ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 319 - (show annotations)
Mon Jan 28 15:27:45 2008 UTC (12 years, 11 months ago) by ph10
File size: 150740 byte(s)
File tidy-up for 7.6 release.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.6 28-Jan-08
5 ---------------------
7 1. A character class containing a very large number of characters with
8 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
9 overflow.
11 2. Patch to cut out the "long long" test in pcrecpp_unittest when
12 HAVE_LONG_LONG is not defined.
14 3. Applied Christian Ehrlicher's patch to update the CMake build files to
15 bring them up to date and include new features. This patch includes:
17 - Fixed PH's badly added libz and libbz2 support.
18 - Fixed a problem with static linking.
19 - Added pcredemo. [But later removed - see 7 below.]
20 - Fixed dftables problem and added an option.
21 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
23 - Added readline support for pcretest.
24 - Added an listing of the option settings after cmake has run.
26 4. A user submitted a patch to Makefile that makes it easy to create
27 "pcre.dll" under mingw when using Configure/Make. I added stuff to
28 Makefile.am that cause it to include this special target, without
29 affecting anything else. Note that the same mingw target plus all
30 the other distribution libraries and programs are now supported
31 when configuring with CMake (see 6 below) instead of with
32 Configure/Make.
34 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
35 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
36 exported in the Windows port". It has not yet been confirmed that the patch
37 solves the problem, but it does no harm.
39 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
40 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
41 with CMake, and also correct the comment about stack recursion.
43 7. Remove the automatic building of pcredemo from the ./configure system and
44 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
45 of a program that users should build themselves after PCRE is installed, so
46 building it automatically is not really right. What is more, it gave
47 trouble in some build environments.
49 8. Further tidies to CMakeLists.txt from Sheri and Christian.
52 Version 7.5 10-Jan-08
53 ---------------------
55 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
56 values in parens when parsing an RE using the C++ wrapper."
58 2. Negative specials like \S did not work in character classes in UTF-8 mode.
59 Characters greater than 255 were excluded from the class instead of being
60 included.
62 3. The same bug as (2) above applied to negated POSIX classes such as
63 [:^space:].
65 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
66 defined or documented. It seems to have been a typo for PCRE_STATIC, so
67 I have changed it.
69 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
70 first named subpattern) and a construct such as (?&a) would reference the
71 first named subpattern whose name started with "a" (in other words, the
72 length check was missing). Both these problems are fixed. "Subpattern name
73 expected" is now given for (?&) (a zero-length name), and this patch also
74 makes it give the same error for \k'' (previously it complained that that
75 was a reference to a non-existent subpattern).
77 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
78 this is right because (?- can be followed by option settings as well as by
79 digits. I have, however, made the messages clearer.
81 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
82 than the number used in the conditional) now cause a compile-time error.
83 This is actually not compatible with Perl, which accepts such patterns, but
84 treats the conditional as always being FALSE (as PCRE used to), but it
85 seems to me that giving a diagnostic is better.
87 8. Change "alphameric" to the more common word "alphanumeric" in comments
88 and messages.
90 9. Fix two occurrences of "backslash" in comments that should have been
91 "backspace".
93 10. Remove two redundant lines of code that can never be obeyed (their function
94 was moved elsewhere).
96 11. The program that makes PCRE's Unicode character property table had a bug
97 which caused it to generate incorrect table entries for sequences of
98 characters that have the same character type, but are in different scripts.
99 It amalgamated them into a single range, with the script of the first of
100 them. In other words, some characters were in the wrong script. There were
101 thirteen such cases, affecting characters in the following ranges:
103 U+002b0 - U+002c1
104 U+0060c - U+0060d
105 U+0061e - U+00612
106 U+0064b - U+0065e
107 U+0074d - U+0076d
108 U+01800 - U+01805
109 U+01d00 - U+01d77
110 U+01d9b - U+01dbf
111 U+0200b - U+0200f
112 U+030fc - U+030fe
113 U+03260 - U+0327f
114 U+0fb46 - U+0fbb1
115 U+10450 - U+1049d
117 12. The -o option (show only the matching part of a line) for pcregrep was not
118 compatible with GNU grep in that, if there was more than one match in a
119 line, it showed only the first of them. It now behaves in the same way as
120 GNU grep.
122 13. If the -o and -v options were combined for pcregrep, it printed a blank
123 line for every non-matching line. GNU grep prints nothing, and pcregrep now
124 does the same. The return code can be used to tell if there were any
125 non-matching lines.
127 14. Added --file-offsets and --line-offsets to pcregrep.
129 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
130 infinitely looping recursion. The bug was that positive lookaheads were not
131 being skipped when checking for a possible empty match (negative lookaheads
132 and both kinds of lookbehind were skipped).
134 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
135 inclusion of <windows.h> to before rather than after the definition of
136 INVALID_FILE_ATTRIBUTES (patch from David Byron).
138 17. Specifying a possessive quantifier with a specific limit for a Unicode
139 character property caused pcre_compile() to compile bad code, which led at
140 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
141 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
142 caused the error; without that there was no problem.
144 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
146 19. Added --enable-pcretest-libreadline.
148 20. In pcrecpp.cc, the variable 'count' was incremented twice in
149 RE::GlobalReplace(). As a result, the number of replacements returned was
150 double what it should be. I removed one of the increments, but Craig sent a
151 later patch that removed the other one (the right fix) and added unit tests
152 that check the return values (which was not done before).
154 21. Several CMake things:
156 (1) Arranged that, when cmake is used on Unix, the libraries end up with
157 the names libpcre and libpcreposix, not just pcre and pcreposix.
159 (2) The above change means that pcretest and pcregrep are now correctly
160 linked with the newly-built libraries, not previously installed ones.
164 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
165 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
166 UTF-8 newline character). The key issue is that the pattern starts .*;
167 this means that the match must be either at the beginning, or after a
168 newline. The bug was in the code for advancing after a failed match and
169 checking that the new position followed a newline. It was not taking
170 account of UTF-8 characters correctly.
172 23. PCRE was behaving differently from Perl in the way it recognized POSIX
173 character classes. PCRE was not treating the sequence [:...:] as a
174 character class unless the ... were all letters. Perl, however, seems to
175 allow any characters between [: and :], though of course it rejects as
176 unknown any "names" that contain non-letters, because all the known class
177 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
178 for example, whereas PCRE did not - it did not recognize a POSIX character
179 class. This seemed a bit dangerous, so the code has been changed to be
180 closer to Perl. The behaviour is not identical to Perl, because PCRE will
181 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
182 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
183 Perl does, and where it didn't before.
185 24. Rewrite so as to remove the single use of %n from pcregrep because in some
186 Windows environments %n is disabled by default.
189 Version 7.4 21-Sep-07
190 ---------------------
192 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
193 means that a class such as [\s] counted as "explicit reference to CR or
194 LF". That isn't really right - the whole point of the change was to try to
195 help when there was an actual mention of one of the two characters. So now
196 the change happens only if \r or \n (or a literal CR or LF) character is
197 encountered.
199 2. The 32-bit options word was also used for 6 internal flags, but the numbers
200 of both had grown to the point where there were only 3 bits left.
201 Fortunately, there was spare space in the data structure, and so I have
202 moved the internal flags into a new 16-bit field to free up more option
203 bits.
205 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
206 but did not set the internal JCHANGED flag - either of these is enough to
207 control the way the "get" function works - but the PCRE_INFO_JCHANGED
208 facility is supposed to tell if (?J) was ever used, so now (?J) at the
209 start sets both bits.
211 4. Added options (at build time, compile time, exec time) to change \R from
212 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
214 5. doc/pcresyntax.html was missing from the distribution.
216 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
217 compatibility, even though it is no longer used.
219 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
220 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
221 windows.h file is present (where different names are used). [This was
222 reversed later after testing - see 16 below.]
224 8. Changed all #include <config.h> to #include "config.h". There were also
225 some further <pcre.h> cases that I changed to "pcre.h".
227 9. When pcregrep was used with the --colour option, it missed the line ending
228 sequence off the lines that it output.
230 10. It was pointed out to me that arrays of string pointers cause lots of
231 relocations when a shared library is dynamically loaded. A technique of
232 using a single long string with a table of offsets can drastically reduce
233 these. I have refactored PCRE in four places to do this. The result is
234 dramatic:
236 Originally: 290
237 After changing UCP table: 187
238 After changing error message table: 43
239 After changing table of "verbs" 36
240 After changing table of Posix names 22
242 Thanks to the folks working on Gregex for glib for this insight.
244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
245 unicode-properties was also set.
247 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
249 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
250 checked only for CRLF.
252 14. Added casts to pcretest.c to avoid compiler warnings.
254 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
256 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
257 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
258 entirely. This removes changes made in 7 above.
260 17. The CMake files have been updated, and there is now more information about
261 building with CMake in the NON-UNIX-USE document.
264 Version 7.3 28-Aug-07
265 ---------------------
267 1. In the rejigging of the build system that eventually resulted in 7.1, the
268 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
269 brackets there is not right, since it causes compilers to look for an
270 installed pcre.h, not the version that is in the source that is being
271 compiled (which of course may be different). I have changed it back to:
273 #include "pcre.h"
275 I have a vague recollection that the change was concerned with compiling in
276 different directories, but in the new build system, that is taken care of
277 by the VPATH setting the Makefile.
279 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
280 when the subject happened to end in the byte 0x85 (e.g. if the last
281 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
282 characters but of course it shouldn't be taken as a newline when it is part
283 of another character. The bug was that, for an unlimited repeat of . in
284 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
285 characters when looking for a newline.
287 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
289 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
290 in debug output.
292 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
293 long printing in the pcrecpp unittest when running under MinGW.
295 6. ESC_K was left out of the EBCDIC table.
297 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
298 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
299 limit also applies to "virtual nesting" when a pattern is recursive, and in
300 this case 1000 isn't so big. I have been able to remove this limit at the
301 expense of backing off one optimization in certain circumstances. Normally,
302 when pcre_exec() would call its internal match() function recursively and
303 immediately return the result unconditionally, it uses a "tail recursion"
304 feature to save stack. However, when a subpattern that can match an empty
305 string has an unlimited repetition quantifier, it no longer makes this
306 optimization. That gives it a stack frame in which to save the data for
307 checking that an empty string has been matched. Previously this was taken
308 from the 1000-entry workspace that had been reserved. So now there is no
309 explicit limit, but more stack is used.
311 8. Applied Daniel's patches to solve problems with the import/export magic
312 syntax that is required for Windows, and which was going wrong for the
313 pcreposix and pcrecpp parts of the library. These were overlooked when this
314 problem was solved for the main library.
316 9. There were some crude static tests to avoid integer overflow when computing
317 the size of patterns that contain repeated groups with explicit upper
318 limits. As the maximum quantifier is 65535, the maximum group length was
319 set at 30,000 so that the product of these two numbers did not overflow a
320 32-bit integer. However, it turns out that people want to use groups that
321 are longer than 30,000 bytes (though not repeat them that many times).
322 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
323 made it possible to implement the integer overflow checks in a much more
324 dynamic way, which I have now done. The artificial limitation on group
325 length has been removed - we now have only the limit on the total length of
326 the compiled pattern, which depends on the LINK_SIZE setting.
328 10. Fixed a bug in the documentation for get/copy named substring when
329 duplicate names are permitted. If none of the named substrings are set, the
330 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
331 empty string.
333 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
334 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
335 because the ] is interpreted as the first data character and the
336 terminating ] is not found. PCRE has been made compatible with Perl in this
337 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
338 cause memory overwriting.
340 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
341 string has been matched (to stop an infinite loop). It was not recognizing
342 a conditional subpattern that could match an empty string if that
343 subpattern was within another subpattern. For example, it looped when
344 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
345 condition was not nested. This bug has been fixed.
347 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
348 past the start of the subject in the presence of bytes with the top bit
349 set, for example "\x8aBCD".
351 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
352 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
354 14. Optimized (?!) to (*FAIL).
356 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
357 This restricts code points to be within the range 0 to 0x10FFFF, excluding
358 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
359 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
360 does: it's just the validity check that is more restrictive.
362 16. Inserted checks for integer overflows during escape sequence (backslash)
363 processing, and also fixed erroneous offset values for syntax errors during
364 backslash processing.
366 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
367 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
369 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
370 caused an overrun.
372 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
373 something other than just ASCII characters) inside a group that had an
374 unlimited repeat caused a loop at compile time (while checking to see
375 whether the group could match an empty string).
377 20. Debugging a pattern containing \p or \P could cause a crash. For example,
378 [\P{Any}] did so. (Error in the code for printing property names.)
380 21. An orphan \E inside a character class could cause a crash.
382 22. A repeated capturing bracket such as (A)? could cause a wild memory
383 reference during compilation.
385 23. There are several functions in pcre_compile() that scan along a compiled
386 expression for various reasons (e.g. to see if it's fixed length for look
387 behind). There were bugs in these functions when a repeated \p or \P was
388 present in the pattern. These operators have additional parameters compared
389 with \d, etc, and these were not being taken into account when moving along
390 the compiled data. Specifically:
392 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
393 length.
395 (b) An item such as \pL+ within a repeated group could cause crashes or
396 loops.
398 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
399 "reference to non-existent subpattern" error.
401 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
403 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
404 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
406 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
408 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
409 character were causing crashes (broken optimization).
411 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
412 \p or \P) caused a compile-time loop.
414 28. More problems have arisen in unanchored patterns when CRLF is a valid line
415 break. For example, the unstudied pattern [\r\n]A does not match the string
416 "\r\nA" because change 7.0/46 below moves the current point on by two
417 characters after failing to match at the start. However, the pattern \nA
418 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
419 the same is true. There doesn't seem any very clean way out of this, but
420 what I have chosen to do makes the common cases work: PCRE now takes note
421 of whether there can be an explicit match for \r or \n anywhere in the
422 pattern, and if so, 7.0/46 no longer applies. As part of this change,
423 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
424 pattern has explicit CR or LF references.
426 29. Added (*CR) etc for changing newline setting at start of pattern.
429 Version 7.2 19-Jun-07
430 ---------------------
432 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
433 which is apparently normally available under Windows.
435 2. Re-jig the pcregrep tests with different newline settings in an attempt
436 to make them independent of the local environment's newline setting.
438 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
440 4. Some of the "internals" tests were previously cut out when the link size
441 was not 2, because the output contained actual offsets. The recent new
442 "Z" feature of pcretest means that these can be cut out, making the tests
443 usable with all link sizes.
445 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
446 stack recursion. This gives a massive performance boost under BSD, but just
447 a small improvement under Linux. However, it saves one field in the frame
448 in all cases.
450 6. Added more features from the forthcoming Perl 5.10:
452 (a) (?-n) (where n is a string of digits) is a relative subroutine or
453 recursion call. It refers to the nth most recently opened parentheses.
455 (b) (?+n) is also a relative subroutine call; it refers to the nth next
456 to be opened parentheses.
458 (c) Conditions that refer to capturing parentheses can be specified
459 relatively, for example, (?(-2)... or (?(+3)...
461 (d) \K resets the start of the current match so that everything before
462 is not part of it.
464 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
466 (f) \g{name} is another synonym - part of Perl 5.10's unification of
467 reference syntax.
469 (g) (?| introduces a group in which the numbering of parentheses in each
470 alternative starts with the same number.
472 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
474 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
477 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
478 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
479 for detecting groups that can match an empty string.
481 9. A pattern with a very large number of alternatives (more than several
482 hundred) was running out of internal workspace during the pre-compile
483 phase, where pcre_compile() figures out how much memory will be needed. A
484 bit of new cunning has reduced the workspace needed for groups with
485 alternatives. The 1000-alternative test pattern now uses 12 bytes of
486 workspace instead of running out of the 4096 that are available.
488 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
490 11. Applied patch from Google to remove an optimization that didn't quite work.
491 The report of the bug said:
493 pcrecpp::RE("a*").FullMatch("aaa") matches, while
494 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
495 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
497 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
498 it matched the wrong number of bytes.
501 Version 7.1 24-Apr-07
502 ---------------------
504 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
505 that is more "standard", making use of automake and other Autotools. There
506 is some re-arrangement of the files and adjustment of comments consequent
507 on this.
509 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
510 for recursive directory scanning broke on some systems because the files
511 are not scanned in any specific order and on different systems the order
512 was different. A call to "sort" has been inserted into RunGrepTest for the
513 approprate test as a short-term fix. In the longer term there may be an
514 alternative.
516 3. I had an email from Eric Raymond about problems translating some of PCRE's
517 man pages to HTML (despite the fact that I distribute HTML pages, some
518 people do their own conversions for various reasons). The problems
519 concerned the use of low-level troff macros .br and .in. I have therefore
520 removed all such uses from the man pages (some were redundant, some could
521 be replaced by .nf/.fi pairs). The 132html script that I use to generate
522 HTML has been updated to handle .nf/.fi and to complain if it encounters
523 .br or .in.
525 4. Updated comments in configure.ac that get placed in config.h.in and also
526 arranged for config.h to be included in the distribution, with the name
527 config.h.generic, for the benefit of those who have to compile without
528 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
530 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
531 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
532 makevp.bat for the current PCRE, using the additional files
533 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
535 6. A Windows user reported a minor discrepancy with test 2, which turned out
536 to be caused by a trailing space on an input line that had got lost in his
537 copy. The trailing space was an accident, so I've just removed it.
539 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
540 that is needed.
542 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
543 as "const" (a) because they are and (b) because it helps the PHP
544 maintainers who have recently made a script to detect big data structures
545 in the php code that should be moved to the .rodata section. I remembered
546 to update Builducptable as well, so it won't revert if ucptable.h is ever
547 re-created.
549 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
550 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
551 order to be able to cut out the UTF-8 tables in the latter when UTF-8
552 support is not required. This saves 1.5-2K of code, which is important in
553 some applications.
555 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
556 so as not to refer to the tables, even though these functions will never be
557 called when UTF-8 support is disabled. Otherwise there are problems with a
558 shared library.
560 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
562 (a) It was defining its arguments as char * instead of void *.
564 (b) It was assuming that all moves were upwards in memory; this was true
565 a long time ago when I wrote it, but is no longer the case.
567 The emulated memove() is provided for those environments that have neither
568 memmove() nor bcopy(). I didn't think anyone used it these days, but that
569 is clearly not the case, as these two bugs were recently reported.
571 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
572 and Detrail to create the HTML documentation, the .txt form of the man
573 pages, and it removes trailing spaces from listed files. It also creates
574 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
575 case, it wraps all the #defines with #ifndefs. This script should be run
576 before "make dist".
578 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
579 with Unicode property support.
581 (a) For a maximizing quantifier, if the two different cases of the
582 character were of different lengths in their UTF-8 codings (there are
583 some cases like this - I found 11), and the matching function had to
584 back up over a mixture of the two cases, it incorrectly assumed they
585 were both the same length.
587 (b) When PCRE was configured to use the heap rather than the stack for
588 recursion during matching, it was not correctly preserving the data for
589 the other case of a UTF-8 character when checking ahead for a match
590 while processing a minimizing repeat. If the check also involved
591 matching a wide character, but failed, corruption could cause an
592 erroneous result when trying to check for a repeat of the original
593 character.
595 13. Some tidying changes to the testing mechanism:
597 (a) The RunTest script now detects the internal link size and whether there
598 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
599 values substituted by "configure". (The RunGrepTest script already did
600 this for UTF-8.) The configure.ac script no longer substitutes the
601 relevant variables.
603 (b) The debugging options /B and /D in pcretest show the compiled bytecode
604 with length and offset values. This means that the output is different
605 for different internal link sizes. Test 2 is skipped for link sizes
606 other than 2 because of this, bypassing the problem. Unfortunately,
607 there was also a test in test 3 (the locale tests) that used /B and
608 failed for link sizes other than 2. Rather than cut the whole test out,
609 I have added a new /Z option to pcretest that replaces the length and
610 offset values with spaces. This is now used to make test 3 independent
611 of link size. (Test 2 will be tidied up later.)
613 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
614 segmentation fault instead of returning the appropriate error message.
616 15. In multiline mode when the newline sequence was set to "any", the pattern
617 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
618 This doesn't seem right; it now treats the CRLF combination as the line
619 ending, and so does not match in that case. It's only a pattern such as ^$
620 that would hit this one: something like ^ABC$ would have failed after \r
621 and then tried again after \r\n.
623 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
624 in an attempt to make files that differ only in their line terminators
625 compare equal. This works on Linux.
627 17. Under certain error circumstances pcregrep might try to free random memory
628 as it exited. This is now fixed, thanks to valgrind.
630 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
631 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
632 was because its rules for how to advance for /g after matching an empty
633 string at the end of a line did not allow for this case. They now check for
634 it specially.
636 20. pcretest is supposed to handle patterns and data of any length, by
637 extending its buffers when necessary. It was getting this wrong when the
638 buffer for a data line had to be extended.
640 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
641 CRLF as a newline sequence.
643 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
644 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
645 I have nevertheless tidied it up.
647 23. Added some casts to kill warnings from HP-UX ia64 compiler.
649 24. Added a man page for pcre-config.
652 Version 7.0 19-Dec-06
653 ---------------------
655 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
656 moving to gcc 4.1.1.
658 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
659 sys/time.h, which is documented as needed for this function. It doesn't
660 seem to matter on Linux, but it showed up on some releases of OS X.
662 3. It seems that there are systems where bytes whose values are greater than
663 127 match isprint() in the "C" locale. The "C" locale should be the
664 default when a C program starts up. In most systems, only ASCII printing
665 characters match isprint(). This difference caused the output from pcretest
666 to vary, making some of the tests fail. I have changed pcretest so that:
668 (a) When it is outputting text in the compiled version of a pattern, bytes
669 other than 32-126 are always shown as hex escapes.
671 (b) When it is outputting text that is a matched part of a subject string,
672 it does the same, unless a different locale has been set for the match
673 (using the /L modifier). In this case, it uses isprint() to decide.
675 4. Fixed a major bug that caused incorrect computation of the amount of memory
676 required for a compiled pattern when options that changed within the
677 pattern affected the logic of the preliminary scan that determines the
678 length. The relevant options are -x, and -i in UTF-8 mode. The result was
679 that the computed length was too small. The symptoms of this bug were
680 either the PCRE error "internal error: code overflow" from pcre_compile(),
681 or a glibc crash with a message such as "pcretest: free(): invalid next
682 size (fast)". Examples of patterns that provoked this bug (shown in
683 pcretest format) are:
685 /(?-x: )/x
686 /(?x)(?-x: \s*#\s*)/
687 /((?i)[\x{c0}])/8
688 /(?i:[\x{c0}])/8
690 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
691 is now done differently.
693 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
694 wrapper classes; (b) implement a new function in the C++ scanner that is
695 more efficient than the old way of doing things because it avoids levels of
696 recursion in the regex matching; (c) add a paragraph to the documentation
697 for the FullMatch() function.
699 6. The escape sequence \n was being treated as whatever was defined as
700 "newline". Not only was this contrary to the documentation, which states
701 that \n is character 10 (hex 0A), but it also went horribly wrong when
702 "newline" was defined as CRLF. This has been fixed.
704 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
705 was being set to -1 for the "end of line" case (supposedly a value that no
706 character can have). Though this value is never used (the check for end of
707 line is "zero bytes in current character"), it caused compiler complaints.
708 I've changed it to 0xffffffff.
710 8. In pcre_version.c, the version string was being built by a sequence of
711 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
712 string (as it is for production releases) called a macro with an empty
713 argument. The C standard says the result of this is undefined. The gcc
714 compiler treats it as an empty string (which was what was wanted) but it is
715 reported that Visual C gives an error. The source has been hacked around to
716 avoid this problem.
718 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
719 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
720 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
721 of them did).
723 10. Originally, pcretest opened its input and output without "b"; then I was
724 told that "b" was needed in some environments, so it was added for release
725 5.0 to both the input and output. (It makes no difference on Unix-like
726 systems.) Later I was told that it is wrong for the input on Windows. I've
727 now abstracted the modes into two macros, to make it easier to fiddle with
728 them, and removed "b" from the input mode under Windows.
730 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
732 12. Added -help and --help to pcretest as an official way of being reminded
733 of the options.
735 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
736 and pcrecpp.cc because they annoy compilers at high warning levels.
738 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
740 15. Fixed an occurrence of == in configure.ac that should have been = (shell
741 scripts are not C programs :-) and which was not noticed because it works
742 on Linux.
744 16. pcretest is supposed to handle any length of pattern and data line (as one
745 line or as a continued sequence of lines) by extending its input buffer if
746 necessary. This feature was broken for very long pattern lines, leading to
747 a string of junk being passed to pcre_compile() if the pattern was longer
748 than about 50K.
750 17. I have done a major re-factoring of the way pcre_compile() computes the
751 amount of memory needed for a compiled pattern. Previously, there was code
752 that made a preliminary scan of the pattern in order to do this. That was
753 OK when PCRE was new, but as the facilities have expanded, it has become
754 harder and harder to keep it in step with the real compile phase, and there
755 have been a number of bugs (see for example, 4 above). I have now found a
756 cunning way of running the real compile function in a "fake" mode that
757 enables it to compute how much memory it would need, while actually only
758 ever using a few hundred bytes of working memory and without too many
759 tests of the mode. This should make future maintenance and development
760 easier. A side effect of this work is that the limit of 200 on the nesting
761 depth of parentheses has been removed (though this was never a serious
762 limitation, I suspect). However, there is a downside: pcre_compile() now
763 runs more slowly than before (30% or more, depending on the pattern). I
764 hope this isn't a big issue. There is no effect on runtime performance.
766 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
767 newline (only possible for the last line of a file) and it was a
768 pattern that set a locale (followed by /Lsomething), pcretest crashed.
770 19. Added additional timing features to pcretest. (1) The -tm option now times
771 matching only, not compiling. (2) Both -t and -tm can be followed, as a
772 separate command line item, by a number that specifies the number of
773 repeats to use when timing. The default is 50000; this gives better
774 precision, but takes uncomfortably long for very large patterns.
776 20. Extended pcre_study() to be more clever in cases where a branch of a
777 subpattern has no definite first character. For example, (a*|b*)[cd] would
778 previously give no result from pcre_study(). Now it recognizes that the
779 first character must be a, b, c, or d.
781 21. There was an incorrect error "recursive call could loop indefinitely" if
782 a subpattern (or the entire pattern) that was being tested for matching an
783 empty string contained only one non-empty item after a nested subpattern.
784 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
785 incorrectly, because the \d was being skipped in the check.
787 22. The pcretest program now has a new pattern option /B and a command line
788 option -b, which is equivalent to adding /B to every pattern. This causes
789 it to show the compiled bytecode, without the additional information that
790 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
791 is the same as /B/I).
793 23. A new optimization is now able automatically to treat some sequences such
794 as a*b as a*+b. More specifically, if something simple (such as a character
795 or a simple class like \d) has an unlimited quantifier, and is followed by
796 something that cannot possibly match the quantified thing, the quantifier
797 is automatically "possessified".
799 24. A recursive reference to a subpattern whose number was greater than 39
800 went wrong under certain circumstances in UTF-8 mode. This bug could also
801 have affected the operation of pcre_study().
803 25. Realized that a little bit of performance could be had by replacing
804 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
806 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
808 27. Possessive quantifiers such as a++ were previously implemented by turning
809 them into atomic groups such as ($>a+). Now they have their own opcodes,
810 which improves performance. This includes the automatically created ones
811 from 23 above.
813 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
814 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
815 the first matched character to be a colon. This applied both to named and
816 numbered groups.
818 29. The ucpinternal.h header file was missing its idempotency #ifdef.
820 30. I was sent a "project" file called libpcre.a.dev which I understand makes
821 building PCRE on Windows easier, so I have included it in the distribution.
823 31. There is now a check in pcretest against a ridiculously large number being
824 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
825 loop, the loop is abandoned.
827 32. Forward references to subpatterns in conditions such as (?(2)...) where
828 subpattern 2 is defined later cause pcre_compile() to search forwards in
829 the pattern for the relevant set of parentheses. This search went wrong
830 when there were unescaped parentheses in a character class, parentheses
831 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
833 33. "Subroutine" calls and backreferences were previously restricted to
834 referencing subpatterns earlier in the regex. This restriction has now
835 been removed.
837 34. Added a number of extra features that are going to be in Perl 5.10. On the
838 whole, these are just syntactic alternatives for features that PCRE had
839 previously implemented using the Python syntax or my own invention. The
840 other formats are all retained for compatibility.
842 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
843 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
844 also .NET compatible.
846 (b) A recursion or subroutine call to a named group can now be defined as
847 (?&name) as well as (?P>name).
849 (c) A backreference to a named group can now be defined as \k<name> or
850 \k'name' as well as (?P=name). The new forms, as well as being in Perl
851 5.10, are also .NET compatible.
853 (d) A conditional reference to a named group can now use the syntax
854 (?(<name>) or (?('name') as well as (?(name).
856 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
857 groups (named and numbered) that are never evaluated inline, but can be
858 called as "subroutines" from elsewhere. In effect, the DEFINE condition
859 is always false. There may be only one alternative in such a group.
861 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
862 as the simple (?(R). The condition is true only if the most recent
863 recursion is that of the given number or name. It does not search out
864 through the entire recursion stack.
866 (g) The escape \gN or \g{N} has been added, where N is a positive or
867 negative number, specifying an absolute or relative reference.
869 35. Tidied to get rid of some further signed/unsigned compiler warnings and
870 some "unreachable code" warnings.
872 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
873 things, this adds five new scripts.
875 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
876 There were also incompatibilities regarding the handling of \Q..\E inside
877 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
878 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
880 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
881 matches an empty string, and forcibly breaks the loop. There were bugs in
882 this code in non-simple cases. For a pattern such as ^(a()*)* matched
883 against aaaa the result was just "a" rather than "aaaa", for example. Two
884 separate and independent bugs (that affected different cases) have been
885 fixed.
887 39. Refactored the code to abolish the use of different opcodes for small
888 capturing bracket numbers. This is a tidy that I avoided doing when I
889 removed the limit on the number of capturing brackets for 3.5 back in 2001.
890 The new approach is not only tidier, it makes it possible to reduce the
891 memory needed to fix the previous bug (38).
893 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
894 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
895 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
896 mode.
898 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
899 report.
901 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
902 copy construction and assignment in the C++ wrapper.
904 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
905 couple of bugs that could have given wrong results in the "--newline=crlf"
906 case.
908 44. Added a number of casts and did some reorganization of signed/unsigned int
909 variables following suggestions from Dair Grant. Also renamed the variable
910 "this" as "item" because it is a C++ keyword.
912 45. Arranged for dftables to add
914 #include "pcre_internal.h"
916 to pcre_chartables.c because without it, gcc 4.x may remove the array
917 definition from the final binary if PCRE is built into a static library and
918 dead code stripping is activated.
920 46. For an unanchored pattern, if a match attempt fails at the start of a
921 newline sequence, and the newline setting is CRLF or ANY, and the next two
922 characters are CRLF, advance by two characters instead of one.
925 Version 6.7 04-Jul-06
926 ---------------------
928 1. In order to handle tests when input lines are enormously long, pcretest has
929 been re-factored so that it automatically extends its buffers when
930 necessary. The code is crude, but this _is_ just a test program. The
931 default size has been increased from 32K to 50K.
933 2. The code in pcre_study() was using the value of the re argument before
934 testing it for NULL. (Of course, in any sensible call of the function, it
935 won't be NULL.)
937 3. The memmove() emulation function in pcre_internal.h, which is used on
938 systems that lack both memmove() and bcopy() - that is, hardly ever -
939 was missing a "static" storage class specifier.
941 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
942 containing an extended class (one that cannot be represented by a bitmap
943 because it contains high-valued characters or Unicode property items, e.g.
944 [\pZ]). Almost always one would set UTF-8 mode when processing such a
945 pattern, but PCRE should not loop if you do not (it no longer does).
946 [Detail: two cases were found: (a) a repeated subpattern containing an
947 extended class; (b) a recursive reference to a subpattern that followed a
948 previous extended class. It wasn't skipping over the extended class
949 correctly when UTF-8 mode was not set.]
951 5. A negated single-character class was not being recognized as fixed-length
952 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
953 compile error "lookbehind assertion is not fixed length".
955 6. The RunPerlTest auxiliary script was showing an unexpected difference
956 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
957 write a Perl script that can interpret lines of an input file either as
958 byte characters or as UTF-8, which is what "perltest" was being required to
959 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
960 can't do is switch easily at run time between having the "use utf8;" pragma
961 or not. In the end, I fudged it by using the RunPerlTest script to insert
962 "use utf8;" explicitly for the UTF-8 tests.
964 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
965 the end of the subject string, contrary to the documentation and to what
966 Perl does. This was true of both matching functions. Now it matches only at
967 the start of the subject and immediately after *internal* newlines.
969 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
970 a pointer to an int instead of a pointer to an unsigned long int. This
971 caused problems on 64-bit systems.
973 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
974 instance of the 'standard' template library not being so standard".
976 10. There was no check on the number of named subpatterns nor the maximum
977 length of a subpattern name. The product of these values is used to compute
978 the size of the memory block for a compiled pattern. By supplying a very
979 long subpattern name and a large number of named subpatterns, the size
980 computation could be caused to overflow. This is now prevented by limiting
981 the length of names to 32 characters, and the number of named subpatterns
982 to 10,000.
984 11. Subpatterns that are repeated with specific counts have to be replicated in
985 the compiled pattern. The size of memory for this was computed from the
986 length of the subpattern and the repeat count. The latter is limited to
987 65535, but there was no limit on the former, meaning that integer overflow
988 could in principle occur. The compiled length of a repeated subpattern is
989 now limited to 30,000 bytes in order to prevent this.
991 12. Added the optional facility to have named substrings with the same name.
993 13. Added the ability to use a named substring as a condition, using the
994 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
995 are numbers (not recommended). Forward references are permitted.
997 14. Added forward references in named backreferences (if you see what I mean).
999 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1000 pattern could run off the end of the subject. For example, the pattern
1001 "(?s)(.{1,5})"8 did this with the subject "ab".
1003 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1004 PCRE_CASELESS was set when matching characters that were quantified with ?
1005 or *.
1007 17. A character class other than a single negated character that had a minimum
1008 but no maximum quantifier - for example [ab]{6,} - was not handled
1009 correctly by pce_dfa_exec(). It would match only one character.
1011 18. A valid (though odd) pattern that looked like a POSIX character
1012 class but used an invalid character after [ (for example [[,abc,]]) caused
1013 pcre_compile() to give the error "Failed: internal error: code overflow" or
1014 in some cases to crash with a glibc free() error. This could even happen if
1015 the pattern terminated after [[ but there just happened to be a sequence of
1016 letters, a binary zero, and a closing ] in the memory that followed.
1018 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1019 over the years. Originally (before any Unicode support), just the bottom 8
1020 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1021 output from "man perlunicode" includes this:
1023 The regular expression compiler produces polymorphic opcodes. That
1024 is, the pattern adapts to the data and automatically switches to
1025 the Unicode character scheme when presented with Unicode data--or
1026 instead uses a traditional byte scheme when presented with byte
1027 data.
1029 Sadly, a wide octal escape does not cause a switch, and in a string with
1030 no other multibyte characters, these octal escapes are treated as before.
1031 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1032 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1033 Unicode string.
1035 I have not perpetrated such confusion in PCRE. Up till now, it took just
1036 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1037 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1038 translate to the appropriate multibyte character.
1040 29. Applied some refactoring to reduce the number of warnings from Microsoft
1041 and Borland compilers. This has included removing the fudge introduced
1042 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1043 a warning about an unused variable.
1045 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1046 characters since release 4.0, because Perl (from release 5.004) does not.
1047 [Or at least, is documented not to: some releases seem to be in conflict
1048 with the documentation.] However, when a pattern was studied with
1049 pcre_study() and all its branches started with \s, PCRE still included VT
1050 as a possible starting character. Of course, this did no harm; it just
1051 caused an unnecessary match attempt.
1053 22. Removed a now-redundant internal flag bit that recorded the fact that case
1054 dependency changed within the pattern. This was once needed for "required
1055 byte" processing, but is no longer used. This recovers a now-scarce options
1056 bit. Also moved the least significant internal flag bit to the most-
1057 significant bit of the word, which was not previously used (hangover from
1058 the days when it was an int rather than a uint) to free up another bit for
1059 the future.
1061 23. Added support for CRLF line endings as well as CR and LF. As well as the
1062 default being selectable at build time, it can now be changed at runtime
1063 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1064 specify that it is scanning data with non-default line endings.
1066 24. Changed the definition of CXXLINK to make it agree with the definition of
1067 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1069 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1070 recursions. This makes a big different to stack usage for some patterns.
1072 26. If a subpattern containing a named recursion or subroutine reference such
1073 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1074 the space required for the compiled pattern went wrong and gave too small a
1075 value. Depending on the environment, this could lead to "Failed: internal
1076 error: code overflow at offset 49" or "glibc detected double free or
1077 corruption" errors.
1079 27. Applied patches from Google (a) to support the new newline modes and (b) to
1080 advance over multibyte UTF-8 characters in GlobalReplace.
1082 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1083 difference for some implementation of PCRE in some Windows version.
1085 29. Added some extra testing facilities to pcretest:
1087 \q<number> in a data line sets the "match limit" value
1088 \Q<number> in a data line sets the "match recursion limt" value
1089 -S <number> sets the stack size, where <number> is in megabytes
1091 The -S option isn't available for Windows.
1094 Version 6.6 06-Feb-06
1095 ---------------------
1097 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1098 in pcreposix.h. I have copied the definition from pcre.h.
1100 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1101 because pcre.h is no longer a built file.
1103 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1104 not normally included in the compiled code.
1107 Version 6.5 01-Feb-06
1108 ---------------------
1110 1. When using the partial match feature with pcre_dfa_exec(), it was not
1111 anchoring the second and subsequent partial matches at the new starting
1112 point. This could lead to incorrect results. For example, with the pattern
1113 /1234/, partially matching against "123" and then "a4" gave a match.
1115 2. Changes to pcregrep:
1117 (a) All non-match returns from pcre_exec() were being treated as failures
1118 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1119 error message is output. Some extra information is given for the
1121 probably the only errors that are likely to be caused by users (by
1122 specifying a regex that has nested indefinite repeats, for instance).
1123 If there are more than 20 of these errors, pcregrep is abandoned.
1125 (b) A binary zero was treated as data while matching, but terminated the
1126 output line if it was written out. This has been fixed: binary zeroes
1127 are now no different to any other data bytes.
1129 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1130 used to set a locale for matching. The --locale=xxxx long option has
1131 been added (no short equivalent) to specify a locale explicitly on the
1132 pcregrep command, overriding the environment variables.
1134 (d) When -B was used with -n, some line numbers in the output were one less
1135 than they should have been.
1137 (e) Added the -o (--only-matching) option.
1139 (f) If -A or -C was used with -c (count only), some lines of context were
1140 accidentally printed for the final match.
1142 (g) Added the -H (--with-filename) option.
1144 (h) The combination of options -rh failed to suppress file names for files
1145 that were found from directory arguments.
1147 (i) Added the -D (--devices) and -d (--directories) options.
1149 (j) Added the -F (--fixed-strings) option.
1151 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1153 (l) Added the --colo(u)r option.
1155 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1156 is not present by default.
1158 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1159 items such as (?R) or (?1), when the recursion could match a number of
1160 alternatives. If it matched one of the alternatives, but subsequently,
1161 outside the recursion, there was a failure, the code tried to back up into
1162 the recursion. However, because of the way PCRE is implemented, this is not
1163 possible, and the result was an incorrect result from the match.
1165 In order to prevent this happening, the specification of recursion has
1166 been changed so that all such subpatterns are automatically treated as
1167 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1169 4. I had overlooked the fact that, in some locales, there are characters for
1170 which isalpha() is true but neither isupper() nor islower() are true. In
1171 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1172 and ordfeminine) are like this. This affected the treatment of \w and \W
1173 when they appeared in character classes, but not when they appeared outside
1174 a character class. The bit map for "word" characters is now created
1175 separately from the results of isalnum() instead of just taking it from the
1176 upper, lower, and digit maps. (Plus the underscore character, of course.)
1178 5. The above bug also affected the handling of POSIX character classes such as
1179 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1180 permanent tables. Instead, the bit maps for such a class were previously
1181 created as the appropriate unions of the upper, lower, and digit bitmaps.
1182 Now they are created by subtraction from the [[:word:]] class, which has
1183 its own bitmap.
1185 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1186 It is created by subtracting the vertical space characters (\x09, \x0a,
1187 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1188 subtraction was done in the overall bitmap for a character class, meaning
1189 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1190 be recognized. This bug has been fixed.
1192 7. Patches from the folks at Google:
1194 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1195 real life, but is still worth protecting against".
1197 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1198 regular expressions".
1200 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1201 have it.
1203 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1204 "configure" and the latter not, in order to fix a problem somebody had
1205 with compiling the Arg class on HP-UX.
1207 (e) Improve the error-handling of the C++ wrapper a little bit.
1209 (f) New tests for checking recursion limiting.
1211 8. The pcre_memmove() function, which is used only if the environment does not
1212 have a standard memmove() function (and is therefore rarely compiled),
1213 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1214 returning a result (though PCRE never actually uses the result).
1216 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1217 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1218 returned instead of calling malloc() with an overflowing number that would
1219 most likely cause subsequent chaos.
1221 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1223 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1224 with this option is matched, the nmatch and pmatch options of regexec() are
1225 ignored.
1227 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1228 provided in case anyone wants to the the POSIX interface with UTF-8
1229 strings.
1231 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1232 C++ linking (needed for some HP-UX environments).
1234 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1235 (unused parameter) and in the pcre_printint() function (omitted "default"
1236 switch label when the default is to do nothing).
1238 15. Added some code to make it possible, when PCRE is compiled as a C++
1239 library, to replace subject pointers for pcre_exec() with a smart pointer
1240 class, thus making it possible to process discontinuous strings.
1242 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1243 much the same function. They were added by different people who were trying
1244 to make PCRE easy to compile on non-Unix systems. It has been suggested
1245 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1246 for compiling on Windows systems. I have therefore replaced it with
1247 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1248 defaults to "extern" for C or "extern C" for C++, which works fine on
1249 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1250 SCOPE with something explicit in config.h. In addition:
1252 (a) pcreposix.h still had just "extern" instead of either of these macros;
1253 I have replaced it with PCRE_DATA_SCOPE.
1255 (b) Functions such as _pcre_xclass(), which are internal to the library,
1256 but external in the C sense, all had PCRE_EXPORT in their definitions.
1257 This is apparently wrong for the Windows case, so I have removed it.
1258 (It makes no difference on Unix-like systems.)
1260 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1261 of recursive calls to match(). This is different to MATCH_LIMIT because
1262 that limits the total number of calls to match(), not all of which increase
1263 the depth of recursion. Limiting the recursion depth limits the amount of
1264 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1265 when PCRE is compiled, and changed at run time. A patch from Google adds
1266 this functionality to the C++ interface.
1268 18. Changes to the handling of Unicode character properties:
1270 (a) Updated the table to Unicode 4.1.0.
1272 (b) Recognize characters that are not in the table as "Cn" (undefined).
1274 (c) I revised the way the table is implemented to a much improved format
1275 which includes recognition of ranges. It now supports the ranges that
1276 are defined in UnicodeData.txt, and it also amalgamates other
1277 characters into ranges. This has reduced the number of entries in the
1278 table from around 16,000 to around 3,000, thus reducing its size
1279 considerably. I realized I did not need to use a tree structure after
1280 all - a binary chop search is just as efficient. Having reduced the
1281 number of entries, I extended their size from 6 bytes to 8 bytes to
1282 allow for more data.
1284 (d) Added support for Unicode script names via properties such as \p{Han}.
1286 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1287 matching that character.
1289 20. When matching a repeated Unicode property with a minimum greater than zero,
1290 (for example \pL{2,}), PCRE could look past the end of the subject if it
1291 reached it while seeking the minimum number of characters. This could
1292 happen only if some of the characters were more than one byte long, because
1293 there is a check for at least the minimum number of bytes.
1295 21. Refactored the implementation of \p and \P so as to be more general, to
1296 allow for more different types of property in future. This has changed the
1297 compiled form incompatibly. Anybody with saved compiled patterns that use
1298 \p or \P will have to recompile them.
1300 22. Added "Any" and "L&" to the supported property types.
1302 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1303 but give a compile time error if the value is greater than 0xff.
1305 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1306 accidentally not being installed or uninstalled.
1308 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1309 made were to insert the current release number. This seemed silly, because
1310 it made things harder for people building PCRE on systems that don't run
1311 "configure". I have turned pcre.h into a distributed file, no longer built
1312 by "configure", with the version identification directly included. There is
1313 no longer a pcre.h.in file.
1315 However, this change necessitated a change to the pcre-config script as
1316 well. It is built from pcre-config.in, and one of the substitutions was the
1317 release number. I have updated configure.ac so that ./configure now finds
1318 the release number by grepping pcre.h.
1320 26. Added the ability to run the tests under valgrind.
1323 Version 6.4 05-Sep-05
1324 ---------------------
1326 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1327 "--" to be printed when multiple files were scanned, even when none of the
1328 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1329 consider it to be a bug, and have restored the previous behaviour.
1331 2. A couple of code tidies to get rid of compiler warnings.
1333 3. The pcretest program used to cheat by referring to symbols in the library
1334 whose names begin with _pcre_. These are internal symbols that are not
1335 really supposed to be visible externally, and in some environments it is
1336 possible to suppress them. The cheating is now confined to including
1337 certain files from the library's source, which is a bit cleaner.
1339 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1340 file's purpose clearer.
1342 5. Reorganized pcre_ucp_findchar().
1345 Version 6.3 15-Aug-05
1346 ---------------------
1348 1. The file libpcre.pc.in did not have general read permission in the tarball.
1350 2. There were some problems when building without C++ support:
1352 (a) If C++ support was not built, "make install" and "make test" still
1353 tried to test it.
1355 (b) There were problems when the value of CXX was explicitly set. Some
1356 changes have been made to try to fix these, and ...
1358 (c) --disable-cpp can now be used to explicitly disable C++ support.
1360 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1361 backslash in a target when C++ was disabled. This confuses some
1362 versions of "make", apparently. Using an intermediate variable solves
1363 this. (Same for CPP_LOBJ.)
1365 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1366 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1367 necessary on certain architectures.
1369 4. Added a setting of -export-symbols-regex to the link command to remove
1370 those symbols that are exported in the C sense, but actually are local
1371 within the library, and not documented. Their names all begin with
1372 "_pcre_". This is not a perfect job, because (a) we have to except some
1373 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1374 available (and never for static libraries). I have made a note to try to
1375 find a way round (a) in the future.
1378 Version 6.2 01-Aug-05
1379 ---------------------
1381 1. There was no test for integer overflow of quantifier values. A construction
1382 such as {1111111111111111} would give undefined results. What is worse, if
1383 a minimum quantifier for a parenthesized subpattern overflowed and became
1384 negative, the calculation of the memory size went wrong. This could have
1385 led to memory overwriting.
1387 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1389 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1390 operating environments where this matters.
1392 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1393 PCRE options from within the C++ wrapper.
1395 5. Named capturing subpatterns were not being correctly counted when a pattern
1396 was compiled. This caused two problems: (a) If there were more than 100
1397 such subpatterns, the calculation of the memory needed for the whole
1398 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1399 back references of the form \12, where the number was greater than 9, were
1400 not recognized as back references, even though there were sufficient
1401 previous subpatterns.
1403 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1404 versions of gcc, e.g. 2.95.4.
1407 Version 6.1 21-Jun-05
1408 ---------------------
1410 1. There was one reference to the variable "posix" in pcretest.c that was not
1411 surrounded by "#if !defined NOPOSIX".
1413 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1414 the cross-check on the old pcre_info() function, for the benefit of the
1415 cut-down version of PCRE that is currently imported into Exim.
1417 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1418 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1419 patterns that start (?i) but I don't think that matters. The use of (?i) is
1420 just an example; this all applies to the other options as well.
1422 4. Since libtool seems to echo the compile commands it is issuing, the output
1423 from "make" can be reduced a bit by putting "@" in front of each libtool
1424 compile command.
1426 5. Patch from the folks at Google for configure.in to be a bit more thorough
1427 in checking for a suitable C++ installation before trying to compile the
1428 C++ stuff. This should fix a reported problem when a compiler was present,
1429 but no suitable headers.
1431 6. The man pages all had just "PCRE" as their title. I have changed them to
1432 be the relevant file name. I have also arranged that these names are
1433 retained in the file doc/pcre.txt, which is a concatenation in text format
1434 of all the man pages except the little individual ones for each function.
1436 7. The NON-UNIX-USE file had not been updated for the different set of source
1437 files that come with release 6. I also added a few comments about the C++
1438 wrapper.
1441 Version 6.0 07-Jun-05
1442 ---------------------
1444 1. Some minor internal re-organization to help with my DFA experiments.
1446 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1447 didn't matter for the library itself when fully configured, but did matter
1448 when compiling without UCP support, or within Exim, where the ucp files are
1449 not imported.
1451 3. Refactoring of the library code to split up the various functions into
1452 different source modules. The addition of the new DFA matching code (see
1453 below) to a single monolithic source would have made it really too
1454 unwieldy, quite apart from causing all the code to be include in a
1455 statically linked application, when only some functions are used. This is
1456 relevant even without the DFA addition now that patterns can be compiled in
1457 one application and matched in another.
1459 The downside of splitting up is that there have to be some external
1460 functions and data tables that are used internally in different modules of
1461 the library but which are not part of the API. These have all had their
1462 names changed to start with "_pcre_" so that they are unlikely to clash
1463 with other external names.
1465 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1466 a different (DFA) algorithm. Although it is slower than the original
1467 function, it does have some advantages for certain types of matching
1468 problem.
1470 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1471 including restarting after a partial match.
1473 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1474 defined when compiling for Windows was sent to me. I have put it into the
1475 code, though I have no means of testing or verifying it.
1477 7. Added the pcre_refcount() auxiliary function.
1479 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1480 match before or at the first newline in the subject string. In pcretest,
1481 the /f option on a pattern can be used to set this.
1483 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1484 would behave wrongly. This has been present in PCRE since release 4.0.
1486 10. A number of changes to the pcregrep command:
1488 (a) Refactored how -x works; insert ^(...)$ instead of setting
1489 PCRE_ANCHORED and checking the length, in preparation for adding
1490 something similar for -w.
1492 (b) Added the -w (match as a word) option.
1494 (c) Refactored the way lines are read and buffered so as to have more
1495 than one at a time available.
1497 (d) Implemented a pcregrep test script.
1499 (e) Added the -M (multiline match) option. This allows patterns to match
1500 over several lines of the subject. The buffering ensures that at least
1501 8K, or the rest of the document (whichever is the shorter) is available
1502 for matching (and similarly the previous 8K for lookbehind assertions).
1504 (f) Changed the --help output so that it now says
1506 -w, --word-regex(p)
1508 instead of two lines, one with "regex" and the other with "regexp"
1509 because that confused at least one person since the short forms are the
1510 same. (This required a bit of code, as the output is generated
1511 automatically from a table. It wasn't just a text change.)
1513 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1514 option but starts with a hyphen. Could be a pattern or a path name
1515 starting with a hyphen, for instance.
1517 (h) "-" can be given as a file name to represent stdin.
1519 (i) When file names are being printed, "(standard input)" is used for
1520 the standard input, for compatibility with GNU grep. Previously
1521 "<stdin>" was used.
1523 (j) The option --label=xxx can be used to supply a name to be used for
1524 stdin when file names are being printed. There is no short form.
1526 (k) Re-factored the options decoding logic because we are going to add
1527 two more options that take data. Such options can now be given in four
1528 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1530 (l) Added the -A, -B, and -C options for requesting that lines of context
1531 around matches be printed.
1533 (m) Added the -L option to print the names of files that do not contain
1534 any matching lines, that is, the complement of -l.
1536 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1537 continue to scan other files.
1539 (o) The -s option was incorrectly implemented. For compatibility with other
1540 greps, it now suppresses the error message for a non-existent or non-
1541 accessible file (but not the return code). There is a new option called
1542 -q that suppresses the output of matching lines, which was what -s was
1543 previously doing.
1545 (p) Added --include and --exclude options to specify files for inclusion
1546 and exclusion when recursing.
1548 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1549 Hopefully, it now does.
1551 12. Missing cast in pcre_study().
1553 13. Added an "uninstall" target to the makefile.
1555 14. Replaced "extern" in the function prototypes in Makefile.in with
1556 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1557 world, but is set differently for Windows.
1559 15. Added a second compiling function called pcre_compile2(). The only
1560 difference is that it has an extra argument, which is a pointer to an
1561 integer error code. When there is a compile-time failure, this is set
1562 non-zero, in addition to the error test pointer being set to point to an
1563 error message. The new argument may be NULL if no error number is required
1564 (but then you may as well call pcre_compile(), which is now just a
1565 wrapper). This facility is provided because some applications need a
1566 numeric error indication, but it has also enabled me to tidy up the way
1567 compile-time errors are handled in the POSIX wrapper.
1569 16. Added VPATH=.libs to the makefile; this should help when building with one
1570 prefix path and installing with another. (Or so I'm told by someone who
1571 knows more about this stuff than I do.)
1573 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1574 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1575 match everything, including newlines. This is not POSIX-compatible, but
1576 somebody wanted the feature. From pcretest it can be activated by using
1577 both the P and the s flags.
1579 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1581 19. libpcre.pc was being incorrectly installed as executable.
1583 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1584 it now also looks for '\r' so that it will work unmodified on Windows.
1586 21. Added Google's contributed C++ wrapper to the distribution.
1588 22. Added some untidy missing memory free() calls in pcretest, to keep
1589 Electric Fence happy when testing.
1593 Version 5.0 13-Sep-04
1594 ---------------------
1596 1. Internal change: literal characters are no longer packed up into items
1597 containing multiple characters in a single byte-string. Each character
1598 is now matched using a separate opcode. However, there may be more than one
1599 byte in the character in UTF-8 mode.
1601 2. The pcre_callout_block structure has two new fields: pattern_position and
1602 next_item_length. These contain the offset in the pattern to the next match
1603 item, and its length, respectively.
1605 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1606 insertion of callouts before each pattern item. Added the /C option to
1607 pcretest to make use of this.
1609 4. On the advice of a Windows user, the lines
1611 #if defined(_WIN32) || defined(WIN32)
1612 _setmode( _fileno( stdout ), 0x8000 );
1613 #endif /* defined(_WIN32) || defined(WIN32) */
1615 have been added to the source of pcretest. This apparently does useful
1616 magic in relation to line terminators.
1618 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1619 for the benefit of those environments where the "b" makes a difference.
1621 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1622 to know about it. I have put a hack into configure.in that adds in code
1623 to set GCC=yes if CC=icc. This seems to end up at a point in the
1624 generated configure script that is early enough to affect the setting of
1625 compiler options, which is what is needed, but I have no means of testing
1626 whether it really works. (The user who reported this had patched the
1627 generated configure script, which of course I cannot do.)
1629 LATER: After change 22 below (new libtool files), the configure script
1630 seems to know about icc (and also ecc). Therefore, I have commented out
1631 this hack in configure.in.
1633 7. Added support for pkg-config (2 patches were sent in).
1635 8. Negated POSIX character classes that used a combination of internal tables
1636 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1637 [[:^ascii]]. Typically, they would match almost any characters. The other
1638 POSIX classes were not broken in this way.
1640 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1641 to find the match, as PCRE was deluded into thinking that the match had to
1642 start at the start point or following a newline. The same bug applied to
1643 patterns with negative forward assertions or any backward assertions
1644 preceding ".*" at the start, unless the pattern required a fixed first
1645 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1647 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1648 starting at the last subject character, bytes beyond the end of the subject
1649 string were read.
1651 11. Renamed the variable "class" as "classbits" to make life easier for C++
1652 users. (Previously there was a macro definition, but it apparently wasn't
1653 enough.)
1655 12. Added the new field "tables" to the extra data so that tables can be passed
1656 in at exec time, or the internal tables can be re-selected. This allows
1657 a compiled regex to be saved and re-used at a later time by a different
1658 program that might have everything at different addresses.
1660 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1661 -R library as well as a -L library.
1663 14. The debugging options of pcretest (-d on the command line or D on a
1664 pattern) showed incorrect output for anything following an extended class
1665 that contained multibyte characters and which was followed by a quantifier.
1667 15. Added optional support for general category Unicode character properties
1668 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1669 support. It adds about 90K to the size of the library. The meanings of the
1670 inbuilt class escapes such as \d and \s have NOT been changed.
1672 16. Updated pcredemo.c to include calls to free() to release the memory for the
1673 compiled pattern.
1675 17. The generated file chartables.c was being created in the source directory
1676 instead of in the building directory. This caused the build to fail if the
1677 source directory was different from the building directory, and was
1678 read-only.
1680 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1681 file. No doubt somebody will tell me if they don't make sense... Also added
1682 Dan Mooney's comments about building on OpenVMS.
1684 19. Added support for partial matching via the PCRE_PARTIAL option for
1685 pcre_exec() and the \P data escape in pcretest.
1687 20. Extended pcretest with 3 new pattern features:
1689 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1690 write the compiled pattern to the file whose name is "rest-of-line".
1691 This is a straight binary dump of the data, with the saved pointer to
1692 the character tables forced to be NULL. The study data, if any, is
1693 written too. After writing, pcretest reads a new pattern.
1695 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1696 compiled pattern from the given file. There must not be any
1697 occurrences of "<" in the file name (pretty unlikely); if there are,
1698 pcretest will instead treat the initial "<" as a pattern delimiter.
1699 After reading in the pattern, pcretest goes on to read data lines as
1700 usual.
1702 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1703 and 16-bit fields in a compiled pattern, to simulate a pattern that
1704 was compiled on a host of opposite endianness.
1706 21. The pcre-exec() function can now cope with patterns that were compiled on
1707 hosts of opposite endianness, with this restriction:
1709 As for any compiled expression that is saved and used later, the tables
1710 pointer field cannot be preserved; the extra_data field in the arguments
1711 to pcre_exec() should be used to pass in a tables address if a value
1712 other than the default internal tables were used at compile time.
1714 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1715 now diagnosed as an error. Previously, most of the time, a negative number
1716 would have been treated as zero, but if in addition "ovector" was passed as
1717 NULL, a crash could occur.
1719 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1720 new versions from the libtool 1.5 distribution (the last one is a copy of
1721 a file called libtool.m4). This seems to have fixed the need to patch
1722 "configure" to support Darwin 1.3 (which I used to do). However, I still
1723 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1724 workstation).
1726 24. Changed the PCRE licence to be the more standard "BSD" licence.
1729 Version 4.5 01-Dec-03
1730 ---------------------
1732 1. There has been some re-arrangement of the code for the match() function so
1733 that it can be compiled in a version that does not call itself recursively.
1734 Instead, it keeps those local variables that need separate instances for
1735 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1736 needs to "recurse". Keeping track of where control must go is done by means
1737 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1738 hide most of the details from the main code, and operates only if
1739 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1740 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1741 operating.
1743 To make it easier for callers to provide specially tailored get/free
1744 functions for this usage, two new functions, pcre_stack_malloc, and
1745 pcre_stack_free, are used. They are always called in strict stacking order,
1746 and the size of block requested is always the same.
1748 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1749 PCRE has been compiled to use the stack or the heap for recursion. The
1750 -C option of pcretest uses this to show which version is compiled.
1752 A new data escape \S, is added to pcretest; it causes the amounts of store
1753 obtained and freed by both kinds of malloc/free at match time to be added
1754 to the output.
1756 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1757 what's available on my current Linux desktop machine.
1759 3. When matching a UTF-8 string, the test for a valid string at the start has
1760 been extended. If start_offset is not zero, PCRE now checks that it points
1761 to a byte that is the start of a UTF-8 character. If not, it returns
1762 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1763 this is necessary because there may be backward assertions in the pattern.
1764 When matching the same subject several times, it may save resources to use
1765 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1767 4. The code for checking the validity of UTF-8 strings has been tightened so
1768 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1769 containing "overlong sequences".
1771 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1772 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1773 should have been "&", but it just so happened that all the cases this let
1774 through by mistake were picked up later in the function.
1776 6. I had used a variable called "isblank" - this is a C99 function, causing
1777 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1779 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1780 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1781 so the version is shown; (c) stop "make test" echoing "./RunTest".
1783 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1785 9. The prototype for memmove() for systems that don't have it was using
1786 size_t, but the inclusion of the header that defines size_t was later. I've
1787 moved the #includes for the C headers earlier to avoid this.
1789 10. Added some adjustments to the code to make it easier to compiler on certain
1790 special systems:
1792 (a) Some "const" qualifiers were missing.
1793 (b) Added the macro EXPORT before all exported functions; by default this
1794 is defined to be empty.
1795 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1796 that it reads its output file name as an argument instead of writing
1797 to the standard output and assuming this can be redirected.
1799 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1800 class containing characters with values greater than 255, PCRE compilation
1801 went into a loop.
1803 12. A recursive reference to a subpattern that was within another subpattern
1804 that had a minimum quantifier of zero caused PCRE to crash. For example,
1805 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1806 recursion. If the recursively-called subpattern itself had a zero repeat,
1807 that was OK.
1809 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1810 buffer into which it was copied (for escape processing) was still set at
1811 1024, so long lines caused crashes.
1813 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1814 "internal error: code overflow...". This applied to any character class
1815 that was followed by a possessive quantifier.
1817 15. Modified the Makefile to add libpcre.la as a prerequisite for
1818 libpcreposix.la because I was told this is needed for a parallel build to
1819 work.
1821 16. If a pattern that contained .* following optional items at the start was
1822 studied, the wrong optimizing data was generated, leading to matching
1823 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1824 matching string must start with a or b or c. The correct conclusion for
1825 this pattern is that a match can start with any character.
1828 Version 4.4 13-Aug-03
1829 ---------------------
1831 1. In UTF-8 mode, a character class containing characters with values between
1832 127 and 255 was not handled correctly if the compiled pattern was studied.
1833 In fixing this, I have also improved the studying algorithm for such
1834 classes (slightly).
1836 2. Three internal functions had redundant arguments passed to them. Removal
1837 might give a very teeny performance improvement.
1839 3. Documentation bug: the value of the capture_top field in a callout is *one
1840 more than* the number of the hightest numbered captured substring.
1842 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1843 in incorrectly linking with a previously installed version. They now link
1844 explicitly with libpcre.la.
1846 5. configure.in no longer needs to recognize Cygwin specially.
1848 6. A problem in pcre.in for Windows platforms is fixed.
1850 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1851 pcretest, it used to include the size of the study block as part of its
1852 output. Unfortunately, the structure contains a field that has a different
1853 size on different hardware architectures. This meant that the tests that
1854 showed this size failed. As the block is currently always of a fixed size,
1855 this information isn't actually particularly useful in pcretest output, so
1856 I have just removed it.
1858 8. Three pre-processor statements accidentally did not start in column 1.
1859 Sadly, there are *still* compilers around that complain, even though
1860 standard C has not required this for well over a decade. Sigh.
1862 9. In pcretest, the code for checking callouts passed small integers in the
1863 callout_data field, which is a void * field. However, some picky compilers
1864 complained about the casts involved for this on 64-bit systems. Now
1865 pcretest passes the address of the small integer instead, which should get
1866 rid of the warnings.
1868 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1869 both compile and run time, and gives an error if an invalid UTF-8 sequence
1870 is found. There is a option for disabling this check in cases where the
1871 string is known to be correct and/or the maximum performance is wanted.
1873 11. In response to a bug report, I changed one line in Makefile.in from
1875 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1876 to
1877 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1879 to look similar to other lines, but I have no way of telling whether this
1880 is the right thing to do, as I do not use Windows. No doubt I'll get told
1881 if it's wrong...
1884 Version 4.3 21-May-03
1885 ---------------------
1887 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1888 Makefile.
1890 2. Some refactoring to improve the quality of the code:
1892 (i) The utf8_table... variables are now declared "const".
1894 (ii) The code for \cx, which used the "case flipping" table to upper case
1895 lower case letters, now just substracts 32. This is ASCII-specific,
1896 but the whole concept of \cx is ASCII-specific, so it seems
1897 reasonable.
1899 (iii) PCRE was using its character types table to recognize decimal and
1900 hexadecimal digits in the pattern. This is silly, because it handles
1901 only 0-9, a-f, and A-F, but the character types table is locale-
1902 specific, which means strange things might happen. A private
1903 table is now used for this - though it costs 256 bytes, a table is
1904 much faster than multiple explicit tests. Of course, the standard
1905 character types table is still used for matching digits in subject
1906 strings against \d.
1908 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1909 ending in _t are). So I've renamed it as ESC_tee.
1911 3. The first argument for regexec() in the POSIX wrapper should have been
1912 defined as "const".
1914 4. Changed pcretest to use malloc() for its buffers so that they can be
1915 Electric Fenced for debugging.
1917 5. There were several places in the code where, in UTF-8 mode, PCRE would try
1918 to read one or more bytes before the start of the subject string. Often this
1919 had no effect on PCRE's behaviour, but in some circumstances it could
1920 provoke a segmentation fault.
1922 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1923 to try to read one or more bytes before the start of the subject string.
1925 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1926 UTF-8 support could misbehave in various ways if the subject string
1927 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1928 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1929 back over UTF-8 characters.)
1932 Version 4.2 14-Apr-03
1933 ---------------------
1935 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1937 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1938 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1939 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1940 [WIN_PREFIX]: new variable, "cyg" for Cygwin
1941 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1943 Note: automatic setting of the BUILD variables is not yet working
1944 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1945 compile-time but not at link-time
1946 [LINK]: use for linking executables only
1947 make different versions for Windows and non-Windows
1948 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1949 libraries
1950 [LINK_FOR_BUILD]: new variable
1951 [OBJEXT]: use throughout
1952 [EXEEXT]: use throughout
1953 <winshared>: new target
1954 <wininstall>: new target
1955 <dftables.o>: use native compiler
1956 <dftables>: use native linker
1957 <install>: handle Windows platform correctly
1958 <clean>: ditto
1959 <check>: ditto
1960 copy DLL to top builddir before testing
1962 As part of these changes, -no-undefined was removed again. This was reported
1963 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1964 in any case.
1966 3. Some tidies to get rid of compiler warnings:
1968 . In the match_data structure, match_limit was an unsigned long int, whereas
1969 match_call_count was an int. I've made them both unsigned long ints.
1971 . In pcretest the fact that a const uschar * doesn't automatically cast to
1972 a void * provoked a warning.
1974 . Turning on some more compiler warnings threw up some "shadow" variables
1975 and a few more missing casts.
1977 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1978 option, a class that contained a single character with a value between 128
1979 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1981 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1982 option, a class that contained several characters, but with at least one
1983 whose value was between 128 and 255 caused PCRE to crash.
1986 Version 4.1 12-Mar-03
1987 ---------------------
1989 1. Compiling with gcc -pedantic found a couple of places where casts were
1990 needed, and a string in dftables.c that was longer than standard compilers are
1991 required to support.
1993 2. Compiling with Sun's compiler found a few more places where the code could
1994 be tidied up in order to avoid warnings.
1996 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1997 first of these names is deprecated in the latest Autoconf in favour of the name
1998 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1999 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2000 analogy I have changed it to CFLAGS_FOR_BUILD.
2002 4. Added -no-undefined to the linking command in the Makefile, because this is
2003 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2004 linking step for the pcreposix library.
2006 5. PCRE was failing to diagnose the case of two named groups with the same
2007 name.
2009 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2010 literal character that is needed in the subject for a match, and scans along to
2011 ensure that it is present before embarking on the full matching process. This
2012 saves time in cases of nested unlimited repeats that are never going to match.
2013 Problem: the scan can take a lot of time if the subject is very long (e.g.
2014 megabytes), thus penalizing straightforward matches. It is now done only if the
2015 amount of subject to be scanned is less than 1000 bytes.
2017 7. A lesser problem with the same optimization is that it was recording the
2018 first character of an anchored pattern as "needed", thus provoking a search
2019 right along the subject, even when the first match of the pattern was going to
2020 fail. The "needed" character is now not set for anchored patterns, unless it
2021 follows something in the pattern that is of non-fixed length. Thus, it still
2022 fulfils its original purpose of finding quick non-matches in cases of nested
2023 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2026 Version 4.0 17-Feb-03
2027 ---------------------
2029 1. If a comment in an extended regex that started immediately after a meta-item
2030 extended to the end of string, PCRE compiled incorrect data. This could lead to
2031 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2033 2. Moved to autoconf 2.53 and libtool 1.4.2.
2035 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2036 the special perltest8 script is no longer needed - all the tests can be run
2037 from a single perltest script.
2039 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2040 by \s. It has now been removed in PCRE. This means it isn't recognized as
2041 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2042 class [:space:] *does* include VT, thereby creating a mess.
2044 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2045 space and tab.
2047 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2048 its new features into the main test script, reducing the number of scripts.
2050 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2051 were backward compatible, and made the (?i) apply to the whole pattern, as if
2052 /i were given. Now it behaves more logically, and applies the option setting
2053 only to what follows. PCRE has been changed to follow suit. However, if it
2054 finds options settings right at the start of the pattern, it extracts them into
2055 the global options, as before. Thus, they show up in the info data.
2057 8. Added support for the \Q...\E escape sequence. Characters in between are
2058 treated as literals. This is slightly different from Perl in that $ and @ are
2059 also handled as literals inside the quotes. In Perl, they will cause variable
2060 interpolation. Note the following examples:
2062 Pattern PCRE matches Perl matches
2064 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2065 \Qabc\$xyz\E abc\$xyz abc\$xyz
2066 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2068 For compatibility with Perl, \Q...\E sequences are recognized inside character
2069 classes as well as outside them.
2071 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2072 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2073 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2074 signed/unsigned warnings.
2076 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2077 option for pcretest, so I've replaced it by a simple function that does just
2078 that job.
2080 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2081 "pcregrep -".
2083 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2084 Java package. This provides some syntactic sugar for simple cases of what my
2085 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2086 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2087 item, you can use this simplified notation. Note that only makes sense with
2088 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2089 greediness, whatever the setting of the PCRE_UNGREEDY option.
2091 13. A change of greediness default within a pattern was not taking effect at
2092 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2093 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2094 was abstracted outside.
2096 14. PCRE now supports the \G assertion. It is true when the current matching
2097 position is at the start point of the match. This differs from \A when the
2098 starting offset is non-zero. Used with the /g option of pcretest (or similar
2099 code), it works in the same way as it does for Perl's /g option. If all
2100 alternatives of a regex begin with \G, the expression is anchored to the start
2101 match position, and the "anchored" flag is set in the compiled expression.
2103 15. Some bugs concerning the handling of certain option changes within patterns
2104 have been fixed. These applied to options other than (?ims). For example,
2105 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2106 been the other way round. Some of this was related to change 7 above.
2108 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2109 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2110 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2111 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2113 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2114 mode. Unlike ".", it always matches newline, whatever the setting of
2115 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2116 assertions. Perl allows it, but it doesn't (in general) work because it can't
2117 calculate the length of the lookbehind. At least, that's the case for Perl
2118 5.8.0 - I've been told they are going to document that it doesn't work in
2119 future.
2121 18. Added an error diagnosis for escapes that PCRE does not support: these are
2122 \L, \l, \N, \P, \p, \U, \u, and \X.
2124 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2125 reading past the end of the pattern in cases such as /[abcd/.
2127 20. PCRE was getting more memory than necessary for patterns with classes that
2128 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2130 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2131 compiling PCRE for use with Virtual Pascal.
2133 22. Small fix to the Makefile to make it work properly if the build is done
2134 outside the source tree.
2136 23. Added a new extension: a condition to go with recursion. If a conditional
2137 subpattern starts with (?(R) the "true" branch is used if recursion has
2138 happened, whereas the "false" branch is used only at the top level.
2140 24. When there was a very long string of literal characters (over 255 bytes
2141 without UTF support, over 250 bytes with UTF support), the computation of how
2142 much memory was required could be incorrect, leading to segfaults or other
2143 strange effects.
2145 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2146 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2147 there was a subsequent back reference to those brackets. This meant that, for
2148 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2149 possible to check for precisely this case. All we can do is abandon the
2150 optimization if .* occurs inside capturing brackets when there are any back
2151 references whatsoever. (See below for a better fix that came later.)
2153 26. The handling of the optimization for finding the first character of a
2154 non-anchored pattern, and for finding a character that is required later in the
2155 match were failing in some cases. This didn't break the matching; it just
2156 failed to optimize when it could. The way this is done has been re-implemented.
2158 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2160 28. Added a new feature that provides some of the functionality that Perl
2161 provides with (?{...}). The facility is termed a "callout". The way it is done
2162 in PCRE is for the caller to provide an optional function, by setting
2163 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2164 global variable. By default it is unset, which disables all calling out. To get
2165 the function called, the regex must include (?C) at appropriate points. This
2166 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2167 This provides a means of identifying different callout points. When PCRE
2168 reaches such a point in the regex, if pcre_callout has been set, the external
2169 function is called. It is provided with data in a structure called
2170 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2171 matching continues; if it returns a non-zero value, the match at the current
2172 point fails. However, backtracking will occur if possible. [This was changed
2173 later and other features added - see item 49 below.]
2175 29. pcretest is upgraded to test the callout functionality. It provides a
2176 callout function that displays information. By default, it shows the start of
2177 the match and the current position in the text. There are some new data escapes
2178 to vary what happens:
2180 \C+ in addition, show current contents of captured substrings
2181 \C- do not supply a callout function
2182 \C!n return 1 when callout number n is reached
2183 \C!n!m return 1 when callout number n is reached for the mth time
2185 30. If pcregrep was called with the -l option and just a single file name, it
2186 output "<stdin>" if a match was found, instead of the file name.
2188 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2189 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2190 pcre_exec(). This saves a malloc/free per call. The default value of
2191 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2192 when configuring.
2194 32. The default maximum size of a compiled pattern is 64K. There have been a
2195 few cases of people hitting this limit. The code now uses macros to handle the
2196 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2197 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2198 configuring. Tests 2 and 5 work only with 2-byte links because they output
2199 debugging information about compiled patterns.
2201 33. Internal code re-arrangements:
2203 (a) Moved the debugging function for printing out a compiled regex into
2204 its own source file (printint.c) and used #include to pull it into
2205 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2206 separate copies.
2208 (b) Defined the list of op-code names for debugging as a macro in
2209 internal.h so that it is next to the definition of the opcodes.
2211 (c) Defined a table of op-code lengths for simpler skipping along compiled
2212 code. This is again a macro in internal.h so that it is next to the
2213 definition of the opcodes.
2215 34. Added support for recursive calls to individual subpatterns, along the
2216 lines of Robin Houston's patch (but implemented somewhat differently).
2218 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2219 allow it to read and process whole directories in Win32. This code was
2220 contributed by Lionel Fourquaux; it has not been tested by me.
2222 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2223 used to name a group. Names consist of alphanumerics and underscores, and must
2224 be unique. Back references use the syntax (?P=name) and recursive calls use
2225 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2226 numbers. The function pcre_fullinfo() can be used after compilation to extract
2227 a name/number map. There are three relevant calls:
2229 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2230 PCRE_INFO_NAMECOUNT yields the number of entries
2231 PCRE_INFO_NAMETABLE yields a pointer to the map.
2233 The map is a vector of fixed-size entries. The size of each entry depends on
2234 the length of the longest name used. The first two bytes of each entry are the
2235 group number, most significant byte first. There follows the corresponding
2236 name, zero terminated. The names are in alphabetical order.
2238 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2239 case instead of 255. Making it the same both with and without UTF-8 support
2240 means that the same test output works with both.
2242 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2243 calling malloc() with a zero argument.
2245 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2246 optimization. I've improved things by keeping a bitmap of backreferences with
2247 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2248 fact referenced, the optimization can be applied. It is unlikely that a
2249 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2250 the match to follow \n) will appear inside brackets with a number greater than
2251 31, but if it does, any back reference > 31 suppresses the optimization.
2253 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2254 of disabling numbered capturing parentheses. Any opening parenthesis that is
2255 not followed by ? behaves as if it were followed by ?: but named parentheses
2256 can still be used for capturing (and they will acquire numbers in the usual
2257 way).
2259 41. Redesigned the return codes from the match() function into yes/no/error so
2260 that errors can be passed back from deep inside the nested calls. A malloc
2261 failure while inside a recursive subpattern call now causes the
2262 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2264 42. It is now possible to set a limit on the number of times the match()
2265 function is called in a call to pcre_exec(). This facility makes it possible to
2266 limit the amount of recursion and backtracking, though not in a directly
2267 obvious way, because the match() function is used in a number of different
2268 circumstances. The count starts from zero for each position in the subject
2269 string (for non-anchored patterns). The default limit is, for compatibility, a
2270 large number, namely 10 000 000. You can change this in two ways:
2272 (a) When configuring PCRE before making, you can use --with-match-limit=n
2273 to set a default value for the compiled library.
2275 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2276 a different value is set. See 45 below.
2278 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2280 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2281 of things that can be changed at compile time. The first argument specifies
2282 what is wanted and the second points to where the information is to be placed.
2283 The current list of available information is:
2287 The output is an integer that is set to one if UTF-8 support is available;
2288 otherwise it is set to zero.
2292 The output is an integer that it set to the value of the code that is used for
2293 newline. It is either LF (10) or CR (13).
2297 The output is an integer that contains the number of bytes used for internal
2298 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2302 The output is an integer that contains the threshold above which the POSIX
2303 interface uses malloc() for output vectors. See item 31 above.
2307 The output is an unsigned integer that contains the default limit of the number
2308 of match() calls in a pcre_exec() execution. See 42 above.
2310 44. pcretest has been upgraded by the addition of the -C option. This causes it
2311 to extract all the available output from the new pcre_config() function, and to
2312 output it. The program then exits immediately.
2314 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2315 order to support additional features. One way would have been to define
2316 pcre_exec2() (for example) with extra arguments, but this would not have been
2317 extensible, and would also have required all calls to the original function to
2318 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2319 is used for passing in "extra" data from pcre_study().
2321 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2322 contains the following fields:
2324 flags a bitmap indicating which of the following fields are set
2325 study_data opaque data from pcre_study()
2326 match_limit a way of specifying a limit on match() calls for a specific
2327 call to pcre_exec()
2328 callout_data data for callouts (see 49 below)
2330 The flag bits are also defined in pcre.h, and are
2336 The pcre_study() function now returns one of these new pcre_extra blocks, with
2337 the actual study data pointed to by the study_data field, and the
2338 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2339 before. That is, this change is entirely upwards-compatible and requires no
2340 change to existing code.
2342 If you want to pass in additional data to pcre_exec(), you can either place it
2343 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2344 block.
2346 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2347 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2348 times with different match limits, until it finds the minimum value needed for
2349 pcre_exec() to complete. The value is then output. This can be instructive; for
2350 most simple matches the number is quite small, but for pathological cases it
2351 gets very large very quickly.
2353 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2354 returns the size of the data block pointed to by the study_data field in a
2355 pcre_extra block, that is, the value that was passed as the argument to
2356 pcre_malloc() when PCRE was getting memory in which to place the information
2357 created by pcre_study(). The fourth argument should point to a size_t variable.
2358 pcretest has been extended so that this information is shown after a successful
2359 pcre_study() call when information about the compiled regex is being displayed.
2361 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2362 because what follows is always an absolute path. (Later: it turns out that this
2363 is more than cosmetic for MinGW, because it doesn't like empty path
2364 components.)
2366 49. Some changes have been made to the callout feature (see 28 above):
2368 (i) A callout function now has three choices for what it returns:
2370 0 => success, carry on matching
2371 > 0 => failure at this point, but backtrack if possible
2372 < 0 => serious error, return this value from pcre_exec()
2374 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2375 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2376 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2377 use by callout functions. It will never be used by PCRE itself.
2379 (ii) The pcre_extra structure (see 45 above) has a void * field called
2380 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2381 pcre_callout_block structure has a field of the same name. The contents of
2382 the field passed in the pcre_extra structure are passed to the callout
2383 function in the corresponding field in the callout block. This makes it
2384 easier to use the same callout-containing regex from multiple threads. For
2385 testing, the pcretest program has a new data escape
2387 \C*n pass the number n (may be negative) as callout_data
2389 If the callout function in pcretest receives a non-zero value as
2390 callout_data, it returns that value.
2392 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2393 there were some redundant $(CFLAGS) in commands that are now specified as
2394 $(LINK), which already includes $(CFLAGS).
2396 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2397 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2398 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2399 one-byte characters throughout. Note that case-insensitive matching applies
2400 only to characters whose values are less than 256. PCRE doesn't support the
2401 notion of cases for higher-valued characters.
2403 (i) A character class whose characters are all within 0-255 is handled as
2404 a bit map, and the map is inverted for negative classes. Previously, a
2405 character > 255 always failed to match such a class; however it should
2406 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2408 (ii) A negated character class with a single character < 255 is coded as
2409 "not this character" (OP_NOT). This wasn't working properly when the test
2410 character was multibyte, either singly or repeated.
2412 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2413 mode, for example: \x{100}{2,3}.
2415 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2416 singly or repeated) now correctly test multibyte characters. However,
2417 PCRE doesn't recognize any characters with values greater than 255 as
2418 digits, spaces, or word characters. Such characters always match \D, \S,
2419 and \W, and never match \d, \s, or \w.
2421 (v) Classes may now contain characters and character ranges with values
2422 greater than 255. For example: [ab\x{100}-\x{400}].
2424 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2425 PCRE in UTF-8 mode.
2427 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2428 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2429 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2430 value.)
2432 53. The single man page has become too large. I have therefore split it up into
2433 a number of separate man pages. These also give rise to individual HTML pages;
2434 these are now put in a separate directory, and there is an index.html page that
2435 lists them all. Some hyperlinking between the pages has been installed.
2437 54. Added convenience functions for handling named capturing parentheses.
2439 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2440 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2441 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2442 are faulted.
2444 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2445 calling configure. These values are used when compiling the dftables.c program
2446 which is run to generate the source of the default character tables. They
2447 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2448 you will need to set these values.
2450 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2453 Version 3.9 02-Jan-02
2454 ---------------------
2456 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2458 2. If --disable-static was given, the building process failed when trying to
2459 build pcretest and pcregrep. (For some reason it was using libtool to compile
2460 them, which is not right, as they aren't part of the library.)
2463 Version 3.8 18-Dec-01
2464 ---------------------
2466 1. The experimental UTF-8 code was completely screwed up. It was packing the
2467 bytes in the wrong order. How dumb can you get?
2470 Version 3.7 29-Oct-01
2471 ---------------------
2473 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2474 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2475 this didn't happen under Solaris 8, where I normally test things.
2477 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2478 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2479 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2480 this fix an hour or so after the initial 3.7 release.)
2483 Version 3.6 23-Oct-01
2484 ---------------------
2486 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2487 offsets passed as NULL with zero offset count.
2489 2. The config.guess and config.sub files had not been updated when I moved to
2490 the latest autoconf.
2493 Version 3.5 15-Aug-01
2494 ---------------------
2496 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2497 had been forgotten.
2499 2. By using declared but undefined structures, we can avoid using "void"
2500 definitions in pcre.h while keeping the internal definitions of the structures
2501 private.
2503 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2504 user point of view, this means that both static and shared libraries are built
2505 by default, but this can be individually controlled. More of the work of
2506 handling this static/shared cases is now inside libtool instead of PCRE's make
2507 file.
2509 4. The pcretest utility is now installed along with pcregrep because it is
2510 useful for users (to test regexs) and by doing this, it automatically gets
2511 relinked by libtool. The documentation has been turned into a man page, so
2512 there are now .1, .txt, and .html versions in /doc.
2514 5. Upgrades to pcregrep:
2515 (i) Added long-form option names like gnu grep.
2516 (ii) Added --help to list all options with an explanatory phrase.
2517 (iii) Added -r, --recursive to recurse into sub-directories.
2518 (iv) Added -f, --file to read patterns from a file.
2520 6. pcre_exec() was referring to its "code" argument before testing that
2521 argument for NULL (and giving an error if it was NULL).
2523 7. Upgraded Makefile.in to allow for compiling in a different directory from
2524 the source directory.
2526 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2527 options bits, the pointer it was passed was to an int instead of to an unsigned
2528 long int. This mattered only on 64-bit systems.
2530 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2531 generated) instead of pcre.in, which it its source. Also made the same change
2532 in several of the .c files.
2534 10. A new release of gcc defines printf() as a macro, which broke pcretest
2535 because it had an ifdef in the middle of a string argument for printf(). Fixed
2536 by using separate calls to printf().
2538 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2539 script, to force use of CR or LF instead of \n in the source. On non-Unix
2540 systems, the value can be set in config.h.
2542 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2543 absolute limit. Changed the text of the error message to make this clear, and
2544 likewise updated the man page.
2546 13. The limit of 99 on the number of capturing subpatterns has been removed.
2547 The new limit is 65535, which I hope will not be a "real" limit.
2550 Version 3.4 22-Aug-00
2551 ---------------------
2553 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2555 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2558 Version 3.3 01-Aug-00
2559 ---------------------
2561 1. If an octal character was given, but the value was greater than \377, it
2562 was not getting masked to the least significant bits, as documented. This could
2563 lead to crashes in some systems.
2565 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2566 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2568 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2569 These just pass their arguments on to (pcre_free)(), but they are provided
2570 because some uses of PCRE bind it to non-C systems that can call its functions,
2571 but cannot call free() or pcre_free() directly.
2573 4. Add "make test" as a synonym for "make check". Corrected some comments in
2574 the Makefile.
2576 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2577 Makefile.
2579 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2580 command called pgrep for grepping around the active processes.
2582 7. Added the beginnings of support for UTF-8 character strings.
2584 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2585 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2586 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2587 out for the ar command.)
2590 Version 3.2 12-May-00
2591 ---------------------
2593 This is purely a bug fixing release.
2595 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2596 of ZA. This was just one example of several cases that could provoke this bug,
2597 which was introduced by change 9 of version 2.00. The code for breaking
2598 infinite loops after an iteration that matches an empty string was't working
2599 correctly.
2601 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2602 when matched against abbab (for example). After matching an empty string, it
2603 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2604 caused it to match further down the string than it should.
2606 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2607 was there because it doesn't seem to be needed, and it causes trouble on some
2608 systems, as it is not a Standard C header. It has been removed.
2610 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2611 were reported on the Macintosh. The changes were from
2613 while ((c = *(++ptr)) != 0 && c != '\n');
2614 to
2615 while ((c = *(++ptr)) != 0 && c != '\n') ;
2617 Totally extraordinary, but if that's what it takes...
2619 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2620 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2621 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2622 assumes the way PCRE uses memmove() (always moving upwards).
2624 6. PCRE is being used in one environment where strchr() is not available. There
2625 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2626 faster code anyway.
2629 Version 3.1 09-Feb-00
2630 ---------------------
2632 The only change in this release is the fixing of some bugs in Makefile.in for
2633 the "install" target:
2635 (1) It was failing to install pcreposix.h.
2637 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2640 Version 3.0 01-Feb-00
2641 ---------------------
2643 1. Add support for the /+ modifier to perltest (to output $` like it does in
2644 pcretest).
2646 2. Add support for the /g modifier to perltest.
2648 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2649 matches null strings.
2651 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2652 pattern. Perl treats empty patterns specially - it reuses the most recent
2653 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2654 effect.
2656 5. The POSIX interface was broken in that it was just handing over the POSIX
2657 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2658 required a bigger vector, with some working space on the end. This means that
2659 the POSIX wrapper now has to get and free some memory, and copy the results.
2661 6. Added some simple autoconf support, placing the test data and the
2662 documentation in separate directories, re-organizing some of the
2663 information files, and making it build pcre-config (a GNU standard). Also added
2664 libtool support for building PCRE as a shared library, which is now the
2665 default.
2667 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2668 09 are not valid octal constants. Single digits will be used for minor values
2669 less than 10.
2671 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2672 existing programs that set these in the POSIX interface can use PCRE without
2673 modification.
2675 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2676 return all that pcre_info() returns, plus additional data. The pcre_info()
2677 function is retained for compatibility, but is considered to be obsolete.
2679 10. Added experimental recursion feature (?R) to handle one common case that
2680 Perl 5.6 will be able to do with (?p{...}).
2682 11. Added support for POSIX character classes like [:alpha:], which Perl is
2683 adopting.
2686 Version 2.08 31-Aug-99
2687 ----------------------
2689 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2690 trying to match at the startoffset position, but instead was moving forward to
2691 the next newline as if a previous match had failed.
2693 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2694 and could get into a loop if a null string was matched other than at the start
2695 of the subject.
2697 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2698 be distinguished at compile time, and for completeness also added PCRE_DATE.
2700 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2701 in GnuWin32 environments.
2704 Version 2.07 29-Jul-99
2705 ----------------------
2707 1. The documentation is now supplied in plain text form and HTML as well as in
2708 the form of man page sources.
2710 2. C++ compilers don't like assigning (void *) values to other pointer types.
2711 In particular this affects malloc(). Although there is no problem in Standard
2712 C, I've put in casts to keep C++ compilers happy.
2714 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2715 should be (const char *).
2717 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2718 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2719 However, I haven't made this a standard facility. The documentation doesn't
2720 mention it, and the Makefile doesn't support it.
2722 5. The Makefile now contains an "install" target, with editable destinations at
2723 the top of the file. The pcretest program is not installed.
2725 6. pgrep -V now gives the PCRE version number and date.
2727 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2728 causing the entire string to be ignored, instead of just the last character.
2730 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2731 non-matching string, it can take a very, very long time, even for strings of
2732 quite modest length, because of the nested recursion. PCRE now does better in
2733 some of these cases. It does this by remembering the last required literal
2734 character in the pattern, and pre-searching the subject to ensure it is present
2735 before running the real match. In other words, it applies a heuristic to detect
2736 some types of certain failure quickly, and in the above example, if presented
2737 with a string that has no trailing " it gives "no match" very quickly.
2739 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2740 other alternatives are tried instead.
2743 Version 2.06 09-Jun-99
2744 ----------------------
2746 1. Change pcretest's output for amount of store used to show just the code
2747 space, because the remainder (the data block) varies in size between 32-bit and
2748 64-bit systems.
2750 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2751 start matching at. This allows lookbehinds to work when searching for multiple
2752 occurrences in a string.
2754 3. Added additional options to pcretest for testing multiple occurrences:
2756 /+ outputs the rest of the string that follows a match
2757 /g loops for multiple occurrences, using the new startoffset argument
2758 /G loops for multiple occurrences by passing an incremented pointer
2760 4. PCRE wasn't doing the "first character" optimization for patterns starting
2761 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2762 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2763 the letter 'x'. On long subject strings, this gives a significant speed-up.
2766 Version 2.05 21-Apr-99
2767 ----------------------
2769 1. Changed the type of magic_number from int to long int so that it works
2770 properly on 16-bit systems.
2772 2. Fixed a bug which caused patterns starting with .* not to work correctly
2773 when the subject string contained newline characters. PCRE was assuming
2774 anchoring for such patterns in all cases, which is not correct because .* will
2775 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2776 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2777 must be retried after every newline in the subject.
2780 Version 2.04 18-Feb-99
2781 ----------------------
2783 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2784 computation of the store needed to hold the pattern was incorrect (too large).
2785 If such patterns were nested a few deep, this could multiply and become a real
2786 problem.
2788 2. Added /M option to pcretest to show the memory requirement of a specific
2789 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2791 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2792 compiled in such a way that the backtracking after subsequent failure was
2793 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2794 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2797 Version 2.03 02-Feb-99
2798 ----------------------
2800 1. Fixed typo and small mistake in man page.
2802 2. Added 4th condition (GPL supersedes if conflict) and created separate
2803 LICENCE file containing the conditions.
2805 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2806 Perl, that is the internal \ allows the delimiter to be included in the
2807 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2808 the final delimiter, add \ to the end of the pattern (to test the error).
2810 4. Added the convenience functions for extracting substrings after a successful
2811 match. Updated pcretest to make it able to test these functions.
2814 Version 2.02 14-Jan-99
2815 ----------------------
2817 1. Initialized the working variables associated with each extraction so that
2818 their saving and restoring doesn't refer to uninitialized store.
2820 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2821 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2822 fix the problem.
2824 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2825 calls, and wasn't printing the correct value for compiling calls. Increased the
2826 default value of LOOPREPEAT, and the number of significant figures in the
2827 times.
2829 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2831 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2832 a building problem on Windows NT with a FAT file system.
2835 Version 2.01 21-Oct-98
2836 ----------------------
2838 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2839 to character tables built by pcre_maketables() in the current locale. If NULL
2840 is passed, the default tables are used.
2843 Version 2.00 24-Sep-98
2844 ----------------------
2846 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2847 it any more.
2849 2. Allow quantification of (?>) groups, and make it work correctly.
2851 3. The first character computation wasn't working for (?>) groups.
2853 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2854 end of the subject) and add 5.005's \z, which really does match only at the
2855 very end of the subject.
2857 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2859 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2860 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2861 localized options. All options to pcre_study() were also removed.
2863 7. Add other new features from 5.005:
2865 $(?<= positive lookbehind
2866 $(?<! negative lookbehind
2867 (?imsx-imsx) added the unsetting capability
2868 such a setting is global if at outer level; local otherwise
2869 (?imsx-imsx:) non-capturing groups with option setting
2870 (?(cond)re|re) conditional pattern matching
2872 A backreference to itself in a repeated group matches the previous
2873 captured string.
2875 8. General tidying up of studying (both automatic and via "study")
2876 consequential on the addition of new assertions.
2878 9. As in 5.005, unlimited repeated groups that could match an empty substring
2879 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2880 runtime if any iteration does actually match an empty substring.
2882 10. Include the RunTest script in the distribution.
2884 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2885 discrepancies, some of which were old and were also with respect to 5.004. They
2886 have now been fixed.
2889 Version 1.09 28-Apr-98
2890 ----------------------
2892 1. A negated single character class followed by a quantifier with a minimum
2893 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2894 program crashes, or just wrong answers. This did not apply to negated classes
2895 containing more than one character, or to minima other than one.
2898 Version 1.08 27-Mar-98
2899 ----------------------
2901 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2903 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2904 latter must appear before anything that relies on it in the pattern.
2907 Version 1.07 16-Feb-98
2908 ----------------------
2910 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2911 repeat of a potentially empty string).
2914 Version 1.06 23-Jan-98
2915 ----------------------
2917 1. Added Markus Oberhumer's little patches for C++.
2919 2. Literal strings longer than 255 characters were broken.
2922 Version 1.05 23-Dec-97
2923 ----------------------
2925 1. Negated character classes containing more than one character were failing if
2926 PCRE_CASELESS was set at run time.
2929 Version 1.04 19-Dec-97
2930 ----------------------
2932 1. Corrected the man page, where some "const" qualifiers had been omitted.
2934 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
2935 input syntax.
2937 3. Fixed memory leak which occurred when a regex with back references was
2938 matched with an offsets vector that wasn't big enough. The temporary memory
2939 that is used in this case wasn't being freed if the match failed.
2941 4. Tidied pcretest to ensure it frees memory that it gets.
2943 5. Temporary memory was being obtained in the case where the passed offsets
2944 vector was exactly big enough.
2946 6. Corrected definition of offsetof() from change 5 below.
2948 7. I had screwed up change 6 below and broken the rules for the use of
2949 setjmp(). Now fixed.
2952 Version 1.03 18-Dec-97
2953 ----------------------
2955 1. A erroneous regex with a missing opening parenthesis was correctly
2956 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
2957 on some systems.
2959 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
2960 it was reported that one broken compiler failed on the former because "code" is
2961 also an independent variable.
2963 3. The erroneous regex a[]b caused an array overrun reference.
2965 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
2966 fail on data ending with that character. (It was going on too far, and checking
2967 the next character, typically a binary zero.) This was specific to the
2968 optimized code for single-character negative classes.
2970 5. Added a contributed patch from the TIN world which does the following:
2972 + Add an undef for memmove, in case the the system defines a macro for it.
2974 + Add a definition of offsetof(), in case there isn't one. (I don't know
2975 the reason behind this - offsetof() is part of the ANSI standard - but
2976 it does no harm).
2978 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
2979 most of the places where whitespace preceded '#'. I have given up and
2980 allowed the remaining 2 cases to be at the margin.
2982 + Rename some variables in pcre to eliminate shadowing. This seems very
2983 pedantic, but does no harm, of course.
2985 6. Moved the call to setjmp() into its own function, to get rid of warnings
2986 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
2988 7. Constructs such as \d{8,} were compiling into the equivalent of
2989 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
2990 outcome, but in this particular case used more store than had been allocated,
2991 which caused the bug to be discovered because it threw up an internal error.
2993 8. The debugging code in both pcre and pcretest for outputting the compiled
2994 form of a regex was going wrong in the case of back references followed by
2995 curly-bracketed repeats.
2998 Version 1.02 12-Dec-97
2999 ----------------------
3001 1. Typos in pcre.3 and comments in the source fixed.
3003 2. Applied a contributed patch to get rid of places where it used to remove
3004 'const' from variables, and fixed some signed/unsigned and uninitialized
3005 variable warnings.
3007 3. Added the "runtest" target to Makefile.
3009 4. Set default compiler flag to -O2 rather than just -O.
3012 Version 1.01 19-Nov-97
3013 ----------------------
3015 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3016 like /([ab]*)*/, that is, for classes with more than one character in them.
3018 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3019 as /((?>a*))*/ (a PCRE_EXTRA facility).
3022 Version 1.00 18-Nov-97
3023 ----------------------
3025 1. Added compile-time macros to support systems such as SunOS4 which don't have
3026 memmove() or strerror() but have other things that can be used instead.
3028 2. Arranged that "make clean" removes the executables.
3031 Version 0.99 27-Oct-97
3032 ----------------------
3034 1. Fixed bug in code for optimizing classes with only one character. It was
3035 initializing a 32-byte map regardless, which could cause it to run off the end
3036 of the memory it had got.
3038 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3041 Version 0.98 22-Oct-97
3042 ----------------------
3044 1. Fixed bug in code for handling temporary memory usage when there are more
3045 back references than supplied space in the ovector. This could cause segfaults.
3048 Version 0.97 21-Oct-97
3049 ----------------------
3051 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3053 2. Optimized negated single characters not to use a bit map.
3055 3. Brought error texts together as macro definitions; clarified some of them;
3056 fixed one that was wrong - it said "range out of order" when it meant "invalid
3057 escape sequence".
3059 4. Changed some char * arguments to const char *.
3061 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3063 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3064 pcretest.
3067 Version 0.96 16-Oct-97
3068 ----------------------
3070 1. Added a simple "pgrep" utility to the distribution.
3072 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3073 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3074 where "ddd" means "one or more decimal digits".
3076 3. Fixed serious bug. If a pattern had a back reference, but the call to
3077 pcre_exec() didn't supply a large enough ovector to record the related
3078 identifying subpattern, the match always failed. PCRE now remembers the number
3079 of the largest back reference, and gets some temporary memory in which to save
3080 the offsets during matching if necessary, in order to ensure that
3081 backreferences always work.
3083 4. Increased the compatibility with Perl in a number of ways:
3085 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3086 to request this handling. The option can be set at compile or exec time.
3088 (b) $ matches before a terminating newline by default; an option
3089 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3090 mode). The option can be set at compile or exec time.
3092 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3093 the same as Perl's. If the decimal number it represents is less than 10
3094 or there aren't that many previous left capturing parentheses, an octal
3095 escape is read. Inside a character class, it's always an octal escape,
3096 even if it is a single digit.
3098 (d) An escaped but undefined alphabetic character is taken as a literal,
3099 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3100 escapes.
3102 (e) {0} is now permitted. (The previous item is removed from the compiled
3103 pattern).
3105 5. Changed all the names of code files so that the basic parts are no longer
3106 than 10 characters, and abolished the teeny "globals.c" file.
3108 6. Changed the handling of character classes; they are now done with a 32-byte
3109 bit map always.
3111 7. Added the -d and /D options to pcretest to make it possible to look at the
3112 internals of compilation without having to recompile pcre.
3115 Version 0.95 23-Sep-97
3116 ----------------------
3118 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3119 \x20 at the start of a run of normal characters. These were being treated as
3120 real characters, instead of the source characters being re-checked.
3123 Version 0.94 18-Sep-97
3124 ----------------------
3126 1. The functions are now thread-safe, with the caveat that the global variables
3127 containing pointers to malloc() and free() or alternative functions are the
3128 same for all threads.
3130 2. Get pcre_study() to generate a bitmap of initial characters for non-
3131 anchored patterns when this is possible, and use it if passed to pcre_exec().
3134 Version 0.93 15-Sep-97
3135 ----------------------
3137 1. /(b)|(:+)/ was computing an incorrect first character.
3139 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3140 but not actually doing anything yet.
3142 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3143 as Perl does (e.g. [-az] or [az-]).
3145 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3146 all possible positions.
3148 5. Split up into different modules to avoid including unneeded functions in a
3149 compiled binary. However, compile and exec are still in one module. The "study"
3150 function is split off.
3152 6. The character tables are now in a separate module whose source is generated
3153 by an auxiliary program - but can then be edited by hand if required. There are
3154 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3155 toupper() in the code.
3157 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3158 make them global. Abolish the function for setting them, as the caller can now
3159 set them directly.
3162 Version 0.92 11-Sep-97
3163 ----------------------
3165 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3166 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3168 2. Caseless matching was not working in character classes if the characters in
3169 the pattern were in upper case.
3171 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3173 4. Make PCRE_ANCHORED public and accept as a compile option.
3175 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3176 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3177 pass them.
3179 6. Give an error if bad option bits passed at compile or run time.
3181 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3182 pcretest to cause it to pass that flag.
3184 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3185 options, and the first character, if set.
3187 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3190 Version 0.91 10-Sep-97
3191 ----------------------
3193 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3194 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3196 2. PCRE was looping on encountering an indefinitely repeated back reference to
3197 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3198 Perl does - treats the match as successful.
3200 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5