ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 345 - (show annotations)
Mon Apr 28 15:10:02 2008 UTC (13 years, 1 month ago) by ph10
File size: 155838 byte(s)
Tidies for the 7.7-RC1 distribution.
1 ChangeLog for PCRE
2 ------------------
4 Version 7.7 28-Apr-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.
97 17. Compiling pcretest on Windows with readline support failed without the
98 following two fixes: (1) Make the unistd.h include conditional on
99 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
101 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
102 ncurses library to be included for pcretest when ReadLine support is
103 requested, but also to allow for it to be overridden. This patch came from
104 Daniel Bergström.
107 Version 7.6 28-Jan-08
108 ---------------------
110 1. A character class containing a very large number of characters with
111 codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
112 overflow.
114 2. Patch to cut out the "long long" test in pcrecpp_unittest when
115 HAVE_LONG_LONG is not defined.
117 3. Applied Christian Ehrlicher's patch to update the CMake build files to
118 bring them up to date and include new features. This patch includes:
120 - Fixed PH's badly added libz and libbz2 support.
121 - Fixed a problem with static linking.
122 - Added pcredemo. [But later removed - see 7 below.]
123 - Fixed dftables problem and added an option.
124 - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
126 - Added readline support for pcretest.
127 - Added an listing of the option settings after cmake has run.
129 4. A user submitted a patch to Makefile that makes it easy to create
130 "pcre.dll" under mingw when using Configure/Make. I added stuff to
131 Makefile.am that cause it to include this special target, without
132 affecting anything else. Note that the same mingw target plus all
133 the other distribution libraries and programs are now supported
134 when configuring with CMake (see 6 below) instead of with
135 Configure/Make.
137 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
138 This is an attempt to solve the reported problem "pcrecpp::no_arg is not
139 exported in the Windows port". It has not yet been confirmed that the patch
140 solves the problem, but it does no harm.
142 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
143 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
144 with CMake, and also correct the comment about stack recursion.
146 7. Remove the automatic building of pcredemo from the ./configure system and
147 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
148 of a program that users should build themselves after PCRE is installed, so
149 building it automatically is not really right. What is more, it gave
150 trouble in some build environments.
152 8. Further tidies to CMakeLists.txt from Sheri and Christian.
155 Version 7.5 10-Jan-08
156 ---------------------
158 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
159 values in parens when parsing an RE using the C++ wrapper."
161 2. Negative specials like \S did not work in character classes in UTF-8 mode.
162 Characters greater than 255 were excluded from the class instead of being
163 included.
165 3. The same bug as (2) above applied to negated POSIX classes such as
166 [:^space:].
168 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
169 defined or documented. It seems to have been a typo for PCRE_STATIC, so
170 I have changed it.
172 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
173 first named subpattern) and a construct such as (?&a) would reference the
174 first named subpattern whose name started with "a" (in other words, the
175 length check was missing). Both these problems are fixed. "Subpattern name
176 expected" is now given for (?&) (a zero-length name), and this patch also
177 makes it give the same error for \k'' (previously it complained that that
178 was a reference to a non-existent subpattern).
180 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
181 this is right because (?- can be followed by option settings as well as by
182 digits. I have, however, made the messages clearer.
184 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
185 than the number used in the conditional) now cause a compile-time error.
186 This is actually not compatible with Perl, which accepts such patterns, but
187 treats the conditional as always being FALSE (as PCRE used to), but it
188 seems to me that giving a diagnostic is better.
190 8. Change "alphameric" to the more common word "alphanumeric" in comments
191 and messages.
193 9. Fix two occurrences of "backslash" in comments that should have been
194 "backspace".
196 10. Remove two redundant lines of code that can never be obeyed (their function
197 was moved elsewhere).
199 11. The program that makes PCRE's Unicode character property table had a bug
200 which caused it to generate incorrect table entries for sequences of
201 characters that have the same character type, but are in different scripts.
202 It amalgamated them into a single range, with the script of the first of
203 them. In other words, some characters were in the wrong script. There were
204 thirteen such cases, affecting characters in the following ranges:
206 U+002b0 - U+002c1
207 U+0060c - U+0060d
208 U+0061e - U+00612
209 U+0064b - U+0065e
210 U+0074d - U+0076d
211 U+01800 - U+01805
212 U+01d00 - U+01d77
213 U+01d9b - U+01dbf
214 U+0200b - U+0200f
215 U+030fc - U+030fe
216 U+03260 - U+0327f
217 U+0fb46 - U+0fbb1
218 U+10450 - U+1049d
220 12. The -o option (show only the matching part of a line) for pcregrep was not
221 compatible with GNU grep in that, if there was more than one match in a
222 line, it showed only the first of them. It now behaves in the same way as
223 GNU grep.
225 13. If the -o and -v options were combined for pcregrep, it printed a blank
226 line for every non-matching line. GNU grep prints nothing, and pcregrep now
227 does the same. The return code can be used to tell if there were any
228 non-matching lines.
230 14. Added --file-offsets and --line-offsets to pcregrep.
232 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
233 infinitely looping recursion. The bug was that positive lookaheads were not
234 being skipped when checking for a possible empty match (negative lookaheads
235 and both kinds of lookbehind were skipped).
237 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
238 inclusion of <windows.h> to before rather than after the definition of
239 INVALID_FILE_ATTRIBUTES (patch from David Byron).
241 17. Specifying a possessive quantifier with a specific limit for a Unicode
242 character property caused pcre_compile() to compile bad code, which led at
243 runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
244 are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
245 caused the error; without that there was no problem.
247 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
249 19. Added --enable-pcretest-libreadline.
251 20. In pcrecpp.cc, the variable 'count' was incremented twice in
252 RE::GlobalReplace(). As a result, the number of replacements returned was
253 double what it should be. I removed one of the increments, but Craig sent a
254 later patch that removed the other one (the right fix) and added unit tests
255 that check the return values (which was not done before).
257 21. Several CMake things:
259 (1) Arranged that, when cmake is used on Unix, the libraries end up with
260 the names libpcre and libpcreposix, not just pcre and pcreposix.
262 (2) The above change means that pcretest and pcregrep are now correctly
263 linked with the newly-built libraries, not previously installed ones.
267 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
268 crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
269 UTF-8 newline character). The key issue is that the pattern starts .*;
270 this means that the match must be either at the beginning, or after a
271 newline. The bug was in the code for advancing after a failed match and
272 checking that the new position followed a newline. It was not taking
273 account of UTF-8 characters correctly.
275 23. PCRE was behaving differently from Perl in the way it recognized POSIX
276 character classes. PCRE was not treating the sequence [:...:] as a
277 character class unless the ... were all letters. Perl, however, seems to
278 allow any characters between [: and :], though of course it rejects as
279 unknown any "names" that contain non-letters, because all the known class
280 names consist only of letters. Thus, Perl gives an error for [[:1234:]],
281 for example, whereas PCRE did not - it did not recognize a POSIX character
282 class. This seemed a bit dangerous, so the code has been changed to be
283 closer to Perl. The behaviour is not identical to Perl, because PCRE will
284 diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
285 treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
286 Perl does, and where it didn't before.
288 24. Rewrite so as to remove the single use of %n from pcregrep because in some
289 Windows environments %n is disabled by default.
292 Version 7.4 21-Sep-07
293 ---------------------
295 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
296 means that a class such as [\s] counted as "explicit reference to CR or
297 LF". That isn't really right - the whole point of the change was to try to
298 help when there was an actual mention of one of the two characters. So now
299 the change happens only if \r or \n (or a literal CR or LF) character is
300 encountered.
302 2. The 32-bit options word was also used for 6 internal flags, but the numbers
303 of both had grown to the point where there were only 3 bits left.
304 Fortunately, there was spare space in the data structure, and so I have
305 moved the internal flags into a new 16-bit field to free up more option
306 bits.
308 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
309 but did not set the internal JCHANGED flag - either of these is enough to
310 control the way the "get" function works - but the PCRE_INFO_JCHANGED
311 facility is supposed to tell if (?J) was ever used, so now (?J) at the
312 start sets both bits.
314 4. Added options (at build time, compile time, exec time) to change \R from
315 matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
317 5. doc/pcresyntax.html was missing from the distribution.
319 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
320 compatibility, even though it is no longer used.
322 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
323 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
324 windows.h file is present (where different names are used). [This was
325 reversed later after testing - see 16 below.]
327 8. Changed all #include <config.h> to #include "config.h". There were also
328 some further <pcre.h> cases that I changed to "pcre.h".
330 9. When pcregrep was used with the --colour option, it missed the line ending
331 sequence off the lines that it output.
333 10. It was pointed out to me that arrays of string pointers cause lots of
334 relocations when a shared library is dynamically loaded. A technique of
335 using a single long string with a table of offsets can drastically reduce
336 these. I have refactored PCRE in four places to do this. The result is
337 dramatic:
339 Originally: 290
340 After changing UCP table: 187
341 After changing error message table: 43
342 After changing table of "verbs" 36
343 After changing table of Posix names 22
345 Thanks to the folks working on Gregex for glib for this insight.
347 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
348 unicode-properties was also set.
350 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
352 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
353 checked only for CRLF.
355 14. Added casts to pcretest.c to avoid compiler warnings.
357 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
359 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
360 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
361 entirely. This removes changes made in 7 above.
363 17. The CMake files have been updated, and there is now more information about
364 building with CMake in the NON-UNIX-USE document.
367 Version 7.3 28-Aug-07
368 ---------------------
370 1. In the rejigging of the build system that eventually resulted in 7.1, the
371 line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
372 brackets there is not right, since it causes compilers to look for an
373 installed pcre.h, not the version that is in the source that is being
374 compiled (which of course may be different). I have changed it back to:
376 #include "pcre.h"
378 I have a vague recollection that the change was concerned with compiling in
379 different directories, but in the new build system, that is taken care of
380 by the VPATH setting the Makefile.
382 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
383 when the subject happened to end in the byte 0x85 (e.g. if the last
384 character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
385 characters but of course it shouldn't be taken as a newline when it is part
386 of another character. The bug was that, for an unlimited repeat of . in
387 not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
388 characters when looking for a newline.
390 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
392 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
393 in debug output.
395 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
396 long printing in the pcrecpp unittest when running under MinGW.
398 6. ESC_K was left out of the EBCDIC table.
400 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
401 parentheses; I made it 1000, which seemed large enough. Unfortunately, the
402 limit also applies to "virtual nesting" when a pattern is recursive, and in
403 this case 1000 isn't so big. I have been able to remove this limit at the
404 expense of backing off one optimization in certain circumstances. Normally,
405 when pcre_exec() would call its internal match() function recursively and
406 immediately return the result unconditionally, it uses a "tail recursion"
407 feature to save stack. However, when a subpattern that can match an empty
408 string has an unlimited repetition quantifier, it no longer makes this
409 optimization. That gives it a stack frame in which to save the data for
410 checking that an empty string has been matched. Previously this was taken
411 from the 1000-entry workspace that had been reserved. So now there is no
412 explicit limit, but more stack is used.
414 8. Applied Daniel's patches to solve problems with the import/export magic
415 syntax that is required for Windows, and which was going wrong for the
416 pcreposix and pcrecpp parts of the library. These were overlooked when this
417 problem was solved for the main library.
419 9. There were some crude static tests to avoid integer overflow when computing
420 the size of patterns that contain repeated groups with explicit upper
421 limits. As the maximum quantifier is 65535, the maximum group length was
422 set at 30,000 so that the product of these two numbers did not overflow a
423 32-bit integer. However, it turns out that people want to use groups that
424 are longer than 30,000 bytes (though not repeat them that many times).
425 Change 7.0/17 (the refactoring of the way the pattern size is computed) has
426 made it possible to implement the integer overflow checks in a much more
427 dynamic way, which I have now done. The artificial limitation on group
428 length has been removed - we now have only the limit on the total length of
429 the compiled pattern, which depends on the LINK_SIZE setting.
431 10. Fixed a bug in the documentation for get/copy named substring when
432 duplicate names are permitted. If none of the named substrings are set, the
433 functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
434 empty string.
436 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
437 instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
438 because the ] is interpreted as the first data character and the
439 terminating ] is not found. PCRE has been made compatible with Perl in this
440 regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
441 cause memory overwriting.
443 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
444 string has been matched (to stop an infinite loop). It was not recognizing
445 a conditional subpattern that could match an empty string if that
446 subpattern was within another subpattern. For example, it looped when
447 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
448 condition was not nested. This bug has been fixed.
450 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
451 past the start of the subject in the presence of bytes with the top bit
452 set, for example "\x8aBCD".
454 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
455 (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
457 14. Optimized (?!) to (*FAIL).
459 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
460 This restricts code points to be within the range 0 to 0x10FFFF, excluding
461 the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
462 full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
463 does: it's just the validity check that is more restrictive.
465 16. Inserted checks for integer overflows during escape sequence (backslash)
466 processing, and also fixed erroneous offset values for syntax errors during
467 backslash processing.
469 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
470 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
472 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
473 caused an overrun.
475 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
476 something other than just ASCII characters) inside a group that had an
477 unlimited repeat caused a loop at compile time (while checking to see
478 whether the group could match an empty string).
480 20. Debugging a pattern containing \p or \P could cause a crash. For example,
481 [\P{Any}] did so. (Error in the code for printing property names.)
483 21. An orphan \E inside a character class could cause a crash.
485 22. A repeated capturing bracket such as (A)? could cause a wild memory
486 reference during compilation.
488 23. There are several functions in pcre_compile() that scan along a compiled
489 expression for various reasons (e.g. to see if it's fixed length for look
490 behind). There were bugs in these functions when a repeated \p or \P was
491 present in the pattern. These operators have additional parameters compared
492 with \d, etc, and these were not being taken into account when moving along
493 the compiled data. Specifically:
495 (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
496 length.
498 (b) An item such as \pL+ within a repeated group could cause crashes or
499 loops.
501 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
502 "reference to non-existent subpattern" error.
504 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
506 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
507 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
509 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
511 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
512 character were causing crashes (broken optimization).
514 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
515 \p or \P) caused a compile-time loop.
517 28. More problems have arisen in unanchored patterns when CRLF is a valid line
518 break. For example, the unstudied pattern [\r\n]A does not match the string
519 "\r\nA" because change 7.0/46 below moves the current point on by two
520 characters after failing to match at the start. However, the pattern \nA
521 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
522 the same is true. There doesn't seem any very clean way out of this, but
523 what I have chosen to do makes the common cases work: PCRE now takes note
524 of whether there can be an explicit match for \r or \n anywhere in the
525 pattern, and if so, 7.0/46 no longer applies. As part of this change,
526 there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
527 pattern has explicit CR or LF references.
529 29. Added (*CR) etc for changing newline setting at start of pattern.
532 Version 7.2 19-Jun-07
533 ---------------------
535 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
536 which is apparently normally available under Windows.
538 2. Re-jig the pcregrep tests with different newline settings in an attempt
539 to make them independent of the local environment's newline setting.
541 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
543 4. Some of the "internals" tests were previously cut out when the link size
544 was not 2, because the output contained actual offsets. The recent new
545 "Z" feature of pcretest means that these can be cut out, making the tests
546 usable with all link sizes.
548 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
549 stack recursion. This gives a massive performance boost under BSD, but just
550 a small improvement under Linux. However, it saves one field in the frame
551 in all cases.
553 6. Added more features from the forthcoming Perl 5.10:
555 (a) (?-n) (where n is a string of digits) is a relative subroutine or
556 recursion call. It refers to the nth most recently opened parentheses.
558 (b) (?+n) is also a relative subroutine call; it refers to the nth next
559 to be opened parentheses.
561 (c) Conditions that refer to capturing parentheses can be specified
562 relatively, for example, (?(-2)... or (?(+3)...
564 (d) \K resets the start of the current match so that everything before
565 is not part of it.
567 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
569 (f) \g{name} is another synonym - part of Perl 5.10's unification of
570 reference syntax.
572 (g) (?| introduces a group in which the numbering of parentheses in each
573 alternative starts with the same number.
575 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
577 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
580 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
581 terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
582 for detecting groups that can match an empty string.
584 9. A pattern with a very large number of alternatives (more than several
585 hundred) was running out of internal workspace during the pre-compile
586 phase, where pcre_compile() figures out how much memory will be needed. A
587 bit of new cunning has reduced the workspace needed for groups with
588 alternatives. The 1000-alternative test pattern now uses 12 bytes of
589 workspace instead of running out of the 4096 that are available.
591 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
593 11. Applied patch from Google to remove an optimization that didn't quite work.
594 The report of the bug said:
596 pcrecpp::RE("a*").FullMatch("aaa") matches, while
597 pcrecpp::RE("a*?").FullMatch("aaa") does not, and
598 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
600 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
601 it matched the wrong number of bytes.
604 Version 7.1 24-Apr-07
605 ---------------------
607 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
608 that is more "standard", making use of automake and other Autotools. There
609 is some re-arrangement of the files and adjustment of comments consequent
610 on this.
612 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
613 for recursive directory scanning broke on some systems because the files
614 are not scanned in any specific order and on different systems the order
615 was different. A call to "sort" has been inserted into RunGrepTest for the
616 approprate test as a short-term fix. In the longer term there may be an
617 alternative.
619 3. I had an email from Eric Raymond about problems translating some of PCRE's
620 man pages to HTML (despite the fact that I distribute HTML pages, some
621 people do their own conversions for various reasons). The problems
622 concerned the use of low-level troff macros .br and .in. I have therefore
623 removed all such uses from the man pages (some were redundant, some could
624 be replaced by .nf/.fi pairs). The 132html script that I use to generate
625 HTML has been updated to handle .nf/.fi and to complain if it encounters
626 .br or .in.
628 4. Updated comments in configure.ac that get placed in config.h.in and also
629 arranged for config.h to be included in the distribution, with the name
630 config.h.generic, for the benefit of those who have to compile without
631 Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
633 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
634 Weber: (1) pcre_internal.h was missing some function renames; (2) updated
635 makevp.bat for the current PCRE, using the additional files
636 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
638 6. A Windows user reported a minor discrepancy with test 2, which turned out
639 to be caused by a trailing space on an input line that had got lost in his
640 copy. The trailing space was an accident, so I've just removed it.
642 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
643 that is needed.
645 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
646 as "const" (a) because they are and (b) because it helps the PHP
647 maintainers who have recently made a script to detect big data structures
648 in the php code that should be moved to the .rodata section. I remembered
649 to update Builducptable as well, so it won't revert if ucptable.h is ever
650 re-created.
652 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
653 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
654 order to be able to cut out the UTF-8 tables in the latter when UTF-8
655 support is not required. This saves 1.5-2K of code, which is important in
656 some applications.
658 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
659 so as not to refer to the tables, even though these functions will never be
660 called when UTF-8 support is disabled. Otherwise there are problems with a
661 shared library.
663 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
665 (a) It was defining its arguments as char * instead of void *.
667 (b) It was assuming that all moves were upwards in memory; this was true
668 a long time ago when I wrote it, but is no longer the case.
670 The emulated memove() is provided for those environments that have neither
671 memmove() nor bcopy(). I didn't think anyone used it these days, but that
672 is clearly not the case, as these two bugs were recently reported.
674 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
675 and Detrail to create the HTML documentation, the .txt form of the man
676 pages, and it removes trailing spaces from listed files. It also creates
677 pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
678 case, it wraps all the #defines with #ifndefs. This script should be run
679 before "make dist".
681 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
682 with Unicode property support.
684 (a) For a maximizing quantifier, if the two different cases of the
685 character were of different lengths in their UTF-8 codings (there are
686 some cases like this - I found 11), and the matching function had to
687 back up over a mixture of the two cases, it incorrectly assumed they
688 were both the same length.
690 (b) When PCRE was configured to use the heap rather than the stack for
691 recursion during matching, it was not correctly preserving the data for
692 the other case of a UTF-8 character when checking ahead for a match
693 while processing a minimizing repeat. If the check also involved
694 matching a wide character, but failed, corruption could cause an
695 erroneous result when trying to check for a repeat of the original
696 character.
698 13. Some tidying changes to the testing mechanism:
700 (a) The RunTest script now detects the internal link size and whether there
701 is UTF-8 and UCP support by running ./pcretest -C instead of relying on
702 values substituted by "configure". (The RunGrepTest script already did
703 this for UTF-8.) The configure.ac script no longer substitutes the
704 relevant variables.
706 (b) The debugging options /B and /D in pcretest show the compiled bytecode
707 with length and offset values. This means that the output is different
708 for different internal link sizes. Test 2 is skipped for link sizes
709 other than 2 because of this, bypassing the problem. Unfortunately,
710 there was also a test in test 3 (the locale tests) that used /B and
711 failed for link sizes other than 2. Rather than cut the whole test out,
712 I have added a new /Z option to pcretest that replaces the length and
713 offset values with spaces. This is now used to make test 3 independent
714 of link size. (Test 2 will be tidied up later.)
716 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
717 segmentation fault instead of returning the appropriate error message.
719 15. In multiline mode when the newline sequence was set to "any", the pattern
720 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
721 This doesn't seem right; it now treats the CRLF combination as the line
722 ending, and so does not match in that case. It's only a pattern such as ^$
723 that would hit this one: something like ^ABC$ would have failed after \r
724 and then tried again after \r\n.
726 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
727 in an attempt to make files that differ only in their line terminators
728 compare equal. This works on Linux.
730 17. Under certain error circumstances pcregrep might try to free random memory
731 as it exited. This is now fixed, thanks to valgrind.
733 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
734 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
735 was because its rules for how to advance for /g after matching an empty
736 string at the end of a line did not allow for this case. They now check for
737 it specially.
739 20. pcretest is supposed to handle patterns and data of any length, by
740 extending its buffers when necessary. It was getting this wrong when the
741 buffer for a data line had to be extended.
743 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
744 CRLF as a newline sequence.
746 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
747 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
748 I have nevertheless tidied it up.
750 23. Added some casts to kill warnings from HP-UX ia64 compiler.
752 24. Added a man page for pcre-config.
755 Version 7.0 19-Dec-06
756 ---------------------
758 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
759 moving to gcc 4.1.1.
761 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
762 sys/time.h, which is documented as needed for this function. It doesn't
763 seem to matter on Linux, but it showed up on some releases of OS X.
765 3. It seems that there are systems where bytes whose values are greater than
766 127 match isprint() in the "C" locale. The "C" locale should be the
767 default when a C program starts up. In most systems, only ASCII printing
768 characters match isprint(). This difference caused the output from pcretest
769 to vary, making some of the tests fail. I have changed pcretest so that:
771 (a) When it is outputting text in the compiled version of a pattern, bytes
772 other than 32-126 are always shown as hex escapes.
774 (b) When it is outputting text that is a matched part of a subject string,
775 it does the same, unless a different locale has been set for the match
776 (using the /L modifier). In this case, it uses isprint() to decide.
778 4. Fixed a major bug that caused incorrect computation of the amount of memory
779 required for a compiled pattern when options that changed within the
780 pattern affected the logic of the preliminary scan that determines the
781 length. The relevant options are -x, and -i in UTF-8 mode. The result was
782 that the computed length was too small. The symptoms of this bug were
783 either the PCRE error "internal error: code overflow" from pcre_compile(),
784 or a glibc crash with a message such as "pcretest: free(): invalid next
785 size (fast)". Examples of patterns that provoked this bug (shown in
786 pcretest format) are:
788 /(?-x: )/x
789 /(?x)(?-x: \s*#\s*)/
790 /((?i)[\x{c0}])/8
791 /(?i:[\x{c0}])/8
793 HOWEVER: Change 17 below makes this fix obsolete as the memory computation
794 is now done differently.
796 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
797 wrapper classes; (b) implement a new function in the C++ scanner that is
798 more efficient than the old way of doing things because it avoids levels of
799 recursion in the regex matching; (c) add a paragraph to the documentation
800 for the FullMatch() function.
802 6. The escape sequence \n was being treated as whatever was defined as
803 "newline". Not only was this contrary to the documentation, which states
804 that \n is character 10 (hex 0A), but it also went horribly wrong when
805 "newline" was defined as CRLF. This has been fixed.
807 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
808 was being set to -1 for the "end of line" case (supposedly a value that no
809 character can have). Though this value is never used (the check for end of
810 line is "zero bytes in current character"), it caused compiler complaints.
811 I've changed it to 0xffffffff.
813 8. In pcre_version.c, the version string was being built by a sequence of
814 C macros that, in the event of PCRE_PRERELEASE being defined as an empty
815 string (as it is for production releases) called a macro with an empty
816 argument. The C standard says the result of this is undefined. The gcc
817 compiler treats it as an empty string (which was what was wanted) but it is
818 reported that Visual C gives an error. The source has been hacked around to
819 avoid this problem.
821 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
822 builds of pcretest, and changed the call to _setmode() to use _O_BINARY
823 instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
824 of them did).
826 10. Originally, pcretest opened its input and output without "b"; then I was
827 told that "b" was needed in some environments, so it was added for release
828 5.0 to both the input and output. (It makes no difference on Unix-like
829 systems.) Later I was told that it is wrong for the input on Windows. I've
830 now abstracted the modes into two macros, to make it easier to fiddle with
831 them, and removed "b" from the input mode under Windows.
833 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
835 12. Added -help and --help to pcretest as an official way of being reminded
836 of the options.
838 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
839 and pcrecpp.cc because they annoy compilers at high warning levels.
841 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
843 15. Fixed an occurrence of == in configure.ac that should have been = (shell
844 scripts are not C programs :-) and which was not noticed because it works
845 on Linux.
847 16. pcretest is supposed to handle any length of pattern and data line (as one
848 line or as a continued sequence of lines) by extending its input buffer if
849 necessary. This feature was broken for very long pattern lines, leading to
850 a string of junk being passed to pcre_compile() if the pattern was longer
851 than about 50K.
853 17. I have done a major re-factoring of the way pcre_compile() computes the
854 amount of memory needed for a compiled pattern. Previously, there was code
855 that made a preliminary scan of the pattern in order to do this. That was
856 OK when PCRE was new, but as the facilities have expanded, it has become
857 harder and harder to keep it in step with the real compile phase, and there
858 have been a number of bugs (see for example, 4 above). I have now found a
859 cunning way of running the real compile function in a "fake" mode that
860 enables it to compute how much memory it would need, while actually only
861 ever using a few hundred bytes of working memory and without too many
862 tests of the mode. This should make future maintenance and development
863 easier. A side effect of this work is that the limit of 200 on the nesting
864 depth of parentheses has been removed (though this was never a serious
865 limitation, I suspect). However, there is a downside: pcre_compile() now
866 runs more slowly than before (30% or more, depending on the pattern). I
867 hope this isn't a big issue. There is no effect on runtime performance.
869 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
870 newline (only possible for the last line of a file) and it was a
871 pattern that set a locale (followed by /Lsomething), pcretest crashed.
873 19. Added additional timing features to pcretest. (1) The -tm option now times
874 matching only, not compiling. (2) Both -t and -tm can be followed, as a
875 separate command line item, by a number that specifies the number of
876 repeats to use when timing. The default is 50000; this gives better
877 precision, but takes uncomfortably long for very large patterns.
879 20. Extended pcre_study() to be more clever in cases where a branch of a
880 subpattern has no definite first character. For example, (a*|b*)[cd] would
881 previously give no result from pcre_study(). Now it recognizes that the
882 first character must be a, b, c, or d.
884 21. There was an incorrect error "recursive call could loop indefinitely" if
885 a subpattern (or the entire pattern) that was being tested for matching an
886 empty string contained only one non-empty item after a nested subpattern.
887 For example, the pattern (?>\x{100}*)\d(?R) provoked this error
888 incorrectly, because the \d was being skipped in the check.
890 22. The pcretest program now has a new pattern option /B and a command line
891 option -b, which is equivalent to adding /B to every pattern. This causes
892 it to show the compiled bytecode, without the additional information that
893 -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
894 is the same as /B/I).
896 23. A new optimization is now able automatically to treat some sequences such
897 as a*b as a*+b. More specifically, if something simple (such as a character
898 or a simple class like \d) has an unlimited quantifier, and is followed by
899 something that cannot possibly match the quantified thing, the quantifier
900 is automatically "possessified".
902 24. A recursive reference to a subpattern whose number was greater than 39
903 went wrong under certain circumstances in UTF-8 mode. This bug could also
904 have affected the operation of pcre_study().
906 25. Realized that a little bit of performance could be had by replacing
907 (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
909 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
911 27. Possessive quantifiers such as a++ were previously implemented by turning
912 them into atomic groups such as ($>a+). Now they have their own opcodes,
913 which improves performance. This includes the automatically created ones
914 from 23 above.
916 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
917 lookahead was broken if it was not anchored. PCRE was mistakenly expecting
918 the first matched character to be a colon. This applied both to named and
919 numbered groups.
921 29. The ucpinternal.h header file was missing its idempotency #ifdef.
923 30. I was sent a "project" file called libpcre.a.dev which I understand makes
924 building PCRE on Windows easier, so I have included it in the distribution.
926 31. There is now a check in pcretest against a ridiculously large number being
927 returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
928 loop, the loop is abandoned.
930 32. Forward references to subpatterns in conditions such as (?(2)...) where
931 subpattern 2 is defined later cause pcre_compile() to search forwards in
932 the pattern for the relevant set of parentheses. This search went wrong
933 when there were unescaped parentheses in a character class, parentheses
934 escaped with \Q...\E, or parentheses in a #-comment in /x mode.
936 33. "Subroutine" calls and backreferences were previously restricted to
937 referencing subpatterns earlier in the regex. This restriction has now
938 been removed.
940 34. Added a number of extra features that are going to be in Perl 5.10. On the
941 whole, these are just syntactic alternatives for features that PCRE had
942 previously implemented using the Python syntax or my own invention. The
943 other formats are all retained for compatibility.
945 (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
946 as (?P<name>...). The new forms, as well as being in Perl 5.10, are
947 also .NET compatible.
949 (b) A recursion or subroutine call to a named group can now be defined as
950 (?&name) as well as (?P>name).
952 (c) A backreference to a named group can now be defined as \k<name> or
953 \k'name' as well as (?P=name). The new forms, as well as being in Perl
954 5.10, are also .NET compatible.
956 (d) A conditional reference to a named group can now use the syntax
957 (?(<name>) or (?('name') as well as (?(name).
959 (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
960 groups (named and numbered) that are never evaluated inline, but can be
961 called as "subroutines" from elsewhere. In effect, the DEFINE condition
962 is always false. There may be only one alternative in such a group.
964 (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
965 as the simple (?(R). The condition is true only if the most recent
966 recursion is that of the given number or name. It does not search out
967 through the entire recursion stack.
969 (g) The escape \gN or \g{N} has been added, where N is a positive or
970 negative number, specifying an absolute or relative reference.
972 35. Tidied to get rid of some further signed/unsigned compiler warnings and
973 some "unreachable code" warnings.
975 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
976 things, this adds five new scripts.
978 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
979 There were also incompatibilities regarding the handling of \Q..\E inside
980 character classes, for example with patterns like [\Qa\E-\Qz\E] where the
981 hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
983 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
984 matches an empty string, and forcibly breaks the loop. There were bugs in
985 this code in non-simple cases. For a pattern such as ^(a()*)* matched
986 against aaaa the result was just "a" rather than "aaaa", for example. Two
987 separate and independent bugs (that affected different cases) have been
988 fixed.
990 39. Refactored the code to abolish the use of different opcodes for small
991 capturing bracket numbers. This is a tidy that I avoided doing when I
992 removed the limit on the number of capturing brackets for 3.5 back in 2001.
993 The new approach is not only tidier, it makes it possible to reduce the
994 memory needed to fix the previous bug (38).
996 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
997 sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
998 processing dot, circumflex, or dollar metacharacters, or #-comments in /x
999 mode.
1001 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1002 report.
1004 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1005 copy construction and assignment in the C++ wrapper.
1007 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1008 couple of bugs that could have given wrong results in the "--newline=crlf"
1009 case.
1011 44. Added a number of casts and did some reorganization of signed/unsigned int
1012 variables following suggestions from Dair Grant. Also renamed the variable
1013 "this" as "item" because it is a C++ keyword.
1015 45. Arranged for dftables to add
1017 #include "pcre_internal.h"
1019 to pcre_chartables.c because without it, gcc 4.x may remove the array
1020 definition from the final binary if PCRE is built into a static library and
1021 dead code stripping is activated.
1023 46. For an unanchored pattern, if a match attempt fails at the start of a
1024 newline sequence, and the newline setting is CRLF or ANY, and the next two
1025 characters are CRLF, advance by two characters instead of one.
1028 Version 6.7 04-Jul-06
1029 ---------------------
1031 1. In order to handle tests when input lines are enormously long, pcretest has
1032 been re-factored so that it automatically extends its buffers when
1033 necessary. The code is crude, but this _is_ just a test program. The
1034 default size has been increased from 32K to 50K.
1036 2. The code in pcre_study() was using the value of the re argument before
1037 testing it for NULL. (Of course, in any sensible call of the function, it
1038 won't be NULL.)
1040 3. The memmove() emulation function in pcre_internal.h, which is used on
1041 systems that lack both memmove() and bcopy() - that is, hardly ever -
1042 was missing a "static" storage class specifier.
1044 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1045 containing an extended class (one that cannot be represented by a bitmap
1046 because it contains high-valued characters or Unicode property items, e.g.
1047 [\pZ]). Almost always one would set UTF-8 mode when processing such a
1048 pattern, but PCRE should not loop if you do not (it no longer does).
1049 [Detail: two cases were found: (a) a repeated subpattern containing an
1050 extended class; (b) a recursive reference to a subpattern that followed a
1051 previous extended class. It wasn't skipping over the extended class
1052 correctly when UTF-8 mode was not set.]
1054 5. A negated single-character class was not being recognized as fixed-length
1055 in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1056 compile error "lookbehind assertion is not fixed length".
1058 6. The RunPerlTest auxiliary script was showing an unexpected difference
1059 between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1060 write a Perl script that can interpret lines of an input file either as
1061 byte characters or as UTF-8, which is what "perltest" was being required to
1062 do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1063 can't do is switch easily at run time between having the "use utf8;" pragma
1064 or not. In the end, I fudged it by using the RunPerlTest script to insert
1065 "use utf8;" explicitly for the UTF-8 tests.
1067 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1068 the end of the subject string, contrary to the documentation and to what
1069 Perl does. This was true of both matching functions. Now it matches only at
1070 the start of the subject and immediately after *internal* newlines.
1072 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1073 a pointer to an int instead of a pointer to an unsigned long int. This
1074 caused problems on 64-bit systems.
1076 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1077 instance of the 'standard' template library not being so standard".
1079 10. There was no check on the number of named subpatterns nor the maximum
1080 length of a subpattern name. The product of these values is used to compute
1081 the size of the memory block for a compiled pattern. By supplying a very
1082 long subpattern name and a large number of named subpatterns, the size
1083 computation could be caused to overflow. This is now prevented by limiting
1084 the length of names to 32 characters, and the number of named subpatterns
1085 to 10,000.
1087 11. Subpatterns that are repeated with specific counts have to be replicated in
1088 the compiled pattern. The size of memory for this was computed from the
1089 length of the subpattern and the repeat count. The latter is limited to
1090 65535, but there was no limit on the former, meaning that integer overflow
1091 could in principle occur. The compiled length of a repeated subpattern is
1092 now limited to 30,000 bytes in order to prevent this.
1094 12. Added the optional facility to have named substrings with the same name.
1096 13. Added the ability to use a named substring as a condition, using the
1097 Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1098 are numbers (not recommended). Forward references are permitted.
1100 14. Added forward references in named backreferences (if you see what I mean).
1102 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1103 pattern could run off the end of the subject. For example, the pattern
1104 "(?s)(.{1,5})"8 did this with the subject "ab".
1106 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1107 PCRE_CASELESS was set when matching characters that were quantified with ?
1108 or *.
1110 17. A character class other than a single negated character that had a minimum
1111 but no maximum quantifier - for example [ab]{6,} - was not handled
1112 correctly by pce_dfa_exec(). It would match only one character.
1114 18. A valid (though odd) pattern that looked like a POSIX character
1115 class but used an invalid character after [ (for example [[,abc,]]) caused
1116 pcre_compile() to give the error "Failed: internal error: code overflow" or
1117 in some cases to crash with a glibc free() error. This could even happen if
1118 the pattern terminated after [[ but there just happened to be a sequence of
1119 letters, a binary zero, and a closing ] in the memory that followed.
1121 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1122 over the years. Originally (before any Unicode support), just the bottom 8
1123 bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1124 output from "man perlunicode" includes this:
1126 The regular expression compiler produces polymorphic opcodes. That
1127 is, the pattern adapts to the data and automatically switches to
1128 the Unicode character scheme when presented with Unicode data--or
1129 instead uses a traditional byte scheme when presented with byte
1130 data.
1132 Sadly, a wide octal escape does not cause a switch, and in a string with
1133 no other multibyte characters, these octal escapes are treated as before.
1134 Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1135 /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1136 Unicode string.
1138 I have not perpetrated such confusion in PCRE. Up till now, it took just
1139 the bottom 8 bits, as in old Perl. I have now made octal escapes with
1140 values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1141 translate to the appropriate multibyte character.
1143 29. Applied some refactoring to reduce the number of warnings from Microsoft
1144 and Borland compilers. This has included removing the fudge introduced
1145 seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1146 a warning about an unused variable.
1148 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1149 characters since release 4.0, because Perl (from release 5.004) does not.
1150 [Or at least, is documented not to: some releases seem to be in conflict
1151 with the documentation.] However, when a pattern was studied with
1152 pcre_study() and all its branches started with \s, PCRE still included VT
1153 as a possible starting character. Of course, this did no harm; it just
1154 caused an unnecessary match attempt.
1156 22. Removed a now-redundant internal flag bit that recorded the fact that case
1157 dependency changed within the pattern. This was once needed for "required
1158 byte" processing, but is no longer used. This recovers a now-scarce options
1159 bit. Also moved the least significant internal flag bit to the most-
1160 significant bit of the word, which was not previously used (hangover from
1161 the days when it was an int rather than a uint) to free up another bit for
1162 the future.
1164 23. Added support for CRLF line endings as well as CR and LF. As well as the
1165 default being selectable at build time, it can now be changed at runtime
1166 via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1167 specify that it is scanning data with non-default line endings.
1169 24. Changed the definition of CXXLINK to make it agree with the definition of
1170 LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1172 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1173 recursions. This makes a big different to stack usage for some patterns.
1175 26. If a subpattern containing a named recursion or subroutine reference such
1176 as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1177 the space required for the compiled pattern went wrong and gave too small a
1178 value. Depending on the environment, this could lead to "Failed: internal
1179 error: code overflow at offset 49" or "glibc detected double free or
1180 corruption" errors.
1182 27. Applied patches from Google (a) to support the new newline modes and (b) to
1183 advance over multibyte UTF-8 characters in GlobalReplace.
1185 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1186 difference for some implementation of PCRE in some Windows version.
1188 29. Added some extra testing facilities to pcretest:
1190 \q<number> in a data line sets the "match limit" value
1191 \Q<number> in a data line sets the "match recursion limt" value
1192 -S <number> sets the stack size, where <number> is in megabytes
1194 The -S option isn't available for Windows.
1197 Version 6.6 06-Feb-06
1198 ---------------------
1200 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1201 in pcreposix.h. I have copied the definition from pcre.h.
1203 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1204 because pcre.h is no longer a built file.
1206 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1207 not normally included in the compiled code.
1210 Version 6.5 01-Feb-06
1211 ---------------------
1213 1. When using the partial match feature with pcre_dfa_exec(), it was not
1214 anchoring the second and subsequent partial matches at the new starting
1215 point. This could lead to incorrect results. For example, with the pattern
1216 /1234/, partially matching against "123" and then "a4" gave a match.
1218 2. Changes to pcregrep:
1220 (a) All non-match returns from pcre_exec() were being treated as failures
1221 to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1222 error message is output. Some extra information is given for the
1224 probably the only errors that are likely to be caused by users (by
1225 specifying a regex that has nested indefinite repeats, for instance).
1226 If there are more than 20 of these errors, pcregrep is abandoned.
1228 (b) A binary zero was treated as data while matching, but terminated the
1229 output line if it was written out. This has been fixed: binary zeroes
1230 are now no different to any other data bytes.
1232 (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1233 used to set a locale for matching. The --locale=xxxx long option has
1234 been added (no short equivalent) to specify a locale explicitly on the
1235 pcregrep command, overriding the environment variables.
1237 (d) When -B was used with -n, some line numbers in the output were one less
1238 than they should have been.
1240 (e) Added the -o (--only-matching) option.
1242 (f) If -A or -C was used with -c (count only), some lines of context were
1243 accidentally printed for the final match.
1245 (g) Added the -H (--with-filename) option.
1247 (h) The combination of options -rh failed to suppress file names for files
1248 that were found from directory arguments.
1250 (i) Added the -D (--devices) and -d (--directories) options.
1252 (j) Added the -F (--fixed-strings) option.
1254 (k) Allow "-" to be used as a file name for -f as well as for a data file.
1256 (l) Added the --colo(u)r option.
1258 (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1259 is not present by default.
1261 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1262 items such as (?R) or (?1), when the recursion could match a number of
1263 alternatives. If it matched one of the alternatives, but subsequently,
1264 outside the recursion, there was a failure, the code tried to back up into
1265 the recursion. However, because of the way PCRE is implemented, this is not
1266 possible, and the result was an incorrect result from the match.
1268 In order to prevent this happening, the specification of recursion has
1269 been changed so that all such subpatterns are automatically treated as
1270 atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1272 4. I had overlooked the fact that, in some locales, there are characters for
1273 which isalpha() is true but neither isupper() nor islower() are true. In
1274 the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1275 and ordfeminine) are like this. This affected the treatment of \w and \W
1276 when they appeared in character classes, but not when they appeared outside
1277 a character class. The bit map for "word" characters is now created
1278 separately from the results of isalnum() instead of just taking it from the
1279 upper, lower, and digit maps. (Plus the underscore character, of course.)
1281 5. The above bug also affected the handling of POSIX character classes such as
1282 [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1283 permanent tables. Instead, the bit maps for such a class were previously
1284 created as the appropriate unions of the upper, lower, and digit bitmaps.
1285 Now they are created by subtraction from the [[:word:]] class, which has
1286 its own bitmap.
1288 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1289 It is created by subtracting the vertical space characters (\x09, \x0a,
1290 \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1291 subtraction was done in the overall bitmap for a character class, meaning
1292 that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1293 be recognized. This bug has been fixed.
1295 7. Patches from the folks at Google:
1297 (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1298 real life, but is still worth protecting against".
1300 (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1301 regular expressions".
1303 (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1304 have it.
1306 (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1307 "configure" and the latter not, in order to fix a problem somebody had
1308 with compiling the Arg class on HP-UX.
1310 (e) Improve the error-handling of the C++ wrapper a little bit.
1312 (f) New tests for checking recursion limiting.
1314 8. The pcre_memmove() function, which is used only if the environment does not
1315 have a standard memmove() function (and is therefore rarely compiled),
1316 contained two bugs: (a) use of int instead of size_t, and (b) it was not
1317 returning a result (though PCRE never actually uses the result).
1319 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1320 large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1321 returned instead of calling malloc() with an overflowing number that would
1322 most likely cause subsequent chaos.
1324 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1326 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1327 with this option is matched, the nmatch and pmatch options of regexec() are
1328 ignored.
1330 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1331 provided in case anyone wants to the the POSIX interface with UTF-8
1332 strings.
1334 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1335 C++ linking (needed for some HP-UX environments).
1337 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1338 (unused parameter) and in the pcre_printint() function (omitted "default"
1339 switch label when the default is to do nothing).
1341 15. Added some code to make it possible, when PCRE is compiled as a C++
1342 library, to replace subject pointers for pcre_exec() with a smart pointer
1343 class, thus making it possible to process discontinuous strings.
1345 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1346 much the same function. They were added by different people who were trying
1347 to make PCRE easy to compile on non-Unix systems. It has been suggested
1348 that PCRE_EXPORT be abolished now that there is more automatic apparatus
1349 for compiling on Windows systems. I have therefore replaced it with
1350 PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1351 defaults to "extern" for C or "extern C" for C++, which works fine on
1352 Unix-like systems. It is now possible to override the value of PCRE_DATA_
1353 SCOPE with something explicit in config.h. In addition:
1355 (a) pcreposix.h still had just "extern" instead of either of these macros;
1356 I have replaced it with PCRE_DATA_SCOPE.
1358 (b) Functions such as _pcre_xclass(), which are internal to the library,
1359 but external in the C sense, all had PCRE_EXPORT in their definitions.
1360 This is apparently wrong for the Windows case, so I have removed it.
1361 (It makes no difference on Unix-like systems.)
1363 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1364 of recursive calls to match(). This is different to MATCH_LIMIT because
1365 that limits the total number of calls to match(), not all of which increase
1366 the depth of recursion. Limiting the recursion depth limits the amount of
1367 stack (or heap if NO_RECURSE is set) that is used. The default can be set
1368 when PCRE is compiled, and changed at run time. A patch from Google adds
1369 this functionality to the C++ interface.
1371 18. Changes to the handling of Unicode character properties:
1373 (a) Updated the table to Unicode 4.1.0.
1375 (b) Recognize characters that are not in the table as "Cn" (undefined).
1377 (c) I revised the way the table is implemented to a much improved format
1378 which includes recognition of ranges. It now supports the ranges that
1379 are defined in UnicodeData.txt, and it also amalgamates other
1380 characters into ranges. This has reduced the number of entries in the
1381 table from around 16,000 to around 3,000, thus reducing its size
1382 considerably. I realized I did not need to use a tree structure after
1383 all - a binary chop search is just as efficient. Having reduced the
1384 number of entries, I extended their size from 6 bytes to 8 bytes to
1385 allow for more data.
1387 (d) Added support for Unicode script names via properties such as \p{Han}.
1389 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1390 matching that character.
1392 20. When matching a repeated Unicode property with a minimum greater than zero,
1393 (for example \pL{2,}), PCRE could look past the end of the subject if it
1394 reached it while seeking the minimum number of characters. This could
1395 happen only if some of the characters were more than one byte long, because
1396 there is a check for at least the minimum number of bytes.
1398 21. Refactored the implementation of \p and \P so as to be more general, to
1399 allow for more different types of property in future. This has changed the
1400 compiled form incompatibly. Anybody with saved compiled patterns that use
1401 \p or \P will have to recompile them.
1403 22. Added "Any" and "L&" to the supported property types.
1405 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1406 but give a compile time error if the value is greater than 0xff.
1408 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1409 accidentally not being installed or uninstalled.
1411 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1412 made were to insert the current release number. This seemed silly, because
1413 it made things harder for people building PCRE on systems that don't run
1414 "configure". I have turned pcre.h into a distributed file, no longer built
1415 by "configure", with the version identification directly included. There is
1416 no longer a pcre.h.in file.
1418 However, this change necessitated a change to the pcre-config script as
1419 well. It is built from pcre-config.in, and one of the substitutions was the
1420 release number. I have updated configure.ac so that ./configure now finds
1421 the release number by grepping pcre.h.
1423 26. Added the ability to run the tests under valgrind.
1426 Version 6.4 05-Sep-05
1427 ---------------------
1429 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1430 "--" to be printed when multiple files were scanned, even when none of the
1431 -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1432 consider it to be a bug, and have restored the previous behaviour.
1434 2. A couple of code tidies to get rid of compiler warnings.
1436 3. The pcretest program used to cheat by referring to symbols in the library
1437 whose names begin with _pcre_. These are internal symbols that are not
1438 really supposed to be visible externally, and in some environments it is
1439 possible to suppress them. The cheating is now confined to including
1440 certain files from the library's source, which is a bit cleaner.
1442 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1443 file's purpose clearer.
1445 5. Reorganized pcre_ucp_findchar().
1448 Version 6.3 15-Aug-05
1449 ---------------------
1451 1. The file libpcre.pc.in did not have general read permission in the tarball.
1453 2. There were some problems when building without C++ support:
1455 (a) If C++ support was not built, "make install" and "make test" still
1456 tried to test it.
1458 (b) There were problems when the value of CXX was explicitly set. Some
1459 changes have been made to try to fix these, and ...
1461 (c) --disable-cpp can now be used to explicitly disable C++ support.
1463 (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1464 backslash in a target when C++ was disabled. This confuses some
1465 versions of "make", apparently. Using an intermediate variable solves
1466 this. (Same for CPP_LOBJ.)
1468 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1469 (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1470 necessary on certain architectures.
1472 4. Added a setting of -export-symbols-regex to the link command to remove
1473 those symbols that are exported in the C sense, but actually are local
1474 within the library, and not documented. Their names all begin with
1475 "_pcre_". This is not a perfect job, because (a) we have to except some
1476 symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1477 available (and never for static libraries). I have made a note to try to
1478 find a way round (a) in the future.
1481 Version 6.2 01-Aug-05
1482 ---------------------
1484 1. There was no test for integer overflow of quantifier values. A construction
1485 such as {1111111111111111} would give undefined results. What is worse, if
1486 a minimum quantifier for a parenthesized subpattern overflowed and became
1487 negative, the calculation of the memory size went wrong. This could have
1488 led to memory overwriting.
1490 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1492 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1493 operating environments where this matters.
1495 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1496 PCRE options from within the C++ wrapper.
1498 5. Named capturing subpatterns were not being correctly counted when a pattern
1499 was compiled. This caused two problems: (a) If there were more than 100
1500 such subpatterns, the calculation of the memory needed for the whole
1501 compiled pattern went wrong, leading to an overflow error. (b) Numerical
1502 back references of the form \12, where the number was greater than 9, were
1503 not recognized as back references, even though there were sufficient
1504 previous subpatterns.
1506 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1507 versions of gcc, e.g. 2.95.4.
1510 Version 6.1 21-Jun-05
1511 ---------------------
1513 1. There was one reference to the variable "posix" in pcretest.c that was not
1514 surrounded by "#if !defined NOPOSIX".
1516 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1517 the cross-check on the old pcre_info() function, for the benefit of the
1518 cut-down version of PCRE that is currently imported into Exim.
1520 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1521 allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1522 patterns that start (?i) but I don't think that matters. The use of (?i) is
1523 just an example; this all applies to the other options as well.
1525 4. Since libtool seems to echo the compile commands it is issuing, the output
1526 from "make" can be reduced a bit by putting "@" in front of each libtool
1527 compile command.
1529 5. Patch from the folks at Google for configure.in to be a bit more thorough
1530 in checking for a suitable C++ installation before trying to compile the
1531 C++ stuff. This should fix a reported problem when a compiler was present,
1532 but no suitable headers.
1534 6. The man pages all had just "PCRE" as their title. I have changed them to
1535 be the relevant file name. I have also arranged that these names are
1536 retained in the file doc/pcre.txt, which is a concatenation in text format
1537 of all the man pages except the little individual ones for each function.
1539 7. The NON-UNIX-USE file had not been updated for the different set of source
1540 files that come with release 6. I also added a few comments about the C++
1541 wrapper.
1544 Version 6.0 07-Jun-05
1545 ---------------------
1547 1. Some minor internal re-organization to help with my DFA experiments.
1549 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1550 didn't matter for the library itself when fully configured, but did matter
1551 when compiling without UCP support, or within Exim, where the ucp files are
1552 not imported.
1554 3. Refactoring of the library code to split up the various functions into
1555 different source modules. The addition of the new DFA matching code (see
1556 below) to a single monolithic source would have made it really too
1557 unwieldy, quite apart from causing all the code to be include in a
1558 statically linked application, when only some functions are used. This is
1559 relevant even without the DFA addition now that patterns can be compiled in
1560 one application and matched in another.
1562 The downside of splitting up is that there have to be some external
1563 functions and data tables that are used internally in different modules of
1564 the library but which are not part of the API. These have all had their
1565 names changed to start with "_pcre_" so that they are unlikely to clash
1566 with other external names.
1568 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1569 a different (DFA) algorithm. Although it is slower than the original
1570 function, it does have some advantages for certain types of matching
1571 problem.
1573 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1574 including restarting after a partial match.
1576 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1577 defined when compiling for Windows was sent to me. I have put it into the
1578 code, though I have no means of testing or verifying it.
1580 7. Added the pcre_refcount() auxiliary function.
1582 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1583 match before or at the first newline in the subject string. In pcretest,
1584 the /f option on a pattern can be used to set this.
1586 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1587 would behave wrongly. This has been present in PCRE since release 4.0.
1589 10. A number of changes to the pcregrep command:
1591 (a) Refactored how -x works; insert ^(...)$ instead of setting
1592 PCRE_ANCHORED and checking the length, in preparation for adding
1593 something similar for -w.
1595 (b) Added the -w (match as a word) option.
1597 (c) Refactored the way lines are read and buffered so as to have more
1598 than one at a time available.
1600 (d) Implemented a pcregrep test script.
1602 (e) Added the -M (multiline match) option. This allows patterns to match
1603 over several lines of the subject. The buffering ensures that at least
1604 8K, or the rest of the document (whichever is the shorter) is available
1605 for matching (and similarly the previous 8K for lookbehind assertions).
1607 (f) Changed the --help output so that it now says
1609 -w, --word-regex(p)
1611 instead of two lines, one with "regex" and the other with "regexp"
1612 because that confused at least one person since the short forms are the
1613 same. (This required a bit of code, as the output is generated
1614 automatically from a table. It wasn't just a text change.)
1616 (g) -- can be used to terminate pcregrep options if the next thing isn't an
1617 option but starts with a hyphen. Could be a pattern or a path name
1618 starting with a hyphen, for instance.
1620 (h) "-" can be given as a file name to represent stdin.
1622 (i) When file names are being printed, "(standard input)" is used for
1623 the standard input, for compatibility with GNU grep. Previously
1624 "<stdin>" was used.
1626 (j) The option --label=xxx can be used to supply a name to be used for
1627 stdin when file names are being printed. There is no short form.
1629 (k) Re-factored the options decoding logic because we are going to add
1630 two more options that take data. Such options can now be given in four
1631 different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1633 (l) Added the -A, -B, and -C options for requesting that lines of context
1634 around matches be printed.
1636 (m) Added the -L option to print the names of files that do not contain
1637 any matching lines, that is, the complement of -l.
1639 (n) The return code is 2 if any file cannot be opened, but pcregrep does
1640 continue to scan other files.
1642 (o) The -s option was incorrectly implemented. For compatibility with other
1643 greps, it now suppresses the error message for a non-existent or non-
1644 accessible file (but not the return code). There is a new option called
1645 -q that suppresses the output of matching lines, which was what -s was
1646 previously doing.
1648 (p) Added --include and --exclude options to specify files for inclusion
1649 and exclusion when recursing.
1651 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1652 Hopefully, it now does.
1654 12. Missing cast in pcre_study().
1656 13. Added an "uninstall" target to the makefile.
1658 14. Replaced "extern" in the function prototypes in Makefile.in with
1659 "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1660 world, but is set differently for Windows.
1662 15. Added a second compiling function called pcre_compile2(). The only
1663 difference is that it has an extra argument, which is a pointer to an
1664 integer error code. When there is a compile-time failure, this is set
1665 non-zero, in addition to the error test pointer being set to point to an
1666 error message. The new argument may be NULL if no error number is required
1667 (but then you may as well call pcre_compile(), which is now just a
1668 wrapper). This facility is provided because some applications need a
1669 numeric error indication, but it has also enabled me to tidy up the way
1670 compile-time errors are handled in the POSIX wrapper.
1672 16. Added VPATH=.libs to the makefile; this should help when building with one
1673 prefix path and installing with another. (Or so I'm told by someone who
1674 knows more about this stuff than I do.)
1676 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1677 passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1678 match everything, including newlines. This is not POSIX-compatible, but
1679 somebody wanted the feature. From pcretest it can be activated by using
1680 both the P and the s flags.
1682 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1684 19. libpcre.pc was being incorrectly installed as executable.
1686 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1687 it now also looks for '\r' so that it will work unmodified on Windows.
1689 21. Added Google's contributed C++ wrapper to the distribution.
1691 22. Added some untidy missing memory free() calls in pcretest, to keep
1692 Electric Fence happy when testing.
1696 Version 5.0 13-Sep-04
1697 ---------------------
1699 1. Internal change: literal characters are no longer packed up into items
1700 containing multiple characters in a single byte-string. Each character
1701 is now matched using a separate opcode. However, there may be more than one
1702 byte in the character in UTF-8 mode.
1704 2. The pcre_callout_block structure has two new fields: pattern_position and
1705 next_item_length. These contain the offset in the pattern to the next match
1706 item, and its length, respectively.
1708 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1709 insertion of callouts before each pattern item. Added the /C option to
1710 pcretest to make use of this.
1712 4. On the advice of a Windows user, the lines
1714 #if defined(_WIN32) || defined(WIN32)
1715 _setmode( _fileno( stdout ), 0x8000 );
1716 #endif /* defined(_WIN32) || defined(WIN32) */
1718 have been added to the source of pcretest. This apparently does useful
1719 magic in relation to line terminators.
1721 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1722 for the benefit of those environments where the "b" makes a difference.
1724 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1725 to know about it. I have put a hack into configure.in that adds in code
1726 to set GCC=yes if CC=icc. This seems to end up at a point in the
1727 generated configure script that is early enough to affect the setting of
1728 compiler options, which is what is needed, but I have no means of testing
1729 whether it really works. (The user who reported this had patched the
1730 generated configure script, which of course I cannot do.)
1732 LATER: After change 22 below (new libtool files), the configure script
1733 seems to know about icc (and also ecc). Therefore, I have commented out
1734 this hack in configure.in.
1736 7. Added support for pkg-config (2 patches were sent in).
1738 8. Negated POSIX character classes that used a combination of internal tables
1739 were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1740 [[:^ascii]]. Typically, they would match almost any characters. The other
1741 POSIX classes were not broken in this way.
1743 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1744 to find the match, as PCRE was deluded into thinking that the match had to
1745 start at the start point or following a newline. The same bug applied to
1746 patterns with negative forward assertions or any backward assertions
1747 preceding ".*" at the start, unless the pattern required a fixed first
1748 character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1750 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1751 starting at the last subject character, bytes beyond the end of the subject
1752 string were read.
1754 11. Renamed the variable "class" as "classbits" to make life easier for C++
1755 users. (Previously there was a macro definition, but it apparently wasn't
1756 enough.)
1758 12. Added the new field "tables" to the extra data so that tables can be passed
1759 in at exec time, or the internal tables can be re-selected. This allows
1760 a compiled regex to be saved and re-used at a later time by a different
1761 program that might have everything at different addresses.
1763 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1764 -R library as well as a -L library.
1766 14. The debugging options of pcretest (-d on the command line or D on a
1767 pattern) showed incorrect output for anything following an extended class
1768 that contained multibyte characters and which was followed by a quantifier.
1770 15. Added optional support for general category Unicode character properties
1771 via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1772 support. It adds about 90K to the size of the library. The meanings of the
1773 inbuilt class escapes such as \d and \s have NOT been changed.
1775 16. Updated pcredemo.c to include calls to free() to release the memory for the
1776 compiled pattern.
1778 17. The generated file chartables.c was being created in the source directory
1779 instead of in the building directory. This caused the build to fail if the
1780 source directory was different from the building directory, and was
1781 read-only.
1783 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1784 file. No doubt somebody will tell me if they don't make sense... Also added
1785 Dan Mooney's comments about building on OpenVMS.
1787 19. Added support for partial matching via the PCRE_PARTIAL option for
1788 pcre_exec() and the \P data escape in pcretest.
1790 20. Extended pcretest with 3 new pattern features:
1792 (i) A pattern option of the form ">rest-of-line" causes pcretest to
1793 write the compiled pattern to the file whose name is "rest-of-line".
1794 This is a straight binary dump of the data, with the saved pointer to
1795 the character tables forced to be NULL. The study data, if any, is
1796 written too. After writing, pcretest reads a new pattern.
1798 (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1799 compiled pattern from the given file. There must not be any
1800 occurrences of "<" in the file name (pretty unlikely); if there are,
1801 pcretest will instead treat the initial "<" as a pattern delimiter.
1802 After reading in the pattern, pcretest goes on to read data lines as
1803 usual.
1805 (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1806 and 16-bit fields in a compiled pattern, to simulate a pattern that
1807 was compiled on a host of opposite endianness.
1809 21. The pcre-exec() function can now cope with patterns that were compiled on
1810 hosts of opposite endianness, with this restriction:
1812 As for any compiled expression that is saved and used later, the tables
1813 pointer field cannot be preserved; the extra_data field in the arguments
1814 to pcre_exec() should be used to pass in a tables address if a value
1815 other than the default internal tables were used at compile time.
1817 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1818 now diagnosed as an error. Previously, most of the time, a negative number
1819 would have been treated as zero, but if in addition "ovector" was passed as
1820 NULL, a crash could occur.
1822 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1823 new versions from the libtool 1.5 distribution (the last one is a copy of
1824 a file called libtool.m4). This seems to have fixed the need to patch
1825 "configure" to support Darwin 1.3 (which I used to do). However, I still
1826 had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1827 workstation).
1829 24. Changed the PCRE licence to be the more standard "BSD" licence.
1832 Version 4.5 01-Dec-03
1833 ---------------------
1835 1. There has been some re-arrangement of the code for the match() function so
1836 that it can be compiled in a version that does not call itself recursively.
1837 Instead, it keeps those local variables that need separate instances for
1838 each "recursion" in a frame on the heap, and gets/frees frames whenever it
1839 needs to "recurse". Keeping track of where control must go is done by means
1840 of setjmp/longjmp. The whole thing is implemented by a set of macros that
1841 hide most of the details from the main code, and operates only if
1842 NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1843 "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1844 operating.
1846 To make it easier for callers to provide specially tailored get/free
1847 functions for this usage, two new functions, pcre_stack_malloc, and
1848 pcre_stack_free, are used. They are always called in strict stacking order,
1849 and the size of block requested is always the same.
1851 The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1852 PCRE has been compiled to use the stack or the heap for recursion. The
1853 -C option of pcretest uses this to show which version is compiled.
1855 A new data escape \S, is added to pcretest; it causes the amounts of store
1856 obtained and freed by both kinds of malloc/free at match time to be added
1857 to the output.
1859 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1860 what's available on my current Linux desktop machine.
1862 3. When matching a UTF-8 string, the test for a valid string at the start has
1863 been extended. If start_offset is not zero, PCRE now checks that it points
1864 to a byte that is the start of a UTF-8 character. If not, it returns
1865 PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1866 this is necessary because there may be backward assertions in the pattern.
1867 When matching the same subject several times, it may save resources to use
1868 PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1870 4. The code for checking the validity of UTF-8 strings has been tightened so
1871 that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1872 containing "overlong sequences".
1874 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1875 I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1876 should have been "&", but it just so happened that all the cases this let
1877 through by mistake were picked up later in the function.
1879 6. I had used a variable called "isblank" - this is a C99 function, causing
1880 some compilers to warn. To avoid this, I renamed it (as "blankclass").
1882 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1883 prompting; (b) run "./pcretest /dev/null" at the start of the test script
1884 so the version is shown; (c) stop "make test" echoing "./RunTest".
1886 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1888 9. The prototype for memmove() for systems that don't have it was using
1889 size_t, but the inclusion of the header that defines size_t was later. I've
1890 moved the #includes for the C headers earlier to avoid this.
1892 10. Added some adjustments to the code to make it easier to compiler on certain
1893 special systems:
1895 (a) Some "const" qualifiers were missing.
1896 (b) Added the macro EXPORT before all exported functions; by default this
1897 is defined to be empty.
1898 (c) Changed the dftables auxiliary program (that builds chartables.c) so
1899 that it reads its output file name as an argument instead of writing
1900 to the standard output and assuming this can be redirected.
1902 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1903 class containing characters with values greater than 255, PCRE compilation
1904 went into a loop.
1906 12. A recursive reference to a subpattern that was within another subpattern
1907 that had a minimum quantifier of zero caused PCRE to crash. For example,
1908 (x(y(?2))z)? provoked this bug with a subject that got as far as the
1909 recursion. If the recursively-called subpattern itself had a zero repeat,
1910 that was OK.
1912 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1913 buffer into which it was copied (for escape processing) was still set at
1914 1024, so long lines caused crashes.
1916 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1917 "internal error: code overflow...". This applied to any character class
1918 that was followed by a possessive quantifier.
1920 15. Modified the Makefile to add libpcre.la as a prerequisite for
1921 libpcreposix.la because I was told this is needed for a parallel build to
1922 work.
1924 16. If a pattern that contained .* following optional items at the start was
1925 studied, the wrong optimizing data was generated, leading to matching
1926 errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1927 matching string must start with a or b or c. The correct conclusion for
1928 this pattern is that a match can start with any character.
1931 Version 4.4 13-Aug-03
1932 ---------------------
1934 1. In UTF-8 mode, a character class containing characters with values between
1935 127 and 255 was not handled correctly if the compiled pattern was studied.
1936 In fixing this, I have also improved the studying algorithm for such
1937 classes (slightly).
1939 2. Three internal functions had redundant arguments passed to them. Removal
1940 might give a very teeny performance improvement.
1942 3. Documentation bug: the value of the capture_top field in a callout is *one
1943 more than* the number of the hightest numbered captured substring.
1945 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1946 in incorrectly linking with a previously installed version. They now link
1947 explicitly with libpcre.la.
1949 5. configure.in no longer needs to recognize Cygwin specially.
1951 6. A problem in pcre.in for Windows platforms is fixed.
1953 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1954 pcretest, it used to include the size of the study block as part of its
1955 output. Unfortunately, the structure contains a field that has a different
1956 size on different hardware architectures. This meant that the tests that
1957 showed this size failed. As the block is currently always of a fixed size,
1958 this information isn't actually particularly useful in pcretest output, so
1959 I have just removed it.
1961 8. Three pre-processor statements accidentally did not start in column 1.
1962 Sadly, there are *still* compilers around that complain, even though
1963 standard C has not required this for well over a decade. Sigh.
1965 9. In pcretest, the code for checking callouts passed small integers in the
1966 callout_data field, which is a void * field. However, some picky compilers
1967 complained about the casts involved for this on 64-bit systems. Now
1968 pcretest passes the address of the small integer instead, which should get
1969 rid of the warnings.
1971 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1972 both compile and run time, and gives an error if an invalid UTF-8 sequence
1973 is found. There is a option for disabling this check in cases where the
1974 string is known to be correct and/or the maximum performance is wanted.
1976 11. In response to a bug report, I changed one line in Makefile.in from
1978 -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1979 to
1980 -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1982 to look similar to other lines, but I have no way of telling whether this
1983 is the right thing to do, as I do not use Windows. No doubt I'll get told
1984 if it's wrong...
1987 Version 4.3 21-May-03
1988 ---------------------
1990 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1991 Makefile.
1993 2. Some refactoring to improve the quality of the code:
1995 (i) The utf8_table... variables are now declared "const".
1997 (ii) The code for \cx, which used the "case flipping" table to upper case
1998 lower case letters, now just substracts 32. This is ASCII-specific,
1999 but the whole concept of \cx is ASCII-specific, so it seems
2000 reasonable.
2002 (iii) PCRE was using its character types table to recognize decimal and
2003 hexadecimal digits in the pattern. This is silly, because it handles
2004 only 0-9, a-f, and A-F, but the character types table is locale-
2005 specific, which means strange things might happen. A private
2006 table is now used for this - though it costs 256 bytes, a table is
2007 much faster than multiple explicit tests. Of course, the standard
2008 character types table is still used for matching digits in subject
2009 strings against \d.
2011 (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2012 ending in _t are). So I've renamed it as ESC_tee.
2014 3. The first argument for regexec() in the POSIX wrapper should have been
2015 defined as "const".
2017 4. Changed pcretest to use malloc() for its buffers so that they can be
2018 Electric Fenced for debugging.
2020 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2021 to read one or more bytes before the start of the subject string. Often this
2022 had no effect on PCRE's behaviour, but in some circumstances it could
2023 provoke a segmentation fault.
2025 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2026 to try to read one or more bytes before the start of the subject string.
2028 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2029 UTF-8 support could misbehave in various ways if the subject string
2030 contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2031 area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2032 back over UTF-8 characters.)
2035 Version 4.2 14-Apr-03
2036 ---------------------
2038 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2040 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2041 [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2042 [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2043 [WIN_PREFIX]: new variable, "cyg" for Cygwin
2044 * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2046 Note: automatic setting of the BUILD variables is not yet working
2047 set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2048 compile-time but not at link-time
2049 [LINK]: use for linking executables only
2050 make different versions for Windows and non-Windows
2051 [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2052 libraries
2053 [LINK_FOR_BUILD]: new variable
2054 [OBJEXT]: use throughout
2055 [EXEEXT]: use throughout
2056 <winshared>: new target
2057 <wininstall>: new target
2058 <dftables.o>: use native compiler
2059 <dftables>: use native linker
2060 <install>: handle Windows platform correctly
2061 <clean>: ditto
2062 <check>: ditto
2063 copy DLL to top builddir before testing
2065 As part of these changes, -no-undefined was removed again. This was reported
2066 to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2067 in any case.
2069 3. Some tidies to get rid of compiler warnings:
2071 . In the match_data structure, match_limit was an unsigned long int, whereas
2072 match_call_count was an int. I've made them both unsigned long ints.
2074 . In pcretest the fact that a const uschar * doesn't automatically cast to
2075 a void * provoked a warning.
2077 . Turning on some more compiler warnings threw up some "shadow" variables
2078 and a few more missing casts.
2080 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2081 option, a class that contained a single character with a value between 128
2082 and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2084 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2085 option, a class that contained several characters, but with at least one
2086 whose value was between 128 and 255 caused PCRE to crash.
2089 Version 4.1 12-Mar-03
2090 ---------------------
2092 1. Compiling with gcc -pedantic found a couple of places where casts were
2093 needed, and a string in dftables.c that was longer than standard compilers are
2094 required to support.
2096 2. Compiling with Sun's compiler found a few more places where the code could
2097 be tidied up in order to avoid warnings.
2099 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2100 first of these names is deprecated in the latest Autoconf in favour of the name
2101 CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2102 compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2103 analogy I have changed it to CFLAGS_FOR_BUILD.
2105 4. Added -no-undefined to the linking command in the Makefile, because this is
2106 apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2107 linking step for the pcreposix library.
2109 5. PCRE was failing to diagnose the case of two named groups with the same
2110 name.
2112 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2113 literal character that is needed in the subject for a match, and scans along to
2114 ensure that it is present before embarking on the full matching process. This
2115 saves time in cases of nested unlimited repeats that are never going to match.
2116 Problem: the scan can take a lot of time if the subject is very long (e.g.
2117 megabytes), thus penalizing straightforward matches. It is now done only if the
2118 amount of subject to be scanned is less than 1000 bytes.
2120 7. A lesser problem with the same optimization is that it was recording the
2121 first character of an anchored pattern as "needed", thus provoking a search
2122 right along the subject, even when the first match of the pattern was going to
2123 fail. The "needed" character is now not set for anchored patterns, unless it
2124 follows something in the pattern that is of non-fixed length. Thus, it still
2125 fulfils its original purpose of finding quick non-matches in cases of nested
2126 unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2129 Version 4.0 17-Feb-03
2130 ---------------------
2132 1. If a comment in an extended regex that started immediately after a meta-item
2133 extended to the end of string, PCRE compiled incorrect data. This could lead to
2134 all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2136 2. Moved to autoconf 2.53 and libtool 1.4.2.
2138 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2139 the special perltest8 script is no longer needed - all the tests can be run
2140 from a single perltest script.
2142 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2143 by \s. It has now been removed in PCRE. This means it isn't recognized as
2144 whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2145 class [:space:] *does* include VT, thereby creating a mess.
2147 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2148 space and tab.
2150 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2151 its new features into the main test script, reducing the number of scripts.
2153 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2154 were backward compatible, and made the (?i) apply to the whole pattern, as if
2155 /i were given. Now it behaves more logically, and applies the option setting
2156 only to what follows. PCRE has been changed to follow suit. However, if it
2157 finds options settings right at the start of the pattern, it extracts them into
2158 the global options, as before. Thus, they show up in the info data.
2160 8. Added support for the \Q...\E escape sequence. Characters in between are
2161 treated as literals. This is slightly different from Perl in that $ and @ are
2162 also handled as literals inside the quotes. In Perl, they will cause variable
2163 interpolation. Note the following examples:
2165 Pattern PCRE matches Perl matches
2167 \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2168 \Qabc\$xyz\E abc\$xyz abc\$xyz
2169 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2171 For compatibility with Perl, \Q...\E sequences are recognized inside character
2172 classes as well as outside them.
2174 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2175 floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2176 (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2177 signed/unsigned warnings.
2179 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2180 option for pcretest, so I've replaced it by a simple function that does just
2181 that job.
2183 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2184 "pcregrep -".
2186 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2187 Java package. This provides some syntactic sugar for simple cases of what my
2188 documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2189 as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2190 item, you can use this simplified notation. Note that only makes sense with
2191 greedy quantifiers. Consequently, the use of the possessive quantifier forces
2192 greediness, whatever the setting of the PCRE_UNGREEDY option.
2194 13. A change of greediness default within a pattern was not taking effect at
2195 the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2196 subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2197 was abstracted outside.
2199 14. PCRE now supports the \G assertion. It is true when the current matching
2200 position is at the start point of the match. This differs from \A when the
2201 starting offset is non-zero. Used with the /g option of pcretest (or similar
2202 code), it works in the same way as it does for Perl's /g option. If all
2203 alternatives of a regex begin with \G, the expression is anchored to the start
2204 match position, and the "anchored" flag is set in the compiled expression.
2206 15. Some bugs concerning the handling of certain option changes within patterns
2207 have been fixed. These applied to options other than (?ims). For example,
2208 "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2209 been the other way round. Some of this was related to change 7 above.
2211 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2212 features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2213 and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2214 POSIX classes only within a class (e.g. /[[:alpha:]]/).
2216 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2217 mode. Unlike ".", it always matches newline, whatever the setting of
2218 PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2219 assertions. Perl allows it, but it doesn't (in general) work because it can't
2220 calculate the length of the lookbehind. At least, that's the case for Perl
2221 5.8.0 - I've been told they are going to document that it doesn't work in
2222 future.
2224 18. Added an error diagnosis for escapes that PCRE does not support: these are
2225 \L, \l, \N, \P, \p, \U, \u, and \X.
2227 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2228 reading past the end of the pattern in cases such as /[abcd/.
2230 20. PCRE was getting more memory than necessary for patterns with classes that
2231 contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2233 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2234 compiling PCRE for use with Virtual Pascal.
2236 22. Small fix to the Makefile to make it work properly if the build is done
2237 outside the source tree.
2239 23. Added a new extension: a condition to go with recursion. If a conditional
2240 subpattern starts with (?(R) the "true" branch is used if recursion has
2241 happened, whereas the "false" branch is used only at the top level.
2243 24. When there was a very long string of literal characters (over 255 bytes
2244 without UTF support, over 250 bytes with UTF support), the computation of how
2245 much memory was required could be incorrect, leading to segfaults or other
2246 strange effects.
2248 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2249 start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2250 there was a subsequent back reference to those brackets. This meant that, for
2251 example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2252 possible to check for precisely this case. All we can do is abandon the
2253 optimization if .* occurs inside capturing brackets when there are any back
2254 references whatsoever. (See below for a better fix that came later.)
2256 26. The handling of the optimization for finding the first character of a
2257 non-anchored pattern, and for finding a character that is required later in the
2258 match were failing in some cases. This didn't break the matching; it just
2259 failed to optimize when it could. The way this is done has been re-implemented.
2261 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2263 28. Added a new feature that provides some of the functionality that Perl
2264 provides with (?{...}). The facility is termed a "callout". The way it is done
2265 in PCRE is for the caller to provide an optional function, by setting
2266 pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2267 global variable. By default it is unset, which disables all calling out. To get
2268 the function called, the regex must include (?C) at appropriate points. This
2269 is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2270 This provides a means of identifying different callout points. When PCRE
2271 reaches such a point in the regex, if pcre_callout has been set, the external
2272 function is called. It is provided with data in a structure called
2273 pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2274 matching continues; if it returns a non-zero value, the match at the current
2275 point fails. However, backtracking will occur if possible. [This was changed
2276 later and other features added - see item 49 below.]
2278 29. pcretest is upgraded to test the callout functionality. It provides a
2279 callout function that displays information. By default, it shows the start of
2280 the match and the current position in the text. There are some new data escapes
2281 to vary what happens:
2283 \C+ in addition, show current contents of captured substrings
2284 \C- do not supply a callout function
2285 \C!n return 1 when callout number n is reached
2286 \C!n!m return 1 when callout number n is reached for the mth time
2288 30. If pcregrep was called with the -l option and just a single file name, it
2289 output "<stdin>" if a match was found, instead of the file name.
2291 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2292 slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2293 pcre_exec(). This saves a malloc/free per call. The default value of
2294 POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2295 when configuring.
2297 32. The default maximum size of a compiled pattern is 64K. There have been a
2298 few cases of people hitting this limit. The code now uses macros to handle the
2299 storing of links as offsets within the compiled pattern. It defaults to 2-byte
2300 links, but this can be changed to 3 or 4 bytes by --with-link-size when
2301 configuring. Tests 2 and 5 work only with 2-byte links because they output
2302 debugging information about compiled patterns.
2304 33. Internal code re-arrangements:
2306 (a) Moved the debugging function for printing out a compiled regex into
2307 its own source file (printint.c) and used #include to pull it into
2308 pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2309 separate copies.
2311 (b) Defined the list of op-code names for debugging as a macro in
2312 internal.h so that it is next to the definition of the opcodes.
2314 (c) Defined a table of op-code lengths for simpler skipping along compiled
2315 code. This is again a macro in internal.h so that it is next to the
2316 definition of the opcodes.
2318 34. Added support for recursive calls to individual subpatterns, along the
2319 lines of Robin Houston's patch (but implemented somewhat differently).
2321 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2322 allow it to read and process whole directories in Win32. This code was
2323 contributed by Lionel Fourquaux; it has not been tested by me.
2325 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2326 used to name a group. Names consist of alphanumerics and underscores, and must
2327 be unique. Back references use the syntax (?P=name) and recursive calls use
2328 (?P>name) which is a PCRE extension to the Python extension. Groups still have
2329 numbers. The function pcre_fullinfo() can be used after compilation to extract
2330 a name/number map. There are three relevant calls:
2332 PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2333 PCRE_INFO_NAMECOUNT yields the number of entries
2334 PCRE_INFO_NAMETABLE yields a pointer to the map.
2336 The map is a vector of fixed-size entries. The size of each entry depends on
2337 the length of the longest name used. The first two bytes of each entry are the
2338 group number, most significant byte first. There follows the corresponding
2339 name, zero terminated. The names are in alphabetical order.
2341 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2342 case instead of 255. Making it the same both with and without UTF-8 support
2343 means that the same test output works with both.
2345 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2346 calling malloc() with a zero argument.
2348 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2349 optimization. I've improved things by keeping a bitmap of backreferences with
2350 numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2351 fact referenced, the optimization can be applied. It is unlikely that a
2352 relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2353 the match to follow \n) will appear inside brackets with a number greater than
2354 31, but if it does, any back reference > 31 suppresses the optimization.
2356 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2357 of disabling numbered capturing parentheses. Any opening parenthesis that is
2358 not followed by ? behaves as if it were followed by ?: but named parentheses
2359 can still be used for capturing (and they will acquire numbers in the usual
2360 way).
2362 41. Redesigned the return codes from the match() function into yes/no/error so
2363 that errors can be passed back from deep inside the nested calls. A malloc
2364 failure while inside a recursive subpattern call now causes the
2365 PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2367 42. It is now possible to set a limit on the number of times the match()
2368 function is called in a call to pcre_exec(). This facility makes it possible to
2369 limit the amount of recursion and backtracking, though not in a directly
2370 obvious way, because the match() function is used in a number of different
2371 circumstances. The count starts from zero for each position in the subject
2372 string (for non-anchored patterns). The default limit is, for compatibility, a
2373 large number, namely 10 000 000. You can change this in two ways:
2375 (a) When configuring PCRE before making, you can use --with-match-limit=n
2376 to set a default value for the compiled library.
2378 (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2379 a different value is set. See 45 below.
2381 If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2383 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2384 of things that can be changed at compile time. The first argument specifies
2385 what is wanted and the second points to where the information is to be placed.
2386 The current list of available information is:
2390 The output is an integer that is set to one if UTF-8 support is available;
2391 otherwise it is set to zero.
2395 The output is an integer that it set to the value of the code that is used for
2396 newline. It is either LF (10) or CR (13).
2400 The output is an integer that contains the number of bytes used for internal
2401 linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2405 The output is an integer that contains the threshold above which the POSIX
2406 interface uses malloc() for output vectors. See item 31 above.
2410 The output is an unsigned integer that contains the default limit of the number
2411 of match() calls in a pcre_exec() execution. See 42 above.
2413 44. pcretest has been upgraded by the addition of the -C option. This causes it
2414 to extract all the available output from the new pcre_config() function, and to
2415 output it. The program then exits immediately.
2417 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2418 order to support additional features. One way would have been to define
2419 pcre_exec2() (for example) with extra arguments, but this would not have been
2420 extensible, and would also have required all calls to the original function to
2421 be mapped to the new one. Instead, I have chosen to extend the mechanism that
2422 is used for passing in "extra" data from pcre_study().
2424 The pcre_extra structure is now exposed and defined in pcre.h. It currently
2425 contains the following fields:
2427 flags a bitmap indicating which of the following fields are set
2428 study_data opaque data from pcre_study()
2429 match_limit a way of specifying a limit on match() calls for a specific
2430 call to pcre_exec()
2431 callout_data data for callouts (see 49 below)
2433 The flag bits are also defined in pcre.h, and are
2439 The pcre_study() function now returns one of these new pcre_extra blocks, with
2440 the actual study data pointed to by the study_data field, and the
2441 PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2442 before. That is, this change is entirely upwards-compatible and requires no
2443 change to existing code.
2445 If you want to pass in additional data to pcre_exec(), you can either place it
2446 in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2447 block.
2449 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2450 data string contains the escape sequence \M, pcretest calls pcre_exec() several
2451 times with different match limits, until it finds the minimum value needed for
2452 pcre_exec() to complete. The value is then output. This can be instructive; for
2453 most simple matches the number is quite small, but for pathological cases it
2454 gets very large very quickly.
2456 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2457 returns the size of the data block pointed to by the study_data field in a
2458 pcre_extra block, that is, the value that was passed as the argument to
2459 pcre_malloc() when PCRE was getting memory in which to place the information
2460 created by pcre_study(). The fourth argument should point to a size_t variable.
2461 pcretest has been extended so that this information is shown after a successful
2462 pcre_study() call when information about the compiled regex is being displayed.
2464 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2465 because what follows is always an absolute path. (Later: it turns out that this
2466 is more than cosmetic for MinGW, because it doesn't like empty path
2467 components.)
2469 49. Some changes have been made to the callout feature (see 28 above):
2471 (i) A callout function now has three choices for what it returns:
2473 0 => success, carry on matching
2474 > 0 => failure at this point, but backtrack if possible
2475 < 0 => serious error, return this value from pcre_exec()
2477 Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2478 values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2479 "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2480 use by callout functions. It will never be used by PCRE itself.
2482 (ii) The pcre_extra structure (see 45 above) has a void * field called
2483 callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2484 pcre_callout_block structure has a field of the same name. The contents of
2485 the field passed in the pcre_extra structure are passed to the callout
2486 function in the corresponding field in the callout block. This makes it
2487 easier to use the same callout-containing regex from multiple threads. For
2488 testing, the pcretest program has a new data escape
2490 \C*n pass the number n (may be negative) as callout_data
2492 If the callout function in pcretest receives a non-zero value as
2493 callout_data, it returns that value.
2495 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2496 there were some redundant $(CFLAGS) in commands that are now specified as
2497 $(LINK), which already includes $(CFLAGS).
2499 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2500 has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2501 with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2502 one-byte characters throughout. Note that case-insensitive matching applies
2503 only to characters whose values are less than 256. PCRE doesn't support the
2504 notion of cases for higher-valued characters.
2506 (i) A character class whose characters are all within 0-255 is handled as
2507 a bit map, and the map is inverted for negative classes. Previously, a
2508 character > 255 always failed to match such a class; however it should
2509 match if the class was a negative one (e.g. [^ab]). This has been fixed.
2511 (ii) A negated character class with a single character < 255 is coded as
2512 "not this character" (OP_NOT). This wasn't working properly when the test
2513 character was multibyte, either singly or repeated.
2515 (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2516 mode, for example: \x{100}{2,3}.
2518 (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2519 singly or repeated) now correctly test multibyte characters. However,
2520 PCRE doesn't recognize any characters with values greater than 255 as
2521 digits, spaces, or word characters. Such characters always match \D, \S,
2522 and \W, and never match \d, \s, or \w.
2524 (v) Classes may now contain characters and character ranges with values
2525 greater than 255. For example: [ab\x{100}-\x{400}].
2527 (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2528 PCRE in UTF-8 mode.
2530 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2531 PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2532 retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2533 value.)
2535 53. The single man page has become too large. I have therefore split it up into
2536 a number of separate man pages. These also give rise to individual HTML pages;
2537 these are now put in a separate directory, and there is an index.html page that
2538 lists them all. Some hyperlinking between the pages has been installed.
2540 54. Added convenience functions for handling named capturing parentheses.
2542 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2543 aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2544 true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2545 are faulted.
2547 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2548 calling configure. These values are used when compiling the dftables.c program
2549 which is run to generate the source of the default character tables. They
2550 default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2551 you will need to set these values.
2553 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2556 Version 3.9 02-Jan-02
2557 ---------------------
2559 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2561 2. If --disable-static was given, the building process failed when trying to
2562 build pcretest and pcregrep. (For some reason it was using libtool to compile
2563 them, which is not right, as they aren't part of the library.)
2566 Version 3.8 18-Dec-01
2567 ---------------------
2569 1. The experimental UTF-8 code was completely screwed up. It was packing the
2570 bytes in the wrong order. How dumb can you get?
2573 Version 3.7 29-Oct-01
2574 ---------------------
2576 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2577 This caused pcretest, when used on the test data, to segfault. Unfortunately,
2578 this didn't happen under Solaris 8, where I normally test things.
2580 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2581 doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2582 isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2583 this fix an hour or so after the initial 3.7 release.)
2586 Version 3.6 23-Oct-01
2587 ---------------------
2589 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2590 offsets passed as NULL with zero offset count.
2592 2. The config.guess and config.sub files had not been updated when I moved to
2593 the latest autoconf.
2596 Version 3.5 15-Aug-01
2597 ---------------------
2599 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2600 had been forgotten.
2602 2. By using declared but undefined structures, we can avoid using "void"
2603 definitions in pcre.h while keeping the internal definitions of the structures
2604 private.
2606 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2607 user point of view, this means that both static and shared libraries are built
2608 by default, but this can be individually controlled. More of the work of
2609 handling this static/shared cases is now inside libtool instead of PCRE's make
2610 file.
2612 4. The pcretest utility is now installed along with pcregrep because it is
2613 useful for users (to test regexs) and by doing this, it automatically gets
2614 relinked by libtool. The documentation has been turned into a man page, so
2615 there are now .1, .txt, and .html versions in /doc.
2617 5. Upgrades to pcregrep:
2618 (i) Added long-form option names like gnu grep.
2619 (ii) Added --help to list all options with an explanatory phrase.
2620 (iii) Added -r, --recursive to recurse into sub-directories.
2621 (iv) Added -f, --file to read patterns from a file.
2623 6. pcre_exec() was referring to its "code" argument before testing that
2624 argument for NULL (and giving an error if it was NULL).
2626 7. Upgraded Makefile.in to allow for compiling in a different directory from
2627 the source directory.
2629 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2630 options bits, the pointer it was passed was to an int instead of to an unsigned
2631 long int. This mattered only on 64-bit systems.
2633 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2634 generated) instead of pcre.in, which it its source. Also made the same change
2635 in several of the .c files.
2637 10. A new release of gcc defines printf() as a macro, which broke pcretest
2638 because it had an ifdef in the middle of a string argument for printf(). Fixed
2639 by using separate calls to printf().
2641 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2642 script, to force use of CR or LF instead of \n in the source. On non-Unix
2643 systems, the value can be set in config.h.
2645 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2646 absolute limit. Changed the text of the error message to make this clear, and
2647 likewise updated the man page.
2649 13. The limit of 99 on the number of capturing subpatterns has been removed.
2650 The new limit is 65535, which I hope will not be a "real" limit.
2653 Version 3.4 22-Aug-00
2654 ---------------------
2656 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2658 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2661 Version 3.3 01-Aug-00
2662 ---------------------
2664 1. If an octal character was given, but the value was greater than \377, it
2665 was not getting masked to the least significant bits, as documented. This could
2666 lead to crashes in some systems.
2668 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2669 the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2671 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2672 These just pass their arguments on to (pcre_free)(), but they are provided
2673 because some uses of PCRE bind it to non-C systems that can call its functions,
2674 but cannot call free() or pcre_free() directly.
2676 4. Add "make test" as a synonym for "make check". Corrected some comments in
2677 the Makefile.
2679 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2680 Makefile.
2682 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2683 command called pgrep for grepping around the active processes.
2685 7. Added the beginnings of support for UTF-8 character strings.
2687 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2688 RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2689 the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2690 out for the ar command.)
2693 Version 3.2 12-May-00
2694 ---------------------
2696 This is purely a bug fixing release.
2698 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2699 of ZA. This was just one example of several cases that could provoke this bug,
2700 which was introduced by change 9 of version 2.00. The code for breaking
2701 infinite loops after an iteration that matches an empty string was't working
2702 correctly.
2704 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2705 when matched against abbab (for example). After matching an empty string, it
2706 wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2707 caused it to match further down the string than it should.
2709 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2710 was there because it doesn't seem to be needed, and it causes trouble on some
2711 systems, as it is not a Standard C header. It has been removed.
2713 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2714 were reported on the Macintosh. The changes were from
2716 while ((c = *(++ptr)) != 0 && c != '\n');
2717 to
2718 while ((c = *(++ptr)) != 0 && c != '\n') ;
2720 Totally extraordinary, but if that's what it takes...
2722 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2723 available. Added HAVE_BCOPY and an autoconf test for it; if neither
2724 HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2725 assumes the way PCRE uses memmove() (always moving upwards).
2727 6. PCRE is being used in one environment where strchr() is not available. There
2728 was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2729 faster code anyway.
2732 Version 3.1 09-Feb-00
2733 ---------------------
2735 The only change in this release is the fixing of some bugs in Makefile.in for
2736 the "install" target:
2738 (1) It was failing to install pcreposix.h.
2740 (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2743 Version 3.0 01-Feb-00
2744 ---------------------
2746 1. Add support for the /+ modifier to perltest (to output $` like it does in
2747 pcretest).
2749 2. Add support for the /g modifier to perltest.
2751 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2752 matches null strings.
2754 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2755 pattern. Perl treats empty patterns specially - it reuses the most recent
2756 pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2757 effect.
2759 5. The POSIX interface was broken in that it was just handing over the POSIX
2760 captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2761 required a bigger vector, with some working space on the end. This means that
2762 the POSIX wrapper now has to get and free some memory, and copy the results.
2764 6. Added some simple autoconf support, placing the test data and the
2765 documentation in separate directories, re-organizing some of the
2766 information files, and making it build pcre-config (a GNU standard). Also added
2767 libtool support for building PCRE as a shared library, which is now the
2768 default.
2770 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2771 09 are not valid octal constants. Single digits will be used for minor values
2772 less than 10.
2774 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2775 existing programs that set these in the POSIX interface can use PCRE without
2776 modification.
2778 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2779 return all that pcre_info() returns, plus additional data. The pcre_info()
2780 function is retained for compatibility, but is considered to be obsolete.
2782 10. Added experimental recursion feature (?R) to handle one common case that
2783 Perl 5.6 will be able to do with (?p{...}).
2785 11. Added support for POSIX character classes like [:alpha:], which Perl is
2786 adopting.
2789 Version 2.08 31-Aug-99
2790 ----------------------
2792 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2793 trying to match at the startoffset position, but instead was moving forward to
2794 the next newline as if a previous match had failed.
2796 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2797 and could get into a loop if a null string was matched other than at the start
2798 of the subject.
2800 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2801 be distinguished at compile time, and for completeness also added PCRE_DATE.
2803 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2804 in GnuWin32 environments.
2807 Version 2.07 29-Jul-99
2808 ----------------------
2810 1. The documentation is now supplied in plain text form and HTML as well as in
2811 the form of man page sources.
2813 2. C++ compilers don't like assigning (void *) values to other pointer types.
2814 In particular this affects malloc(). Although there is no problem in Standard
2815 C, I've put in casts to keep C++ compilers happy.
2817 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2818 should be (const char *).
2820 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2821 be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2822 However, I haven't made this a standard facility. The documentation doesn't
2823 mention it, and the Makefile doesn't support it.
2825 5. The Makefile now contains an "install" target, with editable destinations at
2826 the top of the file. The pcretest program is not installed.
2828 6. pgrep -V now gives the PCRE version number and date.
2830 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2831 causing the entire string to be ignored, instead of just the last character.
2833 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2834 non-matching string, it can take a very, very long time, even for strings of
2835 quite modest length, because of the nested recursion. PCRE now does better in
2836 some of these cases. It does this by remembering the last required literal
2837 character in the pattern, and pre-searching the subject to ensure it is present
2838 before running the real match. In other words, it applies a heuristic to detect
2839 some types of certain failure quickly, and in the above example, if presented
2840 with a string that has no trailing " it gives "no match" very quickly.
2842 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2843 other alternatives are tried instead.
2846 Version 2.06 09-Jun-99
2847 ----------------------
2849 1. Change pcretest's output for amount of store used to show just the code
2850 space, because the remainder (the data block) varies in size between 32-bit and
2851 64-bit systems.
2853 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2854 start matching at. This allows lookbehinds to work when searching for multiple
2855 occurrences in a string.
2857 3. Added additional options to pcretest for testing multiple occurrences:
2859 /+ outputs the rest of the string that follows a match
2860 /g loops for multiple occurrences, using the new startoffset argument
2861 /G loops for multiple occurrences by passing an incremented pointer
2863 4. PCRE wasn't doing the "first character" optimization for patterns starting
2864 with \b or \B, though it was doing it for other lookbehind assertions. That is,
2865 it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2866 the letter 'x'. On long subject strings, this gives a significant speed-up.
2869 Version 2.05 21-Apr-99
2870 ----------------------
2872 1. Changed the type of magic_number from int to long int so that it works
2873 properly on 16-bit systems.
2875 2. Fixed a bug which caused patterns starting with .* not to work correctly
2876 when the subject string contained newline characters. PCRE was assuming
2877 anchoring for such patterns in all cases, which is not correct because .* will
2878 not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2879 DOTALL is set at top level; otherwise it knows that patterns starting with .*
2880 must be retried after every newline in the subject.
2883 Version 2.04 18-Feb-99
2884 ----------------------
2886 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2887 computation of the store needed to hold the pattern was incorrect (too large).
2888 If such patterns were nested a few deep, this could multiply and become a real
2889 problem.
2891 2. Added /M option to pcretest to show the memory requirement of a specific
2892 pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2894 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2895 compiled in such a way that the backtracking after subsequent failure was
2896 pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2897 ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2900 Version 2.03 02-Feb-99
2901 ----------------------
2903 1. Fixed typo and small mistake in man page.
2905 2. Added 4th condition (GPL supersedes if conflict) and created separate
2906 LICENCE file containing the conditions.
2908 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2909 Perl, that is the internal \ allows the delimiter to be included in the
2910 pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2911 the final delimiter, add \ to the end of the pattern (to test the error).
2913 4. Added the convenience functions for extracting substrings after a successful
2914 match. Updated pcretest to make it able to test these functions.
2917 Version 2.02 14-Jan-99
2918 ----------------------
2920 1. Initialized the working variables associated with each extraction so that
2921 their saving and restoring doesn't refer to uninitialized store.
2923 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2924 compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2925 fix the problem.
2927 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2928 calls, and wasn't printing the correct value for compiling calls. Increased the
2929 default value of LOOPREPEAT, and the number of significant figures in the
2930 times.
2932 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2934 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2935 a building problem on Windows NT with a FAT file system.
2938 Version 2.01 21-Oct-98
2939 ----------------------
2941 1. Changed the API for pcre_compile() to allow for the provision of a pointer
2942 to character tables built by pcre_maketables() in the current locale. If NULL
2943 is passed, the default tables are used.
2946 Version 2.00 24-Sep-98
2947 ----------------------
2949 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
2950 it any more.
2952 2. Allow quantification of (?>) groups, and make it work correctly.
2954 3. The first character computation wasn't working for (?>) groups.
2956 4. Correct the implementation of \Z (it is permitted to match on the \n at the
2957 end of the subject) and add 5.005's \z, which really does match only at the
2958 very end of the subject.
2960 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
2962 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
2963 DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
2964 localized options. All options to pcre_study() were also removed.
2966 7. Add other new features from 5.005:
2968 $(?<= positive lookbehind
2969 $(?<! negative lookbehind
2970 (?imsx-imsx) added the unsetting capability
2971 such a setting is global if at outer level; local otherwise
2972 (?imsx-imsx:) non-capturing groups with option setting
2973 (?(cond)re|re) conditional pattern matching
2975 A backreference to itself in a repeated group matches the previous
2976 captured string.
2978 8. General tidying up of studying (both automatic and via "study")
2979 consequential on the addition of new assertions.
2981 9. As in 5.005, unlimited repeated groups that could match an empty substring
2982 are no longer faulted at compile time. Instead, the loop is forcibly broken at
2983 runtime if any iteration does actually match an empty substring.
2985 10. Include the RunTest script in the distribution.
2987 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
2988 discrepancies, some of which were old and were also with respect to 5.004. They
2989 have now been fixed.
2992 Version 1.09 28-Apr-98
2993 ----------------------
2995 1. A negated single character class followed by a quantifier with a minimum
2996 value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
2997 program crashes, or just wrong answers. This did not apply to negated classes
2998 containing more than one character, or to minima other than one.
3001 Version 1.08 27-Mar-98
3002 ----------------------
3004 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3006 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3007 latter must appear before anything that relies on it in the pattern.
3010 Version 1.07 16-Feb-98
3011 ----------------------
3013 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3014 repeat of a potentially empty string).
3017 Version 1.06 23-Jan-98
3018 ----------------------
3020 1. Added Markus Oberhumer's little patches for C++.
3022 2. Literal strings longer than 255 characters were broken.
3025 Version 1.05 23-Dec-97
3026 ----------------------
3028 1. Negated character classes containing more than one character were failing if
3029 PCRE_CASELESS was set at run time.
3032 Version 1.04 19-Dec-97
3033 ----------------------
3035 1. Corrected the man page, where some "const" qualifiers had been omitted.
3037 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3038 input syntax.
3040 3. Fixed memory leak which occurred when a regex with back references was
3041 matched with an offsets vector that wasn't big enough. The temporary memory
3042 that is used in this case wasn't being freed if the match failed.
3044 4. Tidied pcretest to ensure it frees memory that it gets.
3046 5. Temporary memory was being obtained in the case where the passed offsets
3047 vector was exactly big enough.
3049 6. Corrected definition of offsetof() from change 5 below.
3051 7. I had screwed up change 6 below and broken the rules for the use of
3052 setjmp(). Now fixed.
3055 Version 1.03 18-Dec-97
3056 ----------------------
3058 1. A erroneous regex with a missing opening parenthesis was correctly
3059 diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3060 on some systems.
3062 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3063 it was reported that one broken compiler failed on the former because "code" is
3064 also an independent variable.
3066 3. The erroneous regex a[]b caused an array overrun reference.
3068 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3069 fail on data ending with that character. (It was going on too far, and checking
3070 the next character, typically a binary zero.) This was specific to the
3071 optimized code for single-character negative classes.
3073 5. Added a contributed patch from the TIN world which does the following:
3075 + Add an undef for memmove, in case the the system defines a macro for it.
3077 + Add a definition of offsetof(), in case there isn't one. (I don't know
3078 the reason behind this - offsetof() is part of the ANSI standard - but
3079 it does no harm).
3081 + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3082 most of the places where whitespace preceded '#'. I have given up and
3083 allowed the remaining 2 cases to be at the margin.
3085 + Rename some variables in pcre to eliminate shadowing. This seems very
3086 pedantic, but does no harm, of course.
3088 6. Moved the call to setjmp() into its own function, to get rid of warnings
3089 from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3091 7. Constructs such as \d{8,} were compiling into the equivalent of
3092 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3093 outcome, but in this particular case used more store than had been allocated,
3094 which caused the bug to be discovered because it threw up an internal error.
3096 8. The debugging code in both pcre and pcretest for outputting the compiled
3097 form of a regex was going wrong in the case of back references followed by
3098 curly-bracketed repeats.
3101 Version 1.02 12-Dec-97
3102 ----------------------
3104 1. Typos in pcre.3 and comments in the source fixed.
3106 2. Applied a contributed patch to get rid of places where it used to remove
3107 'const' from variables, and fixed some signed/unsigned and uninitialized
3108 variable warnings.
3110 3. Added the "runtest" target to Makefile.
3112 4. Set default compiler flag to -O2 rather than just -O.
3115 Version 1.01 19-Nov-97
3116 ----------------------
3118 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3119 like /([ab]*)*/, that is, for classes with more than one character in them.
3121 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3122 as /((?>a*))*/ (a PCRE_EXTRA facility).
3125 Version 1.00 18-Nov-97
3126 ----------------------
3128 1. Added compile-time macros to support systems such as SunOS4 which don't have
3129 memmove() or strerror() but have other things that can be used instead.
3131 2. Arranged that "make clean" removes the executables.
3134 Version 0.99 27-Oct-97
3135 ----------------------
3137 1. Fixed bug in code for optimizing classes with only one character. It was
3138 initializing a 32-byte map regardless, which could cause it to run off the end
3139 of the memory it had got.
3141 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3144 Version 0.98 22-Oct-97
3145 ----------------------
3147 1. Fixed bug in code for handling temporary memory usage when there are more
3148 back references than supplied space in the ovector. This could cause segfaults.
3151 Version 0.97 21-Oct-97
3152 ----------------------
3154 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3156 2. Optimized negated single characters not to use a bit map.
3158 3. Brought error texts together as macro definitions; clarified some of them;
3159 fixed one that was wrong - it said "range out of order" when it meant "invalid
3160 escape sequence".
3162 4. Changed some char * arguments to const char *.
3164 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3166 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3167 pcretest.
3170 Version 0.96 16-Oct-97
3171 ----------------------
3173 1. Added a simple "pgrep" utility to the distribution.
3175 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3176 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3177 where "ddd" means "one or more decimal digits".
3179 3. Fixed serious bug. If a pattern had a back reference, but the call to
3180 pcre_exec() didn't supply a large enough ovector to record the related
3181 identifying subpattern, the match always failed. PCRE now remembers the number
3182 of the largest back reference, and gets some temporary memory in which to save
3183 the offsets during matching if necessary, in order to ensure that
3184 backreferences always work.
3186 4. Increased the compatibility with Perl in a number of ways:
3188 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3189 to request this handling. The option can be set at compile or exec time.
3191 (b) $ matches before a terminating newline by default; an option
3192 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3193 mode). The option can be set at compile or exec time.
3195 (c) The handling of \ followed by a digit other than 0 is now supposed to be
3196 the same as Perl's. If the decimal number it represents is less than 10
3197 or there aren't that many previous left capturing parentheses, an octal
3198 escape is read. Inside a character class, it's always an octal escape,
3199 even if it is a single digit.
3201 (d) An escaped but undefined alphabetic character is taken as a literal,
3202 unless PCRE_EXTRA is set. Currently this just reserves the remaining
3203 escapes.
3205 (e) {0} is now permitted. (The previous item is removed from the compiled
3206 pattern).
3208 5. Changed all the names of code files so that the basic parts are no longer
3209 than 10 characters, and abolished the teeny "globals.c" file.
3211 6. Changed the handling of character classes; they are now done with a 32-byte
3212 bit map always.
3214 7. Added the -d and /D options to pcretest to make it possible to look at the
3215 internals of compilation without having to recompile pcre.
3218 Version 0.95 23-Sep-97
3219 ----------------------
3221 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3222 \x20 at the start of a run of normal characters. These were being treated as
3223 real characters, instead of the source characters being re-checked.
3226 Version 0.94 18-Sep-97
3227 ----------------------
3229 1. The functions are now thread-safe, with the caveat that the global variables
3230 containing pointers to malloc() and free() or alternative functions are the
3231 same for all threads.
3233 2. Get pcre_study() to generate a bitmap of initial characters for non-
3234 anchored patterns when this is possible, and use it if passed to pcre_exec().
3237 Version 0.93 15-Sep-97
3238 ----------------------
3240 1. /(b)|(:+)/ was computing an incorrect first character.
3242 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3243 but not actually doing anything yet.
3245 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3246 as Perl does (e.g. [-az] or [az-]).
3248 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3249 all possible positions.
3251 5. Split up into different modules to avoid including unneeded functions in a
3252 compiled binary. However, compile and exec are still in one module. The "study"
3253 function is split off.
3255 6. The character tables are now in a separate module whose source is generated
3256 by an auxiliary program - but can then be edited by hand if required. There are
3257 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3258 toupper() in the code.
3260 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3261 make them global. Abolish the function for setting them, as the caller can now
3262 set them directly.
3265 Version 0.92 11-Sep-97
3266 ----------------------
3268 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3269 (e.g. /a{1,3}/) was broken (I mis-optimized it).
3271 2. Caseless matching was not working in character classes if the characters in
3272 the pattern were in upper case.
3274 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3276 4. Make PCRE_ANCHORED public and accept as a compile option.
3278 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3279 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3280 pass them.
3282 6. Give an error if bad option bits passed at compile or run time.
3284 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3285 pcretest to cause it to pass that flag.
3287 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3288 options, and the first character, if set.
3290 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3293 Version 0.91 10-Sep-97
3294 ----------------------
3296 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3297 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3299 2. PCRE was looping on encountering an indefinitely repeated back reference to
3300 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3301 Perl does - treats the match as successful.
3303 ****


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

  ViewVC Help
Powered by ViewVC 1.1.5