ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 341 - (show annotations)
Sat Apr 19 16:41:04 2008 UTC (12 years, 10 months ago) by ph10
File size: 155078 byte(s)
Error occurred while calculating annotation data.
Fix DFA (?!) bug; add support for JavaScript empty classes.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.7 05-Mar-08
5 ---------------------
7 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
8 a string to a long long, pretend we don't even have a long long." This is
9 done by checking for the strtoq, strtoll, and _strtoi64 functions.
11 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
12 pre-7.6 versions, which defined a global no_arg variable instead of putting
13 it in the RE class. (See also #8 below.)
15 3. Remove a line of dead code, identified by coverity and reported by Nuno
16 Lopes.
18 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
20 (1) The include/exclude patterns were being applied to the whole pathnames
21 of files, instead of just to the final components.
23 (2) If there was more than one level of directory, the subdirectories were
24 skipped unless they satisfied the include/exclude conditions. This is
25 inconsistent with GNU grep (and could even be seen as contrary to the
26 pcregrep specification - which I improved to make it absolutely clear).
27 The action now is always to scan all levels of directory, and just
28 apply the include/exclude patterns to regular files.
30 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
31 --exclude_dir in the tests to avoid scanning .svn directories.
33 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
34 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
35 doesn't support NULs in patterns.
37 7. Added some missing "const"s to declarations of static tables in
38 pcre_compile.c and pcre_dfa_exec.c.
40 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
41 caused by fix #2 above. (Subsequently also a second patch to fix the
42 first patch. And a third patch - this was a messy problem.)
44 9. Applied Craig's patch to remove the use of push_back().
46 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
47 matching function regexec().
49 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
50 which, however, unlike Perl's \g{...}, are subroutine calls, not back
51 references. PCRE supports relative numbers with this syntax (I don't think
52 Oniguruma does).
54 12. Previously, a group with a zero repeat such as (...){0} was completely
55 omitted from the compiled regex. However, this means that if the group
56 was called as a subroutine from elsewhere in the pattern, things went wrong
57 (an internal error was given). Such groups are now left in the compiled
58 pattern, with a new opcode that causes them to be skipped at execution
59 time.
61 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
62 to the way PCRE behaves:
64 (a) A lone ] character is dis-allowed (Perl treats it as data).
66 (b) A back reference to an unmatched subpattern matches an empty string
67 (Perl fails the current match path).
69 (c) A data ] in a character class must be notated as \] because if the
70 first data character in a class is ], it defines an empty class. (In
71 Perl it is not possible to have an empty class.) The empty class []
72 never matches; it forces failure and is equivalent to (*FAIL) or (?!).
73 The negative empty class [^] matches any one character, independently
74 of the DOTALL setting.
76 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
77 non-existent subpattern following a character class starting with ']' and
78 containing () gave an internal compiling error instead of "reference to
79 non-existent subpattern". Fortunately, when the pattern did exist, the
80 compiled code was correct. (When scanning forwards to check for the
81 existencd of the subpattern, it was treating the data ']' as terminating
82 the class, so got the count wrong. When actually compiling, the reference
83 was subsequently set up correctly.)
85 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
86 it was being rejected as not supported by pcre_dfa_exec(), even though
87 other assertions are supported. I have made pcre_dfa_exec() support
88 (*FAIL).
91 Version 7.6 28-Jan-08
92 ---------------------
94 1. A character class containing a very large number of characters with
95 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
96 overflow.
98 2. Patch to cut out the "long long" test in pcrecpp_unittest when
99 HAVE_LONG_LONG is not defined.
101 3. Applied Christian Ehrlicher's patch to update the CMake build files to
102 bring them up to date and include new features. This patch includes:
104 - Fixed PH's badly added libz and libbz2 support.
105 - Fixed a problem with static linking.
106 - Added pcredemo. [But later removed - see 7 below.]
107 - Fixed dftables problem and added an option.
108 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
110 - Added readline support for pcretest.
111 - Added an listing of the option settings after cmake has run.
113 4. A user submitted a patch to Makefile that makes it easy to create
114 "pcre.dll" under mingw when using Configure/Make. I added stuff to
115 Makefile.am that cause it to include this special target, without
116 affecting anything else. Note that the same mingw target plus all
117 the other distribution libraries and programs are now supported
118 when configuring with CMake (see 6 below) instead of with
119 Configure/Make.
121 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
122 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
123 exported in the Windows port". It has not yet been confirmed that the patch
124 solves the problem, but it does no harm.
126 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
127 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
128 with CMake, and also correct the comment about stack recursion.
130 7. Remove the automatic building of pcredemo from the ./configure system and
131 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
132 of a program that users should build themselves after PCRE is installed, so
133 building it automatically is not really right. What is more, it gave
134 trouble in some build environments.
136 8. Further tidies to CMakeLists.txt from Sheri and Christian.
139 Version 7.5 10-Jan-08
140 ---------------------
142 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
143 values in parens when parsing an RE using the C++ wrapper."
145 2. Negative specials like \S did not work in character classes in UTF-8 mode.
146 Characters greater than 255 were excluded from the class instead of being
147 included.
149 3. The same bug as (2) above applied to negated POSIX classes such as
150 [:^space:].
152 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
153 defined or documented. It seems to have been a typo for PCRE_STATIC, so
154 I have changed it.
156 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
157 first named subpattern) and a construct such as (?&a) would reference the
158 first named subpattern whose name started with "a" (in other words, the
159 length check was missing). Both these problems are fixed. "Subpattern name
160 expected" is now given for (?&) (a zero-length name), and this patch also
161 makes it give the same error for \k'' (previously it complained that that
162 was a reference to a non-existent subpattern).
164 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
165 this is right because (?- can be followed by option settings as well as by
166 digits. I have, however, made the messages clearer.
168 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
169 than the number used in the conditional) now cause a compile-time error.
170 This is actually not compatible with Perl, which accepts such patterns, but
171 treats the conditional as always being FALSE (as PCRE used to), but it
172 seems to me that giving a diagnostic is better.
174 8. Change "alphameric" to the more common word "alphanumeric" in comments
175 and messages.
177 9. Fix two occurrences of "backslash" in comments that should have been
178 "backspace".
180 10. Remove two redundant lines of code that can never be obeyed (their function
181 was moved elsewhere).
183 11. The program that makes PCRE's Unicode character property table had a bug
184 which caused it to generate incorrect table entries for sequences of
185 characters that have the same character type, but are in different scripts.
186 It amalgamated them into a single range, with the script of the first of
187 them. In other words, some characters were in the wrong script. There were
188 thirteen such cases, affecting characters in the following ranges:
190 U+002b0 - U+002c1
191 U+0060c - U+0060d
192 U+0061e - U+00612
193 U+0064b - U+0065e
194 U+0074d - U+0076d
195 U+01800 - U+01805
196 U+01d00 - U+01d77
197 U+01d9b - U+01dbf
198 U+0200b - U+0200f
199 U+030fc - U+030fe
200 U+03260 - U+0327f
201 U+0fb46 - U+0fbb1
202 U+10450 - U+1049d
204 12. The -o option (show only the matching part of a line) for pcregrep was not
205 compatible with GNU grep in that, if there was more than one match in a
206 line, it showed only the first of them. It now behaves in the same way as
207 GNU grep.
209 13. If the -o and -v options were combined for pcregrep, it printed a blank
210 line for every non-matching line. GNU grep prints nothing, and pcregrep now
211 does the same. The return code can be used to tell if there were any
212 non-matching lines.
214 14. Added --file-offsets and --line-offsets to pcregrep.
216 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
217 infinitely looping recursion. The bug was that positive lookaheads were not
218 being skipped when checking for a possible empty match (negative lookaheads
219 and both kinds of lookbehind were skipped).
221 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
222 inclusion of <windows.h> to before rather than after the definition of
223 INVALID_FILE_ATTRIBUTES (patch from David Byron).
225 17. Specifying a possessive quantifier with a specific limit for a Unicode
226 character property caused pcre_compile() to compile bad code, which led at
227 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
228 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
229 caused the error; without that there was no problem.
231 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
233 19. Added --enable-pcretest-libreadline.
235 20. In pcrecpp.cc, the variable 'count' was incremented twice in
236 RE::GlobalReplace(). As a result, the number of replacements returned was
237 double what it should be. I removed one of the increments, but Craig sent a
238 later patch that removed the other one (the right fix) and added unit tests
239 that check the return values (which was not done before).
241 21. Several CMake things:
243 (1) Arranged that, when cmake is used on Unix, the libraries end up with
244 the names libpcre and libpcreposix, not just pcre and pcreposix.
246 (2) The above change means that pcretest and pcregrep are now correctly
247 linked with the newly-built libraries, not previously installed ones.
251 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
252 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
253 UTF-8 newline character). The key issue is that the pattern starts .*;
254 this means that the match must be either at the beginning, or after a
255 newline. The bug was in the code for advancing after a failed match and
256 checking that the new position followed a newline. It was not taking
257 account of UTF-8 characters correctly.
259 23. PCRE was behaving differently from Perl in the way it recognized POSIX
260 character classes. PCRE was not treating the sequence [:...:] as a
261 character class unless the ... were all letters. Perl, however, seems to
262 allow any characters between [: and :], though of course it rejects as
263 unknown any "names" that contain non-letters, because all the known class
264 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
265 for example, whereas PCRE did not - it did not recognize a POSIX character
266 class. This seemed a bit dangerous, so the code has been changed to be
267 closer to Perl. The behaviour is not identical to Perl, because PCRE will
268 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
269 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
270 Perl does, and where it didn't before.
272 24. Rewrite so as to remove the single use of %n from pcregrep because in some
273 Windows environments %n is disabled by default.
276 Version 7.4 21-Sep-07
277 ---------------------
279 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
280 means that a class such as [\s] counted as "explicit reference to CR or
281 LF". That isn't really right - the whole point of the change was to try to
282 help when there was an actual mention of one of the two characters. So now
283 the change happens only if \r or \n (or a literal CR or LF) character is
284 encountered.
286 2. The 32-bit options word was also used for 6 internal flags, but the numbers
287 of both had grown to the point where there were only 3 bits left.
288 Fortunately, there was spare space in the data structure, and so I have
289 moved the internal flags into a new 16-bit field to free up more option
290 bits.
292 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
293 but did not set the internal JCHANGED flag - either of these is enough to
294 control the way the "get" function works - but the PCRE_INFO_JCHANGED
295 facility is supposed to tell if (?J) was ever used, so now (?J) at the
296 start sets both bits.
298 4. Added options (at build time, compile time, exec time) to change \R from
299 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
301 5. doc/pcresyntax.html was missing from the distribution.
303 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
304 compatibility, even though it is no longer used.
306 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
307 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
308 windows.h file is present (where different names are used). [This was
309 reversed later after testing - see 16 below.]
311 8. Changed all #include <config.h> to #include "config.h". There were also
312 some further <pcre.h> cases that I changed to "pcre.h".
314 9. When pcregrep was used with the --colour option, it missed the line ending
315 sequence off the lines that it output.
317 10. It was pointed out to me that arrays of string pointers cause lots of
318 relocations when a shared library is dynamically loaded. A technique of
319 using a single long string with a table of offsets can drastically reduce
320 these. I have refactored PCRE in four places to do this. The result is
321 dramatic:
323 Originally: 290
324 After changing UCP table: 187
325 After changing error message table: 43
326 After changing table of "verbs" 36
327 After changing table of Posix names 22
329 Thanks to the folks working on Gregex for glib for this insight.
331 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
332 unicode-properties was also set.
334 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
336 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
337 checked only for CRLF.
339 14. Added casts to pcretest.c to avoid compiler warnings.
341 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
343 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
344 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
345 entirely. This removes changes made in 7 above.
347 17. The CMake files have been updated, and there is now more information about
348 building with CMake in the NON-UNIX-USE document.
351 Version 7.3 28-Aug-07
352 ---------------------
354 1. In the rejigging of the build system that eventually resulted in 7.1, the
355 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
356 brackets there is not right, since it causes compilers to look for an
357 installed pcre.h, not the version that is in the source that is being
358 compiled (which of course may be different). I have changed it back to:
360 #include "pcre.h"
362 I have a vague recollection that the change was concerned with compiling in
363 different directories, but in the new build system, that is taken care of
364 by the VPATH setting the Makefile.
366 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
367 when the subject happened to end in the byte 0x85 (e.g. if the last
368 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
369 characters but of course it shouldn't be taken as a newline when it is part
370 of another character. The bug was that, for an unlimited repeat of . in
371 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
372 characters when looking for a newline.
374 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
376 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
377 in debug output.
379 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
380 long printing in the pcrecpp unittest when running under MinGW.
382 6. ESC_K was left out of the EBCDIC table.
384 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
385 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
386 limit also applies to "virtual nesting" when a pattern is recursive, and in
387 this case 1000 isn't so big. I have been able to remove this limit at the
388 expense of backing off one optimization in certain circumstances. Normally,
389 when pcre_exec() would call its internal match() function recursively and
390 immediately return the result unconditionally, it uses a "tail recursion"
391 feature to save stack. However, when a subpattern that can match an empty
392 string has an unlimited repetition quantifier, it no longer makes this
393 optimization. That gives it a stack frame in which to save the data for
394 checking that an empty string has been matched. Previously this was taken
395 from the 1000-entry workspace that had been reserved. So now there is no
396 explicit limit, but more stack is used.
398 8. Applied Daniel's patches to solve problems with the import/export magic
399 syntax that is required for Windows, and which was going wrong for the
400 pcreposix and pcrecpp parts of the library. These were overlooked when this
401 problem was solved for the main library.
403 9. There were some crude static tests to avoid integer overflow when computing
404 the size of patterns that contain repeated groups with explicit upper
405 limits. As the maximum quantifier is 65535, the maximum group length was
406 set at 30,000 so that the product of these two numbers did not overflow a
407 32-bit integer. However, it turns out that people want to use groups that
408 are longer than 30,000 bytes (though not repeat them that many times).
409 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
410 made it possible to implement the integer overflow checks in a much more
411 dynamic way, which I have now done. The artificial limitation on group
412 length has been removed - we now have only the limit on the total length of
413 the compiled pattern, which depends on the LINK_SIZE setting.
415 10. Fixed a bug in the documentation for get/copy named substring when
416 duplicate names are permitted. If none of the named substrings are set, the
417 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
418 empty string.
420 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
421 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
422 because the ] is interpreted as the first data character and the
423 terminating ] is not found. PCRE has been made compatible with Perl in this
424 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
425 cause memory overwriting.
427 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
428 string has been matched (to stop an infinite loop). It was not recognizing
429 a conditional subpattern that could match an empty string if that
430 subpattern was within another subpattern. For example, it looped when
431 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
432 condition was not nested. This bug has been fixed.
434 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
435 past the start of the subject in the presence of bytes with the top bit
436 set, for example "\x8aBCD".
438 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
439 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
441 14. Optimized (?!) to (*FAIL).
443 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
444 This restricts code points to be within the range 0 to 0x10FFFF, excluding
445 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
446 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
447 does: it's just the validity check that is more restrictive.
449 16. Inserted checks for integer overflows during escape sequence (backslash)
450 processing, and also fixed erroneous offset values for syntax errors during
451 backslash processing.
453 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
454 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
456 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
457 caused an overrun.
459 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
460 something other than just ASCII characters) inside a group that had an
461 unlimited repeat caused a loop at compile time (while checking to see
462 whether the group could match an empty string).
464 20. Debugging a pattern containing \p or \P could cause a crash. For example,
465 [\P{Any}] did so. (Error in the code for printing property names.)
467 21. An orphan \E inside a character class could cause a crash.
469 22. A repeated capturing bracket such as (A)? could cause a wild memory
470 reference during compilation.
472 23. There are several functions in pcre_compile() that scan along a compiled
473 expression for various reasons (e.g. to see if it's fixed length for look
474 behind). There were bugs in these functions when a repeated \p or \P was
475 present in the pattern. These operators have additional parameters compared
476 with \d, etc, and these were not being taken into account when moving along
477 the compiled data. Specifically:
479 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
480 length.
482 (b) An item such as \pL+ within a repeated group could cause crashes or
483 loops.
485 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
486 "reference to non-existent subpattern" error.
488 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
490 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
491 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
493 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
495 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
496 character were causing crashes (broken optimization).
498 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
499 \p or \P) caused a compile-time loop.
501 28. More problems have arisen in unanchored patterns when CRLF is a valid line
502 break. For example, the unstudied pattern [\r\n]A does not match the string
503 "\r\nA" because change 7.0/46 below moves the current point on by two
504 characters after failing to match at the start. However, the pattern \nA
505 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
506 the same is true. There doesn't seem any very clean way out of this, but
507 what I have chosen to do makes the common cases work: PCRE now takes note
508 of whether there can be an explicit match for \r or \n anywhere in the
509 pattern, and if so, 7.0/46 no longer applies. As part of this change,
510 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
511 pattern has explicit CR or LF references.
513 29. Added (*CR) etc for changing newline setting at start of pattern.
516 Version 7.2 19-Jun-07
517 ---------------------
519 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
520 which is apparently normally available under Windows.
522 2. Re-jig the pcregrep tests with different newline settings in an attempt
523 to make them independent of the local environment's newline setting.
525 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
527 4. Some of the "internals" tests were previously cut out when the link size
528 was not 2, because the output contained actual offsets. The recent new
529 "Z" feature of pcretest means that these can be cut out, making the tests
530 usable with all link sizes.
532 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
533 stack recursion. This gives a massive performance boost under BSD, but just
534 a small improvement under Linux. However, it saves one field in the frame
535 in all cases.
537 6. Added more features from the forthcoming Perl 5.10:
539 (a) (?-n) (where n is a string of digits) is a relative subroutine or
540 recursion call. It refers to the nth most recently opened parentheses.
542 (b) (?+n) is also a relative subroutine call; it refers to the nth next
543 to be opened parentheses.
545 (c) Conditions that refer to capturing parentheses can be specified
546 relatively, for example, (?(-2)... or (?(+3)...
548 (d) \K resets the start of the current match so that everything before
549 is not part of it.
551 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
553 (f) \g{name} is another synonym - part of Perl 5.10's unification of
554 reference syntax.
556 (g) (?| introduces a group in which the numbering of parentheses in each
557 alternative starts with the same number.
559 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
561 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
564 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
565 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
566 for detecting groups that can match an empty string.
568 9. A pattern with a very large number of alternatives (more than several
569 hundred) was running out of internal workspace during the pre-compile
570 phase, where pcre_compile() figures out how much memory will be needed. A
571 bit of new cunning has reduced the workspace needed for groups with
572 alternatives. The 1000-alternative test pattern now uses 12 bytes of
573 workspace instead of running out of the 4096 that are available.
575 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
577 11. Applied patch from Google to remove an optimization that didn't quite work.
578 The report of the bug said:
580 pcrecpp::RE("a*").FullMatch("aaa") matches, while
581 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
582 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
584 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
585 it matched the wrong number of bytes.
588 Version 7.1 24-Apr-07
589 ---------------------
591 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
592 that is more "standard", making use of automake and other Autotools. There
593 is some re-arrangement of the files and adjustment of comments consequent
594 on this.
596 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
597 for recursive directory scanning broke on some systems because the files
598 are not scanned in any specific order and on different systems the order
599 was different. A call to "sort" has been inserted into RunGrepTest for the
600 approprate test as a short-term fix. In the longer term there may be an
601 alternative.
603 3. I had an email from Eric Raymond about problems translating some of PCRE's
604 man pages to HTML (despite the fact that I distribute HTML pages, some
605 people do their own conversions for various reasons). The problems
606 concerned the use of low-level troff macros .br and .in. I have therefore
607 removed all such uses from the man pages (some were redundant, some could
608 be replaced by .nf/.fi pairs). The 132html script that I use to generate
609 HTML has been updated to handle .nf/.fi and to complain if it encounters
610 .br or .in.
612 4. Updated comments in configure.ac that get placed in config.h.in and also
613 arranged for config.h to be included in the distribution, with the name
614 config.h.generic, for the benefit of those who have to compile without
615 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
617 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
618 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
619 makevp.bat for the current PCRE, using the additional files
620 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
622 6. A Windows user reported a minor discrepancy with test 2, which turned out
623 to be caused by a trailing space on an input line that had got lost in his
624 copy. The trailing space was an accident, so I've just removed it.
626 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
627 that is needed.
629 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
630 as "const" (a) because they are and (b) because it helps the PHP
631 maintainers who have recently made a script to detect big data structures
632 in the php code that should be moved to the .rodata section. I remembered
633 to update Builducptable as well, so it won't revert if ucptable.h is ever
634 re-created.
636 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
637 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
638 order to be able to cut out the UTF-8 tables in the latter when UTF-8
639 support is not required. This saves 1.5-2K of code, which is important in
640 some applications.
642 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
643 so as not to refer to the tables, even though these functions will never be
644 called when UTF-8 support is disabled. Otherwise there are problems with a
645 shared library.
647 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
649 (a) It was defining its arguments as char * instead of void *.
651 (b) It was assuming that all moves were upwards in memory; this was true
652 a long time ago when I wrote it, but is no longer the case.
654 The emulated memove() is provided for those environments that have neither
655 memmove() nor bcopy(). I didn't think anyone used it these days, but that
656 is clearly not the case, as these two bugs were recently reported.
658 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
659 and Detrail to create the HTML documentation, the .txt form of the man
660 pages, and it removes trailing spaces from listed files. It also creates
661 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
662 case, it wraps all the #defines with #ifndefs. This script should be run
663 before "make dist".
665 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
666 with Unicode property support.
668 (a) For a maximizing quantifier, if the two different cases of the
669 character were of different lengths in their UTF-8 codings (there are
670 some cases like this - I found 11), and the matching function had to
671 back up over a mixture of the two cases, it incorrectly assumed they
672 were both the same length.
674 (b) When PCRE was configured to use the heap rather than the stack for
675 recursion during matching, it was not correctly preserving the data for
676 the other case of a UTF-8 character when checking ahead for a match
677 while processing a minimizing repeat. If the check also involved
678 matching a wide character, but failed, corruption could cause an
679 erroneous result when trying to check for a repeat of the original
680 character.
682 13. Some tidying changes to the testing mechanism:
684 (a) The RunTest script now detects the internal link size and whether there
685 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
686 values substituted by "configure". (The RunGrepTest script already did
687 this for UTF-8.) The configure.ac script no longer substitutes the
688 relevant variables.
690 (b) The debugging options /B and /D in pcretest show the compiled bytecode
691 with length and offset values. This means that the output is different
692 for different internal link sizes. Test 2 is skipped for link sizes
693 other than 2 because of this, bypassing the problem. Unfortunately,
694 there was also a test in test 3 (the locale tests) that used /B and
695 failed for link sizes other than 2. Rather than cut the whole test out,
696 I have added a new /Z option to pcretest that replaces the length and
697 offset values with spaces. This is now used to make test 3 independent
698 of link size. (Test 2 will be tidied up later.)
700 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
701 segmentation fault instead of returning the appropriate error message.
703 15. In multiline mode when the newline sequence was set to "any", the pattern
704 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
705 This doesn't seem right; it now treats the CRLF combination as the line
706 ending, and so does not match in that case. It's only a pattern such as ^$
707 that would hit this one: something like ^ABC$ would have failed after \r
708 and then tried again after \r\n.
710 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
711 in an attempt to make files that differ only in their line terminators
712 compare equal. This works on Linux.
714 17. Under certain error circumstances pcregrep might try to free random memory
715 as it exited. This is now fixed, thanks to valgrind.
717 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
718 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
719 was because its rules for how to advance for /g after matching an empty
720 string at the end of a line did not allow for this case. They now check for
721 it specially.
723 20. pcretest is supposed to handle patterns and data of any length, by
724 extending its buffers when necessary. It was getting this wrong when the
725 buffer for a data line had to be extended.
727 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
728 CRLF as a newline sequence.
730 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
731 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
732 I have nevertheless tidied it up.
734 23. Added some casts to kill warnings from HP-UX ia64 compiler.
736 24. Added a man page for pcre-config.
739 Version 7.0 19-Dec-06
740 ---------------------
742 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
743 moving to gcc 4.1.1.
745 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
746 sys/time.h, which is documented as needed for this function. It doesn't
747 seem to matter on Linux, but it showed up on some releases of OS X.
749 3. It seems that there are systems where bytes whose values are greater than
750 127 match isprint() in the "C" locale. The "C" locale should be the
751 default when a C program starts up. In most systems, only ASCII printing
752 characters match isprint(). This difference caused the output from pcretest
753 to vary, making some of the tests fail. I have changed pcretest so that:
755 (a) When it is outputting text in the compiled version of a pattern, bytes
756 other than 32-126 are always shown as hex escapes.
758 (b) When it is outputting text that is a matched part of a subject string,
759 it does the same, unless a different locale has been set for the match
760 (using the /L modifier). In this case, it uses isprint() to decide.
762 4. Fixed a major bug that caused incorrect computation of the amount of memory
763 required for a compiled pattern when options that changed within the
764 pattern affected the logic of the preliminary scan that determines the
765 length. The relevant options are -x, and -i in UTF-8 mode. The result was
766 that the computed length was too small. The symptoms of this bug were
767 either the PCRE error "internal error: code overflow" from pcre_compile(),
768 or a glibc crash with a message such as "pcretest: free(): invalid next
769 size (fast)". Examples of patterns that provoked this bug (shown in
770 pcretest format) are:
772 /(?-x: )/x
773 /(?x)(?-x: \s*#\s*)/
774 /((?i)[\x{c0}])/8
775 /(?i:[\x{c0}])/8
777 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
778 is now done differently.
780 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
781 wrapper classes; (b) implement a new function in the C++ scanner that is
782 more efficient than the old way of doing things because it avoids levels of
783 recursion in the regex matching; (c) add a paragraph to the documentation
784 for the FullMatch() function.
786 6. The escape sequence \n was being treated as whatever was defined as
787 "newline". Not only was this contrary to the documentation, which states
788 that \n is character 10 (hex 0A), but it also went horribly wrong when
789 "newline" was defined as CRLF. This has been fixed.
791 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
792 was being set to -1 for the "end of line" case (supposedly a value that no
793 character can have). Though this value is never used (the check for end of
794 line is "zero bytes in current character"), it caused compiler complaints.
795 I've changed it to 0xffffffff.
797 8. In pcre_version.c, the version string was being built by a sequence of
798 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
799 string (as it is for production releases) called a macro with an empty
800 argument. The C standard says the result of this is undefined. The gcc
801 compiler treats it as an empty string (which was what was wanted) but it is
802 reported that Visual C gives an error. The source has been hacked around to
803 avoid this problem.
805 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
806 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
807 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
808 of them did).
810 10. Originally, pcretest opened its input and output without "b"; then I was
811 told that "b" was needed in some environments, so it was added for release
812 5.0 to both the input and output. (It makes no difference on Unix-like
813 systems.) Later I was told that it is wrong for the input on Windows. I've
814 now abstracted the modes into two macros, to make it easier to fiddle with
815 them, and removed "b" from the input mode under Windows.
817 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
819 12. Added -help and --help to pcretest as an official way of being reminded
820 of the options.
822 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
823 and pcrecpp.cc because they annoy compilers at high warning levels.
825 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
827 15. Fixed an occurrence of == in configure.ac that should have been = (shell
828 scripts are not C programs :-) and which was not noticed because it works
829 on Linux.
831 16. pcretest is supposed to handle any length of pattern and data line (as one
832 line or as a continued sequence of lines) by extending its input buffer if
833 necessary. This feature was broken for very long pattern lines, leading to
834 a string of junk being passed to pcre_compile() if the pattern was longer
835 than about 50K.
837 17. I have done a major re-factoring of the way pcre_compile() computes the
838 amount of memory needed for a compiled pattern. Previously, there was code
839 that made a preliminary scan of the pattern in order to do this. That was
840 OK when PCRE was new, but as the facilities have expanded, it has become
841 harder and harder to keep it in step with the real compile phase, and there
842 have been a number of bugs (see for example, 4 above). I have now found a
843 cunning way of running the real compile function in a "fake" mode that
844 enables it to compute how much memory it would need, while actually only
845 ever using a few hundred bytes of working memory and without too many
846 tests of the mode. This should make future maintenance and development
847 easier. A side effect of this work is that the limit of 200 on the nesting
848 depth of parentheses has been removed (though this was never a serious
849 limitation, I suspect). However, there is a downside: pcre_compile() now
850 runs more slowly than before (30% or more, depending on the pattern). I
851 hope this isn't a big issue. There is no effect on runtime performance.
853 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
854 newline (only possible for the last line of a file) and it was a
855 pattern that set a locale (followed by /Lsomething), pcretest crashed.
857 19. Added additional timing features to pcretest. (1) The -tm option now times
858 matching only, not compiling. (2) Both -t and -tm can be followed, as a
859 separate command line item, by a number that specifies the number of
860 repeats to use when timing. The default is 50000; this gives better
861 precision, but takes uncomfortably long for very large patterns.
863 20. Extended pcre_study() to be more clever in cases where a branch of a
864 subpattern has no definite first character. For example, (a*|b*)[cd] would
865 previously give no result from pcre_study(). Now it recognizes that the
866 first character must be a, b, c, or d.
868 21. There was an incorrect error "recursive call could loop indefinitely" if
869 a subpattern (or the entire pattern) that was being tested for matching an
870 empty string contained only one non-empty item after a nested subpattern.
871 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
872 incorrectly, because the \d was being skipped in the check.
874 22. The pcretest program now has a new pattern option /B and a command line
875 option -b, which is equivalent to adding /B to every pattern. This causes
876 it to show the compiled bytecode, without the additional information that
877 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
878 is the same as /B/I).
880 23. A new optimization is now able automatically to treat some sequences such
881 as a*b as a*+b. More specifically, if something simple (such as a character
882 or a simple class like \d) has an unlimited quantifier, and is followed by
883 something that cannot possibly match the quantified thing, the quantifier
884 is automatically "possessified".
886 24. A recursive reference to a subpattern whose number was greater than 39
887 went wrong under certain circumstances in UTF-8 mode. This bug could also
888 have affected the operation of pcre_study().
890 25. Realized that a little bit of performance could be had by replacing
891 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
893 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
895 27. Possessive quantifiers such as a++ were previously implemented by turning
896 them into atomic groups such as ($>a+). Now they have their own opcodes,
897 which improves performance. This includes the automatically created ones
898 from 23 above.
900 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
901 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
902 the first matched character to be a colon. This applied both to named and
903 numbered groups.
905 29. The ucpinternal.h header file was missing its idempotency #ifdef.
907 30. I was sent a "project" file called libpcre.a.dev which I understand makes
908 building PCRE on Windows easier, so I have included it in the distribution.
910 31. There is now a check in pcretest against a ridiculously large number being
911 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
912 loop, the loop is abandoned.
914 32. Forward references to subpatterns in conditions such as (?(2)...) where
915 subpattern 2 is defined later cause pcre_compile() to search forwards in
916 the pattern for the relevant set of parentheses. This search went wrong
917 when there were unescaped parentheses in a character class, parentheses
918 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
920 33. "Subroutine" calls and backreferences were previously restricted to
921 referencing subpatterns earlier in the regex. This restriction has now
922 been removed.
924 34. Added a number of extra features that are going to be in Perl 5.10. On the
925 whole, these are just syntactic alternatives for features that PCRE had
926 previously implemented using the Python syntax or my own invention. The
927 other formats are all retained for compatibility.
929 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
930 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
931 also .NET compatible.
933 (b) A recursion or subroutine call to a named group can now be defined as
934 (?&name) as well as (?P>name).
936 (c) A backreference to a named group can now be defined as \k<name> or
937 \k'name' as well as (?P=name). The new forms, as well as being in Perl
938 5.10, are also .NET compatible.
940 (d) A conditional reference to a named group can now use the syntax
941 (?(<name>) or (?('name') as well as (?(name).
943 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
944 groups (named and numbered) that are never evaluated inline, but can be
945 called as "subroutines" from elsewhere. In effect, the DEFINE condition
946 is always false. There may be only one alternative in such a group.
948 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
949 as the simple (?(R). The condition is true only if the most recent
950 recursion is that of the given number or name. It does not search out
951 through the entire recursion stack.
953 (g) The escape \gN or \g{N} has been added, where N is a positive or
954 negative number, specifying an absolute or relative reference.
956 35. Tidied to get rid of some further signed/unsigned compiler warnings and
957 some "unreachable code" warnings.
959 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
960 things, this adds five new scripts.
962 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
963 There were also incompatibilities regarding the handling of \Q..\E inside
964 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
965 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
967 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
968 matches an empty string, and forcibly breaks the loop. There were bugs in
969 this code in non-simple cases. For a pattern such as ^(a()*)* matched
970 against aaaa the result was just "a" rather than "aaaa", for example. Two
971 separate and independent bugs (that affected different cases) have been
972 fixed.
974 39. Refactored the code to abolish the use of different opcodes for small
975 capturing bracket numbers. This is a tidy that I avoided doing when I
976 removed the limit on the number of capturing brackets for 3.5 back in 2001.
977 The new approach is not only tidier, it makes it possible to reduce the
978 memory needed to fix the previous bug (38).
980 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
981 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
982 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
983 mode.
985 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
986 report.
988 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
989 copy construction and assignment in the C++ wrapper.
991 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
992 couple of bugs that could have given wrong results in the "--newline=crlf"
993 case.
995 44. Added a number of casts and did some reorganization of signed/unsigned int
996 variables following suggestions from Dair Grant. Also renamed the variable
997 "this" as "item" because it is a C++ keyword.
999 45. Arranged for dftables to add
1001 #include "pcre_internal.h"
1003 to pcre_chartables.c because without it, gcc 4.x may remove the array
1004 definition from the final binary if PCRE is built into a static library and
1005 dead code stripping is activated.
1007 46. For an unanchored pattern, if a match attempt fails at the start of a
1008 newline sequence, and the newline setting is CRLF or ANY, and the next two
1009 characters are CRLF, advance by two characters instead of one.
1012 Version 6.7 04-Jul-06
1013 ---------------------
1015 1. In order to handle tests when input lines are enormously long, pcretest has
1016 been re-factored so that it automatically extends its buffers when
1017 necessary. The code is crude, but this _is_ just a test program. The
1018 default size has been increased from 32K to 50K.
1020 2. The code in pcre_study() was using the value of the re argument before
1021 testing it for NULL. (Of course, in any sensible call of the function, it
1022 won't be NULL.)
1024 3. The memmove() emulation function in pcre_internal.h, which is used on
1025 systems that lack both memmove() and bcopy() - that is, hardly ever -
1026 was missing a "static" storage class specifier.
1028 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1029 containing an extended class (one that cannot be represented by a bitmap
1030 because it contains high-valued characters or Unicode property items, e.g.
1031 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1032 pattern, but PCRE should not loop if you do not (it no longer does).
1033 [Detail: two cases were found: (a) a repeated subpattern containing an
1034 extended class; (b) a recursive reference to a subpattern that followed a
1035 previous extended class. It wasn't skipping over the extended class
1036 correctly when UTF-8 mode was not set.]
1038 5. A negated single-character class was not being recognized as fixed-length
1039 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1040 compile error "lookbehind assertion is not fixed length".
1042 6. The RunPerlTest auxiliary script was showing an unexpected difference
1043 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1044 write a Perl script that can interpret lines of an input file either as
1045 byte characters or as UTF-8, which is what "perltest" was being required to
1046 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1047 can't do is switch easily at run time between having the "use utf8;" pragma
1048 or not. In the end, I fudged it by using the RunPerlTest script to insert
1049 "use utf8;" explicitly for the UTF-8 tests.
1051 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1052 the end of the subject string, contrary to the documentation and to what
1053 Perl does. This was true of both matching functions. Now it matches only at
1054 the start of the subject and immediately after *internal* newlines.
1056 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1057 a pointer to an int instead of a pointer to an unsigned long int. This
1058 caused problems on 64-bit systems.
1060 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1061 instance of the 'standard' template library not being so standard".
1063 10. There was no check on the number of named subpatterns nor the maximum
1064 length of a subpattern name. The product of these values is used to compute
1065 the size of the memory block for a compiled pattern. By supplying a very
1066 long subpattern name and a large number of named subpatterns, the size
1067 computation could be caused to overflow. This is now prevented by limiting
1068 the length of names to 32 characters, and the number of named subpatterns
1069 to 10,000.
1071 11. Subpatterns that are repeated with specific counts have to be replicated in
1072 the compiled pattern. The size of memory for this was computed from the
1073 length of the subpattern and the repeat count. The latter is limited to
1074 65535, but there was no limit on the former, meaning that integer overflow
1075 could in principle occur. The compiled length of a repeated subpattern is
1076 now limited to 30,000 bytes in order to prevent this.
1078 12. Added the optional facility to have named substrings with the same name.
1080 13. Added the ability to use a named substring as a condition, using the
1081 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1082 are numbers (not recommended). Forward references are permitted.
1084 14. Added forward references in named backreferences (if you see what I mean).
1086 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1087 pattern could run off the end of the subject. For example, the pattern
1088 "(?s)(.{1,5})"8 did this with the subject "ab".
1090 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1091 PCRE_CASELESS was set when matching characters that were quantified with ?
1092 or *.
1094 17. A character class other than a single negated character that had a minimum
1095 but no maximum quantifier - for example [ab]{6,} - was not handled
1096 correctly by pce_dfa_exec(). It would match only one character.
1098 18. A valid (though odd) pattern that looked like a POSIX character
1099 class but used an invalid character after [ (for example [[,abc,]]) caused
1100 pcre_compile() to give the error "Failed: internal error: code overflow" or
1101 in some cases to crash with a glibc free() error. This could even happen if
1102 the pattern terminated after [[ but there just happened to be a sequence of
1103 letters, a binary zero, and a closing ] in the memory that followed.
1105 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1106 over the years. Originally (before any Unicode support), just the bottom 8
1107 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1108 output from "man perlunicode" includes this:
1110 The regular expression compiler produces polymorphic opcodes. That
1111 is, the pattern adapts to the data and automatically switches to
1112 the Unicode character scheme when presented with Unicode data--or
1113 instead uses a traditional byte scheme when presented with byte
1114 data.
1116 Sadly, a wide octal escape does not cause a switch, and in a string with
1117 no other multibyte characters, these octal escapes are treated as before.
1118 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1119 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1120 Unicode string.
1122 I have not perpetrated such confusion in PCRE. Up till now, it took just
1123 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1124 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1125 translate to the appropriate multibyte character.
1127 29. Applied some refactoring to reduce the number of warnings from Microsoft
1128 and Borland compilers. This has included removing the fudge introduced
1129 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1130 a warning about an unused variable.
1132 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1133 characters since release 4.0, because Perl (from release 5.004) does not.
1134 [Or at least, is documented not to: some releases seem to be in conflict
1135 with the documentation.] However, when a pattern was studied with
1136 pcre_study() and all its branches started with \s, PCRE still included VT
1137 as a possible starting character. Of course, this did no harm; it just
1138 caused an unnecessary match attempt.
1140 22. Removed a now-redundant internal flag bit that recorded the fact that case
1141 dependency changed within the pattern. This was once needed for "required
1142 byte" processing, but is no longer used. This recovers a now-scarce options
1143 bit. Also moved the least significant internal flag bit to the most-
1144 significant bit of the word, which was not previously used (hangover from
1145 the days when it was an int rather than a uint) to free up another bit for
1146 the future.
1148 23. Added support for CRLF line endings as well as CR and LF. As well as the
1149 default being selectable at build time, it can now be changed at runtime
1150 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1151 specify that it is scanning data with non-default line endings.
1153 24. Changed the definition of CXXLINK to make it agree with the definition of
1154 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1156 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1157 recursions. This makes a big different to stack usage for some patterns.
1159 26. If a subpattern containing a named recursion or subroutine reference such
1160 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1161 the space required for the compiled pattern went wrong and gave too small a
1162 value. Depending on the environment, this could lead to "Failed: internal
1163 error: code overflow at offset 49" or "glibc detected double free or
1164 corruption" errors.
1166 27. Applied patches from Google (a) to support the new newline modes and (b) to
1167 advance over multibyte UTF-8 characters in GlobalReplace.
1169 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1170 difference for some implementation of PCRE in some Windows version.
1172 29. Added some extra testing facilities to pcretest:
1174 \q<number> in a data line sets the "match limit" value
1175 \Q<number> in a data line sets the "match recursion limt" value
1176 -S <number> sets the stack size, where <number> is in megabytes
1178 The -S option isn't available for Windows.
1181 Version 6.6 06-Feb-06
1182 ---------------------
1184 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1185 in pcreposix.h. I have copied the definition from pcre.h.
1187 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1188 because pcre.h is no longer a built file.
1190 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1191 not normally included in the compiled code.
1194 Version 6.5 01-Feb-06
1195 ---------------------
1197 1. When using the partial match feature with pcre_dfa_exec(), it was not
1198 anchoring the second and subsequent partial matches at the new starting
1199 point. This could lead to incorrect results. For example, with the pattern
1200 /1234/, partially matching against "123" and then "a4" gave a match.
1202 2. Changes to pcregrep:
1204 (a) All non-match returns from pcre_exec() were being treated as failures
1205 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1206 error message is output. Some extra information is given for the
1208 probably the only errors that are likely to be caused by users (by
1209 specifying a regex that has nested indefinite repeats, for instance).
1210 If there are more than 20 of these errors, pcregrep is abandoned.
1212 (b) A binary zero was treated as data while matching, but terminated the
1213 output line if it was written out. This has been fixed: binary zeroes
1214 are now no different to any other data bytes.
1216 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1217 used to set a locale for matching. The --locale=xxxx long option has
1218 been added (no short equivalent) to specify a locale explicitly on the
1219 pcregrep command, overriding the environment variables.
1221 (d) When -B was used with -n, some line numbers in the output were one less
1222 than they should have been.
1224 (e) Added the -o (--only-matching) option.
1226 (f) If -A or -C was used with -c (count only), some lines of context were
1227 accidentally printed for the final match.
1229 (g) Added the -H (--with-filename) option.
1231 (h) The combination of options -rh failed to suppress file names for files
1232 that were found from directory arguments.
1234 (i) Added the -D (--devices) and -d (--directories) options.
1236 (j) Added the -F (--fixed-strings) option.
1238 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1240 (l) Added the --colo(u)r option.
1242 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1243 is not present by default.
1245 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1246 items such as (?R) or (?1), when the recursion could match a number of
1247 alternatives. If it matched one of the alternatives, but subsequently,
1248 outside the recursion, there was a failure, the code tried to back up into
1249 the recursion. However, because of the way PCRE is implemented, this is not
1250 possible, and the result was an incorrect result from the match.
1252 In order to prevent this happening, the specification of recursion has
1253 been changed so that all such subpatterns are automatically treated as
1254 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1256 4. I had overlooked the fact that, in some locales, there are characters for
1257 which isalpha() is true but neither isupper() nor islower() are true. In
1258 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1259 and ordfeminine) are like this. This affected the treatment of \w and \W
1260 when they appeared in character classes, but not when they appeared outside
1261 a character class. The bit map for "word" characters is now created
1262 separately from the results of isalnum() instead of just taking it from the
1263 upper, lower, and digit maps. (Plus the underscore character, of course.)
1265 5. The above bug also affected the handling of POSIX character classes such as
1266 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1267 permanent tables. Instead, the bit maps for such a class were previously
1268 created as the appropriate unions of the upper, lower, and digit bitmaps.
1269 Now they are created by subtraction from the [[:word:]] class, which has
1270 its own bitmap.
1272 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1273 It is created by subtracting the vertical space characters (\x09, \x0a,
1274 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1275 subtraction was done in the overall bitmap for a character class, meaning
1276 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1277 be recognized. This bug has been fixed.
1279 7. Patches from the folks at Google:
1281 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1282 real life, but is still worth protecting against".
1284 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1285 regular expressions".
1287 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1288 have it.
1290 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1291 "configure" and the latter not, in order to fix a problem somebody had
1292 with compiling the Arg class on HP-UX.
1294 (e) Improve the error-handling of the C++ wrapper a little bit.
1296 (f) New tests for checking recursion limiting.
1298 8. The pcre_memmove() function, which is used only if the environment does not
1299 have a standard memmove() function (and is therefore rarely compiled),
1300 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1301 returning a result (though PCRE never actually uses the result).
1303 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1304 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1305 returned instead of calling malloc() with an overflowing number that would
1306 most likely cause subsequent chaos.
1308 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1310 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1311 with this option is matched, the nmatch and pmatch options of regexec() are
1312 ignored.
1314 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1315 provided in case anyone wants to the the POSIX interface with UTF-8
1316 strings.
1318 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1319 C++ linking (needed for some HP-UX environments).
1321 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1322 (unused parameter) and in the pcre_printint() function (omitted "default"
1323 switch label when the default is to do nothing).
1325 15. Added some code to make it possible, when PCRE is compiled as a C++
1326 library, to replace subject pointers for pcre_exec() with a smart pointer
1327 class, thus making it possible to process discontinuous strings.
1329 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1330 much the same function. They were added by different people who were trying
1331 to make PCRE easy to compile on non-Unix systems. It has been suggested
1332 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1333 for compiling on Windows systems. I have therefore replaced it with
1334 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1335 defaults to "extern" for C or "extern C" for C++, which works fine on
1336 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1337 SCOPE with something explicit in config.h. In addition:
1339 (a) pcreposix.h still had just "extern" instead of either of these macros;
1340 I have replaced it with PCRE_DATA_SCOPE.
1342 (b) Functions such as _pcre_xclass(), which are internal to the library,
1343 but external in the C sense, all had PCRE_EXPORT in their definitions.
1344 This is apparently wrong for the Windows case, so I have removed it.
1345 (It makes no difference on Unix-like systems.)
1347 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1348 of recursive calls to match(). This is different to MATCH_LIMIT because
1349 that limits the total number of calls to match(), not all of which increase
1350 the depth of recursion. Limiting the recursion depth limits the amount of
1351 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1352 when PCRE is compiled, and changed at run time. A patch from Google adds
1353 this functionality to the C++ interface.
1355 18. Changes to the handling of Unicode character properties:
1357 (a) Updated the table to Unicode 4.1.0.
1359 (b) Recognize characters that are not in the table as "Cn" (undefined).
1361 (c) I revised the way the table is implemented to a much improved format
1362 which includes recognition of ranges. It now supports the ranges that
1363 are defined in UnicodeData.txt, and it also amalgamates other
1364 characters into ranges. This has reduced the number of entries in the
1365 table from around 16,000 to around 3,000, thus reducing its size
1366 considerably. I realized I did not need to use a tree structure after
1367 all - a binary chop search is just as efficient. Having reduced the
1368 number of entries, I extended their size from 6 bytes to 8 bytes to
1369 allow for more data.
1371 (d) Added support for Unicode script names via properties such as \p{Han}.
1373 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1374 matching that character.
1376 20. When matching a repeated Unicode property with a minimum greater than zero,
1377 (for example \pL{2,}), PCRE could look past the end of the subject if it
1378 reached it while seeking the minimum number of characters. This could
1379 happen only if some of the characters were more than one byte long, because
1380 there is a check for at least the minimum number of bytes.
1382 21. Refactored the implementation of \p and \P so as to be more general, to
1383 allow for more different types of property in future. This has changed the
1384 compiled form incompatibly. Anybody with saved compiled patterns that use
1385 \p or \P will have to recompile them.
1387 22. Added "Any" and "L&" to the supported property types.
1389 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1390 but give a compile time error if the value is greater than 0xff.
1392 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1393 accidentally not being installed or uninstalled.
1395 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1396 made were to insert the current release number. This seemed silly, because
1397 it made things harder for people building PCRE on systems that don't run
1398 "configure". I have turned pcre.h into a distributed file, no longer built
1399 by "configure", with the version identification directly included. There is
1400 no longer a pcre.h.in file.
1402 However, this change necessitated a change to the pcre-config script as
1403 well. It is built from pcre-config.in, and one of the substitutions was the
1404 release number. I have updated configure.ac so that ./configure now finds
1405 the release number by grepping pcre.h.
1407 26. Added the ability to run the tests under valgrind.
1410 Version 6.4 05-Sep-05
1411 ---------------------
1413 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1414 "--" to be printed when multiple files were scanned, even when none of the
1415 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1416 consider it to be a bug, and have restored the previous behaviour.
1418 2. A couple of code tidies to get rid of compiler warnings.
1420 3. The pcretest program used to cheat by referring to symbols in the library
1421 whose names begin with _pcre_. These are internal symbols that are not
1422 really supposed to be visible externally, and in some environments it is
1423 possible to suppress them. The cheating is now confined to including
1424 certain files from the library's source, which is a bit cleaner.
1426 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1427 file's purpose clearer.
1429 5. Reorganized pcre_ucp_findchar().
1432 Version 6.3 15-Aug-05
1433 ---------------------
1435 1. The file libpcre.pc.in did not have general read permission in the tarball.
1437 2. There were some problems when building without C++ support:
1439 (a) If C++ support was not built, "make install" and "make test" still
1440 tried to test it.
1442 (b) There were problems when the value of CXX was explicitly set. Some
1443 changes have been made to try to fix these, and ...
1445 (c) --disable-cpp can now be used to explicitly disable C++ support.
1447 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1448 backslash in a target when C++ was disabled. This confuses some
1449 versions of "make", apparently. Using an intermediate variable solves
1450 this. (Same for CPP_LOBJ.)
1452 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1453 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1454 necessary on certain architectures.
1456 4. Added a setting of -export-symbols-regex to the link command to remove
1457 those symbols that are exported in the C sense, but actually are local
1458 within the library, and not documented. Their names all begin with
1459 "_pcre_". This is not a perfect job, because (a) we have to except some
1460 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1461 available (and never for static libraries). I have made a note to try to
1462 find a way round (a) in the future.
1465 Version 6.2 01-Aug-05
1466 ---------------------
1468 1. There was no test for integer overflow of quantifier values. A construction
1469 such as {1111111111111111} would give undefined results. What is worse, if
1470 a minimum quantifier for a parenthesized subpattern overflowed and became
1471 negative, the calculation of the memory size went wrong. This could have
1472 led to memory overwriting.
1474 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1476 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1477 operating environments where this matters.
1479 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1480 PCRE options from within the C++ wrapper.
1482 5. Named capturing subpatterns were not being correctly counted when a pattern
1483 was compiled. This caused two problems: (a) If there were more than 100
1484 such subpatterns, the calculation of the memory needed for the whole
1485 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1486 back references of the form \12, where the number was greater than 9, were
1487 not recognized as back references, even though there were sufficient
1488 previous subpatterns.
1490 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1491 versions of gcc, e.g. 2.95.4.
1494 Version 6.1 21-Jun-05
1495 ---------------------
1497 1. There was one reference to the variable "posix" in pcretest.c that was not
1498 surrounded by "#if !defined NOPOSIX".
1500 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1501 the cross-check on the old pcre_info() function, for the benefit of the
1502 cut-down version of PCRE that is currently imported into Exim.
1504 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1505 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1506 patterns that start (?i) but I don't think that matters. The use of (?i) is
1507 just an example; this all applies to the other options as well.
1509 4. Since libtool seems to echo the compile commands it is issuing, the output
1510 from "make" can be reduced a bit by putting "@" in front of each libtool
1511 compile command.
1513 5. Patch from the folks at Google for configure.in to be a bit more thorough
1514 in checking for a suitable C++ installation before trying to compile the
1515 C++ stuff. This should fix a reported problem when a compiler was present,
1516 but no suitable headers.
1518 6. The man pages all had just "PCRE" as their title. I have changed them to
1519 be the relevant file name. I have also arranged that these names are
1520 retained in the file doc/pcre.txt, which is a concatenation in text format
1521 of all the man pages except the little individual ones for each function.
1523 7. The NON-UNIX-USE file had not been updated for the different set of source
1524 files that come with release 6. I also added a few comments about the C++
1525 wrapper.
1528 Version 6.0 07-Jun-05
1529 ---------------------
1531 1. Some minor internal re-organization to help with my DFA experiments.
1533 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1534 didn't matter for the library itself when fully configured, but did matter
1535 when compiling without UCP support, or within Exim, where the ucp files are
1536 not imported.
1538 3. Refactoring of the library code to split up the various functions into
1539 different source modules. The addition of the new DFA matching code (see
1540 below) to a single monolithic source would have made it really too
1541 unwieldy, quite apart from causing all the code to be include in a
1542 statically linked application, when only some functions are used. This is
1543 relevant even without the DFA addition now that patterns can be compiled in
1544 one application and matched in another.
1546 The downside of splitting up is that there have to be some external
1547 functions and data tables that are used internally in different modules of
1548 the library but which are not part of the API. These have all had their
1549 names changed to start with "_pcre_" so that they are unlikely to clash
1550 with other external names.
1552 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1553 a different (DFA) algorithm. Although it is slower than the original
1554 function, it does have some advantages for certain types of matching
1555 problem.
1557 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1558 including restarting after a partial match.
1560 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1561 defined when compiling for Windows was sent to me. I have put it into the
1562 code, though I have no means of testing or verifying it.
1564 7. Added the pcre_refcount() auxiliary function.
1566 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1567 match before or at the first newline in the subject string. In pcretest,
1568 the /f option on a pattern can be used to set this.
1570 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1571 would behave wrongly. This has been present in PCRE since release 4.0.
1573 10. A number of changes to the pcregrep command:
1575 (a) Refactored how -x works; insert ^(...)$ instead of setting
1576 PCRE_ANCHORED and checking the length, in preparation for adding
1577 something similar for -w.
1579 (b) Added the -w (match as a word) option.
1581 (c) Refactored the way lines are read and buffered so as to have more
1582 than one at a time available.
1584 (d) Implemented a pcregrep test script.
1586 (e) Added the -M (multiline match) option. This allows patterns to match
1587 over several lines of the subject. The buffering ensures that at least
1588 8K, or the rest of the document (whichever is the shorter) is available
1589 for matching (and similarly the previous 8K for lookbehind assertions).
1591 (f) Changed the --help output so that it now says
1593 -w, --word-regex(p)
1595 instead of two lines, one with "regex" and the other with "regexp"
1596 because that confused at least one person since the short forms are the
1597 same. (This required a bit of code, as the output is generated
1598 automatically from a table. It wasn't just a text change.)
1600 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1601 option but starts with a hyphen. Could be a pattern or a path name
1602 starting with a hyphen, for instance.
1604 (h) "-" can be given as a file name to represent stdin.
1606 (i) When file names are being printed, "(standard input)" is used for
1607 the standard input, for compatibility with GNU grep. Previously
1608 "<stdin>" was used.
1610 (j) The option --label=xxx can be used to supply a name to be used for
1611 stdin when file names are being printed. There is no short form.
1613 (k) Re-factored the options decoding logic because we are going to add
1614 two more options that take data. Such options can now be given in four
1615 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1617 (l) Added the -A, -B, and -C options for requesting that lines of context
1618 around matches be printed.
1620 (m) Added the -L option to print the names of files that do not contain
1621 any matching lines, that is, the complement of -l.
1623 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1624 continue to scan other files.
1626 (o) The -s option was incorrectly implemented. For compatibility with other
1627 greps, it now suppresses the error message for a non-existent or non-
1628 accessible file (but not the return code). There is a new option called
1629 -q that suppresses the output of matching lines, which was what -s was
1630 previously doing.
1632 (p) Added --include and --exclude options to specify files for inclusion
1633 and exclusion when recursing.
1635 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1636 Hopefully, it now does.
1638 12. Missing cast in pcre_study().
1640 13. Added an "uninstall" target to the makefile.
1642 14. Replaced "extern" in the function prototypes in Makefile.in with
1643 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1644 world, but is set differently for Windows.
1646 15. Added a second compiling function called pcre_compile2(). The only
1647 difference is that it has an extra argument, which is a pointer to an
1648 integer error code. When there is a compile-time failure, this is set
1649 non-zero, in addition to the error test pointer being set to point to an
1650 error message. The new argument may be NULL if no error number is required
1651 (but then you may as well call pcre_compile(), which is now just a
1652 wrapper). This facility is provided because some applications need a
1653 numeric error indication, but it has also enabled me to tidy up the way
1654 compile-time errors are handled in the POSIX wrapper.
1656 16. Added VPATH=.libs to the makefile; this should help when building with one
1657 prefix path and installing with another. (Or so I'm told by someone who
1658 knows more about this stuff than I do.)
1660 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1661 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1662 match everything, including newlines. This is not POSIX-compatible, but
1663 somebody wanted the feature. From pcretest it can be activated by using
1664 both the P and the s flags.
1666 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1668 19. libpcre.pc was being incorrectly installed as executable.
1670 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1671 it now also looks for '\r' so that it will work unmodified on Windows.
1673 21. Added Google's contributed C++ wrapper to the distribution.
1675 22. Added some untidy missing memory free() calls in pcretest, to keep
1676 Electric Fence happy when testing.
1680 Version 5.0 13-Sep-04
1681 ---------------------
1683 1. Internal change: literal characters are no longer packed up into items
1684 containing multiple characters in a single byte-string. Each character
1685 is now matched using a separate opcode. However, there may be more than one
1686 byte in the character in UTF-8 mode.
1688 2. The pcre_callout_block structure has two new fields: pattern_position and
1689 next_item_length. These contain the offset in the pattern to the next match
1690 item, and its length, respectively.
1692 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1693 insertion of callouts before each pattern item. Added the /C option to
1694 pcretest to make use of this.
1696 4. On the advice of a Windows user, the lines
1698 #if defined(_WIN32) || defined(WIN32)
1699 _setmode( _fileno( stdout ), 0x8000 );
1700 #endif /* defined(_WIN32) || defined(WIN32) */
1702 have been added to the source of pcretest. This apparently does useful
1703 magic in relation to line terminators.
1705 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1706 for the benefit of those environments where the "b" makes a difference.
1708 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1709 to know about it. I have put a hack into configure.in that adds in code
1710 to set GCC=yes if CC=icc. This seems to end up at a point in the
1711 generated configure script that is early enough to affect the setting of
1712 compiler options, which is what is needed, but I have no means of testing
1713 whether it really works. (The user who reported this had patched the
1714 generated configure script, which of course I cannot do.)
1716 LATER: After change 22 below (new libtool files), the configure script
1717 seems to know about icc (and also ecc). Therefore, I have commented out
1718 this hack in configure.in.
1720 7. Added support for pkg-config (2 patches were sent in).
1722 8. Negated POSIX character classes that used a combination of internal tables
1723 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1724 [[:^ascii]]. Typically, they would match almost any characters. The other
1725 POSIX classes were not broken in this way.
1727 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1728 to find the match, as PCRE was deluded into thinking that the match had to
1729 start at the start point or following a newline. The same bug applied to
1730 patterns with negative forward assertions or any backward assertions
1731 preceding ".*" at the start, unless the pattern required a fixed first
1732 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1734 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1735 starting at the last subject character, bytes beyond the end of the subject
1736 string were read.
1738 11. Renamed the variable "class" as "classbits" to make life easier for C++
1739 users. (Previously there was a macro definition, but it apparently wasn't
1740 enough.)
1742 12. Added the new field "tables" to the extra data so that tables can be passed
1743 in at exec time, or the internal tables can be re-selected. This allows
1744 a compiled regex to be saved and re-used at a later time by a different
1745 program that might have everything at different addresses.
1747 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1748 -R library as well as a -L library.
1750 14. The debugging options of pcretest (-d on the command line or D on a
1751 pattern) showed incorrect output for anything following an extended class
1752 that contained multibyte characters and which was followed by a quantifier.
1754 15. Added optional support for general category Unicode character properties
1755 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1756 support. It adds about 90K to the size of the library. The meanings of the
1757 inbuilt class escapes such as \d and \s have NOT been changed.
1759 16. Updated pcredemo.c to include calls to free() to release the memory for the
1760 compiled pattern.
1762 17. The generated file chartables.c was being created in the source directory
1763 instead of in the building directory. This caused the build to fail if the
1764 source directory was different from the building directory, and was
1765 read-only.
1767 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1768 file. No doubt somebody will tell me if they don't make sense... Also added
1769 Dan Mooney's comments about building on OpenVMS.
1771 19. Added support for partial matching via the PCRE_PARTIAL option for
1772 pcre_exec() and the \P data escape in pcretest.
1774 20. Extended pcretest with 3 new pattern features:
1776 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1777 write the compiled pattern to the file whose name is "rest-of-line".
1778 This is a straight binary dump of the data, with the saved pointer to
1779 the character tables forced to be NULL. The study data, if any, is
1780 written too. After writing, pcretest reads a new pattern.
1782 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1783 compiled pattern from the given file. There must not be any
1784 occurrences of "<" in the file name (pretty unlikely); if there are,
1785 pcretest will instead treat the initial "<" as a pattern delimiter.
1786 After reading in the pattern, pcretest goes on to read data lines as
1787 usual.
1789 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1790 and 16-bit fields in a compiled pattern, to simulate a pattern that
1791 was compiled on a host of opposite endianness.
1793 21. The pcre-exec() function can now cope with patterns that were compiled on
1794 hosts of opposite endianness, with this restriction:
1796 As for any compiled expression that is saved and used later, the tables
1797 pointer field cannot be preserved; the extra_data field in the arguments
1798 to pcre_exec() should be used to pass in a tables address if a value
1799 other than the default internal tables were used at compile time.
1801 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1802 now diagnosed as an error. Previously, most of the time, a negative number
1803 would have been treated as zero, but if in addition "ovector" was passed as
1804 NULL, a crash could occur.
1806 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1807 new versions from the libtool 1.5 distribution (the last one is a copy of
1808 a file called libtool.m4). This seems to have fixed the need to patch
1809 "configure" to support Darwin 1.3 (which I used to do). However, I still
1810 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1811 workstation).
1813 24. Changed the PCRE licence to be the more standard "BSD" licence.
1816 Version 4.5 01-Dec-03
1817 ---------------------
1819 1. There has been some re-arrangement of the code for the match() function so
1820 that it can be compiled in a version that does not call itself recursively.
1821 Instead, it keeps those local variables that need separate instances for
1822 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1823 needs to "recurse". Keeping track of where control must go is done by means
1824 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1825 hide most of the details from the main code, and operates only if
1826 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1827 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1828 operating.
1830 To make it easier for callers to provide specially tailored get/free
1831 functions for this usage, two new functions, pcre_stack_malloc, and
1832 pcre_stack_free, are used. They are always called in strict stacking order,
1833 and the size of block requested is always the same.
1835 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1836 PCRE has been compiled to use the stack or the heap for recursion. The
1837 -C option of pcretest uses this to show which version is compiled.
1839 A new data escape \S, is added to pcretest; it causes the amounts of store
1840 obtained and freed by both kinds of malloc/free at match time to be added
1841 to the output.
1843 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1844 what's available on my current Linux desktop machine.
1846 3. When matching a UTF-8 string, the test for a valid string at the start has
1847 been extended. If start_offset is not zero, PCRE now checks that it points
1848 to a byte that is the start of a UTF-8 character. If not, it returns
1849 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1850 this is necessary because there may be backward assertions in the pattern.
1851 When matching the same subject several times, it may save resources to use
1852 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1854 4. The code for checking the validity of UTF-8 strings has been tightened so
1855 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1856 containing "overlong sequences".
1858 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1859 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1860 should have been "&", but it just so happened that all the cases this let
1861 through by mistake were picked up later in the function.
1863 6. I had used a variable called "isblank" - this is a C99 function, causing
1864 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1866 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1867 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1868 so the version is shown; (c) stop "make test" echoing "./RunTest".
1870 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1872 9. The prototype for memmove() for systems that don't have it was using
1873 size_t, but the inclusion of the header that defines size_t was later. I've
1874 moved the #includes for the C headers earlier to avoid this.
1876 10. Added some adjustments to the code to make it easier to compiler on certain
1877 special systems:
1879 (a) Some "const" qualifiers were missing.
1880 (b) Added the macro EXPORT before all exported functions; by default this
1881 is defined to be empty.
1882 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1883 that it reads its output file name as an argument instead of writing
1884 to the standard output and assuming this can be redirected.
1886 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1887 class containing characters with values greater than 255, PCRE compilation
1888 went into a loop.
1890 12. A recursive reference to a subpattern that was within another subpattern
1891 that had a minimum quantifier of zero caused PCRE to crash. For example,
1892 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1893 recursion. If the recursively-called subpattern itself had a zero repeat,
1894 that was OK.
1896 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1897 buffer into which it was copied (for escape processing) was still set at
1898 1024, so long lines caused crashes.
1900 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1901 "internal error: code overflow...". This applied to any character class
1902 that was followed by a possessive quantifier.
1904 15. Modified the Makefile to add libpcre.la as a prerequisite for
1905 libpcreposix.la because I was told this is needed for a parallel build to
1906 work.
1908 16. If a pattern that contained .* following optional items at the start was
1909 studied, the wrong optimizing data was generated, leading to matching
1910 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1911 matching string must start with a or b or c. The correct conclusion for
1912 this pattern is that a match can start with any character.
1915 Version 4.4 13-Aug-03
1916 ---------------------
1918 1. In UTF-8 mode, a character class containing characters with values between
1919 127 and 255 was not handled correctly if the compiled pattern was studied.
1920 In fixing this, I have also improved the studying algorithm for such
1921 classes (slightly).
1923 2. Three internal functions had redundant arguments passed to them. Removal
1924 might give a very teeny performance improvement.
1926 3. Documentation bug: the value of the capture_top field in a callout is *one
1927 more than* the number of the hightest numbered captured substring.
1929 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1930 in incorrectly linking with a previously installed version. They now link
1931 explicitly with libpcre.la.
1933 5. configure.in no longer needs to recognize Cygwin specially.
1935 6. A problem in pcre.in for Windows platforms is fixed.
1937 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1938 pcretest, it used to include the size of the study block as part of its
1939 output. Unfortunately, the structure contains a field that has a different
1940 size on different hardware architectures. This meant that the tests that
1941 showed this size failed. As the block is currently always of a fixed size,
1942 this information isn't actually particularly useful in pcretest output, so
1943 I have just removed it.
1945 8. Three pre-processor statements accidentally did not start in column 1.
1946 Sadly, there are *still* compilers around that complain, even though
1947 standard C has not required this for well over a decade. Sigh.
1949 9. In pcretest, the code for checking callouts passed small integers in the
1950 callout_data field, which is a void * field. However, some picky compilers
1951 complained about the casts involved for this on 64-bit systems. Now
1952 pcretest passes the address of the small integer instead, which should get
1953 rid of the warnings.
1955 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1956 both compile and run time, and gives an error if an invalid UTF-8 sequence
1957 is found. There is a option for disabling this check in cases where the
1958 string is known to be correct and/or the maximum performance is wanted.
1960 11. In response to a bug report, I changed one line in Makefile.in from
1962 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1963 to
1964 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1966 to look similar to other lines, but I have no way of telling whether this
1967 is the right thing to do, as I do not use Windows. No doubt I'll get told
1968 if it's wrong...
1971 Version 4.3 21-May-03
1972 ---------------------
1974 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1975 Makefile.
1977 2. Some refactoring to improve the quality of the code:
1979 (i) The utf8_table... variables are now declared "const".
1981 (ii) The code for \cx, which used the "case flipping" table to upper case
1982 lower case letters, now just substracts 32. This is ASCII-specific,
1983 but the whole concept of \cx is ASCII-specific, so it seems
1984 reasonable.
1986 (iii) PCRE was using its character types table to recognize decimal and
1987 hexadecimal digits in the pattern. This is silly, because it handles
1988 only 0-9, a-f, and A-F, but the character types table is locale-
1989 specific, which means strange things might happen. A private
1990 table is now used for this - though it costs 256 bytes, a table is
1991 much faster than multiple explicit tests. Of course, the standard
1992 character types table is still used for matching digits in subject
1993 strings against \d.
1995 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1996 ending in _t are). So I've renamed it as ESC_tee.
1998 3. The first argument for regexec() in the POSIX wrapper should have been
1999 defined as "const".
2001 4. Changed pcretest to use malloc() for its buffers so that they can be
2002 Electric Fenced for debugging.
2004 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2005 to read one or more bytes before the start of the subject string. Often this
2006 had no effect on PCRE's behaviour, but in some circumstances it could
2007 provoke a segmentation fault.
2009 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2010 to try to read one or more bytes before the start of the subject string.
2012 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2013 UTF-8 support could misbehave in various ways if the subject string
2014 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2015 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2016 back over UTF-8 characters.)
2019 Version 4.2 14-Apr-03
2020 ---------------------
2022 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2024 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2025 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2026 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2027 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2028 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2030 Note: automatic setting of the BUILD variables is not yet working
2031 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2032 compile-time but not at link-time
2033 [LINK]: use for linking executables only
2034 make different versions for Windows and non-Windows
2035 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2036 libraries
2037 [LINK_FOR_BUILD]: new variable
2038 [OBJEXT]: use throughout
2039 [EXEEXT]: use throughout
2040 <winshared>: new target
2041 <wininstall>: new target
2042 <dftables.o>: use native compiler
2043 <dftables>: use native linker
2044 <install>: handle Windows platform correctly
2045 <clean>: ditto
2046 <check>: ditto
2047 copy DLL to top builddir before testing
2049 As part of these changes, -no-undefined was removed again. This was reported
2050 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2051 in any case.
2053 3. Some tidies to get rid of compiler warnings:
2055 . In the match_data structure, match_limit was an unsigned long int, whereas
2056 match_call_count was an int. I've made them both unsigned long ints.
2058 . In pcretest the fact that a const uschar * doesn't automatically cast to
2059 a void * provoked a warning.
2061 . Turning on some more compiler warnings threw up some "shadow" variables
2062 and a few more missing casts.
2064 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2065 option, a class that contained a single character with a value between 128
2066 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2068 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2069 option, a class that contained several characters, but with at least one
2070 whose value was between 128 and 255 caused PCRE to crash.
2073 Version 4.1 12-Mar-03
2074 ---------------------
2076 1. Compiling with gcc -pedantic found a couple of places where casts were
2077 needed, and a string in dftables.c that was longer than standard compilers are
2078 required to support.
2080 2. Compiling with Sun's compiler found a few more places where the code could
2081 be tidied up in order to avoid warnings.
2083 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2084 first of these names is deprecated in the latest Autoconf in favour of the name
2085 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2086 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2087 analogy I have changed it to CFLAGS_FOR_BUILD.
2089 4. Added -no-undefined to the linking command in the Makefile, because this is
2090 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2091 linking step for the pcreposix library.
2093 5. PCRE was failing to diagnose the case of two named groups with the same
2094 name.
2096 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2097 literal character that is needed in the subject for a match, and scans along to
2098 ensure that it is present before embarking on the full matching process. This
2099 saves time in cases of nested unlimited repeats that are never going to match.
2100 Problem: the scan can take a lot of time if the subject is very long (e.g.
2101 megabytes), thus penalizing straightforward matches. It is now done only if the
2102 amount of subject to be scanned is less than 1000 bytes.
2104 7. A lesser problem with the same optimization is that it was recording the
2105 first character of an anchored pattern as "needed", thus provoking a search
2106 right along the subject, even when the first match of the pattern was going to
2107 fail. The "needed" character is now not set for anchored patterns, unless it
2108 follows something in the pattern that is of non-fixed length. Thus, it still
2109 fulfils its original purpose of finding quick non-matches in cases of nested
2110 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2113 Version 4.0 17-Feb-03
2114 ---------------------
2116 1. If a comment in an extended regex that started immediately after a meta-item
2117 extended to the end of string, PCRE compiled incorrect data. This could lead to
2118 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2120 2. Moved to autoconf 2.53 and libtool 1.4.2.
2122 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2123 the special perltest8 script is no longer needed - all the tests can be run
2124 from a single perltest script.
2126 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2127 by \s. It has now been removed in PCRE. This means it isn't recognized as
2128 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2129 class [:space:] *does* include VT, thereby creating a mess.
2131 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2132 space and tab.
2134 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2135 its new features into the main test script, reducing the number of scripts.
2137 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2138 were backward compatible, and made the (?i) apply to the whole pattern, as if
2139 /i were given. Now it behaves more logically, and applies the option setting
2140 only to what follows. PCRE has been changed to follow suit. However, if it
2141 finds options settings right at the start of the pattern, it extracts them into
2142 the global options, as before. Thus, they show up in the info data.
2144 8. Added support for the \Q...\E escape sequence. Characters in between are
2145 treated as literals. This is slightly different from Perl in that $ and @ are
2146 also handled as literals inside the quotes. In Perl, they will cause variable
2147 interpolation. Note the following examples:
2149 Pattern PCRE matches Perl matches
2151 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2152 \Qabc\$xyz\E abc\$xyz abc\$xyz
2153 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2155 For compatibility with Perl, \Q...\E sequences are recognized inside character
2156 classes as well as outside them.
2158 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2159 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2160 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2161 signed/unsigned warnings.
2163 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2164 option for pcretest, so I've replaced it by a simple function that does just
2165 that job.
2167 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2168 "pcregrep -".
2170 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2171 Java package. This provides some syntactic sugar for simple cases of what my
2172 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2173 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2174 item, you can use this simplified notation. Note that only makes sense with
2175 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2176 greediness, whatever the setting of the PCRE_UNGREEDY option.
2178 13. A change of greediness default within a pattern was not taking effect at
2179 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2180 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2181 was abstracted outside.
2183 14. PCRE now supports the \G assertion. It is true when the current matching
2184 position is at the start point of the match. This differs from \A when the
2185 starting offset is non-zero. Used with the /g option of pcretest (or similar
2186 code), it works in the same way as it does for Perl's /g option. If all
2187 alternatives of a regex begin with \G, the expression is anchored to the start
2188 match position, and the "anchored" flag is set in the compiled expression.
2190 15. Some bugs concerning the handling of certain option changes within patterns
2191 have been fixed. These applied to options other than (?ims). For example,
2192 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2193 been the other way round. Some of this was related to change 7 above.
2195 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2196 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2197 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2198 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2200 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2201 mode. Unlike ".", it always matches newline, whatever the setting of
2202 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2203 assertions. Perl allows it, but it doesn't (in general) work because it can't
2204 calculate the length of the lookbehind. At least, that's the case for Perl
2205 5.8.0 - I've been told they are going to document that it doesn't work in
2206 future.
2208 18. Added an error diagnosis for escapes that PCRE does not support: these are
2209 \L, \l, \N, \P, \p, \U, \u, and \X.
2211 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2212 reading past the end of the pattern in cases such as /[abcd/.
2214 20. PCRE was getting more memory than necessary for patterns with classes that
2215 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2217 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2218 compiling PCRE for use with Virtual Pascal.
2220 22. Small fix to the Makefile to make it work properly if the build is done
2221 outside the source tree.
2223 23. Added a new extension: a condition to go with recursion. If a conditional
2224 subpattern starts with (?(R) the "true" branch is used if recursion has
2225 happened, whereas the "false" branch is used only at the top level.
2227 24. When there was a very long string of literal characters (over 255 bytes
2228 without UTF support, over 250 bytes with UTF support), the computation of how
2229 much memory was required could be incorrect, leading to segfaults or other
2230 strange effects.
2232 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2233 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2234 there was a subsequent back reference to those brackets. This meant that, for
2235 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2236 possible to check for precisely this case. All we can do is abandon the
2237 optimization if .* occurs inside capturing brackets when there are any back
2238 references whatsoever. (See below for a better fix that came later.)
2240 26. The handling of the optimization for finding the first character of a
2241 non-anchored pattern, and for finding a character that is required later in the
2242 match were failing in some cases. This didn't break the matching; it just
2243 failed to optimize when it could. The way this is done has been re-implemented.
2245 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2247 28. Added a new feature that provides some of the functionality that Perl
2248 provides with (?{...}). The facility is termed a "callout". The way it is done
2249 in PCRE is for the caller to provide an optional function, by setting
2250 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2251 global variable. By default it is unset, which disables all calling out. To get
2252 the function called, the regex must include (?C) at appropriate points. This
2253 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2254 This provides a means of identifying different callout points. When PCRE
2255 reaches such a point in the regex, if pcre_callout has been set, the external
2256 function is called. It is provided with data in a structure called
2257 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2258 matching continues; if it returns a non-zero value, the match at the current
2259 point fails. However, backtracking will occur if possible. [This was changed
2260 later and other features added - see item 49 below.]
2262 29. pcretest is upgraded to test the callout functionality. It provides a
2263 callout function that displays information. By default, it shows the start of
2264 the match and the current position in the text. There are some new data escapes
2265 to vary what happens:
2267 \C+ in addition, show current contents of captured substrings
2268 \C- do not supply a callout function
2269 \C!n return 1 when callout number n is reached
2270 \C!n!m return 1 when callout number n is reached for the mth time
2272 30. If pcregrep was called with the -l option and just a single file name, it
2273 output "<stdin>" if a match was found, instead of the file name.
2275 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2276 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2277 pcre_exec(). This saves a malloc/free per call. The default value of
2278 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2279 when configuring.
2281 32. The default maximum size of a compiled pattern is 64K. There have been a
2282 few cases of people hitting this limit. The code now uses macros to handle the
2283 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2284 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2285 configuring. Tests 2 and 5 work only with 2-byte links because they output
2286 debugging information about compiled patterns.
2288 33. Internal code re-arrangements:
2290 (a) Moved the debugging function for printing out a compiled regex into
2291 its own source file (printint.c) and used #include to pull it into
2292 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2293 separate copies.
2295 (b) Defined the list of op-code names for debugging as a macro in
2296 internal.h so that it is next to the definition of the opcodes.
2298 (c) Defined a table of op-code lengths for simpler skipping along compiled
2299 code. This is again a macro in internal.h so that it is next to the
2300 definition of the opcodes.
2302 34. Added support for recursive calls to individual subpatterns, along the
2303 lines of Robin Houston's patch (but implemented somewhat differently).
2305 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2306 allow it to read and process whole directories in Win32. This code was
2307 contributed by Lionel Fourquaux; it has not been tested by me.
2309 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2310 used to name a group. Names consist of alphanumerics and underscores, and must
2311 be unique. Back references use the syntax (?P=name) and recursive calls use
2312 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2313 numbers. The function pcre_fullinfo() can be used after compilation to extract
2314 a name/number map. There are three relevant calls:
2316 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2317 PCRE_INFO_NAMECOUNT yields the number of entries
2318 PCRE_INFO_NAMETABLE yields a pointer to the map.
2320 The map is a vector of fixed-size entries. The size of each entry depends on
2321 the length of the longest name used. The first two bytes of each entry are the
2322 group number, most significant byte first. There follows the corresponding
2323 name, zero terminated. The names are in alphabetical order.
2325 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2326 case instead of 255. Making it the same both with and without UTF-8 support
2327 means that the same test output works with both.
2329 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2330 calling malloc() with a zero argument.
2332 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2333 optimization. I've improved things by keeping a bitmap of backreferences with
2334 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2335 fact referenced, the optimization can be applied. It is unlikely that a
2336 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2337 the match to follow \n) will appear inside brackets with a number greater than
2338 31, but if it does, any back reference > 31 suppresses the optimization.
2340 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2341 of disabling numbered capturing parentheses. Any opening parenthesis that is
2342 not followed by ? behaves as if it were followed by ?: but named parentheses
2343 can still be used for capturing (and they will acquire numbers in the usual
2344 way).
2346 41. Redesigned the return codes from the match() function into yes/no/error so
2347 that errors can be passed back from deep inside the nested calls. A malloc
2348 failure while inside a recursive subpattern call now causes the
2349 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2351 42. It is now possible to set a limit on the number of times the match()
2352 function is called in a call to pcre_exec(). This facility makes it possible to
2353 limit the amount of recursion and backtracking, though not in a directly
2354 obvious way, because the match() function is used in a number of different
2355 circumstances. The count starts from zero for each position in the subject
2356 string (for non-anchored patterns). The default limit is, for compatibility, a
2357 large number, namely 10 000 000. You can change this in two ways:
2359 (a) When configuring PCRE before making, you can use --with-match-limit=n
2360 to set a default value for the compiled library.
2362 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2363 a different value is set. See 45 below.
2365 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2367 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2368 of things that can be changed at compile time. The first argument specifies
2369 what is wanted and the second points to where the information is to be placed.
2370 The current list of available information is:
2374 The output is an integer that is set to one if UTF-8 support is available;
2375 otherwise it is set to zero.
2379 The output is an integer that it set to the value of the code that is used for
2380 newline. It is either LF (10) or CR (13).
2384 The output is an integer that contains the number of bytes used for internal
2385 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2389 The output is an integer that contains the threshold above which the POSIX
2390 interface uses malloc() for output vectors. See item 31 above.
2394 The output is an unsigned integer that contains the default limit of the number
2395 of match() calls in a pcre_exec() execution. See 42 above.
2397 44. pcretest has been upgraded by the addition of the -C option. This causes it
2398 to extract all the available output from the new pcre_config() function, and to
2399 output it. The program then exits immediately.
2401 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2402 order to support additional features. One way would have been to define
2403 pcre_exec2() (for example) with extra arguments, but this would not have been
2404 extensible, and would also have required all calls to the original function to
2405 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2406 is used for passing in "extra" data from pcre_study().
2408 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2409 contains the following fields:
2411 flags a bitmap indicating which of the following fields are set
2412 study_data opaque data from pcre_study()
2413 match_limit a way of specifying a limit on match() calls for a specific
2414 call to pcre_exec()
2415 callout_data data for callouts (see 49 below)
2417 The flag bits are also defined in pcre.h, and are
2423 The pcre_study() function now returns one of these new pcre_extra blocks, with
2424 the actual study data pointed to by the study_data field, and the
2425 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2426 before. That is, this change is entirely upwards-compatible and requires no
2427 change to existing code.
2429 If you want to pass in additional data to pcre_exec(), you can either place it
2430 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2431 block.
2433 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2434 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2435 times with different match limits, until it finds the minimum value needed for
2436 pcre_exec() to complete. The value is then output. This can be instructive; for
2437 most simple matches the number is quite small, but for pathological cases it
2438 gets very large very quickly.
2440 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2441 returns the size of the data block pointed to by the study_data field in a
2442 pcre_extra block, that is, the value that was passed as the argument to
2443 pcre_malloc() when PCRE was getting memory in which to place the information
2444 created by pcre_study(). The fourth argument should point to a size_t variable.
2445 pcretest has been extended so that this information is shown after a successful
2446 pcre_study() call when information about the compiled regex is being displayed.
2448 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2449 because what follows is always an absolute path. (Later: it turns out that this
2450 is more than cosmetic for MinGW, because it doesn't like empty path
2451 components.)
2453 49. Some changes have been made to the callout feature (see 28 above):
2455 (i) A callout function now has three choices for what it returns:
2457 0 => success, carry on matching
2458 > 0 => failure at this point, but backtrack if possible
2459 < 0 => serious error, return this value from pcre_exec()
2461 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2462 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2463 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2464 use by callout functions. It will never be used by PCRE itself.
2466 (ii) The pcre_extra structure (see 45 above) has a void * field called
2467 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2468 pcre_callout_block structure has a field of the same name. The contents of
2469 the field passed in the pcre_extra structure are passed to the callout
2470 function in the corresponding field in the callout block. This makes it
2471 easier to use the same callout-containing regex from multiple threads. For
2472 testing, the pcretest program has a new data escape
2474 \C*n pass the number n (may be negative) as callout_data
2476 If the callout function in pcretest receives a non-zero value as
2477 callout_data, it returns that value.
2479 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2480 there were some redundant $(CFLAGS) in commands that are now specified as
2481 $(LINK), which already includes $(CFLAGS).
2483 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2484 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2485 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2486 one-byte characters throughout. Note that case-insensitive matching applies
2487 only to characters whose values are less than 256. PCRE doesn't support the
2488 notion of cases for higher-valued characters.
2490 (i) A character class whose characters are all within 0-255 is handled as
2491 a bit map, and the map is inverted for negative classes. Previously, a
2492 character > 255 always failed to match such a class; however it should
2493 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2495 (ii) A negated character class with a single character < 255 is coded as
2496 "not this character" (OP_NOT). This wasn't working properly when the test
2497 character was multibyte, either singly or repeated.
2499 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2500 mode, for example: \x{100}{2,3}.
2502 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2503 singly or repeated) now correctly test multibyte characters. However,
2504 PCRE doesn't recognize any characters with values greater than 255 as
2505 digits, spaces, or word characters. Such characters always match \D, \S,
2506 and \W, and never match \d, \s, or \w.
2508 (v) Classes may now contain characters and character ranges with values
2509 greater than 255. For example: [ab\x{100}-\x{400}].
2511 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2512 PCRE in UTF-8 mode.
2514 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2515 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2516 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2517 value.)
2519 53. The single man page has become too large. I have therefore split it up into
2520 a number of separate man pages. These also give rise to individual HTML pages;
2521 these are now put in a separate directory, and there is an index.html page that
2522 lists them all. Some hyperlinking between the pages has been installed.
2524 54. Added convenience functions for handling named capturing parentheses.
2526 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2527 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2528 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2529 are faulted.
2531 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2532 calling configure. These values are used when compiling the dftables.c program
2533 which is run to generate the source of the default character tables. They
2534 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2535 you will need to set these values.
2537 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2540 Version 3.9 02-Jan-02
2541 ---------------------
2543 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2545 2. If --disable-static was given, the building process failed when trying to
2546 build pcretest and pcregrep. (For some reason it was using libtool to compile
2547 them, which is not right, as they aren't part of the library.)
2550 Version 3.8 18-Dec-01
2551 ---------------------
2553 1. The experimental UTF-8 code was completely screwed up. It was packing the
2554 bytes in the wrong order. How dumb can you get?
2557 Version 3.7 29-Oct-01
2558 ---------------------
2560 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2561 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2562 this didn't happen under Solaris 8, where I normally test things.
2564 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2565 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2566 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2567 this fix an hour or so after the initial 3.7 release.)
2570 Version 3.6 23-Oct-01
2571 ---------------------
2573 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2574 offsets passed as NULL with zero offset count.
2576 2. The config.guess and config.sub files had not been updated when I moved to
2577 the latest autoconf.
2580 Version 3.5 15-Aug-01
2581 ---------------------
2583 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2584 had been forgotten.
2586 2. By using declared but undefined structures, we can avoid using "void"
2587 definitions in pcre.h while keeping the internal definitions of the structures
2588 private.
2590 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2591 user point of view, this means that both static and shared libraries are built
2592 by default, but this can be individually controlled. More of the work of
2593 handling this static/shared cases is now inside libtool instead of PCRE's make
2594 file.
2596 4. The pcretest utility is now installed along with pcregrep because it is
2597 useful for users (to test regexs) and by doing this, it automatically gets
2598 relinked by libtool. The documentation has been turned into a man page, so
2599 there are now .1, .txt, and .html versions in /doc.
2601 5. Upgrades to pcregrep:
2602 (i) Added long-form option names like gnu grep.
2603 (ii) Added --help to list all options with an explanatory phrase.
2604 (iii) Added -r, --recursive to recurse into sub-directories.
2605 (iv) Added -f, --file to read patterns from a file.
2607 6. pcre_exec() was referring to its "code" argument before testing that
2608 argument for NULL (and giving an error if it was NULL).
2610 7. Upgraded Makefile.in to allow for compiling in a different directory from
2611 the source directory.
2613 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2614 options bits, the pointer it was passed was to an int instead of to an unsigned
2615 long int. This mattered only on 64-bit systems.
2617 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2618 generated) instead of pcre.in, which it its source. Also made the same change
2619 in several of the .c files.
2621 10. A new release of gcc defines printf() as a macro, which broke pcretest
2622 because it had an ifdef in the middle of a string argument for printf(). Fixed
2623 by using separate calls to printf().
2625 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2626 script, to force use of CR or LF instead of \n in the source. On non-Unix
2627 systems, the value can be set in config.h.
2629 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2630 absolute limit. Changed the text of the error message to make this clear, and
2631 likewise updated the man page.
2633 13. The limit of 99 on the number of capturing subpatterns has been removed.
2634 The new limit is 65535, which I hope will not be a "real" limit.
2637 Version 3.4 22-Aug-00
2638 ---------------------
2640 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2642 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2645 Version 3.3 01-Aug-00
2646 ---------------------
2648 1. If an octal character was given, but the value was greater than \377, it
2649 was not getting masked to the least significant bits, as documented. This could
2650 lead to crashes in some systems.
2652 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2653 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2655 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2656 These just pass their arguments on to (pcre_free)(), but they are provided
2657 because some uses of PCRE bind it to non-C systems that can call its functions,
2658 but cannot call free() or pcre_free() directly.
2660 4. Add "make test" as a synonym for "make check". Corrected some comments in
2661 the Makefile.
2663 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2664 Makefile.
2666 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2667 command called pgrep for grepping around the active processes.
2669 7. Added the beginnings of support for UTF-8 character strings.
2671 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2672 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2673 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2674 out for the ar command.)
2677 Version 3.2 12-May-00
2678 ---------------------
2680 This is purely a bug fixing release.
2682 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2683 of ZA. This was just one example of several cases that could provoke this bug,
2684 which was introduced by change 9 of version 2.00. The code for breaking
2685 infinite loops after an iteration that matches an empty string was't working
2686 correctly.
2688 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2689 when matched against abbab (for example). After matching an empty string, it
2690 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2691 caused it to match further down the string than it should.
2693 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2694 was there because it doesn't seem to be needed, and it causes trouble on some
2695 systems, as it is not a Standard C header. It has been removed.
2697 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2698 were reported on the Macintosh. The changes were from
2700 while ((c = *(++ptr)) != 0 && c != '\n');
2701 to
2702 while ((c = *(++ptr)) != 0 && c != '\n') ;
2704 Totally extraordinary, but if that's what it takes...
2706 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2707 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2708 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2709 assumes the way PCRE uses memmove() (always moving upwards).
2711 6. PCRE is being used in one environment where strchr() is not available. There
2712 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2713 faster code anyway.
2716 Version 3.1 09-Feb-00
2717 ---------------------
2719 The only change in this release is the fixing of some bugs in Makefile.in for
2720 the "install" target:
2722 (1) It was failing to install pcreposix.h.
2724 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2727 Version 3.0 01-Feb-00
2728 ---------------------
2730 1. Add support for the /+ modifier to perltest (to output $` like it does in
2731 pcretest).
2733 2. Add support for the /g modifier to perltest.
2735 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2736 matches null strings.
2738 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2739 pattern. Perl treats empty patterns specially - it reuses the most recent
2740 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2741 effect.
2743 5. The POSIX interface was broken in that it was just handing over the POSIX
2744 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2745 required a bigger vector, with some working space on the end. This means that
2746 the POSIX wrapper now has to get and free some memory, and copy the results.
2748 6. Added some simple autoconf support, placing the test data and the
2749 documentation in separate directories, re-organizing some of the
2750 information files, and making it build pcre-config (a GNU standard). Also added
2751 libtool support for building PCRE as a shared library, which is now the
2752 default.
2754 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2755 09 are not valid octal constants. Single digits will be used for minor values
2756 less than 10.
2758 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2759 existing programs that set these in the POSIX interface can use PCRE without
2760 modification.
2762 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2763 return all that pcre_info() returns, plus additional data. The pcre_info()
2764 function is retained for compatibility, but is considered to be obsolete.
2766 10. Added experimental recursion feature (?R) to handle one common case that
2767 Perl 5.6 will be able to do with (?p{...}).
2769 11. Added support for POSIX character classes like [:alpha:], which Perl is
2770 adopting.
2773 Version 2.08 31-Aug-99
2774 ----------------------
2776 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2777 trying to match at the startoffset position, but instead was moving forward to
2778 the next newline as if a previous match had failed.
2780 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2781 and could get into a loop if a null string was matched other than at the start
2782 of the subject.
2784 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2785 be distinguished at compile time, and for completeness also added PCRE_DATE.
2787 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2788 in GnuWin32 environments.
2791 Version 2.07 29-Jul-99
2792 ----------------------
2794 1. The documentation is now supplied in plain text form and HTML as well as in
2795 the form of man page sources.
2797 2. C++ compilers don't like assigning (void *) values to other pointer types.
2798 In particular this affects malloc(). Although there is no problem in Standard
2799 C, I've put in casts to keep C++ compilers happy.
2801 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2802 should be (const char *).
2804 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2805 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2806 However, I haven't made this a standard facility. The documentation doesn't
2807 mention it, and the Makefile doesn't support it.
2809 5. The Makefile now contains an "install" target, with editable destinations at
2810 the top of the file. The pcretest program is not installed.
2812 6. pgrep -V now gives the PCRE version number and date.
2814 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2815 causing the entire string to be ignored, instead of just the last character.
2817 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2818 non-matching string, it can take a very, very long time, even for strings of
2819 quite modest length, because of the nested recursion. PCRE now does better in
2820 some of these cases. It does this by remembering the last required literal
2821 character in the pattern, and pre-searching the subject to ensure it is present
2822 before running the real match. In other words, it applies a heuristic to detect
2823 some types of certain failure quickly, and in the above example, if presented
2824 with a string that has no trailing " it gives "no match" very quickly.
2826 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2827 other alternatives are tried instead.
2830 Version 2.06 09-Jun-99
2831 ----------------------
2833 1. Change pcretest's output for amount of store used to show just the code
2834 space, because the remainder (the data block) varies in size between 32-bit and
2835 64-bit systems.
2837 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2838 start matching at. This allows lookbehinds to work when searching for multiple
2839 occurrences in a string.
2841 3. Added additional options to pcretest for testing multiple occurrences:
2843 /+ outputs the rest of the string that follows a match
2844 /g loops for multiple occurrences, using the new startoffset argument
2845 /G loops for multiple occurrences by passing an incremented pointer
2847 4. PCRE wasn't doing the "first character" optimization for patterns starting
2848 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2849 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2850 the letter 'x'. On long subject strings, this gives a significant speed-up.
2853 Version 2.05 21-Apr-99
2854 ----------------------
2856 1. Changed the type of magic_number from int to long int so that it works
2857 properly on 16-bit systems.
2859 2. Fixed a bug which caused patterns starting with .* not to work correctly
2860 when the subject string contained newline characters. PCRE was assuming
2861 anchoring for such patterns in all cases, which is not correct because .* will
2862 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2863 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2864 must be retried after every newline in the subject.
2867 Version 2.04 18-Feb-99
2868 ----------------------
2870 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2871 computation of the store needed to hold the pattern was incorrect (too large).
2872 If such patterns were nested a few deep, this could multiply and become a real
2873 problem.
2875 2. Added /M option to pcretest to show the memory requirement of a specific
2876 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2878 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2879 compiled in such a way that the backtracking after subsequent failure was
2880 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2881 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2884 Version 2.03 02-Feb-99
2885 ----------------------
2887 1. Fixed typo and small mistake in man page.
2889 2. Added 4th condition (GPL supersedes if conflict) and created separate
2890 LICENCE file containing the conditions.
2892 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2893 Perl, that is the internal \ allows the delimiter to be included in the
2894 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2895 the final delimiter, add \ to the end of the pattern (to test the error).
2897 4. Added the convenience functions for extracting substrings after a successful
2898 match. Updated pcretest to make it able to test these functions.
2901 Version 2.02 14-Jan-99
2902 ----------------------
2904 1. Initialized the working variables associated with each extraction so that
2905 their saving and restoring doesn't refer to uninitialized store.
2907 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2908 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2909 fix the problem.
2911 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2912 calls, and wasn't printing the correct value for compiling calls. Increased the
2913 default value of LOOPREPEAT, and the number of significant figures in the
2914 times.
2916 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2918 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2919 a building problem on Windows NT with a FAT file system.
2922 Version 2.01 21-Oct-98
2923 ----------------------
2925 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2926 to character tables built by pcre_maketables() in the current locale. If NULL
2927 is passed, the default tables are used.
2930 Version 2.00 24-Sep-98
2931 ----------------------
2933 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2934 it any more.
2936 2. Allow quantification of (?>) groups, and make it work correctly.
2938 3. The first character computation wasn't working for (?>) groups.
2940 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2941 end of the subject) and add 5.005's \z, which really does match only at the
2942 very end of the subject.
2944 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2946 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2947 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2948 localized options. All options to pcre_study() were also removed.
2950 7. Add other new features from 5.005:
2952 $(?<= positive lookbehind
2953 $(?<! negative lookbehind
2954 (?imsx-imsx) added the unsetting capability
2955 such a setting is global if at outer level; local otherwise
2956 (?imsx-imsx:) non-capturing groups with option setting
2957 (?(cond)re|re) conditional pattern matching
2959 A backreference to itself in a repeated group matches the previous
2960 captured string.
2962 8. General tidying up of studying (both automatic and via "study")
2963 consequential on the addition of new assertions.
2965 9. As in 5.005, unlimited repeated groups that could match an empty substring
2966 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2967 runtime if any iteration does actually match an empty substring.
2969 10. Include the RunTest script in the distribution.
2971 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2972 discrepancies, some of which were old and were also with respect to 5.004. They
2973 have now been fixed.
2976 Version 1.09 28-Apr-98
2977 ----------------------
2979 1. A negated single character class followed by a quantifier with a minimum
2980 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2981 program crashes, or just wrong answers. This did not apply to negated classes
2982 containing more than one character, or to minima other than one.
2985 Version 1.08 27-Mar-98
2986 ----------------------
2988 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2990 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2991 latter must appear before anything that relies on it in the pattern.
2994 Version 1.07 16-Feb-98
2995 ----------------------
2997 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
2998 repeat of a potentially empty string).
3001 Version 1.06 23-Jan-98
3002 ----------------------
3004 1. Added Markus Oberhumer's little patches for C++.
3006 2. Literal strings longer than 255 characters were broken.
3009 Version 1.05 23-Dec-97
3010 ----------------------
3012 1. Negated character classes containing more than one character were failing if
3013 PCRE_CASELESS was set at run time.
3016 Version 1.04 19-Dec-97
3017 ----------------------
3019 1. Corrected the man page, where some "const" qualifiers had been omitted.
3021 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3022 input syntax.
3024 3. Fixed memory leak which occurred when a regex with back references was
3025 matched with an offsets vector that wasn't big enough. The temporary memory
3026 that is used in this case wasn't being freed if the match failed.
3028 4. Tidied pcretest to ensure it frees memory that it gets.
3030 5. Temporary memory was being obtained in the case where the passed offsets
3031 vector was exactly big enough.
3033 6. Corrected definition of offsetof() from change 5 below.
3035 7. I had screwed up change 6 below and broken the rules for the use of
3036 setjmp(). Now fixed.
3039 Version 1.03 18-Dec-97
3040 ----------------------
3042 1. A erroneous regex with a missing opening parenthesis was correctly
3043 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3044 on some systems.
3046 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3047 it was reported that one broken compiler failed on the former because "code" is
3048 also an independent variable.
3050 3. The erroneous regex a[]b caused an array overrun reference.
3052 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3053 fail on data ending with that character. (It was going on too far, and checking
3054 the next character, typically a binary zero.) This was specific to the
3055 optimized code for single-character negative classes.
3057 5. Added a contributed patch from the TIN world which does the following:
3059 + Add an undef for memmove, in case the the system defines a macro for it.
3061 + Add a definition of offsetof(), in case there isn't one. (I don't know
3062 the reason behind this - offsetof() is part of the ANSI standard - but
3063 it does no harm).
3065 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3066 most of the places where whitespace preceded '#'. I have given up and
3067 allowed the remaining 2 cases to be at the margin.
3069 + Rename some variables in pcre to eliminate shadowing. This seems very
3070 pedantic, but does no harm, of course.
3072 6. Moved the call to setjmp() into its own function, to get rid of warnings
3073 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3075 7. Constructs such as \d{8,} were compiling into the equivalent of
3076 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3077 outcome, but in this particular case used more store than had been allocated,
3078 which caused the bug to be discovered because it threw up an internal error.
3080 8. The debugging code in both pcre and pcretest for outputting the compiled
3081 form of a regex was going wrong in the case of back references followed by
3082 curly-bracketed repeats.
3085 Version 1.02 12-Dec-97
3086 ----------------------
3088 1. Typos in pcre.3 and comments in the source fixed.
3090 2. Applied a contributed patch to get rid of places where it used to remove
3091 'const' from variables, and fixed some signed/unsigned and uninitialized
3092 variable warnings.
3094 3. Added the "runtest" target to Makefile.
3096 4. Set default compiler flag to -O2 rather than just -O.
3099 Version 1.01 19-Nov-97
3100 ----------------------
3102 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3103 like /([ab]*)*/, that is, for classes with more than one character in them.
3105 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3106 as /((?>a*))*/ (a PCRE_EXTRA facility).
3109 Version 1.00 18-Nov-97
3110 ----------------------
3112 1. Added compile-time macros to support systems such as SunOS4 which don't have
3113 memmove() or strerror() but have other things that can be used instead.
3115 2. Arranged that "make clean" removes the executables.
3118 Version 0.99 27-Oct-97
3119 ----------------------
3121 1. Fixed bug in code for optimizing classes with only one character. It was
3122 initializing a 32-byte map regardless, which could cause it to run off the end
3123 of the memory it had got.
3125 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3128 Version 0.98 22-Oct-97
3129 ----------------------
3131 1. Fixed bug in code for handling temporary memory usage when there are more
3132 back references than supplied space in the ovector. This could cause segfaults.
3135 Version 0.97 21-Oct-97
3136 ----------------------
3138 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3140 2. Optimized negated single characters not to use a bit map.
3142 3. Brought error texts together as macro definitions; clarified some of them;
3143 fixed one that was wrong - it said "range out of order" when it meant "invalid
3144 escape sequence".
3146 4. Changed some char * arguments to const char *.
3148 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3150 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3151 pcretest.
3154 Version 0.96 16-Oct-97
3155 ----------------------
3157 1. Added a simple "pgrep" utility to the distribution.
3159 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3160 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3161 where "ddd" means "one or more decimal digits".
3163 3. Fixed serious bug. If a pattern had a back reference, but the call to
3164 pcre_exec() didn't supply a large enough ovector to record the related
3165 identifying subpattern, the match always failed. PCRE now remembers the number
3166 of the largest back reference, and gets some temporary memory in which to save
3167 the offsets during matching if necessary, in order to ensure that
3168 backreferences always work.
3170 4. Increased the compatibility with Perl in a number of ways:
3172 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3173 to request this handling. The option can be set at compile or exec time.
3175 (b) $ matches before a terminating newline by default; an option
3176 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3177 mode). The option can be set at compile or exec time.
3179 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3180 the same as Perl's. If the decimal number it represents is less than 10
3181 or there aren't that many previous left capturing parentheses, an octal
3182 escape is read. Inside a character class, it's always an octal escape,
3183 even if it is a single digit.
3185 (d) An escaped but undefined alphabetic character is taken as a literal,
3186 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3187 escapes.
3189 (e) {0} is now permitted. (The previous item is removed from the compiled
3190 pattern).
3192 5. Changed all the names of code files so that the basic parts are no longer
3193 than 10 characters, and abolished the teeny "globals.c" file.
3195 6. Changed the handling of character classes; they are now done with a 32-byte
3196 bit map always.
3198 7. Added the -d and /D options to pcretest to make it possible to look at the
3199 internals of compilation without having to recompile pcre.
3202 Version 0.95 23-Sep-97
3203 ----------------------
3205 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3206 \x20 at the start of a run of normal characters. These were being treated as
3207 real characters, instead of the source characters being re-checked.
3210 Version 0.94 18-Sep-97
3211 ----------------------
3213 1. The functions are now thread-safe, with the caveat that the global variables
3214 containing pointers to malloc() and free() or alternative functions are the
3215 same for all threads.
3217 2. Get pcre_study() to generate a bitmap of initial characters for non-
3218 anchored patterns when this is possible, and use it if passed to pcre_exec().
3221 Version 0.93 15-Sep-97
3222 ----------------------
3224 1. /(b)|(:+)/ was computing an incorrect first character.
3226 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3227 but not actually doing anything yet.
3229 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3230 as Perl does (e.g. [-az] or [az-]).
3232 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3233 all possible positions.
3235 5. Split up into different modules to avoid including unneeded functions in a
3236 compiled binary. However, compile and exec are still in one module. The "study"
3237 function is split off.
3239 6. The character tables are now in a separate module whose source is generated
3240 by an auxiliary program - but can then be edited by hand if required. There are
3241 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3242 toupper() in the code.
3244 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3245 make them global. Abolish the function for setting them, as the caller can now
3246 set them directly.
3249 Version 0.92 11-Sep-97
3250 ----------------------
3252 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3253 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3255 2. Caseless matching was not working in character classes if the characters in
3256 the pattern were in upper case.
3258 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3260 4. Make PCRE_ANCHORED public and accept as a compile option.
3262 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3263 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3264 pass them.
3266 6. Give an error if bad option bits passed at compile or run time.
3268 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3269 pcretest to cause it to pass that flag.
3271 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3272 options, and the first character, if set.
3274 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3277 Version 0.91 10-Sep-97
3278 ----------------------
3280 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3281 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3283 2. PCRE was looping on encountering an indefinitely repeated back reference to
3284 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3285 Perl does - treats the match as successful.
3287 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5