest it can be activated by using
2268     both the P and the s flags.
2269    
2270     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
2271    
2272     19. libpcre.pc was being incorrectly installed as executable.
2273    
2274     20. A couple of places in pcretest check for end-of-line by looking for '\n';
2275     it now also looks for '\r' so that it will work unmodified on Windows.
2276    
2277     21. Added Google's contributed C++ wrapper to the distribution.
2278    
2279     22. Added some untidy missing memory free() calls in pcretest, to keep
2280     Electric Fence happy when testing.
2281    
2282    
2283    
2284 nigel 75 Version 5.0 13-Sep-04
2285     ---------------------
2286    
2287     1. Internal change: literal characters are no longer packed up into items
2288     containing multiple characters in a single byte-string. Each character
2289     is now matched using a separate opcode. However, there may be more than one
2290     byte in the character in UTF-8 mode.
2291    
2292     2. The pcre_callout_block structure has two new fields: pattern_position and
2293     next_item_length. These contain the offset in the pattern to the next match
2294     item, and its length, respectively.
2295    
2296     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
2297     insertion of callouts before each pattern item. Added the /C option to
2298     pcretest to make use of this.
2299    
2300     4. On the advice of a Windows user, the lines
2301    
2302     #if defined(_WIN32) || defined(WIN32)
2303     _setmode( _fileno( stdout ), 0x8000 );
2304     #endif /* defined(_WIN32) || defined(WIN32) */
2305    
2306     have been added to the source of pcretest. This apparently does useful
2307     magic in relation to line terminators.
2308    
2309     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
2310     for the benefit of those environments where the "b" makes a difference.
2311    
2312     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
2313     to know about it. I have put a hack into configure.in that adds in code
2314     to set GCC=yes if CC=icc. This seems to end up at a point in the
2315     generated configure script that is early enough to affect the setting of
2316     compiler options, which is what is needed, but I have no means of testing
2317     whether it really works. (The user who reported this had patched the
2318     generated configure script, which of course I cannot do.)
2319    
2320     LATER: After change 22 below (new libtool files), the configure script
2321     seems to know about icc (and also ecc). Therefore, I have commented out
2322     this hack in configure.in.
2323    
2324     7. Added support for pkg-config (2 patches were sent in).
2325    
2326     8. Negated POSIX character classes that used a combination of internal tables
2327     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
2328     [[:^ascii]]. Typically, they would match almost any characters. The other
2329     POSIX classes were not broken in this way.
2330    
2331     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
2332     to find the match, as PCRE was deluded into thinking that the match had to
2333     start at the start point or following a newline. The same bug applied to
2334     patterns with negative forward assertions or any backward assertions
2335     preceding ".*" at the start, unless the pattern required a fixed first
2336     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
2337    
2338     10. In UTF-8 mode, when moving forwards in the subject after a failed match
2339     starting at the last subject character, bytes beyond the end of the subject
2340     string were read.
2341    
2342     11. Renamed the variable "class" as "classbits" to make life easier for C++
2343     users. (Previously there was a macro definition, but it apparently wasn't
2344     enough.)
2345    
2346     12. Added the new field "tables" to the extra data so that tables can be passed
2347     in at exec time, or the internal tables can be re-selected. This allows
2348     a compiled regex to be saved and re-used at a later time by a different
2349     program that might have everything at different addresses.
2350    
2351     13. Modified the pcre-config script so that, when run on Solaris, it shows a
2352     -R library as well as a -L library.
2353    
2354     14. The debugging options of pcretest (-d on the command line or D on a
2355     pattern) showed incorrect output for anything following an extended class
2356     that contained multibyte characters and which was followed by a quantifier.
2357    
2358     15. Added optional support for general category Unicode character properties
2359     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
2360     support. It adds about 90K to the size of the library. The meanings of the
2361     inbuilt class escapes such as \d and \s have NOT been changed.
2362    
2363     16. Updated pcredemo.c to include calls to free() to release the memory for the
2364     compiled pattern.
2365    
2366     17. The generated file chartables.c was being created in the source directory
2367     instead of in the building directory. This caused the build to fail if the
2368     source directory was different from the building directory, and was
2369     read-only.
2370    
2371     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
2372     file. No doubt somebody will tell me if they don't make sense... Also added
2373     Dan Mooney's comments about building on OpenVMS.
2374    
2375     19. Added support for partial matching via the PCRE_PARTIAL option for
2376     pcre_exec() and the \P data escape in pcretest.
2377    
2378     20. Extended pcretest with 3 new pattern features:
2379    
2380     (i) A pattern option of the form ">rest-of-line" causes pcretest to
2381     write the compiled pattern to the file whose name is "rest-of-line".
2382     This is a straight binary dump of the data, with the saved pointer to
2383     the character tables forced to be NULL. The study data, if any, is
2384     written too. After writing, pcretest reads a new pattern.
2385    
2386     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2387     compiled pattern from the given file. There must not be any
2388     occurrences of "<" in the file name (pretty unlikely); if there are,
2389     pcretest will instead treat the initial "<" as a pattern delimiter.
2390     After reading in the pattern, pcretest goes on to read data lines as
2391     usual.
2392    
2393     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2394     and 16-bit fields in a compiled pattern, to simulate a pattern that
2395     was compiled on a host of opposite endianness.
2396    
2397     21. The pcre-exec() function can now cope with patterns that were compiled on
2398     hosts of opposite endianness, with this restriction:
2399    
2400     As for any compiled expression that is saved and used later, the tables
2401     pointer field cannot be preserved; the extra_data field in the arguments
2402     to pcre_exec() should be used to pass in a tables address if a value
2403     other than the default internal tables were used at compile time.
2404    
2405     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2406     now diagnosed as an error. Previously, most of the time, a negative number
2407     would have been treated as zero, but if in addition "ovector" was passed as
2408     NULL, a crash could occur.
2409    
2410     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2411     new versions from the libtool 1.5 distribution (the last one is a copy of
2412     a file called libtool.m4). This seems to have fixed the need to patch
2413     "configure" to support Darwin 1.3 (which I used to do). However, I still
2414     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2415     workstation).
2416    
2417     24. Changed the PCRE licence to be the more standard "BSD" licence.
2418    
2419    
2420 nigel 73 Version 4.5 01-Dec-03
2421     ---------------------
2422    
2423     1. There has been some re-arrangement of the code for the match() function so
2424     that it can be compiled in a version that does not call itself recursively.
2425     Instead, it keeps those local variables that need separate instances for
2426     each "recursion" in a frame on the heap, and gets/frees frames whenever it
2427     needs to "recurse". Keeping track of where control must go is done by means
2428     of setjmp/longjmp. The whole thing is implemented by a set of macros that
2429     hide most of the details from the main code, and operates only if
2430     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2431     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2432     operating.
2433    
2434     To make it easier for callers to provide specially tailored get/free
2435     functions for this usage, two new functions, pcre_stack_malloc, and
2436     pcre_stack_free, are used. They are always called in strict stacking order,
2437     and the size of block requested is always the same.
2438    
2439     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2440     PCRE has been compiled to use the stack or the heap for recursion. The
2441     -C option of pcretest uses this to show which version is compiled.
2442    
2443     A new data escape \S, is added to pcretest; it causes the amounts of store
2444     obtained and freed by both kinds of malloc/free at match time to be added
2445     to the output.
2446    
2447     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2448     what's available on my current Linux desktop machine.
2449    
2450     3. When matching a UTF-8 string, the test for a valid string at the start has
2451     been extended. If start_offset is not zero, PCRE now checks that it points
2452     to a byte that is the start of a UTF-8 character. If not, it returns
2453     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2454     this is necessary because there may be backward assertions in the pattern.
2455     When matching the same subject several times, it may save resources to use
2456     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2457    
2458     4. The code for checking the validity of UTF-8 strings has been tightened so
2459     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2460     containing "overlong sequences".
2461    
2462     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2463     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2464     should have been "&", but it just so happened that all the cases this let
2465     through by mistake were picked up later in the function.
2466    
2467     6. I had used a variable called "isblank" - this is a C99 function, causing
2468     some compilers to warn. To avoid this, I renamed it (as "blankclass").
2469    
2470     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2471     prompting; (b) run "./pcretest /dev/null" at the start of the test script
2472     so the version is shown; (c) stop "make test" echoing "./RunTest".
2473    
2474     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2475    
2476     9. The prototype for memmove() for systems that don't have it was using
2477     size_t, but the inclusion of the header that defines size_t was later. I've
2478     moved the #includes for the C headers earlier to avoid this.
2479    
2480     10. Added some adjustments to the code to make it easier to compiler on certain
2481     special systems:
2482    
2483     (a) Some "const" qualifiers were missing.
2484     (b) Added the macro EXPORT before all exported functions; by default this
2485     is defined to be empty.
2486     (c) Changed the dftables auxiliary program (that builds chartables.c) so
2487     that it reads its output file name as an argument instead of writing
2488     to the standard output and assuming this can be redirected.
2489    
2490     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2491     class containing characters with values greater than 255, PCRE compilation
2492     went into a loop.
2493    
2494     12. A recursive reference to a subpattern that was within another subpattern
2495     that had a minimum quantifier of zero caused PCRE to crash. For example,
2496     (x(y(?2))z)? provoked this bug with a subject that got as far as the
2497     recursion. If the recursively-called subpattern itself had a zero repeat,
2498     that was OK.
2499    
2500     13. In pcretest, the buffer for reading a data line was set at 30K, but the
2501     buffer into which it was copied (for escape processing) was still set at
2502     1024, so long lines caused crashes.
2503    
2504     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2505     "internal error: code overflow...". This applied to any character class
2506     that was followed by a possessive quantifier.
2507    
2508     15. Modified the Makefile to add libpcre.la as a prerequisite for
2509     libpcreposix.la because I was told this is needed for a parallel build to
2510     work.
2511    
2512     16. If a pattern that contained .* following optional items at the start was
2513     studied, the wrong optimizing data was generated, leading to matching
2514     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2515     matching string must start with a or b or c. The correct conclusion for
2516     this pattern is that a match can start with any character.
2517    
2518    
2519 nigel 71 Version 4.4 13-Aug-03
2520     ---------------------
2521    
2522     1. In UTF-8 mode, a character class containing characters with values between
2523     127 and 255 was not handled correctly if the compiled pattern was studied.
2524     In fixing this, I have also improved the studying algorithm for such
2525     classes (slightly).
2526    
2527     2. Three internal functions had redundant arguments passed to them. Removal
2528     might give a very teeny performance improvement.
2529    
2530     3. Documentation bug: the value of the capture_top field in a callout is *one
2531     more than* the number of the hightest numbered captured substring.
2532    
2533     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2534     in incorrectly linking with a previously installed version. They now link
2535     explicitly with libpcre.la.
2536    
2537     5. configure.in no longer needs to recognize Cygwin specially.
2538    
2539     6. A problem in pcre.in for Windows platforms is fixed.
2540    
2541     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2542     pcretest, it used to include the size of the study block as part of its
2543     output. Unfortunately, the structure contains a field that has a different
2544     size on different hardware architectures. This meant that the tests that
2545     showed this size failed. As the block is currently always of a fixed size,
2546     this information isn't actually particularly useful in pcretest output, so
2547     I have just removed it.
2548    
2549     8. Three pre-processor statements accidentally did not start in column 1.
2550     Sadly, there are *still* compilers around that complain, even though
2551     standard C has not required this for well over a decade. Sigh.
2552    
2553     9. In pcretest, the code for checking callouts passed small integers in the
2554     callout_data field, which is a void * field. However, some picky compilers
2555     complained about the casts involved for this on 64-bit systems. Now
2556     pcretest passes the address of the small integer instead, which should get
2557     rid of the warnings.
2558    
2559     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2560     both compile and run time, and gives an error if an invalid UTF-8 sequence
2561     is found. There is a option for disabling this check in cases where the
2562     string is known to be correct and/or the maximum performance is wanted.
2563    
2564     11. In response to a bug report, I changed one line in Makefile.in from
2565    
2566     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2567     to
2568     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2569    
2570     to look similar to other lines, but I have no way of telling whether this
2571     is the right thing to do, as I do not use Windows. No doubt I'll get told
2572     if it's wrong...
2573    
2574    
2575 nigel 69 Version 4.3 21-May-03
2576     ---------------------
2577    
2578     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2579     Makefile.
2580    
2581     2. Some refactoring to improve the quality of the code:
2582    
2583     (i) The utf8_table... variables are now declared "const".
2584    
2585     (ii) The code for \cx, which used the "case flipping" table to upper case
2586     lower case letters, now just substracts 32. This is ASCII-specific,
2587     but the whole concept of \cx is ASCII-specific, so it seems
2588     reasonable.
2589    
2590     (iii) PCRE was using its character types table to recognize decimal and
2591     hexadecimal digits in the pattern. This is silly, because it handles
2592     only 0-9, a-f, and A-F, but the character types table is locale-
2593     specific, which means strange things might happen. A private
2594     table is now used for this - though it costs 256 bytes, a table is
2595     much faster than multiple explicit tests. Of course, the standard
2596     character types table is still used for matching digits in subject
2597     strings against \d.
2598    
2599     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2600     ending in _t are). So I've renamed it as ESC_tee.
2601    
2602     3. The first argument for regexec() in the POSIX wrapper should have been
2603     defined as "const".
2604    
2605     4. Changed pcretest to use malloc() for its buffers so that they can be
2606     Electric Fenced for debugging.
2607    
2608     5. There were several places in the code where, in UTF-8 mode, PCRE would try
2609     to read one or more bytes before the start of the subject string. Often this
2610     had no effect on PCRE's behaviour, but in some circumstances it could
2611     provoke a segmentation fault.
2612    
2613     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2614     to try to read one or more bytes before the start of the subject string.
2615    
2616     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2617     UTF-8 support could misbehave in various ways if the subject string
2618     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2619     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2620     back over UTF-8 characters.)
2621    
2622    
2623 nigel 67 Version 4.2 14-Apr-03
2624     ---------------------
2625    
2626     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2627    
2628     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2629     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2630     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2631     [WIN_PREFIX]: new variable, "cyg" for Cygwin
2632     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2633     and BUILD_EXEEXT
2634     Note: automatic setting of the BUILD variables is not yet working
2635     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2636     compile-time but not at link-time
2637     [LINK]: use for linking executables only
2638     make different versions for Windows and non-Windows
2639     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2640     libraries
2641     [LINK_FOR_BUILD]: new variable
2642     [OBJEXT]: use throughout
2643     [EXEEXT]: use throughout
2644     <winshared>: new target
2645     <wininstall>: new target
2646     <dftables.o>: use native compiler
2647     <dftables>: use native linker
2648     <install>: handle Windows platform correctly
2649     <clean>: ditto
2650     <check>: ditto
2651     copy DLL to top builddir before testing
2652    
2653     As part of these changes, -no-undefined was removed again. This was reported
2654     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2655     in any case.
2656    
2657     3. Some tidies to get rid of compiler warnings:
2658    
2659     . In the match_data structure, match_limit was an unsigned long int, whereas
2660     match_call_count was an int. I've made them both unsigned long ints.
2661    
2662     . In pcretest the fact that a const uschar * doesn't automatically cast to
2663     a void * provoked a warning.
2664    
2665     . Turning on some more compiler warnings threw up some "shadow" variables
2666     and a few more missing casts.
2667    
2668     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2669     option, a class that contained a single character with a value between 128
2670     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2671    
2672     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2673     option, a class that contained several characters, but with at least one
2674     whose value was between 128 and 255 caused PCRE to crash.
2675    
2676    
2677 nigel 65 Version 4.1 12-Mar-03
2678     ---------------------
2679 nigel 63
2680 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
2681     needed, and a string in dftables.c that was longer than standard compilers are
2682     required to support.
2683    
2684     2. Compiling with Sun's compiler found a few more places where the code could
2685     be tidied up in order to avoid warnings.
2686    
2687     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2688     first of these names is deprecated in the latest Autoconf in favour of the name
2689     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2690     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2691     analogy I have changed it to CFLAGS_FOR_BUILD.
2692    
2693     4. Added -no-undefined to the linking command in the Makefile, because this is
2694     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2695     linking step for the pcreposix library.
2696    
2697     5. PCRE was failing to diagnose the case of two named groups with the same
2698     name.
2699    
2700     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2701     literal character that is needed in the subject for a match, and scans along to
2702     ensure that it is present before embarking on the full matching process. This
2703     saves time in cases of nested unlimited repeats that are never going to match.
2704     Problem: the scan can take a lot of time if the subject is very long (e.g.
2705     megabytes), thus penalizing straightforward matches. It is now done only if the
2706     amount of subject to be scanned is less than 1000 bytes.
2707    
2708     7. A lesser problem with the same optimization is that it was recording the
2709     first character of an anchored pattern as "needed", thus provoking a search
2710     right along the subject, even when the first match of the pattern was going to
2711     fail. The "needed" character is now not set for anchored patterns, unless it
2712     follows something in the pattern that is of non-fixed length. Thus, it still
2713     fulfils its original purpose of finding quick non-matches in cases of nested
2714     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2715    
2716    
2717     Version 4.0 17-Feb-03
2718     ---------------------
2719    
2720 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
2721     extended to the end of string, PCRE compiled incorrect data. This could lead to
2722     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2723    
2724     2. Moved to autoconf 2.53 and libtool 1.4.2.
2725    
2726     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2727     the special perltest8 script is no longer needed - all the tests can be run
2728     from a single perltest script.
2729    
2730     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2731     by \s. It has now been removed in PCRE. This means it isn't recognized as
2732     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2733     class [:space:] *does* include VT, thereby creating a mess.
2734    
2735     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2736     space and tab.
2737    
2738     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2739     its new features into the main test script, reducing the number of scripts.
2740    
2741     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2742     were backward compatible, and made the (?i) apply to the whole pattern, as if
2743     /i were given. Now it behaves more logically, and applies the option setting
2744     only to what follows. PCRE has been changed to follow suit. However, if it
2745     finds options settings right at the start of the pattern, it extracts them into
2746     the global options, as before. Thus, they show up in the info data.
2747    
2748     8. Added support for the \Q...\E escape sequence. Characters in between are
2749     treated as literals. This is slightly different from Perl in that $ and @ are
2750     also handled as literals inside the quotes. In Perl, they will cause variable
2751     interpolation. Note the following examples:
2752    
2753     Pattern PCRE matches Perl matches
2754    
2755     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2756     \Qabc\$xyz\E abc\$xyz abc\$xyz
2757     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2758    
2759     For compatibility with Perl, \Q...\E sequences are recognized inside character
2760     classes as well as outside them.
2761    
2762     9. Re-organized 3 code statements in pcretest to avoid "overflow in
2763     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2764     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2765     signed/unsigned warnings.
2766    
2767     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2768     option for pcretest, so I've replaced it by a simple function that does just
2769     that job.
2770    
2771     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2772     "pcregrep -".
2773    
2774     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2775     Java package. This provides some syntactic sugar for simple cases of what my
2776     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2777     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2778     item, you can use this simplified notation. Note that only makes sense with
2779     greedy quantifiers. Consequently, the use of the possessive quantifier forces
2780     greediness, whatever the setting of the PCRE_UNGREEDY option.
2781    
2782     13. A change of greediness default within a pattern was not taking effect at
2783     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2784     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2785     was abstracted outside.
2786    
2787     14. PCRE now supports the \G assertion. It is true when the current matching
2788     position is at the start point of the match. This differs from \A when the
2789     starting offset is non-zero. Used with the /g option of pcretest (or similar
2790     code), it works in the same way as it does for Perl's /g option. If all
2791     alternatives of a regex begin with \G, the expression is anchored to the start
2792     match position, and the "anchored" flag is set in the compiled expression.
2793    
2794     15. Some bugs concerning the handling of certain option changes within patterns
2795     have been fixed. These applied to options other than (?ims). For example,
2796     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2797     been the other way round. Some of this was related to change 7 above.
2798    
2799     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2800     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2801     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2802     POSIX classes only within a class (e.g. /[[:alpha:]]/).
2803    
2804     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2805     mode. Unlike ".", it always matches newline, whatever the setting of
2806     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2807     assertions. Perl allows it, but it doesn't (in general) work because it can't
2808     calculate the length of the lookbehind. At least, that's the case for Perl
2809     5.8.0 - I've been told they are going to document that it doesn't work in
2810     future.
2811    
2812     18. Added an error diagnosis for escapes that PCRE does not support: these are
2813     \L, \l, \N, \P, \p, \U, \u, and \X.
2814    
2815     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2816     reading past the end of the pattern in cases such as /[abcd/.
2817    
2818     20. PCRE was getting more memory than necessary for patterns with classes that
2819     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2820    
2821     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2822     compiling PCRE for use with Virtual Pascal.
2823    
2824     22. Small fix to the Makefile to make it work properly if the build is done
2825     outside the source tree.
2826    
2827     23. Added a new extension: a condition to go with recursion. If a conditional
2828     subpattern starts with (?(R) the "true" branch is used if recursion has
2829     happened, whereas the "false" branch is used only at the top level.
2830    
2831     24. When there was a very long string of literal characters (over 255 bytes
2832     without UTF support, over 250 bytes with UTF support), the computation of how
2833     much memory was required could be incorrect, leading to segfaults or other
2834     strange effects.
2835    
2836     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2837     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2838     there was a subsequent back reference to those brackets. This meant that, for
2839     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2840     possible to check for precisely this case. All we can do is abandon the
2841     optimization if .* occurs inside capturing brackets when there are any back
2842     references whatsoever. (See below for a better fix that came later.)
2843    
2844     26. The handling of the optimization for finding the first character of a
2845     non-anchored pattern, and for finding a character that is required later in the
2846     match were failing in some cases. This didn't break the matching; it just
2847     failed to optimize when it could. The way this is done has been re-implemented.
2848    
2849     27. Fixed typo in error message for invalid (?R item (it said "(?p").
2850    
2851     28. Added a new feature that provides some of the functionality that Perl
2852     provides with (?{...}). The facility is termed a "callout". The way it is done
2853     in PCRE is for the caller to provide an optional function, by setting
2854     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2855     global variable. By default it is unset, which disables all calling out. To get
2856     the function called, the regex must include (?C) at appropriate points. This
2857     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2858     This provides a means of identifying different callout points. When PCRE
2859     reaches such a point in the regex, if pcre_callout has been set, the external
2860     function is called. It is provided with data in a structure called
2861     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2862     matching continues; if it returns a non-zero value, the match at the current
2863     point fails. However, backtracking will occur if possible. [This was changed
2864     later and other features added - see item 49 below.]
2865    
2866     29. pcretest is upgraded to test the callout functionality. It provides a
2867     callout function that displays information. By default, it shows the start of
2868     the match and the current position in the text. There are some new data escapes
2869     to vary what happens:
2870    
2871     \C+ in addition, show current contents of captured substrings
2872     \C- do not supply a callout function
2873     \C!n return 1 when callout number n is reached
2874     \C!n!m return 1 when callout number n is reached for the mth time
2875    
2876     30. If pcregrep was called with the -l option and just a single file name, it
2877     output "<stdin>" if a match was found, instead of the file name.
2878    
2879     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2880     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2881     pcre_exec(). This saves a malloc/free per call. The default value of
2882     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2883     when configuring.
2884    
2885     32. The default maximum size of a compiled pattern is 64K. There have been a
2886     few cases of people hitting this limit. The code now uses macros to handle the
2887     storing of links as offsets within the compiled pattern. It defaults to 2-byte
2888     links, but this can be changed to 3 or 4 bytes by --with-link-size when
2889     configuring. Tests 2 and 5 work only with 2-byte links because they output
2890     debugging information about compiled patterns.
2891    
2892     33. Internal code re-arrangements:
2893    
2894     (a) Moved the debugging function for printing out a compiled regex into
2895     its own source file (printint.c) and used #include to pull it into
2896     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2897     separate copies.
2898    
2899     (b) Defined the list of op-code names for debugging as a macro in
2900     internal.h so that it is next to the definition of the opcodes.
2901    
2902     (c) Defined a table of op-code lengths for simpler skipping along compiled
2903     code. This is again a macro in internal.h so that it is next to the
2904     definition of the opcodes.
2905    
2906     34. Added support for recursive calls to individual subpatterns, along the
2907     lines of Robin Houston's patch (but implemented somewhat differently).
2908    
2909     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2910     allow it to read and process whole directories in Win32. This code was
2911     contributed by Lionel Fourquaux; it has not been tested by me.
2912    
2913     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2914     used to name a group. Names consist of alphanumerics and underscores, and must
2915     be unique. Back references use the syntax (?P=name) and recursive calls use
2916     (?P>name) which is a PCRE extension to the Python extension. Groups still have
2917     numbers. The function pcre_fullinfo() can be used after compilation to extract
2918     a name/number map. There are three relevant calls:
2919    
2920     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2921     PCRE_INFO_NAMECOUNT yields the number of entries
2922     PCRE_INFO_NAMETABLE yields a pointer to the map.
2923    
2924     The map is a vector of fixed-size entries. The size of each entry depends on
2925     the length of the longest name used. The first two bytes of each entry are the
2926     group number, most significant byte first. There follows the corresponding
2927     name, zero terminated. The names are in alphabetical order.
2928    
2929     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2930     case instead of 255. Making it the same both with and without UTF-8 support
2931     means that the same test output works with both.
2932    
2933     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2934     calling malloc() with a zero argument.
2935    
2936     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2937     optimization. I've improved things by keeping a bitmap of backreferences with
2938     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2939     fact referenced, the optimization can be applied. It is unlikely that a
2940     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2941     the match to follow \n) will appear inside brackets with a number greater than
2942     31, but if it does, any back reference > 31 suppresses the optimization.
2943    
2944     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2945     of disabling numbered capturing parentheses. Any opening parenthesis that is
2946     not followed by ? behaves as if it were followed by ?: but named parentheses
2947     can still be used for capturing (and they will acquire numbers in the usual
2948     way).
2949    
2950     41. Redesigned the return codes from the match() function into yes/no/error so
2951     that errors can be passed back from deep inside the nested calls. A malloc
2952     failure while inside a recursive subpattern call now causes the
2953     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2954    
2955     42. It is now possible to set a limit on the number of times the match()
2956     function is called in a call to pcre_exec(). This facility makes it possible to
2957     limit the amount of recursion and backtracking, though not in a directly
2958     obvious way, because the match() function is used in a number of different
2959     circumstances. The count starts from zero for each position in the subject
2960     string (for non-anchored patterns). The default limit is, for compatibility, a
2961     large number, namely 10 000 000. You can change this in two ways:
2962    
2963     (a) When configuring PCRE before making, you can use --with-match-limit=n
2964     to set a default value for the compiled library.
2965    
2966     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2967     a different value is set. See 45 below.
2968    
2969     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2970    
2971     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2972     of things that can be changed at compile time. The first argument specifies
2973     what is wanted and the second points to where the information is to be placed.
2974     The current list of available information is:
2975    
2976     PCRE_CONFIG_UTF8
2977    
2978     The output is an integer that is set to one if UTF-8 support is available;
2979     otherwise it is set to zero.
2980    
2981     PCRE_CONFIG_NEWLINE
2982    
2983     The output is an integer that it set to the value of the code that is used for
2984     newline. It is either LF (10) or CR (13).
2985    
2986     PCRE_CONFIG_LINK_SIZE
2987    
2988     The output is an integer that contains the number of bytes used for internal
2989     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2990    
2991     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2992    
2993     The output is an integer that contains the threshold above which the POSIX
2994     interface uses malloc() for output vectors. See item 31 above.
2995    
2996     PCRE_CONFIG_MATCH_LIMIT
2997    
2998     The output is an unsigned integer that contains the default limit of the number
2999     of match() calls in a pcre_exec() execution. See 42 above.
3000    
3001     44. pcretest has been upgraded by the addition of the -C option. This causes it
3002     to extract all the available output from the new pcre_config() function, and to
3003     output it. The program then exits immediately.
3004    
3005     45. A need has arisen to pass over additional data with calls to pcre_exec() in
3006     order to support additional features. One way would have been to define
3007     pcre_exec2() (for example) with extra arguments, but this would not have been
3008     extensible, and would also have required all calls to the original function to
3009     be mapped to the new one. Instead, I have chosen to extend the mechanism that
3010     is used for passing in "extra" data from pcre_study().
3011    
3012     The pcre_extra structure is now exposed and defined in pcre.h. It currently
3013     contains the following fields:
3014    
3015     flags a bitmap indicating which of the following fields are set
3016     study_data opaque data from pcre_study()
3017     match_limit a way of specifying a limit on match() calls for a specific
3018     call to pcre_exec()
3019     callout_data data for callouts (see 49 below)
3020    
3021     The flag bits are also defined in pcre.h, and are
3022    
3023     PCRE_EXTRA_STUDY_DATA
3024     PCRE_EXTRA_MATCH_LIMIT
3025     PCRE_EXTRA_CALLOUT_DATA
3026    
3027     The pcre_study() function now returns one of these new pcre_extra blocks, with
3028     the actual study data pointed to by the study_data field, and the
3029     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
3030     before. That is, this change is entirely upwards-compatible and requires no
3031     change to existing code.
3032    
3033     If you want to pass in additional data to pcre_exec(), you can either place it
3034     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
3035     block.
3036    
3037     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
3038     data string contains the escape sequence \M, pcretest calls pcre_exec() several
3039     times with different match limits, until it finds the minimum value needed for
3040     pcre_exec() to complete. The value is then output. This can be instructive; for
3041     most simple matches the number is quite small, but for pathological cases it
3042     gets very large very quickly.
3043    
3044     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
3045     returns the size of the data block pointed to by the study_data field in a
3046     pcre_extra block, that is, the value that was passed as the argument to
3047     pcre_malloc() when PCRE was getting memory in which to place the information
3048     created by pcre_study(). The fourth argument should point to a size_t variable.
3049     pcretest has been extended so that this information is shown after a successful
3050     pcre_study() call when information about the compiled regex is being displayed.
3051    
3052     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
3053     because what follows is always an absolute path. (Later: it turns out that this
3054     is more than cosmetic for MinGW, because it doesn't like empty path
3055     components.)
3056    
3057     49. Some changes have been made to the callout feature (see 28 above):
3058    
3059     (i) A callout function now has three choices for what it returns:
3060    
3061     0 => success, carry on matching
3062     > 0 => failure at this point, but backtrack if possible
3063     < 0 => serious error, return this value from pcre_exec()
3064    
3065     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
3066     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
3067     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
3068     use by callout functions. It will never be used by PCRE itself.
3069    
3070     (ii) The pcre_extra structure (see 45 above) has a void * field called
3071     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
3072     pcre_callout_block structure has a field of the same name. The contents of
3073     the field passed in the pcre_extra structure are passed to the callout
3074     function in the corresponding field in the callout block. This makes it
3075     easier to use the same callout-containing regex from multiple threads. For
3076     testing, the pcretest program has a new data escape
3077    
3078     \C*n pass the number n (may be negative) as callout_data
3079    
3080     If the callout function in pcretest receives a non-zero value as
3081     callout_data, it returns that value.
3082    
3083     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
3084     there were some redundant $(CFLAGS) in commands that are now specified as
3085     $(LINK), which already includes $(CFLAGS).
3086    
3087     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
3088     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
3089     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
3090     one-byte characters throughout. Note that case-insensitive matching applies
3091     only to characters whose values are less than 256. PCRE doesn't support the
3092     notion of cases for higher-valued characters.
3093    
3094     (i) A character class whose characters are all within 0-255 is handled as
3095     a bit map, and the map is inverted for negative classes. Previously, a
3096     character > 255 always failed to match such a class; however it should
3097     match if the class was a negative one (e.g. [^ab]). This has been fixed.
3098    
3099     (ii) A negated character class with a single character < 255 is coded as
3100     "not this character" (OP_NOT). This wasn't working properly when the test
3101     character was multibyte, either singly or repeated.
3102    
3103     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
3104     mode, for example: \x{100}{2,3}.
3105    
3106     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
3107     singly or repeated) now correctly test multibyte characters. However,
3108     PCRE doesn't recognize any characters with values greater than 255 as
3109     digits, spaces, or word characters. Such characters always match \D, \S,
3110     and \W, and never match \d, \s, or \w.
3111    
3112     (v) Classes may now contain characters and character ranges with values
3113     greater than 255. For example: [ab\x{100}-\x{400}].
3114    
3115     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
3116     PCRE in UTF-8 mode.
3117    
3118     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
3119     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
3120     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
3121     value.)
3122    
3123     53. The single man page has become too large. I have therefore split it up into
3124     a number of separate man pages. These also give rise to individual HTML pages;
3125     these are now put in a separate directory, and there is an index.html page that
3126     lists them all. Some hyperlinking between the pages has been installed.
3127    
3128     54. Added convenience functions for handling named capturing parentheses.
3129    
3130     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
3131     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
3132     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
3133     are faulted.
3134    
3135     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
3136     calling configure. These values are used when compiling the dftables.c program
3137     which is run to generate the source of the default character tables. They
3138     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
3139     you will need to set these values.
3140    
3141     57. Updated the building process for Windows DLL, as provided by Fred Cox.
3142    
3143    
3144     Version 3.9 02-Jan-02
3145 nigel 61 ---------------------
3146    
3147     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
3148    
3149     2. If --disable-static was given, the building process failed when trying to
3150     build pcretest and pcregrep. (For some reason it was using libtool to compile
3151     them, which is not right, as they aren't part of the library.)
3152    
3153    
3154 nigel 59 Version 3.8 18-Dec-01
3155     ---------------------
3156    
3157     1. The experimental UTF-8 code was completely screwed up. It was packing the
3158     bytes in the wrong order. How dumb can you get?
3159    
3160    
3161 nigel 57 Version 3.7 29-Oct-01
3162     ---------------------
3163    
3164     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
3165     This caused pcretest, when used on the test data, to segfault. Unfortunately,
3166     this didn't happen under Solaris 8, where I normally test things.
3167    
3168 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
3169     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
3170     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
3171     this fix an hour or so after the initial 3.7 release.)
3172 nigel 57
3173 nigel 59
3174 nigel 55 Version 3.6 23-Oct-01
3175     ---------------------
3176    
3177     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
3178     offsets passed as NULL with zero offset count.
3179    
3180     2. The config.guess and config.sub files had not been updated when I moved to
3181     the latest autoconf.
3182    
3183    
3184 nigel 53 Version 3.5 15-Aug-01
3185     ---------------------
3186 nigel 5
3187 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
3188     had been forgotten.
3189    
3190     2. By using declared but undefined structures, we can avoid using "void"
3191     definitions in pcre.h while keeping the internal definitions of the structures
3192     private.
3193    
3194     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
3195     user point of view, this means that both static and shared libraries are built
3196     by default, but this can be individually controlled. More of the work of
3197     handling this static/shared cases is now inside libtool instead of PCRE's make
3198     file.
3199    
3200     4. The pcretest utility is now installed along with pcregrep because it is
3201     useful for users (to test regexs) and by doing this, it automatically gets
3202     relinked by libtool. The documentation has been turned into a man page, so
3203     there are now .1, .txt, and .html versions in /doc.
3204    
3205     5. Upgrades to pcregrep:
3206     (i) Added long-form option names like gnu grep.
3207     (ii) Added --help to list all options with an explanatory phrase.
3208     (iii) Added -r, --recursive to recurse into sub-directories.
3209     (iv) Added -f, --file to read patterns from a file.
3210    
3211     6. pcre_exec() was referring to its "code" argument before testing that
3212     argument for NULL (and giving an error if it was NULL).
3213    
3214     7. Upgraded Makefile.in to allow for compiling in a different directory from
3215     the source directory.
3216    
3217     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
3218     options bits, the pointer it was passed was to an int instead of to an unsigned
3219     long int. This mattered only on 64-bit systems.
3220    
3221     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
3222     generated) instead of pcre.in, which it its source. Also made the same change
3223     in several of the .c files.
3224    
3225     10. A new release of gcc defines printf() as a macro, which broke pcretest
3226     because it had an ifdef in the middle of a string argument for printf(). Fixed
3227     by using separate calls to printf().
3228    
3229     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
3230     script, to force use of CR or LF instead of \n in the source. On non-Unix
3231     systems, the value can be set in config.h.
3232    
3233     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
3234     absolute limit. Changed the text of the error message to make this clear, and
3235     likewise updated the man page.
3236    
3237     13. The limit of 99 on the number of capturing subpatterns has been removed.
3238     The new limit is 65535, which I hope will not be a "real" limit.
3239    
3240    
3241 nigel 51 Version 3.4 22-Aug-00
3242     ---------------------
3243    
3244     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
3245    
3246     2. Diagnose condition (?(0) as an error instead of crashing on matching.
3247    
3248    
3249 nigel 49 Version 3.3 01-Aug-00
3250     ---------------------
3251    
3252     1. If an octal character was given, but the value was greater than \377, it
3253     was not getting masked to the least significant bits, as documented. This could
3254     lead to crashes in some systems.
3255    
3256     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
3257     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
3258    
3259     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
3260     These just pass their arguments on to (pcre_free)(), but they are provided
3261     because some uses of PCRE bind it to non-C systems that can call its functions,
3262     but cannot call free() or pcre_free() directly.
3263    
3264     4. Add "make test" as a synonym for "make check". Corrected some comments in
3265     the Makefile.
3266    
3267     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
3268     Makefile.
3269    
3270     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
3271     command called pgrep for grepping around the active processes.
3272    
3273     7. Added the beginnings of support for UTF-8 character strings.
3274    
3275     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
3276     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
3277     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
3278     out for the ar command.)
3279    
3280    
3281 nigel 47 Version 3.2 12-May-00
3282     ---------------------
3283    
3284     This is purely a bug fixing release.
3285    
3286     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
3287     of ZA. This was just one example of several cases that could provoke this bug,
3288     which was introduced by change 9 of version 2.00. The code for breaking
3289     infinite loops after an iteration that matches an empty string was't working
3290     correctly.
3291    
3292     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
3293     when matched against abbab (for example). After matching an empty string, it
3294     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
3295     caused it to match further down the string than it should.
3296    
3297     3. The code contained an inclusion of sys/types.h. It isn't clear why this
3298     was there because it doesn't seem to be needed, and it causes trouble on some
3299     systems, as it is not a Standard C header. It has been removed.
3300    
3301     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
3302     were reported on the Macintosh. The changes were from
3303    
3304     while ((c = *(++ptr)) != 0 && c != '\n');
3305     to
3306     while ((c = *(++ptr)) != 0 && c != '\n') ;
3307    
3308     Totally extraordinary, but if that's what it takes...
3309    
3310     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
3311     available. Added HAVE_BCOPY and an autoconf test for it; if neither
3312     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
3313     assumes the way PCRE uses memmove() (always moving upwards).
3314    
3315     6. PCRE is being used in one environment where strchr() is not available. There
3316     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
3317     faster code anyway.
3318    
3319    
3320 nigel 45 Version 3.1 09-Feb-00
3321     ---------------------
3322    
3323     The only change in this release is the fixing of some bugs in Makefile.in for
3324     the "install" target:
3325    
3326     (1) It was failing to install pcreposix.h.
3327    
3328     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
3329    
3330    
3331 nigel 43 Version 3.0 01-Feb-00
3332     ---------------------
3333 nigel 41
3334     1. Add support for the /+ modifier to perltest (to output $` like it does in
3335     pcretest).
3336    
3337     2. Add support for the /g modifier to perltest.
3338    
3339     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
3340     matches null strings.
3341    
3342     4. Fix perltest so that it doesn't do unwanted things when fed an empty
3343     pattern. Perl treats empty patterns specially - it reuses the most recent
3344     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
3345     effect.
3346    
3347     5. The POSIX interface was broken in that it was just handing over the POSIX
3348     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
3349     required a bigger vector, with some working space on the end. This means that
3350     the POSIX wrapper now has to get and free some memory, and copy the results.
3351    
3352 nigel 43 6. Added some simple autoconf support, placing the test data and the
3353     documentation in separate directories, re-organizing some of the
3354     information files, and making it build pcre-config (a GNU standard). Also added
3355     libtool support for building PCRE as a shared library, which is now the
3356     default.
3357 nigel 41
3358 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
3359     09 are not valid octal constants. Single digits will be used for minor values
3360     less than 10.
3361    
3362     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
3363     existing programs that set these in the POSIX interface can use PCRE without
3364     modification.
3365    
3366     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
3367     return all that pcre_info() returns, plus additional data. The pcre_info()
3368     function is retained for compatibility, but is considered to be obsolete.
3369    
3370     10. Added experimental recursion feature (?R) to handle one common case that
3371     Perl 5.6 will be able to do with (?p{...}).
3372    
3373     11. Added support for POSIX character classes like [:alpha:], which Perl is
3374     adopting.
3375    
3376    
3377 nigel 39 Version 2.08 31-Aug-99
3378     ----------------------
3379    
3380     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
3381     trying to match at the startoffset position, but instead was moving forward to
3382     the next newline as if a previous match had failed.
3383    
3384     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
3385     and could get into a loop if a null string was matched other than at the start
3386     of the subject.
3387    
3388     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3389     be distinguished at compile time, and for completeness also added PCRE_DATE.
3390    
3391     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3392     in GnuWin32 environments.
3393    
3394    
3395 nigel 37 Version 2.07 29-Jul-99
3396     ----------------------
3397    
3398     1. The documentation is now supplied in plain text form and HTML as well as in
3399     the form of man page sources.
3400    
3401     2. C++ compilers don't like assigning (void *) values to other pointer types.
3402     In particular this affects malloc(). Although there is no problem in Standard
3403     C, I've put in casts to keep C++ compilers happy.
3404    
3405     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3406     should be (const char *).
3407    
3408     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3409     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3410     However, I haven't made this a standard facility. The documentation doesn't
3411     mention it, and the Makefile doesn't support it.
3412    
3413     5. The Makefile now contains an "install" target, with editable destinations at
3414     the top of the file. The pcretest program is not installed.
3415    
3416     6. pgrep -V now gives the PCRE version number and date.
3417    
3418     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3419     causing the entire string to be ignored, instead of just the last character.
3420    
3421     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3422     non-matching string, it can take a very, very long time, even for strings of
3423     quite modest length, because of the nested recursion. PCRE now does better in
3424     some of these cases. It does this by remembering the last required literal
3425     character in the pattern, and pre-searching the subject to ensure it is present
3426     before running the real match. In other words, it applies a heuristic to detect
3427     some types of certain failure quickly, and in the above example, if presented
3428     with a string that has no trailing " it gives "no match" very quickly.
3429    
3430     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3431     other alternatives are tried instead.
3432    
3433    
3434 nigel 35 Version 2.06 09-Jun-99
3435     ----------------------
3436    
3437     1. Change pcretest's output for amount of store used to show just the code
3438     space, because the remainder (the data block) varies in size between 32-bit and
3439     64-bit systems.
3440    
3441     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3442     start matching at. This allows lookbehinds to work when searching for multiple
3443     occurrences in a string.
3444    
3445     3. Added additional options to pcretest for testing multiple occurrences:
3446    
3447     /+ outputs the rest of the string that follows a match
3448     /g loops for multiple occurrences, using the new startoffset argument
3449     /G loops for multiple occurrences by passing an incremented pointer
3450    
3451     4. PCRE wasn't doing the "first character" optimization for patterns starting
3452     with \b or \B, though it was doing it for other lookbehind assertions. That is,
3453     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3454     the letter 'x'. On long subject strings, this gives a significant speed-up.
3455    
3456    
3457 nigel 33 Version 2.05 21-Apr-99
3458     ----------------------
3459    
3460     1. Changed the type of magic_number from int to long int so that it works
3461     properly on 16-bit systems.
3462    
3463     2. Fixed a bug which caused patterns starting with .* not to work correctly
3464     when the subject string contained newline characters. PCRE was assuming
3465     anchoring for such patterns in all cases, which is not correct because .* will
3466     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3467     DOTALL is set at top level; otherwise it knows that patterns starting with .*
3468     must be retried after every newline in the subject.
3469    
3470    
3471 nigel 31 Version 2.04 18-Feb-99
3472     ----------------------
3473    
3474     1. For parenthesized subpatterns with repeats whose minimum was zero, the
3475     computation of the store needed to hold the pattern was incorrect (too large).
3476     If such patterns were nested a few deep, this could multiply and become a real
3477     problem.
3478    
3479     2. Added /M option to pcretest to show the memory requirement of a specific
3480     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3481    
3482     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3483     compiled in such a way that the backtracking after subsequent failure was
3484     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3485     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3486    
3487    
3488 nigel 29 Version 2.03 02-Feb-99
3489     ----------------------
3490    
3491     1. Fixed typo and small mistake in man page.
3492    
3493 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
3494     LICENCE file containing the conditions.
3495 nigel 29
3496     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3497     Perl, that is the internal \ allows the delimiter to be included in the
3498     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3499     the final delimiter, add \ to the end of the pattern (to test the error).
3500    
3501     4. Added the convenience functions for extracting substrings after a successful
3502     match. Updated pcretest to make it able to test these functions.
3503    
3504    
3505 nigel 27 Version 2.02 14-Jan-99
3506     ----------------------
3507    
3508     1. Initialized the working variables associated with each extraction so that
3509     their saving and restoring doesn't refer to uninitialized store.
3510    
3511     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3512     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3513     fix the problem.
3514    
3515     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3516     calls, and wasn't printing the correct value for compiling calls. Increased the
3517     default value of LOOPREPEAT, and the number of significant figures in the
3518     times.
3519    
3520     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3521    
3522     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3523     a building problem on Windows NT with a FAT file system.
3524    
3525    
3526 nigel 25 Version 2.01 21-Oct-98
3527     ----------------------
3528    
3529     1. Changed the API for pcre_compile() to allow for the provision of a pointer
3530     to character tables built by pcre_maketables() in the current locale. If NULL
3531     is passed, the default tables are used.
3532    
3533    
3534 nigel 23 Version 2.00 24-Sep-98
3535 nigel 21 ----------------------
3536    
3537 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3538     it any more.
3539    
3540     2. Allow quantification of (?>) groups, and make it work correctly.
3541    
3542     3. The first character computation wasn't working for (?>) groups.
3543    
3544     4. Correct the implementation of \Z (it is permitted to match on the \n at the
3545     end of the subject) and add 5.005's \z, which really does match only at the
3546     very end of the subject.
3547    
3548     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3549    
3550     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3551     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3552     localized options. All options to pcre_study() were also removed.
3553    
3554     7. Add other new features from 5.005:
3555    
3556     $(?<= positive lookbehind
3557     $(?<! negative lookbehind
3558     (?imsx-imsx) added the unsetting capability
3559     such a setting is global if at outer level; local otherwise
3560     (?imsx-imsx:) non-capturing groups with option setting
3561     (?(cond)re|re) conditional pattern matching
3562    
3563     A backreference to itself in a repeated group matches the previous
3564     captured string.
3565    
3566     8. General tidying up of studying (both automatic and via "study")
3567     consequential on the addition of new assertions.
3568    
3569     9. As in 5.005, unlimited repeated groups that could match an empty substring
3570     are no longer faulted at compile time. Instead, the loop is forcibly broken at
3571     runtime if any iteration does actually match an empty substring.
3572    
3573     10. Include the RunTest script in the distribution.
3574    
3575     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3576     discrepancies, some of which were old and were also with respect to 5.004. They
3577     have now been fixed.
3578    
3579    
3580     Version 1.09 28-Apr-98
3581     ----------------------
3582    
3583 nigel 21 1. A negated single character class followed by a quantifier with a minimum
3584     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3585     program crashes, or just wrong answers. This did not apply to negated classes
3586     containing more than one character, or to minima other than one.
3587    
3588    
3589 nigel 19 Version 1.08 27-Mar-98
3590     ----------------------
3591    
3592     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3593    
3594     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3595     latter must appear before anything that relies on it in the pattern.
3596    
3597    
3598 nigel 17 Version 1.07 16-Feb-98
3599     ----------------------
3600    
3601     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3602     repeat of a potentially empty string).
3603    
3604    
3605 nigel 15 Version 1.06 23-Jan-98
3606     ----------------------
3607    
3608     1. Added Markus Oberhumer's little patches for C++.
3609    
3610     2. Literal strings longer than 255 characters were broken.
3611    
3612    
3613 nigel 13 Version 1.05 23-Dec-97
3614     ----------------------
3615    
3616     1. Negated character classes containing more than one character were failing if
3617     PCRE_CASELESS was set at run time.
3618    
3619    
3620 nigel 11 Version 1.04 19-Dec-97
3621     ----------------------
3622    
3623     1. Corrected the man page, where some "const" qualifiers had been omitted.
3624    
3625     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3626     input syntax.
3627    
3628     3. Fixed memory leak which occurred when a regex with back references was
3629     matched with an offsets vector that wasn't big enough. The temporary memory
3630     that is used in this case wasn't being freed if the match failed.
3631    
3632     4. Tidied pcretest to ensure it frees memory that it gets.
3633    
3634     5. Temporary memory was being obtained in the case where the passed offsets
3635     vector was exactly big enough.
3636    
3637     6. Corrected definition of offsetof() from change 5 below.
3638    
3639     7. I had screwed up change 6 below and broken the rules for the use of
3640     setjmp(). Now fixed.
3641    
3642    
3643 nigel 9 Version 1.03 18-Dec-97
3644     ----------------------
3645    
3646     1. A erroneous regex with a missing opening parenthesis was correctly
3647     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3648     on some systems.
3649    
3650     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3651     it was reported that one broken compiler failed on the former because "code" is
3652     also an independent variable.
3653    
3654     3. The erroneous regex a[]b caused an array overrun reference.
3655    
3656     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3657     fail on data ending with that character. (It was going on too far, and checking
3658     the next character, typically a binary zero.) This was specific to the
3659     optimized code for single-character negative classes.
3660    
3661     5. Added a contributed patch from the TIN world which does the following:
3662    
3663     + Add an undef for memmove, in case the the system defines a macro for it.
3664    
3665     + Add a definition of offsetof(), in case there isn't one. (I don't know
3666     the reason behind this - offsetof() is part of the ANSI standard - but
3667     it does no harm).
3668    
3669     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3670     most of the places where whitespace preceded '#'. I have given up and
3671     allowed the remaining 2 cases to be at the margin.
3672    
3673     + Rename some variables in pcre to eliminate shadowing. This seems very
3674     pedantic, but does no harm, of course.
3675    
3676     6. Moved the call to setjmp() into its own function, to get rid of warnings
3677     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3678    
3679     7. Constructs such as \d{8,} were compiling into the equivalent of
3680 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3681 nigel 9 outcome, but in this particular case used more store than had been allocated,
3682     which caused the bug to be discovered because it threw up an internal error.
3683    
3684     8. The debugging code in both pcre and pcretest for outputting the compiled
3685     form of a regex was going wrong in the case of back references followed by
3686     curly-bracketed repeats.
3687    
3688    
3689 nigel 7 Version 1.02 12-Dec-97
3690     ----------------------
3691    
3692     1. Typos in pcre.3 and comments in the source fixed.
3693    
3694     2. Applied a contributed patch to get rid of places where it used to remove
3695     'const' from variables, and fixed some signed/unsigned and uninitialized
3696     variable warnings.
3697    
3698     3. Added the "runtest" target to Makefile.
3699    
3700     4. Set default compiler flag to -O2 rather than just -O.
3701    
3702    
3703 nigel 5 Version 1.01 19-Nov-97
3704     ----------------------
3705    
3706     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3707     like /([ab]*)*/, that is, for classes with more than one character in them.
3708    
3709     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3710     as /((?>a*))*/ (a PCRE_EXTRA facility).
3711    
3712    
3713     Version 1.00 18-Nov-97
3714     ----------------------
3715    
3716     1. Added compile-time macros to support systems such as SunOS4 which don't have
3717     memmove() or strerror() but have other things that can be used instead.
3718    
3719     2. Arranged that "make clean" removes the executables.
3720    
3721    
3722 nigel 3 Version 0.99 27-Oct-97
3723     ----------------------
3724    
3725     1. Fixed bug in code for optimizing classes with only one character. It was
3726     initializing a 32-byte map regardless, which could cause it to run off the end
3727     of the memory it had got.
3728    
3729     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3730    
3731    
3732     Version 0.98 22-Oct-97
3733     ----------------------
3734    
3735     1. Fixed bug in code for handling temporary memory usage when there are more
3736     back references than supplied space in the ovector. This could cause segfaults.
3737    
3738    
3739     Version 0.97 21-Oct-97
3740     ----------------------
3741    
3742     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3743    
3744     2. Optimized negated single characters not to use a bit map.
3745    </