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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 366 - (show annotations) (download)
Mon Jul 14 15:45:32 2008 UTC (6 years ago) by ph10
File size: 159253 byte(s)
Lazy qualifiers were not always working in UTF-8 mode (8.0/15).

1 ChangeLog for PCRE
2 ------------------
3
4 Version 8.0 02 Jul-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
70 Version 7.7 07-May-08
71 ---------------------
72
73 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
74 a string to a long long, pretend we don't even have a long long." This is
75 done by checking for the strtoq, strtoll, and _strtoi64 functions.
76
77 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
78 pre-7.6 versions, which defined a global no_arg variable instead of putting
79 it in the RE class. (See also #8 below.)
80
81 3. Remove a line of dead code, identified by coverity and reported by Nuno
82 Lopes.
83
84 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
85
86 (1) The include/exclude patterns were being applied to the whole pathnames
87 of files, instead of just to the final components.
88
89 (2) If there was more than one level of directory, the subdirectories were
90 skipped unless they satisfied the include/exclude conditions. This is
91 inconsistent with GNU grep (and could even be seen as contrary to the
92 pcregrep specification - which I improved to make it absolutely clear).
93 The action now is always to scan all levels of directory, and just
94 apply the include/exclude patterns to regular files.
95
96 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
97 --exclude_dir in the tests to avoid scanning .svn directories.
98
99 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
100 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
101 doesn't support NULs in patterns.
102
103 7. Added some missing "const"s to declarations of static tables in
104 pcre_compile.c and pcre_dfa_exec.c.
105
106 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
107 caused by fix #2 above. (Subsequently also a second patch to fix the
108 first patch. And a third patch - this was a messy problem.)
109
110 9. Applied Craig's patch to remove the use of push_back().
111
112 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
113 matching function regexec().
114
115 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
116 which, however, unlike Perl's \g{...}, are subroutine calls, not back
117 references. PCRE supports relative numbers with this syntax (I don't think
118 Oniguruma does).
119
120 12. Previously, a group with a zero repeat such as (...){0} was completely
121 omitted from the compiled regex. However, this means that if the group
122 was called as a subroutine from elsewhere in the pattern, things went wrong
123 (an internal error was given). Such groups are now left in the compiled
124 pattern, with a new opcode that causes them to be skipped at execution
125 time.
126
127 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
128 to the way PCRE behaves:
129
130 (a) A lone ] character is dis-allowed (Perl treats it as data).
131
132 (b) A back reference to an unmatched subpattern matches an empty string
133 (Perl fails the current match path).
134
135 (c) A data ] in a character class must be notated as \] because if the
136 first data character in a class is ], it defines an empty class. (In
137 Perl it is not possible to have an empty class.) The empty class []
138 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
139 The negative empty class [^] matches any one character, independently
140 of the DOTALL setting.
141
142 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
143 non-existent subpattern following a character class starting with ']' and
144 containing () gave an internal compiling error instead of "reference to
145 non-existent subpattern". Fortunately, when the pattern did exist, the
146 compiled code was correct. (When scanning forwards to check for the
147 existencd of the subpattern, it was treating the data ']' as terminating
148 the class, so got the count wrong. When actually compiling, the reference
149 was subsequently set up correctly.)
150
151 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
152 it was being rejected as not supported by pcre_dfa_exec(), even though
153 other assertions are supported. I have made pcre_dfa_exec() support
154 (*FAIL).
155
156 16. The implementation of 13c above involved the invention of a new opcode,
157 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
158 cannot be changed at match time, I realized I could make a small
159 improvement to matching performance by compiling OP_ALLANY instead of
160 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
161 on the OP_ANY path.
162
163 17. Compiling pcretest on Windows with readline support failed without the
164 following two fixes: (1) Make the unistd.h include conditional on
165 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
166
167 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
168 ncurses library to be included for pcretest when ReadLine support is
169 requested, but also to allow for it to be overridden. This patch came from
170 Daniel Bergström.
171
172 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
173 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
174 any errors with the current Unicode tables. Thanks to Peter Kankowski for
175 spotting this.
176
177
178 Version 7.6 28-Jan-08
179 ---------------------
180
181 1. A character class containing a very large number of characters with
182 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
183 overflow.
184
185 2. Patch to cut out the "long long" test in pcrecpp_unittest when
186 HAVE_LONG_LONG is not defined.
187
188 3. Applied Christian Ehrlicher's patch to update the CMake build files to
189 bring them up to date and include new features. This patch includes:
190
191 - Fixed PH's badly added libz and libbz2 support.
192 - Fixed a problem with static linking.
193 - Added pcredemo. [But later removed - see 7 below.]
194 - Fixed dftables problem and added an option.
195 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
196 HAVE_LONG_LONG.
197 - Added readline support for pcretest.
198 - Added an listing of the option settings after cmake has run.
199
200 4. A user submitted a patch to Makefile that makes it easy to create
201 "pcre.dll" under mingw when using Configure/Make. I added stuff to
202 Makefile.am that cause it to include this special target, without
203 affecting anything else. Note that the same mingw target plus all
204 the other distribution libraries and programs are now supported
205 when configuring with CMake (see 6 below) instead of with
206 Configure/Make.
207
208 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
209 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
210 exported in the Windows port". It has not yet been confirmed that the patch
211 solves the problem, but it does no harm.
212
213 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
214 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
215 with CMake, and also correct the comment about stack recursion.
216
217 7. Remove the automatic building of pcredemo from the ./configure system and
218 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
219 of a program that users should build themselves after PCRE is installed, so
220 building it automatically is not really right. What is more, it gave
221 trouble in some build environments.
222
223 8. Further tidies to CMakeLists.txt from Sheri and Christian.
224
225
226 Version 7.5 10-Jan-08
227 ---------------------
228
229 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
230 values in parens when parsing an RE using the C++ wrapper."
231
232 2. Negative specials like \S did not work in character classes in UTF-8 mode.
233 Characters greater than 255 were excluded from the class instead of being
234 included.
235
236 3. The same bug as (2) above applied to negated POSIX classes such as
237 [:^space:].
238
239 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
240 defined or documented. It seems to have been a typo for PCRE_STATIC, so
241 I have changed it.
242
243 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
244 first named subpattern) and a construct such as (?&a) would reference the
245 first named subpattern whose name started with "a" (in other words, the
246 length check was missing). Both these problems are fixed. "Subpattern name
247 expected" is now given for (?&) (a zero-length name), and this patch also
248 makes it give the same error for \k'' (previously it complained that that
249 was a reference to a non-existent subpattern).
250
251 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
252 this is right because (?- can be followed by option settings as well as by
253 digits. I have, however, made the messages clearer.
254
255 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
256 than the number used in the conditional) now cause a compile-time error.
257 This is actually not compatible with Perl, which accepts such patterns, but
258 treats the conditional as always being FALSE (as PCRE used to), but it
259 seems to me that giving a diagnostic is better.
260
261 8. Change "alphameric" to the more common word "alphanumeric" in comments
262 and messages.
263
264 9. Fix two occurrences of "backslash" in comments that should have been
265 "backspace".
266
267 10. Remove two redundant lines of code that can never be obeyed (their function
268 was moved elsewhere).
269
270 11. The program that makes PCRE's Unicode character property table had a bug
271 which caused it to generate incorrect table entries for sequences of
272 characters that have the same character type, but are in different scripts.
273 It amalgamated them into a single range, with the script of the first of
274 them. In other words, some characters were in the wrong script. There were
275 thirteen such cases, affecting characters in the following ranges:
276
277 U+002b0 - U+002c1
278 U+0060c - U+0060d
279 U+0061e - U+00612
280 U+0064b - U+0065e
281 U+0074d - U+0076d
282 U+01800 - U+01805
283 U+01d00 - U+01d77
284 U+01d9b - U+01dbf
285 U+0200b - U+0200f
286 U+030fc - U+030fe
287 U+03260 - U+0327f
288 U+0fb46 - U+0fbb1
289 U+10450 - U+1049d
290
291 12. The -o option (show only the matching part of a line) for pcregrep was not
292 compatible with GNU grep in that, if there was more than one match in a
293 line, it showed only the first of them. It now behaves in the same way as
294 GNU grep.
295
296 13. If the -o and -v options were combined for pcregrep, it printed a blank
297 line for every non-matching line. GNU grep prints nothing, and pcregrep now
298 does the same. The return code can be used to tell if there were any
299 non-matching lines.
300
301 14. Added --file-offsets and --line-offsets to pcregrep.
302
303 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
304 infinitely looping recursion. The bug was that positive lookaheads were not
305 being skipped when checking for a possible empty match (negative lookaheads
306 and both kinds of lookbehind were skipped).
307
308 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
309 inclusion of <windows.h> to before rather than after the definition of
310 INVALID_FILE_ATTRIBUTES (patch from David Byron).
311
312 17. Specifying a possessive quantifier with a specific limit for a Unicode
313 character property caused pcre_compile() to compile bad code, which led at
314 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
315 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
316 caused the error; without that there was no problem.
317
318 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
319
320 19. Added --enable-pcretest-libreadline.
321
322 20. In pcrecpp.cc, the variable 'count' was incremented twice in
323 RE::GlobalReplace(). As a result, the number of replacements returned was
324 double what it should be. I removed one of the increments, but Craig sent a
325 later patch that removed the other one (the right fix) and added unit tests
326 that check the return values (which was not done before).
327
328 21. Several CMake things:
329
330 (1) Arranged that, when cmake is used on Unix, the libraries end up with
331 the names libpcre and libpcreposix, not just pcre and pcreposix.
332
333 (2) The above change means that pcretest and pcregrep are now correctly
334 linked with the newly-built libraries, not previously installed ones.
335
336 (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
337
338 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
339 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
340 UTF-8 newline character). The key issue is that the pattern starts .*;
341 this means that the match must be either at the beginning, or after a
342 newline. The bug was in the code for advancing after a failed match and
343 checking that the new position followed a newline. It was not taking
344 account of UTF-8 characters correctly.
345
346 23. PCRE was behaving differently from Perl in the way it recognized POSIX
347 character classes. PCRE was not treating the sequence [:...:] as a
348 character class unless the ... were all letters. Perl, however, seems to
349 allow any characters between [: and :], though of course it rejects as
350 unknown any "names" that contain non-letters, because all the known class
351 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
352 for example, whereas PCRE did not - it did not recognize a POSIX character
353 class. This seemed a bit dangerous, so the code has been changed to be
354 closer to Perl. The behaviour is not identical to Perl, because PCRE will
355 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
356 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
357 Perl does, and where it didn't before.
358
359 24. Rewrite so as to remove the single use of %n from pcregrep because in some
360 Windows environments %n is disabled by default.
361
362
363 Version 7.4 21-Sep-07
364 ---------------------
365
366 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
367 means that a class such as [\s] counted as "explicit reference to CR or
368 LF". That isn't really right - the whole point of the change was to try to
369 help when there was an actual mention of one of the two characters. So now
370 the change happens only if \r or \n (or a literal CR or LF) character is
371 encountered.
372
373 2. The 32-bit options word was also used for 6 internal flags, but the numbers
374 of both had grown to the point where there were only 3 bits left.
375 Fortunately, there was spare space in the data structure, and so I have
376 moved the internal flags into a new 16-bit field to free up more option
377 bits.
378
379 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
380 but did not set the internal JCHANGED flag - either of these is enough to
381 control the way the "get" function works - but the PCRE_INFO_JCHANGED
382 facility is supposed to tell if (?J) was ever used, so now (?J) at the
383 start sets both bits.
384
385 4. Added options (at build time, compile time, exec time) to change \R from
386 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
387
388 5. doc/pcresyntax.html was missing from the distribution.
389
390 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
391 compatibility, even though it is no longer used.
392
393 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
394 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
395 windows.h file is present (where different names are used). [This was
396 reversed later after testing - see 16 below.]
397
398 8. Changed all #include <config.h> to #include "config.h". There were also
399 some further <pcre.h> cases that I changed to "pcre.h".
400
401 9. When pcregrep was used with the --colour option, it missed the line ending
402 sequence off the lines that it output.
403
404 10. It was pointed out to me that arrays of string pointers cause lots of
405 relocations when a shared library is dynamically loaded. A technique of
406 using a single long string with a table of offsets can drastically reduce
407 these. I have refactored PCRE in four places to do this. The result is
408 dramatic:
409
410 Originally: 290
411 After changing UCP table: 187
412 After changing error message table: 43
413 After changing table of "verbs" 36
414 After changing table of Posix names 22
415
416 Thanks to the folks working on Gregex for glib for this insight.
417
418 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
419 unicode-properties was also set.
420
421 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
422
423 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
424 checked only for CRLF.
425
426 14. Added casts to pcretest.c to avoid compiler warnings.
427
428 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
429
430 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
431 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
432 entirely. This removes changes made in 7 above.
433
434 17. The CMake files have been updated, and there is now more information about
435 building with CMake in the NON-UNIX-USE document.
436
437
438 Version 7.3 28-Aug-07
439 ---------------------
440
441 1. In the rejigging of the build system that eventually resulted in 7.1, the
442 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
443 brackets there is not right, since it causes compilers to look for an
444 installed pcre.h, not the version that is in the source that is being
445 compiled (which of course may be different). I have changed it back to:
446
447 #include "pcre.h"
448
449 I have a vague recollection that the change was concerned with compiling in
450 different directories, but in the new build system, that is taken care of
451 by the VPATH setting the Makefile.
452
453 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
454 when the subject happened to end in the byte 0x85 (e.g. if the last
455 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
456 characters but of course it shouldn't be taken as a newline when it is part
457 of another character. The bug was that, for an unlimited repeat of . in
458 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
459 characters when looking for a newline.
460
461 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
462
463 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
464 in debug output.
465
466 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
467 long printing in the pcrecpp unittest when running under MinGW.
468
469 6. ESC_K was left out of the EBCDIC table.
470
471 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
472 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
473 limit also applies to "virtual nesting" when a pattern is recursive, and in
474 this case 1000 isn't so big. I have been able to remove this limit at the
475 expense of backing off one optimization in certain circumstances. Normally,
476 when pcre_exec() would call its internal match() function recursively and
477 immediately return the result unconditionally, it uses a "tail recursion"
478 feature to save stack. However, when a subpattern that can match an empty
479 string has an unlimited repetition quantifier, it no longer makes this
480 optimization. That gives it a stack frame in which to save the data for
481 checking that an empty string has been matched. Previously this was taken
482 from the 1000-entry workspace that had been reserved. So now there is no
483 explicit limit, but more stack is used.
484
485 8. Applied Daniel's patches to solve problems with the import/export magic
486 syntax that is required for Windows, and which was going wrong for the
487 pcreposix and pcrecpp parts of the library. These were overlooked when this
488 problem was solved for the main library.
489
490 9. There were some crude static tests to avoid integer overflow when computing
491 the size of patterns that contain repeated groups with explicit upper
492 limits. As the maximum quantifier is 65535, the maximum group length was
493 set at 30,000 so that the product of these two numbers did not overflow a
494 32-bit integer. However, it turns out that people want to use groups that
495 are longer than 30,000 bytes (though not repeat them that many times).
496 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
497 made it possible to implement the integer overflow checks in a much more
498 dynamic way, which I have now done. The artificial limitation on group
499 length has been removed - we now have only the limit on the total length of
500 the compiled pattern, which depends on the LINK_SIZE setting.
501
502 10. Fixed a bug in the documentation for get/copy named substring when
503 duplicate names are permitted. If none of the named substrings are set, the
504 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
505 empty string.
506
507 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
508 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
509 because the ] is interpreted as the first data character and the
510 terminating ] is not found. PCRE has been made compatible with Perl in this
511 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
512 cause memory overwriting.
513
514 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
515 string has been matched (to stop an infinite loop). It was not recognizing
516 a conditional subpattern that could match an empty string if that
517 subpattern was within another subpattern. For example, it looped when
518 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
519 condition was not nested. This bug has been fixed.
520
521 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
522 past the start of the subject in the presence of bytes with the top bit
523 set, for example "\x8aBCD".
524
525 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
526 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
527
528 14. Optimized (?!) to (*FAIL).
529
530 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
531 This restricts code points to be within the range 0 to 0x10FFFF, excluding
532 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
533 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
534 does: it's just the validity check that is more restrictive.
535
536 16. Inserted checks for integer overflows during escape sequence (backslash)
537 processing, and also fixed erroneous offset values for syntax errors during
538 backslash processing.
539
540 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
541 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
542
543 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
544 caused an overrun.
545
546 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
547 something other than just ASCII characters) inside a group that had an
548 unlimited repeat caused a loop at compile time (while checking to see
549 whether the group could match an empty string).
550
551 20. Debugging a pattern containing \p or \P could cause a crash. For example,
552 [\P{Any}] did so. (Error in the code for printing property names.)
553
554 21. An orphan \E inside a character class could cause a crash.
555
556 22. A repeated capturing bracket such as (A)? could cause a wild memory
557 reference during compilation.
558
559 23. There are several functions in pcre_compile() that scan along a compiled
560 expression for various reasons (e.g. to see if it's fixed length for look
561 behind). There were bugs in these functions when a repeated \p or \P was
562 present in the pattern. These operators have additional parameters compared
563 with \d, etc, and these were not being taken into account when moving along
564 the compiled data. Specifically:
565
566 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
567 length.
568
569 (b) An item such as \pL+ within a repeated group could cause crashes or
570 loops.
571
572 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
573 "reference to non-existent subpattern" error.
574
575 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
576
577 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
578 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
579
580 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
581
582 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
583 character were causing crashes (broken optimization).
584
585 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
586 \p or \P) caused a compile-time loop.
587
588 28. More problems have arisen in unanchored patterns when CRLF is a valid line
589 break. For example, the unstudied pattern [\r\n]A does not match the string
590 "\r\nA" because change 7.0/46 below moves the current point on by two
591 characters after failing to match at the start. However, the pattern \nA
592 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
593 the same is true. There doesn't seem any very clean way out of this, but
594 what I have chosen to do makes the common cases work: PCRE now takes note
595 of whether there can be an explicit match for \r or \n anywhere in the
596 pattern, and if so, 7.0/46 no longer applies. As part of this change,
597 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
598 pattern has explicit CR or LF references.
599
600 29. Added (*CR) etc for changing newline setting at start of pattern.
601
602
603 Version 7.2 19-Jun-07
604 ---------------------
605
606 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
607 which is apparently normally available under Windows.
608
609 2. Re-jig the pcregrep tests with different newline settings in an attempt
610 to make them independent of the local environment's newline setting.
611
612 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
613
614 4. Some of the "internals" tests were previously cut out when the link size
615 was not 2, because the output contained actual offsets. The recent new
616 "Z" feature of pcretest means that these can be cut out, making the tests
617 usable with all link sizes.
618
619 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
620 stack recursion. This gives a massive performance boost under BSD, but just
621 a small improvement under Linux. However, it saves one field in the frame
622 in all cases.
623
624 6. Added more features from the forthcoming Perl 5.10:
625
626 (a) (?-n) (where n is a string of digits) is a relative subroutine or
627 recursion call. It refers to the nth most recently opened parentheses.
628
629 (b) (?+n) is also a relative subroutine call; it refers to the nth next
630 to be opened parentheses.
631
632 (c) Conditions that refer to capturing parentheses can be specified
633 relatively, for example, (?(-2)... or (?(+3)...
634
635 (d) \K resets the start of the current match so that everything before
636 is not part of it.
637
638 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
639
640 (f) \g{name} is another synonym - part of Perl 5.10's unification of
641 reference syntax.
642
643 (g) (?| introduces a group in which the numbering of parentheses in each
644 alternative starts with the same number.
645
646 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
647
648 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
649 PCRE_INFO_JCHANGED.
650
651 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
652 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
653 for detecting groups that can match an empty string.
654
655 9. A pattern with a very large number of alternatives (more than several
656 hundred) was running out of internal workspace during the pre-compile
657 phase, where pcre_compile() figures out how much memory will be needed. A
658 bit of new cunning has reduced the workspace needed for groups with
659 alternatives. The 1000-alternative test pattern now uses 12 bytes of
660 workspace instead of running out of the 4096 that are available.
661
662 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
663
664 11. Applied patch from Google to remove an optimization that didn't quite work.
665 The report of the bug said:
666
667 pcrecpp::RE("a*").FullMatch("aaa") matches, while
668 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
669 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
670
671 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
672 it matched the wrong number of bytes.
673
674
675 Version 7.1 24-Apr-07
676 ---------------------
677
678 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
679 that is more "standard", making use of automake and other Autotools. There
680 is some re-arrangement of the files and adjustment of comments consequent
681 on this.
682
683 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
684 for recursive directory scanning broke on some systems because the files
685 are not scanned in any specific order and on different systems the order
686 was different. A call to "sort" has been inserted into RunGrepTest for the
687 approprate test as a short-term fix. In the longer term there may be an
688 alternative.
689
690 3. I had an email from Eric Raymond about problems translating some of PCRE's
691 man pages to HTML (despite the fact that I distribute HTML pages, some
692 people do their own conversions for various reasons). The problems
693 concerned the use of low-level troff macros .br and .in. I have therefore
694 removed all such uses from the man pages (some were redundant, some could
695 be replaced by .nf/.fi pairs). The 132html script that I use to generate
696 HTML has been updated to handle .nf/.fi and to complain if it encounters
697 .br or .in.
698
699 4. Updated comments in configure.ac that get placed in config.h.in and also
700 arranged for config.h to be included in the distribution, with the name
701 config.h.generic, for the benefit of those who have to compile without
702 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
703
704 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
705 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
706 makevp.bat for the current PCRE, using the additional files
707 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
708
709 6. A Windows user reported a minor discrepancy with test 2, which turned out
710 to be caused by a trailing space on an input line that had got lost in his
711 copy. The trailing space was an accident, so I've just removed it.
712
713 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
714 that is needed.
715
716 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
717 as "const" (a) because they are and (b) because it helps the PHP
718 maintainers who have recently made a script to detect big data structures
719 in the php code that should be moved to the .rodata section. I remembered
720 to update Builducptable as well, so it won't revert if ucptable.h is ever
721 re-created.
722
723 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
724 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
725 order to be able to cut out the UTF-8 tables in the latter when UTF-8
726 support is not required. This saves 1.5-2K of code, which is important in
727 some applications.
728
729 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
730 so as not to refer to the tables, even though these functions will never be
731 called when UTF-8 support is disabled. Otherwise there are problems with a
732 shared library.
733
734 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
735
736 (a) It was defining its arguments as char * instead of void *.
737
738 (b) It was assuming that all moves were upwards in memory; this was true
739 a long time ago when I wrote it, but is no longer the case.
740
741 The emulated memove() is provided for those environments that have neither
742 memmove() nor bcopy(). I didn't think anyone used it these days, but that
743 is clearly not the case, as these two bugs were recently reported.
744
745 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
746 and Detrail to create the HTML documentation, the .txt form of the man
747 pages, and it removes trailing spaces from listed files. It also creates
748 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
749 case, it wraps all the #defines with #ifndefs. This script should be run
750 before "make dist".
751
752 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
753 with Unicode property support.
754
755 (a) For a maximizing quantifier, if the two different cases of the
756 character were of different lengths in their UTF-8 codings (there are
757 some cases like this - I found 11), and the matching function had to
758 back up over a mixture of the two cases, it incorrectly assumed they
759 were both the same length.
760
761 (b) When PCRE was configured to use the heap rather than the stack for
762 recursion during matching, it was not correctly preserving the data for
763 the other case of a UTF-8 character when checking ahead for a match
764 while processing a minimizing repeat. If the check also involved
765 matching a wide character, but failed, corruption could cause an
766 erroneous result when trying to check for a repeat of the original
767 character.
768
769 13. Some tidying changes to the testing mechanism:
770
771 (a) The RunTest script now detects the internal link size and whether there
772 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
773 values substituted by "configure". (The RunGrepTest script already did
774 this for UTF-8.) The configure.ac script no longer substitutes the
775 relevant variables.
776
777 (b) The debugging options /B and /D in pcretest show the compiled bytecode
778 with length and offset values. This means that the output is different
779 for different internal link sizes. Test 2 is skipped for link sizes
780 other than 2 because of this, bypassing the problem. Unfortunately,
781 there was also a test in test 3 (the locale tests) that used /B and
782 failed for link sizes other than 2. Rather than cut the whole test out,
783 I have added a new /Z option to pcretest that replaces the length and
784 offset values with spaces. This is now used to make test 3 independent
785 of link size. (Test 2 will be tidied up later.)
786
787 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
788 segmentation fault instead of returning the appropriate error message.
789
790 15. In multiline mode when the newline sequence was set to "any", the pattern
791 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
792 This doesn't seem right; it now treats the CRLF combination as the line
793 ending, and so does not match in that case. It's only a pattern such as ^$
794 that would hit this one: something like ^ABC$ would have failed after \r
795 and then tried again after \r\n.
796
797 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
798 in an attempt to make files that differ only in their line terminators
799 compare equal. This works on Linux.
800
801 17. Under certain error circumstances pcregrep might try to free random memory
802 as it exited. This is now fixed, thanks to valgrind.
803
804 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
805 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
806 was because its rules for how to advance for /g after matching an empty
807 string at the end of a line did not allow for this case. They now check for
808 it specially.
809
810 20. pcretest is supposed to handle patterns and data of any length, by
811 extending its buffers when necessary. It was getting this wrong when the
812 buffer for a data line had to be extended.
813
814 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
815 CRLF as a newline sequence.
816
817 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
818 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
819 I have nevertheless tidied it up.
820
821 23. Added some casts to kill warnings from HP-UX ia64 compiler.
822
823 24. Added a man page for pcre-config.
824
825
826 Version 7.0 19-Dec-06
827 ---------------------
828
829 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
830 moving to gcc 4.1.1.
831
832 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
833 sys/time.h, which is documented as needed for this function. It doesn't
834 seem to matter on Linux, but it showed up on some releases of OS X.
835
836 3. It seems that there are systems where bytes whose values are greater than
837 127 match isprint() in the "C" locale. The "C" locale should be the
838 default when a C program starts up. In most systems, only ASCII printing
839 characters match isprint(). This difference caused the output from pcretest
840 to vary, making some of the tests fail. I have changed pcretest so that:
841
842 (a) When it is outputting text in the compiled version of a pattern, bytes
843 other than 32-126 are always shown as hex escapes.
844
845 (b) When it is outputting text that is a matched part of a subject string,
846 it does the same, unless a different locale has been set for the match
847 (using the /L modifier). In this case, it uses isprint() to decide.
848
849 4. Fixed a major bug that caused incorrect computation of the amount of memory
850 required for a compiled pattern when options that changed within the
851 pattern affected the logic of the preliminary scan that determines the
852 length. The relevant options are -x, and -i in UTF-8 mode. The result was
853 that the computed length was too small. The symptoms of this bug were
854 either the PCRE error "internal error: code overflow" from pcre_compile(),
855 or a glibc crash with a message such as "pcretest: free(): invalid next
856 size (fast)". Examples of patterns that provoked this bug (shown in
857 pcretest format) are:
858
859 /(?-x: )/x
860 /(?x)(?-x: \s*#\s*)/
861 /((?i)[\x{c0}])/8
862 /(?i:[\x{c0}])/8
863
864 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
865 is now done differently.
866
867 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
868 wrapper classes; (b) implement a new function in the C++ scanner that is
869 more efficient than the old way of doing things because it avoids levels of
870 recursion in the regex matching; (c) add a paragraph to the documentation
871 for the FullMatch() function.
872
873 6. The escape sequence \n was being treated as whatever was defined as
874 "newline". Not only was this contrary to the documentation, which states
875 that \n is character 10 (hex 0A), but it also went horribly wrong when
876 "newline" was defined as CRLF. This has been fixed.
877
878 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
879 was being set to -1 for the "end of line" case (supposedly a value that no
880 character can have). Though this value is never used (the check for end of
881 line is "zero bytes in current character"), it caused compiler complaints.
882 I've changed it to 0xffffffff.
883
884 8. In pcre_version.c, the version string was being built by a sequence of
885 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
886 string (as it is for production releases) called a macro with an empty
887 argument. The C standard says the result of this is undefined. The gcc
888 compiler treats it as an empty string (which was what was wanted) but it is
889 reported that Visual C gives an error. The source has been hacked around to
890 avoid this problem.
891
892 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
893 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
894 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
895 of them did).
896
897 10. Originally, pcretest opened its input and output without "b"; then I was
898 told that "b" was needed in some environments, so it was added for release
899 5.0 to both the input and output. (It makes no difference on Unix-like
900 systems.) Later I was told that it is wrong for the input on Windows. I've
901 now abstracted the modes into two macros, to make it easier to fiddle with
902 them, and removed "b" from the input mode under Windows.
903
904 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
905
906 12. Added -help and --help to pcretest as an official way of being reminded
907 of the options.
908
909 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
910 and pcrecpp.cc because they annoy compilers at high warning levels.
911
912 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
913
914 15. Fixed an occurrence of == in configure.ac that should have been = (shell
915 scripts are not C programs :-) and which was not noticed because it works
916 on Linux.
917
918 16. pcretest is supposed to handle any length of pattern and data line (as one
919 line or as a continued sequence of lines) by extending its input buffer if
920 necessary. This feature was broken for very long pattern lines, leading to
921 a string of junk being passed to pcre_compile() if the pattern was longer
922 than about 50K.
923
924 17. I have done a major re-factoring of the way pcre_compile() computes the
925 amount of memory needed for a compiled pattern. Previously, there was code
926 that made a preliminary scan of the pattern in order to do this. That was
927 OK when PCRE was new, but as the facilities have expanded, it has become
928 harder and harder to keep it in step with the real compile phase, and there
929 have been a number of bugs (see for example, 4 above). I have now found a
930 cunning way of running the real compile function in a "fake" mode that
931 enables it to compute how much memory it would need, while actually only
932 ever using a few hundred bytes of working memory and without too many
933 tests of the mode. This should make future maintenance and development
934 easier. A side effect of this work is that the limit of 200 on the nesting
935 depth of parentheses has been removed (though this was never a serious
936 limitation, I suspect). However, there is a downside: pcre_compile() now
937 runs more slowly than before (30% or more, depending on the pattern). I
938 hope this isn't a big issue. There is no effect on runtime performance.
939
940 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
941 newline (only possible for the last line of a file) and it was a
942 pattern that set a locale (followed by /Lsomething), pcretest crashed.
943
944 19. Added additional timing features to pcretest. (1) The -tm option now times
945 matching only, not compiling. (2) Both -t and -tm can be followed, as a
946 separate command line item, by a number that specifies the number of
947 repeats to use when timing. The default is 50000; this gives better
948 precision, but takes uncomfortably long for very large patterns.
949
950 20. Extended pcre_study() to be more clever in cases where a branch of a
951 subpattern has no definite first character. For example, (a*|b*)[cd] would
952 previously give no result from pcre_study(). Now it recognizes that the
953 first character must be a, b, c, or d.
954
955 21. There was an incorrect error "recursive call could loop indefinitely" if
956 a subpattern (or the entire pattern) that was being tested for matching an
957 empty string contained only one non-empty item after a nested subpattern.
958 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
959 incorrectly, because the \d was being skipped in the check.
960
961 22. The pcretest program now has a new pattern option /B and a command line
962 option -b, which is equivalent to adding /B to every pattern. This causes
963 it to show the compiled bytecode, without the additional information that
964 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
965 is the same as /B/I).
966
967 23. A new optimization is now able automatically to treat some sequences such
968 as a*b as a*+b. More specifically, if something simple (such as a character
969 or a simple class like \d) has an unlimited quantifier, and is followed by
970 something that cannot possibly match the quantified thing, the quantifier
971 is automatically "possessified".
972
973 24. A recursive reference to a subpattern whose number was greater than 39
974 went wrong under certain circumstances in UTF-8 mode. This bug could also
975 have affected the operation of pcre_study().
976
977 25. Realized that a little bit of performance could be had by replacing
978 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
979
980 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
981
982 27. Possessive quantifiers such as a++ were previously implemented by turning
983 them into atomic groups such as ($>a+). Now they have their own opcodes,
984 which improves performance. This includes the automatically created ones
985 from 23 above.
986
987 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
988 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
989 the first matched character to be a colon. This applied both to named and
990 numbered groups.
991
992 29. The ucpinternal.h header file was missing its idempotency #ifdef.
993
994 30. I was sent a "project" file called libpcre.a.dev which I understand makes
995 building PCRE on Windows easier, so I have included it in the distribution.
996
997 31. There is now a check in pcretest against a ridiculously large number being
998 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
999 loop, the loop is abandoned.
1000
1001 32. Forward references to subpatterns in conditions such as (?(2)...) where
1002 subpattern 2 is defined later cause pcre_compile() to search forwards in
1003 the pattern for the relevant set of parentheses. This search went wrong
1004 when there were unescaped parentheses in a character class, parentheses
1005 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1006
1007 33. "Subroutine" calls and backreferences were previously restricted to
1008 referencing subpatterns earlier in the regex. This restriction has now
1009 been removed.
1010
1011 34. Added a number of extra features that are going to be in Perl 5.10. On the
1012 whole, these are just syntactic alternatives for features that PCRE had
1013 previously implemented using the Python syntax or my own invention. The
1014 other formats are all retained for compatibility.
1015
1016 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1017 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1018 also .NET compatible.
1019
1020 (b) A recursion or subroutine call to a named group can now be defined as
1021 (?&name) as well as (?P>name).
1022
1023 (c) A backreference to a named group can now be defined as \k<name> or
1024 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1025 5.10, are also .NET compatible.
1026
1027 (d) A conditional reference to a named group can now use the syntax
1028 (?(<name>) or (?('name') as well as (?(name).
1029
1030 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1031 groups (named and numbered) that are never evaluated inline, but can be
1032 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1033 is always false. There may be only one alternative in such a group.
1034
1035 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1036 as the simple (?(R). The condition is true only if the most recent
1037 recursion is that of the given number or name. It does not search out
1038 through the entire recursion stack.
1039
1040 (g) The escape \gN or \g{N} has been added, where N is a positive or
1041 negative number, specifying an absolute or relative reference.
1042
1043 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1044 some "unreachable code" warnings.
1045
1046 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1047 things, this adds five new scripts.
1048
1049 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1050 There were also incompatibilities regarding the handling of \Q..\E inside
1051 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1052 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1053
1054 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1055 matches an empty string, and forcibly breaks the loop. There were bugs in
1056 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1057 against aaaa the result was just "a" rather than "aaaa", for example. Two
1058 separate and independent bugs (that affected different cases) have been
1059 fixed.
1060
1061 39. Refactored the code to abolish the use of different opcodes for small
1062 capturing bracket numbers. This is a tidy that I avoided doing when I
1063 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1064 The new approach is not only tidier, it makes it possible to reduce the
1065 memory needed to fix the previous bug (38).
1066
1067 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1068 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1069 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1070 mode.
1071
1072 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1073 report.
1074
1075 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1076 copy construction and assignment in the C++ wrapper.
1077
1078 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1079 couple of bugs that could have given wrong results in the "--newline=crlf"
1080 case.
1081
1082 44. Added a number of casts and did some reorganization of signed/unsigned int
1083 variables following suggestions from Dair Grant. Also renamed the variable
1084 "this" as "item" because it is a C++ keyword.
1085
1086 45. Arranged for dftables to add
1087
1088 #include "pcre_internal.h"
1089
1090 to pcre_chartables.c because without it, gcc 4.x may remove the array
1091 definition from the final binary if PCRE is built into a static library and
1092 dead code stripping is activated.
1093
1094 46. For an unanchored pattern, if a match attempt fails at the start of a
1095 newline sequence, and the newline setting is CRLF or ANY, and the next two
1096 characters are CRLF, advance by two characters instead of one.
1097
1098
1099 Version 6.7 04-Jul-06
1100 ---------------------
1101
1102 1. In order to handle tests when input lines are enormously long, pcretest has
1103 been re-factored so that it automatically extends its buffers when
1104 necessary. The code is crude, but this _is_ just a test program. The
1105 default size has been increased from 32K to 50K.
1106
1107 2. The code in pcre_study() was using the value of the re argument before
1108 testing it for NULL. (Of course, in any sensible call of the function, it
1109 won't be NULL.)
1110
1111 3. The memmove() emulation function in pcre_internal.h, which is used on
1112 systems that lack both memmove() and bcopy() - that is, hardly ever -
1113 was missing a "static" storage class specifier.
1114
1115 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1116 containing an extended class (one that cannot be represented by a bitmap
1117 because it contains high-valued characters or Unicode property items, e.g.
1118 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1119 pattern, but PCRE should not loop if you do not (it no longer does).
1120 [Detail: two cases were found: (a) a repeated subpattern containing an
1121 extended class; (b) a recursive reference to a subpattern that followed a
1122 previous extended class. It wasn't skipping over the extended class
1123 correctly when UTF-8 mode was not set.]
1124
1125 5. A negated single-character class was not being recognized as fixed-length
1126 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1127 compile error "lookbehind assertion is not fixed length".
1128
1129 6. The RunPerlTest auxiliary script was showing an unexpected difference
1130 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1131 write a Perl script that can interpret lines of an input file either as
1132 byte characters or as UTF-8, which is what "perltest" was being required to
1133 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1134 can't do is switch easily at run time between having the "use utf8;" pragma
1135 or not. In the end, I fudged it by using the RunPerlTest script to insert
1136 "use utf8;" explicitly for the UTF-8 tests.
1137
1138 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1139 the end of the subject string, contrary to the documentation and to what
1140 Perl does. This was true of both matching functions. Now it matches only at
1141 the start of the subject and immediately after *internal* newlines.
1142
1143 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1144 a pointer to an int instead of a pointer to an unsigned long int. This
1145 caused problems on 64-bit systems.
1146
1147 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1148 instance of the 'standard' template library not being so standard".
1149
1150 10. There was no check on the number of named subpatterns nor the maximum
1151 length of a subpattern name. The product of these values is used to compute
1152 the size of the memory block for a compiled pattern. By supplying a very
1153 long subpattern name and a large number of named subpatterns, the size
1154 computation could be caused to overflow. This is now prevented by limiting
1155 the length of names to 32 characters, and the number of named subpatterns
1156 to 10,000.
1157
1158 11. Subpatterns that are repeated with specific counts have to be replicated in
1159 the compiled pattern. The size of memory for this was computed from the
1160 length of the subpattern and the repeat count. The latter is limited to
1161 65535, but there was no limit on the former, meaning that integer overflow
1162 could in principle occur. The compiled length of a repeated subpattern is
1163 now limited to 30,000 bytes in order to prevent this.
1164
1165 12. Added the optional facility to have named substrings with the same name.
1166
1167 13. Added the ability to use a named substring as a condition, using the
1168 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1169 are numbers (not recommended). Forward references are permitted.
1170
1171 14. Added forward references in named backreferences (if you see what I mean).
1172
1173 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1174 pattern could run off the end of the subject. For example, the pattern
1175 "(?s)(.{1,5})"8 did this with the subject "ab".
1176
1177 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1178 PCRE_CASELESS was set when matching characters that were quantified with ?
1179 or *.
1180
1181 17. A character class other than a single negated character that had a minimum
1182 but no maximum quantifier - for example [ab]{6,} - was not handled
1183 correctly by pce_dfa_exec(). It would match only one character.
1184
1185 18. A valid (though odd) pattern that looked like a POSIX character
1186 class but used an invalid character after [ (for example [[,abc,]]) caused
1187 pcre_compile() to give the error "Failed: internal error: code overflow" or
1188 in some cases to crash with a glibc free() error. This could even happen if
1189 the pattern terminated after [[ but there just happened to be a sequence of
1190 letters, a binary zero, and a closing ] in the memory that followed.
1191
1192 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1193 over the years. Originally (before any Unicode support), just the bottom 8
1194 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1195 output from "man perlunicode" includes this:
1196
1197 The regular expression compiler produces polymorphic opcodes. That
1198 is, the pattern adapts to the data and automatically switches to
1199 the Unicode character scheme when presented with Unicode data--or
1200 instead uses a traditional byte scheme when presented with byte
1201 data.
1202
1203 Sadly, a wide octal escape does not cause a switch, and in a string with
1204 no other multibyte characters, these octal escapes are treated as before.
1205 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1206 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1207 Unicode string.
1208
1209 I have not perpetrated such confusion in PCRE. Up till now, it took just
1210 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1211 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1212 translate to the appropriate multibyte character.
1213
1214 29. Applied some refactoring to reduce the number of warnings from Microsoft
1215 and Borland compilers. This has included removing the fudge introduced
1216 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1217 a warning about an unused variable.
1218
1219 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1220 characters since release 4.0, because Perl (from release 5.004) does not.
1221 [Or at least, is documented not to: some releases seem to be in conflict
1222 with the documentation.] However, when a pattern was studied with
1223 pcre_study() and all its branches started with \s, PCRE still included VT
1224 as a possible starting character. Of course, this did no harm; it just
1225 caused an unnecessary match attempt.
1226
1227 22. Removed a now-redundant internal flag bit that recorded the fact that case
1228 dependency changed within the pattern. This was once needed for "required
1229 byte" processing, but is no longer used. This recovers a now-scarce options
1230 bit. Also moved the least significant internal flag bit to the most-
1231 significant bit of the word, which was not previously used (hangover from
1232 the days when it was an int rather than a uint) to free up another bit for
1233 the future.
1234
1235 23. Added support for CRLF line endings as well as CR and LF. As well as the
1236 default being selectable at build time, it can now be changed at runtime
1237 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1238 specify that it is scanning data with non-default line endings.
1239
1240 24. Changed the definition of CXXLINK to make it agree with the definition of
1241 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1242
1243 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1244 recursions. This makes a big different to stack usage for some patterns.
1245
1246 26. If a subpattern containing a named recursion or subroutine reference such
1247 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1248 the space required for the compiled pattern went wrong and gave too small a
1249 value. Depending on the environment, this could lead to "Failed: internal
1250 error: code overflow at offset 49" or "glibc detected double free or
1251 corruption" errors.
1252
1253 27. Applied patches from Google (a) to support the new newline modes and (b) to
1254 advance over multibyte UTF-8 characters in GlobalReplace.
1255
1256 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1257 difference for some implementation of PCRE in some Windows version.
1258
1259 29. Added some extra testing facilities to pcretest:
1260
1261 \q<number> in a data line sets the "match limit" value
1262 \Q<number> in a data line sets the "match recursion limt" value
1263 -S <number> sets the stack size, where <number> is in megabytes
1264
1265 The -S option isn't available for Windows.
1266
1267
1268 Version 6.6 06-Feb-06
1269 ---------------------
1270
1271 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1272 in pcreposix.h. I have copied the definition from pcre.h.
1273
1274 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1275 because pcre.h is no longer a built file.
1276
1277 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1278 not normally included in the compiled code.
1279
1280
1281 Version 6.5 01-Feb-06
1282 ---------------------
1283
1284 1. When using the partial match feature with pcre_dfa_exec(), it was not
1285 anchoring the second and subsequent partial matches at the new starting
1286 point. This could lead to incorrect results. For example, with the pattern
1287 /1234/, partially matching against "123" and then "a4" gave a match.
1288
1289 2. Changes to pcregrep:
1290
1291 (a) All non-match returns from pcre_exec() were being treated as failures
1292 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1293 error message is output. Some extra information is given for the
1294 PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1295 probably the only errors that are likely to be caused by users (by
1296 specifying a regex that has nested indefinite repeats, for instance).
1297 If there are more than 20 of these errors, pcregrep is abandoned.
1298
1299 (b) A binary zero was treated as data while matching, but terminated the
1300 output line if it was written out. This has been fixed: binary zeroes
1301 are now no different to any other data bytes.
1302
1303 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1304 used to set a locale for matching. The --locale=xxxx long option has
1305 been added (no short equivalent) to specify a locale explicitly on the
1306 pcregrep command, overriding the environment variables.
1307
1308 (d) When -B was used with -n, some line numbers in the output were one less
1309 than they should have been.
1310
1311 (e) Added the -o (--only-matching) option.
1312
1313 (f) If -A or -C was used with -c (count only), some lines of context were
1314 accidentally printed for the final match.
1315
1316 (g) Added the -H (--with-filename) option.
1317
1318 (h) The combination of options -rh failed to suppress file names for files
1319 that were found from directory arguments.
1320
1321 (i) Added the -D (--devices) and -d (--directories) options.
1322
1323 (j) Added the -F (--fixed-strings) option.
1324
1325 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1326
1327 (l) Added the --colo(u)r option.
1328
1329 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1330 is not present by default.
1331
1332 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1333 items such as (?R) or (?1), when the recursion could match a number of
1334 alternatives. If it matched one of the alternatives, but subsequently,
1335 outside the recursion, there was a failure, the code tried to back up into
1336 the recursion. However, because of the way PCRE is implemented, this is not
1337 possible, and the result was an incorrect result from the match.
1338
1339 In order to prevent this happening, the specification of recursion has
1340 been changed so that all such subpatterns are automatically treated as
1341 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1342
1343 4. I had overlooked the fact that, in some locales, there are characters for
1344 which isalpha() is true but neither isupper() nor islower() are true. In
1345 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1346 and ordfeminine) are like this. This affected the treatment of \w and \W
1347 when they appeared in character classes, but not when they appeared outside
1348 a character class. The bit map for "word" characters is now created
1349 separately from the results of isalnum() instead of just taking it from the
1350 upper, lower, and digit maps. (Plus the underscore character, of course.)
1351
1352 5. The above bug also affected the handling of POSIX character classes such as
1353 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1354 permanent tables. Instead, the bit maps for such a class were previously
1355 created as the appropriate unions of the upper, lower, and digit bitmaps.
1356 Now they are created by subtraction from the [[:word:]] class, which has
1357 its own bitmap.
1358
1359 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1360 It is created by subtracting the vertical space characters (\x09, \x0a,
1361 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1362 subtraction was done in the overall bitmap for a character class, meaning
1363 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1364 be recognized. This bug has been fixed.
1365
1366 7. Patches from the folks at Google:
1367
1368 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1369 real life, but is still worth protecting against".
1370
1371 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1372 regular expressions".
1373
1374 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1375 have it.
1376
1377 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1378 "configure" and the latter not, in order to fix a problem somebody had
1379 with compiling the Arg class on HP-UX.
1380
1381 (e) Improve the error-handling of the C++ wrapper a little bit.
1382
1383 (f) New tests for checking recursion limiting.
1384
1385 8. The pcre_memmove() function, which is used only if the environment does not
1386 have a standard memmove() function (and is therefore rarely compiled),
1387 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1388 returning a result (though PCRE never actually uses the result).
1389
1390 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1391 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1392 returned instead of calling malloc() with an overflowing number that would
1393 most likely cause subsequent chaos.
1394
1395 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1396
1397 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1398 with this option is matched, the nmatch and pmatch options of regexec() are
1399 ignored.
1400
1401 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1402 provided in case anyone wants to the the POSIX interface with UTF-8
1403 strings.
1404
1405 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1406 C++ linking (needed for some HP-UX environments).
1407
1408 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1409 (unused parameter) and in the pcre_printint() function (omitted "default"
1410 switch label when the default is to do nothing).
1411
1412 15. Added some code to make it possible, when PCRE is compiled as a C++
1413 library, to replace subject pointers for pcre_exec() with a smart pointer
1414 class, thus making it possible to process discontinuous strings.
1415
1416 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1417 much the same function. They were added by different people who were trying
1418 to make PCRE easy to compile on non-Unix systems. It has been suggested
1419 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1420 for compiling on Windows systems. I have therefore replaced it with
1421 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1422 defaults to "extern" for C or "extern C" for C++, which works fine on
1423 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1424 SCOPE with something explicit in config.h. In addition:
1425
1426 (a) pcreposix.h still had just "extern" instead of either of these macros;
1427 I have replaced it with PCRE_DATA_SCOPE.
1428
1429 (b) Functions such as _pcre_xclass(), which are internal to the library,
1430 but external in the C sense, all had PCRE_EXPORT in their definitions.
1431 This is apparently wrong for the Windows case, so I have removed it.
1432 (It makes no difference on Unix-like systems.)
1433
1434 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1435 of recursive calls to match(). This is different to MATCH_LIMIT because
1436 that limits the total number of calls to match(), not all of which increase
1437 the depth of recursion. Limiting the recursion depth limits the amount of
1438 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1439 when PCRE is compiled, and changed at run time. A patch from Google adds
1440 this functionality to the C++ interface.
1441
1442 18. Changes to the handling of Unicode character properties:
1443
1444 (a) Updated the table to Unicode 4.1.0.
1445
1446 (b) Recognize characters that are not in the table as "Cn" (undefined).
1447
1448 (c) I revised the way the table is implemented to a much improved format
1449 which includes recognition of ranges. It now supports the ranges that
1450 are defined in UnicodeData.txt, and it also amalgamates other
1451 characters into ranges. This has reduced the number of entries in the
1452 table from around 16,000 to around 3,000, thus reducing its size
1453 considerably. I realized I did not need to use a tree structure after
1454 all - a binary chop search is just as efficient. Having reduced the
1455 number of entries, I extended their size from 6 bytes to 8 bytes to
1456 allow for more data.
1457
1458 (d) Added support for Unicode script names via properties such as \p{Han}.
1459
1460 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1461 matching that character.
1462
1463 20. When matching a repeated Unicode property with a minimum greater than zero,
1464 (for example \pL{2,}), PCRE could look past the end of the subject if it
1465 reached it while seeking the minimum number of characters. This could
1466 happen only if some of the characters were more than one byte long, because
1467 there is a check for at least the minimum number of bytes.
1468
1469 21. Refactored the implementation of \p and \P so as to be more general, to
1470 allow for more different types of property in future. This has changed the
1471 compiled form incompatibly. Anybody with saved compiled patterns that use
1472 \p or \P will have to recompile them.
1473
1474 22. Added "Any" and "L&" to the supported property types.
1475
1476 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1477 but give a compile time error if the value is greater than 0xff.
1478
1479 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1480 accidentally not being installed or uninstalled.
1481
1482 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1483 made were to insert the current release number. This seemed silly, because
1484 it made things harder for people building PCRE on systems that don't run
1485 "configure". I have turned pcre.h into a distributed file, no longer built
1486 by "configure", with the version identification directly included. There is
1487 no longer a pcre.h.in file.
1488
1489 However, this change necessitated a change to the pcre-config script as
1490 well. It is built from pcre-config.in, and one of the substitutions was the
1491 release number. I have updated configure.ac so that ./configure now finds
1492 the release number by grepping pcre.h.
1493
1494 26. Added the ability to run the tests under valgrind.
1495
1496
1497 Version 6.4 05-Sep-05
1498 ---------------------
1499
1500 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1501 "--" to be printed when multiple files were scanned, even when none of the
1502 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1503 consider it to be a bug, and have restored the previous behaviour.
1504
1505 2. A couple of code tidies to get rid of compiler warnings.
1506
1507 3. The pcretest program used to cheat by referring to symbols in the library
1508 whose names begin with _pcre_. These are internal symbols that are not
1509 really supposed to be visible externally, and in some environments it is
1510 possible to suppress them. The cheating is now confined to including
1511 certain files from the library's source, which is a bit cleaner.
1512
1513 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1514 file's purpose clearer.
1515
1516 5. Reorganized pcre_ucp_findchar().
1517
1518
1519 Version 6.3 15-Aug-05
1520 ---------------------
1521
1522 1. The file libpcre.pc.in did not have general read permission in the tarball.
1523
1524 2. There were some problems when building without C++ support:
1525
1526 (a) If C++ support was not built, "make install" and "make test" still
1527 tried to test it.
1528
1529 (b) There were problems when the value of CXX was explicitly set. Some
1530 changes have been made to try to fix these, and ...
1531
1532 (c) --disable-cpp can now be used to explicitly disable C++ support.
1533
1534 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1535 backslash in a target when C++ was disabled. This confuses some
1536 versions of "make", apparently. Using an intermediate variable solves
1537 this. (Same for CPP_LOBJ.)
1538
1539 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1540 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1541 necessary on certain architectures.
1542
1543 4. Added a setting of -export-symbols-regex to the link command to remove
1544 those symbols that are exported in the C sense, but actually are local
1545 within the library, and not documented. Their names all begin with
1546 "_pcre_". This is not a perfect job, because (a) we have to except some
1547 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1548 available (and never for static libraries). I have made a note to try to
1549 find a way round (a) in the future.
1550
1551
1552 Version 6.2 01-Aug-05
1553 ---------------------
1554
1555 1. There was no test for integer overflow of quantifier values. A construction
1556 such as {1111111111111111} would give undefined results. What is worse, if
1557 a minimum quantifier for a parenthesized subpattern overflowed and became
1558 negative, the calculation of the memory size went wrong. This could have
1559 led to memory overwriting.
1560
1561 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1562
1563 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1564 operating environments where this matters.
1565
1566 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1567 PCRE options from within the C++ wrapper.
1568
1569 5. Named capturing subpatterns were not being correctly counted when a pattern
1570 was compiled. This caused two problems: (a) If there were more than 100
1571 such subpatterns, the calculation of the memory needed for the whole
1572 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1573 back references of the form \12, where the number was greater than 9, were
1574 not recognized as back references, even though there were sufficient
1575 previous subpatterns.
1576
1577 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1578 versions of gcc, e.g. 2.95.4.
1579
1580
1581 Version 6.1 21-Jun-05
1582 ---------------------
1583
1584 1. There was one reference to the variable "posix" in pcretest.c that was not
1585 surrounded by "#if !defined NOPOSIX".
1586
1587 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1588 the cross-check on the old pcre_info() function, for the benefit of the
1589 cut-down version of PCRE that is currently imported into Exim.
1590
1591 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1592 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1593 patterns that start (?i) but I don't think that matters. The use of (?i) is
1594 just an example; this all applies to the other options as well.
1595
1596 4. Since libtool seems to echo the compile commands it is issuing, the output
1597 from "make" can be reduced a bit by putting "@" in front of each libtool
1598 compile command.
1599
1600 5. Patch from the folks at Google for configure.in to be a bit more thorough
1601 in checking for a suitable C++ installation before trying to compile the
1602 C++ stuff. This should fix a reported problem when a compiler was present,
1603 but no suitable headers.
1604
1605 6. The man pages all had just "PCRE" as their title. I have changed them to
1606 be the relevant file name. I have also arranged that these names are
1607 retained in the file doc/pcre.txt, which is a concatenation in text format
1608 of all the man pages except the little individual ones for each function.
1609
1610 7. The NON-UNIX-USE file had not been updated for the different set of source
1611 files that come with release 6. I also added a few comments about the C++
1612 wrapper.
1613
1614
1615 Version 6.0 07-Jun-05
1616 ---------------------
1617
1618 1. Some minor internal re-organization to help with my DFA experiments.
1619
1620 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1621 didn't matter for the library itself when fully configured, but did matter
1622 when compiling without UCP support, or within Exim, where the ucp files are
1623 not imported.
1624
1625 3. Refactoring of the library code to split up the various functions into
1626 different source modules. The addition of the new DFA matching code (see
1627 below) to a single monolithic source would have made it really too
1628 unwieldy, quite apart from causing all the code to be include in a
1629 statically linked application, when only some functions are used. This is
1630 relevant even without the DFA addition now that patterns can be compiled in
1631 one application and matched in another.
1632
1633 The downside of splitting up is that there have to be some external
1634 functions and data tables that are used internally in different modules of
1635 the library but which are not part of the API. These have all had their
1636 names changed to start with "_pcre_" so that they are unlikely to clash
1637 with other external names.
1638
1639 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1640 a different (DFA) algorithm. Although it is slower than the original
1641 function, it does have some advantages for certain types of matching
1642 problem.
1643
1644 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1645 including restarting after a partial match.
1646
1647 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1648 defined when compiling for Windows was sent to me. I have put it into the
1649 code, though I have no means of testing or verifying it.
1650
1651 7. Added the pcre_refcount() auxiliary function.
1652
1653 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1654 match before or at the first newline in the subject string. In pcretest,
1655 the /f option on a pattern can be used to set this.
1656
1657 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1658 would behave wrongly. This has been present in PCRE since release 4.0.
1659
1660 10. A number of changes to the pcregrep command:
1661
1662 (a) Refactored how -x works; insert ^(...)$ instead of setting
1663 PCRE_ANCHORED and checking the length, in preparation for adding
1664 something similar for -w.
1665
1666 (b) Added the -w (match as a word) option.
1667
1668 (c) Refactored the way lines are read and buffered so as to have more
1669 than one at a time available.
1670
1671 (d) Implemented a pcregrep test script.
1672
1673 (e) Added the -M (multiline match) option. This allows patterns to match
1674 over several lines of the subject. The buffering ensures that at least
1675 8K, or the rest of the document (whichever is the shorter) is available
1676 for matching (and similarly the previous 8K for lookbehind assertions).
1677
1678 (f) Changed the --help output so that it now says
1679
1680 -w, --word-regex(p)
1681
1682 instead of two lines, one with "regex" and the other with "regexp"
1683 because that confused at least one person since the short forms are the
1684 same. (This required a bit of code, as the output is generated
1685 automatically from a table. It wasn't just a text change.)
1686
1687 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1688 option but starts with a hyphen. Could be a pattern or a path name
1689 starting with a hyphen, for instance.
1690
1691 (h) "-" can be given as a file name to represent stdin.
1692
1693 (i) When file names are being printed, "(standard input)" is used for
1694 the standard input, for compatibility with GNU grep. Previously
1695 "<stdin>" was used.
1696
1697 (j) The option --label=xxx can be used to supply a name to be used for
1698 stdin when file names are being printed. There is no short form.
1699
1700 (k) Re-factored the options decoding logic because we are going to add
1701 two more options that take data. Such options can now be given in four
1702 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1703
1704 (l) Added the -A, -B, and -C options for requesting that lines of context
1705 around matches be printed.
1706
1707 (m) Added the -L option to print the names of files that do not contain
1708 any matching lines, that is, the complement of -l.
1709
1710 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1711 continue to scan other files.
1712
1713 (o) The -s option was incorrectly implemented. For compatibility with other
1714 greps, it now suppresses the error message for a non-existent or non-
1715 accessible file (but not the return code). There is a new option called
1716 -q that suppresses the output of matching lines, which was what -s was
1717 previously doing.
1718
1719 (p) Added --include and --exclude options to specify files for inclusion
1720 and exclusion when recursing.
1721
1722 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1723 Hopefully, it now does.
1724
1725 12. Missing cast in pcre_study().
1726
1727 13. Added an "uninstall" target to the makefile.
1728
1729 14. Replaced "extern" in the function prototypes in Makefile.in with
1730 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1731 world, but is set differently for Windows.
1732
1733 15. Added a second compiling function called pcre_compile2(). The only
1734 difference is that it has an extra argument, which is a pointer to an
1735 integer error code. When there is a compile-time failure, this is set
1736 non-zero, in addition to the error test pointer being set to point to an
1737 error message. The new argument may be NULL if no error number is required
1738 (but then you may as well call pcre_compile(), which is now just a
1739 wrapper). This facility is provided because some applications need a
1740 numeric error indication, but it has also enabled me to tidy up the way
1741 compile-time errors are handled in the POSIX wrapper.
1742
1743 16. Added VPATH=.libs to the makefile; this should help when building with one
1744 prefix path and installing with another. (Or so I'm told by someone who
1745 knows more about this stuff than I do.)
1746
1747 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1748 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1749 match everything, including newlines. This is not POSIX-compatible, but
1750 somebody wanted the feature. From pcretest it can be activated by using
1751 both the P and the s flags.
1752
1753 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1754
1755 19. libpcre.pc was being incorrectly installed as executable.
1756
1757 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1758 it now also looks for '\r' so that it will work unmodified on Windows.
1759
1760 21. Added Google's contributed C++ wrapper to the distribution.
1761
1762 22. Added some untidy missing memory free() calls in pcretest, to keep
1763 Electric Fence happy when testing.
1764
1765
1766
1767 Version 5.0 13-Sep-04
1768 ---------------------
1769
1770 1. Internal change: literal characters are no longer packed up into items
1771 containing multiple characters in a single byte-string. Each character
1772 is now matched using a separate opcode. However, there may be more than one
1773 byte in the character in UTF-8 mode.
1774
1775 2. The pcre_callout_block structure has two new fields: pattern_position and
1776 next_item_length. These contain the offset in the pattern to the next match
1777 item, and its length, respectively.
1778
1779 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1780 insertion of callouts before each pattern item. Added the /C option to
1781 pcretest to make use of this.
1782
1783 4. On the advice of a Windows user, the lines
1784
1785 #if defined(_WIN32) || defined(WIN32)
1786 _setmode( _fileno( stdout ), 0x8000 );
1787 #endif /* defined(_WIN32) || defined(WIN32) */
1788
1789 have been added to the source of pcretest. This apparently does useful
1790 magic in relation to line terminators.
1791
1792 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1793 for the benefit of those environments where the "b" makes a difference.
1794
1795 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1796 to know about it. I have put a hack into configure.in that adds in code
1797 to set GCC=yes if CC=icc. This seems to end up at a point in the
1798 generated configure script that is early enough to affect the setting of
1799 compiler options, which is what is needed, but I have no means of testing
1800 whether it really works. (The user who reported this had patched the
1801 generated configure script, which of course I cannot do.)
1802
1803 LATER: After change 22 below (new libtool files), the configure script
1804 seems to know about icc (and also ecc). Therefore, I have commented out
1805 this hack in configure.in.
1806
1807 7. Added support for pkg-config (2 patches were sent in).
1808
1809 8. Negated POSIX character classes that used a combination of internal tables
1810 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1811 [[:^ascii]]. Typically, they would match almost any characters. The other
1812 POSIX classes were not broken in this way.
1813
1814 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1815 to find the match, as PCRE was deluded into thinking that the match had to
1816 start at the start point or following a newline. The same bug applied to
1817 patterns with negative forward assertions or any backward assertions
1818 preceding ".*" at the start, unless the pattern required a fixed first
1819 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1820
1821 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1822 starting at the last subject character, bytes beyond the end of the subject
1823 string were read.
1824
1825 11. Renamed the variable "class" as "classbits" to make life easier for C++
1826 users. (Previously there was a macro definition, but it apparently wasn't
1827 enough.)
1828
1829 12. Added the new field "tables" to the extra data so that tables can be passed
1830 in at exec time, or the internal tables can be re-selected. This allows
1831 a compiled regex to be saved and re-used at a later time by a different
1832 program that might have everything at different addresses.
1833
1834 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1835 -R library as well as a -L library.
1836
1837 14. The debugging options of pcretest (-d on the command line or D on a
1838 pattern) showed incorrect output for anything following an extended class
1839 that contained multibyte characters and which was followed by a quantifier.
1840
1841 15. Added optional support for general category Unicode character properties
1842 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1843 support. It adds about 90K to the size of the library. The meanings of the
1844 inbuilt class escapes such as \d and \s have NOT been changed.
1845
1846 16. Updated pcredemo.c to include calls to free() to release the memory for the
1847 compiled pattern.
1848
1849 17. The generated file chartables.c was being created in the source directory
1850 instead of in the building directory. This caused the build to fail if the
1851 source directory was different from the building directory, and was
1852 read-only.
1853
1854 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1855 file. No doubt somebody will tell me if they don't make sense... Also added
1856 Dan Mooney's comments about building on OpenVMS.
1857
1858 19. Added support for partial matching via the PCRE_PARTIAL option for
1859 pcre_exec() and the \P data escape in pcretest.
1860
1861 20. Extended pcretest with 3 new pattern features:
1862
1863 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1864 write the compiled pattern to the file whose name is "rest-of-line".
1865 This is a straight binary dump of the data, with the saved pointer to
1866 the character tables forced to be NULL. The study data, if any, is
1867 written too. After writing, pcretest reads a new pattern.
1868
1869 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1870 compiled pattern from the given file. There must not be any
1871 occurrences of "<" in the file name (pretty unlikely); if there are,
1872 pcretest will instead treat the initial "<" as a pattern delimiter.
1873 After reading in the pattern, pcretest goes on to read data lines as
1874 usual.
1875
1876 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1877 and 16-bit fields in a compiled pattern, to simulate a pattern that
1878 was compiled on a host of opposite endianness.
1879
1880 21. The pcre-exec() function can now cope with patterns that were compiled on
1881 hosts of opposite endianness, with this restriction:
1882
1883 As for any compiled expression that is saved and used later, the tables
1884 pointer field cannot be preserved; the extra_data field in the arguments
1885 to pcre_exec() should be used to pass in a tables address if a value
1886 other than the default internal tables were used at compile time.
1887
1888 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1889 now diagnosed as an error. Previously, most of the time, a negative number
1890 would have been treated as zero, but if in addition "ovector" was passed as
1891 NULL, a crash could occur.
1892
1893 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1894 new versions from the libtool 1.5 distribution (the last one is a copy of
1895 a file called libtool.m4). This seems to have fixed the need to patch
1896 "configure" to support Darwin 1.3 (which I used to do). However, I still
1897 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1898 workstation).
1899
1900 24. Changed the PCRE licence to be the more standard "BSD" licence.
1901
1902
1903 Version 4.5 01-Dec-03
1904 ---------------------
1905
1906 1. There has been some re-arrangement of the code for the match() function so
1907 that it can be compiled in a version that does not call itself recursively.
1908 Instead, it keeps those local variables that need separate instances for
1909 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1910 needs to "recurse". Keeping track of where control must go is done by means
1911 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1912 hide most of the details from the main code, and operates only if
1913 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1914 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1915 operating.
1916
1917 To make it easier for callers to provide specially tailored get/free
1918 functions for this usage, two new functions, pcre_stack_malloc, and
1919 pcre_stack_free, are used. They are always called in strict stacking order,
1920 and the size of block requested is always the same.
1921
1922 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1923 PCRE has been compiled to use the stack or the heap for recursion. The
1924 -C option of pcretest uses this to show which version is compiled.
1925
1926 A new data escape \S, is added to pcretest; it causes the amounts of store
1927 obtained and freed by both kinds of malloc/free at match time to be added
1928 to the output.
1929
1930 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1931 what's available on my current Linux desktop machine.
1932
1933 3. When matching a UTF-8 string, the test for a valid string at the start has
1934 been extended. If start_offset is not zero, PCRE now checks that it points
1935 to a byte that is the start of a UTF-8 character. If not, it returns
1936 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1937 this is necessary because there may be backward assertions in the pattern.
1938 When matching the same subject several times, it may save resources to use
1939 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1940
1941 4. The code for checking the validity of UTF-8 strings has been tightened so
1942 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1943 containing "overlong sequences".
1944
1945 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1946 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1947 should have been "&", but it just so happened that all the cases this let
1948 through by mistake were picked up later in the function.
1949
1950 6. I had used a variable called "isblank" - this is a C99 function, causing
1951 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1952
1953 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1954 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1955 so the version is shown; (c) stop "make test" echoing "./RunTest".
1956
1957 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1958
1959 9. The prototype for memmove() for systems that don't have it was using
1960 size_t, but the inclusion of the header that defines size_t was later. I've
1961 moved the #includes for the C headers earlier to avoid this.
1962
1963 10. Added some adjustments to the code to make it easier to compiler on certain
1964 special systems:
1965
1966 (a) Some "const" qualifiers were missing.
1967 (b) Added the macro EXPORT before all exported functions; by default this
1968 is defined to be empty.
1969 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1970 that it reads its output file name as an argument instead of writing
1971 to the standard output and assuming this can be redirected.
1972
1973 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1974 class containing characters with values greater than 255, PCRE compilation
1975 went into a loop.
1976
1977 12. A recursive reference to a subpattern that was within another subpattern
1978 that had a minimum quantifier of zero caused PCRE to crash. For example,
1979 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1980 recursion. If the recursively-called subpattern itself had a zero repeat,
1981 that was OK.
1982
1983 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1984 buffer into which it was copied (for escape processing) was still set at
1985 1024, so long lines caused crashes.
1986
1987 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1988 "internal error: code overflow...". This applied to any character class
1989 that was followed by a possessive quantifier.
1990
1991 15. Modified the Makefile to add libpcre.la as a prerequisite for
1992 libpcreposix.la because I was told this is needed for a parallel build to
1993 work.
1994
1995 16. If a pattern that contained .* following optional items at the start was
1996 studied, the wrong optimizing data was generated, leading to matching
1997 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1998 matching string must start with a or b or c. The correct conclusion for
1999 this pattern is that a match can start with any character.
2000
2001
2002 Version 4.4 13-Aug-03
2003 ---------------------
2004
2005 1. In UTF-8 mode, a character class containing characters with values between
2006 127 and 255 was not handled correctly if the compiled pattern was studied.
2007 In fixing this, I have also improved the studying algorithm for such
2008 classes (slightly).
2009
2010 2. Three internal functions had redundant arguments passed to them. Removal
2011 might give a very teeny performance improvement.
2012
2013 3. Documentation bug: the value of the capture_top field in a callout is *one
2014 more than* the number of the hightest numbered captured substring.
2015
2016 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2017 in incorrectly linking with a previously installed version. They now link
2018 explicitly with libpcre.la.
2019
2020 5. configure.in no longer needs to recognize Cygwin specially.
2021
2022 6. A problem in pcre.in for Windows platforms is fixed.
2023
2024 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2025 pcretest, it used to include the size of the study block as part of its
2026 output. Unfortunately, the structure contains a field that has a different
2027 size on different hardware architectures. This meant that the tests that
2028 showed this size failed. As the block is currently always of a fixed size,
2029 this information isn't actually particularly useful in pcretest output, so
2030 I have just removed it.
2031
2032 8. Three pre-processor statements accidentally did not start in column 1.
2033 Sadly, there are *still* compilers around that complain, even though
2034 standard C has not required this for well over a decade. Sigh.
2035
2036 9. In pcretest, the code for checking callouts passed small integers in the
2037 callout_data field, which is a void * field. However, some picky compilers
2038 complained about the casts involved for this on 64-bit systems. Now
2039 pcretest passes the address of the small integer instead, which should get
2040 rid of the warnings.
2041
2042 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2043 both compile and run time, and gives an error if an invalid UTF-8 sequence
2044 is found. There is a option for disabling this check in cases where the
2045 string is known to be correct and/or the maximum performance is wanted.
2046
2047 11. In response to a bug report, I changed one line in Makefile.in from
2048
2049 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2050 to
2051 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2052
2053 to look similar to other lines, but I have no way of telling whether this
2054 is the right thing to do, as I do not use Windows. No doubt I'll get told
2055 if it's wrong...
2056
2057
2058 Version 4.3 21-May-03
2059 ---------------------
2060
2061 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2062 Makefile.
2063
2064 2. Some refactoring to improve the quality of the code:
2065
2066 (i) The utf8_table... variables are now declared "const".
2067
2068 (ii) The code for \cx, which used the "case flipping" table to upper case
2069 lower case letters, now just substracts 32. This is ASCII-specific,
2070 but the whole concept of \cx is ASCII-specific, so it seems
2071 reasonable.
2072
2073 (iii) PCRE was using its character types table to recognize decimal and
2074 hexadecimal digits in the pattern. This is silly, because it handles
2075 only 0-9, a-f, and A-F, but the character types table is locale-
2076 specific, which means strange things might happen. A private
2077 table is now used for this - though it costs 256 bytes, a table is
2078 much faster than multiple explicit tests. Of course, the standard
2079 character types table is still used for matching digits in subject
2080 strings against \d.
2081
2082 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2083 ending in _t are). So I've renamed it as ESC_tee.
2084
2085 3. The first argument for regexec() in the POSIX wrapper should have been
2086 defined as "const".
2087
2088 4. Changed pcretest to use malloc() for its buffers so that they can be
2089 Electric Fenced for debugging.
2090
2091 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2092 to read one or more bytes before the start of the subject string. Often this
2093 had no effect on PCRE's behaviour, but in some circumstances it could
2094 provoke a segmentation fault.
2095
2096 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2097 to try to read one or more bytes before the start of the subject string.
2098
2099 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2100 UTF-8 support could misbehave in various ways if the subject string
2101 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2102 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2103 back over UTF-8 characters.)
2104
2105
2106 Version 4.2 14-Apr-03
2107 ---------------------
2108
2109 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2110
2111 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2112 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2113 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2114 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2115 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2116 and BUILD_EXEEXT
2117 Note: automatic setting of the BUILD variables is not yet working
2118 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2119 compile-time but not at link-time
2120 [LINK]: use for linking executables only
2121 make different versions for Windows and non-Windows
2122 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2123 libraries
2124 [LINK_FOR_BUILD]: new variable
2125 [OBJEXT]: use throughout
2126 [EXEEXT]: use throughout
2127 <winshared>: new target
2128 <wininstall>: new target
2129 <dftables.o>: use native compiler
2130 <dftables>: use native linker
2131 <install>: handle Windows platform correctly
2132 <clean>: ditto
2133 <check>: ditto
2134 copy DLL to top builddir before testing
2135
2136 As part of these changes, -no-undefined was removed again. This was reported
2137 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2138 in any case.
2139
2140 3. Some tidies to get rid of compiler warnings:
2141
2142 . In the match_data structure, match_limit was an unsigned long int, whereas
2143 match_call_count was an int. I've made them both unsigned long ints.
2144
2145 . In pcretest the fact that a const uschar * doesn't automatically cast to
2146 a void * provoked a warning.
2147
2148 . Turning on some more compiler warnings threw up some "shadow" variables
2149 and a few more missing casts.
2150
2151 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2152 option, a class that contained a single character with a value between 128
2153 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2154
2155 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2156 option, a class that contained several characters, but with at least one
2157 whose value was between 128 and 255 caused PCRE to crash.
2158
2159
2160 Version 4.1 12-Mar-03
2161 ---------------------
2162
2163 1. Compiling with gcc -pedantic found a couple of places where casts were
2164 needed, and a string in dftables.c that was longer than standard compilers are
2165 required to support.
2166
2167 2. Compiling with Sun's compiler found a few more places where the code could
2168 be tidied up in order to avoid warnings.
2169
2170 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2171 first of these names is deprecated in the latest Autoconf in favour of the name
2172 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2173 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2174 analogy I have changed it to CFLAGS_FOR_BUILD.
2175
2176 4. Added -no-undefined to the linking command in the Makefile, because this is
2177 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2178 linking step for the pcreposix library.
2179
2180 5. PCRE was failing to diagnose the case of two named groups with the same
2181 name.
2182
2183 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2184 literal character that is needed in the subject for a match, and scans along to
2185 ensure that it is present before embarking on the full matching process. This
2186 saves time in cases of nested unlimited repeats that are never going to match.
2187 Problem: the scan can take a lot of time if the subject is very long (e.g.
2188 megabytes), thus penalizing straightforward matches. It is now done only if the
2189 amount of subject to be scanned is less than 1000 bytes.
2190
2191 7. A lesser problem with the same optimization is that it was recording the
2192 first character of an anchored pattern as "needed", thus provoking a search
2193 right along the subject, even when the first match of the pattern was going to
2194 fail. The "needed" character is now not set for anchored patterns, unless it
2195 follows something in the pattern that is of non-fixed length. Thus, it still
2196 fulfils its original purpose of finding quick non-matches in cases of nested
2197 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2198
2199
2200 Version 4.0 17-Feb-03
2201 ---------------------
2202
2203 1. If a comment in an extended regex that started immediately after a meta-item
2204 extended to the end of string, PCRE compiled incorrect data. This could lead to
2205 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2206
2207 2. Moved to autoconf 2.53 and libtool 1.4.2.
2208
2209 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2210 the special perltest8 script is no longer needed - all the tests can be run
2211 from a single perltest script.
2212
2213 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2214 by \s. It has now been removed in PCRE. This means it isn't recognized as
2215 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2216 class [:space:] *does* include VT, thereby creating a mess.
2217
2218 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2219 space and tab.
2220
2221 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2222 its new features into the main test script, reducing the number of scripts.
2223
2224 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2225 were backward compatible, and made the (?i) apply to the whole pattern, as if
2226 /i were given. Now it behaves more logically, and applies the option setting
2227 only to what follows. PCRE has been changed to follow suit. However, if it
2228 finds options settings right at the start of the pattern, it extracts them into
2229 the global options, as before. Thus, they show up in the info data.
2230
2231 8. Added support for the \Q...\E escape sequence. Characters in between are
2232 treated as literals. This is slightly different from Perl in that $ and @ are
2233 also handled as literals inside the quotes. In Perl, they will cause variable
2234 interpolation. Note the following examples:
2235
2236 Pattern PCRE matches Perl matches
2237
2238 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2239 \Qabc\$xyz\E abc\$xyz abc\$xyz
2240 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2241
2242 For compatibility with Perl, \Q...\E sequences are recognized inside character
2243 classes as well as outside them.
2244
2245 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2246 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2247 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2248 signed/unsigned warnings.
2249
2250 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2251 option for pcretest, so I've replaced it by a simple function that does just
2252 that job.
2253
2254 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2255 "pcregrep -".
2256
2257 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2258 Java package. This provides some syntactic sugar for simple cases of what my
2259 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2260 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2261 item, you can use this simplified notation. Note that only makes sense with
2262 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2263 greediness, whatever the setting of the PCRE_UNGREEDY option.
2264
2265 13. A change of greediness default within a pattern was not taking effect at
2266 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2267 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2268 was abstracted outside.
2269
2270 14. PCRE now supports the \G assertion. It is true when the current matching
2271 position is at the start point of the match. This differs from \A when the
2272 starting offset is non-zero. Used with the /g option of pcretest (or similar
2273 code), it works in the same way as it does for Perl's /g option. If all
2274 alternatives of a regex begin with \G, the expression is anchored to the start
2275 match position, and the "anchored" flag is set in the compiled expression.
2276
2277 15. Some bugs concerning the handling of certain option changes within patterns
2278 have been fixed. These applied to options other than (?ims). For example,
2279 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2280 been the other way round. Some of this was related to change 7 above.
2281
2282 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2283 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2284 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2285 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2286
2287 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2288 mode. Unlike ".", it always matches newline, whatever the setting of
2289 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2290 assertions. Perl allows it, but it doesn't (in general) work because it can't
2291 calculate the length of the lookbehind. At least, that's the case for Perl
2292 5.8.0 - I've been told they are going to document that it doesn't work in
2293 future.
2294
2295 18. Added an error diagnosis for escapes that PCRE does not support: these are
2296 \L, \l, \N, \P, \p, \U, \u, and \X.
2297
2298 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2299 reading past the end of the pattern in cases such as /[abcd/.
2300
2301 20. PCRE was getting more memory than necessary for patterns with classes that
2302 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2303
2304 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2305 compiling PCRE for use with Virtual Pascal.
2306
2307 22. Small fix to the Makefile to make it work properly if the build is done
2308 outside the source tree.
2309
2310 23. Added a new extension: a condition to go with recursion. If a conditional
2311 subpattern starts with (?(R) the "true" branch is used if recursion has
2312 happened, whereas the "false" branch is used only at the top level.
2313
2314 24. When there was a very long string of literal characters (over 255 bytes
2315 without UTF support, over 250 bytes with UTF support), the computation of how
2316 much memory was required could be incorrect, leading to segfaults or other
2317 strange effects.
2318
2319 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2320 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2321 there was a subsequent back reference to those brackets. This meant that, for
2322 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2323 possible to check for precisely this case. All we can do is abandon the
2324 optimization if .* occurs inside capturing brackets when there are any back
2325 references whatsoever. (See below for a better fix that came later.)
2326
2327 26. The handling of the optimization for finding the first character of a
2328 non-anchored pattern, and for finding a character that is required later in the
2329 match were failing in some cases. This didn't break the matching; it just
2330 failed to optimize when it could. The way this is done has been re-implemented.
2331
2332 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2333
2334 28. Added a new feature that provides some of the functionality that Perl
2335 provides with (?{...}). The facility is termed a "callout". The way it is done
2336 in PCRE is for the caller to provide an optional function, by setting
2337 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2338 global variable. By default it is unset, which disables all calling out. To get
2339 the function called, the regex must include (?C) at appropriate points. This
2340 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2341 This provides a means of identifying different callout points. When PCRE
2342 reaches such a point in the regex, if pcre_callout has been set, the external
2343 function is called. It is provided with data in a structure called
2344 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2345 matching continues; if it returns a non-zero value, the match at the current
2346 point fails. However, backtracking will occur if possible. [This was changed
2347 later and other features added - see item 49 below.]
2348
2349 29. pcretest is upgraded to test the callout functionality. It provides a
2350 callout function that displays information. By default, it shows the start of
2351 the match and the current position in the text. There are some new data escapes
2352 to vary what happens:
2353
2354 \C+ in addition, show current contents of captured substrings
2355 \C- do not supply a callout function
2356 \C!n return 1 when callout number n is reached
2357 \C!n!m return 1 when callout number n is reached for the mth time
2358
2359 30. If pcregrep was called with the -l option and just a single file name, it
2360 output "<stdin>" if a match was found, instead of the file name.
2361
2362 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2363 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2364 pcre_exec(). This saves a malloc/free per call. The default value of
2365 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2366 when configuring.
2367
2368 32. The default maximum size of a compiled pattern is 64K. There have been a
2369 few cases of people hitting this limit. The code now uses macros to handle the
2370 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2371 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2372 configuring. Tests 2 and 5 work only with 2-byte links because they output
2373 debugging information about compiled patterns.
2374
2375 33. Internal code re-arrangements:
2376
2377 (a) Moved the debugging function for printing out a compiled regex into
2378 its own source file (printint.c) and used #include to pull it into
2379 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2380 separate copies.
2381
2382 (b) Defined the list of op-code names for debugging as a macro in
2383 internal.h so that it is next to the definition of the opcodes.
2384
2385 (c) Defined a table of op-code lengths for simpler skipping along compiled
2386 code. This is again a macro in internal.h so that it is next to the
2387 definition of the opcodes.
2388
2389 34. Added support for recursive calls to individual subpatterns, along the
2390 lines of Robin Houston's patch (but implemented somewhat differently).
2391
2392 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2393 allow it to read and process whole directories in Win32. This code was
2394 contributed by Lionel Fourquaux; it has not been tested by me.
2395
2396 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2397 used to name a group. Names consist of alphanumerics and underscores, and must
2398 be unique. Back references use the syntax (?P=name) and recursive calls use
2399 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2400 numbers. The function pcre_fullinfo() can be used after compilation to extract
2401 a name/number map. There are three relevant calls:
2402
2403 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2404 PCRE_INFO_NAMECOUNT yields the number of entries
2405 PCRE_INFO_NAMETABLE yields a pointer to the map.
2406
2407 The map is a vector of fixed-size entries. The size of each entry depends on
2408 the length of the longest name used. The first two bytes of each entry are the
2409 group number, most significant byte first. There follows the corresponding
2410 name, zero terminated. The names are in alphabetical order.
2411
2412 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2413 case instead of 255. Making it the same both with and without UTF-8 support
2414 means that the same test output works with both.
2415
2416 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2417 calling malloc() with a zero argument.
2418
2419 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2420 optimization. I've improved things by keeping a bitmap of backreferences with
2421 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2422 fact referenced, the optimization can be applied. It is unlikely that a
2423 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2424 the match to follow \n) will appear inside brackets with a number greater than
2425 31, but if it does, any back reference > 31 suppresses the optimization.
2426
2427 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2428 of disabling numbered capturing parentheses. Any opening parenthesis that is
2429 not followed by ? behaves as if it were followed by ?: but named parentheses
2430 can still be used for capturing (and they will acquire numbers in the usual
2431 way).
2432
2433 41. Redesigned the return codes from the match() function into yes/no/error so
2434 that errors can be passed back from deep inside the nested calls. A malloc
2435 failure while inside a recursive subpattern call now causes the
2436 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2437
2438 42. It is now possible to set a limit on the number of times the match()
2439 function is called in a call to pcre_exec(). This facility makes it possible to
2440 limit the amount of recursion and backtracking, though not in a directly
2441 obvious way, because the match() function is used in a number of different
2442 circumstances. The count starts from zero for each position in the subject
2443 string (for non-anchored patterns). The default limit is, for compatibility, a
2444 large number, namely 10 000 000. You can change this in two ways:
2445
2446 (a) When configuring PCRE before making, you can use --with-match-limit=n
2447 to set a default value for the compiled library.
2448
2449 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2450 a different value is set. See 45 below.
2451
2452 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2453
2454 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2455 of things that can be changed at compile time. The first argument specifies
2456 what is wanted and the second points to where the information is to be placed.
2457 The current list of available information is:
2458
2459 PCRE_CONFIG_UTF8
2460
2461 The output is an integer that is set to one if UTF-8 support is available;
2462 otherwise it is set to zero.
2463
2464 PCRE_CONFIG_NEWLINE
2465
2466 The output is an integer that it set to the value of the code that is used for
2467 newline. It is either LF (10) or CR (13).
2468
2469 PCRE_CONFIG_LINK_SIZE
2470
2471 The output is an integer that contains the number of bytes used for internal
2472 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2473
2474 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2475
2476 The output is an integer that contains the threshold above which the POSIX
2477 interface uses malloc() for output vectors. See item 31 above.
2478
2479 PCRE_CONFIG_MATCH_LIMIT
2480
2481 The output is an unsigned integer that contains the default limit of the number
2482 of match() calls in a pcre_exec() execution. See 42 above.
2483
2484 44. pcretest has been upgraded by the addition of the -C option. This causes it
2485 to extract all the available output from the new pcre_config() function, and to
2486 output it. The program then exits immediately.
2487
2488 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2489 order to support additional features. One way would have been to define
2490 pcre_exec2() (for example) with extra arguments, but this would not have been
2491 extensible, and would also have required all calls to the original function to
2492 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2493 is used for passing in "extra" data from pcre_study().
2494
2495 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2496 contains the following fields:
2497
2498 flags a bitmap indicating which of the following fields are set
2499 study_data opaque data from pcre_study()
2500 match_limit a way of specifying a limit on match() calls for a specific
2501 call to pcre_exec()
2502 callout_data data for callouts (see 49 below)
2503
2504 The flag bits are also defined in pcre.h, and are
2505
2506 PCRE_EXTRA_STUDY_DATA
2507 PCRE_EXTRA_MATCH_LIMIT
2508 PCRE_EXTRA_CALLOUT_DATA
2509
2510 The pcre_study() function now returns one of these new pcre_extra blocks, with
2511 the actual study data pointed to by the study_data field, and the
2512 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2513 before. That is, this change is entirely upwards-compatible and requires no
2514 change to existing code.
2515
2516 If you want to pass in additional data to pcre_exec(), you can either place it
2517 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2518 block.
2519
2520 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2521 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2522 times with different match limits, until it finds the minimum value needed for
2523 pcre_exec() to complete. The value is then output. This can be instructive; for
2524 most simple matches the number is quite small, but for pathological cases it
2525 gets very large very quickly.
2526
2527 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2528 returns the size of the data block pointed to by the study_data field in a
2529 pcre_extra block, that is, the value that was passed as the argument to
2530 pcre_malloc() when PCRE was getting memory in which to place the information
2531 created by pcre_study(). The fourth argument should point to a size_t variable.
2532 pcretest has been extended so that this information is shown after a successful
2533 pcre_study() call when information about the compiled regex is being displayed.
2534
2535 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2536 because what follows is always an absolute path. (Later: it turns out that this
2537 is more than cosmetic for MinGW, because it doesn't like empty path
2538 components.)
2539
2540 49. Some changes have been made to the callout feature (see 28 above):
2541
2542 (i) A callout function now has three choices for what it returns:
2543
2544 0 => success, carry on matching
2545 > 0 => failure at this point, but backtrack if possible
2546 < 0 => serious error, return this value from pcre_exec()
2547
2548 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2549 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2550 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2551 use by callout functions. It will never be used by PCRE itself.
2552
2553 (ii) The pcre_extra structure (see 45 above) has a void * field called
2554 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2555 pcre_callout_block structure has a field of the same name. The contents of
2556 the field passed in the pcre_extra structure are passed to the callout
2557 function in the corresponding field in the callout block. This makes it
2558 easier to use the same callout-containing regex from multiple threads. For
2559 testing, the pcretest program has a new data escape
2560
2561 \C*n pass the number n (may be negative) as callout_data
2562
2563 If the callout function in pcretest receives a non-zero value as
2564 callout_data, it returns that value.
2565
2566 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2567 there were some redundant $(CFLAGS) in commands that are now specified as
2568 $(LINK), which already includes $(CFLAGS).
2569
2570 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2571 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2572 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2573 one-byte characters throughout. Note that case-insensitive matching applies
2574 only to characters whose values are less than 256. PCRE doesn't support the
2575 notion of cases for higher-valued characters.
2576
2577 (i) A character class whose characters are all within 0-255 is handled as
2578 a bit map, and the map is inverted for negative classes. Previously, a
2579 character > 255 always failed to match such a class; however it should
2580 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2581
2582 (ii) A negated character class with a single character < 255 is coded as
2583 "not this character" (OP_NOT). This wasn't working properly when the test
2584 character was multibyte, either singly or repeated.
2585
2586 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2587 mode, for example: \x{100}{2,3}.
2588
2589 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2590 singly or repeated) now correctly test multibyte characters. However,
2591 PCRE doesn't recognize any characters with values greater than 255 as
2592 digits, spaces, or word characters. Such characters always match \D, \S,
2593 and \W, and never match \d, \s, or \w.
2594
2595 (v) Classes may now contain characters and character ranges with values
2596 greater than 255. For example: [ab\x{100}-\x{400}].
2597
2598 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2599 PCRE in UTF-8 mode.
2600
2601 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2602 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2603 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2604 value.)
2605
2606 53. The single man page has become too large. I have therefore split it up into
2607 a number of separate man pages. These also give rise to individual HTML pages;
2608 these are now put in a separate directory, and there is an index.html page that
2609 lists them all. Some hyperlinking between the pages has been installed.
2610
2611 54. Added convenience functions for handling named capturing parentheses.
2612
2613 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2614 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2615 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2616 are faulted.
2617
2618 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2619 calling configure. These values are used when compiling the dftables.c program
2620 which is run to generate the source of the default character tables. They
2621 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2622 you will need to set these values.
2623
2624 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2625
2626
2627 Version 3.9 02-Jan-02
2628 ---------------------
2629
2630 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2631
2632 2. If --disable-static was given, the building process failed when trying to
2633 build pcretest and pcregrep. (For some reason it was using libtool to compile
2634 them, which is not right, as they aren't part of the library.)
2635
2636
2637 Version 3.8 18-Dec-01
2638 ---------------------
2639
2640 1. The experimental UTF-8 code was completely screwed up. It was packing the
2641 bytes in the wrong order. How dumb can you get?
2642
2643
2644 Version 3.7 29-Oct-01
2645 ---------------------
2646
2647 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2648 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2649 this didn't happen under Solaris 8, where I normally test things.
2650
2651 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2652 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2653 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2654 this fix an hour or so after the initial 3.7 release.)
2655
2656
2657 Version 3.6 23-Oct-01
2658 ---------------------
2659
2660 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2661 offsets passed as NULL with zero offset count.
2662
2663 2. The config.guess and config.sub files had not been updated when I moved to
2664 the latest autoconf.
2665
2666
2667 Version 3.5 15-Aug-01
2668 ---------------------
2669
2670 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2671 had been forgotten.
2672
2673 2. By using declared but undefined structures, we can avoid using "void"
2674 definitions in pcre.h while keeping the internal definitions of the structures
2675 private.
2676
2677 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2678 user point of view, this means that both static and shared libraries are built
2679 by default, but this can be individually controlled. More of the work of
2680 handling this static/shared cases is now inside libtool instead of PCRE's make
2681 file.
2682
2683 4. The pcretest utility is now installed along with pcregrep because it is
2684 useful for users (to test regexs) and by doing this, it automatically gets
2685 relinked by libtool. The documentation has been turned into a man page, so
2686 there are now .1, .txt, and .html versions in /doc.
2687
2688 5. Upgrades to pcregrep:
2689 (i) Added long-form option names like gnu grep.
2690 (ii) Added --help to list all options with an explanatory phrase.
2691 (iii) Added -r, --recursive to recurse into sub-directories.
2692 (iv) Added -f, --file to read patterns from a file.
2693
2694 6. pcre_exec() was referring to its "code" argument before testing that
2695 argument for NULL (and giving an error if it was NULL).
2696
2697 7. Upgraded Makefile.in to allow for compiling in a different directory from
2698 the source directory.
2699
2700 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2701 options bits, the pointer it was passed was to an int instead of to an unsigned
2702 long int. This mattered only on 64-bit systems.
2703
2704 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2705 generated) instead of pcre.in, which it its source. Also made the same change
2706 in several of the .c files.
2707
2708 10. A new release of gcc defines printf() as a macro, which broke pcretest
2709 because it had an ifdef in the middle of a string argument for printf(). Fixed
2710 by using separate calls to printf().
2711
2712 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2713 script, to force use of CR or LF instead of \n in the source. On non-Unix
2714 systems, the value can be set in config.h.
2715
2716 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2717 absolute limit. Changed the text of the error message to make this clear, and
2718 likewise updated the man page.
2719
2720 13. The limit of 99 on the number of capturing subpatterns has been removed.
2721 The new limit is 65535, which I hope will not be a "real" limit.
2722
2723
2724 Version 3.4 22-Aug-00
2725 ---------------------
2726
2727 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2728
2729 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2730
2731
2732 Version 3.3 01-Aug-00
2733 ---------------------
2734
2735 1. If an octal character was given, but the value was greater than \377, it
2736 was not getting masked to the least significant bits, as documented. This could
2737 lead to crashes in some systems.
2738
2739 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2740 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2741
2742 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2743 These just pass their arguments on to (pcre_free)(), but they are provided
2744 because some uses of PCRE bind it to non-C systems that can call its functions,
2745 but cannot call free() or pcre_free() directly.
2746
2747 4. Add "make test" as a synonym for "make check". Corrected some comments in
2748 the Makefile.
2749
2750 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2751 Makefile.
2752
2753 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2754 command called pgrep for grepping around the active processes.
2755
2756 7. Added the beginnings of support for UTF-8 character strings.
2757
2758 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2759 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2760 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2761 out for the ar command.)
2762
2763
2764 Version 3.2 12-May-00
2765 ---------------------
2766
2767 This is purely a bug fixing release.
2768
2769 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2770 of ZA. This was just one example of several cases that could provoke this bug,
2771 which was introduced by change 9 of version 2.00. The code for breaking
2772 infinite loops after an iteration that matches an empty string was't working
2773 correctly.
2774
2775 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2776 when matched against abbab (for example). After matching an empty string, it
2777 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2778 caused it to match further down the string than it should.
2779
2780 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2781 was there because it doesn't seem to be needed, and it causes trouble on some
2782 systems, as it is not a Standard C header. It has been removed.
2783
2784 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2785 were reported on the Macintosh. The changes were from
2786
2787 while ((c = *(++ptr)) != 0 && c != '\n');
2788 to
2789 while ((c = *(++ptr)) != 0 && c != '\n') ;
2790
2791 Totally extraordinary, but if that's what it takes...
2792
2793 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2794 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2795 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2796 assumes the way PCRE uses memmove() (always moving upwards).
2797
2798 6. PCRE is being used in one environment where strchr() is not available. There
2799 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2800 faster code anyway.
2801
2802
2803 Version 3.1 09-Feb-00
2804 ---------------------
2805
2806 The only change in this release is the fixing of some bugs in Makefile.in for
2807 the "install" target:
2808
2809 (1) It was failing to install pcreposix.h.
2810
2811 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2812
2813
2814 Version 3.0 01-Feb-00
2815 ---------------------
2816
2817 1. Add support for the /+ modifier to perltest (to output $` like it does in
2818 pcretest).
2819
2820 2. Add support for the /g modifier to perltest.
2821
2822 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2823 matches null strings.
2824
2825 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2826 pattern. Perl treats empty patterns specially - it reuses the most recent
2827 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2828 effect.
2829
2830 5. The POSIX interface was broken in that it was just handing over the POSIX
2831 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2832 required a bigger vector, with some working space on the end. This means that
2833 the POSIX wrapper now has to get and free some memory, and copy the results.
2834
2835 6. Added some simple autoconf support, placing the test data and the
2836 documentation in separate directories, re-organizing some of the
2837 information files, and making it build pcre-config (a GNU standard). Also added
2838 libtool support for building PCRE as a shared library, which is now the
2839 default.
2840
2841 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2842 09 are not valid octal constants. Single digits will be used for minor values
2843 less than 10.
2844
2845 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2846 existing programs that set these in the POSIX interface can use PCRE without
2847 modification.
2848
2849 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2850 return all that pcre_info() returns, plus additional data. The pcre_info()
2851 function is retained for compatibility, but is considered to be obsolete.
2852
2853 10. Added experimental recursion feature (?R) to handle one common case that
2854 Perl 5.6 will be able to do with (?p{...}).
2855
2856 11. Added support for POSIX character classes like [:alpha:], which Perl is
2857 adopting.
2858
2859
2860 Version 2.08 31-Aug-99
2861 ----------------------
2862
2863 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2864 trying to match at the startoffset position, but instead was moving forward to
2865 the next newline as if a previous match had failed.
2866
2867 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2868 and could get into a loop if a null string was matched other than at the start
2869 of the subject.
2870
2871 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2872 be distinguished at compile time, and for completeness also added PCRE_DATE.
2873
2874 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2875 in GnuWin32 environments.
2876
2877
2878 Version 2.07 29-Jul-99
2879 ----------------------
2880
2881 1. The documentation is now supplied in plain text form and HTML as well as in
2882 the form of man page sources.
2883
2884 2. C++ compilers don't like assigning (void *) values to other pointer types.
2885 In particular this affects malloc(). Although there is no problem in Standard
2886 C, I've put in casts to keep C++ compilers happy.
2887
2888 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2889 should be (const char *).
2890
2891 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2892 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2893 However, I haven't made this a standard facility. The documentation doesn't
2894 mention it, and the Makefile doesn't support it.
2895
2896 5. The Makefile now contains an "install" target, with editable destinations at
2897 the top of the file. The pcretest program is not installed.
2898
2899 6. pgrep -V now gives the PCRE version number and date.
2900
2901 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2902 causing the entire string to be ignored, instead of just the last character.
2903
2904 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2905 non-matching string, it can take a very, very long time, even for strings of
2906 quite modest length, because of the nested recursion. PCRE now does better in
2907 some of these cases. It does this by remembering the last required literal
2908 character in the pattern, and pre-searching the subject to ensure it is present
2909 before running the real match. In other words, it applies a heuristic to detect
2910 some types of certain failure quickly, and in the above example, if presented
2911 with a string that has no trailing " it gives "no match" very quickly.
2912
2913 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2914 other alternatives are tried instead.
2915
2916
2917 Version 2.06 09-Jun-99
2918 ----------------------
2919
2920 1. Change pcretest's output for amount of store used to show just the code
2921 space, because the remainder (the data block) varies in size between 32-bit and
2922 64-bit systems.
2923
2924 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2925 start matching at. This allows lookbehinds to work when searching for multiple
2926 occurrences in a string.
2927
2928 3. Added additional options to pcretest for testing multiple occurrences:
2929
2930 /+ outputs the rest of the string that follows a match
2931 /g loops for multiple occurrences, using the new startoffset argument
2932 /G loops for multiple occurrences by passing an incremented pointer
2933
2934 4. PCRE wasn't doing the "first character" optimization for patterns starting
2935 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2936 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2937 the letter 'x'. On long subject strings, this gives a significant speed-up.
2938
2939
2940 Version 2.05 21-Apr-99
2941 ----------------------
2942
2943 1. Changed the type of magic_number from int to long int so that it works
2944 properly on 16-bit systems.
2945
2946 2. Fixed a bug which caused patterns starting with .* not to work correctly
2947 when the subject string contained newline characters. PCRE was assuming
2948 anchoring for such patterns in all cases, which is not correct because .* will
2949 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2950 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2951 must be retried after every newline in the subject.
2952
2953
2954 Version 2.04 18-Feb-99
2955 ----------------------
2956
2957 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2958 computation of the store needed to hold the pattern was incorrect (too large).
2959 If such patterns were nested a few deep, this could multiply and become a real
2960 problem.
2961
2962 2. Added /M option to pcretest to show the memory requirement of a specific
2963 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2964
2965 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2966 compiled in such a way that the backtracking after subsequent failure was
2967 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2968 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2969
2970
2971 Version 2.03 02-Feb-99
2972 ----------------------
2973
2974 1. Fixed typo and small mistake in man page.
2975
2976 2. Added 4th condition (GPL supersedes if conflict) and created separate
2977 LICENCE file containing the conditions.
2978
2979 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2980 Perl, that is the internal \ allows the delimiter to be included in the
2981 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2982 the final delimiter, add \ to the end of the pattern (to test the error).
2983
2984 4. Added the convenience functions for extracting substrings after a successful
2985 match. Updated pcretest to make it able to test these functions.
2986
2987
2988 Version 2.02 14-Jan-99
2989 ----------------------
2990
2991 1. Initialized the working variables associated with each extraction so that
2992 their saving and restoring doesn't refer to uninitialized store.
2993
2994 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2995 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2996 fix the problem.
2997
2998 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2999 calls, and wasn't printing the correct value for compiling calls. Increased the
3000 default value of LOOPREPEAT, and the number of significant figures in the
3001 times.
3002
3003 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3004
3005 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3006 a building problem on Windows NT with a FAT file system.
3007
3008
3009 Version 2.01 21-Oct-98
3010 ----------------------
3011
3012 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3013 to character tables built by pcre_maketables() in the current locale. If NULL
3014 is passed, the default tables are used.
3015
3016
3017 Version 2.00 24-Sep-98
3018 ----------------------
3019
3020 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3021 it any more.
3022
3023 2. Allow quantification of (?>) groups, and make it work correctly.
3024
3025 3. The first character computation wasn't working for (?>) groups.
3026
3027 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3028 end of the subject) and add 5.005's \z, which really does match only at the
3029 very end of the subject.
3030
3031 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3032
3033 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3034 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3035 localized options. All options to pcre_study() were also removed.
3036
3037 7. Add other new features from 5.005:
3038
3039 $(?<= positive lookbehind
3040 $(?<! negative lookbehind
3041 (?imsx-imsx) added the unsetting capability
3042 such a setting is global if at outer level; local otherwise
3043 (?imsx-imsx:) non-capturing groups with option setting
3044 (?(cond)re|re) conditional pattern matching
3045
3046 A backreference to itself in a repeated group matches the previous
3047 captured string.
3048
3049 8. General tidying up of studying (both automatic and via "study")
3050 consequential on the addition of new assertions.
3051
3052 9. As in 5.005, unlimited repeated groups that could match an empty substring
3053 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3054 runtime if any iteration does actually match an empty substring.
3055
3056 10. Include the RunTest script in the distribution.
3057
3058 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3059 discrepancies, some of which were old and were also with respect to 5.004. They
3060 have now been fixed.
3061
3062
3063 Version 1.09 28-Apr-98
3064 ----------------------
3065
3066 1. A negated single character class followed by a quantifier with a minimum
3067 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3068 program crashes, or just wrong answers. This did not apply to negated classes
3069 containing more than one character, or to minima other than one.
3070
3071
3072 Version 1.08 27-Mar-98
3073 ----------------------
3074
3075 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3076
3077 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3078 latter must appear before anything that relies on it in the pattern.
3079
3080
3081 Version 1.07 16-Feb-98
3082 ----------------------
3083
3084 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3085 repeat of a potentially empty string).
3086
3087
3088 Version 1.06 23-Jan-98
3089 ----------------------
3090
3091 1. Added Markus Oberhumer's little patches for C++.
3092
3093 2. Literal strings longer than 255 characters were broken.
3094
3095
3096 Version 1.05 23-Dec-97
3097 ----------------------
3098
3099 1. Negated character classes containing more than one character were failing if
3100 PCRE_CASELESS was set at run time.
3101
3102
3103 Version 1.04 19-Dec-97
3104 ----------------------
3105
3106 1. Corrected the man page, where some "const" qualifiers had been omitted.
3107
3108 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3109 input syntax.
3110
3111 3. Fixed memory leak which occurred when a regex with back references was
3112 matched with an offsets vector that wasn't big enough. The temporary memory
3113 that is used in this case wasn't being freed if the match failed.
3114
3115 4. Tidied pcretest to ensure it frees memory that it gets.
3116
3117 5. Temporary memory was being obtained in the case where the passed offsets
3118 vector was exactly big enough.
3119
3120 6. Corrected definition of offsetof() from change 5 below.
3121
3122 7. I had screwed up change 6 below and broken the rules for the use of
3123 setjmp(). Now fixed.
3124
3125
3126 Version 1.03 18-Dec-97
3127 ----------------------
3128
3129 1. A erroneous regex with a missing opening parenthesis was correctly
3130 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3131 on some systems.
3132
3133 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3134 it was reported that one broken compiler failed on the former because "code" is
3135 also an independent variable.
3136
3137 3. The erroneous regex a[]b caused an array overrun reference.
3138
3139 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3140 fail on data ending with that character. (It was going on too far, and checking
3141 the next character, typically a binary zero.) This was specific to the
3142 optimized code for single-character negative classes.
3143
3144 5. Added a contributed patch from the TIN world which does the following:
3145
3146 + Add an undef for memmove, in case the the system defines a macro for it.
3147
3148 + Add a definition of offsetof(), in case there isn't one. (I don't know
3149 the reason behind this - offsetof() is part of the ANSI standard - but
3150 it does no harm).
3151
3152 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3153 most of the places where whitespace preceded '#'. I have given up and
3154 allowed the remaining 2 cases to be at the margin.
3155
3156 + Rename some variables in pcre to eliminate shadowing. This seems very
3157 pedantic, but does no harm, of course.
3158
3159 6. Moved the call to setjmp() into its own function, to get rid of warnings
3160 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3161
3162 7. Constructs such as \d{8,} were compiling into the equivalent of
3163 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3164 outcome, but in this particular case used more store than had been allocated,
3165 which caused the bug to be discovered because it threw up an internal error.
3166
3167 8. The debugging code in both pcre and pcretest for outputting the compiled
3168 form of a regex was going wrong in the case of back references followed by
3169 curly-bracketed repeats.
3170
3171
3172 Version 1.02 12-Dec-97
3173 ----------------------
3174
3175 1. Typos in pcre.3 and comments in the source fixed.
3176
3177 2. Applied a contributed patch to get rid of places where it used to remove
3178 'const' from variables, and fixed some signed/unsigned and uninitialized
3179 variable warnings.
3180
3181 3. Added the "runtest" target to Makefile.
3182
3183 4. Set default compiler flag to -O2 rather than just -O.
3184
3185
3186 Version 1.01 19-Nov-97
3187 ----------------------
3188
3189 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3190 like /([ab]*)*/, that is, for classes with more than one character in them.
3191
3192 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3193 as /((?>a*))*/ (a PCRE_EXTRA facility).
3194
3195
3196 Version 1.00 18-Nov-97
3197 ----------------------
3198
3199 1. Added compile-time macros to support systems such as SunOS4 which don't have
3200 memmove() or strerror() but have other things that can be used instead.
3201
3202 2. Arranged that "make clean" removes the executables.
3203
3204
3205 Version 0.99 27-Oct-97
3206 ----------------------
3207
3208 1. Fixed bug in code for optimizing classes with only one character. It was
3209 initializing a 32-byte map regardless, which could cause it to run off the end
3210 of the memory it had got.
3211
3212 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3213
3214
3215 Version 0.98 22-Oct-97
3216 ----------------------
3217
3218 1. Fixed bug in code for handling temporary memory usage when there are more
3219 back references than supplied space in the ovector. This could cause segfaults.
3220
3221
3222 Version 0.97 21-Oct-97
3223 ----------------------
3224
3225 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3226
3227 2. Optimized negated single characters not to use a bit map.
3228
3229 3. Brought error texts together as macro definitions; clarified some of them;
3230 fixed one that was wrong - it said "range out of order" when it meant "invalid
3231 escape sequence".
3232
3233 4. Changed some char * arguments to const char *.
3234
3235 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3236
3237 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3238 pcretest.
3239
3240
3241 Version 0.96 16-Oct-97
3242 ----------------------
3243
3244 1. Added a simple "pgrep" utility to the distribution.
3245
3246 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3247 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3248 where "ddd" means "one or more decimal digits".
3249
3250 3. Fixed serious bug. If a pattern had a back reference, but the call to
3251 pcre_exec() didn't supply a large enough ovector to record the related
3252 identifying subpattern, the match always failed. PCRE now remembers the number
3253 of the largest back reference, and gets some temporary memory in which to save
3254 the offsets during matching if necessary, in order to ensure that
3255 backreferences always work.
3256
3257 4. Increased the compatibility with Perl in a number of ways:
3258
3259 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3260 to request this handling. The option can be set at compile or exec time.
3261
3262 (b) $ matches before a terminating newline by default; an option
3263 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3264 mode). The option can be set at compile or exec time.
3265
3266 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3267 the same as Perl's. If the decimal number it represents is less than 10
3268 or there aren't that many previous left capturing parentheses, an octal
3269 escape is read. Inside a character class, it's always an octal escape,
3270 even if it is a single digit.
3271
3272 (d) An escaped but undefined alphabetic character is taken as a literal,
3273 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3274 escapes.
3275
3276 (e) {0} is now permitted. (The previous item is removed from the compiled
3277 pattern).
3278
3279 5. Changed all the names of code files so that the basic parts are no longer
3280 than 10 characters, and abolished the teeny "globals.c" file.
3281
3282 6. Changed the handling of character classes; they are now done with a 32-byte
3283 bit map always.
3284
3285 7. Added the -d and /D options to pcretest to make it possible to look at the
3286 internals of compilation without having to recompile pcre.
3287
3288
3289 Version 0.95 23-Sep-97
3290 ----------------------
3291
3292 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3293 \x20 at the start of a run of normal characters. These were being treated as
3294 real characters, instead of the source characters being re-checked.
3295
3296
3297 Version 0.94 18-Sep-97
3298 ----------------------
3299
3300 1. The functions are now thread-safe, with the caveat that the global variables
3301 containing pointers to malloc() and free() or alternative functions are the
3302 same for all threads.
3303
3304 2. Get pcre_study() to generate a bitmap of initial characters for non-
3305 anchored patterns when this is possible, and use it if passed to pcre_exec().
3306
3307
3308 Version 0.93 15-Sep-97
3309 ----------------------
3310
3311 1. /(b)|(:+)/ was computing an incorrect first character.
3312
3313 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3314 but not actually doing anything yet.
3315
3316 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3317 as Perl does (e.g. [-az] or [az-]).
3318
3319 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3320 all possible positions.
3321
3322 5. Split up into different modules to avoid including unneeded functions in a
3323 compiled binary. However, compile and exec are still in one module. The "study"
3324 function is split off.
3325
3326 6. The character tables are now in a separate module whose source is generated
3327 by an auxiliary program - but can then be edited by hand if required. There are
3328 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3329 toupper() in the code.
3330
3331 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3332 make them global. Abolish the function for setting them, as the caller can now
3333 set them directly.
3334
3335
3336 Version 0.92 11-Sep-97
3337 ----------------------
3338
3339 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3340 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3341
3342 2. Caseless matching was not working in character classes if the characters in
3343 the pattern were in upper case.
3344
3345 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3346
3347 4. Make PCRE_ANCHORED public and accept as a compile option.
3348
3349 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3350 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3351 pass them.
3352
3353 6. Give an error if bad option bits passed at compile or run time.
3354
3355 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3356 pcretest to cause it to pass that flag.
3357
3358 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3359 options, and the first character, if set.
3360
3361 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3362
3363
3364 Version 0.91 10-Sep-97
3365 ----------------------
3366
3367 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3368 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3369
3370 2. PCRE was looping on encountering an indefinitely repeated back reference to
3371 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3372 Perl does - treats the match as successful.
3373
3374 ****

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