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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 359 - (show annotations) (download)
Wed Jul 9 16:20:19 2008 UTC (5 years, 9 months ago) by ph10
File size: 158007 byte(s)
Add PCRE_CALL_CONVENTION for MSVC users; add some missing PCRE_EXP_DEFNs.

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

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