Annotation of win32/pcre/ChangeLog, revision 1.2
1.1 misha 1: ChangeLog for PCRE
2: ------------------
3:
1.2 ! misha 4: Version 7.8 05-Sep-08
! 5: ---------------------
! 6:
! 7: 1. Replaced UCP searching code with optimized version as implemented for Ad
! 8: Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
! 9: stage table and inline lookup instead of a function, giving speed ups of 2
! 10: to 5 times on some simple patterns that I tested. Permission was given to
! 11: distribute the MultiStage2.py script that generates the tables (it's not in
! 12: the tarball, but is in the Subversion repository).
! 13:
! 14: 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
! 15: scripts.
! 16:
! 17: 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
! 18: a group with a zero qualifier. The result of the study could be incorrect,
! 19: or the function might crash, depending on the pattern.
! 20:
! 21: 4. Caseless matching was not working for non-ASCII characters in back
! 22: references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
! 23: It now works when Unicode Property Support is available.
! 24:
! 25: 5. In pcretest, an escape such as \x{de} in the data was always generating
! 26: a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
! 27: non-UTF-8 mode. If the value is greater than 255, it gives a warning about
! 28: truncation.
! 29:
! 30: 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
! 31:
! 32: 7. Added two (int) casts to pcregrep when printing the difference of two
! 33: pointers, in case they are 64-bit values.
! 34:
! 35: 8. Added comments about Mac OS X stack usage to the pcrestack man page and to
! 36: test 2 if it fails.
! 37:
! 38: 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
! 39: and a #define of that name to empty if it is not externally set. This is to
! 40: allow users of MSVC to set it if necessary.
! 41:
! 42: 10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
! 43: the convenience functions in the pcre_get.c source file.
! 44:
! 45: 11. An option change at the start of a pattern that had top-level alternatives
! 46: could cause overwriting and/or a crash. This command provoked a crash in
! 47: some environments:
! 48:
! 49: printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
! 50:
! 51: This potential security problem was recorded as CVE-2008-2371.
! 52:
! 53: 12. For a pattern where the match had to start at the beginning or immediately
! 54: after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
! 55: pcre_dfa_exec() could read past the end of the passed subject if there was
! 56: no match. To help with detecting such bugs (e.g. with valgrind), I modified
! 57: pcretest so that it places the subject at the end of its malloc-ed buffer.
! 58:
! 59: 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
! 60: exec() might read past the end of the data buffer in UTF-8 mode.
! 61:
! 62: 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
! 63: the data contained the byte 0x85 as part of a UTF-8 character within its
! 64: first line. This applied both to normal and DFA matching.
! 65:
! 66: 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
! 67: /^[^d]*?$/8 failed to match "abc".
! 68:
! 69: 16. Added a missing copyright notice to pcrecpp_internal.h.
! 70:
! 71: 17. Make it more clear in the documentation that values returned from
! 72: pcre_exec() in ovector are byte offsets, not character counts.
! 73:
! 74: 18. Tidied a few places to stop certain compilers from issuing warnings.
! 75:
! 76: 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
! 77: supplied by Stefan Weber. I made a further small update for 7.8 because
! 78: there is a change of source arrangements: the pcre_searchfuncs.c module is
! 79: replaced by pcre_ucd.c.
! 80:
! 81:
1.1 misha 82: Version 7.7 07-May-08
83: ---------------------
84:
85: 1. Applied Craig's patch to sort out a long long problem: "If we can't convert
86: a string to a long long, pretend we don't even have a long long." This is
87: done by checking for the strtoq, strtoll, and _strtoi64 functions.
88:
89: 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
90: pre-7.6 versions, which defined a global no_arg variable instead of putting
91: it in the RE class. (See also #8 below.)
92:
93: 3. Remove a line of dead code, identified by coverity and reported by Nuno
94: Lopes.
95:
96: 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
97:
98: (1) The include/exclude patterns were being applied to the whole pathnames
99: of files, instead of just to the final components.
100:
101: (2) If there was more than one level of directory, the subdirectories were
102: skipped unless they satisfied the include/exclude conditions. This is
103: inconsistent with GNU grep (and could even be seen as contrary to the
104: pcregrep specification - which I improved to make it absolutely clear).
105: The action now is always to scan all levels of directory, and just
106: apply the include/exclude patterns to regular files.
107:
108: 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
109: --exclude_dir in the tests to avoid scanning .svn directories.
110:
111: 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
112: NUL character as backslash + 0 rather than backslash + NUL, because PCRE
113: doesn't support NULs in patterns.
114:
115: 7. Added some missing "const"s to declarations of static tables in
116: pcre_compile.c and pcre_dfa_exec.c.
117:
118: 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
119: caused by fix #2 above. (Subsequently also a second patch to fix the
120: first patch. And a third patch - this was a messy problem.)
121:
122: 9. Applied Craig's patch to remove the use of push_back().
123:
124: 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
125: matching function regexec().
126:
127: 11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
128: which, however, unlike Perl's \g{...}, are subroutine calls, not back
129: references. PCRE supports relative numbers with this syntax (I don't think
130: Oniguruma does).
131:
132: 12. Previously, a group with a zero repeat such as (...){0} was completely
133: omitted from the compiled regex. However, this means that if the group
134: was called as a subroutine from elsewhere in the pattern, things went wrong
135: (an internal error was given). Such groups are now left in the compiled
136: pattern, with a new opcode that causes them to be skipped at execution
137: time.
138:
139: 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
140: to the way PCRE behaves:
141:
142: (a) A lone ] character is dis-allowed (Perl treats it as data).
143:
144: (b) A back reference to an unmatched subpattern matches an empty string
145: (Perl fails the current match path).
146:
147: (c) A data ] in a character class must be notated as \] because if the
148: first data character in a class is ], it defines an empty class. (In
149: Perl it is not possible to have an empty class.) The empty class []
150: never matches; it forces failure and is equivalent to (*FAIL) or (?!).
151: The negative empty class [^] matches any one character, independently
152: of the DOTALL setting.
153:
154: 14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
155: non-existent subpattern following a character class starting with ']' and
156: containing () gave an internal compiling error instead of "reference to
157: non-existent subpattern". Fortunately, when the pattern did exist, the
158: compiled code was correct. (When scanning forwards to check for the
159: existencd of the subpattern, it was treating the data ']' as terminating
160: the class, so got the count wrong. When actually compiling, the reference
161: was subsequently set up correctly.)
162:
163: 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
164: it was being rejected as not supported by pcre_dfa_exec(), even though
165: other assertions are supported. I have made pcre_dfa_exec() support
166: (*FAIL).
167:
168: 16. The implementation of 13c above involved the invention of a new opcode,
169: OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
170: cannot be changed at match time, I realized I could make a small
171: improvement to matching performance by compiling OP_ALLANY instead of
172: OP_ANY for "." when DOTALL was set, and then removing the runtime tests
173: on the OP_ANY path.
174:
175: 17. Compiling pcretest on Windows with readline support failed without the
176: following two fixes: (1) Make the unistd.h include conditional on
177: HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
178:
179: 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
180: ncurses library to be included for pcretest when ReadLine support is
181: requested, but also to allow for it to be overridden. This patch came from
182: Daniel Bergström.
183:
184: 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
185: as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
186: any errors with the current Unicode tables. Thanks to Peter Kankowski for
187: spotting this.
188:
189:
190: Version 7.6 28-Jan-08
191: ---------------------
192:
193: 1. A character class containing a very large number of characters with
194: codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
195: overflow.
196:
197: 2. Patch to cut out the "long long" test in pcrecpp_unittest when
198: HAVE_LONG_LONG is not defined.
199:
200: 3. Applied Christian Ehrlicher's patch to update the CMake build files to
201: bring them up to date and include new features. This patch includes:
202:
203: - Fixed PH's badly added libz and libbz2 support.
204: - Fixed a problem with static linking.
205: - Added pcredemo. [But later removed - see 7 below.]
206: - Fixed dftables problem and added an option.
207: - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
208: HAVE_LONG_LONG.
209: - Added readline support for pcretest.
210: - Added an listing of the option settings after cmake has run.
211:
212: 4. A user submitted a patch to Makefile that makes it easy to create
213: "pcre.dll" under mingw when using Configure/Make. I added stuff to
214: Makefile.am that cause it to include this special target, without
215: affecting anything else. Note that the same mingw target plus all
216: the other distribution libraries and programs are now supported
217: when configuring with CMake (see 6 below) instead of with
218: Configure/Make.
219:
220: 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
221: This is an attempt to solve the reported problem "pcrecpp::no_arg is not
222: exported in the Windows port". It has not yet been confirmed that the patch
223: solves the problem, but it does no harm.
224:
225: 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
226: NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
227: with CMake, and also correct the comment about stack recursion.
228:
229: 7. Remove the automatic building of pcredemo from the ./configure system and
230: from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
231: of a program that users should build themselves after PCRE is installed, so
232: building it automatically is not really right. What is more, it gave
233: trouble in some build environments.
234:
235: 8. Further tidies to CMakeLists.txt from Sheri and Christian.
236:
237:
238: Version 7.5 10-Jan-08
239: ---------------------
240:
241: 1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
242: values in parens when parsing an RE using the C++ wrapper."
243:
244: 2. Negative specials like \S did not work in character classes in UTF-8 mode.
245: Characters greater than 255 were excluded from the class instead of being
246: included.
247:
248: 3. The same bug as (2) above applied to negated POSIX classes such as
249: [:^space:].
250:
251: 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
252: defined or documented. It seems to have been a typo for PCRE_STATIC, so
253: I have changed it.
254:
255: 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
256: first named subpattern) and a construct such as (?&a) would reference the
257: first named subpattern whose name started with "a" (in other words, the
258: length check was missing). Both these problems are fixed. "Subpattern name
259: expected" is now given for (?&) (a zero-length name), and this patch also
260: makes it give the same error for \k'' (previously it complained that that
261: was a reference to a non-existent subpattern).
262:
263: 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
264: this is right because (?- can be followed by option settings as well as by
265: digits. I have, however, made the messages clearer.
266:
267: 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
268: than the number used in the conditional) now cause a compile-time error.
269: This is actually not compatible with Perl, which accepts such patterns, but
270: treats the conditional as always being FALSE (as PCRE used to), but it
271: seems to me that giving a diagnostic is better.
272:
273: 8. Change "alphameric" to the more common word "alphanumeric" in comments
274: and messages.
275:
276: 9. Fix two occurrences of "backslash" in comments that should have been
277: "backspace".
278:
279: 10. Remove two redundant lines of code that can never be obeyed (their function
280: was moved elsewhere).
281:
282: 11. The program that makes PCRE's Unicode character property table had a bug
283: which caused it to generate incorrect table entries for sequences of
284: characters that have the same character type, but are in different scripts.
285: It amalgamated them into a single range, with the script of the first of
286: them. In other words, some characters were in the wrong script. There were
287: thirteen such cases, affecting characters in the following ranges:
288:
289: U+002b0 - U+002c1
290: U+0060c - U+0060d
291: U+0061e - U+00612
292: U+0064b - U+0065e
293: U+0074d - U+0076d
294: U+01800 - U+01805
295: U+01d00 - U+01d77
296: U+01d9b - U+01dbf
297: U+0200b - U+0200f
298: U+030fc - U+030fe
299: U+03260 - U+0327f
300: U+0fb46 - U+0fbb1
301: U+10450 - U+1049d
302:
303: 12. The -o option (show only the matching part of a line) for pcregrep was not
304: compatible with GNU grep in that, if there was more than one match in a
305: line, it showed only the first of them. It now behaves in the same way as
306: GNU grep.
307:
308: 13. If the -o and -v options were combined for pcregrep, it printed a blank
309: line for every non-matching line. GNU grep prints nothing, and pcregrep now
310: does the same. The return code can be used to tell if there were any
311: non-matching lines.
312:
313: 14. Added --file-offsets and --line-offsets to pcregrep.
314:
315: 15. The pattern (?=something)(?R) was not being diagnosed as a potentially
316: infinitely looping recursion. The bug was that positive lookaheads were not
317: being skipped when checking for a possible empty match (negative lookaheads
318: and both kinds of lookbehind were skipped).
319:
320: 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
321: inclusion of <windows.h> to before rather than after the definition of
322: INVALID_FILE_ATTRIBUTES (patch from David Byron).
323:
324: 17. Specifying a possessive quantifier with a specific limit for a Unicode
325: character property caused pcre_compile() to compile bad code, which led at
326: runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
327: are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
328: caused the error; without that there was no problem.
329:
330: 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
331:
332: 19. Added --enable-pcretest-libreadline.
333:
334: 20. In pcrecpp.cc, the variable 'count' was incremented twice in
335: RE::GlobalReplace(). As a result, the number of replacements returned was
336: double what it should be. I removed one of the increments, but Craig sent a
337: later patch that removed the other one (the right fix) and added unit tests
338: that check the return values (which was not done before).
339:
340: 21. Several CMake things:
341:
342: (1) Arranged that, when cmake is used on Unix, the libraries end up with
343: the names libpcre and libpcreposix, not just pcre and pcreposix.
344:
345: (2) The above change means that pcretest and pcregrep are now correctly
346: linked with the newly-built libraries, not previously installed ones.
347:
348: (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
349:
350: 22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
351: crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
352: UTF-8 newline character). The key issue is that the pattern starts .*;
353: this means that the match must be either at the beginning, or after a
354: newline. The bug was in the code for advancing after a failed match and
355: checking that the new position followed a newline. It was not taking
356: account of UTF-8 characters correctly.
357:
358: 23. PCRE was behaving differently from Perl in the way it recognized POSIX
359: character classes. PCRE was not treating the sequence [:...:] as a
360: character class unless the ... were all letters. Perl, however, seems to
361: allow any characters between [: and :], though of course it rejects as
362: unknown any "names" that contain non-letters, because all the known class
363: names consist only of letters. Thus, Perl gives an error for [[:1234:]],
364: for example, whereas PCRE did not - it did not recognize a POSIX character
365: class. This seemed a bit dangerous, so the code has been changed to be
366: closer to Perl. The behaviour is not identical to Perl, because PCRE will
367: diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
368: treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
369: Perl does, and where it didn't before.
370:
371: 24. Rewrite so as to remove the single use of %n from pcregrep because in some
372: Windows environments %n is disabled by default.
373:
374:
375: Version 7.4 21-Sep-07
376: ---------------------
377:
378: 1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
379: means that a class such as [\s] counted as "explicit reference to CR or
380: LF". That isn't really right - the whole point of the change was to try to
381: help when there was an actual mention of one of the two characters. So now
382: the change happens only if \r or \n (or a literal CR or LF) character is
383: encountered.
384:
385: 2. The 32-bit options word was also used for 6 internal flags, but the numbers
386: of both had grown to the point where there were only 3 bits left.
387: Fortunately, there was spare space in the data structure, and so I have
388: moved the internal flags into a new 16-bit field to free up more option
389: bits.
390:
391: 3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
392: but did not set the internal JCHANGED flag - either of these is enough to
393: control the way the "get" function works - but the PCRE_INFO_JCHANGED
394: facility is supposed to tell if (?J) was ever used, so now (?J) at the
395: start sets both bits.
396:
397: 4. Added options (at build time, compile time, exec time) to change \R from
398: matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
399:
400: 5. doc/pcresyntax.html was missing from the distribution.
401:
402: 6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
403: compatibility, even though it is no longer used.
404:
405: 7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
406: strtoull to pcrecpp.cc to select the available functions in WIN32 when the
407: windows.h file is present (where different names are used). [This was
408: reversed later after testing - see 16 below.]
409:
410: 8. Changed all #include <config.h> to #include "config.h". There were also
411: some further <pcre.h> cases that I changed to "pcre.h".
412:
413: 9. When pcregrep was used with the --colour option, it missed the line ending
414: sequence off the lines that it output.
415:
416: 10. It was pointed out to me that arrays of string pointers cause lots of
417: relocations when a shared library is dynamically loaded. A technique of
418: using a single long string with a table of offsets can drastically reduce
419: these. I have refactored PCRE in four places to do this. The result is
420: dramatic:
421:
422: Originally: 290
423: After changing UCP table: 187
424: After changing error message table: 43
425: After changing table of "verbs" 36
426: After changing table of Posix names 22
427:
428: Thanks to the folks working on Gregex for glib for this insight.
429:
430: 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
431: unicode-properties was also set.
432:
433: 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
434:
435: 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
436: checked only for CRLF.
437:
438: 14. Added casts to pcretest.c to avoid compiler warnings.
439:
440: 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
441:
442: 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
443: and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
444: entirely. This removes changes made in 7 above.
445:
446: 17. The CMake files have been updated, and there is now more information about
447: building with CMake in the NON-UNIX-USE document.
448:
449:
450: Version 7.3 28-Aug-07
451: ---------------------
452:
453: 1. In the rejigging of the build system that eventually resulted in 7.1, the
454: line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
455: brackets there is not right, since it causes compilers to look for an
456: installed pcre.h, not the version that is in the source that is being
457: compiled (which of course may be different). I have changed it back to:
458:
459: #include "pcre.h"
460:
461: I have a vague recollection that the change was concerned with compiling in
462: different directories, but in the new build system, that is taken care of
463: by the VPATH setting the Makefile.
464:
465: 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
466: when the subject happened to end in the byte 0x85 (e.g. if the last
467: character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
468: characters but of course it shouldn't be taken as a newline when it is part
469: of another character. The bug was that, for an unlimited repeat of . in
470: not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
471: characters when looking for a newline.
472:
473: 3. A small performance improvement in the DOTALL UTF-8 mode .* case.
474:
475: 4. Debugging: adjusted the names of opcodes for different kinds of parentheses
476: in debug output.
477:
478: 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
479: long printing in the pcrecpp unittest when running under MinGW.
480:
481: 6. ESC_K was left out of the EBCDIC table.
482:
483: 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
484: parentheses; I made it 1000, which seemed large enough. Unfortunately, the
485: limit also applies to "virtual nesting" when a pattern is recursive, and in
486: this case 1000 isn't so big. I have been able to remove this limit at the
487: expense of backing off one optimization in certain circumstances. Normally,
488: when pcre_exec() would call its internal match() function recursively and
489: immediately return the result unconditionally, it uses a "tail recursion"
490: feature to save stack. However, when a subpattern that can match an empty
491: string has an unlimited repetition quantifier, it no longer makes this
492: optimization. That gives it a stack frame in which to save the data for
493: checking that an empty string has been matched. Previously this was taken
494: from the 1000-entry workspace that had been reserved. So now there is no
495: explicit limit, but more stack is used.
496:
497: 8. Applied Daniel's patches to solve problems with the import/export magic
498: syntax that is required for Windows, and which was going wrong for the
499: pcreposix and pcrecpp parts of the library. These were overlooked when this
500: problem was solved for the main library.
501:
502: 9. There were some crude static tests to avoid integer overflow when computing
503: the size of patterns that contain repeated groups with explicit upper
504: limits. As the maximum quantifier is 65535, the maximum group length was
505: set at 30,000 so that the product of these two numbers did not overflow a
506: 32-bit integer. However, it turns out that people want to use groups that
507: are longer than 30,000 bytes (though not repeat them that many times).
508: Change 7.0/17 (the refactoring of the way the pattern size is computed) has
509: made it possible to implement the integer overflow checks in a much more
510: dynamic way, which I have now done. The artificial limitation on group
511: length has been removed - we now have only the limit on the total length of
512: the compiled pattern, which depends on the LINK_SIZE setting.
513:
514: 10. Fixed a bug in the documentation for get/copy named substring when
515: duplicate names are permitted. If none of the named substrings are set, the
516: functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
517: empty string.
518:
519: 11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
520: instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
521: because the ] is interpreted as the first data character and the
522: terminating ] is not found. PCRE has been made compatible with Perl in this
523: regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
524: cause memory overwriting.
525:
526: 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
527: string has been matched (to stop an infinite loop). It was not recognizing
528: a conditional subpattern that could match an empty string if that
529: subpattern was within another subpattern. For example, it looped when
530: trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
531: condition was not nested. This bug has been fixed.
532:
533: 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
534: past the start of the subject in the presence of bytes with the top bit
535: set, for example "\x8aBCD".
536:
537: 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
538: (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
539:
540: 14. Optimized (?!) to (*FAIL).
541:
542: 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
543: This restricts code points to be within the range 0 to 0x10FFFF, excluding
544: the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
545: full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
546: does: it's just the validity check that is more restrictive.
547:
548: 16. Inserted checks for integer overflows during escape sequence (backslash)
549: processing, and also fixed erroneous offset values for syntax errors during
550: backslash processing.
551:
552: 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
553: for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
554:
555: 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
556: caused an overrun.
557:
558: 19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
559: something other than just ASCII characters) inside a group that had an
560: unlimited repeat caused a loop at compile time (while checking to see
561: whether the group could match an empty string).
562:
563: 20. Debugging a pattern containing \p or \P could cause a crash. For example,
564: [\P{Any}] did so. (Error in the code for printing property names.)
565:
566: 21. An orphan \E inside a character class could cause a crash.
567:
568: 22. A repeated capturing bracket such as (A)? could cause a wild memory
569: reference during compilation.
570:
571: 23. There are several functions in pcre_compile() that scan along a compiled
572: expression for various reasons (e.g. to see if it's fixed length for look
573: behind). There were bugs in these functions when a repeated \p or \P was
574: present in the pattern. These operators have additional parameters compared
575: with \d, etc, and these were not being taken into account when moving along
576: the compiled data. Specifically:
577:
578: (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
579: length.
580:
581: (b) An item such as \pL+ within a repeated group could cause crashes or
582: loops.
583:
584: (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
585: "reference to non-existent subpattern" error.
586:
587: (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
588:
589: 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
590: characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
591:
592: 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
593:
594: 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
595: character were causing crashes (broken optimization).
596:
597: 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
598: \p or \P) caused a compile-time loop.
599:
600: 28. More problems have arisen in unanchored patterns when CRLF is a valid line
601: break. For example, the unstudied pattern [\r\n]A does not match the string
602: "\r\nA" because change 7.0/46 below moves the current point on by two
603: characters after failing to match at the start. However, the pattern \nA
604: *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
605: the same is true. There doesn't seem any very clean way out of this, but
606: what I have chosen to do makes the common cases work: PCRE now takes note
607: of whether there can be an explicit match for \r or \n anywhere in the
608: pattern, and if so, 7.0/46 no longer applies. As part of this change,
609: there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
610: pattern has explicit CR or LF references.
611:
612: 29. Added (*CR) etc for changing newline setting at start of pattern.
613:
614:
615: Version 7.2 19-Jun-07
616: ---------------------
617:
618: 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
619: which is apparently normally available under Windows.
620:
621: 2. Re-jig the pcregrep tests with different newline settings in an attempt
622: to make them independent of the local environment's newline setting.
623:
624: 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
625:
626: 4. Some of the "internals" tests were previously cut out when the link size
627: was not 2, because the output contained actual offsets. The recent new
628: "Z" feature of pcretest means that these can be cut out, making the tests
629: usable with all link sizes.
630:
631: 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
632: stack recursion. This gives a massive performance boost under BSD, but just
633: a small improvement under Linux. However, it saves one field in the frame
634: in all cases.
635:
636: 6. Added more features from the forthcoming Perl 5.10:
637:
638: (a) (?-n) (where n is a string of digits) is a relative subroutine or
639: recursion call. It refers to the nth most recently opened parentheses.
640:
641: (b) (?+n) is also a relative subroutine call; it refers to the nth next
642: to be opened parentheses.
643:
644: (c) Conditions that refer to capturing parentheses can be specified
645: relatively, for example, (?(-2)... or (?(+3)...
646:
647: (d) \K resets the start of the current match so that everything before
648: is not part of it.
649:
650: (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
651:
652: (f) \g{name} is another synonym - part of Perl 5.10's unification of
653: reference syntax.
654:
655: (g) (?| introduces a group in which the numbering of parentheses in each
656: alternative starts with the same number.
657:
658: (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
659:
660: 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
661: PCRE_INFO_JCHANGED.
662:
663: 8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
664: terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
665: for detecting groups that can match an empty string.
666:
667: 9. A pattern with a very large number of alternatives (more than several
668: hundred) was running out of internal workspace during the pre-compile
669: phase, where pcre_compile() figures out how much memory will be needed. A
670: bit of new cunning has reduced the workspace needed for groups with
671: alternatives. The 1000-alternative test pattern now uses 12 bytes of
672: workspace instead of running out of the 4096 that are available.
673:
674: 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
675:
676: 11. Applied patch from Google to remove an optimization that didn't quite work.
677: The report of the bug said:
678:
679: pcrecpp::RE("a*").FullMatch("aaa") matches, while
680: pcrecpp::RE("a*?").FullMatch("aaa") does not, and
681: pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
682:
683: 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
684: it matched the wrong number of bytes.
685:
686:
687: Version 7.1 24-Apr-07
688: ---------------------
689:
690: 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
691: that is more "standard", making use of automake and other Autotools. There
692: is some re-arrangement of the files and adjustment of comments consequent
693: on this.
694:
695: 2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
696: for recursive directory scanning broke on some systems because the files
697: are not scanned in any specific order and on different systems the order
698: was different. A call to "sort" has been inserted into RunGrepTest for the
699: approprate test as a short-term fix. In the longer term there may be an
700: alternative.
701:
702: 3. I had an email from Eric Raymond about problems translating some of PCRE's
703: man pages to HTML (despite the fact that I distribute HTML pages, some
704: people do their own conversions for various reasons). The problems
705: concerned the use of low-level troff macros .br and .in. I have therefore
706: removed all such uses from the man pages (some were redundant, some could
707: be replaced by .nf/.fi pairs). The 132html script that I use to generate
708: HTML has been updated to handle .nf/.fi and to complain if it encounters
709: .br or .in.
710:
711: 4. Updated comments in configure.ac that get placed in config.h.in and also
712: arranged for config.h to be included in the distribution, with the name
713: config.h.generic, for the benefit of those who have to compile without
714: Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
715:
716: 5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
717: Weber: (1) pcre_internal.h was missing some function renames; (2) updated
718: makevp.bat for the current PCRE, using the additional files
719: makevp_c.txt, makevp_l.txt, and pcregexp.pas.
720:
721: 6. A Windows user reported a minor discrepancy with test 2, which turned out
722: to be caused by a trailing space on an input line that had got lost in his
723: copy. The trailing space was an accident, so I've just removed it.
724:
725: 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
726: that is needed.
727:
728: 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
729: as "const" (a) because they are and (b) because it helps the PHP
730: maintainers who have recently made a script to detect big data structures
731: in the php code that should be moved to the .rodata section. I remembered
732: to update Builducptable as well, so it won't revert if ucptable.h is ever
733: re-created.
734:
735: 9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
736: pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
737: order to be able to cut out the UTF-8 tables in the latter when UTF-8
738: support is not required. This saves 1.5-2K of code, which is important in
739: some applications.
740:
741: Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
742: so as not to refer to the tables, even though these functions will never be
743: called when UTF-8 support is disabled. Otherwise there are problems with a
744: shared library.
745:
746: 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
747:
748: (a) It was defining its arguments as char * instead of void *.
749:
750: (b) It was assuming that all moves were upwards in memory; this was true
751: a long time ago when I wrote it, but is no longer the case.
752:
753: The emulated memove() is provided for those environments that have neither
754: memmove() nor bcopy(). I didn't think anyone used it these days, but that
755: is clearly not the case, as these two bugs were recently reported.
756:
757: 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
758: and Detrail to create the HTML documentation, the .txt form of the man
759: pages, and it removes trailing spaces from listed files. It also creates
760: pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
761: case, it wraps all the #defines with #ifndefs. This script should be run
762: before "make dist".
763:
764: 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
765: with Unicode property support.
766:
767: (a) For a maximizing quantifier, if the two different cases of the
768: character were of different lengths in their UTF-8 codings (there are
769: some cases like this - I found 11), and the matching function had to
770: back up over a mixture of the two cases, it incorrectly assumed they
771: were both the same length.
772:
773: (b) When PCRE was configured to use the heap rather than the stack for
774: recursion during matching, it was not correctly preserving the data for
775: the other case of a UTF-8 character when checking ahead for a match
776: while processing a minimizing repeat. If the check also involved
777: matching a wide character, but failed, corruption could cause an
778: erroneous result when trying to check for a repeat of the original
779: character.
780:
781: 13. Some tidying changes to the testing mechanism:
782:
783: (a) The RunTest script now detects the internal link size and whether there
784: is UTF-8 and UCP support by running ./pcretest -C instead of relying on
785: values substituted by "configure". (The RunGrepTest script already did
786: this for UTF-8.) The configure.ac script no longer substitutes the
787: relevant variables.
788:
789: (b) The debugging options /B and /D in pcretest show the compiled bytecode
790: with length and offset values. This means that the output is different
791: for different internal link sizes. Test 2 is skipped for link sizes
792: other than 2 because of this, bypassing the problem. Unfortunately,
793: there was also a test in test 3 (the locale tests) that used /B and
794: failed for link sizes other than 2. Rather than cut the whole test out,
795: I have added a new /Z option to pcretest that replaces the length and
796: offset values with spaces. This is now used to make test 3 independent
797: of link size. (Test 2 will be tidied up later.)
798:
799: 14. If erroroffset was passed as NULL to pcre_compile, it provoked a
800: segmentation fault instead of returning the appropriate error message.
801:
802: 15. In multiline mode when the newline sequence was set to "any", the pattern
803: ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
804: This doesn't seem right; it now treats the CRLF combination as the line
805: ending, and so does not match in that case. It's only a pattern such as ^$
806: that would hit this one: something like ^ABC$ would have failed after \r
807: and then tried again after \r\n.
808:
809: 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
810: in an attempt to make files that differ only in their line terminators
811: compare equal. This works on Linux.
812:
813: 17. Under certain error circumstances pcregrep might try to free random memory
814: as it exited. This is now fixed, thanks to valgrind.
815:
816: 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
817: "abc\r\n\r\n", it found an unwanted second match after the second \r. This
818: was because its rules for how to advance for /g after matching an empty
819: string at the end of a line did not allow for this case. They now check for
820: it specially.
821:
822: 20. pcretest is supposed to handle patterns and data of any length, by
823: extending its buffers when necessary. It was getting this wrong when the
824: buffer for a data line had to be extended.
825:
826: 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
827: CRLF as a newline sequence.
828:
829: 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
830: out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
831: I have nevertheless tidied it up.
832:
833: 23. Added some casts to kill warnings from HP-UX ia64 compiler.
834:
835: 24. Added a man page for pcre-config.
836:
837:
838: Version 7.0 19-Dec-06
839: ---------------------
840:
841: 1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
842: moving to gcc 4.1.1.
843:
844: 2. The -S option for pcretest uses setrlimit(); I had omitted to #include
845: sys/time.h, which is documented as needed for this function. It doesn't
846: seem to matter on Linux, but it showed up on some releases of OS X.
847:
848: 3. It seems that there are systems where bytes whose values are greater than
849: 127 match isprint() in the "C" locale. The "C" locale should be the
850: default when a C program starts up. In most systems, only ASCII printing
851: characters match isprint(). This difference caused the output from pcretest
852: to vary, making some of the tests fail. I have changed pcretest so that:
853:
854: (a) When it is outputting text in the compiled version of a pattern, bytes
855: other than 32-126 are always shown as hex escapes.
856:
857: (b) When it is outputting text that is a matched part of a subject string,
858: it does the same, unless a different locale has been set for the match
859: (using the /L modifier). In this case, it uses isprint() to decide.
860:
861: 4. Fixed a major bug that caused incorrect computation of the amount of memory
862: required for a compiled pattern when options that changed within the
863: pattern affected the logic of the preliminary scan that determines the
864: length. The relevant options are -x, and -i in UTF-8 mode. The result was
865: that the computed length was too small. The symptoms of this bug were
866: either the PCRE error "internal error: code overflow" from pcre_compile(),
867: or a glibc crash with a message such as "pcretest: free(): invalid next
868: size (fast)". Examples of patterns that provoked this bug (shown in
869: pcretest format) are:
870:
871: /(?-x: )/x
872: /(?x)(?-x: \s*#\s*)/
873: /((?i)[\x{c0}])/8
874: /(?i:[\x{c0}])/8
875:
876: HOWEVER: Change 17 below makes this fix obsolete as the memory computation
877: is now done differently.
878:
879: 5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
880: wrapper classes; (b) implement a new function in the C++ scanner that is
881: more efficient than the old way of doing things because it avoids levels of
882: recursion in the regex matching; (c) add a paragraph to the documentation
883: for the FullMatch() function.
884:
885: 6. The escape sequence \n was being treated as whatever was defined as
886: "newline". Not only was this contrary to the documentation, which states
887: that \n is character 10 (hex 0A), but it also went horribly wrong when
888: "newline" was defined as CRLF. This has been fixed.
889:
890: 7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
891: was being set to -1 for the "end of line" case (supposedly a value that no
892: character can have). Though this value is never used (the check for end of
893: line is "zero bytes in current character"), it caused compiler complaints.
894: I've changed it to 0xffffffff.
895:
896: 8. In pcre_version.c, the version string was being built by a sequence of
897: C macros that, in the event of PCRE_PRERELEASE being defined as an empty
898: string (as it is for production releases) called a macro with an empty
899: argument. The C standard says the result of this is undefined. The gcc
900: compiler treats it as an empty string (which was what was wanted) but it is
901: reported that Visual C gives an error. The source has been hacked around to
902: avoid this problem.
903:
904: 9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
905: builds of pcretest, and changed the call to _setmode() to use _O_BINARY
906: instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
907: of them did).
908:
909: 10. Originally, pcretest opened its input and output without "b"; then I was
910: told that "b" was needed in some environments, so it was added for release
911: 5.0 to both the input and output. (It makes no difference on Unix-like
912: systems.) Later I was told that it is wrong for the input on Windows. I've
913: now abstracted the modes into two macros, to make it easier to fiddle with
914: them, and removed "b" from the input mode under Windows.
915:
916: 11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
917:
918: 12. Added -help and --help to pcretest as an official way of being reminded
919: of the options.
920:
921: 13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
922: and pcrecpp.cc because they annoy compilers at high warning levels.
923:
924: 14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
925:
926: 15. Fixed an occurrence of == in configure.ac that should have been = (shell
927: scripts are not C programs :-) and which was not noticed because it works
928: on Linux.
929:
930: 16. pcretest is supposed to handle any length of pattern and data line (as one
931: line or as a continued sequence of lines) by extending its input buffer if
932: necessary. This feature was broken for very long pattern lines, leading to
933: a string of junk being passed to pcre_compile() if the pattern was longer
934: than about 50K.
935:
936: 17. I have done a major re-factoring of the way pcre_compile() computes the
937: amount of memory needed for a compiled pattern. Previously, there was code
938: that made a preliminary scan of the pattern in order to do this. That was
939: OK when PCRE was new, but as the facilities have expanded, it has become
940: harder and harder to keep it in step with the real compile phase, and there
941: have been a number of bugs (see for example, 4 above). I have now found a
942: cunning way of running the real compile function in a "fake" mode that
943: enables it to compute how much memory it would need, while actually only
944: ever using a few hundred bytes of working memory and without too many
945: tests of the mode. This should make future maintenance and development
946: easier. A side effect of this work is that the limit of 200 on the nesting
947: depth of parentheses has been removed (though this was never a serious
948: limitation, I suspect). However, there is a downside: pcre_compile() now
949: runs more slowly than before (30% or more, depending on the pattern). I
950: hope this isn't a big issue. There is no effect on runtime performance.
951:
952: 18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
953: newline (only possible for the last line of a file) and it was a
954: pattern that set a locale (followed by /Lsomething), pcretest crashed.
955:
956: 19. Added additional timing features to pcretest. (1) The -tm option now times
957: matching only, not compiling. (2) Both -t and -tm can be followed, as a
958: separate command line item, by a number that specifies the number of
959: repeats to use when timing. The default is 50000; this gives better
960: precision, but takes uncomfortably long for very large patterns.
961:
962: 20. Extended pcre_study() to be more clever in cases where a branch of a
963: subpattern has no definite first character. For example, (a*|b*)[cd] would
964: previously give no result from pcre_study(). Now it recognizes that the
965: first character must be a, b, c, or d.
966:
967: 21. There was an incorrect error "recursive call could loop indefinitely" if
968: a subpattern (or the entire pattern) that was being tested for matching an
969: empty string contained only one non-empty item after a nested subpattern.
970: For example, the pattern (?>\x{100}*)\d(?R) provoked this error
971: incorrectly, because the \d was being skipped in the check.
972:
973: 22. The pcretest program now has a new pattern option /B and a command line
974: option -b, which is equivalent to adding /B to every pattern. This causes
975: it to show the compiled bytecode, without the additional information that
976: -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
977: is the same as /B/I).
978:
979: 23. A new optimization is now able automatically to treat some sequences such
980: as a*b as a*+b. More specifically, if something simple (such as a character
981: or a simple class like \d) has an unlimited quantifier, and is followed by
982: something that cannot possibly match the quantified thing, the quantifier
983: is automatically "possessified".
984:
985: 24. A recursive reference to a subpattern whose number was greater than 39
986: went wrong under certain circumstances in UTF-8 mode. This bug could also
987: have affected the operation of pcre_study().
988:
989: 25. Realized that a little bit of performance could be had by replacing
990: (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
991:
992: 26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
993:
994: 27. Possessive quantifiers such as a++ were previously implemented by turning
995: them into atomic groups such as ($>a+). Now they have their own opcodes,
996: which improves performance. This includes the automatically created ones
997: from 23 above.
998:
999: 28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1000: lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1001: the first matched character to be a colon. This applied both to named and
1002: numbered groups.
1003:
1004: 29. The ucpinternal.h header file was missing its idempotency #ifdef.
1005:
1006: 30. I was sent a "project" file called libpcre.a.dev which I understand makes
1007: building PCRE on Windows easier, so I have included it in the distribution.
1008:
1009: 31. There is now a check in pcretest against a ridiculously large number being
1010: returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1011: loop, the loop is abandoned.
1012:
1013: 32. Forward references to subpatterns in conditions such as (?(2)...) where
1014: subpattern 2 is defined later cause pcre_compile() to search forwards in
1015: the pattern for the relevant set of parentheses. This search went wrong
1016: when there were unescaped parentheses in a character class, parentheses
1017: escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1018:
1019: 33. "Subroutine" calls and backreferences were previously restricted to
1020: referencing subpatterns earlier in the regex. This restriction has now
1021: been removed.
1022:
1023: 34. Added a number of extra features that are going to be in Perl 5.10. On the
1024: whole, these are just syntactic alternatives for features that PCRE had
1025: previously implemented using the Python syntax or my own invention. The
1026: other formats are all retained for compatibility.
1027:
1028: (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1029: as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1030: also .NET compatible.
1031:
1032: (b) A recursion or subroutine call to a named group can now be defined as
1033: (?&name) as well as (?P>name).
1034:
1035: (c) A backreference to a named group can now be defined as \k<name> or
1036: \k'name' as well as (?P=name). The new forms, as well as being in Perl
1037: 5.10, are also .NET compatible.
1038:
1039: (d) A conditional reference to a named group can now use the syntax
1040: (?(<name>) or (?('name') as well as (?(name).
1041:
1042: (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1043: groups (named and numbered) that are never evaluated inline, but can be
1044: called as "subroutines" from elsewhere. In effect, the DEFINE condition
1045: is always false. There may be only one alternative in such a group.
1046:
1047: (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1048: as the simple (?(R). The condition is true only if the most recent
1049: recursion is that of the given number or name. It does not search out
1050: through the entire recursion stack.
1051:
1052: (g) The escape \gN or \g{N} has been added, where N is a positive or
1053: negative number, specifying an absolute or relative reference.
1054:
1055: 35. Tidied to get rid of some further signed/unsigned compiler warnings and
1056: some "unreachable code" warnings.
1057:
1058: 36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1059: things, this adds five new scripts.
1060:
1061: 37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1062: There were also incompatibilities regarding the handling of \Q..\E inside
1063: character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1064: hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1065:
1066: 38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1067: matches an empty string, and forcibly breaks the loop. There were bugs in
1068: this code in non-simple cases. For a pattern such as ^(a()*)* matched
1069: against aaaa the result was just "a" rather than "aaaa", for example. Two
1070: separate and independent bugs (that affected different cases) have been
1071: fixed.
1072:
1073: 39. Refactored the code to abolish the use of different opcodes for small
1074: capturing bracket numbers. This is a tidy that I avoided doing when I
1075: removed the limit on the number of capturing brackets for 3.5 back in 2001.
1076: The new approach is not only tidier, it makes it possible to reduce the
1077: memory needed to fix the previous bug (38).
1078:
1079: 40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1080: sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1081: processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1082: mode.
1083:
1084: 41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1085: report.
1086:
1087: 42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1088: copy construction and assignment in the C++ wrapper.
1089:
1090: 43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1091: couple of bugs that could have given wrong results in the "--newline=crlf"
1092: case.
1093:
1094: 44. Added a number of casts and did some reorganization of signed/unsigned int
1095: variables following suggestions from Dair Grant. Also renamed the variable
1096: "this" as "item" because it is a C++ keyword.
1097:
1098: 45. Arranged for dftables to add
1099:
1100: #include "pcre_internal.h"
1101:
1102: to pcre_chartables.c because without it, gcc 4.x may remove the array
1103: definition from the final binary if PCRE is built into a static library and
1104: dead code stripping is activated.
1105:
1106: 46. For an unanchored pattern, if a match attempt fails at the start of a
1107: newline sequence, and the newline setting is CRLF or ANY, and the next two
1108: characters are CRLF, advance by two characters instead of one.
1109:
1110:
1111: Version 6.7 04-Jul-06
1112: ---------------------
1113:
1114: 1. In order to handle tests when input lines are enormously long, pcretest has
1115: been re-factored so that it automatically extends its buffers when
1116: necessary. The code is crude, but this _is_ just a test program. The
1117: default size has been increased from 32K to 50K.
1118:
1119: 2. The code in pcre_study() was using the value of the re argument before
1120: testing it for NULL. (Of course, in any sensible call of the function, it
1121: won't be NULL.)
1122:
1123: 3. The memmove() emulation function in pcre_internal.h, which is used on
1124: systems that lack both memmove() and bcopy() - that is, hardly ever -
1125: was missing a "static" storage class specifier.
1126:
1127: 4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1128: containing an extended class (one that cannot be represented by a bitmap
1129: because it contains high-valued characters or Unicode property items, e.g.
1130: [\pZ]). Almost always one would set UTF-8 mode when processing such a
1131: pattern, but PCRE should not loop if you do not (it no longer does).
1132: [Detail: two cases were found: (a) a repeated subpattern containing an
1133: extended class; (b) a recursive reference to a subpattern that followed a
1134: previous extended class. It wasn't skipping over the extended class
1135: correctly when UTF-8 mode was not set.]
1136:
1137: 5. A negated single-character class was not being recognized as fixed-length
1138: in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1139: compile error "lookbehind assertion is not fixed length".
1140:
1141: 6. The RunPerlTest auxiliary script was showing an unexpected difference
1142: between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1143: write a Perl script that can interpret lines of an input file either as
1144: byte characters or as UTF-8, which is what "perltest" was being required to
1145: do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1146: can't do is switch easily at run time between having the "use utf8;" pragma
1147: or not. In the end, I fudged it by using the RunPerlTest script to insert
1148: "use utf8;" explicitly for the UTF-8 tests.
1149:
1150: 7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1151: the end of the subject string, contrary to the documentation and to what
1152: Perl does. This was true of both matching functions. Now it matches only at
1153: the start of the subject and immediately after *internal* newlines.
1154:
1155: 8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1156: a pointer to an int instead of a pointer to an unsigned long int. This
1157: caused problems on 64-bit systems.
1158:
1159: 9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1160: instance of the 'standard' template library not being so standard".
1161:
1162: 10. There was no check on the number of named subpatterns nor the maximum
1163: length of a subpattern name. The product of these values is used to compute
1164: the size of the memory block for a compiled pattern. By supplying a very
1165: long subpattern name and a large number of named subpatterns, the size
1166: computation could be caused to overflow. This is now prevented by limiting
1167: the length of names to 32 characters, and the number of named subpatterns
1168: to 10,000.
1169:
1170: 11. Subpatterns that are repeated with specific counts have to be replicated in
1171: the compiled pattern. The size of memory for this was computed from the
1172: length of the subpattern and the repeat count. The latter is limited to
1173: 65535, but there was no limit on the former, meaning that integer overflow
1174: could in principle occur. The compiled length of a repeated subpattern is
1175: now limited to 30,000 bytes in order to prevent this.
1176:
1177: 12. Added the optional facility to have named substrings with the same name.
1178:
1179: 13. Added the ability to use a named substring as a condition, using the
1180: Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1181: are numbers (not recommended). Forward references are permitted.
1182:
1183: 14. Added forward references in named backreferences (if you see what I mean).
1184:
1185: 15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1186: pattern could run off the end of the subject. For example, the pattern
1187: "(?s)(.{1,5})"8 did this with the subject "ab".
1188:
1189: 16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1190: PCRE_CASELESS was set when matching characters that were quantified with ?
1191: or *.
1192:
1193: 17. A character class other than a single negated character that had a minimum
1194: but no maximum quantifier - for example [ab]{6,} - was not handled
1195: correctly by pce_dfa_exec(). It would match only one character.
1196:
1197: 18. A valid (though odd) pattern that looked like a POSIX character
1198: class but used an invalid character after [ (for example [[,abc,]]) caused
1199: pcre_compile() to give the error "Failed: internal error: code overflow" or
1200: in some cases to crash with a glibc free() error. This could even happen if
1201: the pattern terminated after [[ but there just happened to be a sequence of
1202: letters, a binary zero, and a closing ] in the memory that followed.
1203:
1204: 19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1205: over the years. Originally (before any Unicode support), just the bottom 8
1206: bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1207: output from "man perlunicode" includes this:
1208:
1209: The regular expression compiler produces polymorphic opcodes. That
1210: is, the pattern adapts to the data and automatically switches to
1211: the Unicode character scheme when presented with Unicode data--or
1212: instead uses a traditional byte scheme when presented with byte
1213: data.
1214:
1215: Sadly, a wide octal escape does not cause a switch, and in a string with
1216: no other multibyte characters, these octal escapes are treated as before.
1217: Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1218: /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1219: Unicode string.
1220:
1221: I have not perpetrated such confusion in PCRE. Up till now, it took just
1222: the bottom 8 bits, as in old Perl. I have now made octal escapes with
1223: values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1224: translate to the appropriate multibyte character.
1225:
1226: 29. Applied some refactoring to reduce the number of warnings from Microsoft
1227: and Borland compilers. This has included removing the fudge introduced
1228: seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1229: a warning about an unused variable.
1230:
1231: 21. PCRE has not included VT (character 0x0b) in the set of whitespace
1232: characters since release 4.0, because Perl (from release 5.004) does not.
1233: [Or at least, is documented not to: some releases seem to be in conflict
1234: with the documentation.] However, when a pattern was studied with
1235: pcre_study() and all its branches started with \s, PCRE still included VT
1236: as a possible starting character. Of course, this did no harm; it just
1237: caused an unnecessary match attempt.
1238:
1239: 22. Removed a now-redundant internal flag bit that recorded the fact that case
1240: dependency changed within the pattern. This was once needed for "required
1241: byte" processing, but is no longer used. This recovers a now-scarce options
1242: bit. Also moved the least significant internal flag bit to the most-
1243: significant bit of the word, which was not previously used (hangover from
1244: the days when it was an int rather than a uint) to free up another bit for
1245: the future.
1246:
1247: 23. Added support for CRLF line endings as well as CR and LF. As well as the
1248: default being selectable at build time, it can now be changed at runtime
1249: via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1250: specify that it is scanning data with non-default line endings.
1251:
1252: 24. Changed the definition of CXXLINK to make it agree with the definition of
1253: LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1254:
1255: 25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1256: recursions. This makes a big different to stack usage for some patterns.
1257:
1258: 26. If a subpattern containing a named recursion or subroutine reference such
1259: as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1260: the space required for the compiled pattern went wrong and gave too small a
1261: value. Depending on the environment, this could lead to "Failed: internal
1262: error: code overflow at offset 49" or "glibc detected double free or
1263: corruption" errors.
1264:
1265: 27. Applied patches from Google (a) to support the new newline modes and (b) to
1266: advance over multibyte UTF-8 characters in GlobalReplace.
1267:
1268: 28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1269: difference for some implementation of PCRE in some Windows version.
1270:
1271: 29. Added some extra testing facilities to pcretest:
1272:
1273: \q<number> in a data line sets the "match limit" value
1274: \Q<number> in a data line sets the "match recursion limt" value
1275: -S <number> sets the stack size, where <number> is in megabytes
1276:
1277: The -S option isn't available for Windows.
1278:
1279:
1280: Version 6.6 06-Feb-06
1281: ---------------------
1282:
1283: 1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1284: in pcreposix.h. I have copied the definition from pcre.h.
1285:
1286: 2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1287: because pcre.h is no longer a built file.
1288:
1289: 3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1290: not normally included in the compiled code.
1291:
1292:
1293: Version 6.5 01-Feb-06
1294: ---------------------
1295:
1296: 1. When using the partial match feature with pcre_dfa_exec(), it was not
1297: anchoring the second and subsequent partial matches at the new starting
1298: point. This could lead to incorrect results. For example, with the pattern
1299: /1234/, partially matching against "123" and then "a4" gave a match.
1300:
1301: 2. Changes to pcregrep:
1302:
1303: (a) All non-match returns from pcre_exec() were being treated as failures
1304: to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1305: error message is output. Some extra information is given for the
1306: PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1307: probably the only errors that are likely to be caused by users (by
1308: specifying a regex that has nested indefinite repeats, for instance).
1309: If there are more than 20 of these errors, pcregrep is abandoned.
1310:
1311: (b) A binary zero was treated as data while matching, but terminated the
1312: output line if it was written out. This has been fixed: binary zeroes
1313: are now no different to any other data bytes.
1314:
1315: (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1316: used to set a locale for matching. The --locale=xxxx long option has
1317: been added (no short equivalent) to specify a locale explicitly on the
1318: pcregrep command, overriding the environment variables.
1319:
1320: (d) When -B was used with -n, some line numbers in the output were one less
1321: than they should have been.
1322:
1323: (e) Added the -o (--only-matching) option.
1324:
1325: (f) If -A or -C was used with -c (count only), some lines of context were
1326: accidentally printed for the final match.
1327:
1328: (g) Added the -H (--with-filename) option.
1329:
1330: (h) The combination of options -rh failed to suppress file names for files
1331: that were found from directory arguments.
1332:
1333: (i) Added the -D (--devices) and -d (--directories) options.
1334:
1335: (j) Added the -F (--fixed-strings) option.
1336:
1337: (k) Allow "-" to be used as a file name for -f as well as for a data file.
1338:
1339: (l) Added the --colo(u)r option.
1340:
1341: (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1342: is not present by default.
1343:
1344: 3. A nasty bug was discovered in the handling of recursive patterns, that is,
1345: items such as (?R) or (?1), when the recursion could match a number of
1346: alternatives. If it matched one of the alternatives, but subsequently,
1347: outside the recursion, there was a failure, the code tried to back up into
1348: the recursion. However, because of the way PCRE is implemented, this is not
1349: possible, and the result was an incorrect result from the match.
1350:
1351: In order to prevent this happening, the specification of recursion has
1352: been changed so that all such subpatterns are automatically treated as
1353: atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1354:
1355: 4. I had overlooked the fact that, in some locales, there are characters for
1356: which isalpha() is true but neither isupper() nor islower() are true. In
1357: the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1358: and ordfeminine) are like this. This affected the treatment of \w and \W
1359: when they appeared in character classes, but not when they appeared outside
1360: a character class. The bit map for "word" characters is now created
1361: separately from the results of isalnum() instead of just taking it from the
1362: upper, lower, and digit maps. (Plus the underscore character, of course.)
1363:
1364: 5. The above bug also affected the handling of POSIX character classes such as
1365: [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1366: permanent tables. Instead, the bit maps for such a class were previously
1367: created as the appropriate unions of the upper, lower, and digit bitmaps.
1368: Now they are created by subtraction from the [[:word:]] class, which has
1369: its own bitmap.
1370:
1371: 6. The [[:blank:]] character class matches horizontal, but not vertical space.
1372: It is created by subtracting the vertical space characters (\x09, \x0a,
1373: \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1374: subtraction was done in the overall bitmap for a character class, meaning
1375: that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1376: be recognized. This bug has been fixed.
1377:
1378: 7. Patches from the folks at Google:
1379:
1380: (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1381: real life, but is still worth protecting against".
1382:
1383: (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1384: regular expressions".
1385:
1386: (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1387: have it.
1388:
1389: (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1390: "configure" and the latter not, in order to fix a problem somebody had
1391: with compiling the Arg class on HP-UX.
1392:
1393: (e) Improve the error-handling of the C++ wrapper a little bit.
1394:
1395: (f) New tests for checking recursion limiting.
1396:
1397: 8. The pcre_memmove() function, which is used only if the environment does not
1398: have a standard memmove() function (and is therefore rarely compiled),
1399: contained two bugs: (a) use of int instead of size_t, and (b) it was not
1400: returning a result (though PCRE never actually uses the result).
1401:
1402: 9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1403: large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1404: returned instead of calling malloc() with an overflowing number that would
1405: most likely cause subsequent chaos.
1406:
1407: 10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1408:
1409: 11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1410: with this option is matched, the nmatch and pmatch options of regexec() are
1411: ignored.
1412:
1413: 12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1414: provided in case anyone wants to the the POSIX interface with UTF-8
1415: strings.
1416:
1417: 13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1418: C++ linking (needed for some HP-UX environments).
1419:
1420: 14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1421: (unused parameter) and in the pcre_printint() function (omitted "default"
1422: switch label when the default is to do nothing).
1423:
1424: 15. Added some code to make it possible, when PCRE is compiled as a C++
1425: library, to replace subject pointers for pcre_exec() with a smart pointer
1426: class, thus making it possible to process discontinuous strings.
1427:
1428: 16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1429: much the same function. They were added by different people who were trying
1430: to make PCRE easy to compile on non-Unix systems. It has been suggested
1431: that PCRE_EXPORT be abolished now that there is more automatic apparatus
1432: for compiling on Windows systems. I have therefore replaced it with
1433: PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1434: defaults to "extern" for C or "extern C" for C++, which works fine on
1435: Unix-like systems. It is now possible to override the value of PCRE_DATA_
1436: SCOPE with something explicit in config.h. In addition:
1437:
1438: (a) pcreposix.h still had just "extern" instead of either of these macros;
1439: I have replaced it with PCRE_DATA_SCOPE.
1440:
1441: (b) Functions such as _pcre_xclass(), which are internal to the library,
1442: but external in the C sense, all had PCRE_EXPORT in their definitions.
1443: This is apparently wrong for the Windows case, so I have removed it.
1444: (It makes no difference on Unix-like systems.)
1445:
1446: 17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1447: of recursive calls to match(). This is different to MATCH_LIMIT because
1448: that limits the total number of calls to match(), not all of which increase
1449: the depth of recursion. Limiting the recursion depth limits the amount of
1450: stack (or heap if NO_RECURSE is set) that is used. The default can be set
1451: when PCRE is compiled, and changed at run time. A patch from Google adds
1452: this functionality to the C++ interface.
1453:
1454: 18. Changes to the handling of Unicode character properties:
1455:
1456: (a) Updated the table to Unicode 4.1.0.
1457:
1458: (b) Recognize characters that are not in the table as "Cn" (undefined).
1459:
1460: (c) I revised the way the table is implemented to a much improved format
1461: which includes recognition of ranges. It now supports the ranges that
1462: are defined in UnicodeData.txt, and it also amalgamates other
1463: characters into ranges. This has reduced the number of entries in the
1464: table from around 16,000 to around 3,000, thus reducing its size
1465: considerably. I realized I did not need to use a tree structure after
1466: all - a binary chop search is just as efficient. Having reduced the
1467: number of entries, I extended their size from 6 bytes to 8 bytes to
1468: allow for more data.
1469:
1470: (d) Added support for Unicode script names via properties such as \p{Han}.
1471:
1472: 19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1473: matching that character.
1474:
1475: 20. When matching a repeated Unicode property with a minimum greater than zero,
1476: (for example \pL{2,}), PCRE could look past the end of the subject if it
1477: reached it while seeking the minimum number of characters. This could
1478: happen only if some of the characters were more than one byte long, because
1479: there is a check for at least the minimum number of bytes.
1480:
1481: 21. Refactored the implementation of \p and \P so as to be more general, to
1482: allow for more different types of property in future. This has changed the
1483: compiled form incompatibly. Anybody with saved compiled patterns that use
1484: \p or \P will have to recompile them.
1485:
1486: 22. Added "Any" and "L&" to the supported property types.
1487:
1488: 23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1489: but give a compile time error if the value is greater than 0xff.
1490:
1491: 24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1492: accidentally not being installed or uninstalled.
1493:
1494: 25. The pcre.h file was built from pcre.h.in, but the only changes that were
1495: made were to insert the current release number. This seemed silly, because
1496: it made things harder for people building PCRE on systems that don't run
1497: "configure". I have turned pcre.h into a distributed file, no longer built
1498: by "configure", with the version identification directly included. There is
1499: no longer a pcre.h.in file.
1500:
1501: However, this change necessitated a change to the pcre-config script as
1502: well. It is built from pcre-config.in, and one of the substitutions was the
1503: release number. I have updated configure.ac so that ./configure now finds
1504: the release number by grepping pcre.h.
1505:
1506: 26. Added the ability to run the tests under valgrind.
1507:
1508:
1509: Version 6.4 05-Sep-05
1510: ---------------------
1511:
1512: 1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1513: "--" to be printed when multiple files were scanned, even when none of the
1514: -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1515: consider it to be a bug, and have restored the previous behaviour.
1516:
1517: 2. A couple of code tidies to get rid of compiler warnings.
1518:
1519: 3. The pcretest program used to cheat by referring to symbols in the library
1520: whose names begin with _pcre_. These are internal symbols that are not
1521: really supposed to be visible externally, and in some environments it is
1522: possible to suppress them. The cheating is now confined to including
1523: certain files from the library's source, which is a bit cleaner.
1524:
1525: 4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1526: file's purpose clearer.
1527:
1528: 5. Reorganized pcre_ucp_findchar().
1529:
1530:
1531: Version 6.3 15-Aug-05
1532: ---------------------
1533:
1534: 1. The file libpcre.pc.in did not have general read permission in the tarball.
1535:
1536: 2. There were some problems when building without C++ support:
1537:
1538: (a) If C++ support was not built, "make install" and "make test" still
1539: tried to test it.
1540:
1541: (b) There were problems when the value of CXX was explicitly set. Some
1542: changes have been made to try to fix these, and ...
1543:
1544: (c) --disable-cpp can now be used to explicitly disable C++ support.
1545:
1546: (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1547: backslash in a target when C++ was disabled. This confuses some
1548: versions of "make", apparently. Using an intermediate variable solves
1549: this. (Same for CPP_LOBJ.)
1550:
1551: 3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1552: (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1553: necessary on certain architectures.
1554:
1555: 4. Added a setting of -export-symbols-regex to the link command to remove
1556: those symbols that are exported in the C sense, but actually are local
1557: within the library, and not documented. Their names all begin with
1558: "_pcre_". This is not a perfect job, because (a) we have to except some
1559: symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1560: available (and never for static libraries). I have made a note to try to
1561: find a way round (a) in the future.
1562:
1563:
1564: Version 6.2 01-Aug-05
1565: ---------------------
1566:
1567: 1. There was no test for integer overflow of quantifier values. A construction
1568: such as {1111111111111111} would give undefined results. What is worse, if
1569: a minimum quantifier for a parenthesized subpattern overflowed and became
1570: negative, the calculation of the memory size went wrong. This could have
1571: led to memory overwriting.
1572:
1573: 2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1574:
1575: 3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1576: operating environments where this matters.
1577:
1578: 4. Applied Giuseppe Maxia's patch to add additional features for controlling
1579: PCRE options from within the C++ wrapper.
1580:
1581: 5. Named capturing subpatterns were not being correctly counted when a pattern
1582: was compiled. This caused two problems: (a) If there were more than 100
1583: such subpatterns, the calculation of the memory needed for the whole
1584: compiled pattern went wrong, leading to an overflow error. (b) Numerical
1585: back references of the form \12, where the number was greater than 9, were
1586: not recognized as back references, even though there were sufficient
1587: previous subpatterns.
1588:
1589: 6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1590: versions of gcc, e.g. 2.95.4.
1591:
1592:
1593: Version 6.1 21-Jun-05
1594: ---------------------
1595:
1596: 1. There was one reference to the variable "posix" in pcretest.c that was not
1597: surrounded by "#if !defined NOPOSIX".
1598:
1599: 2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1600: the cross-check on the old pcre_info() function, for the benefit of the
1601: cut-down version of PCRE that is currently imported into Exim.
1602:
1603: 3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1604: allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1605: patterns that start (?i) but I don't think that matters. The use of (?i) is
1606: just an example; this all applies to the other options as well.
1607:
1608: 4. Since libtool seems to echo the compile commands it is issuing, the output
1609: from "make" can be reduced a bit by putting "@" in front of each libtool
1610: compile command.
1611:
1612: 5. Patch from the folks at Google for configure.in to be a bit more thorough
1613: in checking for a suitable C++ installation before trying to compile the
1614: C++ stuff. This should fix a reported problem when a compiler was present,
1615: but no suitable headers.
1616:
1617: 6. The man pages all had just "PCRE" as their title. I have changed them to
1618: be the relevant file name. I have also arranged that these names are
1619: retained in the file doc/pcre.txt, which is a concatenation in text format
1620: of all the man pages except the little individual ones for each function.
1621:
1622: 7. The NON-UNIX-USE file had not been updated for the different set of source
1623: files that come with release 6. I also added a few comments about the C++
1624: wrapper.
1625:
1626:
1627: Version 6.0 07-Jun-05
1628: ---------------------
1629:
1630: 1. Some minor internal re-organization to help with my DFA experiments.
1631:
1632: 2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1633: didn't matter for the library itself when fully configured, but did matter
1634: when compiling without UCP support, or within Exim, where the ucp files are
1635: not imported.
1636:
1637: 3. Refactoring of the library code to split up the various functions into
1638: different source modules. The addition of the new DFA matching code (see
1639: below) to a single monolithic source would have made it really too
1640: unwieldy, quite apart from causing all the code to be include in a
1641: statically linked application, when only some functions are used. This is
1642: relevant even without the DFA addition now that patterns can be compiled in
1643: one application and matched in another.
1644:
1645: The downside of splitting up is that there have to be some external
1646: functions and data tables that are used internally in different modules of
1647: the library but which are not part of the API. These have all had their
1648: names changed to start with "_pcre_" so that they are unlikely to clash
1649: with other external names.
1650:
1651: 4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1652: a different (DFA) algorithm. Although it is slower than the original
1653: function, it does have some advantages for certain types of matching
1654: problem.
1655:
1656: 5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1657: including restarting after a partial match.
1658:
1659: 6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1660: defined when compiling for Windows was sent to me. I have put it into the
1661: code, though I have no means of testing or verifying it.
1662:
1663: 7. Added the pcre_refcount() auxiliary function.
1664:
1665: 8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1666: match before or at the first newline in the subject string. In pcretest,
1667: the /f option on a pattern can be used to set this.
1668:
1669: 9. A repeated \w when used in UTF-8 mode with characters greater than 256
1670: would behave wrongly. This has been present in PCRE since release 4.0.
1671:
1672: 10. A number of changes to the pcregrep command:
1673:
1674: (a) Refactored how -x works; insert ^(...)$ instead of setting
1675: PCRE_ANCHORED and checking the length, in preparation for adding
1676: something similar for -w.
1677:
1678: (b) Added the -w (match as a word) option.
1679:
1680: (c) Refactored the way lines are read and buffered so as to have more
1681: than one at a time available.
1682:
1683: (d) Implemented a pcregrep test script.
1684:
1685: (e) Added the -M (multiline match) option. This allows patterns to match
1686: over several lines of the subject. The buffering ensures that at least
1687: 8K, or the rest of the document (whichever is the shorter) is available
1688: for matching (and similarly the previous 8K for lookbehind assertions).
1689:
1690: (f) Changed the --help output so that it now says
1691:
1692: -w, --word-regex(p)
1693:
1694: instead of two lines, one with "regex" and the other with "regexp"
1695: because that confused at least one person since the short forms are the
1696: same. (This required a bit of code, as the output is generated
1697: automatically from a table. It wasn't just a text change.)
1698:
1699: (g) -- can be used to terminate pcregrep options if the next thing isn't an
1700: option but starts with a hyphen. Could be a pattern or a path name
1701: starting with a hyphen, for instance.
1702:
1703: (h) "-" can be given as a file name to represent stdin.
1704:
1705: (i) When file names are being printed, "(standard input)" is used for
1706: the standard input, for compatibility with GNU grep. Previously
1707: "<stdin>" was used.
1708:
1709: (j) The option --label=xxx can be used to supply a name to be used for
1710: stdin when file names are being printed. There is no short form.
1711:
1712: (k) Re-factored the options decoding logic because we are going to add
1713: two more options that take data. Such options can now be given in four
1714: different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1715:
1716: (l) Added the -A, -B, and -C options for requesting that lines of context
1717: around matches be printed.
1718:
1719: (m) Added the -L option to print the names of files that do not contain
1720: any matching lines, that is, the complement of -l.
1721:
1722: (n) The return code is 2 if any file cannot be opened, but pcregrep does
1723: continue to scan other files.
1724:
1725: (o) The -s option was incorrectly implemented. For compatibility with other
1726: greps, it now suppresses the error message for a non-existent or non-
1727: accessible file (but not the return code). There is a new option called
1728: -q that suppresses the output of matching lines, which was what -s was
1729: previously doing.
1730:
1731: (p) Added --include and --exclude options to specify files for inclusion
1732: and exclusion when recursing.
1733:
1734: 11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1735: Hopefully, it now does.
1736:
1737: 12. Missing cast in pcre_study().
1738:
1739: 13. Added an "uninstall" target to the makefile.
1740:
1741: 14. Replaced "extern" in the function prototypes in Makefile.in with
1742: "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1743: world, but is set differently for Windows.
1744:
1745: 15. Added a second compiling function called pcre_compile2(). The only
1746: difference is that it has an extra argument, which is a pointer to an
1747: integer error code. When there is a compile-time failure, this is set
1748: non-zero, in addition to the error test pointer being set to point to an
1749: error message. The new argument may be NULL if no error number is required
1750: (but then you may as well call pcre_compile(), which is now just a
1751: wrapper). This facility is provided because some applications need a
1752: numeric error indication, but it has also enabled me to tidy up the way
1753: compile-time errors are handled in the POSIX wrapper.
1754:
1755: 16. Added VPATH=.libs to the makefile; this should help when building with one
1756: prefix path and installing with another. (Or so I'm told by someone who
1757: knows more about this stuff than I do.)
1758:
1759: 17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1760: passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1761: match everything, including newlines. This is not POSIX-compatible, but
1762: somebody wanted the feature. From pcretest it can be activated by using
1763: both the P and the s flags.
1764:
1765: 18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1766:
1767: 19. libpcre.pc was being incorrectly installed as executable.
1768:
1769: 20. A couple of places in pcretest check for end-of-line by looking for '\n';
1770: it now also looks for '\r' so that it will work unmodified on Windows.
1771:
1772: 21. Added Google's contributed C++ wrapper to the distribution.
1773:
1774: 22. Added some untidy missing memory free() calls in pcretest, to keep
1775: Electric Fence happy when testing.
1776:
1777:
1778:
1779: Version 5.0 13-Sep-04
1780: ---------------------
1781:
1782: 1. Internal change: literal characters are no longer packed up into items
1783: containing multiple characters in a single byte-string. Each character
1784: is now matched using a separate opcode. However, there may be more than one
1785: byte in the character in UTF-8 mode.
1786:
1787: 2. The pcre_callout_block structure has two new fields: pattern_position and
1788: next_item_length. These contain the offset in the pattern to the next match
1789: item, and its length, respectively.
1790:
1791: 3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1792: insertion of callouts before each pattern item. Added the /C option to
1793: pcretest to make use of this.
1794:
1795: 4. On the advice of a Windows user, the lines
1796:
1797: #if defined(_WIN32) || defined(WIN32)
1798: _setmode( _fileno( stdout ), 0x8000 );
1799: #endif /* defined(_WIN32) || defined(WIN32) */
1800:
1801: have been added to the source of pcretest. This apparently does useful
1802: magic in relation to line terminators.
1803:
1804: 5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1805: for the benefit of those environments where the "b" makes a difference.
1806:
1807: 6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1808: to know about it. I have put a hack into configure.in that adds in code
1809: to set GCC=yes if CC=icc. This seems to end up at a point in the
1810: generated configure script that is early enough to affect the setting of
1811: compiler options, which is what is needed, but I have no means of testing
1812: whether it really works. (The user who reported this had patched the
1813: generated configure script, which of course I cannot do.)
1814:
1815: LATER: After change 22 below (new libtool files), the configure script
1816: seems to know about icc (and also ecc). Therefore, I have commented out
1817: this hack in configure.in.
1818:
1819: 7. Added support for pkg-config (2 patches were sent in).
1820:
1821: 8. Negated POSIX character classes that used a combination of internal tables
1822: were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1823: [[:^ascii]]. Typically, they would match almost any characters. The other
1824: POSIX classes were not broken in this way.
1825:
1826: 9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1827: to find the match, as PCRE was deluded into thinking that the match had to
1828: start at the start point or following a newline. The same bug applied to
1829: patterns with negative forward assertions or any backward assertions
1830: preceding ".*" at the start, unless the pattern required a fixed first
1831: character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1832:
1833: 10. In UTF-8 mode, when moving forwards in the subject after a failed match
1834: starting at the last subject character, bytes beyond the end of the subject
1835: string were read.
1836:
1837: 11. Renamed the variable "class" as "classbits" to make life easier for C++
1838: users. (Previously there was a macro definition, but it apparently wasn't
1839: enough.)
1840:
1841: 12. Added the new field "tables" to the extra data so that tables can be passed
1842: in at exec time, or the internal tables can be re-selected. This allows
1843: a compiled regex to be saved and re-used at a later time by a different
1844: program that might have everything at different addresses.
1845:
1846: 13. Modified the pcre-config script so that, when run on Solaris, it shows a
1847: -R library as well as a -L library.
1848:
1849: 14. The debugging options of pcretest (-d on the command line or D on a
1850: pattern) showed incorrect output for anything following an extended class
1851: that contained multibyte characters and which was followed by a quantifier.
1852:
1853: 15. Added optional support for general category Unicode character properties
1854: via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1855: support. It adds about 90K to the size of the library. The meanings of the
1856: inbuilt class escapes such as \d and \s have NOT been changed.
1857:
1858: 16. Updated pcredemo.c to include calls to free() to release the memory for the
1859: compiled pattern.
1860:
1861: 17. The generated file chartables.c was being created in the source directory
1862: instead of in the building directory. This caused the build to fail if the
1863: source directory was different from the building directory, and was
1864: read-only.
1865:
1866: 18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1867: file. No doubt somebody will tell me if they don't make sense... Also added
1868: Dan Mooney's comments about building on OpenVMS.
1869:
1870: 19. Added support for partial matching via the PCRE_PARTIAL option for
1871: pcre_exec() and the \P data escape in pcretest.
1872:
1873: 20. Extended pcretest with 3 new pattern features:
1874:
1875: (i) A pattern option of the form ">rest-of-line" causes pcretest to
1876: write the compiled pattern to the file whose name is "rest-of-line".
1877: This is a straight binary dump of the data, with the saved pointer to
1878: the character tables forced to be NULL. The study data, if any, is
1879: written too. After writing, pcretest reads a new pattern.
1880:
1881: (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1882: compiled pattern from the given file. There must not be any
1883: occurrences of "<" in the file name (pretty unlikely); if there are,
1884: pcretest will instead treat the initial "<" as a pattern delimiter.
1885: After reading in the pattern, pcretest goes on to read data lines as
1886: usual.
1887:
1888: (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1889: and 16-bit fields in a compiled pattern, to simulate a pattern that
1890: was compiled on a host of opposite endianness.
1891:
1892: 21. The pcre-exec() function can now cope with patterns that were compiled on
1893: hosts of opposite endianness, with this restriction:
1894:
1895: As for any compiled expression that is saved and used later, the tables
1896: pointer field cannot be preserved; the extra_data field in the arguments
1897: to pcre_exec() should be used to pass in a tables address if a value
1898: other than the default internal tables were used at compile time.
1899:
1900: 22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1901: now diagnosed as an error. Previously, most of the time, a negative number
1902: would have been treated as zero, but if in addition "ovector" was passed as
1903: NULL, a crash could occur.
1904:
1905: 23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1906: new versions from the libtool 1.5 distribution (the last one is a copy of
1907: a file called libtool.m4). This seems to have fixed the need to patch
1908: "configure" to support Darwin 1.3 (which I used to do). However, I still
1909: had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1910: workstation).
1911:
1912: 24. Changed the PCRE licence to be the more standard "BSD" licence.
1913:
1914:
1915: Version 4.5 01-Dec-03
1916: ---------------------
1917:
1918: 1. There has been some re-arrangement of the code for the match() function so
1919: that it can be compiled in a version that does not call itself recursively.
1920: Instead, it keeps those local variables that need separate instances for
1921: each "recursion" in a frame on the heap, and gets/frees frames whenever it
1922: needs to "recurse". Keeping track of where control must go is done by means
1923: of setjmp/longjmp. The whole thing is implemented by a set of macros that
1924: hide most of the details from the main code, and operates only if
1925: NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1926: "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1927: operating.
1928:
1929: To make it easier for callers to provide specially tailored get/free
1930: functions for this usage, two new functions, pcre_stack_malloc, and
1931: pcre_stack_free, are used. They are always called in strict stacking order,
1932: and the size of block requested is always the same.
1933:
1934: The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1935: PCRE has been compiled to use the stack or the heap for recursion. The
1936: -C option of pcretest uses this to show which version is compiled.
1937:
1938: A new data escape \S, is added to pcretest; it causes the amounts of store
1939: obtained and freed by both kinds of malloc/free at match time to be added
1940: to the output.
1941:
1942: 2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1943: what's available on my current Linux desktop machine.
1944:
1945: 3. When matching a UTF-8 string, the test for a valid string at the start has
1946: been extended. If start_offset is not zero, PCRE now checks that it points
1947: to a byte that is the start of a UTF-8 character. If not, it returns
1948: PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1949: this is necessary because there may be backward assertions in the pattern.
1950: When matching the same subject several times, it may save resources to use
1951: PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1952:
1953: 4. The code for checking the validity of UTF-8 strings has been tightened so
1954: that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1955: containing "overlong sequences".
1956:
1957: 5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1958: I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1959: should have been "&", but it just so happened that all the cases this let
1960: through by mistake were picked up later in the function.
1961:
1962: 6. I had used a variable called "isblank" - this is a C99 function, causing
1963: some compilers to warn. To avoid this, I renamed it (as "blankclass").
1964:
1965: 7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1966: prompting; (b) run "./pcretest /dev/null" at the start of the test script
1967: so the version is shown; (c) stop "make test" echoing "./RunTest".
1968:
1969: 8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1970:
1971: 9. The prototype for memmove() for systems that don't have it was using
1972: size_t, but the inclusion of the header that defines size_t was later. I've
1973: moved the #includes for the C headers earlier to avoid this.
1974:
1975: 10. Added some adjustments to the code to make it easier to compiler on certain
1976: special systems:
1977:
1978: (a) Some "const" qualifiers were missing.
1979: (b) Added the macro EXPORT before all exported functions; by default this
1980: is defined to be empty.
1981: (c) Changed the dftables auxiliary program (that builds chartables.c) so
1982: that it reads its output file name as an argument instead of writing
1983: to the standard output and assuming this can be redirected.
1984:
1985: 11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1986: class containing characters with values greater than 255, PCRE compilation
1987: went into a loop.
1988:
1989: 12. A recursive reference to a subpattern that was within another subpattern
1990: that had a minimum quantifier of zero caused PCRE to crash. For example,
1991: (x(y(?2))z)? provoked this bug with a subject that got as far as the
1992: recursion. If the recursively-called subpattern itself had a zero repeat,
1993: that was OK.
1994:
1995: 13. In pcretest, the buffer for reading a data line was set at 30K, but the
1996: buffer into which it was copied (for escape processing) was still set at
1997: 1024, so long lines caused crashes.
1998:
1999: 14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2000: "internal error: code overflow...". This applied to any character class
2001: that was followed by a possessive quantifier.
2002:
2003: 15. Modified the Makefile to add libpcre.la as a prerequisite for
2004: libpcreposix.la because I was told this is needed for a parallel build to
2005: work.
2006:
2007: 16. If a pattern that contained .* following optional items at the start was
2008: studied, the wrong optimizing data was generated, leading to matching
2009: errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2010: matching string must start with a or b or c. The correct conclusion for
2011: this pattern is that a match can start with any character.
2012:
2013:
2014: Version 4.4 13-Aug-03
2015: ---------------------
2016:
2017: 1. In UTF-8 mode, a character class containing characters with values between
2018: 127 and 255 was not handled correctly if the compiled pattern was studied.
2019: In fixing this, I have also improved the studying algorithm for such
2020: classes (slightly).
2021:
2022: 2. Three internal functions had redundant arguments passed to them. Removal
2023: might give a very teeny performance improvement.
2024:
2025: 3. Documentation bug: the value of the capture_top field in a callout is *one
2026: more than* the number of the hightest numbered captured substring.
2027:
2028: 4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2029: in incorrectly linking with a previously installed version. They now link
2030: explicitly with libpcre.la.
2031:
2032: 5. configure.in no longer needs to recognize Cygwin specially.
2033:
2034: 6. A problem in pcre.in for Windows platforms is fixed.
2035:
2036: 7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2037: pcretest, it used to include the size of the study block as part of its
2038: output. Unfortunately, the structure contains a field that has a different
2039: size on different hardware architectures. This meant that the tests that
2040: showed this size failed. As the block is currently always of a fixed size,
2041: this information isn't actually particularly useful in pcretest output, so
2042: I have just removed it.
2043:
2044: 8. Three pre-processor statements accidentally did not start in column 1.
2045: Sadly, there are *still* compilers around that complain, even though
2046: standard C has not required this for well over a decade. Sigh.
2047:
2048: 9. In pcretest, the code for checking callouts passed small integers in the
2049: callout_data field, which is a void * field. However, some picky compilers
2050: complained about the casts involved for this on 64-bit systems. Now
2051: pcretest passes the address of the small integer instead, which should get
2052: rid of the warnings.
2053:
2054: 10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2055: both compile and run time, and gives an error if an invalid UTF-8 sequence
2056: is found. There is a option for disabling this check in cases where the
2057: string is known to be correct and/or the maximum performance is wanted.
2058:
2059: 11. In response to a bug report, I changed one line in Makefile.in from
2060:
2061: -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2062: to
2063: -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2064:
2065: to look similar to other lines, but I have no way of telling whether this
2066: is the right thing to do, as I do not use Windows. No doubt I'll get told
2067: if it's wrong...
2068:
2069:
2070: Version 4.3 21-May-03
2071: ---------------------
2072:
2073: 1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2074: Makefile.
2075:
2076: 2. Some refactoring to improve the quality of the code:
2077:
2078: (i) The utf8_table... variables are now declared "const".
2079:
2080: (ii) The code for \cx, which used the "case flipping" table to upper case
2081: lower case letters, now just substracts 32. This is ASCII-specific,
2082: but the whole concept of \cx is ASCII-specific, so it seems
2083: reasonable.
2084:
2085: (iii) PCRE was using its character types table to recognize decimal and
2086: hexadecimal digits in the pattern. This is silly, because it handles
2087: only 0-9, a-f, and A-F, but the character types table is locale-
2088: specific, which means strange things might happen. A private
2089: table is now used for this - though it costs 256 bytes, a table is
2090: much faster than multiple explicit tests. Of course, the standard
2091: character types table is still used for matching digits in subject
2092: strings against \d.
2093:
2094: (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2095: ending in _t are). So I've renamed it as ESC_tee.
2096:
2097: 3. The first argument for regexec() in the POSIX wrapper should have been
2098: defined as "const".
2099:
2100: 4. Changed pcretest to use malloc() for its buffers so that they can be
2101: Electric Fenced for debugging.
2102:
2103: 5. There were several places in the code where, in UTF-8 mode, PCRE would try
2104: to read one or more bytes before the start of the subject string. Often this
2105: had no effect on PCRE's behaviour, but in some circumstances it could
2106: provoke a segmentation fault.
2107:
2108: 6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2109: to try to read one or more bytes before the start of the subject string.
2110:
2111: 7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2112: UTF-8 support could misbehave in various ways if the subject string
2113: contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2114: area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2115: back over UTF-8 characters.)
2116:
2117:
2118: Version 4.2 14-Apr-03
2119: ---------------------
2120:
2121: 1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2122:
2123: 2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2124: [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2125: [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2126: [WIN_PREFIX]: new variable, "cyg" for Cygwin
2127: * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2128: and BUILD_EXEEXT
2129: Note: automatic setting of the BUILD variables is not yet working
2130: set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2131: compile-time but not at link-time
2132: [LINK]: use for linking executables only
2133: make different versions for Windows and non-Windows
2134: [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2135: libraries
2136: [LINK_FOR_BUILD]: new variable
2137: [OBJEXT]: use throughout
2138: [EXEEXT]: use throughout
2139: <winshared>: new target
2140: <wininstall>: new target
2141: <dftables.o>: use native compiler
2142: <dftables>: use native linker
2143: <install>: handle Windows platform correctly
2144: <clean>: ditto
2145: <check>: ditto
2146: copy DLL to top builddir before testing
2147:
2148: As part of these changes, -no-undefined was removed again. This was reported
2149: to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2150: in any case.
2151:
2152: 3. Some tidies to get rid of compiler warnings:
2153:
2154: . In the match_data structure, match_limit was an unsigned long int, whereas
2155: match_call_count was an int. I've made them both unsigned long ints.
2156:
2157: . In pcretest the fact that a const uschar * doesn't automatically cast to
2158: a void * provoked a warning.
2159:
2160: . Turning on some more compiler warnings threw up some "shadow" variables
2161: and a few more missing casts.
2162:
2163: 4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2164: option, a class that contained a single character with a value between 128
2165: and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2166:
2167: 5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2168: option, a class that contained several characters, but with at least one
2169: whose value was between 128 and 255 caused PCRE to crash.
2170:
2171:
2172: Version 4.1 12-Mar-03
2173: ---------------------
2174:
2175: 1. Compiling with gcc -pedantic found a couple of places where casts were
2176: needed, and a string in dftables.c that was longer than standard compilers are
2177: required to support.
2178:
2179: 2. Compiling with Sun's compiler found a few more places where the code could
2180: be tidied up in order to avoid warnings.
2181:
2182: 3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2183: first of these names is deprecated in the latest Autoconf in favour of the name
2184: CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2185: compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2186: analogy I have changed it to CFLAGS_FOR_BUILD.
2187:
2188: 4. Added -no-undefined to the linking command in the Makefile, because this is
2189: apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2190: linking step for the pcreposix library.
2191:
2192: 5. PCRE was failing to diagnose the case of two named groups with the same
2193: name.
2194:
2195: 6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2196: literal character that is needed in the subject for a match, and scans along to
2197: ensure that it is present before embarking on the full matching process. This
2198: saves time in cases of nested unlimited repeats that are never going to match.
2199: Problem: the scan can take a lot of time if the subject is very long (e.g.
2200: megabytes), thus penalizing straightforward matches. It is now done only if the
2201: amount of subject to be scanned is less than 1000 bytes.
2202:
2203: 7. A lesser problem with the same optimization is that it was recording the
2204: first character of an anchored pattern as "needed", thus provoking a search
2205: right along the subject, even when the first match of the pattern was going to
2206: fail. The "needed" character is now not set for anchored patterns, unless it
2207: follows something in the pattern that is of non-fixed length. Thus, it still
2208: fulfils its original purpose of finding quick non-matches in cases of nested
2209: unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2210:
2211:
2212: Version 4.0 17-Feb-03
2213: ---------------------
2214:
2215: 1. If a comment in an extended regex that started immediately after a meta-item
2216: extended to the end of string, PCRE compiled incorrect data. This could lead to
2217: all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2218:
2219: 2. Moved to autoconf 2.53 and libtool 1.4.2.
2220:
2221: 3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2222: the special perltest8 script is no longer needed - all the tests can be run
2223: from a single perltest script.
2224:
2225: 4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2226: by \s. It has now been removed in PCRE. This means it isn't recognized as
2227: whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2228: class [:space:] *does* include VT, thereby creating a mess.
2229:
2230: 5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2231: space and tab.
2232:
2233: 6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2234: its new features into the main test script, reducing the number of scripts.
2235:
2236: 7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2237: were backward compatible, and made the (?i) apply to the whole pattern, as if
2238: /i were given. Now it behaves more logically, and applies the option setting
2239: only to what follows. PCRE has been changed to follow suit. However, if it
2240: finds options settings right at the start of the pattern, it extracts them into
2241: the global options, as before. Thus, they show up in the info data.
2242:
2243: 8. Added support for the \Q...\E escape sequence. Characters in between are
2244: treated as literals. This is slightly different from Perl in that $ and @ are
2245: also handled as literals inside the quotes. In Perl, they will cause variable
2246: interpolation. Note the following examples:
2247:
2248: Pattern PCRE matches Perl matches
2249:
2250: \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2251: \Qabc\$xyz\E abc\$xyz abc\$xyz
2252: \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2253:
2254: For compatibility with Perl, \Q...\E sequences are recognized inside character
2255: classes as well as outside them.
2256:
2257: 9. Re-organized 3 code statements in pcretest to avoid "overflow in
2258: floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2259: (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2260: signed/unsigned warnings.
2261:
2262: 10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2263: option for pcretest, so I've replaced it by a simple function that does just
2264: that job.
2265:
2266: 11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2267: "pcregrep -".
2268:
2269: 12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2270: Java package. This provides some syntactic sugar for simple cases of what my
2271: documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2272: as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2273: item, you can use this simplified notation. Note that only makes sense with
2274: greedy quantifiers. Consequently, the use of the possessive quantifier forces
2275: greediness, whatever the setting of the PCRE_UNGREEDY option.
2276:
2277: 13. A change of greediness default within a pattern was not taking effect at
2278: the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2279: subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2280: was abstracted outside.
2281:
2282: 14. PCRE now supports the \G assertion. It is true when the current matching
2283: position is at the start point of the match. This differs from \A when the
2284: starting offset is non-zero. Used with the /g option of pcretest (or similar
2285: code), it works in the same way as it does for Perl's /g option. If all
2286: alternatives of a regex begin with \G, the expression is anchored to the start
2287: match position, and the "anchored" flag is set in the compiled expression.
2288:
2289: 15. Some bugs concerning the handling of certain option changes within patterns
2290: have been fixed. These applied to options other than (?ims). For example,
2291: "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2292: been the other way round. Some of this was related to change 7 above.
2293:
2294: 16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2295: features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2296: and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2297: POSIX classes only within a class (e.g. /[[:alpha:]]/).
2298:
2299: 17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2300: mode. Unlike ".", it always matches newline, whatever the setting of
2301: PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2302: assertions. Perl allows it, but it doesn't (in general) work because it can't
2303: calculate the length of the lookbehind. At least, that's the case for Perl
2304: 5.8.0 - I've been told they are going to document that it doesn't work in
2305: future.
2306:
2307: 18. Added an error diagnosis for escapes that PCRE does not support: these are
2308: \L, \l, \N, \P, \p, \U, \u, and \X.
2309:
2310: 19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2311: reading past the end of the pattern in cases such as /[abcd/.
2312:
2313: 20. PCRE was getting more memory than necessary for patterns with classes that
2314: contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2315:
2316: 21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2317: compiling PCRE for use with Virtual Pascal.
2318:
2319: 22. Small fix to the Makefile to make it work properly if the build is done
2320: outside the source tree.
2321:
2322: 23. Added a new extension: a condition to go with recursion. If a conditional
2323: subpattern starts with (?(R) the "true" branch is used if recursion has
2324: happened, whereas the "false" branch is used only at the top level.
2325:
2326: 24. When there was a very long string of literal characters (over 255 bytes
2327: without UTF support, over 250 bytes with UTF support), the computation of how
2328: much memory was required could be incorrect, leading to segfaults or other
2329: strange effects.
2330:
2331: 25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2332: start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2333: there was a subsequent back reference to those brackets. This meant that, for
2334: example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2335: possible to check for precisely this case. All we can do is abandon the
2336: optimization if .* occurs inside capturing brackets when there are any back
2337: references whatsoever. (See below for a better fix that came later.)
2338:
2339: 26. The handling of the optimization for finding the first character of a
2340: non-anchored pattern, and for finding a character that is required later in the
2341: match were failing in some cases. This didn't break the matching; it just
2342: failed to optimize when it could. The way this is done has been re-implemented.
2343:
2344: 27. Fixed typo in error message for invalid (?R item (it said "(?p").
2345:
2346: 28. Added a new feature that provides some of the functionality that Perl
2347: provides with (?{...}). The facility is termed a "callout". The way it is done
2348: in PCRE is for the caller to provide an optional function, by setting
2349: pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2350: global variable. By default it is unset, which disables all calling out. To get
2351: the function called, the regex must include (?C) at appropriate points. This
2352: is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2353: This provides a means of identifying different callout points. When PCRE
2354: reaches such a point in the regex, if pcre_callout has been set, the external
2355: function is called. It is provided with data in a structure called
2356: pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2357: matching continues; if it returns a non-zero value, the match at the current
2358: point fails. However, backtracking will occur if possible. [This was changed
2359: later and other features added - see item 49 below.]
2360:
2361: 29. pcretest is upgraded to test the callout functionality. It provides a
2362: callout function that displays information. By default, it shows the start of
2363: the match and the current position in the text. There are some new data escapes
2364: to vary what happens:
2365:
2366: \C+ in addition, show current contents of captured substrings
2367: \C- do not supply a callout function
2368: \C!n return 1 when callout number n is reached
2369: \C!n!m return 1 when callout number n is reached for the mth time
2370:
2371: 30. If pcregrep was called with the -l option and just a single file name, it
2372: output "<stdin>" if a match was found, instead of the file name.
2373:
2374: 31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2375: slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2376: pcre_exec(). This saves a malloc/free per call. The default value of
2377: POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2378: when configuring.
2379:
2380: 32. The default maximum size of a compiled pattern is 64K. There have been a
2381: few cases of people hitting this limit. The code now uses macros to handle the
2382: storing of links as offsets within the compiled pattern. It defaults to 2-byte
2383: links, but this can be changed to 3 or 4 bytes by --with-link-size when
2384: configuring. Tests 2 and 5 work only with 2-byte links because they output
2385: debugging information about compiled patterns.
2386:
2387: 33. Internal code re-arrangements:
2388:
2389: (a) Moved the debugging function for printing out a compiled regex into
2390: its own source file (printint.c) and used #include to pull it into
2391: pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2392: separate copies.
2393:
2394: (b) Defined the list of op-code names for debugging as a macro in
2395: internal.h so that it is next to the definition of the opcodes.
2396:
2397: (c) Defined a table of op-code lengths for simpler skipping along compiled
2398: code. This is again a macro in internal.h so that it is next to the
2399: definition of the opcodes.
2400:
2401: 34. Added support for recursive calls to individual subpatterns, along the
2402: lines of Robin Houston's patch (but implemented somewhat differently).
2403:
2404: 35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2405: allow it to read and process whole directories in Win32. This code was
2406: contributed by Lionel Fourquaux; it has not been tested by me.
2407:
2408: 36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2409: used to name a group. Names consist of alphanumerics and underscores, and must
2410: be unique. Back references use the syntax (?P=name) and recursive calls use
2411: (?P>name) which is a PCRE extension to the Python extension. Groups still have
2412: numbers. The function pcre_fullinfo() can be used after compilation to extract
2413: a name/number map. There are three relevant calls:
2414:
2415: PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2416: PCRE_INFO_NAMECOUNT yields the number of entries
2417: PCRE_INFO_NAMETABLE yields a pointer to the map.
2418:
2419: The map is a vector of fixed-size entries. The size of each entry depends on
2420: the length of the longest name used. The first two bytes of each entry are the
2421: group number, most significant byte first. There follows the corresponding
2422: name, zero terminated. The names are in alphabetical order.
2423:
2424: 37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2425: case instead of 255. Making it the same both with and without UTF-8 support
2426: means that the same test output works with both.
2427:
2428: 38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2429: calling malloc() with a zero argument.
2430:
2431: 39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2432: optimization. I've improved things by keeping a bitmap of backreferences with
2433: numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2434: fact referenced, the optimization can be applied. It is unlikely that a
2435: relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2436: the match to follow \n) will appear inside brackets with a number greater than
2437: 31, but if it does, any back reference > 31 suppresses the optimization.
2438:
2439: 40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2440: of disabling numbered capturing parentheses. Any opening parenthesis that is
2441: not followed by ? behaves as if it were followed by ?: but named parentheses
2442: can still be used for capturing (and they will acquire numbers in the usual
2443: way).
2444:
2445: 41. Redesigned the return codes from the match() function into yes/no/error so
2446: that errors can be passed back from deep inside the nested calls. A malloc
2447: failure while inside a recursive subpattern call now causes the
2448: PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2449:
2450: 42. It is now possible to set a limit on the number of times the match()
2451: function is called in a call to pcre_exec(). This facility makes it possible to
2452: limit the amount of recursion and backtracking, though not in a directly
2453: obvious way, because the match() function is used in a number of different
2454: circumstances. The count starts from zero for each position in the subject
2455: string (for non-anchored patterns). The default limit is, for compatibility, a
2456: large number, namely 10 000 000. You can change this in two ways:
2457:
2458: (a) When configuring PCRE before making, you can use --with-match-limit=n
2459: to set a default value for the compiled library.
2460:
2461: (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2462: a different value is set. See 45 below.
2463:
2464: If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2465:
2466: 43. Added a new function pcre_config(int, void *) to enable run-time extraction
2467: of things that can be changed at compile time. The first argument specifies
2468: what is wanted and the second points to where the information is to be placed.
2469: The current list of available information is:
2470:
2471: PCRE_CONFIG_UTF8
2472:
2473: The output is an integer that is set to one if UTF-8 support is available;
2474: otherwise it is set to zero.
2475:
2476: PCRE_CONFIG_NEWLINE
2477:
2478: The output is an integer that it set to the value of the code that is used for
2479: newline. It is either LF (10) or CR (13).
2480:
2481: PCRE_CONFIG_LINK_SIZE
2482:
2483: The output is an integer that contains the number of bytes used for internal
2484: linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2485:
2486: PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2487:
2488: The output is an integer that contains the threshold above which the POSIX
2489: interface uses malloc() for output vectors. See item 31 above.
2490:
2491: PCRE_CONFIG_MATCH_LIMIT
2492:
2493: The output is an unsigned integer that contains the default limit of the number
2494: of match() calls in a pcre_exec() execution. See 42 above.
2495:
2496: 44. pcretest has been upgraded by the addition of the -C option. This causes it
2497: to extract all the available output from the new pcre_config() function, and to
2498: output it. The program then exits immediately.
2499:
2500: 45. A need has arisen to pass over additional data with calls to pcre_exec() in
2501: order to support additional features. One way would have been to define
2502: pcre_exec2() (for example) with extra arguments, but this would not have been
2503: extensible, and would also have required all calls to the original function to
2504: be mapped to the new one. Instead, I have chosen to extend the mechanism that
2505: is used for passing in "extra" data from pcre_study().
2506:
2507: The pcre_extra structure is now exposed and defined in pcre.h. It currently
2508: contains the following fields:
2509:
2510: flags a bitmap indicating which of the following fields are set
2511: study_data opaque data from pcre_study()
2512: match_limit a way of specifying a limit on match() calls for a specific
2513: call to pcre_exec()
2514: callout_data data for callouts (see 49 below)
2515:
2516: The flag bits are also defined in pcre.h, and are
2517:
2518: PCRE_EXTRA_STUDY_DATA
2519: PCRE_EXTRA_MATCH_LIMIT
2520: PCRE_EXTRA_CALLOUT_DATA
2521:
2522: The pcre_study() function now returns one of these new pcre_extra blocks, with
2523: the actual study data pointed to by the study_data field, and the
2524: PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2525: before. That is, this change is entirely upwards-compatible and requires no
2526: change to existing code.
2527:
2528: If you want to pass in additional data to pcre_exec(), you can either place it
2529: in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2530: block.
2531:
2532: 46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2533: data string contains the escape sequence \M, pcretest calls pcre_exec() several
2534: times with different match limits, until it finds the minimum value needed for
2535: pcre_exec() to complete. The value is then output. This can be instructive; for
2536: most simple matches the number is quite small, but for pathological cases it
2537: gets very large very quickly.
2538:
2539: 47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2540: returns the size of the data block pointed to by the study_data field in a
2541: pcre_extra block, that is, the value that was passed as the argument to
2542: pcre_malloc() when PCRE was getting memory in which to place the information
2543: created by pcre_study(). The fourth argument should point to a size_t variable.
2544: pcretest has been extended so that this information is shown after a successful
2545: pcre_study() call when information about the compiled regex is being displayed.
2546:
2547: 48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2548: because what follows is always an absolute path. (Later: it turns out that this
2549: is more than cosmetic for MinGW, because it doesn't like empty path
2550: components.)
2551:
2552: 49. Some changes have been made to the callout feature (see 28 above):
2553:
2554: (i) A callout function now has three choices for what it returns:
2555:
2556: 0 => success, carry on matching
2557: > 0 => failure at this point, but backtrack if possible
2558: < 0 => serious error, return this value from pcre_exec()
2559:
2560: Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2561: values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2562: "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2563: use by callout functions. It will never be used by PCRE itself.
2564:
2565: (ii) The pcre_extra structure (see 45 above) has a void * field called
2566: callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2567: pcre_callout_block structure has a field of the same name. The contents of
2568: the field passed in the pcre_extra structure are passed to the callout
2569: function in the corresponding field in the callout block. This makes it
2570: easier to use the same callout-containing regex from multiple threads. For
2571: testing, the pcretest program has a new data escape
2572:
2573: \C*n pass the number n (may be negative) as callout_data
2574:
2575: If the callout function in pcretest receives a non-zero value as
2576: callout_data, it returns that value.
2577:
2578: 50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2579: there were some redundant $(CFLAGS) in commands that are now specified as
2580: $(LINK), which already includes $(CFLAGS).
2581:
2582: 51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2583: has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2584: with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2585: one-byte characters throughout. Note that case-insensitive matching applies
2586: only to characters whose values are less than 256. PCRE doesn't support the
2587: notion of cases for higher-valued characters.
2588:
2589: (i) A character class whose characters are all within 0-255 is handled as
2590: a bit map, and the map is inverted for negative classes. Previously, a
2591: character > 255 always failed to match such a class; however it should
2592: match if the class was a negative one (e.g. [^ab]). This has been fixed.
2593:
2594: (ii) A negated character class with a single character < 255 is coded as
2595: "not this character" (OP_NOT). This wasn't working properly when the test
2596: character was multibyte, either singly or repeated.
2597:
2598: (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2599: mode, for example: \x{100}{2,3}.
2600:
2601: (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2602: singly or repeated) now correctly test multibyte characters. However,
2603: PCRE doesn't recognize any characters with values greater than 255 as
2604: digits, spaces, or word characters. Such characters always match \D, \S,
2605: and \W, and never match \d, \s, or \w.
2606:
2607: (v) Classes may now contain characters and character ranges with values
2608: greater than 255. For example: [ab\x{100}-\x{400}].
2609:
2610: (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2611: PCRE in UTF-8 mode.
2612:
2613: 52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2614: PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2615: retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2616: value.)
2617:
2618: 53. The single man page has become too large. I have therefore split it up into
2619: a number of separate man pages. These also give rise to individual HTML pages;
2620: these are now put in a separate directory, and there is an index.html page that
2621: lists them all. Some hyperlinking between the pages has been installed.
2622:
2623: 54. Added convenience functions for handling named capturing parentheses.
2624:
2625: 55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2626: aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2627: true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2628: are faulted.
2629:
2630: 56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2631: calling configure. These values are used when compiling the dftables.c program
2632: which is run to generate the source of the default character tables. They
2633: default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2634: you will need to set these values.
2635:
2636: 57. Updated the building process for Windows DLL, as provided by Fred Cox.
2637:
2638:
2639: Version 3.9 02-Jan-02
2640: ---------------------
2641:
2642: 1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2643:
2644: 2. If --disable-static was given, the building process failed when trying to
2645: build pcretest and pcregrep. (For some reason it was using libtool to compile
2646: them, which is not right, as they aren't part of the library.)
2647:
2648:
2649: Version 3.8 18-Dec-01
2650: ---------------------
2651:
2652: 1. The experimental UTF-8 code was completely screwed up. It was packing the
2653: bytes in the wrong order. How dumb can you get?
2654:
2655:
2656: Version 3.7 29-Oct-01
2657: ---------------------
2658:
2659: 1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2660: This caused pcretest, when used on the test data, to segfault. Unfortunately,
2661: this didn't happen under Solaris 8, where I normally test things.
2662:
2663: 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2664: doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2665: isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2666: this fix an hour or so after the initial 3.7 release.)
2667:
2668:
2669: Version 3.6 23-Oct-01
2670: ---------------------
2671:
2672: 1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2673: offsets passed as NULL with zero offset count.
2674:
2675: 2. The config.guess and config.sub files had not been updated when I moved to
2676: the latest autoconf.
2677:
2678:
2679: Version 3.5 15-Aug-01
2680: ---------------------
2681:
2682: 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2683: had been forgotten.
2684:
2685: 2. By using declared but undefined structures, we can avoid using "void"
2686: definitions in pcre.h while keeping the internal definitions of the structures
2687: private.
2688:
2689: 3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2690: user point of view, this means that both static and shared libraries are built
2691: by default, but this can be individually controlled. More of the work of
2692: handling this static/shared cases is now inside libtool instead of PCRE's make
2693: file.
2694:
2695: 4. The pcretest utility is now installed along with pcregrep because it is
2696: useful for users (to test regexs) and by doing this, it automatically gets
2697: relinked by libtool. The documentation has been turned into a man page, so
2698: there are now .1, .txt, and .html versions in /doc.
2699:
2700: 5. Upgrades to pcregrep:
2701: (i) Added long-form option names like gnu grep.
2702: (ii) Added --help to list all options with an explanatory phrase.
2703: (iii) Added -r, --recursive to recurse into sub-directories.
2704: (iv) Added -f, --file to read patterns from a file.
2705:
2706: 6. pcre_exec() was referring to its "code" argument before testing that
2707: argument for NULL (and giving an error if it was NULL).
2708:
2709: 7. Upgraded Makefile.in to allow for compiling in a different directory from
2710: the source directory.
2711:
2712: 8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2713: options bits, the pointer it was passed was to an int instead of to an unsigned
2714: long int. This mattered only on 64-bit systems.
2715:
2716: 9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2717: generated) instead of pcre.in, which it its source. Also made the same change
2718: in several of the .c files.
2719:
2720: 10. A new release of gcc defines printf() as a macro, which broke pcretest
2721: because it had an ifdef in the middle of a string argument for printf(). Fixed
2722: by using separate calls to printf().
2723:
2724: 11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2725: script, to force use of CR or LF instead of \n in the source. On non-Unix
2726: systems, the value can be set in config.h.
2727:
2728: 12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2729: absolute limit. Changed the text of the error message to make this clear, and
2730: likewise updated the man page.
2731:
2732: 13. The limit of 99 on the number of capturing subpatterns has been removed.
2733: The new limit is 65535, which I hope will not be a "real" limit.
2734:
2735:
2736: Version 3.4 22-Aug-00
2737: ---------------------
2738:
2739: 1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2740:
2741: 2. Diagnose condition (?(0) as an error instead of crashing on matching.
2742:
2743:
2744: Version 3.3 01-Aug-00
2745: ---------------------
2746:
2747: 1. If an octal character was given, but the value was greater than \377, it
2748: was not getting masked to the least significant bits, as documented. This could
2749: lead to crashes in some systems.
2750:
2751: 2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2752: the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2753:
2754: 3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2755: These just pass their arguments on to (pcre_free)(), but they are provided
2756: because some uses of PCRE bind it to non-C systems that can call its functions,
2757: but cannot call free() or pcre_free() directly.
2758:
2759: 4. Add "make test" as a synonym for "make check". Corrected some comments in
2760: the Makefile.
2761:
2762: 5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2763: Makefile.
2764:
2765: 6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2766: command called pgrep for grepping around the active processes.
2767:
2768: 7. Added the beginnings of support for UTF-8 character strings.
2769:
2770: 8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2771: RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2772: the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2773: out for the ar command.)
2774:
2775:
2776: Version 3.2 12-May-00
2777: ---------------------
2778:
2779: This is purely a bug fixing release.
2780:
2781: 1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2782: of ZA. This was just one example of several cases that could provoke this bug,
2783: which was introduced by change 9 of version 2.00. The code for breaking
2784: infinite loops after an iteration that matches an empty string was't working
2785: correctly.
2786:
2787: 2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2788: when matched against abbab (for example). After matching an empty string, it
2789: wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2790: caused it to match further down the string than it should.
2791:
2792: 3. The code contained an inclusion of sys/types.h. It isn't clear why this
2793: was there because it doesn't seem to be needed, and it causes trouble on some
2794: systems, as it is not a Standard C header. It has been removed.
2795:
2796: 4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2797: were reported on the Macintosh. The changes were from
2798:
2799: while ((c = *(++ptr)) != 0 && c != '\n');
2800: to
2801: while ((c = *(++ptr)) != 0 && c != '\n') ;
2802:
2803: Totally extraordinary, but if that's what it takes...
2804:
2805: 5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2806: available. Added HAVE_BCOPY and an autoconf test for it; if neither
2807: HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2808: assumes the way PCRE uses memmove() (always moving upwards).
2809:
2810: 6. PCRE is being used in one environment where strchr() is not available. There
2811: was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2812: faster code anyway.
2813:
2814:
2815: Version 3.1 09-Feb-00
2816: ---------------------
2817:
2818: The only change in this release is the fixing of some bugs in Makefile.in for
2819: the "install" target:
2820:
2821: (1) It was failing to install pcreposix.h.
2822:
2823: (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2824:
2825:
2826: Version 3.0 01-Feb-00
2827: ---------------------
2828:
2829: 1. Add support for the /+ modifier to perltest (to output $` like it does in
2830: pcretest).
2831:
2832: 2. Add support for the /g modifier to perltest.
2833:
2834: 3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2835: matches null strings.
2836:
2837: 4. Fix perltest so that it doesn't do unwanted things when fed an empty
2838: pattern. Perl treats empty patterns specially - it reuses the most recent
2839: pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2840: effect.
2841:
2842: 5. The POSIX interface was broken in that it was just handing over the POSIX
2843: captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2844: required a bigger vector, with some working space on the end. This means that
2845: the POSIX wrapper now has to get and free some memory, and copy the results.
2846:
2847: 6. Added some simple autoconf support, placing the test data and the
2848: documentation in separate directories, re-organizing some of the
2849: information files, and making it build pcre-config (a GNU standard). Also added
2850: libtool support for building PCRE as a shared library, which is now the
2851: default.
2852:
2853: 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2854: 09 are not valid octal constants. Single digits will be used for minor values
2855: less than 10.
2856:
2857: 8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2858: existing programs that set these in the POSIX interface can use PCRE without
2859: modification.
2860:
2861: 9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2862: return all that pcre_info() returns, plus additional data. The pcre_info()
2863: function is retained for compatibility, but is considered to be obsolete.
2864:
2865: 10. Added experimental recursion feature (?R) to handle one common case that
2866: Perl 5.6 will be able to do with (?p{...}).
2867:
2868: 11. Added support for POSIX character classes like [:alpha:], which Perl is
2869: adopting.
2870:
2871:
2872: Version 2.08 31-Aug-99
2873: ----------------------
2874:
2875: 1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2876: trying to match at the startoffset position, but instead was moving forward to
2877: the next newline as if a previous match had failed.
2878:
2879: 2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2880: and could get into a loop if a null string was matched other than at the start
2881: of the subject.
2882:
2883: 3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2884: be distinguished at compile time, and for completeness also added PCRE_DATE.
2885:
2886: 5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2887: in GnuWin32 environments.
2888:
2889:
2890: Version 2.07 29-Jul-99
2891: ----------------------
2892:
2893: 1. The documentation is now supplied in plain text form and HTML as well as in
2894: the form of man page sources.
2895:
2896: 2. C++ compilers don't like assigning (void *) values to other pointer types.
2897: In particular this affects malloc(). Although there is no problem in Standard
2898: C, I've put in casts to keep C++ compilers happy.
2899:
2900: 3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2901: should be (const char *).
2902:
2903: 4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2904: be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2905: However, I haven't made this a standard facility. The documentation doesn't
2906: mention it, and the Makefile doesn't support it.
2907:
2908: 5. The Makefile now contains an "install" target, with editable destinations at
2909: the top of the file. The pcretest program is not installed.
2910:
2911: 6. pgrep -V now gives the PCRE version number and date.
2912:
2913: 7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2914: causing the entire string to be ignored, instead of just the last character.
2915:
2916: 8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2917: non-matching string, it can take a very, very long time, even for strings of
2918: quite modest length, because of the nested recursion. PCRE now does better in
2919: some of these cases. It does this by remembering the last required literal
2920: character in the pattern, and pre-searching the subject to ensure it is present
2921: before running the real match. In other words, it applies a heuristic to detect
2922: some types of certain failure quickly, and in the above example, if presented
2923: with a string that has no trailing " it gives "no match" very quickly.
2924:
2925: 9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2926: other alternatives are tried instead.
2927:
2928:
2929: Version 2.06 09-Jun-99
2930: ----------------------
2931:
2932: 1. Change pcretest's output for amount of store used to show just the code
2933: space, because the remainder (the data block) varies in size between 32-bit and
2934: 64-bit systems.
2935:
2936: 2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2937: start matching at. This allows lookbehinds to work when searching for multiple
2938: occurrences in a string.
2939:
2940: 3. Added additional options to pcretest for testing multiple occurrences:
2941:
2942: /+ outputs the rest of the string that follows a match
2943: /g loops for multiple occurrences, using the new startoffset argument
2944: /G loops for multiple occurrences by passing an incremented pointer
2945:
2946: 4. PCRE wasn't doing the "first character" optimization for patterns starting
2947: with \b or \B, though it was doing it for other lookbehind assertions. That is,
2948: it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2949: the letter 'x'. On long subject strings, this gives a significant speed-up.
2950:
2951:
2952: Version 2.05 21-Apr-99
2953: ----------------------
2954:
2955: 1. Changed the type of magic_number from int to long int so that it works
2956: properly on 16-bit systems.
2957:
2958: 2. Fixed a bug which caused patterns starting with .* not to work correctly
2959: when the subject string contained newline characters. PCRE was assuming
2960: anchoring for such patterns in all cases, which is not correct because .* will
2961: not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2962: DOTALL is set at top level; otherwise it knows that patterns starting with .*
2963: must be retried after every newline in the subject.
2964:
2965:
2966: Version 2.04 18-Feb-99
2967: ----------------------
2968:
2969: 1. For parenthesized subpatterns with repeats whose minimum was zero, the
2970: computation of the store needed to hold the pattern was incorrect (too large).
2971: If such patterns were nested a few deep, this could multiply and become a real
2972: problem.
2973:
2974: 2. Added /M option to pcretest to show the memory requirement of a specific
2975: pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2976:
2977: 3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2978: compiled in such a way that the backtracking after subsequent failure was
2979: pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2980: ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2981:
2982:
2983: Version 2.03 02-Feb-99
2984: ----------------------
2985:
2986: 1. Fixed typo and small mistake in man page.
2987:
2988: 2. Added 4th condition (GPL supersedes if conflict) and created separate
2989: LICENCE file containing the conditions.
2990:
2991: 3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2992: Perl, that is the internal \ allows the delimiter to be included in the
2993: pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2994: the final delimiter, add \ to the end of the pattern (to test the error).
2995:
2996: 4. Added the convenience functions for extracting substrings after a successful
2997: match. Updated pcretest to make it able to test these functions.
2998:
2999:
3000: Version 2.02 14-Jan-99
3001: ----------------------
3002:
3003: 1. Initialized the working variables associated with each extraction so that
3004: their saving and restoring doesn't refer to uninitialized store.
3005:
3006: 2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3007: compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3008: fix the problem.
3009:
3010: 3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3011: calls, and wasn't printing the correct value for compiling calls. Increased the
3012: default value of LOOPREPEAT, and the number of significant figures in the
3013: times.
3014:
3015: 4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3016:
3017: 5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3018: a building problem on Windows NT with a FAT file system.
3019:
3020:
3021: Version 2.01 21-Oct-98
3022: ----------------------
3023:
3024: 1. Changed the API for pcre_compile() to allow for the provision of a pointer
3025: to character tables built by pcre_maketables() in the current locale. If NULL
3026: is passed, the default tables are used.
3027:
3028:
3029: Version 2.00 24-Sep-98
3030: ----------------------
3031:
3032: 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3033: it any more.
3034:
3035: 2. Allow quantification of (?>) groups, and make it work correctly.
3036:
3037: 3. The first character computation wasn't working for (?>) groups.
3038:
3039: 4. Correct the implementation of \Z (it is permitted to match on the \n at the
3040: end of the subject) and add 5.005's \z, which really does match only at the
3041: very end of the subject.
3042:
3043: 5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3044:
3045: 6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3046: DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3047: localized options. All options to pcre_study() were also removed.
3048:
3049: 7. Add other new features from 5.005:
3050:
3051: $(?<= positive lookbehind
3052: $(?<! negative lookbehind
3053: (?imsx-imsx) added the unsetting capability
3054: such a setting is global if at outer level; local otherwise
3055: (?imsx-imsx:) non-capturing groups with option setting
3056: (?(cond)re|re) conditional pattern matching
3057:
3058: A backreference to itself in a repeated group matches the previous
3059: captured string.
3060:
3061: 8. General tidying up of studying (both automatic and via "study")
3062: consequential on the addition of new assertions.
3063:
3064: 9. As in 5.005, unlimited repeated groups that could match an empty substring
3065: are no longer faulted at compile time. Instead, the loop is forcibly broken at
3066: runtime if any iteration does actually match an empty substring.
3067:
3068: 10. Include the RunTest script in the distribution.
3069:
3070: 11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3071: discrepancies, some of which were old and were also with respect to 5.004. They
3072: have now been fixed.
3073:
3074:
3075: Version 1.09 28-Apr-98
3076: ----------------------
3077:
3078: 1. A negated single character class followed by a quantifier with a minimum
3079: value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3080: program crashes, or just wrong answers. This did not apply to negated classes
3081: containing more than one character, or to minima other than one.
3082:
3083:
3084: Version 1.08 27-Mar-98
3085: ----------------------
3086:
3087: 1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3088:
3089: 2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3090: latter must appear before anything that relies on it in the pattern.
3091:
3092:
3093: Version 1.07 16-Feb-98
3094: ----------------------
3095:
3096: 1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3097: repeat of a potentially empty string).
3098:
3099:
3100: Version 1.06 23-Jan-98
3101: ----------------------
3102:
3103: 1. Added Markus Oberhumer's little patches for C++.
3104:
3105: 2. Literal strings longer than 255 characters were broken.
3106:
3107:
3108: Version 1.05 23-Dec-97
3109: ----------------------
3110:
3111: 1. Negated character classes containing more than one character were failing if
3112: PCRE_CASELESS was set at run time.
3113:
3114:
3115: Version 1.04 19-Dec-97
3116: ----------------------
3117:
3118: 1. Corrected the man page, where some "const" qualifiers had been omitted.
3119:
3120: 2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3121: input syntax.
3122:
3123: 3. Fixed memory leak which occurred when a regex with back references was
3124: matched with an offsets vector that wasn't big enough. The temporary memory
3125: that is used in this case wasn't being freed if the match failed.
3126:
3127: 4. Tidied pcretest to ensure it frees memory that it gets.
3128:
3129: 5. Temporary memory was being obtained in the case where the passed offsets
3130: vector was exactly big enough.
3131:
3132: 6. Corrected definition of offsetof() from change 5 below.
3133:
3134: 7. I had screwed up change 6 below and broken the rules for the use of
3135: setjmp(). Now fixed.
3136:
3137:
3138: Version 1.03 18-Dec-97
3139: ----------------------
3140:
3141: 1. A erroneous regex with a missing opening parenthesis was correctly
3142: diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3143: on some systems.
3144:
3145: 2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3146: it was reported that one broken compiler failed on the former because "code" is
3147: also an independent variable.
3148:
3149: 3. The erroneous regex a[]b caused an array overrun reference.
3150:
3151: 4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3152: fail on data ending with that character. (It was going on too far, and checking
3153: the next character, typically a binary zero.) This was specific to the
3154: optimized code for single-character negative classes.
3155:
3156: 5. Added a contributed patch from the TIN world which does the following:
3157:
3158: + Add an undef for memmove, in case the the system defines a macro for it.
3159:
3160: + Add a definition of offsetof(), in case there isn't one. (I don't know
3161: the reason behind this - offsetof() is part of the ANSI standard - but
3162: it does no harm).
3163:
3164: + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3165: most of the places where whitespace preceded '#'. I have given up and
3166: allowed the remaining 2 cases to be at the margin.
3167:
3168: + Rename some variables in pcre to eliminate shadowing. This seems very
3169: pedantic, but does no harm, of course.
3170:
3171: 6. Moved the call to setjmp() into its own function, to get rid of warnings
3172: from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3173:
3174: 7. Constructs such as \d{8,} were compiling into the equivalent of
3175: \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3176: outcome, but in this particular case used more store than had been allocated,
3177: which caused the bug to be discovered because it threw up an internal error.
3178:
3179: 8. The debugging code in both pcre and pcretest for outputting the compiled
3180: form of a regex was going wrong in the case of back references followed by
3181: curly-bracketed repeats.
3182:
3183:
3184: Version 1.02 12-Dec-97
3185: ----------------------
3186:
3187: 1. Typos in pcre.3 and comments in the source fixed.
3188:
3189: 2. Applied a contributed patch to get rid of places where it used to remove
3190: 'const' from variables, and fixed some signed/unsigned and uninitialized
3191: variable warnings.
3192:
3193: 3. Added the "runtest" target to Makefile.
3194:
3195: 4. Set default compiler flag to -O2 rather than just -O.
3196:
3197:
3198: Version 1.01 19-Nov-97
3199: ----------------------
3200:
3201: 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3202: like /([ab]*)*/, that is, for classes with more than one character in them.
3203:
3204: 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3205: as /((?>a*))*/ (a PCRE_EXTRA facility).
3206:
3207:
3208: Version 1.00 18-Nov-97
3209: ----------------------
3210:
3211: 1. Added compile-time macros to support systems such as SunOS4 which don't have
3212: memmove() or strerror() but have other things that can be used instead.
3213:
3214: 2. Arranged that "make clean" removes the executables.
3215:
3216:
3217: Version 0.99 27-Oct-97
3218: ----------------------
3219:
3220: 1. Fixed bug in code for optimizing classes with only one character. It was
3221: initializing a 32-byte map regardless, which could cause it to run off the end
3222: of the memory it had got.
3223:
3224: 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3225:
3226:
3227: Version 0.98 22-Oct-97
3228: ----------------------
3229:
3230: 1. Fixed bug in code for handling temporary memory usage when there are more
3231: back references than supplied space in the ovector. This could cause segfaults.
3232:
3233:
3234: Version 0.97 21-Oct-97
3235: ----------------------
3236:
3237: 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3238:
3239: 2. Optimized negated single characters not to use a bit map.
3240:
3241: 3. Brought error texts together as macro definitions; clarified some of them;
3242: fixed one that was wrong - it said "range out of order" when it meant "invalid
3243: escape sequence".
3244:
3245: 4. Changed some char * arguments to const char *.
3246:
3247: 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3248:
3249: 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3250: pcretest.
3251:
3252:
3253: Version 0.96 16-Oct-97
3254: ----------------------
3255:
3256: 1. Added a simple "pgrep" utility to the distribution.
3257:
3258: 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3259: unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3260: where "ddd" means "one or more decimal digits".
3261:
3262: 3. Fixed serious bug. If a pattern had a back reference, but the call to
3263: pcre_exec() didn't supply a large enough ovector to record the related
3264: identifying subpattern, the match always failed. PCRE now remembers the number
3265: of the largest back reference, and gets some temporary memory in which to save
3266: the offsets during matching if necessary, in order to ensure that
3267: backreferences always work.
3268:
3269: 4. Increased the compatibility with Perl in a number of ways:
3270:
3271: (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3272: to request this handling. The option can be set at compile or exec time.
3273:
3274: (b) $ matches before a terminating newline by default; an option
3275: PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3276: mode). The option can be set at compile or exec time.
3277:
3278: (c) The handling of \ followed by a digit other than 0 is now supposed to be
3279: the same as Perl's. If the decimal number it represents is less than 10
3280: or there aren't that many previous left capturing parentheses, an octal
3281: escape is read. Inside a character class, it's always an octal escape,
3282: even if it is a single digit.
3283:
3284: (d) An escaped but undefined alphabetic character is taken as a literal,
3285: unless PCRE_EXTRA is set. Currently this just reserves the remaining
3286: escapes.
3287:
3288: (e) {0} is now permitted. (The previous item is removed from the compiled
3289: pattern).
3290:
3291: 5. Changed all the names of code files so that the basic parts are no longer
3292: than 10 characters, and abolished the teeny "globals.c" file.
3293:
3294: 6. Changed the handling of character classes; they are now done with a 32-byte
3295: bit map always.
3296:
3297: 7. Added the -d and /D options to pcretest to make it possible to look at the
3298: internals of compilation without having to recompile pcre.
3299:
3300:
3301: Version 0.95 23-Sep-97
3302: ----------------------
3303:
3304: 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3305: \x20 at the start of a run of normal characters. These were being treated as
3306: real characters, instead of the source characters being re-checked.
3307:
3308:
3309: Version 0.94 18-Sep-97
3310: ----------------------
3311:
3312: 1. The functions are now thread-safe, with the caveat that the global variables
3313: containing pointers to malloc() and free() or alternative functions are the
3314: same for all threads.
3315:
3316: 2. Get pcre_study() to generate a bitmap of initial characters for non-
3317: anchored patterns when this is possible, and use it if passed to pcre_exec().
3318:
3319:
3320: Version 0.93 15-Sep-97
3321: ----------------------
3322:
3323: 1. /(b)|(:+)/ was computing an incorrect first character.
3324:
3325: 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3326: but not actually doing anything yet.
3327:
3328: 3. Treat "-" characters in classes that cannot be part of ranges as literals,
3329: as Perl does (e.g. [-az] or [az-]).
3330:
3331: 4. Set the anchored flag if a branch starts with .* or .*? because that tests
3332: all possible positions.
3333:
3334: 5. Split up into different modules to avoid including unneeded functions in a
3335: compiled binary. However, compile and exec are still in one module. The "study"
3336: function is split off.
3337:
3338: 6. The character tables are now in a separate module whose source is generated
3339: by an auxiliary program - but can then be edited by hand if required. There are
3340: now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3341: toupper() in the code.
3342:
3343: 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3344: make them global. Abolish the function for setting them, as the caller can now
3345: set them directly.
3346:
3347:
3348: Version 0.92 11-Sep-97
3349: ----------------------
3350:
3351: 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3352: (e.g. /a{1,3}/) was broken (I mis-optimized it).
3353:
3354: 2. Caseless matching was not working in character classes if the characters in
3355: the pattern were in upper case.
3356:
3357: 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3358:
3359: 4. Make PCRE_ANCHORED public and accept as a compile option.
3360:
3361: 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3362: PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3363: pass them.
3364:
3365: 6. Give an error if bad option bits passed at compile or run time.
3366:
3367: 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3368: pcretest to cause it to pass that flag.
3369:
3370: 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3371: options, and the first character, if set.
3372:
3373: 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3374:
3375:
3376: Version 0.91 10-Sep-97
3377: ----------------------
3378:
3379: 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3380: match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3381:
3382: 2. PCRE was looping on encountering an indefinitely repeated back reference to
3383: a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3384: Perl does - treats the match as successful.
3385:
3386: ****
E-mail:
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to ChangeLog CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [parser3project] / win32 / pcre