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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 369 - (show annotations) (download)
Sun Aug 24 16:53:47 2008 UTC (6 years ago) by ph10
File size: 159537 byte(s)
Patch to reduce warnings from certain compilers.

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

Properties

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12