ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 342 - (show annotations)
Sun Apr 20 17:10:13 2008 UTC (13 years, 6 months ago) by ph10
File size: 155483 byte(s)
Slight performance improvement by using the new OP_ALLANY opcode for cases of 
the metacharacter "." when DOTALL is set. Also, some tidies consequent upon its 
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).
90 16. The implementation of 13c above involved the invention of a new opcode,
91 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
92 cannot be changed at match time, I realized I could make a small
93 improvement to matching performance by compiling OP_ALLANY instead of
94 OP_ANY for "." when DOTALL was set, and then removing the runtime tests
95 on the OP_ANY path.
98 Version 7.6 28-Jan-08
99 ---------------------
101 1. A character class containing a very large number of characters with
102 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
103 overflow.
105 2. Patch to cut out the "long long" test in pcrecpp_unittest when
106 HAVE_LONG_LONG is not defined.
108 3. Applied Christian Ehrlicher's patch to update the CMake build files to
109 bring them up to date and include new features. This patch includes:
111 - Fixed PH's badly added libz and libbz2 support.
112 - Fixed a problem with static linking.
113 - Added pcredemo. [But later removed - see 7 below.]
114 - Fixed dftables problem and added an option.
115 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
117 - Added readline support for pcretest.
118 - Added an listing of the option settings after cmake has run.
120 4. A user submitted a patch to Makefile that makes it easy to create
121 "pcre.dll" under mingw when using Configure/Make. I added stuff to
122 Makefile.am that cause it to include this special target, without
123 affecting anything else. Note that the same mingw target plus all
124 the other distribution libraries and programs are now supported
125 when configuring with CMake (see 6 below) instead of with
126 Configure/Make.
128 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
129 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
130 exported in the Windows port". It has not yet been confirmed that the patch
131 solves the problem, but it does no harm.
133 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
134 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
135 with CMake, and also correct the comment about stack recursion.
137 7. Remove the automatic building of pcredemo from the ./configure system and
138 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
139 of a program that users should build themselves after PCRE is installed, so
140 building it automatically is not really right. What is more, it gave
141 trouble in some build environments.
143 8. Further tidies to CMakeLists.txt from Sheri and Christian.
146 Version 7.5 10-Jan-08
147 ---------------------
149 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
150 values in parens when parsing an RE using the C++ wrapper."
152 2. Negative specials like \S did not work in character classes in UTF-8 mode.
153 Characters greater than 255 were excluded from the class instead of being
154 included.
156 3. The same bug as (2) above applied to negated POSIX classes such as
157 [:^space:].
159 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
160 defined or documented. It seems to have been a typo for PCRE_STATIC, so
161 I have changed it.
163 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
164 first named subpattern) and a construct such as (?&a) would reference the
165 first named subpattern whose name started with "a" (in other words, the
166 length check was missing). Both these problems are fixed. "Subpattern name
167 expected" is now given for (?&) (a zero-length name), and this patch also
168 makes it give the same error for \k'' (previously it complained that that
169 was a reference to a non-existent subpattern).
171 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
172 this is right because (?- can be followed by option settings as well as by
173 digits. I have, however, made the messages clearer.
175 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
176 than the number used in the conditional) now cause a compile-time error.
177 This is actually not compatible with Perl, which accepts such patterns, but
178 treats the conditional as always being FALSE (as PCRE used to), but it
179 seems to me that giving a diagnostic is better.
181 8. Change "alphameric" to the more common word "alphanumeric" in comments
182 and messages.
184 9. Fix two occurrences of "backslash" in comments that should have been
185 "backspace".
187 10. Remove two redundant lines of code that can never be obeyed (their function
188 was moved elsewhere).
190 11. The program that makes PCRE's Unicode character property table had a bug
191 which caused it to generate incorrect table entries for sequences of
192 characters that have the same character type, but are in different scripts.
193 It amalgamated them into a single range, with the script of the first of
194 them. In other words, some characters were in the wrong script. There were
195 thirteen such cases, affecting characters in the following ranges:
197 U+002b0 - U+002c1
198 U+0060c - U+0060d
199 U+0061e - U+00612
200 U+0064b - U+0065e
201 U+0074d - U+0076d
202 U+01800 - U+01805
203 U+01d00 - U+01d77
204 U+01d9b - U+01dbf
205 U+0200b - U+0200f
206 U+030fc - U+030fe
207 U+03260 - U+0327f
208 U+0fb46 - U+0fbb1
209 U+10450 - U+1049d
211 12. The -o option (show only the matching part of a line) for pcregrep was not
212 compatible with GNU grep in that, if there was more than one match in a
213 line, it showed only the first of them. It now behaves in the same way as
214 GNU grep.
216 13. If the -o and -v options were combined for pcregrep, it printed a blank
217 line for every non-matching line. GNU grep prints nothing, and pcregrep now
218 does the same. The return code can be used to tell if there were any
219 non-matching lines.
221 14. Added --file-offsets and --line-offsets to pcregrep.
223 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
224 infinitely looping recursion. The bug was that positive lookaheads were not
225 being skipped when checking for a possible empty match (negative lookaheads
226 and both kinds of lookbehind were skipped).
228 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
229 inclusion of <windows.h> to before rather than after the definition of
230 INVALID_FILE_ATTRIBUTES (patch from David Byron).
232 17. Specifying a possessive quantifier with a specific limit for a Unicode
233 character property caused pcre_compile() to compile bad code, which led at
234 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
235 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
236 caused the error; without that there was no problem.
238 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
240 19. Added --enable-pcretest-libreadline.
242 20. In pcrecpp.cc, the variable 'count' was incremented twice in
243 RE::GlobalReplace(). As a result, the number of replacements returned was
244 double what it should be. I removed one of the increments, but Craig sent a
245 later patch that removed the other one (the right fix) and added unit tests
246 that check the return values (which was not done before).
248 21. Several CMake things:
250 (1) Arranged that, when cmake is used on Unix, the libraries end up with
251 the names libpcre and libpcreposix, not just pcre and pcreposix.
253 (2) The above change means that pcretest and pcregrep are now correctly
254 linked with the newly-built libraries, not previously installed ones.
258 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
259 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
260 UTF-8 newline character). The key issue is that the pattern starts .*;
261 this means that the match must be either at the beginning, or after a
262 newline. The bug was in the code for advancing after a failed match and
263 checking that the new position followed a newline. It was not taking
264 account of UTF-8 characters correctly.
266 23. PCRE was behaving differently from Perl in the way it recognized POSIX
267 character classes. PCRE was not treating the sequence [:...:] as a
268 character class unless the ... were all letters. Perl, however, seems to
269 allow any characters between [: and :], though of course it rejects as
270 unknown any "names" that contain non-letters, because all the known class
271 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
272 for example, whereas PCRE did not - it did not recognize a POSIX character
273 class. This seemed a bit dangerous, so the code has been changed to be
274 closer to Perl. The behaviour is not identical to Perl, because PCRE will
275 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
276 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
277 Perl does, and where it didn't before.
279 24. Rewrite so as to remove the single use of %n from pcregrep because in some
280 Windows environments %n is disabled by default.
283 Version 7.4 21-Sep-07
284 ---------------------
286 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
287 means that a class such as [\s] counted as "explicit reference to CR or
288 LF". That isn't really right - the whole point of the change was to try to
289 help when there was an actual mention of one of the two characters. So now
290 the change happens only if \r or \n (or a literal CR or LF) character is
291 encountered.
293 2. The 32-bit options word was also used for 6 internal flags, but the numbers
294 of both had grown to the point where there were only 3 bits left.
295 Fortunately, there was spare space in the data structure, and so I have
296 moved the internal flags into a new 16-bit field to free up more option
297 bits.
299 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
300 but did not set the internal JCHANGED flag - either of these is enough to
301 control the way the "get" function works - but the PCRE_INFO_JCHANGED
302 facility is supposed to tell if (?J) was ever used, so now (?J) at the
303 start sets both bits.
305 4. Added options (at build time, compile time, exec time) to change \R from
306 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
308 5. doc/pcresyntax.html was missing from the distribution.
310 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
311 compatibility, even though it is no longer used.
313 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
314 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
315 windows.h file is present (where different names are used). [This was
316 reversed later after testing - see 16 below.]
318 8. Changed all #include <config.h> to #include "config.h". There were also
319 some further <pcre.h> cases that I changed to "pcre.h".
321 9. When pcregrep was used with the --colour option, it missed the line ending
322 sequence off the lines that it output.
324 10. It was pointed out to me that arrays of string pointers cause lots of
325 relocations when a shared library is dynamically loaded. A technique of
326 using a single long string with a table of offsets can drastically reduce
327 these. I have refactored PCRE in four places to do this. The result is
328 dramatic:
330 Originally: 290
331 After changing UCP table: 187
332 After changing error message table: 43
333 After changing table of "verbs" 36
334 After changing table of Posix names 22
336 Thanks to the folks working on Gregex for glib for this insight.
338 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
339 unicode-properties was also set.
341 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
343 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
344 checked only for CRLF.
346 14. Added casts to pcretest.c to avoid compiler warnings.
348 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
350 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
351 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
352 entirely. This removes changes made in 7 above.
354 17. The CMake files have been updated, and there is now more information about
355 building with CMake in the NON-UNIX-USE document.
358 Version 7.3 28-Aug-07
359 ---------------------
361 1. In the rejigging of the build system that eventually resulted in 7.1, the
362 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
363 brackets there is not right, since it causes compilers to look for an
364 installed pcre.h, not the version that is in the source that is being
365 compiled (which of course may be different). I have changed it back to:
367 #include "pcre.h"
369 I have a vague recollection that the change was concerned with compiling in
370 different directories, but in the new build system, that is taken care of
371 by the VPATH setting the Makefile.
373 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
374 when the subject happened to end in the byte 0x85 (e.g. if the last
375 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
376 characters but of course it shouldn't be taken as a newline when it is part
377 of another character. The bug was that, for an unlimited repeat of . in
378 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
379 characters when looking for a newline.
381 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
383 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
384 in debug output.
386 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
387 long printing in the pcrecpp unittest when running under MinGW.
389 6. ESC_K was left out of the EBCDIC table.
391 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
392 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
393 limit also applies to "virtual nesting" when a pattern is recursive, and in
394 this case 1000 isn't so big. I have been able to remove this limit at the
395 expense of backing off one optimization in certain circumstances. Normally,
396 when pcre_exec() would call its internal match() function recursively and
397 immediately return the result unconditionally, it uses a "tail recursion"
398 feature to save stack. However, when a subpattern that can match an empty
399 string has an unlimited repetition quantifier, it no longer makes this
400 optimization. That gives it a stack frame in which to save the data for
401 checking that an empty string has been matched. Previously this was taken
402 from the 1000-entry workspace that had been reserved. So now there is no
403 explicit limit, but more stack is used.
405 8. Applied Daniel's patches to solve problems with the import/export magic
406 syntax that is required for Windows, and which was going wrong for the
407 pcreposix and pcrecpp parts of the library. These were overlooked when this
408 problem was solved for the main library.
410 9. There were some crude static tests to avoid integer overflow when computing
411 the size of patterns that contain repeated groups with explicit upper
412 limits. As the maximum quantifier is 65535, the maximum group length was
413 set at 30,000 so that the product of these two numbers did not overflow a
414 32-bit integer. However, it turns out that people want to use groups that
415 are longer than 30,000 bytes (though not repeat them that many times).
416 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
417 made it possible to implement the integer overflow checks in a much more
418 dynamic way, which I have now done. The artificial limitation on group
419 length has been removed - we now have only the limit on the total length of
420 the compiled pattern, which depends on the LINK_SIZE setting.
422 10. Fixed a bug in the documentation for get/copy named substring when
423 duplicate names are permitted. If none of the named substrings are set, the
424 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
425 empty string.
427 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
428 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
429 because the ] is interpreted as the first data character and the
430 terminating ] is not found. PCRE has been made compatible with Perl in this
431 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
432 cause memory overwriting.
434 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
435 string has been matched (to stop an infinite loop). It was not recognizing
436 a conditional subpattern that could match an empty string if that
437 subpattern was within another subpattern. For example, it looped when
438 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
439 condition was not nested. This bug has been fixed.
441 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
442 past the start of the subject in the presence of bytes with the top bit
443 set, for example "\x8aBCD".
445 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
446 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
448 14. Optimized (?!) to (*FAIL).
450 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
451 This restricts code points to be within the range 0 to 0x10FFFF, excluding
452 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
453 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
454 does: it's just the validity check that is more restrictive.
456 16. Inserted checks for integer overflows during escape sequence (backslash)
457 processing, and also fixed erroneous offset values for syntax errors during
458 backslash processing.
460 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
461 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
463 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
464 caused an overrun.
466 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
467 something other than just ASCII characters) inside a group that had an
468 unlimited repeat caused a loop at compile time (while checking to see
469 whether the group could match an empty string).
471 20. Debugging a pattern containing \p or \P could cause a crash. For example,
472 [\P{Any}] did so. (Error in the code for printing property names.)
474 21. An orphan \E inside a character class could cause a crash.
476 22. A repeated capturing bracket such as (A)? could cause a wild memory
477 reference during compilation.
479 23. There are several functions in pcre_compile() that scan along a compiled
480 expression for various reasons (e.g. to see if it's fixed length for look
481 behind). There were bugs in these functions when a repeated \p or \P was
482 present in the pattern. These operators have additional parameters compared
483 with \d, etc, and these were not being taken into account when moving along
484 the compiled data. Specifically:
486 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
487 length.
489 (b) An item such as \pL+ within a repeated group could cause crashes or
490 loops.
492 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
493 "reference to non-existent subpattern" error.
495 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
497 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
498 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
500 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
502 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
503 character were causing crashes (broken optimization).
505 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
506 \p or \P) caused a compile-time loop.
508 28. More problems have arisen in unanchored patterns when CRLF is a valid line
509 break. For example, the unstudied pattern [\r\n]A does not match the string
510 "\r\nA" because change 7.0/46 below moves the current point on by two
511 characters after failing to match at the start. However, the pattern \nA
512 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
513 the same is true. There doesn't seem any very clean way out of this, but
514 what I have chosen to do makes the common cases work: PCRE now takes note
515 of whether there can be an explicit match for \r or \n anywhere in the
516 pattern, and if so, 7.0/46 no longer applies. As part of this change,
517 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
518 pattern has explicit CR or LF references.
520 29. Added (*CR) etc for changing newline setting at start of pattern.
523 Version 7.2 19-Jun-07
524 ---------------------
526 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
527 which is apparently normally available under Windows.
529 2. Re-jig the pcregrep tests with different newline settings in an attempt
530 to make them independent of the local environment's newline setting.
532 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
534 4. Some of the "internals" tests were previously cut out when the link size
535 was not 2, because the output contained actual offsets. The recent new
536 "Z" feature of pcretest means that these can be cut out, making the tests
537 usable with all link sizes.
539 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
540 stack recursion. This gives a massive performance boost under BSD, but just
541 a small improvement under Linux. However, it saves one field in the frame
542 in all cases.
544 6. Added more features from the forthcoming Perl 5.10:
546 (a) (?-n) (where n is a string of digits) is a relative subroutine or
547 recursion call. It refers to the nth most recently opened parentheses.
549 (b) (?+n) is also a relative subroutine call; it refers to the nth next
550 to be opened parentheses.
552 (c) Conditions that refer to capturing parentheses can be specified
553 relatively, for example, (?(-2)... or (?(+3)...
555 (d) \K resets the start of the current match so that everything before
556 is not part of it.
558 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
560 (f) \g{name} is another synonym - part of Perl 5.10's unification of
561 reference syntax.
563 (g) (?| introduces a group in which the numbering of parentheses in each
564 alternative starts with the same number.
566 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
568 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
571 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
572 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
573 for detecting groups that can match an empty string.
575 9. A pattern with a very large number of alternatives (more than several
576 hundred) was running out of internal workspace during the pre-compile
577 phase, where pcre_compile() figures out how much memory will be needed. A
578 bit of new cunning has reduced the workspace needed for groups with
579 alternatives. The 1000-alternative test pattern now uses 12 bytes of
580 workspace instead of running out of the 4096 that are available.
582 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
584 11. Applied patch from Google to remove an optimization that didn't quite work.
585 The report of the bug said:
587 pcrecpp::RE("a*").FullMatch("aaa") matches, while
588 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
589 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
591 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
592 it matched the wrong number of bytes.
595 Version 7.1 24-Apr-07
596 ---------------------
598 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
599 that is more "standard", making use of automake and other Autotools. There
600 is some re-arrangement of the files and adjustment of comments consequent
601 on this.
603 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
604 for recursive directory scanning broke on some systems because the files
605 are not scanned in any specific order and on different systems the order
606 was different. A call to "sort" has been inserted into RunGrepTest for the
607 approprate test as a short-term fix. In the longer term there may be an
608 alternative.
610 3. I had an email from Eric Raymond about problems translating some of PCRE's
611 man pages to HTML (despite the fact that I distribute HTML pages, some
612 people do their own conversions for various reasons). The problems
613 concerned the use of low-level troff macros .br and .in. I have therefore
614 removed all such uses from the man pages (some were redundant, some could
615 be replaced by .nf/.fi pairs). The 132html script that I use to generate
616 HTML has been updated to handle .nf/.fi and to complain if it encounters
617 .br or .in.
619 4. Updated comments in configure.ac that get placed in config.h.in and also
620 arranged for config.h to be included in the distribution, with the name
621 config.h.generic, for the benefit of those who have to compile without
622 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
624 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
625 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
626 makevp.bat for the current PCRE, using the additional files
627 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
629 6. A Windows user reported a minor discrepancy with test 2, which turned out
630 to be caused by a trailing space on an input line that had got lost in his
631 copy. The trailing space was an accident, so I've just removed it.
633 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
634 that is needed.
636 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
637 as "const" (a) because they are and (b) because it helps the PHP
638 maintainers who have recently made a script to detect big data structures
639 in the php code that should be moved to the .rodata section. I remembered
640 to update Builducptable as well, so it won't revert if ucptable.h is ever
641 re-created.
643 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
644 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
645 order to be able to cut out the UTF-8 tables in the latter when UTF-8
646 support is not required. This saves 1.5-2K of code, which is important in
647 some applications.
649 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
650 so as not to refer to the tables, even though these functions will never be
651 called when UTF-8 support is disabled. Otherwise there are problems with a
652 shared library.
654 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
656 (a) It was defining its arguments as char * instead of void *.
658 (b) It was assuming that all moves were upwards in memory; this was true
659 a long time ago when I wrote it, but is no longer the case.
661 The emulated memove() is provided for those environments that have neither
662 memmove() nor bcopy(). I didn't think anyone used it these days, but that
663 is clearly not the case, as these two bugs were recently reported.
665 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
666 and Detrail to create the HTML documentation, the .txt form of the man
667 pages, and it removes trailing spaces from listed files. It also creates
668 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
669 case, it wraps all the #defines with #ifndefs. This script should be run
670 before "make dist".
672 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
673 with Unicode property support.
675 (a) For a maximizing quantifier, if the two different cases of the
676 character were of different lengths in their UTF-8 codings (there are
677 some cases like this - I found 11), and the matching function had to
678 back up over a mixture of the two cases, it incorrectly assumed they
679 were both the same length.
681 (b) When PCRE was configured to use the heap rather than the stack for
682 recursion during matching, it was not correctly preserving the data for
683 the other case of a UTF-8 character when checking ahead for a match
684 while processing a minimizing repeat. If the check also involved
685 matching a wide character, but failed, corruption could cause an
686 erroneous result when trying to check for a repeat of the original
687 character.
689 13. Some tidying changes to the testing mechanism:
691 (a) The RunTest script now detects the internal link size and whether there
692 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
693 values substituted by "configure". (The RunGrepTest script already did
694 this for UTF-8.) The configure.ac script no longer substitutes the
695 relevant variables.
697 (b) The debugging options /B and /D in pcretest show the compiled bytecode
698 with length and offset values. This means that the output is different
699 for different internal link sizes. Test 2 is skipped for link sizes
700 other than 2 because of this, bypassing the problem. Unfortunately,
701 there was also a test in test 3 (the locale tests) that used /B and
702 failed for link sizes other than 2. Rather than cut the whole test out,
703 I have added a new /Z option to pcretest that replaces the length and
704 offset values with spaces. This is now used to make test 3 independent
705 of link size. (Test 2 will be tidied up later.)
707 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
708 segmentation fault instead of returning the appropriate error message.
710 15. In multiline mode when the newline sequence was set to "any", the pattern
711 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
712 This doesn't seem right; it now treats the CRLF combination as the line
713 ending, and so does not match in that case. It's only a pattern such as ^$
714 that would hit this one: something like ^ABC$ would have failed after \r
715 and then tried again after \r\n.
717 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
718 in an attempt to make files that differ only in their line terminators
719 compare equal. This works on Linux.
721 17. Under certain error circumstances pcregrep might try to free random memory
722 as it exited. This is now fixed, thanks to valgrind.
724 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
725 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
726 was because its rules for how to advance for /g after matching an empty
727 string at the end of a line did not allow for this case. They now check for
728 it specially.
730 20. pcretest is supposed to handle patterns and data of any length, by
731 extending its buffers when necessary. It was getting this wrong when the
732 buffer for a data line had to be extended.
734 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
735 CRLF as a newline sequence.
737 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
738 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
739 I have nevertheless tidied it up.
741 23. Added some casts to kill warnings from HP-UX ia64 compiler.
743 24. Added a man page for pcre-config.
746 Version 7.0 19-Dec-06
747 ---------------------
749 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
750 moving to gcc 4.1.1.
752 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
753 sys/time.h, which is documented as needed for this function. It doesn't
754 seem to matter on Linux, but it showed up on some releases of OS X.
756 3. It seems that there are systems where bytes whose values are greater than
757 127 match isprint() in the "C" locale. The "C" locale should be the
758 default when a C program starts up. In most systems, only ASCII printing
759 characters match isprint(). This difference caused the output from pcretest
760 to vary, making some of the tests fail. I have changed pcretest so that:
762 (a) When it is outputting text in the compiled version of a pattern, bytes
763 other than 32-126 are always shown as hex escapes.
765 (b) When it is outputting text that is a matched part of a subject string,
766 it does the same, unless a different locale has been set for the match
767 (using the /L modifier). In this case, it uses isprint() to decide.
769 4. Fixed a major bug that caused incorrect computation of the amount of memory
770 required for a compiled pattern when options that changed within the
771 pattern affected the logic of the preliminary scan that determines the
772 length. The relevant options are -x, and -i in UTF-8 mode. The result was
773 that the computed length was too small. The symptoms of this bug were
774 either the PCRE error "internal error: code overflow" from pcre_compile(),
775 or a glibc crash with a message such as "pcretest: free(): invalid next
776 size (fast)". Examples of patterns that provoked this bug (shown in
777 pcretest format) are:
779 /(?-x: )/x
780 /(?x)(?-x: \s*#\s*)/
781 /((?i)[\x{c0}])/8
782 /(?i:[\x{c0}])/8
784 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
785 is now done differently.
787 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
788 wrapper classes; (b) implement a new function in the C++ scanner that is
789 more efficient than the old way of doing things because it avoids levels of
790 recursion in the regex matching; (c) add a paragraph to the documentation
791 for the FullMatch() function.
793 6. The escape sequence \n was being treated as whatever was defined as
794 "newline". Not only was this contrary to the documentation, which states
795 that \n is character 10 (hex 0A), but it also went horribly wrong when
796 "newline" was defined as CRLF. This has been fixed.
798 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
799 was being set to -1 for the "end of line" case (supposedly a value that no
800 character can have). Though this value is never used (the check for end of
801 line is "zero bytes in current character"), it caused compiler complaints.
802 I've changed it to 0xffffffff.
804 8. In pcre_version.c, the version string was being built by a sequence of
805 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
806 string (as it is for production releases) called a macro with an empty
807 argument. The C standard says the result of this is undefined. The gcc
808 compiler treats it as an empty string (which was what was wanted) but it is
809 reported that Visual C gives an error. The source has been hacked around to
810 avoid this problem.
812 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
813 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
814 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
815 of them did).
817 10. Originally, pcretest opened its input and output without "b"; then I was
818 told that "b" was needed in some environments, so it was added for release
819 5.0 to both the input and output. (It makes no difference on Unix-like
820 systems.) Later I was told that it is wrong for the input on Windows. I've
821 now abstracted the modes into two macros, to make it easier to fiddle with
822 them, and removed "b" from the input mode under Windows.
824 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
826 12. Added -help and --help to pcretest as an official way of being reminded
827 of the options.
829 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
830 and pcrecpp.cc because they annoy compilers at high warning levels.
832 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
834 15. Fixed an occurrence of == in configure.ac that should have been = (shell
835 scripts are not C programs :-) and which was not noticed because it works
836 on Linux.
838 16. pcretest is supposed to handle any length of pattern and data line (as one
839 line or as a continued sequence of lines) by extending its input buffer if
840 necessary. This feature was broken for very long pattern lines, leading to
841 a string of junk being passed to pcre_compile() if the pattern was longer
842 than about 50K.
844 17. I have done a major re-factoring of the way pcre_compile() computes the
845 amount of memory needed for a compiled pattern. Previously, there was code
846 that made a preliminary scan of the pattern in order to do this. That was
847 OK when PCRE was new, but as the facilities have expanded, it has become
848 harder and harder to keep it in step with the real compile phase, and there
849 have been a number of bugs (see for example, 4 above). I have now found a
850 cunning way of running the real compile function in a "fake" mode that
851 enables it to compute how much memory it would need, while actually only
852 ever using a few hundred bytes of working memory and without too many
853 tests of the mode. This should make future maintenance and development
854 easier. A side effect of this work is that the limit of 200 on the nesting
855 depth of parentheses has been removed (though this was never a serious
856 limitation, I suspect). However, there is a downside: pcre_compile() now
857 runs more slowly than before (30% or more, depending on the pattern). I
858 hope this isn't a big issue. There is no effect on runtime performance.
860 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
861 newline (only possible for the last line of a file) and it was a
862 pattern that set a locale (followed by /Lsomething), pcretest crashed.
864 19. Added additional timing features to pcretest. (1) The -tm option now times
865 matching only, not compiling. (2) Both -t and -tm can be followed, as a
866 separate command line item, by a number that specifies the number of
867 repeats to use when timing. The default is 50000; this gives better
868 precision, but takes uncomfortably long for very large patterns.
870 20. Extended pcre_study() to be more clever in cases where a branch of a
871 subpattern has no definite first character. For example, (a*|b*)[cd] would
872 previously give no result from pcre_study(). Now it recognizes that the
873 first character must be a, b, c, or d.
875 21. There was an incorrect error "recursive call could loop indefinitely" if
876 a subpattern (or the entire pattern) that was being tested for matching an
877 empty string contained only one non-empty item after a nested subpattern.
878 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
879 incorrectly, because the \d was being skipped in the check.
881 22. The pcretest program now has a new pattern option /B and a command line
882 option -b, which is equivalent to adding /B to every pattern. This causes
883 it to show the compiled bytecode, without the additional information that
884 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
885 is the same as /B/I).
887 23. A new optimization is now able automatically to treat some sequences such
888 as a*b as a*+b. More specifically, if something simple (such as a character
889 or a simple class like \d) has an unlimited quantifier, and is followed by
890 something that cannot possibly match the quantified thing, the quantifier
891 is automatically "possessified".
893 24. A recursive reference to a subpattern whose number was greater than 39
894 went wrong under certain circumstances in UTF-8 mode. This bug could also
895 have affected the operation of pcre_study().
897 25. Realized that a little bit of performance could be had by replacing
898 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
900 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
902 27. Possessive quantifiers such as a++ were previously implemented by turning
903 them into atomic groups such as ($>a+). Now they have their own opcodes,
904 which improves performance. This includes the automatically created ones
905 from 23 above.
907 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
908 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
909 the first matched character to be a colon. This applied both to named and
910 numbered groups.
912 29. The ucpinternal.h header file was missing its idempotency #ifdef.
914 30. I was sent a "project" file called libpcre.a.dev which I understand makes
915 building PCRE on Windows easier, so I have included it in the distribution.
917 31. There is now a check in pcretest against a ridiculously large number being
918 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
919 loop, the loop is abandoned.
921 32. Forward references to subpatterns in conditions such as (?(2)...) where
922 subpattern 2 is defined later cause pcre_compile() to search forwards in
923 the pattern for the relevant set of parentheses. This search went wrong
924 when there were unescaped parentheses in a character class, parentheses
925 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
927 33. "Subroutine" calls and backreferences were previously restricted to
928 referencing subpatterns earlier in the regex. This restriction has now
929 been removed.
931 34. Added a number of extra features that are going to be in Perl 5.10. On the
932 whole, these are just syntactic alternatives for features that PCRE had
933 previously implemented using the Python syntax or my own invention. The
934 other formats are all retained for compatibility.
936 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
937 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
938 also .NET compatible.
940 (b) A recursion or subroutine call to a named group can now be defined as
941 (?&name) as well as (?P>name).
943 (c) A backreference to a named group can now be defined as \k<name> or
944 \k'name' as well as (?P=name). The new forms, as well as being in Perl
945 5.10, are also .NET compatible.
947 (d) A conditional reference to a named group can now use the syntax
948 (?(<name>) or (?('name') as well as (?(name).
950 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
951 groups (named and numbered) that are never evaluated inline, but can be
952 called as "subroutines" from elsewhere. In effect, the DEFINE condition
953 is always false. There may be only one alternative in such a group.
955 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
956 as the simple (?(R). The condition is true only if the most recent
957 recursion is that of the given number or name. It does not search out
958 through the entire recursion stack.
960 (g) The escape \gN or \g{N} has been added, where N is a positive or
961 negative number, specifying an absolute or relative reference.
963 35. Tidied to get rid of some further signed/unsigned compiler warnings and
964 some "unreachable code" warnings.
966 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
967 things, this adds five new scripts.
969 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
970 There were also incompatibilities regarding the handling of \Q..\E inside
971 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
972 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
974 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
975 matches an empty string, and forcibly breaks the loop. There were bugs in
976 this code in non-simple cases. For a pattern such as ^(a()*)* matched
977 against aaaa the result was just "a" rather than "aaaa", for example. Two
978 separate and independent bugs (that affected different cases) have been
979 fixed.
981 39. Refactored the code to abolish the use of different opcodes for small
982 capturing bracket numbers. This is a tidy that I avoided doing when I
983 removed the limit on the number of capturing brackets for 3.5 back in 2001.
984 The new approach is not only tidier, it makes it possible to reduce the
985 memory needed to fix the previous bug (38).
987 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
988 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
989 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
990 mode.
992 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
993 report.
995 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
996 copy construction and assignment in the C++ wrapper.
998 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
999 couple of bugs that could have given wrong results in the "--newline=crlf"
1000 case.
1002 44. Added a number of casts and did some reorganization of signed/unsigned int
1003 variables following suggestions from Dair Grant. Also renamed the variable
1004 "this" as "item" because it is a C++ keyword.
1006 45. Arranged for dftables to add
1008 #include "pcre_internal.h"
1010 to pcre_chartables.c because without it, gcc 4.x may remove the array
1011 definition from the final binary if PCRE is built into a static library and
1012 dead code stripping is activated.
1014 46. For an unanchored pattern, if a match attempt fails at the start of a
1015 newline sequence, and the newline setting is CRLF or ANY, and the next two
1016 characters are CRLF, advance by two characters instead of one.
1019 Version 6.7 04-Jul-06
1020 ---------------------
1022 1. In order to handle tests when input lines are enormously long, pcretest has
1023 been re-factored so that it automatically extends its buffers when
1024 necessary. The code is crude, but this _is_ just a test program. The
1025 default size has been increased from 32K to 50K.
1027 2. The code in pcre_study() was using the value of the re argument before
1028 testing it for NULL. (Of course, in any sensible call of the function, it
1029 won't be NULL.)
1031 3. The memmove() emulation function in pcre_internal.h, which is used on
1032 systems that lack both memmove() and bcopy() - that is, hardly ever -
1033 was missing a "static" storage class specifier.
1035 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1036 containing an extended class (one that cannot be represented by a bitmap
1037 because it contains high-valued characters or Unicode property items, e.g.
1038 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1039 pattern, but PCRE should not loop if you do not (it no longer does).
1040 [Detail: two cases were found: (a) a repeated subpattern containing an
1041 extended class; (b) a recursive reference to a subpattern that followed a
1042 previous extended class. It wasn't skipping over the extended class
1043 correctly when UTF-8 mode was not set.]
1045 5. A negated single-character class was not being recognized as fixed-length
1046 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1047 compile error "lookbehind assertion is not fixed length".
1049 6. The RunPerlTest auxiliary script was showing an unexpected difference
1050 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1051 write a Perl script that can interpret lines of an input file either as
1052 byte characters or as UTF-8, which is what "perltest" was being required to
1053 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1054 can't do is switch easily at run time between having the "use utf8;" pragma
1055 or not. In the end, I fudged it by using the RunPerlTest script to insert
1056 "use utf8;" explicitly for the UTF-8 tests.
1058 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1059 the end of the subject string, contrary to the documentation and to what
1060 Perl does. This was true of both matching functions. Now it matches only at
1061 the start of the subject and immediately after *internal* newlines.
1063 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1064 a pointer to an int instead of a pointer to an unsigned long int. This
1065 caused problems on 64-bit systems.
1067 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1068 instance of the 'standard' template library not being so standard".
1070 10. There was no check on the number of named subpatterns nor the maximum
1071 length of a subpattern name. The product of these values is used to compute
1072 the size of the memory block for a compiled pattern. By supplying a very
1073 long subpattern name and a large number of named subpatterns, the size
1074 computation could be caused to overflow. This is now prevented by limiting
1075 the length of names to 32 characters, and the number of named subpatterns
1076 to 10,000.
1078 11. Subpatterns that are repeated with specific counts have to be replicated in
1079 the compiled pattern. The size of memory for this was computed from the
1080 length of the subpattern and the repeat count. The latter is limited to
1081 65535, but there was no limit on the former, meaning that integer overflow
1082 could in principle occur. The compiled length of a repeated subpattern is
1083 now limited to 30,000 bytes in order to prevent this.
1085 12. Added the optional facility to have named substrings with the same name.
1087 13. Added the ability to use a named substring as a condition, using the
1088 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1089 are numbers (not recommended). Forward references are permitted.
1091 14. Added forward references in named backreferences (if you see what I mean).
1093 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1094 pattern could run off the end of the subject. For example, the pattern
1095 "(?s)(.{1,5})"8 did this with the subject "ab".
1097 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1098 PCRE_CASELESS was set when matching characters that were quantified with ?
1099 or *.
1101 17. A character class other than a single negated character that had a minimum
1102 but no maximum quantifier - for example [ab]{6,} - was not handled
1103 correctly by pce_dfa_exec(). It would match only one character.
1105 18. A valid (though odd) pattern that looked like a POSIX character
1106 class but used an invalid character after [ (for example [[,abc,]]) caused
1107 pcre_compile() to give the error "Failed: internal error: code overflow" or
1108 in some cases to crash with a glibc free() error. This could even happen if
1109 the pattern terminated after [[ but there just happened to be a sequence of
1110 letters, a binary zero, and a closing ] in the memory that followed.
1112 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1113 over the years. Originally (before any Unicode support), just the bottom 8
1114 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1115 output from "man perlunicode" includes this:
1117 The regular expression compiler produces polymorphic opcodes. That
1118 is, the pattern adapts to the data and automatically switches to
1119 the Unicode character scheme when presented with Unicode data--or
1120 instead uses a traditional byte scheme when presented with byte
1121 data.
1123 Sadly, a wide octal escape does not cause a switch, and in a string with
1124 no other multibyte characters, these octal escapes are treated as before.
1125 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1126 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1127 Unicode string.
1129 I have not perpetrated such confusion in PCRE. Up till now, it took just
1130 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1131 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1132 translate to the appropriate multibyte character.
1134 29. Applied some refactoring to reduce the number of warnings from Microsoft
1135 and Borland compilers. This has included removing the fudge introduced
1136 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1137 a warning about an unused variable.
1139 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1140 characters since release 4.0, because Perl (from release 5.004) does not.
1141 [Or at least, is documented not to: some releases seem to be in conflict
1142 with the documentation.] However, when a pattern was studied with
1143 pcre_study() and all its branches started with \s, PCRE still included VT
1144 as a possible starting character. Of course, this did no harm; it just
1145 caused an unnecessary match attempt.
1147 22. Removed a now-redundant internal flag bit that recorded the fact that case
1148 dependency changed within the pattern. This was once needed for "required
1149 byte" processing, but is no longer used. This recovers a now-scarce options
1150 bit. Also moved the least significant internal flag bit to the most-
1151 significant bit of the word, which was not previously used (hangover from
1152 the days when it was an int rather than a uint) to free up another bit for
1153 the future.
1155 23. Added support for CRLF line endings as well as CR and LF. As well as the
1156 default being selectable at build time, it can now be changed at runtime
1157 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1158 specify that it is scanning data with non-default line endings.
1160 24. Changed the definition of CXXLINK to make it agree with the definition of
1161 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1163 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1164 recursions. This makes a big different to stack usage for some patterns.
1166 26. If a subpattern containing a named recursion or subroutine reference such
1167 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1168 the space required for the compiled pattern went wrong and gave too small a
1169 value. Depending on the environment, this could lead to "Failed: internal
1170 error: code overflow at offset 49" or "glibc detected double free or
1171 corruption" errors.
1173 27. Applied patches from Google (a) to support the new newline modes and (b) to
1174 advance over multibyte UTF-8 characters in GlobalReplace.
1176 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1177 difference for some implementation of PCRE in some Windows version.
1179 29. Added some extra testing facilities to pcretest:
1181 \q<number> in a data line sets the "match limit" value
1182 \Q<number> in a data line sets the "match recursion limt" value
1183 -S <number> sets the stack size, where <number> is in megabytes
1185 The -S option isn't available for Windows.
1188 Version 6.6 06-Feb-06
1189 ---------------------
1191 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1192 in pcreposix.h. I have copied the definition from pcre.h.
1194 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1195 because pcre.h is no longer a built file.
1197 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1198 not normally included in the compiled code.
1201 Version 6.5 01-Feb-06
1202 ---------------------
1204 1. When using the partial match feature with pcre_dfa_exec(), it was not
1205 anchoring the second and subsequent partial matches at the new starting
1206 point. This could lead to incorrect results. For example, with the pattern
1207 /1234/, partially matching against "123" and then "a4" gave a match.
1209 2. Changes to pcregrep:
1211 (a) All non-match returns from pcre_exec() were being treated as failures
1212 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1213 error message is output. Some extra information is given for the
1215 probably the only errors that are likely to be caused by users (by
1216 specifying a regex that has nested indefinite repeats, for instance).
1217 If there are more than 20 of these errors, pcregrep is abandoned.
1219 (b) A binary zero was treated as data while matching, but terminated the
1220 output line if it was written out. This has been fixed: binary zeroes
1221 are now no different to any other data bytes.
1223 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1224 used to set a locale for matching. The --locale=xxxx long option has
1225 been added (no short equivalent) to specify a locale explicitly on the
1226 pcregrep command, overriding the environment variables.
1228 (d) When -B was used with -n, some line numbers in the output were one less
1229 than they should have been.
1231 (e) Added the -o (--only-matching) option.
1233 (f) If -A or -C was used with -c (count only), some lines of context were
1234 accidentally printed for the final match.
1236 (g) Added the -H (--with-filename) option.
1238 (h) The combination of options -rh failed to suppress file names for files
1239 that were found from directory arguments.
1241 (i) Added the -D (--devices) and -d (--directories) options.
1243 (j) Added the -F (--fixed-strings) option.
1245 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1247 (l) Added the --colo(u)r option.
1249 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1250 is not present by default.
1252 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1253 items such as (?R) or (?1), when the recursion could match a number of
1254 alternatives. If it matched one of the alternatives, but subsequently,
1255 outside the recursion, there was a failure, the code tried to back up into
1256 the recursion. However, because of the way PCRE is implemented, this is not
1257 possible, and the result was an incorrect result from the match.
1259 In order to prevent this happening, the specification of recursion has
1260 been changed so that all such subpatterns are automatically treated as
1261 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1263 4. I had overlooked the fact that, in some locales, there are characters for
1264 which isalpha() is true but neither isupper() nor islower() are true. In
1265 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1266 and ordfeminine) are like this. This affected the treatment of \w and \W
1267 when they appeared in character classes, but not when they appeared outside
1268 a character class. The bit map for "word" characters is now created
1269 separately from the results of isalnum() instead of just taking it from the
1270 upper, lower, and digit maps. (Plus the underscore character, of course.)
1272 5. The above bug also affected the handling of POSIX character classes such as
1273 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1274 permanent tables. Instead, the bit maps for such a class were previously
1275 created as the appropriate unions of the upper, lower, and digit bitmaps.
1276 Now they are created by subtraction from the [[:word:]] class, which has
1277 its own bitmap.
1279 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1280 It is created by subtracting the vertical space characters (\x09, \x0a,
1281 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1282 subtraction was done in the overall bitmap for a character class, meaning
1283 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1284 be recognized. This bug has been fixed.
1286 7. Patches from the folks at Google:
1288 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1289 real life, but is still worth protecting against".
1291 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1292 regular expressions".
1294 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1295 have it.
1297 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1298 "configure" and the latter not, in order to fix a problem somebody had
1299 with compiling the Arg class on HP-UX.
1301 (e) Improve the error-handling of the C++ wrapper a little bit.
1303 (f) New tests for checking recursion limiting.
1305 8. The pcre_memmove() function, which is used only if the environment does not
1306 have a standard memmove() function (and is therefore rarely compiled),
1307 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1308 returning a result (though PCRE never actually uses the result).
1310 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1311 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1312 returned instead of calling malloc() with an overflowing number that would
1313 most likely cause subsequent chaos.
1315 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1317 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1318 with this option is matched, the nmatch and pmatch options of regexec() are
1319 ignored.
1321 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1322 provided in case anyone wants to the the POSIX interface with UTF-8
1323 strings.
1325 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1326 C++ linking (needed for some HP-UX environments).
1328 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1329 (unused parameter) and in the pcre_printint() function (omitted "default"
1330 switch label when the default is to do nothing).
1332 15. Added some code to make it possible, when PCRE is compiled as a C++
1333 library, to replace subject pointers for pcre_exec() with a smart pointer
1334 class, thus making it possible to process discontinuous strings.
1336 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1337 much the same function. They were added by different people who were trying
1338 to make PCRE easy to compile on non-Unix systems. It has been suggested
1339 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1340 for compiling on Windows systems. I have therefore replaced it with
1341 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1342 defaults to "extern" for C or "extern C" for C++, which works fine on
1343 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1344 SCOPE with something explicit in config.h. In addition:
1346 (a) pcreposix.h still had just "extern" instead of either of these macros;
1347 I have replaced it with PCRE_DATA_SCOPE.
1349 (b) Functions such as _pcre_xclass(), which are internal to the library,
1350 but external in the C sense, all had PCRE_EXPORT in their definitions.
1351 This is apparently wrong for the Windows case, so I have removed it.
1352 (It makes no difference on Unix-like systems.)
1354 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1355 of recursive calls to match(). This is different to MATCH_LIMIT because
1356 that limits the total number of calls to match(), not all of which increase
1357 the depth of recursion. Limiting the recursion depth limits the amount of
1358 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1359 when PCRE is compiled, and changed at run time. A patch from Google adds
1360 this functionality to the C++ interface.
1362 18. Changes to the handling of Unicode character properties:
1364 (a) Updated the table to Unicode 4.1.0.
1366 (b) Recognize characters that are not in the table as "Cn" (undefined).
1368 (c) I revised the way the table is implemented to a much improved format
1369 which includes recognition of ranges. It now supports the ranges that
1370 are defined in UnicodeData.txt, and it also amalgamates other
1371 characters into ranges. This has reduced the number of entries in the
1372 table from around 16,000 to around 3,000, thus reducing its size
1373 considerably. I realized I did not need to use a tree structure after
1374 all - a binary chop search is just as efficient. Having reduced the
1375 number of entries, I extended their size from 6 bytes to 8 bytes to
1376 allow for more data.
1378 (d) Added support for Unicode script names via properties such as \p{Han}.
1380 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1381 matching that character.
1383 20. When matching a repeated Unicode property with a minimum greater than zero,
1384 (for example \pL{2,}), PCRE could look past the end of the subject if it
1385 reached it while seeking the minimum number of characters. This could
1386 happen only if some of the characters were more than one byte long, because
1387 there is a check for at least the minimum number of bytes.
1389 21. Refactored the implementation of \p and \P so as to be more general, to
1390 allow for more different types of property in future. This has changed the
1391 compiled form incompatibly. Anybody with saved compiled patterns that use
1392 \p or \P will have to recompile them.
1394 22. Added "Any" and "L&" to the supported property types.
1396 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1397 but give a compile time error if the value is greater than 0xff.
1399 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1400 accidentally not being installed or uninstalled.
1402 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1403 made were to insert the current release number. This seemed silly, because
1404 it made things harder for people building PCRE on systems that don't run
1405 "configure". I have turned pcre.h into a distributed file, no longer built
1406 by "configure", with the version identification directly included. There is
1407 no longer a pcre.h.in file.
1409 However, this change necessitated a change to the pcre-config script as
1410 well. It is built from pcre-config.in, and one of the substitutions was the
1411 release number. I have updated configure.ac so that ./configure now finds
1412 the release number by grepping pcre.h.
1414 26. Added the ability to run the tests under valgrind.
1417 Version 6.4 05-Sep-05
1418 ---------------------
1420 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1421 "--" to be printed when multiple files were scanned, even when none of the
1422 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1423 consider it to be a bug, and have restored the previous behaviour.
1425 2. A couple of code tidies to get rid of compiler warnings.
1427 3. The pcretest program used to cheat by referring to symbols in the library
1428 whose names begin with _pcre_. These are internal symbols that are not
1429 really supposed to be visible externally, and in some environments it is
1430 possible to suppress them. The cheating is now confined to including
1431 certain files from the library's source, which is a bit cleaner.
1433 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1434 file's purpose clearer.
1436 5. Reorganized pcre_ucp_findchar().
1439 Version 6.3 15-Aug-05
1440 ---------------------
1442 1. The file libpcre.pc.in did not have general read permission in the tarball.
1444 2. There were some problems when building without C++ support:
1446 (a) If C++ support was not built, "make install" and "make test" still
1447 tried to test it.
1449 (b) There were problems when the value of CXX was explicitly set. Some
1450 changes have been made to try to fix these, and ...
1452 (c) --disable-cpp can now be used to explicitly disable C++ support.
1454 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1455 backslash in a target when C++ was disabled. This confuses some
1456 versions of "make", apparently. Using an intermediate variable solves
1457 this. (Same for CPP_LOBJ.)
1459 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1460 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1461 necessary on certain architectures.
1463 4. Added a setting of -export-symbols-regex to the link command to remove
1464 those symbols that are exported in the C sense, but actually are local
1465 within the library, and not documented. Their names all begin with
1466 "_pcre_". This is not a perfect job, because (a) we have to except some
1467 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1468 available (and never for static libraries). I have made a note to try to
1469 find a way round (a) in the future.
1472 Version 6.2 01-Aug-05
1473 ---------------------
1475 1. There was no test for integer overflow of quantifier values. A construction
1476 such as {1111111111111111} would give undefined results. What is worse, if
1477 a minimum quantifier for a parenthesized subpattern overflowed and became
1478 negative, the calculation of the memory size went wrong. This could have
1479 led to memory overwriting.
1481 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1483 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1484 operating environments where this matters.
1486 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1487 PCRE options from within the C++ wrapper.
1489 5. Named capturing subpatterns were not being correctly counted when a pattern
1490 was compiled. This caused two problems: (a) If there were more than 100
1491 such subpatterns, the calculation of the memory needed for the whole
1492 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1493 back references of the form \12, where the number was greater than 9, were
1494 not recognized as back references, even though there were sufficient
1495 previous subpatterns.
1497 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1498 versions of gcc, e.g. 2.95.4.
1501 Version 6.1 21-Jun-05
1502 ---------------------
1504 1. There was one reference to the variable "posix" in pcretest.c that was not
1505 surrounded by "#if !defined NOPOSIX".
1507 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1508 the cross-check on the old pcre_info() function, for the benefit of the
1509 cut-down version of PCRE that is currently imported into Exim.
1511 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1512 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1513 patterns that start (?i) but I don't think that matters. The use of (?i) is
1514 just an example; this all applies to the other options as well.
1516 4. Since libtool seems to echo the compile commands it is issuing, the output
1517 from "make" can be reduced a bit by putting "@" in front of each libtool
1518 compile command.
1520 5. Patch from the folks at Google for configure.in to be a bit more thorough
1521 in checking for a suitable C++ installation before trying to compile the
1522 C++ stuff. This should fix a reported problem when a compiler was present,
1523 but no suitable headers.
1525 6. The man pages all had just "PCRE" as their title. I have changed them to
1526 be the relevant file name. I have also arranged that these names are
1527 retained in the file doc/pcre.txt, which is a concatenation in text format
1528 of all the man pages except the little individual ones for each function.
1530 7. The NON-UNIX-USE file had not been updated for the different set of source
1531 files that come with release 6. I also added a few comments about the C++
1532 wrapper.
1535 Version 6.0 07-Jun-05
1536 ---------------------
1538 1. Some minor internal re-organization to help with my DFA experiments.
1540 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1541 didn't matter for the library itself when fully configured, but did matter
1542 when compiling without UCP support, or within Exim, where the ucp files are
1543 not imported.
1545 3. Refactoring of the library code to split up the various functions into
1546 different source modules. The addition of the new DFA matching code (see
1547 below) to a single monolithic source would have made it really too
1548 unwieldy, quite apart from causing all the code to be include in a
1549 statically linked application, when only some functions are used. This is
1550 relevant even without the DFA addition now that patterns can be compiled in
1551 one application and matched in another.
1553 The downside of splitting up is that there have to be some external
1554 functions and data tables that are used internally in different modules of
1555 the library but which are not part of the API. These have all had their
1556 names changed to start with "_pcre_" so that they are unlikely to clash
1557 with other external names.
1559 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1560 a different (DFA) algorithm. Although it is slower than the original
1561 function, it does have some advantages for certain types of matching
1562 problem.
1564 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1565 including restarting after a partial match.
1567 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1568 defined when compiling for Windows was sent to me. I have put it into the
1569 code, though I have no means of testing or verifying it.
1571 7. Added the pcre_refcount() auxiliary function.
1573 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1574 match before or at the first newline in the subject string. In pcretest,
1575 the /f option on a pattern can be used to set this.
1577 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1578 would behave wrongly. This has been present in PCRE since release 4.0.
1580 10. A number of changes to the pcregrep command:
1582 (a) Refactored how -x works; insert ^(...)$ instead of setting
1583 PCRE_ANCHORED and checking the length, in preparation for adding
1584 something similar for -w.
1586 (b) Added the -w (match as a word) option.
1588 (c) Refactored the way lines are read and buffered so as to have more
1589 than one at a time available.
1591 (d) Implemented a pcregrep test script.
1593 (e) Added the -M (multiline match) option. This allows patterns to match
1594 over several lines of the subject. The buffering ensures that at least
1595 8K, or the rest of the document (whichever is the shorter) is available
1596 for matching (and similarly the previous 8K for lookbehind assertions).
1598 (f) Changed the --help output so that it now says
1600 -w, --word-regex(p)
1602 instead of two lines, one with "regex" and the other with "regexp"
1603 because that confused at least one person since the short forms are the
1604 same. (This required a bit of code, as the output is generated
1605 automatically from a table. It wasn't just a text change.)
1607 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1608 option but starts with a hyphen. Could be a pattern or a path name
1609 starting with a hyphen, for instance.
1611 (h) "-" can be given as a file name to represent stdin.
1613 (i) When file names are being printed, "(standard input)" is used for
1614 the standard input, for compatibility with GNU grep. Previously
1615 "<stdin>" was used.
1617 (j) The option --label=xxx can be used to supply a name to be used for
1618 stdin when file names are being printed. There is no short form.
1620 (k) Re-factored the options decoding logic because we are going to add
1621 two more options that take data. Such options can now be given in four
1622 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1624 (l) Added the -A, -B, and -C options for requesting that lines of context
1625 around matches be printed.
1627 (m) Added the -L option to print the names of files that do not contain
1628 any matching lines, that is, the complement of -l.
1630 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1631 continue to scan other files.
1633 (o) The -s option was incorrectly implemented. For compatibility with other
1634 greps, it now suppresses the error message for a non-existent or non-
1635 accessible file (but not the return code). There is a new option called
1636 -q that suppresses the output of matching lines, which was what -s was
1637 previously doing.
1639 (p) Added --include and --exclude options to specify files for inclusion
1640 and exclusion when recursing.
1642 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1643 Hopefully, it now does.
1645 12. Missing cast in pcre_study().
1647 13. Added an "uninstall" target to the makefile.
1649 14. Replaced "extern" in the function prototypes in Makefile.in with
1650 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1651 world, but is set differently for Windows.
1653 15. Added a second compiling function called pcre_compile2(). The only
1654 difference is that it has an extra argument, which is a pointer to an
1655 integer error code. When there is a compile-time failure, this is set
1656 non-zero, in addition to the error test pointer being set to point to an
1657 error message. The new argument may be NULL if no error number is required
1658 (but then you may as well call pcre_compile(), which is now just a
1659 wrapper). This facility is provided because some applications need a
1660 numeric error indication, but it has also enabled me to tidy up the way
1661 compile-time errors are handled in the POSIX wrapper.
1663 16. Added VPATH=.libs to the makefile; this should help when building with one
1664 prefix path and installing with another. (Or so I'm told by someone who
1665 knows more about this stuff than I do.)
1667 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1668 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1669 match everything, including newlines. This is not POSIX-compatible, but
1670 somebody wanted the feature. From pcretest it can be activated by using
1671 both the P and the s flags.
1673 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1675 19. libpcre.pc was being incorrectly installed as executable.
1677 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1678 it now also looks for '\r' so that it will work unmodified on Windows.
1680 21. Added Google's contributed C++ wrapper to the distribution.
1682 22. Added some untidy missing memory free() calls in pcretest, to keep
1683 Electric Fence happy when testing.
1687 Version 5.0 13-Sep-04
1688 ---------------------
1690 1. Internal change: literal characters are no longer packed up into items
1691 containing multiple characters in a single byte-string. Each character
1692 is now matched using a separate opcode. However, there may be more than one
1693 byte in the character in UTF-8 mode.
1695 2. The pcre_callout_block structure has two new fields: pattern_position and
1696 next_item_length. These contain the offset in the pattern to the next match
1697 item, and its length, respectively.
1699 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1700 insertion of callouts before each pattern item. Added the /C option to
1701 pcretest to make use of this.
1703 4. On the advice of a Windows user, the lines
1705 #if defined(_WIN32) || defined(WIN32)
1706 _setmode( _fileno( stdout ), 0x8000 );
1707 #endif /* defined(_WIN32) || defined(WIN32) */
1709 have been added to the source of pcretest. This apparently does useful
1710 magic in relation to line terminators.
1712 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1713 for the benefit of those environments where the "b" makes a difference.
1715 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1716 to know about it. I have put a hack into configure.in that adds in code
1717 to set GCC=yes if CC=icc. This seems to end up at a point in the
1718 generated configure script that is early enough to affect the setting of
1719 compiler options, which is what is needed, but I have no means of testing
1720 whether it really works. (The user who reported this had patched the
1721 generated configure script, which of course I cannot do.)
1723 LATER: After change 22 below (new libtool files), the configure script
1724 seems to know about icc (and also ecc). Therefore, I have commented out
1725 this hack in configure.in.
1727 7. Added support for pkg-config (2 patches were sent in).
1729 8. Negated POSIX character classes that used a combination of internal tables
1730 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1731 [[:^ascii]]. Typically, they would match almost any characters. The other
1732 POSIX classes were not broken in this way.
1734 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1735 to find the match, as PCRE was deluded into thinking that the match had to
1736 start at the start point or following a newline. The same bug applied to
1737 patterns with negative forward assertions or any backward assertions
1738 preceding ".*" at the start, unless the pattern required a fixed first
1739 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1741 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1742 starting at the last subject character, bytes beyond the end of the subject
1743 string were read.
1745 11. Renamed the variable "class" as "classbits" to make life easier for C++
1746 users. (Previously there was a macro definition, but it apparently wasn't
1747 enough.)
1749 12. Added the new field "tables" to the extra data so that tables can be passed
1750 in at exec time, or the internal tables can be re-selected. This allows
1751 a compiled regex to be saved and re-used at a later time by a different
1752 program that might have everything at different addresses.
1754 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1755 -R library as well as a -L library.
1757 14. The debugging options of pcretest (-d on the command line or D on a
1758 pattern) showed incorrect output for anything following an extended class
1759 that contained multibyte characters and which was followed by a quantifier.
1761 15. Added optional support for general category Unicode character properties
1762 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1763 support. It adds about 90K to the size of the library. The meanings of the
1764 inbuilt class escapes such as \d and \s have NOT been changed.
1766 16. Updated pcredemo.c to include calls to free() to release the memory for the
1767 compiled pattern.
1769 17. The generated file chartables.c was being created in the source directory
1770 instead of in the building directory. This caused the build to fail if the
1771 source directory was different from the building directory, and was
1772 read-only.
1774 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1775 file. No doubt somebody will tell me if they don't make sense... Also added
1776 Dan Mooney's comments about building on OpenVMS.
1778 19. Added support for partial matching via the PCRE_PARTIAL option for
1779 pcre_exec() and the \P data escape in pcretest.
1781 20. Extended pcretest with 3 new pattern features:
1783 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1784 write the compiled pattern to the file whose name is "rest-of-line".
1785 This is a straight binary dump of the data, with the saved pointer to
1786 the character tables forced to be NULL. The study data, if any, is
1787 written too. After writing, pcretest reads a new pattern.
1789 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1790 compiled pattern from the given file. There must not be any
1791 occurrences of "<" in the file name (pretty unlikely); if there are,
1792 pcretest will instead treat the initial "<" as a pattern delimiter.
1793 After reading in the pattern, pcretest goes on to read data lines as
1794 usual.
1796 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1797 and 16-bit fields in a compiled pattern, to simulate a pattern that
1798 was compiled on a host of opposite endianness.
1800 21. The pcre-exec() function can now cope with patterns that were compiled on
1801 hosts of opposite endianness, with this restriction:
1803 As for any compiled expression that is saved and used later, the tables
1804 pointer field cannot be preserved; the extra_data field in the arguments
1805 to pcre_exec() should be used to pass in a tables address if a value
1806 other than the default internal tables were used at compile time.
1808 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1809 now diagnosed as an error. Previously, most of the time, a negative number
1810 would have been treated as zero, but if in addition "ovector" was passed as
1811 NULL, a crash could occur.
1813 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1814 new versions from the libtool 1.5 distribution (the last one is a copy of
1815 a file called libtool.m4). This seems to have fixed the need to patch
1816 "configure" to support Darwin 1.3 (which I used to do). However, I still
1817 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1818 workstation).
1820 24. Changed the PCRE licence to be the more standard "BSD" licence.
1823 Version 4.5 01-Dec-03
1824 ---------------------
1826 1. There has been some re-arrangement of the code for the match() function so
1827 that it can be compiled in a version that does not call itself recursively.
1828 Instead, it keeps those local variables that need separate instances for
1829 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1830 needs to "recurse". Keeping track of where control must go is done by means
1831 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1832 hide most of the details from the main code, and operates only if
1833 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1834 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1835 operating.
1837 To make it easier for callers to provide specially tailored get/free
1838 functions for this usage, two new functions, pcre_stack_malloc, and
1839 pcre_stack_free, are used. They are always called in strict stacking order,
1840 and the size of block requested is always the same.
1842 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1843 PCRE has been compiled to use the stack or the heap for recursion. The
1844 -C option of pcretest uses this to show which version is compiled.
1846 A new data escape \S, is added to pcretest; it causes the amounts of store
1847 obtained and freed by both kinds of malloc/free at match time to be added
1848 to the output.
1850 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1851 what's available on my current Linux desktop machine.
1853 3. When matching a UTF-8 string, the test for a valid string at the start has
1854 been extended. If start_offset is not zero, PCRE now checks that it points
1855 to a byte that is the start of a UTF-8 character. If not, it returns
1856 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1857 this is necessary because there may be backward assertions in the pattern.
1858 When matching the same subject several times, it may save resources to use
1859 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1861 4. The code for checking the validity of UTF-8 strings has been tightened so
1862 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1863 containing "overlong sequences".
1865 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1866 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1867 should have been "&", but it just so happened that all the cases this let
1868 through by mistake were picked up later in the function.
1870 6. I had used a variable called "isblank" - this is a C99 function, causing
1871 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1873 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1874 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1875 so the version is shown; (c) stop "make test" echoing "./RunTest".
1877 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1879 9. The prototype for memmove() for systems that don't have it was using
1880 size_t, but the inclusion of the header that defines size_t was later. I've
1881 moved the #includes for the C headers earlier to avoid this.
1883 10. Added some adjustments to the code to make it easier to compiler on certain
1884 special systems:
1886 (a) Some "const" qualifiers were missing.
1887 (b) Added the macro EXPORT before all exported functions; by default this
1888 is defined to be empty.
1889 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1890 that it reads its output file name as an argument instead of writing
1891 to the standard output and assuming this can be redirected.
1893 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1894 class containing characters with values greater than 255, PCRE compilation
1895 went into a loop.
1897 12. A recursive reference to a subpattern that was within another subpattern
1898 that had a minimum quantifier of zero caused PCRE to crash. For example,
1899 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1900 recursion. If the recursively-called subpattern itself had a zero repeat,
1901 that was OK.
1903 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1904 buffer into which it was copied (for escape processing) was still set at
1905 1024, so long lines caused crashes.
1907 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1908 "internal error: code overflow...". This applied to any character class
1909 that was followed by a possessive quantifier.
1911 15. Modified the Makefile to add libpcre.la as a prerequisite for
1912 libpcreposix.la because I was told this is needed for a parallel build to
1913 work.
1915 16. If a pattern that contained .* following optional items at the start was
1916 studied, the wrong optimizing data was generated, leading to matching
1917 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1918 matching string must start with a or b or c. The correct conclusion for
1919 this pattern is that a match can start with any character.
1922 Version 4.4 13-Aug-03
1923 ---------------------
1925 1. In UTF-8 mode, a character class containing characters with values between
1926 127 and 255 was not handled correctly if the compiled pattern was studied.
1927 In fixing this, I have also improved the studying algorithm for such
1928 classes (slightly).
1930 2. Three internal functions had redundant arguments passed to them. Removal
1931 might give a very teeny performance improvement.
1933 3. Documentation bug: the value of the capture_top field in a callout is *one
1934 more than* the number of the hightest numbered captured substring.
1936 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1937 in incorrectly linking with a previously installed version. They now link
1938 explicitly with libpcre.la.
1940 5. configure.in no longer needs to recognize Cygwin specially.
1942 6. A problem in pcre.in for Windows platforms is fixed.
1944 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1945 pcretest, it used to include the size of the study block as part of its
1946 output. Unfortunately, the structure contains a field that has a different
1947 size on different hardware architectures. This meant that the tests that
1948 showed this size failed. As the block is currently always of a fixed size,
1949 this information isn't actually particularly useful in pcretest output, so
1950 I have just removed it.
1952 8. Three pre-processor statements accidentally did not start in column 1.
1953 Sadly, there are *still* compilers around that complain, even though
1954 standard C has not required this for well over a decade. Sigh.
1956 9. In pcretest, the code for checking callouts passed small integers in the
1957 callout_data field, which is a void * field. However, some picky compilers
1958 complained about the casts involved for this on 64-bit systems. Now
1959 pcretest passes the address of the small integer instead, which should get
1960 rid of the warnings.
1962 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1963 both compile and run time, and gives an error if an invalid UTF-8 sequence
1964 is found. There is a option for disabling this check in cases where the
1965 string is known to be correct and/or the maximum performance is wanted.
1967 11. In response to a bug report, I changed one line in Makefile.in from
1969 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1970 to
1971 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1973 to look similar to other lines, but I have no way of telling whether this
1974 is the right thing to do, as I do not use Windows. No doubt I'll get told
1975 if it's wrong...
1978 Version 4.3 21-May-03
1979 ---------------------
1981 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1982 Makefile.
1984 2. Some refactoring to improve the quality of the code:
1986 (i) The utf8_table... variables are now declared "const".
1988 (ii) The code for \cx, which used the "case flipping" table to upper case
1989 lower case letters, now just substracts 32. This is ASCII-specific,
1990 but the whole concept of \cx is ASCII-specific, so it seems
1991 reasonable.
1993 (iii) PCRE was using its character types table to recognize decimal and
1994 hexadecimal digits in the pattern. This is silly, because it handles
1995 only 0-9, a-f, and A-F, but the character types table is locale-
1996 specific, which means strange things might happen. A private
1997 table is now used for this - though it costs 256 bytes, a table is
1998 much faster than multiple explicit tests. Of course, the standard
1999 character types table is still used for matching digits in subject
2000 strings against \d.
2002 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2003 ending in _t are). So I've renamed it as ESC_tee.
2005 3. The first argument for regexec() in the POSIX wrapper should have been
2006 defined as "const".
2008 4. Changed pcretest to use malloc() for its buffers so that they can be
2009 Electric Fenced for debugging.
2011 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2012 to read one or more bytes before the start of the subject string. Often this
2013 had no effect on PCRE's behaviour, but in some circumstances it could
2014 provoke a segmentation fault.
2016 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2017 to try to read one or more bytes before the start of the subject string.
2019 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2020 UTF-8 support could misbehave in various ways if the subject string
2021 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2022 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2023 back over UTF-8 characters.)
2026 Version 4.2 14-Apr-03
2027 ---------------------
2029 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2031 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2032 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2033 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2034 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2035 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2037 Note: automatic setting of the BUILD variables is not yet working
2038 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2039 compile-time but not at link-time
2040 [LINK]: use for linking executables only
2041 make different versions for Windows and non-Windows
2042 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2043 libraries
2044 [LINK_FOR_BUILD]: new variable
2045 [OBJEXT]: use throughout
2046 [EXEEXT]: use throughout
2047 <winshared>: new target
2048 <wininstall>: new target
2049 <dftables.o>: use native compiler
2050 <dftables>: use native linker
2051 <install>: handle Windows platform correctly
2052 <clean>: ditto
2053 <check>: ditto
2054 copy DLL to top builddir before testing
2056 As part of these changes, -no-undefined was removed again. This was reported
2057 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2058 in any case.
2060 3. Some tidies to get rid of compiler warnings:
2062 . In the match_data structure, match_limit was an unsigned long int, whereas
2063 match_call_count was an int. I've made them both unsigned long ints.
2065 . In pcretest the fact that a const uschar * doesn't automatically cast to
2066 a void * provoked a warning.
2068 . Turning on some more compiler warnings threw up some "shadow" variables
2069 and a few more missing casts.
2071 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2072 option, a class that contained a single character with a value between 128
2073 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2075 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2076 option, a class that contained several characters, but with at least one
2077 whose value was between 128 and 255 caused PCRE to crash.
2080 Version 4.1 12-Mar-03
2081 ---------------------
2083 1. Compiling with gcc -pedantic found a couple of places where casts were
2084 needed, and a string in dftables.c that was longer than standard compilers are
2085 required to support.
2087 2. Compiling with Sun's compiler found a few more places where the code could
2088 be tidied up in order to avoid warnings.
2090 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2091 first of these names is deprecated in the latest Autoconf in favour of the name
2092 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2093 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2094 analogy I have changed it to CFLAGS_FOR_BUILD.
2096 4. Added -no-undefined to the linking command in the Makefile, because this is
2097 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2098 linking step for the pcreposix library.
2100 5. PCRE was failing to diagnose the case of two named groups with the same
2101 name.
2103 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2104 literal character that is needed in the subject for a match, and scans along to
2105 ensure that it is present before embarking on the full matching process. This
2106 saves time in cases of nested unlimited repeats that are never going to match.
2107 Problem: the scan can take a lot of time if the subject is very long (e.g.
2108 megabytes), thus penalizing straightforward matches. It is now done only if the
2109 amount of subject to be scanned is less than 1000 bytes.
2111 7. A lesser problem with the same optimization is that it was recording the
2112 first character of an anchored pattern as "needed", thus provoking a search
2113 right along the subject, even when the first match of the pattern was going to
2114 fail. The "needed" character is now not set for anchored patterns, unless it
2115 follows something in the pattern that is of non-fixed length. Thus, it still
2116 fulfils its original purpose of finding quick non-matches in cases of nested
2117 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2120 Version 4.0 17-Feb-03
2121 ---------------------
2123 1. If a comment in an extended regex that started immediately after a meta-item
2124 extended to the end of string, PCRE compiled incorrect data. This could lead to
2125 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2127 2. Moved to autoconf 2.53 and libtool 1.4.2.
2129 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2130 the special perltest8 script is no longer needed - all the tests can be run
2131 from a single perltest script.
2133 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2134 by \s. It has now been removed in PCRE. This means it isn't recognized as
2135 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2136 class [:space:] *does* include VT, thereby creating a mess.
2138 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2139 space and tab.
2141 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2142 its new features into the main test script, reducing the number of scripts.
2144 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2145 were backward compatible, and made the (?i) apply to the whole pattern, as if
2146 /i were given. Now it behaves more logically, and applies the option setting
2147 only to what follows. PCRE has been changed to follow suit. However, if it
2148 finds options settings right at the start of the pattern, it extracts them into
2149 the global options, as before. Thus, they show up in the info data.
2151 8. Added support for the \Q...\E escape sequence. Characters in between are
2152 treated as literals. This is slightly different from Perl in that $ and @ are
2153 also handled as literals inside the quotes. In Perl, they will cause variable
2154 interpolation. Note the following examples:
2156 Pattern PCRE matches Perl matches
2158 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2159 \Qabc\$xyz\E abc\$xyz abc\$xyz
2160 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2162 For compatibility with Perl, \Q...\E sequences are recognized inside character
2163 classes as well as outside them.
2165 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2166 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2167 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2168 signed/unsigned warnings.
2170 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2171 option for pcretest, so I've replaced it by a simple function that does just
2172 that job.
2174 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2175 "pcregrep -".
2177 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2178 Java package. This provides some syntactic sugar for simple cases of what my
2179 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2180 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2181 item, you can use this simplified notation. Note that only makes sense with
2182 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2183 greediness, whatever the setting of the PCRE_UNGREEDY option.
2185 13. A change of greediness default within a pattern was not taking effect at
2186 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2187 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2188 was abstracted outside.
2190 14. PCRE now supports the \G assertion. It is true when the current matching
2191 position is at the start point of the match. This differs from \A when the
2192 starting offset is non-zero. Used with the /g option of pcretest (or similar
2193 code), it works in the same way as it does for Perl's /g option. If all
2194 alternatives of a regex begin with \G, the expression is anchored to the start
2195 match position, and the "anchored" flag is set in the compiled expression.
2197 15. Some bugs concerning the handling of certain option changes within patterns
2198 have been fixed. These applied to options other than (?ims). For example,
2199 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2200 been the other way round. Some of this was related to change 7 above.
2202 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2203 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2204 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2205 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2207 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2208 mode. Unlike ".", it always matches newline, whatever the setting of
2209 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2210 assertions. Perl allows it, but it doesn't (in general) work because it can't
2211 calculate the length of the lookbehind. At least, that's the case for Perl
2212 5.8.0 - I've been told they are going to document that it doesn't work in
2213 future.
2215 18. Added an error diagnosis for escapes that PCRE does not support: these are
2216 \L, \l, \N, \P, \p, \U, \u, and \X.
2218 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2219 reading past the end of the pattern in cases such as /[abcd/.
2221 20. PCRE was getting more memory than necessary for patterns with classes that
2222 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2224 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2225 compiling PCRE for use with Virtual Pascal.
2227 22. Small fix to the Makefile to make it work properly if the build is done
2228 outside the source tree.
2230 23. Added a new extension: a condition to go with recursion. If a conditional
2231 subpattern starts with (?(R) the "true" branch is used if recursion has
2232 happened, whereas the "false" branch is used only at the top level.
2234 24. When there was a very long string of literal characters (over 255 bytes
2235 without UTF support, over 250 bytes with UTF support), the computation of how
2236 much memory was required could be incorrect, leading to segfaults or other
2237 strange effects.
2239 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2240 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2241 there was a subsequent back reference to those brackets. This meant that, for
2242 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2243 possible to check for precisely this case. All we can do is abandon the
2244 optimization if .* occurs inside capturing brackets when there are any back
2245 references whatsoever. (See below for a better fix that came later.)
2247 26. The handling of the optimization for finding the first character of a
2248 non-anchored pattern, and for finding a character that is required later in the
2249 match were failing in some cases. This didn't break the matching; it just
2250 failed to optimize when it could. The way this is done has been re-implemented.
2252 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2254 28. Added a new feature that provides some of the functionality that Perl
2255 provides with (?{...}). The facility is termed a "callout". The way it is done
2256 in PCRE is for the caller to provide an optional function, by setting
2257 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2258 global variable. By default it is unset, which disables all calling out. To get
2259 the function called, the regex must include (?C) at appropriate points. This
2260 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2261 This provides a means of identifying different callout points. When PCRE
2262 reaches such a point in the regex, if pcre_callout has been set, the external
2263 function is called. It is provided with data in a structure called
2264 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2265 matching continues; if it returns a non-zero value, the match at the current
2266 point fails. However, backtracking will occur if possible. [This was changed
2267 later and other features added - see item 49 below.]
2269 29. pcretest is upgraded to test the callout functionality. It provides a
2270 callout function that displays information. By default, it shows the start of
2271 the match and the current position in the text. There are some new data escapes
2272 to vary what happens:
2274 \C+ in addition, show current contents of captured substrings
2275 \C- do not supply a callout function
2276 \C!n return 1 when callout number n is reached
2277 \C!n!m return 1 when callout number n is reached for the mth time
2279 30. If pcregrep was called with the -l option and just a single file name, it
2280 output "<stdin>" if a match was found, instead of the file name.
2282 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2283 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2284 pcre_exec(). This saves a malloc/free per call. The default value of
2285 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2286 when configuring.
2288 32. The default maximum size of a compiled pattern is 64K. There have been a
2289 few cases of people hitting this limit. The code now uses macros to handle the
2290 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2291 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2292 configuring. Tests 2 and 5 work only with 2-byte links because they output
2293 debugging information about compiled patterns.
2295 33. Internal code re-arrangements:
2297 (a) Moved the debugging function for printing out a compiled regex into
2298 its own source file (printint.c) and used #include to pull it into
2299 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2300 separate copies.
2302 (b) Defined the list of op-code names for debugging as a macro in
2303 internal.h so that it is next to the definition of the opcodes.
2305 (c) Defined a table of op-code lengths for simpler skipping along compiled
2306 code. This is again a macro in internal.h so that it is next to the
2307 definition of the opcodes.
2309 34. Added support for recursive calls to individual subpatterns, along the
2310 lines of Robin Houston's patch (but implemented somewhat differently).
2312 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2313 allow it to read and process whole directories in Win32. This code was
2314 contributed by Lionel Fourquaux; it has not been tested by me.
2316 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2317 used to name a group. Names consist of alphanumerics and underscores, and must
2318 be unique. Back references use the syntax (?P=name) and recursive calls use
2319 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2320 numbers. The function pcre_fullinfo() can be used after compilation to extract
2321 a name/number map. There are three relevant calls:
2323 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2324 PCRE_INFO_NAMECOUNT yields the number of entries
2325 PCRE_INFO_NAMETABLE yields a pointer to the map.
2327 The map is a vector of fixed-size entries. The size of each entry depends on
2328 the length of the longest name used. The first two bytes of each entry are the
2329 group number, most significant byte first. There follows the corresponding
2330 name, zero terminated. The names are in alphabetical order.
2332 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2333 case instead of 255. Making it the same both with and without UTF-8 support
2334 means that the same test output works with both.
2336 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2337 calling malloc() with a zero argument.
2339 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2340 optimization. I've improved things by keeping a bitmap of backreferences with
2341 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2342 fact referenced, the optimization can be applied. It is unlikely that a
2343 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2344 the match to follow \n) will appear inside brackets with a number greater than
2345 31, but if it does, any back reference > 31 suppresses the optimization.
2347 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2348 of disabling numbered capturing parentheses. Any opening parenthesis that is
2349 not followed by ? behaves as if it were followed by ?: but named parentheses
2350 can still be used for capturing (and they will acquire numbers in the usual
2351 way).
2353 41. Redesigned the return codes from the match() function into yes/no/error so
2354 that errors can be passed back from deep inside the nested calls. A malloc
2355 failure while inside a recursive subpattern call now causes the
2356 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2358 42. It is now possible to set a limit on the number of times the match()
2359 function is called in a call to pcre_exec(). This facility makes it possible to
2360 limit the amount of recursion and backtracking, though not in a directly
2361 obvious way, because the match() function is used in a number of different
2362 circumstances. The count starts from zero for each position in the subject
2363 string (for non-anchored patterns). The default limit is, for compatibility, a
2364 large number, namely 10 000 000. You can change this in two ways:
2366 (a) When configuring PCRE before making, you can use --with-match-limit=n
2367 to set a default value for the compiled library.
2369 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2370 a different value is set. See 45 below.
2372 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2374 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2375 of things that can be changed at compile time. The first argument specifies
2376 what is wanted and the second points to where the information is to be placed.
2377 The current list of available information is:
2381 The output is an integer that is set to one if UTF-8 support is available;
2382 otherwise it is set to zero.
2386 The output is an integer that it set to the value of the code that is used for
2387 newline. It is either LF (10) or CR (13).
2391 The output is an integer that contains the number of bytes used for internal
2392 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2396 The output is an integer that contains the threshold above which the POSIX
2397 interface uses malloc() for output vectors. See item 31 above.
2401 The output is an unsigned integer that contains the default limit of the number
2402 of match() calls in a pcre_exec() execution. See 42 above.
2404 44. pcretest has been upgraded by the addition of the -C option. This causes it
2405 to extract all the available output from the new pcre_config() function, and to
2406 output it. The program then exits immediately.
2408 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2409 order to support additional features. One way would have been to define
2410 pcre_exec2() (for example) with extra arguments, but this would not have been
2411 extensible, and would also have required all calls to the original function to
2412 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2413 is used for passing in "extra" data from pcre_study().
2415 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2416 contains the following fields:
2418 flags a bitmap indicating which of the following fields are set
2419 study_data opaque data from pcre_study()
2420 match_limit a way of specifying a limit on match() calls for a specific
2421 call to pcre_exec()
2422 callout_data data for callouts (see 49 below)
2424 The flag bits are also defined in pcre.h, and are
2430 The pcre_study() function now returns one of these new pcre_extra blocks, with
2431 the actual study data pointed to by the study_data field, and the
2432 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2433 before. That is, this change is entirely upwards-compatible and requires no
2434 change to existing code.
2436 If you want to pass in additional data to pcre_exec(), you can either place it
2437 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2438 block.
2440 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2441 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2442 times with different match limits, until it finds the minimum value needed for
2443 pcre_exec() to complete. The value is then output. This can be instructive; for
2444 most simple matches the number is quite small, but for pathological cases it
2445 gets very large very quickly.
2447 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2448 returns the size of the data block pointed to by the study_data field in a
2449 pcre_extra block, that is, the value that was passed as the argument to
2450 pcre_malloc() when PCRE was getting memory in which to place the information
2451 created by pcre_study(). The fourth argument should point to a size_t variable.
2452 pcretest has been extended so that this information is shown after a successful
2453 pcre_study() call when information about the compiled regex is being displayed.
2455 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2456 because what follows is always an absolute path. (Later: it turns out that this
2457 is more than cosmetic for MinGW, because it doesn't like empty path
2458 components.)
2460 49. Some changes have been made to the callout feature (see 28 above):
2462 (i) A callout function now has three choices for what it returns:
2464 0 => success, carry on matching
2465 > 0 => failure at this point, but backtrack if possible
2466 < 0 => serious error, return this value from pcre_exec()
2468 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2469 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2470 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2471 use by callout functions. It will never be used by PCRE itself.
2473 (ii) The pcre_extra structure (see 45 above) has a void * field called
2474 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2475 pcre_callout_block structure has a field of the same name. The contents of
2476 the field passed in the pcre_extra structure are passed to the callout
2477 function in the corresponding field in the callout block. This makes it
2478 easier to use the same callout-containing regex from multiple threads. For
2479 testing, the pcretest program has a new data escape
2481 \C*n pass the number n (may be negative) as callout_data
2483 If the callout function in pcretest receives a non-zero value as
2484 callout_data, it returns that value.
2486 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2487 there were some redundant $(CFLAGS) in commands that are now specified as
2488 $(LINK), which already includes $(CFLAGS).
2490 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2491 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2492 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2493 one-byte characters throughout. Note that case-insensitive matching applies
2494 only to characters whose values are less than 256. PCRE doesn't support the
2495 notion of cases for higher-valued characters.
2497 (i) A character class whose characters are all within 0-255 is handled as
2498 a bit map, and the map is inverted for negative classes. Previously, a
2499 character > 255 always failed to match such a class; however it should
2500 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2502 (ii) A negated character class with a single character < 255 is coded as
2503 "not this character" (OP_NOT). This wasn't working properly when the test
2504 character was multibyte, either singly or repeated.
2506 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2507 mode, for example: \x{100}{2,3}.
2509 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2510 singly or repeated) now correctly test multibyte characters. However,
2511 PCRE doesn't recognize any characters with values greater than 255 as
2512 digits, spaces, or word characters. Such characters always match \D, \S,
2513 and \W, and never match \d, \s, or \w.
2515 (v) Classes may now contain characters and character ranges with values
2516 greater than 255. For example: [ab\x{100}-\x{400}].
2518 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2519 PCRE in UTF-8 mode.
2521 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2522 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2523 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2524 value.)
2526 53. The single man page has become too large. I have therefore split it up into
2527 a number of separate man pages. These also give rise to individual HTML pages;
2528 these are now put in a separate directory, and there is an index.html page that
2529 lists them all. Some hyperlinking between the pages has been installed.
2531 54. Added convenience functions for handling named capturing parentheses.
2533 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2534 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2535 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2536 are faulted.
2538 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2539 calling configure. These values are used when compiling the dftables.c program
2540 which is run to generate the source of the default character tables. They
2541 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2542 you will need to set these values.
2544 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2547 Version 3.9 02-Jan-02
2548 ---------------------
2550 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2552 2. If --disable-static was given, the building process failed when trying to
2553 build pcretest and pcregrep. (For some reason it was using libtool to compile
2554 them, which is not right, as they aren't part of the library.)
2557 Version 3.8 18-Dec-01
2558 ---------------------
2560 1. The experimental UTF-8 code was completely screwed up. It was packing the
2561 bytes in the wrong order. How dumb can you get?
2564 Version 3.7 29-Oct-01
2565 ---------------------
2567 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2568 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2569 this didn't happen under Solaris 8, where I normally test things.
2571 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2572 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2573 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2574 this fix an hour or so after the initial 3.7 release.)
2577 Version 3.6 23-Oct-01
2578 ---------------------
2580 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2581 offsets passed as NULL with zero offset count.
2583 2. The config.guess and config.sub files had not been updated when I moved to
2584 the latest autoconf.
2587 Version 3.5 15-Aug-01
2588 ---------------------
2590 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2591 had been forgotten.
2593 2. By using declared but undefined structures, we can avoid using "void"
2594 definitions in pcre.h while keeping the internal definitions of the structures
2595 private.
2597 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2598 user point of view, this means that both static and shared libraries are built
2599 by default, but this can be individually controlled. More of the work of
2600 handling this static/shared cases is now inside libtool instead of PCRE's make
2601 file.
2603 4. The pcretest utility is now installed along with pcregrep because it is
2604 useful for users (to test regexs) and by doing this, it automatically gets
2605 relinked by libtool. The documentation has been turned into a man page, so
2606 there are now .1, .txt, and .html versions in /doc.
2608 5. Upgrades to pcregrep:
2609 (i) Added long-form option names like gnu grep.
2610 (ii) Added --help to list all options with an explanatory phrase.
2611 (iii) Added -r, --recursive to recurse into sub-directories.
2612 (iv) Added -f, --file to read patterns from a file.
2614 6. pcre_exec() was referring to its "code" argument before testing that
2615 argument for NULL (and giving an error if it was NULL).
2617 7. Upgraded Makefile.in to allow for compiling in a different directory from
2618 the source directory.
2620 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2621 options bits, the pointer it was passed was to an int instead of to an unsigned
2622 long int. This mattered only on 64-bit systems.
2624 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2625 generated) instead of pcre.in, which it its source. Also made the same change
2626 in several of the .c files.
2628 10. A new release of gcc defines printf() as a macro, which broke pcretest
2629 because it had an ifdef in the middle of a string argument for printf(). Fixed
2630 by using separate calls to printf().
2632 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2633 script, to force use of CR or LF instead of \n in the source. On non-Unix
2634 systems, the value can be set in config.h.
2636 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2637 absolute limit. Changed the text of the error message to make this clear, and
2638 likewise updated the man page.
2640 13. The limit of 99 on the number of capturing subpatterns has been removed.
2641 The new limit is 65535, which I hope will not be a "real" limit.
2644 Version 3.4 22-Aug-00
2645 ---------------------
2647 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2649 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2652 Version 3.3 01-Aug-00
2653 ---------------------
2655 1. If an octal character was given, but the value was greater than \377, it
2656 was not getting masked to the least significant bits, as documented. This could
2657 lead to crashes in some systems.
2659 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2660 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2662 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2663 These just pass their arguments on to (pcre_free)(), but they are provided
2664 because some uses of PCRE bind it to non-C systems that can call its functions,
2665 but cannot call free() or pcre_free() directly.
2667 4. Add "make test" as a synonym for "make check". Corrected some comments in
2668 the Makefile.
2670 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2671 Makefile.
2673 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2674 command called pgrep for grepping around the active processes.
2676 7. Added the beginnings of support for UTF-8 character strings.
2678 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2679 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2680 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2681 out for the ar command.)
2684 Version 3.2 12-May-00
2685 ---------------------
2687 This is purely a bug fixing release.
2689 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2690 of ZA. This was just one example of several cases that could provoke this bug,
2691 which was introduced by change 9 of version 2.00. The code for breaking
2692 infinite loops after an iteration that matches an empty string was't working
2693 correctly.
2695 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2696 when matched against abbab (for example). After matching an empty string, it
2697 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2698 caused it to match further down the string than it should.
2700 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2701 was there because it doesn't seem to be needed, and it causes trouble on some
2702 systems, as it is not a Standard C header. It has been removed.
2704 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2705 were reported on the Macintosh. The changes were from
2707 while ((c = *(++ptr)) != 0 && c != '\n');
2708 to
2709 while ((c = *(++ptr)) != 0 && c != '\n') ;
2711 Totally extraordinary, but if that's what it takes...
2713 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2714 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2715 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2716 assumes the way PCRE uses memmove() (always moving upwards).
2718 6. PCRE is being used in one environment where strchr() is not available. There
2719 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2720 faster code anyway.
2723 Version 3.1 09-Feb-00
2724 ---------------------
2726 The only change in this release is the fixing of some bugs in Makefile.in for
2727 the "install" target:
2729 (1) It was failing to install pcreposix.h.
2731 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2734 Version 3.0 01-Feb-00
2735 ---------------------
2737 1. Add support for the /+ modifier to perltest (to output $` like it does in
2738 pcretest).
2740 2. Add support for the /g modifier to perltest.
2742 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2743 matches null strings.
2745 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2746 pattern. Perl treats empty patterns specially - it reuses the most recent
2747 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2748 effect.
2750 5. The POSIX interface was broken in that it was just handing over the POSIX
2751 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2752 required a bigger vector, with some working space on the end. This means that
2753 the POSIX wrapper now has to get and free some memory, and copy the results.
2755 6. Added some simple autoconf support, placing the test data and the
2756 documentation in separate directories, re-organizing some of the
2757 information files, and making it build pcre-config (a GNU standard). Also added
2758 libtool support for building PCRE as a shared library, which is now the
2759 default.
2761 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2762 09 are not valid octal constants. Single digits will be used for minor values
2763 less than 10.
2765 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2766 existing programs that set these in the POSIX interface can use PCRE without
2767 modification.
2769 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2770 return all that pcre_info() returns, plus additional data. The pcre_info()
2771 function is retained for compatibility, but is considered to be obsolete.
2773 10. Added experimental recursion feature (?R) to handle one common case that
2774 Perl 5.6 will be able to do with (?p{...}).
2776 11. Added support for POSIX character classes like [:alpha:], which Perl is
2777 adopting.
2780 Version 2.08 31-Aug-99
2781 ----------------------
2783 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2784 trying to match at the startoffset position, but instead was moving forward to
2785 the next newline as if a previous match had failed.
2787 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2788 and could get into a loop if a null string was matched other than at the start
2789 of the subject.
2791 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2792 be distinguished at compile time, and for completeness also added PCRE_DATE.
2794 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2795 in GnuWin32 environments.
2798 Version 2.07 29-Jul-99
2799 ----------------------
2801 1. The documentation is now supplied in plain text form and HTML as well as in
2802 the form of man page sources.
2804 2. C++ compilers don't like assigning (void *) values to other pointer types.
2805 In particular this affects malloc(). Although there is no problem in Standard
2806 C, I've put in casts to keep C++ compilers happy.
2808 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2809 should be (const char *).
2811 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2812 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2813 However, I haven't made this a standard facility. The documentation doesn't
2814 mention it, and the Makefile doesn't support it.
2816 5. The Makefile now contains an "install" target, with editable destinations at
2817 the top of the file. The pcretest program is not installed.
2819 6. pgrep -V now gives the PCRE version number and date.
2821 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2822 causing the entire string to be ignored, instead of just the last character.
2824 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2825 non-matching string, it can take a very, very long time, even for strings of
2826 quite modest length, because of the nested recursion. PCRE now does better in
2827 some of these cases. It does this by remembering the last required literal
2828 character in the pattern, and pre-searching the subject to ensure it is present
2829 before running the real match. In other words, it applies a heuristic to detect
2830 some types of certain failure quickly, and in the above example, if presented
2831 with a string that has no trailing " it gives "no match" very quickly.
2833 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2834 other alternatives are tried instead.
2837 Version 2.06 09-Jun-99
2838 ----------------------
2840 1. Change pcretest's output for amount of store used to show just the code
2841 space, because the remainder (the data block) varies in size between 32-bit and
2842 64-bit systems.
2844 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2845 start matching at. This allows lookbehinds to work when searching for multiple
2846 occurrences in a string.
2848 3. Added additional options to pcretest for testing multiple occurrences:
2850 /+ outputs the rest of the string that follows a match
2851 /g loops for multiple occurrences, using the new startoffset argument
2852 /G loops for multiple occurrences by passing an incremented pointer
2854 4. PCRE wasn't doing the "first character" optimization for patterns starting
2855 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2856 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2857 the letter 'x'. On long subject strings, this gives a significant speed-up.
2860 Version 2.05 21-Apr-99
2861 ----------------------
2863 1. Changed the type of magic_number from int to long int so that it works
2864 properly on 16-bit systems.
2866 2. Fixed a bug which caused patterns starting with .* not to work correctly
2867 when the subject string contained newline characters. PCRE was assuming
2868 anchoring for such patterns in all cases, which is not correct because .* will
2869 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2870 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2871 must be retried after every newline in the subject.
2874 Version 2.04 18-Feb-99
2875 ----------------------
2877 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2878 computation of the store needed to hold the pattern was incorrect (too large).
2879 If such patterns were nested a few deep, this could multiply and become a real
2880 problem.
2882 2. Added /M option to pcretest to show the memory requirement of a specific
2883 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2885 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2886 compiled in such a way that the backtracking after subsequent failure was
2887 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2888 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2891 Version 2.03 02-Feb-99
2892 ----------------------
2894 1. Fixed typo and small mistake in man page.
2896 2. Added 4th condition (GPL supersedes if conflict) and created separate
2897 LICENCE file containing the conditions.
2899 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2900 Perl, that is the internal \ allows the delimiter to be included in the
2901 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2902 the final delimiter, add \ to the end of the pattern (to test the error).
2904 4. Added the convenience functions for extracting substrings after a successful
2905 match. Updated pcretest to make it able to test these functions.
2908 Version 2.02 14-Jan-99
2909 ----------------------
2911 1. Initialized the working variables associated with each extraction so that
2912 their saving and restoring doesn't refer to uninitialized store.
2914 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2915 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2916 fix the problem.
2918 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2919 calls, and wasn't printing the correct value for compiling calls. Increased the
2920 default value of LOOPREPEAT, and the number of significant figures in the
2921 times.
2923 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2925 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2926 a building problem on Windows NT with a FAT file system.
2929 Version 2.01 21-Oct-98
2930 ----------------------
2932 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2933 to character tables built by pcre_maketables() in the current locale. If NULL
2934 is passed, the default tables are used.
2937 Version 2.00 24-Sep-98
2938 ----------------------
2940 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2941 it any more.
2943 2. Allow quantification of (?>) groups, and make it work correctly.
2945 3. The first character computation wasn't working for (?>) groups.
2947 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2948 end of the subject) and add 5.005's \z, which really does match only at the
2949 very end of the subject.
2951 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2953 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2954 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2955 localized options. All options to pcre_study() were also removed.
2957 7. Add other new features from 5.005:
2959 $(?<= positive lookbehind
2960 $(?<! negative lookbehind
2961 (?imsx-imsx) added the unsetting capability
2962 such a setting is global if at outer level; local otherwise
2963 (?imsx-imsx:) non-capturing groups with option setting
2964 (?(cond)re|re) conditional pattern matching
2966 A backreference to itself in a repeated group matches the previous
2967 captured string.
2969 8. General tidying up of studying (both automatic and via "study")
2970 consequential on the addition of new assertions.
2972 9. As in 5.005, unlimited repeated groups that could match an empty substring
2973 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2974 runtime if any iteration does actually match an empty substring.
2976 10. Include the RunTest script in the distribution.
2978 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2979 discrepancies, some of which were old and were also with respect to 5.004. They
2980 have now been fixed.
2983 Version 1.09 28-Apr-98
2984 ----------------------
2986 1. A negated single character class followed by a quantifier with a minimum
2987 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2988 program crashes, or just wrong answers. This did not apply to negated classes
2989 containing more than one character, or to minima other than one.
2992 Version 1.08 27-Mar-98
2993 ----------------------
2995 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
2997 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
2998 latter must appear before anything that relies on it in the pattern.
3001 Version 1.07 16-Feb-98
3002 ----------------------
3004 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3005 repeat of a potentially empty string).
3008 Version 1.06 23-Jan-98
3009 ----------------------
3011 1. Added Markus Oberhumer's little patches for C++.
3013 2. Literal strings longer than 255 characters were broken.
3016 Version 1.05 23-Dec-97
3017 ----------------------
3019 1. Negated character classes containing more than one character were failing if
3020 PCRE_CASELESS was set at run time.
3023 Version 1.04 19-Dec-97
3024 ----------------------
3026 1. Corrected the man page, where some "const" qualifiers had been omitted.
3028 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3029 input syntax.
3031 3. Fixed memory leak which occurred when a regex with back references was
3032 matched with an offsets vector that wasn't big enough. The temporary memory
3033 that is used in this case wasn't being freed if the match failed.
3035 4. Tidied pcretest to ensure it frees memory that it gets.
3037 5. Temporary memory was being obtained in the case where the passed offsets
3038 vector was exactly big enough.
3040 6. Corrected definition of offsetof() from change 5 below.
3042 7. I had screwed up change 6 below and broken the rules for the use of
3043 setjmp(). Now fixed.
3046 Version 1.03 18-Dec-97
3047 ----------------------
3049 1. A erroneous regex with a missing opening parenthesis was correctly
3050 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3051 on some systems.
3053 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3054 it was reported that one broken compiler failed on the former because "code" is
3055 also an independent variable.
3057 3. The erroneous regex a[]b caused an array overrun reference.
3059 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3060 fail on data ending with that character. (It was going on too far, and checking
3061 the next character, typically a binary zero.) This was specific to the
3062 optimized code for single-character negative classes.
3064 5. Added a contributed patch from the TIN world which does the following:
3066 + Add an undef for memmove, in case the the system defines a macro for it.
3068 + Add a definition of offsetof(), in case there isn't one. (I don't know
3069 the reason behind this - offsetof() is part of the ANSI standard - but
3070 it does no harm).
3072 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3073 most of the places where whitespace preceded '#'. I have given up and
3074 allowed the remaining 2 cases to be at the margin.
3076 + Rename some variables in pcre to eliminate shadowing. This seems very
3077 pedantic, but does no harm, of course.
3079 6. Moved the call to setjmp() into its own function, to get rid of warnings
3080 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3082 7. Constructs such as \d{8,} were compiling into the equivalent of
3083 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3084 outcome, but in this particular case used more store than had been allocated,
3085 which caused the bug to be discovered because it threw up an internal error.
3087 8. The debugging code in both pcre and pcretest for outputting the compiled
3088 form of a regex was going wrong in the case of back references followed by
3089 curly-bracketed repeats.
3092 Version 1.02 12-Dec-97
3093 ----------------------
3095 1. Typos in pcre.3 and comments in the source fixed.
3097 2. Applied a contributed patch to get rid of places where it used to remove
3098 'const' from variables, and fixed some signed/unsigned and uninitialized
3099 variable warnings.
3101 3. Added the "runtest" target to Makefile.
3103 4. Set default compiler flag to -O2 rather than just -O.
3106 Version 1.01 19-Nov-97
3107 ----------------------
3109 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3110 like /([ab]*)*/, that is, for classes with more than one character in them.
3112 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3113 as /((?>a*))*/ (a PCRE_EXTRA facility).
3116 Version 1.00 18-Nov-97
3117 ----------------------
3119 1. Added compile-time macros to support systems such as SunOS4 which don't have
3120 memmove() or strerror() but have other things that can be used instead.
3122 2. Arranged that "make clean" removes the executables.
3125 Version 0.99 27-Oct-97
3126 ----------------------
3128 1. Fixed bug in code for optimizing classes with only one character. It was
3129 initializing a 32-byte map regardless, which could cause it to run off the end
3130 of the memory it had got.
3132 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3135 Version 0.98 22-Oct-97
3136 ----------------------
3138 1. Fixed bug in code for handling temporary memory usage when there are more
3139 back references than supplied space in the ovector. This could cause segfaults.
3142 Version 0.97 21-Oct-97
3143 ----------------------
3145 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3147 2. Optimized negated single characters not to use a bit map.
3149 3. Brought error texts together as macro definitions; clarified some of them;
3150 fixed one that was wrong - it said "range out of order" when it meant "invalid
3151 escape sequence".
3153 4. Changed some char * arguments to const char *.
3155 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3157 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3158 pcretest.
3161 Version 0.96 16-Oct-97
3162 ----------------------
3164 1. Added a simple "pgrep" utility to the distribution.
3166 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3167 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3168 where "ddd" means "one or more decimal digits".
3170 3. Fixed serious bug. If a pattern had a back reference, but the call to
3171 pcre_exec() didn't supply a large enough ovector to record the related
3172 identifying subpattern, the match always failed. PCRE now remembers the number
3173 of the largest back reference, and gets some temporary memory in which to save
3174 the offsets during matching if necessary, in order to ensure that
3175 backreferences always work.
3177 4. Increased the compatibility with Perl in a number of ways:
3179 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3180 to request this handling. The option can be set at compile or exec time.
3182 (b) $ matches before a terminating newline by default; an option
3183 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3184 mode). The option can be set at compile or exec time.
3186 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3187 the same as Perl's. If the decimal number it represents is less than 10
3188 or there aren't that many previous left capturing parentheses, an octal
3189 escape is read. Inside a character class, it's always an octal escape,
3190 even if it is a single digit.
3192 (d) An escaped but undefined alphabetic character is taken as a literal,
3193 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3194 escapes.
3196 (e) {0} is now permitted. (The previous item is removed from the compiled
3197 pattern).
3199 5. Changed all the names of code files so that the basic parts are no longer
3200 than 10 characters, and abolished the teeny "globals.c" file.
3202 6. Changed the handling of character classes; they are now done with a 32-byte
3203 bit map always.
3205 7. Added the -d and /D options to pcretest to make it possible to look at the
3206 internals of compilation without having to recompile pcre.
3209 Version 0.95 23-Sep-97
3210 ----------------------
3212 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3213 \x20 at the start of a run of normal characters. These were being treated as
3214 real characters, instead of the source characters being re-checked.
3217 Version 0.94 18-Sep-97
3218 ----------------------
3220 1. The functions are now thread-safe, with the caveat that the global variables
3221 containing pointers to malloc() and free() or alternative functions are the
3222 same for all threads.
3224 2. Get pcre_study() to generate a bitmap of initial characters for non-
3225 anchored patterns when this is possible, and use it if passed to pcre_exec().
3228 Version 0.93 15-Sep-97
3229 ----------------------
3231 1. /(b)|(:+)/ was computing an incorrect first character.
3233 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3234 but not actually doing anything yet.
3236 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3237 as Perl does (e.g. [-az] or [az-]).
3239 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3240 all possible positions.
3242 5. Split up into different modules to avoid including unneeded functions in a
3243 compiled binary. However, compile and exec are still in one module. The "study"
3244 function is split off.
3246 6. The character tables are now in a separate module whose source is generated
3247 by an auxiliary program - but can then be edited by hand if required. There are
3248 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3249 toupper() in the code.
3251 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3252 make them global. Abolish the function for setting them, as the caller can now
3253 set them directly.
3256 Version 0.92 11-Sep-97
3257 ----------------------
3259 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3260 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3262 2. Caseless matching was not working in character classes if the characters in
3263 the pattern were in upper case.
3265 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3267 4. Make PCRE_ANCHORED public and accept as a compile option.
3269 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3270 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3271 pass them.
3273 6. Give an error if bad option bits passed at compile or run time.
3275 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3276 pcretest to cause it to pass that flag.
3278 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3279 options, and the first character, if set.
3281 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3284 Version 0.91 10-Sep-97
3285 ----------------------
3287 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3288 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3290 2. PCRE was looping on encountering an indefinitely repeated back reference to
3291 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3292 Perl does - treats the match as successful.
3294 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5