ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 411 - (show annotations)
Fri Apr 10 15:40:21 2009 UTC (12 years, 1 month ago) by ph10
File size: 165409 byte(s)
Wrap fileno/isatty defs for Windows in #ifndefs
1 ChangeLog for PCRE
2 ------------------
4 Version 7.9 10-Apr-09
5 ---------------------
7 1. When building with support for bzlib/zlib (pcregrep) and/or readline
8 (pcretest), all targets were linked against these libraries. This included
9 libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10 libraries. This caused unwanted dependencies to be created. This problem
11 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12 pcretest is linked with readline.
14 2. The "typedef int BOOL" in pcre_internal.h that was included inside the
15 "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16 moved outside it again, because FALSE and TRUE are already defined in AIX,
17 but BOOL is not.
19 3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20 PCRE_MATCH_LIMIT_RECURSION values as ints, when they should be long ints.
22 4. The pcregrep documentation said spaces were inserted as well as colons (or
23 hyphens) following file names and line numbers when outputting matching
24 lines. This is not true; no spaces are inserted. I have also clarified the
25 wording for the --colour (or --color) option.
27 5. In pcregrep, when --colour was used with -o, the list of matching strings
28 was not coloured; this is different to GNU grep, so I have changed it to be
29 the same.
31 6. When --colo(u)r was used in pcregrep, only the first matching substring in
32 each matching line was coloured. Now it goes on to look for further matches
33 of any of the test patterns, which is the same behaviour as GNU grep.
35 7. A pattern that could match an empty string could cause pcregrep to loop; it
36 doesn't make sense to accept an empty string match in pcregrep, so I have
37 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38 seems to be how GNU grep behaves.
40 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41 start or after a newline", because the conditional assertion was not being
42 correctly handled. The rule now is that both the assertion and what follows
43 in the first alternative must satisfy the test.
45 9. If auto-callout was enabled in a pattern with a conditional group whose
46 condition was an assertion, PCRE could crash during matching, both with
47 pcre_exec() and pcre_dfa_exec().
49 10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
50 used for matching.
52 11. Unicode property support in character classes was not working for
53 characters (bytes) greater than 127 when not in UTF-8 mode.
55 12. Added the -M command line option to pcretest.
57 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
59 15. Added the PCRE_NO_START_OPTIMIZE match-time option.
61 16. Added comments and documentation about mis-use of no_arg in the C++
62 wrapper.
64 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
65 from Martin Jerabek that uses macro names for all relevant character and
66 string constants.
68 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
69 SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
70 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
71 these, but not everybody uses configure.
73 19. A conditional group that had only one branch was not being correctly
74 recognized as an item that could match an empty string. This meant that an
75 enclosing group might also not be so recognized, causing infinite looping
76 (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
77 with the subject "ab", where knowledge that the repeated group can match
78 nothing is needed in order to break the loop.
80 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
81 exec(), but without supplying a callout function, matching went wrong.
83 21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
84 leak if the size of the offset vector was greater than 30. When the vector
85 is smaller, the saved offsets during recursion go onto a local stack
86 vector, but for larger vectors malloc() is used. It was failing to free
87 when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
88 error, in fact).
90 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
91 heapframe that is used only when UTF-8 support is enabled. This caused no
92 problem, but was untidy.
94 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
95 CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
96 included within another project.
98 24. Steven Van Ingelgem's patches to add more options to the CMake support,
99 slightly modified by me:
101 (a) PCRE_BUILD_TESTS can be set OFF not to build the tests, including
102 not building pcregrep.
104 (b) PCRE_BUILD_PCREGREP can be see OFF not to build pcregrep, but only
105 if PCRE_BUILD_TESTS is also set OFF, because the tests use pcregrep.
107 25. Forward references, both numeric and by name, in patterns that made use of
108 duplicate group numbers, could behave incorrectly or give incorrect errors,
109 because when scanning forward to find the reference group, PCRE was not
110 taking into account the duplicate group numbers. A pattern such as
111 ^X(?3)(a)(?|(b)|(q))(Y) is an example.
113 26. Changed a few more instances of "const unsigned char *" to USPTR, making
114 the feature of a custom pointer more persuasive (as requested by a user).
116 27. Wrapped the definitions of fileno and isatty for Windows, which appear in
117 pcretest.c, inside #ifndefs, because it seems they are sometimes already
118 pre-defined.
122 Version 7.8 05-Sep-08
123 ---------------------
125 1. Replaced UCP searching code with optimized version as implemented for Ad
126 Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
127 stage table and inline lookup instead of a function, giving speed ups of 2
128 to 5 times on some simple patterns that I tested. Permission was given to
129 distribute the MultiStage2.py script that generates the tables (it's not in
130 the tarball, but is in the Subversion repository).
132 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
133 scripts.
135 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
136 a group with a zero qualifier. The result of the study could be incorrect,
137 or the function might crash, depending on the pattern.
139 4. Caseless matching was not working for non-ASCII characters in back
140 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
141 It now works when Unicode Property Support is available.
143 5. In pcretest, an escape such as \x{de} in the data was always generating
144 a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
145 non-UTF-8 mode. If the value is greater than 255, it gives a warning about
146 truncation.
148 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
150 7. Added two (int) casts to pcregrep when printing the difference of two
151 pointers, in case they are 64-bit values.
153 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
154 test 2 if it fails.
156 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
157 and a #define of that name to empty if it is not externally set. This is to
158 allow users of MSVC to set it if necessary.
160 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
161 the convenience functions in the pcre_get.c source file.
163 11. An option change at the start of a pattern that had top-level alternatives
164 could cause overwriting and/or a crash. This command provoked a crash in
165 some environments:
167 printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
169 This potential security problem was recorded as CVE-2008-2371.
171 12. For a pattern where the match had to start at the beginning or immediately
172 after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
173 pcre_dfa_exec() could read past the end of the passed subject if there was
174 no match. To help with detecting such bugs (e.g. with valgrind), I modified
175 pcretest so that it places the subject at the end of its malloc-ed buffer.
177 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
178 exec() might read past the end of the data buffer in UTF-8 mode.
180 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
181 the data contained the byte 0x85 as part of a UTF-8 character within its
182 first line. This applied both to normal and DFA matching.
184 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
185 /^[^d]*?$/8 failed to match "abc".
187 16. Added a missing copyright notice to pcrecpp_internal.h.
189 17. Make it more clear in the documentation that values returned from
190 pcre_exec() in ovector are byte offsets, not character counts.
192 18. Tidied a few places to stop certain compilers from issuing warnings.
194 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
195 supplied by Stefan Weber. I made a further small update for 7.8 because
196 there is a change of source arrangements: the pcre_searchfuncs.c module is
197 replaced by pcre_ucd.c.
200 Version 7.7 07-May-08
201 ---------------------
203 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
204 a string to a long long, pretend we don't even have a long long." This is
205 done by checking for the strtoq, strtoll, and _strtoi64 functions.
207 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
208 pre-7.6 versions, which defined a global no_arg variable instead of putting
209 it in the RE class. (See also #8 below.)
211 3. Remove a line of dead code, identified by coverity and reported by Nuno
212 Lopes.
214 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
216 (1) The include/exclude patterns were being applied to the whole pathnames
217 of files, instead of just to the final components.
219 (2) If there was more than one level of directory, the subdirectories were
220 skipped unless they satisfied the include/exclude conditions. This is
221 inconsistent with GNU grep (and could even be seen as contrary to the
222 pcregrep specification - which I improved to make it absolutely clear).
223 The action now is always to scan all levels of directory, and just
224 apply the include/exclude patterns to regular files.
226 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
227 --exclude_dir in the tests to avoid scanning .svn directories.
229 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
230 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
231 doesn't support NULs in patterns.
233 7. Added some missing "const"s to declarations of static tables in
234 pcre_compile.c and pcre_dfa_exec.c.
236 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
237 caused by fix #2 above. (Subsequently also a second patch to fix the
238 first patch. And a third patch - this was a messy problem.)
240 9. Applied Craig's patch to remove the use of push_back().
242 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
243 matching function regexec().
245 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
246 which, however, unlike Perl's \g{...}, are subroutine calls, not back
247 references. PCRE supports relative numbers with this syntax (I don't think
248 Oniguruma does).
250 12. Previously, a group with a zero repeat such as (...){0} was completely
251 omitted from the compiled regex. However, this means that if the group
252 was called as a subroutine from elsewhere in the pattern, things went wrong
253 (an internal error was given). Such groups are now left in the compiled
254 pattern, with a new opcode that causes them to be skipped at execution
255 time.
257 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
258 to the way PCRE behaves:
260 (a) A lone ] character is dis-allowed (Perl treats it as data).
262 (b) A back reference to an unmatched subpattern matches an empty string
263 (Perl fails the current match path).
265 (c) A data ] in a character class must be notated as \] because if the
266 first data character in a class is ], it defines an empty class. (In
267 Perl it is not possible to have an empty class.) The empty class []
268 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
269 The negative empty class [^] matches any one character, independently
270 of the DOTALL setting.
272 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
273 non-existent subpattern following a character class starting with ']' and
274 containing () gave an internal compiling error instead of "reference to
275 non-existent subpattern". Fortunately, when the pattern did exist, the
276 compiled code was correct. (When scanning forwards to check for the
277 existencd of the subpattern, it was treating the data ']' as terminating
278 the class, so got the count wrong. When actually compiling, the reference
279 was subsequently set up correctly.)
281 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
282 it was being rejected as not supported by pcre_dfa_exec(), even though
283 other assertions are supported. I have made pcre_dfa_exec() support
284 (*FAIL).
286 16. The implementation of 13c above involved the invention of a new opcode,
287 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
288 cannot be changed at match time, I realized I could make a small
289 improvement to matching performance by compiling OP_ALLANY instead of
290 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
291 on the OP_ANY path.
293 17. Compiling pcretest on Windows with readline support failed without the
294 following two fixes: (1) Make the unistd.h include conditional on
295 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
297 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
298 ncurses library to be included for pcretest when ReadLine support is
299 requested, but also to allow for it to be overridden. This patch came from
300 Daniel Bergström.
302 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
303 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
304 any errors with the current Unicode tables. Thanks to Peter Kankowski for
305 spotting this.
308 Version 7.6 28-Jan-08
309 ---------------------
311 1. A character class containing a very large number of characters with
312 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
313 overflow.
315 2. Patch to cut out the "long long" test in pcrecpp_unittest when
316 HAVE_LONG_LONG is not defined.
318 3. Applied Christian Ehrlicher's patch to update the CMake build files to
319 bring them up to date and include new features. This patch includes:
321 - Fixed PH's badly added libz and libbz2 support.
322 - Fixed a problem with static linking.
323 - Added pcredemo. [But later removed - see 7 below.]
324 - Fixed dftables problem and added an option.
325 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
327 - Added readline support for pcretest.
328 - Added an listing of the option settings after cmake has run.
330 4. A user submitted a patch to Makefile that makes it easy to create
331 "pcre.dll" under mingw when using Configure/Make. I added stuff to
332 Makefile.am that cause it to include this special target, without
333 affecting anything else. Note that the same mingw target plus all
334 the other distribution libraries and programs are now supported
335 when configuring with CMake (see 6 below) instead of with
336 Configure/Make.
338 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
339 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
340 exported in the Windows port". It has not yet been confirmed that the patch
341 solves the problem, but it does no harm.
343 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
344 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
345 with CMake, and also correct the comment about stack recursion.
347 7. Remove the automatic building of pcredemo from the ./configure system and
348 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
349 of a program that users should build themselves after PCRE is installed, so
350 building it automatically is not really right. What is more, it gave
351 trouble in some build environments.
353 8. Further tidies to CMakeLists.txt from Sheri and Christian.
356 Version 7.5 10-Jan-08
357 ---------------------
359 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
360 values in parens when parsing an RE using the C++ wrapper."
362 2. Negative specials like \S did not work in character classes in UTF-8 mode.
363 Characters greater than 255 were excluded from the class instead of being
364 included.
366 3. The same bug as (2) above applied to negated POSIX classes such as
367 [:^space:].
369 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
370 defined or documented. It seems to have been a typo for PCRE_STATIC, so
371 I have changed it.
373 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
374 first named subpattern) and a construct such as (?&a) would reference the
375 first named subpattern whose name started with "a" (in other words, the
376 length check was missing). Both these problems are fixed. "Subpattern name
377 expected" is now given for (?&) (a zero-length name), and this patch also
378 makes it give the same error for \k'' (previously it complained that that
379 was a reference to a non-existent subpattern).
381 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
382 this is right because (?- can be followed by option settings as well as by
383 digits. I have, however, made the messages clearer.
385 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
386 than the number used in the conditional) now cause a compile-time error.
387 This is actually not compatible with Perl, which accepts such patterns, but
388 treats the conditional as always being FALSE (as PCRE used to), but it
389 seems to me that giving a diagnostic is better.
391 8. Change "alphameric" to the more common word "alphanumeric" in comments
392 and messages.
394 9. Fix two occurrences of "backslash" in comments that should have been
395 "backspace".
397 10. Remove two redundant lines of code that can never be obeyed (their function
398 was moved elsewhere).
400 11. The program that makes PCRE's Unicode character property table had a bug
401 which caused it to generate incorrect table entries for sequences of
402 characters that have the same character type, but are in different scripts.
403 It amalgamated them into a single range, with the script of the first of
404 them. In other words, some characters were in the wrong script. There were
405 thirteen such cases, affecting characters in the following ranges:
407 U+002b0 - U+002c1
408 U+0060c - U+0060d
409 U+0061e - U+00612
410 U+0064b - U+0065e
411 U+0074d - U+0076d
412 U+01800 - U+01805
413 U+01d00 - U+01d77
414 U+01d9b - U+01dbf
415 U+0200b - U+0200f
416 U+030fc - U+030fe
417 U+03260 - U+0327f
418 U+0fb46 - U+0fbb1
419 U+10450 - U+1049d
421 12. The -o option (show only the matching part of a line) for pcregrep was not
422 compatible with GNU grep in that, if there was more than one match in a
423 line, it showed only the first of them. It now behaves in the same way as
424 GNU grep.
426 13. If the -o and -v options were combined for pcregrep, it printed a blank
427 line for every non-matching line. GNU grep prints nothing, and pcregrep now
428 does the same. The return code can be used to tell if there were any
429 non-matching lines.
431 14. Added --file-offsets and --line-offsets to pcregrep.
433 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
434 infinitely looping recursion. The bug was that positive lookaheads were not
435 being skipped when checking for a possible empty match (negative lookaheads
436 and both kinds of lookbehind were skipped).
438 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
439 inclusion of <windows.h> to before rather than after the definition of
440 INVALID_FILE_ATTRIBUTES (patch from David Byron).
442 17. Specifying a possessive quantifier with a specific limit for a Unicode
443 character property caused pcre_compile() to compile bad code, which led at
444 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
445 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
446 caused the error; without that there was no problem.
448 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
450 19. Added --enable-pcretest-libreadline.
452 20. In pcrecpp.cc, the variable 'count' was incremented twice in
453 RE::GlobalReplace(). As a result, the number of replacements returned was
454 double what it should be. I removed one of the increments, but Craig sent a
455 later patch that removed the other one (the right fix) and added unit tests
456 that check the return values (which was not done before).
458 21. Several CMake things:
460 (1) Arranged that, when cmake is used on Unix, the libraries end up with
461 the names libpcre and libpcreposix, not just pcre and pcreposix.
463 (2) The above change means that pcretest and pcregrep are now correctly
464 linked with the newly-built libraries, not previously installed ones.
468 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
469 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
470 UTF-8 newline character). The key issue is that the pattern starts .*;
471 this means that the match must be either at the beginning, or after a
472 newline. The bug was in the code for advancing after a failed match and
473 checking that the new position followed a newline. It was not taking
474 account of UTF-8 characters correctly.
476 23. PCRE was behaving differently from Perl in the way it recognized POSIX
477 character classes. PCRE was not treating the sequence [:...:] as a
478 character class unless the ... were all letters. Perl, however, seems to
479 allow any characters between [: and :], though of course it rejects as
480 unknown any "names" that contain non-letters, because all the known class
481 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
482 for example, whereas PCRE did not - it did not recognize a POSIX character
483 class. This seemed a bit dangerous, so the code has been changed to be
484 closer to Perl. The behaviour is not identical to Perl, because PCRE will
485 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
486 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
487 Perl does, and where it didn't before.
489 24. Rewrite so as to remove the single use of %n from pcregrep because in some
490 Windows environments %n is disabled by default.
493 Version 7.4 21-Sep-07
494 ---------------------
496 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
497 means that a class such as [\s] counted as "explicit reference to CR or
498 LF". That isn't really right - the whole point of the change was to try to
499 help when there was an actual mention of one of the two characters. So now
500 the change happens only if \r or \n (or a literal CR or LF) character is
501 encountered.
503 2. The 32-bit options word was also used for 6 internal flags, but the numbers
504 of both had grown to the point where there were only 3 bits left.
505 Fortunately, there was spare space in the data structure, and so I have
506 moved the internal flags into a new 16-bit field to free up more option
507 bits.
509 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
510 but did not set the internal JCHANGED flag - either of these is enough to
511 control the way the "get" function works - but the PCRE_INFO_JCHANGED
512 facility is supposed to tell if (?J) was ever used, so now (?J) at the
513 start sets both bits.
515 4. Added options (at build time, compile time, exec time) to change \R from
516 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
518 5. doc/pcresyntax.html was missing from the distribution.
520 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
521 compatibility, even though it is no longer used.
523 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
524 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
525 windows.h file is present (where different names are used). [This was
526 reversed later after testing - see 16 below.]
528 8. Changed all #include <config.h> to #include "config.h". There were also
529 some further <pcre.h> cases that I changed to "pcre.h".
531 9. When pcregrep was used with the --colour option, it missed the line ending
532 sequence off the lines that it output.
534 10. It was pointed out to me that arrays of string pointers cause lots of
535 relocations when a shared library is dynamically loaded. A technique of
536 using a single long string with a table of offsets can drastically reduce
537 these. I have refactored PCRE in four places to do this. The result is
538 dramatic:
540 Originally: 290
541 After changing UCP table: 187
542 After changing error message table: 43
543 After changing table of "verbs" 36
544 After changing table of Posix names 22
546 Thanks to the folks working on Gregex for glib for this insight.
548 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
549 unicode-properties was also set.
551 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
553 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
554 checked only for CRLF.
556 14. Added casts to pcretest.c to avoid compiler warnings.
558 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
560 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
561 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
562 entirely. This removes changes made in 7 above.
564 17. The CMake files have been updated, and there is now more information about
565 building with CMake in the NON-UNIX-USE document.
568 Version 7.3 28-Aug-07
569 ---------------------
571 1. In the rejigging of the build system that eventually resulted in 7.1, the
572 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
573 brackets there is not right, since it causes compilers to look for an
574 installed pcre.h, not the version that is in the source that is being
575 compiled (which of course may be different). I have changed it back to:
577 #include "pcre.h"
579 I have a vague recollection that the change was concerned with compiling in
580 different directories, but in the new build system, that is taken care of
581 by the VPATH setting the Makefile.
583 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
584 when the subject happened to end in the byte 0x85 (e.g. if the last
585 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
586 characters but of course it shouldn't be taken as a newline when it is part
587 of another character. The bug was that, for an unlimited repeat of . in
588 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
589 characters when looking for a newline.
591 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
593 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
594 in debug output.
596 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
597 long printing in the pcrecpp unittest when running under MinGW.
599 6. ESC_K was left out of the EBCDIC table.
601 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
602 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
603 limit also applies to "virtual nesting" when a pattern is recursive, and in
604 this case 1000 isn't so big. I have been able to remove this limit at the
605 expense of backing off one optimization in certain circumstances. Normally,
606 when pcre_exec() would call its internal match() function recursively and
607 immediately return the result unconditionally, it uses a "tail recursion"
608 feature to save stack. However, when a subpattern that can match an empty
609 string has an unlimited repetition quantifier, it no longer makes this
610 optimization. That gives it a stack frame in which to save the data for
611 checking that an empty string has been matched. Previously this was taken
612 from the 1000-entry workspace that had been reserved. So now there is no
613 explicit limit, but more stack is used.
615 8. Applied Daniel's patches to solve problems with the import/export magic
616 syntax that is required for Windows, and which was going wrong for the
617 pcreposix and pcrecpp parts of the library. These were overlooked when this
618 problem was solved for the main library.
620 9. There were some crude static tests to avoid integer overflow when computing
621 the size of patterns that contain repeated groups with explicit upper
622 limits. As the maximum quantifier is 65535, the maximum group length was
623 set at 30,000 so that the product of these two numbers did not overflow a
624 32-bit integer. However, it turns out that people want to use groups that
625 are longer than 30,000 bytes (though not repeat them that many times).
626 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
627 made it possible to implement the integer overflow checks in a much more
628 dynamic way, which I have now done. The artificial limitation on group
629 length has been removed - we now have only the limit on the total length of
630 the compiled pattern, which depends on the LINK_SIZE setting.
632 10. Fixed a bug in the documentation for get/copy named substring when
633 duplicate names are permitted. If none of the named substrings are set, the
634 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
635 empty string.
637 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
638 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
639 because the ] is interpreted as the first data character and the
640 terminating ] is not found. PCRE has been made compatible with Perl in this
641 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
642 cause memory overwriting.
644 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
645 string has been matched (to stop an infinite loop). It was not recognizing
646 a conditional subpattern that could match an empty string if that
647 subpattern was within another subpattern. For example, it looped when
648 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
649 condition was not nested. This bug has been fixed.
651 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
652 past the start of the subject in the presence of bytes with the top bit
653 set, for example "\x8aBCD".
655 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
656 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
658 14. Optimized (?!) to (*FAIL).
660 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
661 This restricts code points to be within the range 0 to 0x10FFFF, excluding
662 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
663 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
664 does: it's just the validity check that is more restrictive.
666 16. Inserted checks for integer overflows during escape sequence (backslash)
667 processing, and also fixed erroneous offset values for syntax errors during
668 backslash processing.
670 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
671 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
673 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
674 caused an overrun.
676 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
677 something other than just ASCII characters) inside a group that had an
678 unlimited repeat caused a loop at compile time (while checking to see
679 whether the group could match an empty string).
681 20. Debugging a pattern containing \p or \P could cause a crash. For example,
682 [\P{Any}] did so. (Error in the code for printing property names.)
684 21. An orphan \E inside a character class could cause a crash.
686 22. A repeated capturing bracket such as (A)? could cause a wild memory
687 reference during compilation.
689 23. There are several functions in pcre_compile() that scan along a compiled
690 expression for various reasons (e.g. to see if it's fixed length for look
691 behind). There were bugs in these functions when a repeated \p or \P was
692 present in the pattern. These operators have additional parameters compared
693 with \d, etc, and these were not being taken into account when moving along
694 the compiled data. Specifically:
696 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
697 length.
699 (b) An item such as \pL+ within a repeated group could cause crashes or
700 loops.
702 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
703 "reference to non-existent subpattern" error.
705 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
707 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
708 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
710 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
712 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
713 character were causing crashes (broken optimization).
715 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
716 \p or \P) caused a compile-time loop.
718 28. More problems have arisen in unanchored patterns when CRLF is a valid line
719 break. For example, the unstudied pattern [\r\n]A does not match the string
720 "\r\nA" because change 7.0/46 below moves the current point on by two
721 characters after failing to match at the start. However, the pattern \nA
722 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
723 the same is true. There doesn't seem any very clean way out of this, but
724 what I have chosen to do makes the common cases work: PCRE now takes note
725 of whether there can be an explicit match for \r or \n anywhere in the
726 pattern, and if so, 7.0/46 no longer applies. As part of this change,
727 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
728 pattern has explicit CR or LF references.
730 29. Added (*CR) etc for changing newline setting at start of pattern.
733 Version 7.2 19-Jun-07
734 ---------------------
736 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
737 which is apparently normally available under Windows.
739 2. Re-jig the pcregrep tests with different newline settings in an attempt
740 to make them independent of the local environment's newline setting.
742 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
744 4. Some of the "internals" tests were previously cut out when the link size
745 was not 2, because the output contained actual offsets. The recent new
746 "Z" feature of pcretest means that these can be cut out, making the tests
747 usable with all link sizes.
749 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
750 stack recursion. This gives a massive performance boost under BSD, but just
751 a small improvement under Linux. However, it saves one field in the frame
752 in all cases.
754 6. Added more features from the forthcoming Perl 5.10:
756 (a) (?-n) (where n is a string of digits) is a relative subroutine or
757 recursion call. It refers to the nth most recently opened parentheses.
759 (b) (?+n) is also a relative subroutine call; it refers to the nth next
760 to be opened parentheses.
762 (c) Conditions that refer to capturing parentheses can be specified
763 relatively, for example, (?(-2)... or (?(+3)...
765 (d) \K resets the start of the current match so that everything before
766 is not part of it.
768 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
770 (f) \g{name} is another synonym - part of Perl 5.10's unification of
771 reference syntax.
773 (g) (?| introduces a group in which the numbering of parentheses in each
774 alternative starts with the same number.
776 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
778 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
781 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
782 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
783 for detecting groups that can match an empty string.
785 9. A pattern with a very large number of alternatives (more than several
786 hundred) was running out of internal workspace during the pre-compile
787 phase, where pcre_compile() figures out how much memory will be needed. A
788 bit of new cunning has reduced the workspace needed for groups with
789 alternatives. The 1000-alternative test pattern now uses 12 bytes of
790 workspace instead of running out of the 4096 that are available.
792 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
794 11. Applied patch from Google to remove an optimization that didn't quite work.
795 The report of the bug said:
797 pcrecpp::RE("a*").FullMatch("aaa") matches, while
798 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
799 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
801 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
802 it matched the wrong number of bytes.
805 Version 7.1 24-Apr-07
806 ---------------------
808 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
809 that is more "standard", making use of automake and other Autotools. There
810 is some re-arrangement of the files and adjustment of comments consequent
811 on this.
813 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
814 for recursive directory scanning broke on some systems because the files
815 are not scanned in any specific order and on different systems the order
816 was different. A call to "sort" has been inserted into RunGrepTest for the
817 approprate test as a short-term fix. In the longer term there may be an
818 alternative.
820 3. I had an email from Eric Raymond about problems translating some of PCRE's
821 man pages to HTML (despite the fact that I distribute HTML pages, some
822 people do their own conversions for various reasons). The problems
823 concerned the use of low-level troff macros .br and .in. I have therefore
824 removed all such uses from the man pages (some were redundant, some could
825 be replaced by .nf/.fi pairs). The 132html script that I use to generate
826 HTML has been updated to handle .nf/.fi and to complain if it encounters
827 .br or .in.
829 4. Updated comments in configure.ac that get placed in config.h.in and also
830 arranged for config.h to be included in the distribution, with the name
831 config.h.generic, for the benefit of those who have to compile without
832 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
834 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
835 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
836 makevp.bat for the current PCRE, using the additional files
837 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
839 6. A Windows user reported a minor discrepancy with test 2, which turned out
840 to be caused by a trailing space on an input line that had got lost in his
841 copy. The trailing space was an accident, so I've just removed it.
843 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
844 that is needed.
846 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
847 as "const" (a) because they are and (b) because it helps the PHP
848 maintainers who have recently made a script to detect big data structures
849 in the php code that should be moved to the .rodata section. I remembered
850 to update Builducptable as well, so it won't revert if ucptable.h is ever
851 re-created.
853 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
854 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
855 order to be able to cut out the UTF-8 tables in the latter when UTF-8
856 support is not required. This saves 1.5-2K of code, which is important in
857 some applications.
859 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
860 so as not to refer to the tables, even though these functions will never be
861 called when UTF-8 support is disabled. Otherwise there are problems with a
862 shared library.
864 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
866 (a) It was defining its arguments as char * instead of void *.
868 (b) It was assuming that all moves were upwards in memory; this was true
869 a long time ago when I wrote it, but is no longer the case.
871 The emulated memove() is provided for those environments that have neither
872 memmove() nor bcopy(). I didn't think anyone used it these days, but that
873 is clearly not the case, as these two bugs were recently reported.
875 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
876 and Detrail to create the HTML documentation, the .txt form of the man
877 pages, and it removes trailing spaces from listed files. It also creates
878 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
879 case, it wraps all the #defines with #ifndefs. This script should be run
880 before "make dist".
882 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
883 with Unicode property support.
885 (a) For a maximizing quantifier, if the two different cases of the
886 character were of different lengths in their UTF-8 codings (there are
887 some cases like this - I found 11), and the matching function had to
888 back up over a mixture of the two cases, it incorrectly assumed they
889 were both the same length.
891 (b) When PCRE was configured to use the heap rather than the stack for
892 recursion during matching, it was not correctly preserving the data for
893 the other case of a UTF-8 character when checking ahead for a match
894 while processing a minimizing repeat. If the check also involved
895 matching a wide character, but failed, corruption could cause an
896 erroneous result when trying to check for a repeat of the original
897 character.
899 13. Some tidying changes to the testing mechanism:
901 (a) The RunTest script now detects the internal link size and whether there
902 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
903 values substituted by "configure". (The RunGrepTest script already did
904 this for UTF-8.) The configure.ac script no longer substitutes the
905 relevant variables.
907 (b) The debugging options /B and /D in pcretest show the compiled bytecode
908 with length and offset values. This means that the output is different
909 for different internal link sizes. Test 2 is skipped for link sizes
910 other than 2 because of this, bypassing the problem. Unfortunately,
911 there was also a test in test 3 (the locale tests) that used /B and
912 failed for link sizes other than 2. Rather than cut the whole test out,
913 I have added a new /Z option to pcretest that replaces the length and
914 offset values with spaces. This is now used to make test 3 independent
915 of link size. (Test 2 will be tidied up later.)
917 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
918 segmentation fault instead of returning the appropriate error message.
920 15. In multiline mode when the newline sequence was set to "any", the pattern
921 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
922 This doesn't seem right; it now treats the CRLF combination as the line
923 ending, and so does not match in that case. It's only a pattern such as ^$
924 that would hit this one: something like ^ABC$ would have failed after \r
925 and then tried again after \r\n.
927 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
928 in an attempt to make files that differ only in their line terminators
929 compare equal. This works on Linux.
931 17. Under certain error circumstances pcregrep might try to free random memory
932 as it exited. This is now fixed, thanks to valgrind.
934 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
935 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
936 was because its rules for how to advance for /g after matching an empty
937 string at the end of a line did not allow for this case. They now check for
938 it specially.
940 20. pcretest is supposed to handle patterns and data of any length, by
941 extending its buffers when necessary. It was getting this wrong when the
942 buffer for a data line had to be extended.
944 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
945 CRLF as a newline sequence.
947 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
948 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
949 I have nevertheless tidied it up.
951 23. Added some casts to kill warnings from HP-UX ia64 compiler.
953 24. Added a man page for pcre-config.
956 Version 7.0 19-Dec-06
957 ---------------------
959 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
960 moving to gcc 4.1.1.
962 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
963 sys/time.h, which is documented as needed for this function. It doesn't
964 seem to matter on Linux, but it showed up on some releases of OS X.
966 3. It seems that there are systems where bytes whose values are greater than
967 127 match isprint() in the "C" locale. The "C" locale should be the
968 default when a C program starts up. In most systems, only ASCII printing
969 characters match isprint(). This difference caused the output from pcretest
970 to vary, making some of the tests fail. I have changed pcretest so that:
972 (a) When it is outputting text in the compiled version of a pattern, bytes
973 other than 32-126 are always shown as hex escapes.
975 (b) When it is outputting text that is a matched part of a subject string,
976 it does the same, unless a different locale has been set for the match
977 (using the /L modifier). In this case, it uses isprint() to decide.
979 4. Fixed a major bug that caused incorrect computation of the amount of memory
980 required for a compiled pattern when options that changed within the
981 pattern affected the logic of the preliminary scan that determines the
982 length. The relevant options are -x, and -i in UTF-8 mode. The result was
983 that the computed length was too small. The symptoms of this bug were
984 either the PCRE error "internal error: code overflow" from pcre_compile(),
985 or a glibc crash with a message such as "pcretest: free(): invalid next
986 size (fast)". Examples of patterns that provoked this bug (shown in
987 pcretest format) are:
989 /(?-x: )/x
990 /(?x)(?-x: \s*#\s*)/
991 /((?i)[\x{c0}])/8
992 /(?i:[\x{c0}])/8
994 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
995 is now done differently.
997 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
998 wrapper classes; (b) implement a new function in the C++ scanner that is
999 more efficient than the old way of doing things because it avoids levels of
1000 recursion in the regex matching; (c) add a paragraph to the documentation
1001 for the FullMatch() function.
1003 6. The escape sequence \n was being treated as whatever was defined as
1004 "newline". Not only was this contrary to the documentation, which states
1005 that \n is character 10 (hex 0A), but it also went horribly wrong when
1006 "newline" was defined as CRLF. This has been fixed.
1008 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
1009 was being set to -1 for the "end of line" case (supposedly a value that no
1010 character can have). Though this value is never used (the check for end of
1011 line is "zero bytes in current character"), it caused compiler complaints.
1012 I've changed it to 0xffffffff.
1014 8. In pcre_version.c, the version string was being built by a sequence of
1015 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
1016 string (as it is for production releases) called a macro with an empty
1017 argument. The C standard says the result of this is undefined. The gcc
1018 compiler treats it as an empty string (which was what was wanted) but it is
1019 reported that Visual C gives an error. The source has been hacked around to
1020 avoid this problem.
1022 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1023 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1024 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1025 of them did).
1027 10. Originally, pcretest opened its input and output without "b"; then I was
1028 told that "b" was needed in some environments, so it was added for release
1029 5.0 to both the input and output. (It makes no difference on Unix-like
1030 systems.) Later I was told that it is wrong for the input on Windows. I've
1031 now abstracted the modes into two macros, to make it easier to fiddle with
1032 them, and removed "b" from the input mode under Windows.
1034 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1036 12. Added -help and --help to pcretest as an official way of being reminded
1037 of the options.
1039 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1040 and pcrecpp.cc because they annoy compilers at high warning levels.
1042 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1044 15. Fixed an occurrence of == in configure.ac that should have been = (shell
1045 scripts are not C programs :-) and which was not noticed because it works
1046 on Linux.
1048 16. pcretest is supposed to handle any length of pattern and data line (as one
1049 line or as a continued sequence of lines) by extending its input buffer if
1050 necessary. This feature was broken for very long pattern lines, leading to
1051 a string of junk being passed to pcre_compile() if the pattern was longer
1052 than about 50K.
1054 17. I have done a major re-factoring of the way pcre_compile() computes the
1055 amount of memory needed for a compiled pattern. Previously, there was code
1056 that made a preliminary scan of the pattern in order to do this. That was
1057 OK when PCRE was new, but as the facilities have expanded, it has become
1058 harder and harder to keep it in step with the real compile phase, and there
1059 have been a number of bugs (see for example, 4 above). I have now found a
1060 cunning way of running the real compile function in a "fake" mode that
1061 enables it to compute how much memory it would need, while actually only
1062 ever using a few hundred bytes of working memory and without too many
1063 tests of the mode. This should make future maintenance and development
1064 easier. A side effect of this work is that the limit of 200 on the nesting
1065 depth of parentheses has been removed (though this was never a serious
1066 limitation, I suspect). However, there is a downside: pcre_compile() now
1067 runs more slowly than before (30% or more, depending on the pattern). I
1068 hope this isn't a big issue. There is no effect on runtime performance.
1070 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1071 newline (only possible for the last line of a file) and it was a
1072 pattern that set a locale (followed by /Lsomething), pcretest crashed.
1074 19. Added additional timing features to pcretest. (1) The -tm option now times
1075 matching only, not compiling. (2) Both -t and -tm can be followed, as a
1076 separate command line item, by a number that specifies the number of
1077 repeats to use when timing. The default is 50000; this gives better
1078 precision, but takes uncomfortably long for very large patterns.
1080 20. Extended pcre_study() to be more clever in cases where a branch of a
1081 subpattern has no definite first character. For example, (a*|b*)[cd] would
1082 previously give no result from pcre_study(). Now it recognizes that the
1083 first character must be a, b, c, or d.
1085 21. There was an incorrect error "recursive call could loop indefinitely" if
1086 a subpattern (or the entire pattern) that was being tested for matching an
1087 empty string contained only one non-empty item after a nested subpattern.
1088 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1089 incorrectly, because the \d was being skipped in the check.
1091 22. The pcretest program now has a new pattern option /B and a command line
1092 option -b, which is equivalent to adding /B to every pattern. This causes
1093 it to show the compiled bytecode, without the additional information that
1094 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1095 is the same as /B/I).
1097 23. A new optimization is now able automatically to treat some sequences such
1098 as a*b as a*+b. More specifically, if something simple (such as a character
1099 or a simple class like \d) has an unlimited quantifier, and is followed by
1100 something that cannot possibly match the quantified thing, the quantifier
1101 is automatically "possessified".
1103 24. A recursive reference to a subpattern whose number was greater than 39
1104 went wrong under certain circumstances in UTF-8 mode. This bug could also
1105 have affected the operation of pcre_study().
1107 25. Realized that a little bit of performance could be had by replacing
1108 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1110 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1112 27. Possessive quantifiers such as a++ were previously implemented by turning
1113 them into atomic groups such as ($>a+). Now they have their own opcodes,
1114 which improves performance. This includes the automatically created ones
1115 from 23 above.
1117 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1118 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1119 the first matched character to be a colon. This applied both to named and
1120 numbered groups.
1122 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1124 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1125 building PCRE on Windows easier, so I have included it in the distribution.
1127 31. There is now a check in pcretest against a ridiculously large number being
1128 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1129 loop, the loop is abandoned.
1131 32. Forward references to subpatterns in conditions such as (?(2)...) where
1132 subpattern 2 is defined later cause pcre_compile() to search forwards in
1133 the pattern for the relevant set of parentheses. This search went wrong
1134 when there were unescaped parentheses in a character class, parentheses
1135 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1137 33. "Subroutine" calls and backreferences were previously restricted to
1138 referencing subpatterns earlier in the regex. This restriction has now
1139 been removed.
1141 34. Added a number of extra features that are going to be in Perl 5.10. On the
1142 whole, these are just syntactic alternatives for features that PCRE had
1143 previously implemented using the Python syntax or my own invention. The
1144 other formats are all retained for compatibility.
1146 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1147 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1148 also .NET compatible.
1150 (b) A recursion or subroutine call to a named group can now be defined as
1151 (?&name) as well as (?P>name).
1153 (c) A backreference to a named group can now be defined as \k<name> or
1154 \k'name' as well as (?P=name). The new forms, as well as being in Perl
1155 5.10, are also .NET compatible.
1157 (d) A conditional reference to a named group can now use the syntax
1158 (?(<name>) or (?('name') as well as (?(name).
1160 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1161 groups (named and numbered) that are never evaluated inline, but can be
1162 called as "subroutines" from elsewhere. In effect, the DEFINE condition
1163 is always false. There may be only one alternative in such a group.
1165 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1166 as the simple (?(R). The condition is true only if the most recent
1167 recursion is that of the given number or name. It does not search out
1168 through the entire recursion stack.
1170 (g) The escape \gN or \g{N} has been added, where N is a positive or
1171 negative number, specifying an absolute or relative reference.
1173 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1174 some "unreachable code" warnings.
1176 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1177 things, this adds five new scripts.
1179 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1180 There were also incompatibilities regarding the handling of \Q..\E inside
1181 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1182 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1184 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1185 matches an empty string, and forcibly breaks the loop. There were bugs in
1186 this code in non-simple cases. For a pattern such as ^(a()*)* matched
1187 against aaaa the result was just "a" rather than "aaaa", for example. Two
1188 separate and independent bugs (that affected different cases) have been
1189 fixed.
1191 39. Refactored the code to abolish the use of different opcodes for small
1192 capturing bracket numbers. This is a tidy that I avoided doing when I
1193 removed the limit on the number of capturing brackets for 3.5 back in 2001.
1194 The new approach is not only tidier, it makes it possible to reduce the
1195 memory needed to fix the previous bug (38).
1197 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1198 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1199 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1200 mode.
1202 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1203 report.
1205 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1206 copy construction and assignment in the C++ wrapper.
1208 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1209 couple of bugs that could have given wrong results in the "--newline=crlf"
1210 case.
1212 44. Added a number of casts and did some reorganization of signed/unsigned int
1213 variables following suggestions from Dair Grant. Also renamed the variable
1214 "this" as "item" because it is a C++ keyword.
1216 45. Arranged for dftables to add
1218 #include "pcre_internal.h"
1220 to pcre_chartables.c because without it, gcc 4.x may remove the array
1221 definition from the final binary if PCRE is built into a static library and
1222 dead code stripping is activated.
1224 46. For an unanchored pattern, if a match attempt fails at the start of a
1225 newline sequence, and the newline setting is CRLF or ANY, and the next two
1226 characters are CRLF, advance by two characters instead of one.
1229 Version 6.7 04-Jul-06
1230 ---------------------
1232 1. In order to handle tests when input lines are enormously long, pcretest has
1233 been re-factored so that it automatically extends its buffers when
1234 necessary. The code is crude, but this _is_ just a test program. The
1235 default size has been increased from 32K to 50K.
1237 2. The code in pcre_study() was using the value of the re argument before
1238 testing it for NULL. (Of course, in any sensible call of the function, it
1239 won't be NULL.)
1241 3. The memmove() emulation function in pcre_internal.h, which is used on
1242 systems that lack both memmove() and bcopy() - that is, hardly ever -
1243 was missing a "static" storage class specifier.
1245 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1246 containing an extended class (one that cannot be represented by a bitmap
1247 because it contains high-valued characters or Unicode property items, e.g.
1248 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1249 pattern, but PCRE should not loop if you do not (it no longer does).
1250 [Detail: two cases were found: (a) a repeated subpattern containing an
1251 extended class; (b) a recursive reference to a subpattern that followed a
1252 previous extended class. It wasn't skipping over the extended class
1253 correctly when UTF-8 mode was not set.]
1255 5. A negated single-character class was not being recognized as fixed-length
1256 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1257 compile error "lookbehind assertion is not fixed length".
1259 6. The RunPerlTest auxiliary script was showing an unexpected difference
1260 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1261 write a Perl script that can interpret lines of an input file either as
1262 byte characters or as UTF-8, which is what "perltest" was being required to
1263 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1264 can't do is switch easily at run time between having the "use utf8;" pragma
1265 or not. In the end, I fudged it by using the RunPerlTest script to insert
1266 "use utf8;" explicitly for the UTF-8 tests.
1268 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1269 the end of the subject string, contrary to the documentation and to what
1270 Perl does. This was true of both matching functions. Now it matches only at
1271 the start of the subject and immediately after *internal* newlines.
1273 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1274 a pointer to an int instead of a pointer to an unsigned long int. This
1275 caused problems on 64-bit systems.
1277 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1278 instance of the 'standard' template library not being so standard".
1280 10. There was no check on the number of named subpatterns nor the maximum
1281 length of a subpattern name. The product of these values is used to compute
1282 the size of the memory block for a compiled pattern. By supplying a very
1283 long subpattern name and a large number of named subpatterns, the size
1284 computation could be caused to overflow. This is now prevented by limiting
1285 the length of names to 32 characters, and the number of named subpatterns
1286 to 10,000.
1288 11. Subpatterns that are repeated with specific counts have to be replicated in
1289 the compiled pattern. The size of memory for this was computed from the
1290 length of the subpattern and the repeat count. The latter is limited to
1291 65535, but there was no limit on the former, meaning that integer overflow
1292 could in principle occur. The compiled length of a repeated subpattern is
1293 now limited to 30,000 bytes in order to prevent this.
1295 12. Added the optional facility to have named substrings with the same name.
1297 13. Added the ability to use a named substring as a condition, using the
1298 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1299 are numbers (not recommended). Forward references are permitted.
1301 14. Added forward references in named backreferences (if you see what I mean).
1303 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1304 pattern could run off the end of the subject. For example, the pattern
1305 "(?s)(.{1,5})"8 did this with the subject "ab".
1307 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1308 PCRE_CASELESS was set when matching characters that were quantified with ?
1309 or *.
1311 17. A character class other than a single negated character that had a minimum
1312 but no maximum quantifier - for example [ab]{6,} - was not handled
1313 correctly by pce_dfa_exec(). It would match only one character.
1315 18. A valid (though odd) pattern that looked like a POSIX character
1316 class but used an invalid character after [ (for example [[,abc,]]) caused
1317 pcre_compile() to give the error "Failed: internal error: code overflow" or
1318 in some cases to crash with a glibc free() error. This could even happen if
1319 the pattern terminated after [[ but there just happened to be a sequence of
1320 letters, a binary zero, and a closing ] in the memory that followed.
1322 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1323 over the years. Originally (before any Unicode support), just the bottom 8
1324 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1325 output from "man perlunicode" includes this:
1327 The regular expression compiler produces polymorphic opcodes. That
1328 is, the pattern adapts to the data and automatically switches to
1329 the Unicode character scheme when presented with Unicode data--or
1330 instead uses a traditional byte scheme when presented with byte
1331 data.
1333 Sadly, a wide octal escape does not cause a switch, and in a string with
1334 no other multibyte characters, these octal escapes are treated as before.
1335 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1336 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1337 Unicode string.
1339 I have not perpetrated such confusion in PCRE. Up till now, it took just
1340 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1341 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1342 translate to the appropriate multibyte character.
1344 29. Applied some refactoring to reduce the number of warnings from Microsoft
1345 and Borland compilers. This has included removing the fudge introduced
1346 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1347 a warning about an unused variable.
1349 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1350 characters since release 4.0, because Perl (from release 5.004) does not.
1351 [Or at least, is documented not to: some releases seem to be in conflict
1352 with the documentation.] However, when a pattern was studied with
1353 pcre_study() and all its branches started with \s, PCRE still included VT
1354 as a possible starting character. Of course, this did no harm; it just
1355 caused an unnecessary match attempt.
1357 22. Removed a now-redundant internal flag bit that recorded the fact that case
1358 dependency changed within the pattern. This was once needed for "required
1359 byte" processing, but is no longer used. This recovers a now-scarce options
1360 bit. Also moved the least significant internal flag bit to the most-
1361 significant bit of the word, which was not previously used (hangover from
1362 the days when it was an int rather than a uint) to free up another bit for
1363 the future.
1365 23. Added support for CRLF line endings as well as CR and LF. As well as the
1366 default being selectable at build time, it can now be changed at runtime
1367 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1368 specify that it is scanning data with non-default line endings.
1370 24. Changed the definition of CXXLINK to make it agree with the definition of
1371 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1373 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1374 recursions. This makes a big different to stack usage for some patterns.
1376 26. If a subpattern containing a named recursion or subroutine reference such
1377 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1378 the space required for the compiled pattern went wrong and gave too small a
1379 value. Depending on the environment, this could lead to "Failed: internal
1380 error: code overflow at offset 49" or "glibc detected double free or
1381 corruption" errors.
1383 27. Applied patches from Google (a) to support the new newline modes and (b) to
1384 advance over multibyte UTF-8 characters in GlobalReplace.
1386 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1387 difference for some implementation of PCRE in some Windows version.
1389 29. Added some extra testing facilities to pcretest:
1391 \q<number> in a data line sets the "match limit" value
1392 \Q<number> in a data line sets the "match recursion limt" value
1393 -S <number> sets the stack size, where <number> is in megabytes
1395 The -S option isn't available for Windows.
1398 Version 6.6 06-Feb-06
1399 ---------------------
1401 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1402 in pcreposix.h. I have copied the definition from pcre.h.
1404 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1405 because pcre.h is no longer a built file.
1407 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1408 not normally included in the compiled code.
1411 Version 6.5 01-Feb-06
1412 ---------------------
1414 1. When using the partial match feature with pcre_dfa_exec(), it was not
1415 anchoring the second and subsequent partial matches at the new starting
1416 point. This could lead to incorrect results. For example, with the pattern
1417 /1234/, partially matching against "123" and then "a4" gave a match.
1419 2. Changes to pcregrep:
1421 (a) All non-match returns from pcre_exec() were being treated as failures
1422 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1423 error message is output. Some extra information is given for the
1425 probably the only errors that are likely to be caused by users (by
1426 specifying a regex that has nested indefinite repeats, for instance).
1427 If there are more than 20 of these errors, pcregrep is abandoned.
1429 (b) A binary zero was treated as data while matching, but terminated the
1430 output line if it was written out. This has been fixed: binary zeroes
1431 are now no different to any other data bytes.
1433 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1434 used to set a locale for matching. The --locale=xxxx long option has
1435 been added (no short equivalent) to specify a locale explicitly on the
1436 pcregrep command, overriding the environment variables.
1438 (d) When -B was used with -n, some line numbers in the output were one less
1439 than they should have been.
1441 (e) Added the -o (--only-matching) option.
1443 (f) If -A or -C was used with -c (count only), some lines of context were
1444 accidentally printed for the final match.
1446 (g) Added the -H (--with-filename) option.
1448 (h) The combination of options -rh failed to suppress file names for files
1449 that were found from directory arguments.
1451 (i) Added the -D (--devices) and -d (--directories) options.
1453 (j) Added the -F (--fixed-strings) option.
1455 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1457 (l) Added the --colo(u)r option.
1459 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1460 is not present by default.
1462 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1463 items such as (?R) or (?1), when the recursion could match a number of
1464 alternatives. If it matched one of the alternatives, but subsequently,
1465 outside the recursion, there was a failure, the code tried to back up into
1466 the recursion. However, because of the way PCRE is implemented, this is not
1467 possible, and the result was an incorrect result from the match.
1469 In order to prevent this happening, the specification of recursion has
1470 been changed so that all such subpatterns are automatically treated as
1471 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1473 4. I had overlooked the fact that, in some locales, there are characters for
1474 which isalpha() is true but neither isupper() nor islower() are true. In
1475 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1476 and ordfeminine) are like this. This affected the treatment of \w and \W
1477 when they appeared in character classes, but not when they appeared outside
1478 a character class. The bit map for "word" characters is now created
1479 separately from the results of isalnum() instead of just taking it from the
1480 upper, lower, and digit maps. (Plus the underscore character, of course.)
1482 5. The above bug also affected the handling of POSIX character classes such as
1483 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1484 permanent tables. Instead, the bit maps for such a class were previously
1485 created as the appropriate unions of the upper, lower, and digit bitmaps.
1486 Now they are created by subtraction from the [[:word:]] class, which has
1487 its own bitmap.
1489 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1490 It is created by subtracting the vertical space characters (\x09, \x0a,
1491 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1492 subtraction was done in the overall bitmap for a character class, meaning
1493 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1494 be recognized. This bug has been fixed.
1496 7. Patches from the folks at Google:
1498 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1499 real life, but is still worth protecting against".
1501 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1502 regular expressions".
1504 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1505 have it.
1507 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1508 "configure" and the latter not, in order to fix a problem somebody had
1509 with compiling the Arg class on HP-UX.
1511 (e) Improve the error-handling of the C++ wrapper a little bit.
1513 (f) New tests for checking recursion limiting.
1515 8. The pcre_memmove() function, which is used only if the environment does not
1516 have a standard memmove() function (and is therefore rarely compiled),
1517 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1518 returning a result (though PCRE never actually uses the result).
1520 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1521 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1522 returned instead of calling malloc() with an overflowing number that would
1523 most likely cause subsequent chaos.
1525 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1527 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1528 with this option is matched, the nmatch and pmatch options of regexec() are
1529 ignored.
1531 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1532 provided in case anyone wants to the the POSIX interface with UTF-8
1533 strings.
1535 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1536 C++ linking (needed for some HP-UX environments).
1538 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1539 (unused parameter) and in the pcre_printint() function (omitted "default"
1540 switch label when the default is to do nothing).
1542 15. Added some code to make it possible, when PCRE is compiled as a C++
1543 library, to replace subject pointers for pcre_exec() with a smart pointer
1544 class, thus making it possible to process discontinuous strings.
1546 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1547 much the same function. They were added by different people who were trying
1548 to make PCRE easy to compile on non-Unix systems. It has been suggested
1549 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1550 for compiling on Windows systems. I have therefore replaced it with
1551 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1552 defaults to "extern" for C or "extern C" for C++, which works fine on
1553 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1554 SCOPE with something explicit in config.h. In addition:
1556 (a) pcreposix.h still had just "extern" instead of either of these macros;
1557 I have replaced it with PCRE_DATA_SCOPE.
1559 (b) Functions such as _pcre_xclass(), which are internal to the library,
1560 but external in the C sense, all had PCRE_EXPORT in their definitions.
1561 This is apparently wrong for the Windows case, so I have removed it.
1562 (It makes no difference on Unix-like systems.)
1564 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1565 of recursive calls to match(). This is different to MATCH_LIMIT because
1566 that limits the total number of calls to match(), not all of which increase
1567 the depth of recursion. Limiting the recursion depth limits the amount of
1568 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1569 when PCRE is compiled, and changed at run time. A patch from Google adds
1570 this functionality to the C++ interface.
1572 18. Changes to the handling of Unicode character properties:
1574 (a) Updated the table to Unicode 4.1.0.
1576 (b) Recognize characters that are not in the table as "Cn" (undefined).
1578 (c) I revised the way the table is implemented to a much improved format
1579 which includes recognition of ranges. It now supports the ranges that
1580 are defined in UnicodeData.txt, and it also amalgamates other
1581 characters into ranges. This has reduced the number of entries in the
1582 table from around 16,000 to around 3,000, thus reducing its size
1583 considerably. I realized I did not need to use a tree structure after
1584 all - a binary chop search is just as efficient. Having reduced the
1585 number of entries, I extended their size from 6 bytes to 8 bytes to
1586 allow for more data.
1588 (d) Added support for Unicode script names via properties such as \p{Han}.
1590 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1591 matching that character.
1593 20. When matching a repeated Unicode property with a minimum greater than zero,
1594 (for example \pL{2,}), PCRE could look past the end of the subject if it
1595 reached it while seeking the minimum number of characters. This could
1596 happen only if some of the characters were more than one byte long, because
1597 there is a check for at least the minimum number of bytes.
1599 21. Refactored the implementation of \p and \P so as to be more general, to
1600 allow for more different types of property in future. This has changed the
1601 compiled form incompatibly. Anybody with saved compiled patterns that use
1602 \p or \P will have to recompile them.
1604 22. Added "Any" and "L&" to the supported property types.
1606 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1607 but give a compile time error if the value is greater than 0xff.
1609 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1610 accidentally not being installed or uninstalled.
1612 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1613 made were to insert the current release number. This seemed silly, because
1614 it made things harder for people building PCRE on systems that don't run
1615 "configure". I have turned pcre.h into a distributed file, no longer built
1616 by "configure", with the version identification directly included. There is
1617 no longer a pcre.h.in file.
1619 However, this change necessitated a change to the pcre-config script as
1620 well. It is built from pcre-config.in, and one of the substitutions was the
1621 release number. I have updated configure.ac so that ./configure now finds
1622 the release number by grepping pcre.h.
1624 26. Added the ability to run the tests under valgrind.
1627 Version 6.4 05-Sep-05
1628 ---------------------
1630 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1631 "--" to be printed when multiple files were scanned, even when none of the
1632 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1633 consider it to be a bug, and have restored the previous behaviour.
1635 2. A couple of code tidies to get rid of compiler warnings.
1637 3. The pcretest program used to cheat by referring to symbols in the library
1638 whose names begin with _pcre_. These are internal symbols that are not
1639 really supposed to be visible externally, and in some environments it is
1640 possible to suppress them. The cheating is now confined to including
1641 certain files from the library's source, which is a bit cleaner.
1643 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1644 file's purpose clearer.
1646 5. Reorganized pcre_ucp_findchar().
1649 Version 6.3 15-Aug-05
1650 ---------------------
1652 1. The file libpcre.pc.in did not have general read permission in the tarball.
1654 2. There were some problems when building without C++ support:
1656 (a) If C++ support was not built, "make install" and "make test" still
1657 tried to test it.
1659 (b) There were problems when the value of CXX was explicitly set. Some
1660 changes have been made to try to fix these, and ...
1662 (c) --disable-cpp can now be used to explicitly disable C++ support.
1664 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1665 backslash in a target when C++ was disabled. This confuses some
1666 versions of "make", apparently. Using an intermediate variable solves
1667 this. (Same for CPP_LOBJ.)
1669 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1670 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1671 necessary on certain architectures.
1673 4. Added a setting of -export-symbols-regex to the link command to remove
1674 those symbols that are exported in the C sense, but actually are local
1675 within the library, and not documented. Their names all begin with
1676 "_pcre_". This is not a perfect job, because (a) we have to except some
1677 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1678 available (and never for static libraries). I have made a note to try to
1679 find a way round (a) in the future.
1682 Version 6.2 01-Aug-05
1683 ---------------------
1685 1. There was no test for integer overflow of quantifier values. A construction
1686 such as {1111111111111111} would give undefined results. What is worse, if
1687 a minimum quantifier for a parenthesized subpattern overflowed and became
1688 negative, the calculation of the memory size went wrong. This could have
1689 led to memory overwriting.
1691 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1693 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1694 operating environments where this matters.
1696 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1697 PCRE options from within the C++ wrapper.
1699 5. Named capturing subpatterns were not being correctly counted when a pattern
1700 was compiled. This caused two problems: (a) If there were more than 100
1701 such subpatterns, the calculation of the memory needed for the whole
1702 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1703 back references of the form \12, where the number was greater than 9, were
1704 not recognized as back references, even though there were sufficient
1705 previous subpatterns.
1707 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1708 versions of gcc, e.g. 2.95.4.
1711 Version 6.1 21-Jun-05
1712 ---------------------
1714 1. There was one reference to the variable "posix" in pcretest.c that was not
1715 surrounded by "#if !defined NOPOSIX".
1717 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1718 the cross-check on the old pcre_info() function, for the benefit of the
1719 cut-down version of PCRE that is currently imported into Exim.
1721 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1722 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1723 patterns that start (?i) but I don't think that matters. The use of (?i) is
1724 just an example; this all applies to the other options as well.
1726 4. Since libtool seems to echo the compile commands it is issuing, the output
1727 from "make" can be reduced a bit by putting "@" in front of each libtool
1728 compile command.
1730 5. Patch from the folks at Google for configure.in to be a bit more thorough
1731 in checking for a suitable C++ installation before trying to compile the
1732 C++ stuff. This should fix a reported problem when a compiler was present,
1733 but no suitable headers.
1735 6. The man pages all had just "PCRE" as their title. I have changed them to
1736 be the relevant file name. I have also arranged that these names are
1737 retained in the file doc/pcre.txt, which is a concatenation in text format
1738 of all the man pages except the little individual ones for each function.
1740 7. The NON-UNIX-USE file had not been updated for the different set of source
1741 files that come with release 6. I also added a few comments about the C++
1742 wrapper.
1745 Version 6.0 07-Jun-05
1746 ---------------------
1748 1. Some minor internal re-organization to help with my DFA experiments.
1750 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1751 didn't matter for the library itself when fully configured, but did matter
1752 when compiling without UCP support, or within Exim, where the ucp files are
1753 not imported.
1755 3. Refactoring of the library code to split up the various functions into
1756 different source modules. The addition of the new DFA matching code (see
1757 below) to a single monolithic source would have made it really too
1758 unwieldy, quite apart from causing all the code to be include in a
1759 statically linked application, when only some functions are used. This is
1760 relevant even without the DFA addition now that patterns can be compiled in
1761 one application and matched in another.
1763 The downside of splitting up is that there have to be some external
1764 functions and data tables that are used internally in different modules of
1765 the library but which are not part of the API. These have all had their
1766 names changed to start with "_pcre_" so that they are unlikely to clash
1767 with other external names.
1769 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1770 a different (DFA) algorithm. Although it is slower than the original
1771 function, it does have some advantages for certain types of matching
1772 problem.
1774 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1775 including restarting after a partial match.
1777 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1778 defined when compiling for Windows was sent to me. I have put it into the
1779 code, though I have no means of testing or verifying it.
1781 7. Added the pcre_refcount() auxiliary function.
1783 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1784 match before or at the first newline in the subject string. In pcretest,
1785 the /f option on a pattern can be used to set this.
1787 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1788 would behave wrongly. This has been present in PCRE since release 4.0.
1790 10. A number of changes to the pcregrep command:
1792 (a) Refactored how -x works; insert ^(...)$ instead of setting
1793 PCRE_ANCHORED and checking the length, in preparation for adding
1794 something similar for -w.
1796 (b) Added the -w (match as a word) option.
1798 (c) Refactored the way lines are read and buffered so as to have more
1799 than one at a time available.
1801 (d) Implemented a pcregrep test script.
1803 (e) Added the -M (multiline match) option. This allows patterns to match
1804 over several lines of the subject. The buffering ensures that at least
1805 8K, or the rest of the document (whichever is the shorter) is available
1806 for matching (and similarly the previous 8K for lookbehind assertions).
1808 (f) Changed the --help output so that it now says
1810 -w, --word-regex(p)
1812 instead of two lines, one with "regex" and the other with "regexp"
1813 because that confused at least one person since the short forms are the
1814 same. (This required a bit of code, as the output is generated
1815 automatically from a table. It wasn't just a text change.)
1817 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1818 option but starts with a hyphen. Could be a pattern or a path name
1819 starting with a hyphen, for instance.
1821 (h) "-" can be given as a file name to represent stdin.
1823 (i) When file names are being printed, "(standard input)" is used for
1824 the standard input, for compatibility with GNU grep. Previously
1825 "<stdin>" was used.
1827 (j) The option --label=xxx can be used to supply a name to be used for
1828 stdin when file names are being printed. There is no short form.
1830 (k) Re-factored the options decoding logic because we are going to add
1831 two more options that take data. Such options can now be given in four
1832 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1834 (l) Added the -A, -B, and -C options for requesting that lines of context
1835 around matches be printed.
1837 (m) Added the -L option to print the names of files that do not contain
1838 any matching lines, that is, the complement of -l.
1840 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1841 continue to scan other files.
1843 (o) The -s option was incorrectly implemented. For compatibility with other
1844 greps, it now suppresses the error message for a non-existent or non-
1845 accessible file (but not the return code). There is a new option called
1846 -q that suppresses the output of matching lines, which was what -s was
1847 previously doing.
1849 (p) Added --include and --exclude options to specify files for inclusion
1850 and exclusion when recursing.
1852 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1853 Hopefully, it now does.
1855 12. Missing cast in pcre_study().
1857 13. Added an "uninstall" target to the makefile.
1859 14. Replaced "extern" in the function prototypes in Makefile.in with
1860 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1861 world, but is set differently for Windows.
1863 15. Added a second compiling function called pcre_compile2(). The only
1864 difference is that it has an extra argument, which is a pointer to an
1865 integer error code. When there is a compile-time failure, this is set
1866 non-zero, in addition to the error test pointer being set to point to an
1867 error message. The new argument may be NULL if no error number is required
1868 (but then you may as well call pcre_compile(), which is now just a
1869 wrapper). This facility is provided because some applications need a
1870 numeric error indication, but it has also enabled me to tidy up the way
1871 compile-time errors are handled in the POSIX wrapper.
1873 16. Added VPATH=.libs to the makefile; this should help when building with one
1874 prefix path and installing with another. (Or so I'm told by someone who
1875 knows more about this stuff than I do.)
1877 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1878 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1879 match everything, including newlines. This is not POSIX-compatible, but
1880 somebody wanted the feature. From pcretest it can be activated by using
1881 both the P and the s flags.
1883 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1885 19. libpcre.pc was being incorrectly installed as executable.
1887 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1888 it now also looks for '\r' so that it will work unmodified on Windows.
1890 21. Added Google's contributed C++ wrapper to the distribution.
1892 22. Added some untidy missing memory free() calls in pcretest, to keep
1893 Electric Fence happy when testing.
1897 Version 5.0 13-Sep-04
1898 ---------------------
1900 1. Internal change: literal characters are no longer packed up into items
1901 containing multiple characters in a single byte-string. Each character
1902 is now matched using a separate opcode. However, there may be more than one
1903 byte in the character in UTF-8 mode.
1905 2. The pcre_callout_block structure has two new fields: pattern_position and
1906 next_item_length. These contain the offset in the pattern to the next match
1907 item, and its length, respectively.
1909 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1910 insertion of callouts before each pattern item. Added the /C option to
1911 pcretest to make use of this.
1913 4. On the advice of a Windows user, the lines
1915 #if defined(_WIN32) || defined(WIN32)
1916 _setmode( _fileno( stdout ), 0x8000 );
1917 #endif /* defined(_WIN32) || defined(WIN32) */
1919 have been added to the source of pcretest. This apparently does useful
1920 magic in relation to line terminators.
1922 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1923 for the benefit of those environments where the "b" makes a difference.
1925 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1926 to know about it. I have put a hack into configure.in that adds in code
1927 to set GCC=yes if CC=icc. This seems to end up at a point in the
1928 generated configure script that is early enough to affect the setting of
1929 compiler options, which is what is needed, but I have no means of testing
1930 whether it really works. (The user who reported this had patched the
1931 generated configure script, which of course I cannot do.)
1933 LATER: After change 22 below (new libtool files), the configure script
1934 seems to know about icc (and also ecc). Therefore, I have commented out
1935 this hack in configure.in.
1937 7. Added support for pkg-config (2 patches were sent in).
1939 8. Negated POSIX character classes that used a combination of internal tables
1940 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1941 [[:^ascii]]. Typically, they would match almost any characters. The other
1942 POSIX classes were not broken in this way.
1944 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1945 to find the match, as PCRE was deluded into thinking that the match had to
1946 start at the start point or following a newline. The same bug applied to
1947 patterns with negative forward assertions or any backward assertions
1948 preceding ".*" at the start, unless the pattern required a fixed first
1949 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1951 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1952 starting at the last subject character, bytes beyond the end of the subject
1953 string were read.
1955 11. Renamed the variable "class" as "classbits" to make life easier for C++
1956 users. (Previously there was a macro definition, but it apparently wasn't
1957 enough.)
1959 12. Added the new field "tables" to the extra data so that tables can be passed
1960 in at exec time, or the internal tables can be re-selected. This allows
1961 a compiled regex to be saved and re-used at a later time by a different
1962 program that might have everything at different addresses.
1964 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1965 -R library as well as a -L library.
1967 14. The debugging options of pcretest (-d on the command line or D on a
1968 pattern) showed incorrect output for anything following an extended class
1969 that contained multibyte characters and which was followed by a quantifier.
1971 15. Added optional support for general category Unicode character properties
1972 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1973 support. It adds about 90K to the size of the library. The meanings of the
1974 inbuilt class escapes such as \d and \s have NOT been changed.
1976 16. Updated pcredemo.c to include calls to free() to release the memory for the
1977 compiled pattern.
1979 17. The generated file chartables.c was being created in the source directory
1980 instead of in the building directory. This caused the build to fail if the
1981 source directory was different from the building directory, and was
1982 read-only.
1984 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1985 file. No doubt somebody will tell me if they don't make sense... Also added
1986 Dan Mooney's comments about building on OpenVMS.
1988 19. Added support for partial matching via the PCRE_PARTIAL option for
1989 pcre_exec() and the \P data escape in pcretest.
1991 20. Extended pcretest with 3 new pattern features:
1993 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1994 write the compiled pattern to the file whose name is "rest-of-line".
1995 This is a straight binary dump of the data, with the saved pointer to
1996 the character tables forced to be NULL. The study data, if any, is
1997 written too. After writing, pcretest reads a new pattern.
1999 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
2000 compiled pattern from the given file. There must not be any
2001 occurrences of "<" in the file name (pretty unlikely); if there are,
2002 pcretest will instead treat the initial "<" as a pattern delimiter.
2003 After reading in the pattern, pcretest goes on to read data lines as
2004 usual.
2006 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
2007 and 16-bit fields in a compiled pattern, to simulate a pattern that
2008 was compiled on a host of opposite endianness.
2010 21. The pcre-exec() function can now cope with patterns that were compiled on
2011 hosts of opposite endianness, with this restriction:
2013 As for any compiled expression that is saved and used later, the tables
2014 pointer field cannot be preserved; the extra_data field in the arguments
2015 to pcre_exec() should be used to pass in a tables address if a value
2016 other than the default internal tables were used at compile time.
2018 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
2019 now diagnosed as an error. Previously, most of the time, a negative number
2020 would have been treated as zero, but if in addition "ovector" was passed as
2021 NULL, a crash could occur.
2023 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2024 new versions from the libtool 1.5 distribution (the last one is a copy of
2025 a file called libtool.m4). This seems to have fixed the need to patch
2026 "configure" to support Darwin 1.3 (which I used to do). However, I still
2027 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2028 workstation).
2030 24. Changed the PCRE licence to be the more standard "BSD" licence.
2033 Version 4.5 01-Dec-03
2034 ---------------------
2036 1. There has been some re-arrangement of the code for the match() function so
2037 that it can be compiled in a version that does not call itself recursively.
2038 Instead, it keeps those local variables that need separate instances for
2039 each "recursion" in a frame on the heap, and gets/frees frames whenever it
2040 needs to "recurse". Keeping track of where control must go is done by means
2041 of setjmp/longjmp. The whole thing is implemented by a set of macros that
2042 hide most of the details from the main code, and operates only if
2043 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2044 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2045 operating.
2047 To make it easier for callers to provide specially tailored get/free
2048 functions for this usage, two new functions, pcre_stack_malloc, and
2049 pcre_stack_free, are used. They are always called in strict stacking order,
2050 and the size of block requested is always the same.
2052 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2053 PCRE has been compiled to use the stack or the heap for recursion. The
2054 -C option of pcretest uses this to show which version is compiled.
2056 A new data escape \S, is added to pcretest; it causes the amounts of store
2057 obtained and freed by both kinds of malloc/free at match time to be added
2058 to the output.
2060 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2061 what's available on my current Linux desktop machine.
2063 3. When matching a UTF-8 string, the test for a valid string at the start has
2064 been extended. If start_offset is not zero, PCRE now checks that it points
2065 to a byte that is the start of a UTF-8 character. If not, it returns
2066 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2067 this is necessary because there may be backward assertions in the pattern.
2068 When matching the same subject several times, it may save resources to use
2069 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2071 4. The code for checking the validity of UTF-8 strings has been tightened so
2072 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2073 containing "overlong sequences".
2075 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2076 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2077 should have been "&", but it just so happened that all the cases this let
2078 through by mistake were picked up later in the function.
2080 6. I had used a variable called "isblank" - this is a C99 function, causing
2081 some compilers to warn. To avoid this, I renamed it (as "blankclass").
2083 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2084 prompting; (b) run "./pcretest /dev/null" at the start of the test script
2085 so the version is shown; (c) stop "make test" echoing "./RunTest".
2087 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2089 9. The prototype for memmove() for systems that don't have it was using
2090 size_t, but the inclusion of the header that defines size_t was later. I've
2091 moved the #includes for the C headers earlier to avoid this.
2093 10. Added some adjustments to the code to make it easier to compiler on certain
2094 special systems:
2096 (a) Some "const" qualifiers were missing.
2097 (b) Added the macro EXPORT before all exported functions; by default this
2098 is defined to be empty.
2099 (c) Changed the dftables auxiliary program (that builds chartables.c) so
2100 that it reads its output file name as an argument instead of writing
2101 to the standard output and assuming this can be redirected.
2103 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2104 class containing characters with values greater than 255, PCRE compilation
2105 went into a loop.
2107 12. A recursive reference to a subpattern that was within another subpattern
2108 that had a minimum quantifier of zero caused PCRE to crash. For example,
2109 (x(y(?2))z)? provoked this bug with a subject that got as far as the
2110 recursion. If the recursively-called subpattern itself had a zero repeat,
2111 that was OK.
2113 13. In pcretest, the buffer for reading a data line was set at 30K, but the
2114 buffer into which it was copied (for escape processing) was still set at
2115 1024, so long lines caused crashes.
2117 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2118 "internal error: code overflow...". This applied to any character class
2119 that was followed by a possessive quantifier.
2121 15. Modified the Makefile to add libpcre.la as a prerequisite for
2122 libpcreposix.la because I was told this is needed for a parallel build to
2123 work.
2125 16. If a pattern that contained .* following optional items at the start was
2126 studied, the wrong optimizing data was generated, leading to matching
2127 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2128 matching string must start with a or b or c. The correct conclusion for
2129 this pattern is that a match can start with any character.
2132 Version 4.4 13-Aug-03
2133 ---------------------
2135 1. In UTF-8 mode, a character class containing characters with values between
2136 127 and 255 was not handled correctly if the compiled pattern was studied.
2137 In fixing this, I have also improved the studying algorithm for such
2138 classes (slightly).
2140 2. Three internal functions had redundant arguments passed to them. Removal
2141 might give a very teeny performance improvement.
2143 3. Documentation bug: the value of the capture_top field in a callout is *one
2144 more than* the number of the hightest numbered captured substring.
2146 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2147 in incorrectly linking with a previously installed version. They now link
2148 explicitly with libpcre.la.
2150 5. configure.in no longer needs to recognize Cygwin specially.
2152 6. A problem in pcre.in for Windows platforms is fixed.
2154 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2155 pcretest, it used to include the size of the study block as part of its
2156 output. Unfortunately, the structure contains a field that has a different
2157 size on different hardware architectures. This meant that the tests that
2158 showed this size failed. As the block is currently always of a fixed size,
2159 this information isn't actually particularly useful in pcretest output, so
2160 I have just removed it.
2162 8. Three pre-processor statements accidentally did not start in column 1.
2163 Sadly, there are *still* compilers around that complain, even though
2164 standard C has not required this for well over a decade. Sigh.
2166 9. In pcretest, the code for checking callouts passed small integers in the
2167 callout_data field, which is a void * field. However, some picky compilers
2168 complained about the casts involved for this on 64-bit systems. Now
2169 pcretest passes the address of the small integer instead, which should get
2170 rid of the warnings.
2172 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2173 both compile and run time, and gives an error if an invalid UTF-8 sequence
2174 is found. There is a option for disabling this check in cases where the
2175 string is known to be correct and/or the maximum performance is wanted.
2177 11. In response to a bug report, I changed one line in Makefile.in from
2179 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2180 to
2181 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2183 to look similar to other lines, but I have no way of telling whether this
2184 is the right thing to do, as I do not use Windows. No doubt I'll get told
2185 if it's wrong...
2188 Version 4.3 21-May-03
2189 ---------------------
2191 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2192 Makefile.
2194 2. Some refactoring to improve the quality of the code:
2196 (i) The utf8_table... variables are now declared "const".
2198 (ii) The code for \cx, which used the "case flipping" table to upper case
2199 lower case letters, now just substracts 32. This is ASCII-specific,
2200 but the whole concept of \cx is ASCII-specific, so it seems
2201 reasonable.
2203 (iii) PCRE was using its character types table to recognize decimal and
2204 hexadecimal digits in the pattern. This is silly, because it handles
2205 only 0-9, a-f, and A-F, but the character types table is locale-
2206 specific, which means strange things might happen. A private
2207 table is now used for this - though it costs 256 bytes, a table is
2208 much faster than multiple explicit tests. Of course, the standard
2209 character types table is still used for matching digits in subject
2210 strings against \d.
2212 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2213 ending in _t are). So I've renamed it as ESC_tee.
2215 3. The first argument for regexec() in the POSIX wrapper should have been
2216 defined as "const".
2218 4. Changed pcretest to use malloc() for its buffers so that they can be
2219 Electric Fenced for debugging.
2221 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2222 to read one or more bytes before the start of the subject string. Often this
2223 had no effect on PCRE's behaviour, but in some circumstances it could
2224 provoke a segmentation fault.
2226 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2227 to try to read one or more bytes before the start of the subject string.
2229 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2230 UTF-8 support could misbehave in various ways if the subject string
2231 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2232 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2233 back over UTF-8 characters.)
2236 Version 4.2 14-Apr-03
2237 ---------------------
2239 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2241 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2242 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2243 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2244 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2245 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2247 Note: automatic setting of the BUILD variables is not yet working
2248 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2249 compile-time but not at link-time
2250 [LINK]: use for linking executables only
2251 make different versions for Windows and non-Windows
2252 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2253 libraries
2254 [LINK_FOR_BUILD]: new variable
2255 [OBJEXT]: use throughout
2256 [EXEEXT]: use throughout
2257 <winshared>: new target
2258 <wininstall>: new target
2259 <dftables.o>: use native compiler
2260 <dftables>: use native linker
2261 <install>: handle Windows platform correctly
2262 <clean>: ditto
2263 <check>: ditto
2264 copy DLL to top builddir before testing
2266 As part of these changes, -no-undefined was removed again. This was reported
2267 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2268 in any case.
2270 3. Some tidies to get rid of compiler warnings:
2272 . In the match_data structure, match_limit was an unsigned long int, whereas
2273 match_call_count was an int. I've made them both unsigned long ints.
2275 . In pcretest the fact that a const uschar * doesn't automatically cast to
2276 a void * provoked a warning.
2278 . Turning on some more compiler warnings threw up some "shadow" variables
2279 and a few more missing casts.
2281 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2282 option, a class that contained a single character with a value between 128
2283 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2285 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2286 option, a class that contained several characters, but with at least one
2287 whose value was between 128 and 255 caused PCRE to crash.
2290 Version 4.1 12-Mar-03
2291 ---------------------
2293 1. Compiling with gcc -pedantic found a couple of places where casts were
2294 needed, and a string in dftables.c that was longer than standard compilers are
2295 required to support.
2297 2. Compiling with Sun's compiler found a few more places where the code could
2298 be tidied up in order to avoid warnings.
2300 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2301 first of these names is deprecated in the latest Autoconf in favour of the name
2302 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2303 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2304 analogy I have changed it to CFLAGS_FOR_BUILD.
2306 4. Added -no-undefined to the linking command in the Makefile, because this is
2307 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2308 linking step for the pcreposix library.
2310 5. PCRE was failing to diagnose the case of two named groups with the same
2311 name.
2313 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2314 literal character that is needed in the subject for a match, and scans along to
2315 ensure that it is present before embarking on the full matching process. This
2316 saves time in cases of nested unlimited repeats that are never going to match.
2317 Problem: the scan can take a lot of time if the subject is very long (e.g.
2318 megabytes), thus penalizing straightforward matches. It is now done only if the
2319 amount of subject to be scanned is less than 1000 bytes.
2321 7. A lesser problem with the same optimization is that it was recording the
2322 first character of an anchored pattern as "needed", thus provoking a search
2323 right along the subject, even when the first match of the pattern was going to
2324 fail. The "needed" character is now not set for anchored patterns, unless it
2325 follows something in the pattern that is of non-fixed length. Thus, it still
2326 fulfils its original purpose of finding quick non-matches in cases of nested
2327 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2330 Version 4.0 17-Feb-03
2331 ---------------------
2333 1. If a comment in an extended regex that started immediately after a meta-item
2334 extended to the end of string, PCRE compiled incorrect data. This could lead to
2335 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2337 2. Moved to autoconf 2.53 and libtool 1.4.2.
2339 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2340 the special perltest8 script is no longer needed - all the tests can be run
2341 from a single perltest script.
2343 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2344 by \s. It has now been removed in PCRE. This means it isn't recognized as
2345 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2346 class [:space:] *does* include VT, thereby creating a mess.
2348 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2349 space and tab.
2351 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2352 its new features into the main test script, reducing the number of scripts.
2354 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2355 were backward compatible, and made the (?i) apply to the whole pattern, as if
2356 /i were given. Now it behaves more logically, and applies the option setting
2357 only to what follows. PCRE has been changed to follow suit. However, if it
2358 finds options settings right at the start of the pattern, it extracts them into
2359 the global options, as before. Thus, they show up in the info data.
2361 8. Added support for the \Q...\E escape sequence. Characters in between are
2362 treated as literals. This is slightly different from Perl in that $ and @ are
2363 also handled as literals inside the quotes. In Perl, they will cause variable
2364 interpolation. Note the following examples:
2366 Pattern PCRE matches Perl matches
2368 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2369 \Qabc\$xyz\E abc\$xyz abc\$xyz
2370 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2372 For compatibility with Perl, \Q...\E sequences are recognized inside character
2373 classes as well as outside them.
2375 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2376 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2377 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2378 signed/unsigned warnings.
2380 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2381 option for pcretest, so I've replaced it by a simple function that does just
2382 that job.
2384 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2385 "pcregrep -".
2387 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2388 Java package. This provides some syntactic sugar for simple cases of what my
2389 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2390 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2391 item, you can use this simplified notation. Note that only makes sense with
2392 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2393 greediness, whatever the setting of the PCRE_UNGREEDY option.
2395 13. A change of greediness default within a pattern was not taking effect at
2396 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2397 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2398 was abstracted outside.
2400 14. PCRE now supports the \G assertion. It is true when the current matching
2401 position is at the start point of the match. This differs from \A when the
2402 starting offset is non-zero. Used with the /g option of pcretest (or similar
2403 code), it works in the same way as it does for Perl's /g option. If all
2404 alternatives of a regex begin with \G, the expression is anchored to the start
2405 match position, and the "anchored" flag is set in the compiled expression.
2407 15. Some bugs concerning the handling of certain option changes within patterns
2408 have been fixed. These applied to options other than (?ims). For example,
2409 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2410 been the other way round. Some of this was related to change 7 above.
2412 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2413 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2414 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2415 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2417 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2418 mode. Unlike ".", it always matches newline, whatever the setting of
2419 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2420 assertions. Perl allows it, but it doesn't (in general) work because it can't
2421 calculate the length of the lookbehind. At least, that's the case for Perl
2422 5.8.0 - I've been told they are going to document that it doesn't work in
2423 future.
2425 18. Added an error diagnosis for escapes that PCRE does not support: these are
2426 \L, \l, \N, \P, \p, \U, \u, and \X.
2428 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2429 reading past the end of the pattern in cases such as /[abcd/.
2431 20. PCRE was getting more memory than necessary for patterns with classes that
2432 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2434 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2435 compiling PCRE for use with Virtual Pascal.
2437 22. Small fix to the Makefile to make it work properly if the build is done
2438 outside the source tree.
2440 23. Added a new extension: a condition to go with recursion. If a conditional
2441 subpattern starts with (?(R) the "true" branch is used if recursion has
2442 happened, whereas the "false" branch is used only at the top level.
2444 24. When there was a very long string of literal characters (over 255 bytes
2445 without UTF support, over 250 bytes with UTF support), the computation of how
2446 much memory was required could be incorrect, leading to segfaults or other
2447 strange effects.
2449 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2450 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2451 there was a subsequent back reference to those brackets. This meant that, for
2452 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2453 possible to check for precisely this case. All we can do is abandon the
2454 optimization if .* occurs inside capturing brackets when there are any back
2455 references whatsoever. (See below for a better fix that came later.)
2457 26. The handling of the optimization for finding the first character of a
2458 non-anchored pattern, and for finding a character that is required later in the
2459 match were failing in some cases. This didn't break the matching; it just
2460 failed to optimize when it could. The way this is done has been re-implemented.
2462 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2464 28. Added a new feature that provides some of the functionality that Perl
2465 provides with (?{...}). The facility is termed a "callout". The way it is done
2466 in PCRE is for the caller to provide an optional function, by setting
2467 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2468 global variable. By default it is unset, which disables all calling out. To get
2469 the function called, the regex must include (?C) at appropriate points. This
2470 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2471 This provides a means of identifying different callout points. When PCRE
2472 reaches such a point in the regex, if pcre_callout has been set, the external
2473 function is called. It is provided with data in a structure called
2474 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2475 matching continues; if it returns a non-zero value, the match at the current
2476 point fails. However, backtracking will occur if possible. [This was changed
2477 later and other features added - see item 49 below.]
2479 29. pcretest is upgraded to test the callout functionality. It provides a
2480 callout function that displays information. By default, it shows the start of
2481 the match and the current position in the text. There are some new data escapes
2482 to vary what happens:
2484 \C+ in addition, show current contents of captured substrings
2485 \C- do not supply a callout function
2486 \C!n return 1 when callout number n is reached
2487 \C!n!m return 1 when callout number n is reached for the mth time
2489 30. If pcregrep was called with the -l option and just a single file name, it
2490 output "<stdin>" if a match was found, instead of the file name.
2492 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2493 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2494 pcre_exec(). This saves a malloc/free per call. The default value of
2495 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2496 when configuring.
2498 32. The default maximum size of a compiled pattern is 64K. There have been a
2499 few cases of people hitting this limit. The code now uses macros to handle the
2500 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2501 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2502 configuring. Tests 2 and 5 work only with 2-byte links because they output
2503 debugging information about compiled patterns.
2505 33. Internal code re-arrangements:
2507 (a) Moved the debugging function for printing out a compiled regex into
2508 its own source file (printint.c) and used #include to pull it into
2509 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2510 separate copies.
2512 (b) Defined the list of op-code names for debugging as a macro in
2513 internal.h so that it is next to the definition of the opcodes.
2515 (c) Defined a table of op-code lengths for simpler skipping along compiled
2516 code. This is again a macro in internal.h so that it is next to the
2517 definition of the opcodes.
2519 34. Added support for recursive calls to individual subpatterns, along the
2520 lines of Robin Houston's patch (but implemented somewhat differently).
2522 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2523 allow it to read and process whole directories in Win32. This code was
2524 contributed by Lionel Fourquaux; it has not been tested by me.
2526 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2527 used to name a group. Names consist of alphanumerics and underscores, and must
2528 be unique. Back references use the syntax (?P=name) and recursive calls use
2529 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2530 numbers. The function pcre_fullinfo() can be used after compilation to extract
2531 a name/number map. There are three relevant calls:
2533 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2534 PCRE_INFO_NAMECOUNT yields the number of entries
2535 PCRE_INFO_NAMETABLE yields a pointer to the map.
2537 The map is a vector of fixed-size entries. The size of each entry depends on
2538 the length of the longest name used. The first two bytes of each entry are the
2539 group number, most significant byte first. There follows the corresponding
2540 name, zero terminated. The names are in alphabetical order.
2542 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2543 case instead of 255. Making it the same both with and without UTF-8 support
2544 means that the same test output works with both.
2546 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2547 calling malloc() with a zero argument.
2549 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2550 optimization. I've improved things by keeping a bitmap of backreferences with
2551 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2552 fact referenced, the optimization can be applied. It is unlikely that a
2553 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2554 the match to follow \n) will appear inside brackets with a number greater than
2555 31, but if it does, any back reference > 31 suppresses the optimization.
2557 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2558 of disabling numbered capturing parentheses. Any opening parenthesis that is
2559 not followed by ? behaves as if it were followed by ?: but named parentheses
2560 can still be used for capturing (and they will acquire numbers in the usual
2561 way).
2563 41. Redesigned the return codes from the match() function into yes/no/error so
2564 that errors can be passed back from deep inside the nested calls. A malloc
2565 failure while inside a recursive subpattern call now causes the
2566 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2568 42. It is now possible to set a limit on the number of times the match()
2569 function is called in a call to pcre_exec(). This facility makes it possible to
2570 limit the amount of recursion and backtracking, though not in a directly
2571 obvious way, because the match() function is used in a number of different
2572 circumstances. The count starts from zero for each position in the subject
2573 string (for non-anchored patterns). The default limit is, for compatibility, a
2574 large number, namely 10 000 000. You can change this in two ways:
2576 (a) When configuring PCRE before making, you can use --with-match-limit=n
2577 to set a default value for the compiled library.
2579 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2580 a different value is set. See 45 below.
2582 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2584 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2585 of things that can be changed at compile time. The first argument specifies
2586 what is wanted and the second points to where the information is to be placed.
2587 The current list of available information is:
2591 The output is an integer that is set to one if UTF-8 support is available;
2592 otherwise it is set to zero.
2596 The output is an integer that it set to the value of the code that is used for
2597 newline. It is either LF (10) or CR (13).
2601 The output is an integer that contains the number of bytes used for internal
2602 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2606 The output is an integer that contains the threshold above which the POSIX
2607 interface uses malloc() for output vectors. See item 31 above.
2611 The output is an unsigned integer that contains the default limit of the number
2612 of match() calls in a pcre_exec() execution. See 42 above.
2614 44. pcretest has been upgraded by the addition of the -C option. This causes it
2615 to extract all the available output from the new pcre_config() function, and to
2616 output it. The program then exits immediately.
2618 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2619 order to support additional features. One way would have been to define
2620 pcre_exec2() (for example) with extra arguments, but this would not have been
2621 extensible, and would also have required all calls to the original function to
2622 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2623 is used for passing in "extra" data from pcre_study().
2625 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2626 contains the following fields:
2628 flags a bitmap indicating which of the following fields are set
2629 study_data opaque data from pcre_study()
2630 match_limit a way of specifying a limit on match() calls for a specific
2631 call to pcre_exec()
2632 callout_data data for callouts (see 49 below)
2634 The flag bits are also defined in pcre.h, and are
2640 The pcre_study() function now returns one of these new pcre_extra blocks, with
2641 the actual study data pointed to by the study_data field, and the
2642 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2643 before. That is, this change is entirely upwards-compatible and requires no
2644 change to existing code.
2646 If you want to pass in additional data to pcre_exec(), you can either place it
2647 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2648 block.
2650 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2651 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2652 times with different match limits, until it finds the minimum value needed for
2653 pcre_exec() to complete. The value is then output. This can be instructive; for
2654 most simple matches the number is quite small, but for pathological cases it
2655 gets very large very quickly.
2657 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2658 returns the size of the data block pointed to by the study_data field in a
2659 pcre_extra block, that is, the value that was passed as the argument to
2660 pcre_malloc() when PCRE was getting memory in which to place the information
2661 created by pcre_study(). The fourth argument should point to a size_t variable.
2662 pcretest has been extended so that this information is shown after a successful
2663 pcre_study() call when information about the compiled regex is being displayed.
2665 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2666 because what follows is always an absolute path. (Later: it turns out that this
2667 is more than cosmetic for MinGW, because it doesn't like empty path
2668 components.)
2670 49. Some changes have been made to the callout feature (see 28 above):
2672 (i) A callout function now has three choices for what it returns:
2674 0 => success, carry on matching
2675 > 0 => failure at this point, but backtrack if possible
2676 < 0 => serious error, return this value from pcre_exec()
2678 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2679 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2680 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2681 use by callout functions. It will never be used by PCRE itself.
2683 (ii) The pcre_extra structure (see 45 above) has a void * field called
2684 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2685 pcre_callout_block structure has a field of the same name. The contents of
2686 the field passed in the pcre_extra structure are passed to the callout
2687 function in the corresponding field in the callout block. This makes it
2688 easier to use the same callout-containing regex from multiple threads. For
2689 testing, the pcretest program has a new data escape
2691 \C*n pass the number n (may be negative) as callout_data
2693 If the callout function in pcretest receives a non-zero value as
2694 callout_data, it returns that value.
2696 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2697 there were some redundant $(CFLAGS) in commands that are now specified as
2698 $(LINK), which already includes $(CFLAGS).
2700 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2701 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2702 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2703 one-byte characters throughout. Note that case-insensitive matching applies
2704 only to characters whose values are less than 256. PCRE doesn't support the
2705 notion of cases for higher-valued characters.
2707 (i) A character class whose characters are all within 0-255 is handled as
2708 a bit map, and the map is inverted for negative classes. Previously, a
2709 character > 255 always failed to match such a class; however it should
2710 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2712 (ii) A negated character class with a single character < 255 is coded as
2713 "not this character" (OP_NOT). This wasn't working properly when the test
2714 character was multibyte, either singly or repeated.
2716 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2717 mode, for example: \x{100}{2,3}.
2719 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2720 singly or repeated) now correctly test multibyte characters. However,
2721 PCRE doesn't recognize any characters with values greater than 255 as
2722 digits, spaces, or word characters. Such characters always match \D, \S,
2723 and \W, and never match \d, \s, or \w.
2725 (v) Classes may now contain characters and character ranges with values
2726 greater than 255. For example: [ab\x{100}-\x{400}].
2728 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2729 PCRE in UTF-8 mode.
2731 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2732 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2733 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2734 value.)
2736 53. The single man page has become too large. I have therefore split it up into
2737 a number of separate man pages. These also give rise to individual HTML pages;
2738 these are now put in a separate directory, and there is an index.html page that
2739 lists them all. Some hyperlinking between the pages has been installed.
2741 54. Added convenience functions for handling named capturing parentheses.
2743 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2744 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2745 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2746 are faulted.
2748 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2749 calling configure. These values are used when compiling the dftables.c program
2750 which is run to generate the source of the default character tables. They
2751 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2752 you will need to set these values.
2754 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2757 Version 3.9 02-Jan-02
2758 ---------------------
2760 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2762 2. If --disable-static was given, the building process failed when trying to
2763 build pcretest and pcregrep. (For some reason it was using libtool to compile
2764 them, which is not right, as they aren't part of the library.)
2767 Version 3.8 18-Dec-01
2768 ---------------------
2770 1. The experimental UTF-8 code was completely screwed up. It was packing the
2771 bytes in the wrong order. How dumb can you get?
2774 Version 3.7 29-Oct-01
2775 ---------------------
2777 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2778 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2779 this didn't happen under Solaris 8, where I normally test things.
2781 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2782 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2783 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2784 this fix an hour or so after the initial 3.7 release.)
2787 Version 3.6 23-Oct-01
2788 ---------------------
2790 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2791 offsets passed as NULL with zero offset count.
2793 2. The config.guess and config.sub files had not been updated when I moved to
2794 the latest autoconf.
2797 Version 3.5 15-Aug-01
2798 ---------------------
2800 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2801 had been forgotten.
2803 2. By using declared but undefined structures, we can avoid using "void"
2804 definitions in pcre.h while keeping the internal definitions of the structures
2805 private.
2807 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2808 user point of view, this means that both static and shared libraries are built
2809 by default, but this can be individually controlled. More of the work of
2810 handling this static/shared cases is now inside libtool instead of PCRE's make
2811 file.
2813 4. The pcretest utility is now installed along with pcregrep because it is
2814 useful for users (to test regexs) and by doing this, it automatically gets
2815 relinked by libtool. The documentation has been turned into a man page, so
2816 there are now .1, .txt, and .html versions in /doc.
2818 5. Upgrades to pcregrep:
2819 (i) Added long-form option names like gnu grep.
2820 (ii) Added --help to list all options with an explanatory phrase.
2821 (iii) Added -r, --recursive to recurse into sub-directories.
2822 (iv) Added -f, --file to read patterns from a file.
2824 6. pcre_exec() was referring to its "code" argument before testing that
2825 argument for NULL (and giving an error if it was NULL).
2827 7. Upgraded Makefile.in to allow for compiling in a different directory from
2828 the source directory.
2830 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2831 options bits, the pointer it was passed was to an int instead of to an unsigned
2832 long int. This mattered only on 64-bit systems.
2834 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2835 generated) instead of pcre.in, which it its source. Also made the same change
2836 in several of the .c files.
2838 10. A new release of gcc defines printf() as a macro, which broke pcretest
2839 because it had an ifdef in the middle of a string argument for printf(). Fixed
2840 by using separate calls to printf().
2842 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2843 script, to force use of CR or LF instead of \n in the source. On non-Unix
2844 systems, the value can be set in config.h.
2846 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2847 absolute limit. Changed the text of the error message to make this clear, and
2848 likewise updated the man page.
2850 13. The limit of 99 on the number of capturing subpatterns has been removed.
2851 The new limit is 65535, which I hope will not be a "real" limit.
2854 Version 3.4 22-Aug-00
2855 ---------------------
2857 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2859 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2862 Version 3.3 01-Aug-00
2863 ---------------------
2865 1. If an octal character was given, but the value was greater than \377, it
2866 was not getting masked to the least significant bits, as documented. This could
2867 lead to crashes in some systems.
2869 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2870 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2872 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2873 These just pass their arguments on to (pcre_free)(), but they are provided
2874 because some uses of PCRE bind it to non-C systems that can call its functions,
2875 but cannot call free() or pcre_free() directly.
2877 4. Add "make test" as a synonym for "make check". Corrected some comments in
2878 the Makefile.
2880 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2881 Makefile.
2883 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2884 command called pgrep for grepping around the active processes.
2886 7. Added the beginnings of support for UTF-8 character strings.
2888 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2889 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2890 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2891 out for the ar command.)
2894 Version 3.2 12-May-00
2895 ---------------------
2897 This is purely a bug fixing release.
2899 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2900 of ZA. This was just one example of several cases that could provoke this bug,
2901 which was introduced by change 9 of version 2.00. The code for breaking
2902 infinite loops after an iteration that matches an empty string was't working
2903 correctly.
2905 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2906 when matched against abbab (for example). After matching an empty string, it
2907 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2908 caused it to match further down the string than it should.
2910 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2911 was there because it doesn't seem to be needed, and it causes trouble on some
2912 systems, as it is not a Standard C header. It has been removed.
2914 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2915 were reported on the Macintosh. The changes were from
2917 while ((c = *(++ptr)) != 0 && c != '\n');
2918 to
2919 while ((c = *(++ptr)) != 0 && c != '\n') ;
2921 Totally extraordinary, but if that's what it takes...
2923 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2924 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2925 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2926 assumes the way PCRE uses memmove() (always moving upwards).
2928 6. PCRE is being used in one environment where strchr() is not available. There
2929 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2930 faster code anyway.
2933 Version 3.1 09-Feb-00
2934 ---------------------
2936 The only change in this release is the fixing of some bugs in Makefile.in for
2937 the "install" target:
2939 (1) It was failing to install pcreposix.h.
2941 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2944 Version 3.0 01-Feb-00
2945 ---------------------
2947 1. Add support for the /+ modifier to perltest (to output $` like it does in
2948 pcretest).
2950 2. Add support for the /g modifier to perltest.
2952 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2953 matches null strings.
2955 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2956 pattern. Perl treats empty patterns specially - it reuses the most recent
2957 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2958 effect.
2960 5. The POSIX interface was broken in that it was just handing over the POSIX
2961 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2962 required a bigger vector, with some working space on the end. This means that
2963 the POSIX wrapper now has to get and free some memory, and copy the results.
2965 6. Added some simple autoconf support, placing the test data and the
2966 documentation in separate directories, re-organizing some of the
2967 information files, and making it build pcre-config (a GNU standard). Also added
2968 libtool support for building PCRE as a shared library, which is now the
2969 default.
2971 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2972 09 are not valid octal constants. Single digits will be used for minor values
2973 less than 10.
2975 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2976 existing programs that set these in the POSIX interface can use PCRE without
2977 modification.
2979 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2980 return all that pcre_info() returns, plus additional data. The pcre_info()
2981 function is retained for compatibility, but is considered to be obsolete.
2983 10. Added experimental recursion feature (?R) to handle one common case that
2984 Perl 5.6 will be able to do with (?p{...}).
2986 11. Added support for POSIX character classes like [:alpha:], which Perl is
2987 adopting.
2990 Version 2.08 31-Aug-99
2991 ----------------------
2993 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2994 trying to match at the startoffset position, but instead was moving forward to
2995 the next newline as if a previous match had failed.
2997 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2998 and could get into a loop if a null string was matched other than at the start
2999 of the subject.
3001 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
3002 be distinguished at compile time, and for completeness also added PCRE_DATE.
3004 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
3005 in GnuWin32 environments.
3008 Version 2.07 29-Jul-99
3009 ----------------------
3011 1. The documentation is now supplied in plain text form and HTML as well as in
3012 the form of man page sources.
3014 2. C++ compilers don't like assigning (void *) values to other pointer types.
3015 In particular this affects malloc(). Although there is no problem in Standard
3016 C, I've put in casts to keep C++ compilers happy.
3018 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
3019 should be (const char *).
3021 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3022 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3023 However, I haven't made this a standard facility. The documentation doesn't
3024 mention it, and the Makefile doesn't support it.
3026 5. The Makefile now contains an "install" target, with editable destinations at
3027 the top of the file. The pcretest program is not installed.
3029 6. pgrep -V now gives the PCRE version number and date.
3031 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3032 causing the entire string to be ignored, instead of just the last character.
3034 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3035 non-matching string, it can take a very, very long time, even for strings of
3036 quite modest length, because of the nested recursion. PCRE now does better in
3037 some of these cases. It does this by remembering the last required literal
3038 character in the pattern, and pre-searching the subject to ensure it is present
3039 before running the real match. In other words, it applies a heuristic to detect
3040 some types of certain failure quickly, and in the above example, if presented
3041 with a string that has no trailing " it gives "no match" very quickly.
3043 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3044 other alternatives are tried instead.
3047 Version 2.06 09-Jun-99
3048 ----------------------
3050 1. Change pcretest's output for amount of store used to show just the code
3051 space, because the remainder (the data block) varies in size between 32-bit and
3052 64-bit systems.
3054 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3055 start matching at. This allows lookbehinds to work when searching for multiple
3056 occurrences in a string.
3058 3. Added additional options to pcretest for testing multiple occurrences:
3060 /+ outputs the rest of the string that follows a match
3061 /g loops for multiple occurrences, using the new startoffset argument
3062 /G loops for multiple occurrences by passing an incremented pointer
3064 4. PCRE wasn't doing the "first character" optimization for patterns starting
3065 with \b or \B, though it was doing it for other lookbehind assertions. That is,
3066 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3067 the letter 'x'. On long subject strings, this gives a significant speed-up.
3070 Version 2.05 21-Apr-99
3071 ----------------------
3073 1. Changed the type of magic_number from int to long int so that it works
3074 properly on 16-bit systems.
3076 2. Fixed a bug which caused patterns starting with .* not to work correctly
3077 when the subject string contained newline characters. PCRE was assuming
3078 anchoring for such patterns in all cases, which is not correct because .* will
3079 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3080 DOTALL is set at top level; otherwise it knows that patterns starting with .*
3081 must be retried after every newline in the subject.
3084 Version 2.04 18-Feb-99
3085 ----------------------
3087 1. For parenthesized subpatterns with repeats whose minimum was zero, the
3088 computation of the store needed to hold the pattern was incorrect (too large).
3089 If such patterns were nested a few deep, this could multiply and become a real
3090 problem.
3092 2. Added /M option to pcretest to show the memory requirement of a specific
3093 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3095 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3096 compiled in such a way that the backtracking after subsequent failure was
3097 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3098 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3101 Version 2.03 02-Feb-99
3102 ----------------------
3104 1. Fixed typo and small mistake in man page.
3106 2. Added 4th condition (GPL supersedes if conflict) and created separate
3107 LICENCE file containing the conditions.
3109 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3110 Perl, that is the internal \ allows the delimiter to be included in the
3111 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3112 the final delimiter, add \ to the end of the pattern (to test the error).
3114 4. Added the convenience functions for extracting substrings after a successful
3115 match. Updated pcretest to make it able to test these functions.
3118 Version 2.02 14-Jan-99
3119 ----------------------
3121 1. Initialized the working variables associated with each extraction so that
3122 their saving and restoring doesn't refer to uninitialized store.
3124 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3125 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3126 fix the problem.
3128 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3129 calls, and wasn't printing the correct value for compiling calls. Increased the
3130 default value of LOOPREPEAT, and the number of significant figures in the
3131 times.
3133 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3135 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3136 a building problem on Windows NT with a FAT file system.
3139 Version 2.01 21-Oct-98
3140 ----------------------
3142 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3143 to character tables built by pcre_maketables() in the current locale. If NULL
3144 is passed, the default tables are used.
3147 Version 2.00 24-Sep-98
3148 ----------------------
3150 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3151 it any more.
3153 2. Allow quantification of (?>) groups, and make it work correctly.
3155 3. The first character computation wasn't working for (?>) groups.
3157 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3158 end of the subject) and add 5.005's \z, which really does match only at the
3159 very end of the subject.
3161 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3163 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3164 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3165 localized options. All options to pcre_study() were also removed.
3167 7. Add other new features from 5.005:
3169 $(?<= positive lookbehind
3170 $(?<! negative lookbehind
3171 (?imsx-imsx) added the unsetting capability
3172 such a setting is global if at outer level; local otherwise
3173 (?imsx-imsx:) non-capturing groups with option setting
3174 (?(cond)re|re) conditional pattern matching
3176 A backreference to itself in a repeated group matches the previous
3177 captured string.
3179 8. General tidying up of studying (both automatic and via "study")
3180 consequential on the addition of new assertions.
3182 9. As in 5.005, unlimited repeated groups that could match an empty substring
3183 are no longer faulted at compile time. Instead, the loop is forcibly broken at
3184 runtime if any iteration does actually match an empty substring.
3186 10. Include the RunTest script in the distribution.
3188 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3189 discrepancies, some of which were old and were also with respect to 5.004. They
3190 have now been fixed.
3193 Version 1.09 28-Apr-98
3194 ----------------------
3196 1. A negated single character class followed by a quantifier with a minimum
3197 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3198 program crashes, or just wrong answers. This did not apply to negated classes
3199 containing more than one character, or to minima other than one.
3202 Version 1.08 27-Mar-98
3203 ----------------------
3205 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3207 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3208 latter must appear before anything that relies on it in the pattern.
3211 Version 1.07 16-Feb-98
3212 ----------------------
3214 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3215 repeat of a potentially empty string).
3218 Version 1.06 23-Jan-98
3219 ----------------------
3221 1. Added Markus Oberhumer's little patches for C++.
3223 2. Literal strings longer than 255 characters were broken.
3226 Version 1.05 23-Dec-97
3227 ----------------------
3229 1. Negated character classes containing more than one character were failing if
3230 PCRE_CASELESS was set at run time.
3233 Version 1.04 19-Dec-97
3234 ----------------------
3236 1. Corrected the man page, where some "const" qualifiers had been omitted.
3238 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3239 input syntax.
3241 3. Fixed memory leak which occurred when a regex with back references was
3242 matched with an offsets vector that wasn't big enough. The temporary memory
3243 that is used in this case wasn't being freed if the match failed.
3245 4. Tidied pcretest to ensure it frees memory that it gets.
3247 5. Temporary memory was being obtained in the case where the passed offsets
3248 vector was exactly big enough.
3250 6. Corrected definition of offsetof() from change 5 below.
3252 7. I had screwed up change 6 below and broken the rules for the use of
3253 setjmp(). Now fixed.
3256 Version 1.03 18-Dec-97
3257 ----------------------
3259 1. A erroneous regex with a missing opening parenthesis was correctly
3260 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3261 on some systems.
3263 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3264 it was reported that one broken compiler failed on the former because "code" is
3265 also an independent variable.
3267 3. The erroneous regex a[]b caused an array overrun reference.
3269 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3270 fail on data ending with that character. (It was going on too far, and checking
3271 the next character, typically a binary zero.) This was specific to the
3272 optimized code for single-character negative classes.
3274 5. Added a contributed patch from the TIN world which does the following:
3276 + Add an undef for memmove, in case the the system defines a macro for it.
3278 + Add a definition of offsetof(), in case there isn't one. (I don't know
3279 the reason behind this - offsetof() is part of the ANSI standard - but
3280 it does no harm).
3282 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3283 most of the places where whitespace preceded '#'. I have given up and
3284 allowed the remaining 2 cases to be at the margin.
3286 + Rename some variables in pcre to eliminate shadowing. This seems very
3287 pedantic, but does no harm, of course.
3289 6. Moved the call to setjmp() into its own function, to get rid of warnings
3290 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3292 7. Constructs such as \d{8,} were compiling into the equivalent of
3293 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3294 outcome, but in this particular case used more store than had been allocated,
3295 which caused the bug to be discovered because it threw up an internal error.
3297 8. The debugging code in both pcre and pcretest for outputting the compiled
3298 form of a regex was going wrong in the case of back references followed by
3299 curly-bracketed repeats.
3302 Version 1.02 12-Dec-97
3303 ----------------------
3305 1. Typos in pcre.3 and comments in the source fixed.
3307 2. Applied a contributed patch to get rid of places where it used to remove
3308 'const' from variables, and fixed some signed/unsigned and uninitialized
3309 variable warnings.
3311 3. Added the "runtest" target to Makefile.
3313 4. Set default compiler flag to -O2 rather than just -O.
3316 Version 1.01 19-Nov-97
3317 ----------------------
3319 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3320 like /([ab]*)*/, that is, for classes with more than one character in them.
3322 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3323 as /((?>a*))*/ (a PCRE_EXTRA facility).
3326 Version 1.00 18-Nov-97
3327 ----------------------
3329 1. Added compile-time macros to support systems such as SunOS4 which don't have
3330 memmove() or strerror() but have other things that can be used instead.
3332 2. Arranged that "make clean" removes the executables.
3335 Version 0.99 27-Oct-97
3336 ----------------------
3338 1. Fixed bug in code for optimizing classes with only one character. It was
3339 initializing a 32-byte map regardless, which could cause it to run off the end
3340 of the memory it had got.
3342 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3345 Version 0.98 22-Oct-97
3346 ----------------------
3348 1. Fixed bug in code for handling temporary memory usage when there are more
3349 back references than supplied space in the ovector. This could cause segfaults.
3352 Version 0.97 21-Oct-97
3353 ----------------------
3355 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3357 2. Optimized negated single characters not to use a bit map.
3359 3. Brought error texts together as macro definitions; clarified some of them;
3360 fixed one that was wrong - it said "range out of order" when it meant "invalid
3361 escape sequence".
3363 4. Changed some char * arguments to const char *.
3365 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3367 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3368 pcretest.
3371 Version 0.96 16-Oct-97
3372 ----------------------
3374 1. Added a simple "pgrep" utility to the distribution.
3376 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3377 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3378 where "ddd" means "one or more decimal digits".
3380 3. Fixed serious bug. If a pattern had a back reference, but the call to
3381 pcre_exec() didn't supply a large enough ovector to record the related
3382 identifying subpattern, the match always failed. PCRE now remembers the number
3383 of the largest back reference, and gets some temporary memory in which to save
3384 the offsets during matching if necessary, in order to ensure that
3385 backreferences always work.
3387 4. Increased the compatibility with Perl in a number of ways:
3389 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3390 to request this handling. The option can be set at compile or exec time.
3392 (b) $ matches before a terminating newline by default; an option
3393 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3394 mode). The option can be set at compile or exec time.
3396 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3397 the same as Perl's. If the decimal number it represents is less than 10
3398 or there aren't that many previous left capturing parentheses, an octal
3399 escape is read. Inside a character class, it's always an octal escape,
3400 even if it is a single digit.
3402 (d) An escaped but undefined alphabetic character is taken as a literal,
3403 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3404 escapes.
3406 (e) {0} is now permitted. (The previous item is removed from the compiled
3407 pattern).
3409 5. Changed all the names of code files so that the basic parts are no longer
3410 than 10 characters, and abolished the teeny "globals.c" file.
3412 6. Changed the handling of character classes; they are now done with a 32-byte
3413 bit map always.
3415 7. Added the -d and /D options to pcretest to make it possible to look at the
3416 internals of compilation without having to recompile pcre.
3419 Version 0.95 23-Sep-97
3420 ----------------------
3422 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3423 \x20 at the start of a run of normal characters. These were being treated as
3424 real characters, instead of the source characters being re-checked.
3427 Version 0.94 18-Sep-97
3428 ----------------------
3430 1. The functions are now thread-safe, with the caveat that the global variables
3431 containing pointers to malloc() and free() or alternative functions are the
3432 same for all threads.
3434 2. Get pcre_study() to generate a bitmap of initial characters for non-
3435 anchored patterns when this is possible, and use it if passed to pcre_exec().
3438 Version 0.93 15-Sep-97
3439 ----------------------
3441 1. /(b)|(:+)/ was computing an incorrect first character.
3443 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3444 but not actually doing anything yet.
3446 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3447 as Perl does (e.g. [-az] or [az-]).
3449 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3450 all possible positions.
3452 5. Split up into different modules to avoid including unneeded functions in a
3453 compiled binary. However, compile and exec are still in one module. The "study"
3454 function is split off.
3456 6. The character tables are now in a separate module whose source is generated
3457 by an auxiliary program - but can then be edited by hand if required. There are
3458 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3459 toupper() in the code.
3461 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3462 make them global. Abolish the function for setting them, as the caller can now
3463 set them directly.
3466 Version 0.92 11-Sep-97
3467 ----------------------
3469 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3470 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3472 2. Caseless matching was not working in character classes if the characters in
3473 the pattern were in upper case.
3475 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3477 4. Make PCRE_ANCHORED public and accept as a compile option.
3479 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3480 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3481 pass them.
3483 6. Give an error if bad option bits passed at compile or run time.
3485 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3486 pcretest to cause it to pass that flag.
3488 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3489 options, and the first character, if set.
3491 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3494 Version 0.91 10-Sep-97
3495 ----------------------
3497 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3498 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3500 2. PCRE was looping on encountering an indefinitely repeated back reference to
3501 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3502 Perl does - treats the match as successful.
3504 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5