/[pcre]/code/trunk/ChangeLog
ViewVC logotype

Diff of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 79 by nigel, Sat Feb 24 21:40:52 2007 UTC revision 353 by ph10, Mon Jul 7 15:44:24 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 8.0 02 Jul-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    
22    Version 7.7 07-May-08
23    ---------------------
24    
25    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
26        a string to a long long, pretend we don't even have a long long." This is
27        done by checking for the strtoq, strtoll, and _strtoi64 functions.
28    
29    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
30        pre-7.6 versions, which defined a global no_arg variable instead of putting
31        it in the RE class. (See also #8 below.)
32    
33    3.  Remove a line of dead code, identified by coverity and reported by Nuno
34        Lopes.
35    
36    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
37    
38        (1) The include/exclude patterns were being applied to the whole pathnames
39            of files, instead of just to the final components.
40    
41        (2) If there was more than one level of directory, the subdirectories were
42            skipped unless they satisfied the include/exclude conditions. This is
43            inconsistent with GNU grep (and could even be seen as contrary to the
44            pcregrep specification - which I improved to make it absolutely clear).
45            The action now is always to scan all levels of directory, and just
46            apply the include/exclude patterns to regular files.
47    
48    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
49        --exclude_dir in the tests to avoid scanning .svn directories.
50    
51    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
52        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
53        doesn't support NULs in patterns.
54    
55    7.  Added some missing "const"s to declarations of static tables in
56        pcre_compile.c and pcre_dfa_exec.c.
57    
58    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
59        caused by fix #2  above. (Subsequently also a second patch to fix the
60        first patch. And a third patch - this was a messy problem.)
61    
62    9.  Applied Craig's patch to remove the use of push_back().
63    
64    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
65        matching function regexec().
66    
67    11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
68        which, however, unlike Perl's \g{...}, are subroutine calls, not back
69        references. PCRE supports relative numbers with this syntax (I don't think
70        Oniguruma does).
71    
72    12. Previously, a group with a zero repeat such as (...){0} was completely
73        omitted from the compiled regex. However, this means that if the group
74        was called as a subroutine from elsewhere in the pattern, things went wrong
75        (an internal error was given). Such groups are now left in the compiled
76        pattern, with a new opcode that causes them to be skipped at execution
77        time.
78    
79    13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
80        to the way PCRE behaves:
81    
82        (a) A lone ] character is dis-allowed (Perl treats it as data).
83    
84        (b) A back reference to an unmatched subpattern matches an empty string
85            (Perl fails the current match path).
86    
87        (c) A data ] in a character class must be notated as \] because if the
88            first data character in a class is ], it defines an empty class. (In
89            Perl it is not possible to have an empty class.) The empty class []
90            never matches; it forces failure and is equivalent to (*FAIL) or (?!).
91            The negative empty class [^] matches any one character, independently
92            of the DOTALL setting.
93    
94    14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
95        non-existent subpattern following a character class starting with ']' and
96        containing () gave an internal compiling error instead of "reference to
97        non-existent subpattern". Fortunately, when the pattern did exist, the
98        compiled code was correct. (When scanning forwards to check for the
99        existencd of the subpattern, it was treating the data ']' as terminating
100        the class, so got the count wrong. When actually compiling, the reference
101        was subsequently set up correctly.)
102    
103    15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
104        it was being rejected as not supported by pcre_dfa_exec(), even though
105        other assertions are supported. I have made pcre_dfa_exec() support
106        (*FAIL).
107    
108    16. The implementation of 13c above involved the invention of a new opcode,
109        OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
110        cannot be changed at match time, I realized I could make a small
111        improvement to matching performance by compiling OP_ALLANY instead of
112        OP_ANY for "." when DOTALL was set, and then removing the runtime tests
113        on the OP_ANY path.
114    
115    17. Compiling pcretest on Windows with readline support failed without the
116        following two fixes: (1) Make the unistd.h include conditional on
117        HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
118    
119    18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
120        ncurses library to be included for pcretest when ReadLine support is
121        requested, but also to allow for it to be overridden. This patch came from
122        Daniel Bergström.
123    
124    19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
125        as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
126        any errors with the current Unicode tables. Thanks to Peter Kankowski for
127        spotting this.
128    
129    
130    Version 7.6 28-Jan-08
131    ---------------------
132    
133    1.  A character class containing a very large number of characters with
134        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
135        overflow.
136    
137    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
138        HAVE_LONG_LONG is not defined.
139    
140    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
141        bring them up to date and include new features. This patch includes:
142    
143        - Fixed PH's badly added libz and libbz2 support.
144        - Fixed a problem with static linking.
145        - Added pcredemo. [But later removed - see 7 below.]
146        - Fixed dftables problem and added an option.
147        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
148            HAVE_LONG_LONG.
149        - Added readline support for pcretest.
150        - Added an listing of the option settings after cmake has run.
151    
152    4.  A user submitted a patch to Makefile that makes it easy to create
153        "pcre.dll" under mingw when using Configure/Make. I added stuff to
154        Makefile.am that cause it to include this special target, without
155        affecting anything else. Note that the same mingw target plus all
156        the other distribution libraries and programs are now supported
157        when configuring with CMake (see 6 below) instead of with
158        Configure/Make.
159    
160    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
161        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
162        exported in the Windows port". It has not yet been confirmed that the patch
163        solves the problem, but it does no harm.
164    
165    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
166        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
167        with CMake, and also correct the comment about stack recursion.
168    
169    7.  Remove the automatic building of pcredemo from the ./configure system and
170        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
171        of a program that users should build themselves after PCRE is installed, so
172        building it automatically is not really right. What is more, it gave
173        trouble in some build environments.
174    
175    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
176    
177    
178    Version 7.5 10-Jan-08
179    ---------------------
180    
181    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
182        values in parens when parsing an RE using the C++ wrapper."
183    
184    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
185        Characters greater than 255 were excluded from the class instead of being
186        included.
187    
188    3.  The same bug as (2) above applied to negated POSIX classes such as
189        [:^space:].
190    
191    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
192        defined or documented. It seems to have been a typo for PCRE_STATIC, so
193        I have changed it.
194    
195    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
196        first named subpattern) and a construct such as (?&a) would reference the
197        first named subpattern whose name started with "a" (in other words, the
198        length check was missing). Both these problems are fixed. "Subpattern name
199        expected" is now given for (?&) (a zero-length name), and this patch also
200        makes it give the same error for \k'' (previously it complained that that
201        was a reference to a non-existent subpattern).
202    
203    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
204        this is right because (?- can be followed by option settings as well as by
205        digits. I have, however, made the messages clearer.
206    
207    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
208        than the number used in the conditional) now cause a compile-time error.
209        This is actually not compatible with Perl, which accepts such patterns, but
210        treats the conditional as always being FALSE (as PCRE used to), but it
211        seems to me that giving a diagnostic is better.
212    
213    8.  Change "alphameric" to the more common word "alphanumeric" in comments
214        and messages.
215    
216    9.  Fix two occurrences of "backslash" in comments that should have been
217        "backspace".
218    
219    10. Remove two redundant lines of code that can never be obeyed (their function
220        was moved elsewhere).
221    
222    11. The program that makes PCRE's Unicode character property table had a bug
223        which caused it to generate incorrect table entries for sequences of
224        characters that have the same character type, but are in different scripts.
225        It amalgamated them into a single range, with the script of the first of
226        them. In other words, some characters were in the wrong script. There were
227        thirteen such cases, affecting characters in the following ranges:
228    
229          U+002b0 - U+002c1
230          U+0060c - U+0060d
231          U+0061e - U+00612
232          U+0064b - U+0065e
233          U+0074d - U+0076d
234          U+01800 - U+01805
235          U+01d00 - U+01d77
236          U+01d9b - U+01dbf
237          U+0200b - U+0200f
238          U+030fc - U+030fe
239          U+03260 - U+0327f
240          U+0fb46 - U+0fbb1
241          U+10450 - U+1049d
242    
243    12. The -o option (show only the matching part of a line) for pcregrep was not
244        compatible with GNU grep in that, if there was more than one match in a
245        line, it showed only the first of them. It now behaves in the same way as
246        GNU grep.
247    
248    13. If the -o and -v options were combined for pcregrep, it printed a blank
249        line for every non-matching line. GNU grep prints nothing, and pcregrep now
250        does the same. The return code can be used to tell if there were any
251        non-matching lines.
252    
253    14. Added --file-offsets and --line-offsets to pcregrep.
254    
255    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
256        infinitely looping recursion. The bug was that positive lookaheads were not
257        being skipped when checking for a possible empty match (negative lookaheads
258        and both kinds of lookbehind were skipped).
259    
260    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
261        inclusion of <windows.h> to before rather than after the definition of
262        INVALID_FILE_ATTRIBUTES (patch from David Byron).
263    
264    17. Specifying a possessive quantifier with a specific limit for a Unicode
265        character property caused pcre_compile() to compile bad code, which led at
266        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
267        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
268        caused the error; without that there was no problem.
269    
270    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
271    
272    19. Added --enable-pcretest-libreadline.
273    
274    20. In pcrecpp.cc, the variable 'count' was incremented twice in
275        RE::GlobalReplace(). As a result, the number of replacements returned was
276        double what it should be. I removed one of the increments, but Craig sent a
277        later patch that removed the other one (the right fix) and added unit tests
278        that check the return values (which was not done before).
279    
280    21. Several CMake things:
281    
282        (1) Arranged that, when cmake is used on Unix, the libraries end up with
283            the names libpcre and libpcreposix, not just pcre and pcreposix.
284    
285        (2) The above change means that pcretest and pcregrep are now correctly
286            linked with the newly-built libraries, not previously installed ones.
287    
288        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
289    
290    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
291        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
292        UTF-8 newline character). The key issue is that the pattern starts .*;
293        this means that the match must be either at the beginning, or after a
294        newline. The bug was in the code for advancing after a failed match and
295        checking that the new position followed a newline. It was not taking
296        account of UTF-8 characters correctly.
297    
298    23. PCRE was behaving differently from Perl in the way it recognized POSIX
299        character classes. PCRE was not treating the sequence [:...:] as a
300        character class unless the ... were all letters. Perl, however, seems to
301        allow any characters between [: and :], though of course it rejects as
302        unknown any "names" that contain non-letters, because all the known class
303        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
304        for example, whereas PCRE did not - it did not recognize a POSIX character
305        class. This seemed a bit dangerous, so the code has been changed to be
306        closer to Perl. The behaviour is not identical to Perl, because PCRE will
307        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
308        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
309        Perl does, and where it didn't before.
310    
311    24. Rewrite so as to remove the single use of %n from pcregrep because in some
312        Windows environments %n is disabled by default.
313    
314    
315    Version 7.4 21-Sep-07
316    ---------------------
317    
318    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
319        means that a class such as [\s] counted as "explicit reference to CR or
320        LF". That isn't really right - the whole point of the change was to try to
321        help when there was an actual mention of one of the two characters. So now
322        the change happens only if \r or \n (or a literal CR or LF) character is
323        encountered.
324    
325    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
326        of both had grown to the point where there were only 3 bits left.
327        Fortunately, there was spare space in the data structure, and so I have
328        moved the internal flags into a new 16-bit field to free up more option
329        bits.
330    
331    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
332        but did not set the internal JCHANGED flag - either of these is enough to
333        control the way the "get" function works - but the PCRE_INFO_JCHANGED
334        facility is supposed to tell if (?J) was ever used, so now (?J) at the
335        start sets both bits.
336    
337    4.  Added options (at build time, compile time, exec time) to change \R from
338        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
339    
340    5.  doc/pcresyntax.html was missing from the distribution.
341    
342    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
343        compatibility, even though it is no longer used.
344    
345    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
346        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
347        windows.h file is present (where different names are used). [This was
348        reversed later after testing - see 16 below.]
349    
350    8.  Changed all #include <config.h> to #include "config.h". There were also
351        some further <pcre.h> cases that I changed to "pcre.h".
352    
353    9.  When pcregrep was used with the --colour option, it missed the line ending
354        sequence off the lines that it output.
355    
356    10. It was pointed out to me that arrays of string pointers cause lots of
357        relocations when a shared library is dynamically loaded. A technique of
358        using a single long string with a table of offsets can drastically reduce
359        these. I have refactored PCRE in four places to do this. The result is
360        dramatic:
361    
362          Originally:                          290
363          After changing UCP table:            187
364          After changing error message table:   43
365          After changing table of "verbs"       36
366          After changing table of Posix names   22
367    
368        Thanks to the folks working on Gregex for glib for this insight.
369    
370    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
371        unicode-properties was also set.
372    
373    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
374    
375    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
376        checked only for CRLF.
377    
378    14. Added casts to pcretest.c to avoid compiler warnings.
379    
380    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
381    
382    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
383        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
384        entirely. This removes changes made in 7 above.
385    
386    17. The CMake files have been updated, and there is now more information about
387        building with CMake in the NON-UNIX-USE document.
388    
389    
390    Version 7.3 28-Aug-07
391    ---------------------
392    
393     1. In the rejigging of the build system that eventually resulted in 7.1, the
394        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
395        brackets there is not right, since it causes compilers to look for an
396        installed pcre.h, not the version that is in the source that is being
397        compiled (which of course may be different). I have changed it back to:
398    
399          #include "pcre.h"
400    
401        I have a vague recollection that the change was concerned with compiling in
402        different directories, but in the new build system, that is taken care of
403        by the VPATH setting the Makefile.
404    
405     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
406        when the subject happened to end in the byte 0x85 (e.g. if the last
407        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
408        characters but of course it shouldn't be taken as a newline when it is part
409        of another character. The bug was that, for an unlimited repeat of . in
410        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
411        characters when looking for a newline.
412    
413     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
414    
415     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
416        in debug output.
417    
418     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
419        long printing in the pcrecpp unittest when running under MinGW.
420    
421     6. ESC_K was left out of the EBCDIC table.
422    
423     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
424        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
425        limit also applies to "virtual nesting" when a pattern is recursive, and in
426        this case 1000 isn't so big. I have been able to remove this limit at the
427        expense of backing off one optimization in certain circumstances. Normally,
428        when pcre_exec() would call its internal match() function recursively and
429        immediately return the result unconditionally, it uses a "tail recursion"
430        feature to save stack. However, when a subpattern that can match an empty
431        string has an unlimited repetition quantifier, it no longer makes this
432        optimization. That gives it a stack frame in which to save the data for
433        checking that an empty string has been matched. Previously this was taken
434        from the 1000-entry workspace that had been reserved. So now there is no
435        explicit limit, but more stack is used.
436    
437     8. Applied Daniel's patches to solve problems with the import/export magic
438        syntax that is required for Windows, and which was going wrong for the
439        pcreposix and pcrecpp parts of the library. These were overlooked when this
440        problem was solved for the main library.
441    
442     9. There were some crude static tests to avoid integer overflow when computing
443        the size of patterns that contain repeated groups with explicit upper
444        limits. As the maximum quantifier is 65535, the maximum group length was
445        set at 30,000 so that the product of these two numbers did not overflow a
446        32-bit integer. However, it turns out that people want to use groups that
447        are longer than 30,000 bytes (though not repeat them that many times).
448        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
449        made it possible to implement the integer overflow checks in a much more
450        dynamic way, which I have now done. The artificial limitation on group
451        length has been removed - we now have only the limit on the total length of
452        the compiled pattern, which depends on the LINK_SIZE setting.
453    
454    10. Fixed a bug in the documentation for get/copy named substring when
455        duplicate names are permitted. If none of the named substrings are set, the
456        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
457        empty string.
458    
459    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
460        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
461        because the ] is interpreted as the first data character and the
462        terminating ] is not found. PCRE has been made compatible with Perl in this
463        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
464        cause memory overwriting.
465    
466    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
467        string has been matched (to stop an infinite loop). It was not recognizing
468        a conditional subpattern that could match an empty string if that
469        subpattern was within another subpattern. For example, it looped when
470        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
471        condition was not nested. This bug has been fixed.
472    
473    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
474        past the start of the subject in the presence of bytes with the top bit
475        set, for example "\x8aBCD".
476    
477    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
478        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
479    
480    14. Optimized (?!) to (*FAIL).
481    
482    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
483        This restricts code points to be within the range 0 to 0x10FFFF, excluding
484        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
485        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
486        does: it's just the validity check that is more restrictive.
487    
488    16. Inserted checks for integer overflows during escape sequence (backslash)
489        processing, and also fixed erroneous offset values for syntax errors during
490        backslash processing.
491    
492    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
493        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
494    
495    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
496        caused an overrun.
497    
498    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
499        something other than just ASCII characters) inside a group that had an
500        unlimited repeat caused a loop at compile time (while checking to see
501        whether the group could match an empty string).
502    
503    20. Debugging a pattern containing \p or \P could cause a crash. For example,
504        [\P{Any}] did so. (Error in the code for printing property names.)
505    
506    21. An orphan \E inside a character class could cause a crash.
507    
508    22. A repeated capturing bracket such as (A)? could cause a wild memory
509        reference during compilation.
510    
511    23. There are several functions in pcre_compile() that scan along a compiled
512        expression for various reasons (e.g. to see if it's fixed length for look
513        behind). There were bugs in these functions when a repeated \p or \P was
514        present in the pattern. These operators have additional parameters compared
515        with \d, etc, and these were not being taken into account when moving along
516        the compiled data. Specifically:
517    
518        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
519            length.
520    
521        (b) An item such as \pL+ within a repeated group could cause crashes or
522            loops.
523    
524        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
525            "reference to non-existent subpattern" error.
526    
527        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
528    
529    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
530        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
531    
532    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
533    
534    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
535        character were causing crashes (broken optimization).
536    
537    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
538        \p or \P) caused a compile-time loop.
539    
540    28. More problems have arisen in unanchored patterns when CRLF is a valid line
541        break. For example, the unstudied pattern [\r\n]A does not match the string
542        "\r\nA" because change 7.0/46 below moves the current point on by two
543        characters after failing to match at the start. However, the pattern \nA
544        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
545        the same is true. There doesn't seem any very clean way out of this, but
546        what I have chosen to do makes the common cases work: PCRE now takes note
547        of whether there can be an explicit match for \r or \n anywhere in the
548        pattern, and if so, 7.0/46 no longer applies. As part of this change,
549        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
550        pattern has explicit CR or LF references.
551    
552    29. Added (*CR) etc for changing newline setting at start of pattern.
553    
554    
555    Version 7.2 19-Jun-07
556    ---------------------
557    
558     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
559        which is apparently normally available under Windows.
560    
561     2. Re-jig the pcregrep tests with different newline settings in an attempt
562        to make them independent of the local environment's newline setting.
563    
564     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
565    
566     4. Some of the "internals" tests were previously cut out when the link size
567        was not 2, because the output contained actual offsets. The recent new
568        "Z" feature of pcretest means that these can be cut out, making the tests
569        usable with all link sizes.
570    
571     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
572        stack recursion. This gives a massive performance boost under BSD, but just
573        a small improvement under Linux. However, it saves one field in the frame
574        in all cases.
575    
576     6. Added more features from the forthcoming Perl 5.10:
577    
578        (a) (?-n) (where n is a string of digits) is a relative subroutine or
579            recursion call. It refers to the nth most recently opened parentheses.
580    
581        (b) (?+n) is also a relative subroutine call; it refers to the nth next
582            to be opened parentheses.
583    
584        (c) Conditions that refer to capturing parentheses can be specified
585            relatively, for example, (?(-2)... or (?(+3)...
586    
587        (d) \K resets the start of the current match so that everything before
588            is not part of it.
589    
590        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
591    
592        (f) \g{name} is another synonym - part of Perl 5.10's unification of
593            reference syntax.
594    
595        (g) (?| introduces a group in which the numbering of parentheses in each
596            alternative starts with the same number.
597    
598        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
599    
600     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
601        PCRE_INFO_JCHANGED.
602    
603     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
604        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
605        for detecting groups that can match an empty string.
606    
607     9. A pattern with a very large number of alternatives (more than several
608        hundred) was running out of internal workspace during the pre-compile
609        phase, where pcre_compile() figures out how much memory will be needed. A
610        bit of new cunning has reduced the workspace needed for groups with
611        alternatives. The 1000-alternative test pattern now uses 12 bytes of
612        workspace instead of running out of the 4096 that are available.
613    
614    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
615    
616    11. Applied patch from Google to remove an optimization that didn't quite work.
617        The report of the bug said:
618    
619          pcrecpp::RE("a*").FullMatch("aaa") matches, while
620          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
621          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
622    
623    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
624        it matched the wrong number of bytes.
625    
626    
627    Version 7.1 24-Apr-07
628    ---------------------
629    
630     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
631        that is more "standard", making use of automake and other Autotools. There
632        is some re-arrangement of the files and adjustment of comments consequent
633        on this.
634    
635     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
636        for recursive directory scanning broke on some systems because the files
637        are not scanned in any specific order and on different systems the order
638        was different. A call to "sort" has been inserted into RunGrepTest for the
639        approprate test as a short-term fix. In the longer term there may be an
640        alternative.
641    
642     3. I had an email from Eric Raymond about problems translating some of PCRE's
643        man pages to HTML (despite the fact that I distribute HTML pages, some
644        people do their own conversions for various reasons). The problems
645        concerned the use of low-level troff macros .br and .in. I have therefore
646        removed all such uses from the man pages (some were redundant, some could
647        be replaced by .nf/.fi pairs). The 132html script that I use to generate
648        HTML has been updated to handle .nf/.fi and to complain if it encounters
649        .br or .in.
650    
651     4. Updated comments in configure.ac that get placed in config.h.in and also
652        arranged for config.h to be included in the distribution, with the name
653        config.h.generic, for the benefit of those who have to compile without
654        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
655    
656     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
657        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
658        makevp.bat for the current PCRE, using the additional files
659        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
660    
661     6. A Windows user reported a minor discrepancy with test 2, which turned out
662        to be caused by a trailing space on an input line that had got lost in his
663        copy. The trailing space was an accident, so I've just removed it.
664    
665     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
666        that is needed.
667    
668     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
669        as "const" (a) because they are and (b) because it helps the PHP
670        maintainers who have recently made a script to detect big data structures
671        in the php code that should be moved to the .rodata section. I remembered
672        to update Builducptable as well, so it won't revert if ucptable.h is ever
673        re-created.
674    
675     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
676        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
677        order to be able to cut out the UTF-8 tables in the latter when UTF-8
678        support is not required. This saves 1.5-2K of code, which is important in
679        some applications.
680    
681        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
682        so as not to refer to the tables, even though these functions will never be
683        called when UTF-8 support is disabled. Otherwise there are problems with a
684        shared library.
685    
686    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
687    
688        (a) It was defining its arguments as char * instead of void *.
689    
690        (b) It was assuming that all moves were upwards in memory; this was true
691            a long time ago when I wrote it, but is no longer the case.
692    
693        The emulated memove() is provided for those environments that have neither
694        memmove() nor bcopy(). I didn't think anyone used it these days, but that
695        is clearly not the case, as these two bugs were recently reported.
696    
697    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
698        and Detrail to create the HTML documentation, the .txt form of the man
699        pages, and it removes trailing spaces from listed files. It also creates
700        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
701        case, it wraps all the #defines with #ifndefs. This script should be run
702        before "make dist".
703    
704    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
705        with Unicode property support.
706    
707        (a) For a maximizing quantifier, if the two different cases of the
708            character were of different lengths in their UTF-8 codings (there are
709            some cases like this - I found 11), and the matching function had to
710            back up over a mixture of the two cases, it incorrectly assumed they
711            were both the same length.
712    
713        (b) When PCRE was configured to use the heap rather than the stack for
714            recursion during matching, it was not correctly preserving the data for
715            the other case of a UTF-8 character when checking ahead for a match
716            while processing a minimizing repeat. If the check also involved
717            matching a wide character, but failed, corruption could cause an
718            erroneous result when trying to check for a repeat of the original
719            character.
720    
721    13. Some tidying changes to the testing mechanism:
722    
723        (a) The RunTest script now detects the internal link size and whether there
724            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
725            values substituted by "configure". (The RunGrepTest script already did
726            this for UTF-8.) The configure.ac script no longer substitutes the
727            relevant variables.
728    
729        (b) The debugging options /B and /D in pcretest show the compiled bytecode
730            with length and offset values. This means that the output is different
731            for different internal link sizes. Test 2 is skipped for link sizes
732            other than 2 because of this, bypassing the problem. Unfortunately,
733            there was also a test in test 3 (the locale tests) that used /B and
734            failed for link sizes other than 2. Rather than cut the whole test out,
735            I have added a new /Z option to pcretest that replaces the length and
736            offset values with spaces. This is now used to make test 3 independent
737            of link size. (Test 2 will be tidied up later.)
738    
739    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
740        segmentation fault instead of returning the appropriate error message.
741    
742    15. In multiline mode when the newline sequence was set to "any", the pattern
743        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
744        This doesn't seem right; it now treats the CRLF combination as the line
745        ending, and so does not match in that case. It's only a pattern such as ^$
746        that would hit this one: something like ^ABC$ would have failed after \r
747        and then tried again after \r\n.
748    
749    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
750        in an attempt to make files that differ only in their line terminators
751        compare equal. This works on Linux.
752    
753    17. Under certain error circumstances pcregrep might try to free random memory
754        as it exited. This is now fixed, thanks to valgrind.
755    
756    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
757        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
758        was because its rules for how to advance for /g after matching an empty
759        string at the end of a line did not allow for this case. They now check for
760        it specially.
761    
762    20. pcretest is supposed to handle patterns and data of any length, by
763        extending its buffers when necessary. It was getting this wrong when the
764        buffer for a data line had to be extended.
765    
766    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
767        CRLF as a newline sequence.
768    
769    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
770        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
771        I have nevertheless tidied it up.
772    
773    23. Added some casts to kill warnings from HP-UX ia64 compiler.
774    
775    24. Added a man page for pcre-config.
776    
777    
778    Version 7.0 19-Dec-06
779    ---------------------
780    
781     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
782        moving to gcc 4.1.1.
783    
784     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
785        sys/time.h, which is documented as needed for this function. It doesn't
786        seem to matter on Linux, but it showed up on some releases of OS X.
787    
788     3. It seems that there are systems where bytes whose values are greater than
789        127 match isprint() in the "C" locale. The "C" locale should be the
790        default when a C program starts up. In most systems, only ASCII printing
791        characters match isprint(). This difference caused the output from pcretest
792        to vary, making some of the tests fail. I have changed pcretest so that:
793    
794        (a) When it is outputting text in the compiled version of a pattern, bytes
795            other than 32-126 are always shown as hex escapes.
796    
797        (b) When it is outputting text that is a matched part of a subject string,
798            it does the same, unless a different locale has been set for the match
799            (using the /L modifier). In this case, it uses isprint() to decide.
800    
801     4. Fixed a major bug that caused incorrect computation of the amount of memory
802        required for a compiled pattern when options that changed within the
803        pattern affected the logic of the preliminary scan that determines the
804        length. The relevant options are -x, and -i in UTF-8 mode. The result was
805        that the computed length was too small. The symptoms of this bug were
806        either the PCRE error "internal error: code overflow" from pcre_compile(),
807        or a glibc crash with a message such as "pcretest: free(): invalid next
808        size (fast)". Examples of patterns that provoked this bug (shown in
809        pcretest format) are:
810    
811          /(?-x: )/x
812          /(?x)(?-x: \s*#\s*)/
813          /((?i)[\x{c0}])/8
814          /(?i:[\x{c0}])/8
815    
816        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
817        is now done differently.
818    
819     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
820        wrapper classes; (b) implement a new function in the C++ scanner that is
821        more efficient than the old way of doing things because it avoids levels of
822        recursion in the regex matching; (c) add a paragraph to the documentation
823        for the FullMatch() function.
824    
825     6. The escape sequence \n was being treated as whatever was defined as
826        "newline". Not only was this contrary to the documentation, which states
827        that \n is character 10 (hex 0A), but it also went horribly wrong when
828        "newline" was defined as CRLF. This has been fixed.
829    
830     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
831        was being set to -1 for the "end of line" case (supposedly a value that no
832        character can have). Though this value is never used (the check for end of
833        line is "zero bytes in current character"), it caused compiler complaints.
834        I've changed it to 0xffffffff.
835    
836     8. In pcre_version.c, the version string was being built by a sequence of
837        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
838        string (as it is for production releases) called a macro with an empty
839        argument. The C standard says the result of this is undefined. The gcc
840        compiler treats it as an empty string (which was what was wanted) but it is
841        reported that Visual C gives an error. The source has been hacked around to
842        avoid this problem.
843    
844     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
845        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
846        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
847        of them did).
848    
849    10. Originally, pcretest opened its input and output without "b"; then I was
850        told that "b" was needed in some environments, so it was added for release
851        5.0 to both the input and output. (It makes no difference on Unix-like
852        systems.) Later I was told that it is wrong for the input on Windows. I've
853        now abstracted the modes into two macros, to make it easier to fiddle with
854        them, and removed "b" from the input mode under Windows.
855    
856    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
857    
858    12. Added -help and --help to pcretest as an official way of being reminded
859        of the options.
860    
861    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
862        and pcrecpp.cc because they annoy compilers at high warning levels.
863    
864    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
865    
866    15. Fixed an occurrence of == in configure.ac that should have been = (shell
867        scripts are not C programs :-) and which was not noticed because it works
868        on Linux.
869    
870    16. pcretest is supposed to handle any length of pattern and data line (as one
871        line or as a continued sequence of lines) by extending its input buffer if
872        necessary. This feature was broken for very long pattern lines, leading to
873        a string of junk being passed to pcre_compile() if the pattern was longer
874        than about 50K.
875    
876    17. I have done a major re-factoring of the way pcre_compile() computes the
877        amount of memory needed for a compiled pattern. Previously, there was code
878        that made a preliminary scan of the pattern in order to do this. That was
879        OK when PCRE was new, but as the facilities have expanded, it has become
880        harder and harder to keep it in step with the real compile phase, and there
881        have been a number of bugs (see for example, 4 above). I have now found a
882        cunning way of running the real compile function in a "fake" mode that
883        enables it to compute how much memory it would need, while actually only
884        ever using a few hundred bytes of working memory and without too many
885        tests of the mode. This should make future maintenance and development
886        easier. A side effect of this work is that the limit of 200 on the nesting
887        depth of parentheses has been removed (though this was never a serious
888        limitation, I suspect). However, there is a downside: pcre_compile() now
889        runs more slowly than before (30% or more, depending on the pattern). I
890        hope this isn't a big issue. There is no effect on runtime performance.
891    
892    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
893        newline (only possible for the last line of a file) and it was a
894        pattern that set a locale (followed by /Lsomething), pcretest crashed.
895    
896    19. Added additional timing features to pcretest. (1) The -tm option now times
897        matching only, not compiling. (2) Both -t and -tm can be followed, as a
898        separate command line item, by a number that specifies the number of
899        repeats to use when timing. The default is 50000; this gives better
900        precision, but takes uncomfortably long for very large patterns.
901    
902    20. Extended pcre_study() to be more clever in cases where a branch of a
903        subpattern has no definite first character. For example, (a*|b*)[cd] would
904        previously give no result from pcre_study(). Now it recognizes that the
905        first character must be a, b, c, or d.
906    
907    21. There was an incorrect error "recursive call could loop indefinitely" if
908        a subpattern (or the entire pattern) that was being tested for matching an
909        empty string contained only one non-empty item after a nested subpattern.
910        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
911        incorrectly, because the \d was being skipped in the check.
912    
913    22. The pcretest program now has a new pattern option /B and a command line
914        option -b, which is equivalent to adding /B to every pattern. This causes
915        it to show the compiled bytecode, without the additional information that
916        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
917        is the same as /B/I).
918    
919    23. A new optimization is now able automatically to treat some sequences such
920        as a*b as a*+b. More specifically, if something simple (such as a character
921        or a simple class like \d) has an unlimited quantifier, and is followed by
922        something that cannot possibly match the quantified thing, the quantifier
923        is automatically "possessified".
924    
925    24. A recursive reference to a subpattern whose number was greater than 39
926        went wrong under certain circumstances in UTF-8 mode. This bug could also
927        have affected the operation of pcre_study().
928    
929    25. Realized that a little bit of performance could be had by replacing
930        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
931    
932    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
933    
934    27. Possessive quantifiers such as a++ were previously implemented by turning
935        them into atomic groups such as ($>a+). Now they have their own opcodes,
936        which improves performance. This includes the automatically created ones
937        from 23 above.
938    
939    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
940        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
941        the first matched character to be a colon. This applied both to named and
942        numbered groups.
943    
944    29. The ucpinternal.h header file was missing its idempotency #ifdef.
945    
946    30. I was sent a "project" file called libpcre.a.dev which I understand makes
947        building PCRE on Windows easier, so I have included it in the distribution.
948    
949    31. There is now a check in pcretest against a ridiculously large number being
950        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
951        loop, the loop is abandoned.
952    
953    32. Forward references to subpatterns in conditions such as (?(2)...) where
954        subpattern 2 is defined later cause pcre_compile() to search forwards in
955        the pattern for the relevant set of parentheses. This search went wrong
956        when there were unescaped parentheses in a character class, parentheses
957        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
958    
959    33. "Subroutine" calls and backreferences were previously restricted to
960        referencing subpatterns earlier in the regex. This restriction has now
961        been removed.
962    
963    34. Added a number of extra features that are going to be in Perl 5.10. On the
964        whole, these are just syntactic alternatives for features that PCRE had
965        previously implemented using the Python syntax or my own invention. The
966        other formats are all retained for compatibility.
967    
968        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
969            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
970            also .NET compatible.
971    
972        (b) A recursion or subroutine call to a named group can now be defined as
973            (?&name) as well as (?P>name).
974    
975        (c) A backreference to a named group can now be defined as \k<name> or
976            \k'name' as well as (?P=name). The new forms, as well as being in Perl
977            5.10, are also .NET compatible.
978    
979        (d) A conditional reference to a named group can now use the syntax
980            (?(<name>) or (?('name') as well as (?(name).
981    
982        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
983            groups (named and numbered) that are never evaluated inline, but can be
984            called as "subroutines" from elsewhere. In effect, the DEFINE condition
985            is always false. There may be only one alternative in such a group.
986    
987        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
988            as the simple (?(R). The condition is true only if the most recent
989            recursion is that of the given number or name. It does not search out
990            through the entire recursion stack.
991    
992        (g) The escape \gN or \g{N} has been added, where N is a positive or
993            negative number, specifying an absolute or relative reference.
994    
995    35. Tidied to get rid of some further signed/unsigned compiler warnings and
996        some "unreachable code" warnings.
997    
998    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
999        things, this adds five new scripts.
1000    
1001    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1002        There were also incompatibilities regarding the handling of \Q..\E inside
1003        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1004        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1005    
1006    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1007        matches an empty string, and forcibly breaks the loop. There were bugs in
1008        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
1009        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
1010        separate and independent bugs (that affected different cases) have been
1011        fixed.
1012    
1013    39. Refactored the code to abolish the use of different opcodes for small
1014        capturing bracket numbers. This is a tidy that I avoided doing when I
1015        removed the limit on the number of capturing brackets for 3.5 back in 2001.
1016        The new approach is not only tidier, it makes it possible to reduce the
1017        memory needed to fix the previous bug (38).
1018    
1019    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1020        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1021        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1022        mode.
1023    
1024    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1025        report.
1026    
1027    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1028        copy construction and assignment in the C++ wrapper.
1029    
1030    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1031        couple of bugs that could have given wrong results in the "--newline=crlf"
1032        case.
1033    
1034    44. Added a number of casts and did some reorganization of signed/unsigned int
1035        variables following suggestions from Dair Grant. Also renamed the variable
1036        "this" as "item" because it is a C++ keyword.
1037    
1038    45. Arranged for dftables to add
1039    
1040          #include "pcre_internal.h"
1041    
1042        to pcre_chartables.c because without it, gcc 4.x may remove the array
1043        definition from the final binary if PCRE is built into a static library and
1044        dead code stripping is activated.
1045    
1046    46. For an unanchored pattern, if a match attempt fails at the start of a
1047        newline sequence, and the newline setting is CRLF or ANY, and the next two
1048        characters are CRLF, advance by two characters instead of one.
1049    
1050    
1051    Version 6.7 04-Jul-06
1052    ---------------------
1053    
1054     1. In order to handle tests when input lines are enormously long, pcretest has
1055        been re-factored so that it automatically extends its buffers when
1056        necessary. The code is crude, but this _is_ just a test program. The
1057        default size has been increased from 32K to 50K.
1058    
1059     2. The code in pcre_study() was using the value of the re argument before
1060        testing it for NULL. (Of course, in any sensible call of the function, it
1061        won't be NULL.)
1062    
1063     3. The memmove() emulation function in pcre_internal.h, which is used on
1064        systems that lack both memmove() and bcopy() - that is, hardly ever -
1065        was missing a "static" storage class specifier.
1066    
1067     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1068        containing an extended class (one that cannot be represented by a bitmap
1069        because it contains high-valued characters or Unicode property items, e.g.
1070        [\pZ]). Almost always one would set UTF-8 mode when processing such a
1071        pattern, but PCRE should not loop if you do not (it no longer does).
1072        [Detail: two cases were found: (a) a repeated subpattern containing an
1073        extended class; (b) a recursive reference to a subpattern that followed a
1074        previous extended class. It wasn't skipping over the extended class
1075        correctly when UTF-8 mode was not set.]
1076    
1077     5. A negated single-character class was not being recognized as fixed-length
1078        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1079        compile error "lookbehind assertion is not fixed length".
1080    
1081     6. The RunPerlTest auxiliary script was showing an unexpected difference
1082        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1083        write a Perl script that can interpret lines of an input file either as
1084        byte characters or as UTF-8, which is what "perltest" was being required to
1085        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1086        can't do is switch easily at run time between having the "use utf8;" pragma
1087        or not. In the end, I fudged it by using the RunPerlTest script to insert
1088        "use utf8;" explicitly for the UTF-8 tests.
1089    
1090     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1091        the end of the subject string, contrary to the documentation and to what
1092        Perl does. This was true of both matching functions. Now it matches only at
1093        the start of the subject and immediately after *internal* newlines.
1094    
1095     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1096        a pointer to an int instead of a pointer to an unsigned long int. This
1097        caused problems on 64-bit systems.
1098    
1099     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1100        instance of the 'standard' template library not being so standard".
1101    
1102    10. There was no check on the number of named subpatterns nor the maximum
1103        length of a subpattern name. The product of these values is used to compute
1104        the size of the memory block for a compiled pattern. By supplying a very
1105        long subpattern name and a large number of named subpatterns, the size
1106        computation could be caused to overflow. This is now prevented by limiting
1107        the length of names to 32 characters, and the number of named subpatterns
1108        to 10,000.
1109    
1110    11. Subpatterns that are repeated with specific counts have to be replicated in
1111        the compiled pattern. The size of memory for this was computed from the
1112        length of the subpattern and the repeat count. The latter is limited to
1113        65535, but there was no limit on the former, meaning that integer overflow
1114        could in principle occur. The compiled length of a repeated subpattern is
1115        now limited to 30,000 bytes in order to prevent this.
1116    
1117    12. Added the optional facility to have named substrings with the same name.
1118    
1119    13. Added the ability to use a named substring as a condition, using the
1120        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1121        are numbers (not recommended). Forward references are permitted.
1122    
1123    14. Added forward references in named backreferences (if you see what I mean).
1124    
1125    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1126        pattern could run off the end of the subject. For example, the pattern
1127        "(?s)(.{1,5})"8 did this with the subject "ab".
1128    
1129    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1130        PCRE_CASELESS was set when matching characters that were quantified with ?
1131        or *.
1132    
1133    17. A character class other than a single negated character that had a minimum
1134        but no maximum quantifier - for example [ab]{6,} - was not handled
1135        correctly by pce_dfa_exec(). It would match only one character.
1136    
1137    18. A valid (though odd) pattern that looked like a POSIX character
1138        class but used an invalid character after [ (for example [[,abc,]]) caused
1139        pcre_compile() to give the error "Failed: internal error: code overflow" or
1140        in some cases to crash with a glibc free() error. This could even happen if
1141        the pattern terminated after [[ but there just happened to be a sequence of
1142        letters, a binary zero, and a closing ] in the memory that followed.
1143    
1144    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1145        over the years. Originally (before any Unicode support), just the bottom 8
1146        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1147        output from "man perlunicode" includes this:
1148    
1149          The regular expression compiler produces polymorphic opcodes.  That
1150          is, the pattern adapts to the data and automatically switches to
1151          the Unicode character scheme when presented with Unicode data--or
1152          instead uses a traditional byte scheme when presented with byte
1153          data.
1154    
1155        Sadly, a wide octal escape does not cause a switch, and in a string with
1156        no other multibyte characters, these octal escapes are treated as before.
1157        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1158        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1159        Unicode string.
1160    
1161        I have not perpetrated such confusion in PCRE. Up till now, it took just
1162        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1163        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1164        translate to the appropriate multibyte character.
1165    
1166    29. Applied some refactoring to reduce the number of warnings from Microsoft
1167        and Borland compilers. This has included removing the fudge introduced
1168        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1169        a warning about an unused variable.
1170    
1171    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1172        characters since release 4.0, because Perl (from release 5.004) does not.
1173        [Or at least, is documented not to: some releases seem to be in conflict
1174        with the documentation.] However, when a pattern was studied with
1175        pcre_study() and all its branches started with \s, PCRE still included VT
1176        as a possible starting character. Of course, this did no harm; it just
1177        caused an unnecessary match attempt.
1178    
1179    22. Removed a now-redundant internal flag bit that recorded the fact that case
1180        dependency changed within the pattern. This was once needed for "required
1181        byte" processing, but is no longer used. This recovers a now-scarce options
1182        bit. Also moved the least significant internal flag bit to the most-
1183        significant bit of the word, which was not previously used (hangover from
1184        the days when it was an int rather than a uint) to free up another bit for
1185        the future.
1186    
1187    23. Added support for CRLF line endings as well as CR and LF. As well as the
1188        default being selectable at build time, it can now be changed at runtime
1189        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1190        specify that it is scanning data with non-default line endings.
1191    
1192    24. Changed the definition of CXXLINK to make it agree with the definition of
1193        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1194    
1195    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1196        recursions. This makes a big different to stack usage for some patterns.
1197    
1198    26. If a subpattern containing a named recursion or subroutine reference such
1199        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1200        the space required for the compiled pattern went wrong and gave too small a
1201        value. Depending on the environment, this could lead to "Failed: internal
1202        error: code overflow at offset 49" or "glibc detected double free or
1203        corruption" errors.
1204    
1205    27. Applied patches from Google (a) to support the new newline modes and (b) to
1206        advance over multibyte UTF-8 characters in GlobalReplace.
1207    
1208    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1209        difference for some implementation of PCRE in some Windows version.
1210    
1211    29. Added some extra testing facilities to pcretest:
1212    
1213        \q<number>   in a data line sets the "match limit" value
1214        \Q<number>   in a data line sets the "match recursion limt" value
1215        -S <number>  sets the stack size, where <number> is in megabytes
1216    
1217        The -S option isn't available for Windows.
1218    
1219    
1220    Version 6.6 06-Feb-06
1221    ---------------------
1222    
1223     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1224        in pcreposix.h. I have copied the definition from pcre.h.
1225    
1226     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1227        because pcre.h is no longer a built file.
1228    
1229     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1230        not normally included in the compiled code.
1231    
1232    
1233    Version 6.5 01-Feb-06
1234    ---------------------
1235    
1236     1. When using the partial match feature with pcre_dfa_exec(), it was not
1237        anchoring the second and subsequent partial matches at the new starting
1238        point. This could lead to incorrect results. For example, with the pattern
1239        /1234/, partially matching against "123" and then "a4" gave a match.
1240    
1241     2. Changes to pcregrep:
1242    
1243        (a) All non-match returns from pcre_exec() were being treated as failures
1244            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1245            error message is output. Some extra information is given for the
1246            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1247            probably the only errors that are likely to be caused by users (by
1248            specifying a regex that has nested indefinite repeats, for instance).
1249            If there are more than 20 of these errors, pcregrep is abandoned.
1250    
1251        (b) A binary zero was treated as data while matching, but terminated the
1252            output line if it was written out. This has been fixed: binary zeroes
1253            are now no different to any other data bytes.
1254    
1255        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1256            used to set a locale for matching. The --locale=xxxx long option has
1257            been added (no short equivalent) to specify a locale explicitly on the
1258            pcregrep command, overriding the environment variables.
1259    
1260        (d) When -B was used with -n, some line numbers in the output were one less
1261            than they should have been.
1262    
1263        (e) Added the -o (--only-matching) option.
1264    
1265        (f) If -A or -C was used with -c (count only), some lines of context were
1266            accidentally printed for the final match.
1267    
1268        (g) Added the -H (--with-filename) option.
1269    
1270        (h) The combination of options -rh failed to suppress file names for files
1271            that were found from directory arguments.
1272    
1273        (i) Added the -D (--devices) and -d (--directories) options.
1274    
1275        (j) Added the -F (--fixed-strings) option.
1276    
1277        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1278    
1279        (l) Added the --colo(u)r option.
1280    
1281        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1282            is not present by default.
1283    
1284     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1285        items such as (?R) or (?1), when the recursion could match a number of
1286        alternatives. If it matched one of the alternatives, but subsequently,
1287        outside the recursion, there was a failure, the code tried to back up into
1288        the recursion. However, because of the way PCRE is implemented, this is not
1289        possible, and the result was an incorrect result from the match.
1290    
1291        In order to prevent this happening, the specification of recursion has
1292        been changed so that all such subpatterns are automatically treated as
1293        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1294    
1295     4. I had overlooked the fact that, in some locales, there are characters for
1296        which isalpha() is true but neither isupper() nor islower() are true. In
1297        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1298        and ordfeminine) are like this. This affected the treatment of \w and \W
1299        when they appeared in character classes, but not when they appeared outside
1300        a character class. The bit map for "word" characters is now created
1301        separately from the results of isalnum() instead of just taking it from the
1302        upper, lower, and digit maps. (Plus the underscore character, of course.)
1303    
1304     5. The above bug also affected the handling of POSIX character classes such as
1305        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1306        permanent tables. Instead, the bit maps for such a class were previously
1307        created as the appropriate unions of the upper, lower, and digit bitmaps.
1308        Now they are created by subtraction from the [[:word:]] class, which has
1309        its own bitmap.
1310    
1311     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1312        It is created by subtracting the vertical space characters (\x09, \x0a,
1313        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1314        subtraction was done in the overall bitmap for a character class, meaning
1315        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1316        be recognized. This bug has been fixed.
1317    
1318     7. Patches from the folks at Google:
1319    
1320          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1321          real life, but is still worth protecting against".
1322    
1323          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1324          regular expressions".
1325    
1326          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1327          have it.
1328    
1329          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1330          "configure" and the latter not, in order to fix a problem somebody had
1331          with compiling the Arg class on HP-UX.
1332    
1333          (e) Improve the error-handling of the C++ wrapper a little bit.
1334    
1335          (f) New tests for checking recursion limiting.
1336    
1337     8. The pcre_memmove() function, which is used only if the environment does not
1338        have a standard memmove() function (and is therefore rarely compiled),
1339        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1340        returning a result (though PCRE never actually uses the result).
1341    
1342     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1343        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1344        returned instead of calling malloc() with an overflowing number that would
1345        most likely cause subsequent chaos.
1346    
1347    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1348    
1349    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1350        with this option is matched, the nmatch and pmatch options of regexec() are
1351        ignored.
1352    
1353    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1354        provided in case anyone wants to the the POSIX interface with UTF-8
1355        strings.
1356    
1357    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1358        C++ linking (needed for some HP-UX environments).
1359    
1360    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1361        (unused parameter) and in the pcre_printint() function (omitted "default"
1362        switch label when the default is to do nothing).
1363    
1364    15. Added some code to make it possible, when PCRE is compiled as a C++
1365        library, to replace subject pointers for pcre_exec() with a smart pointer
1366        class, thus making it possible to process discontinuous strings.
1367    
1368    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1369        much the same function. They were added by different people who were trying
1370        to make PCRE easy to compile on non-Unix systems. It has been suggested
1371        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1372        for compiling on Windows systems. I have therefore replaced it with
1373        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1374        defaults to "extern" for C or "extern C" for C++, which works fine on
1375        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1376        SCOPE with something explicit in config.h. In addition:
1377    
1378        (a) pcreposix.h still had just "extern" instead of either of these macros;
1379            I have replaced it with PCRE_DATA_SCOPE.
1380    
1381        (b) Functions such as _pcre_xclass(), which are internal to the library,
1382            but external in the C sense, all had PCRE_EXPORT in their definitions.
1383            This is apparently wrong for the Windows case, so I have removed it.
1384            (It makes no difference on Unix-like systems.)
1385    
1386    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1387        of recursive calls to match(). This is different to MATCH_LIMIT because
1388        that limits the total number of calls to match(), not all of which increase
1389        the depth of recursion. Limiting the recursion depth limits the amount of
1390        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1391        when PCRE is compiled, and changed at run time. A patch from Google adds
1392        this functionality to the C++ interface.
1393    
1394    18. Changes to the handling of Unicode character properties:
1395    
1396        (a) Updated the table to Unicode 4.1.0.
1397    
1398        (b) Recognize characters that are not in the table as "Cn" (undefined).
1399    
1400        (c) I revised the way the table is implemented to a much improved format
1401            which includes recognition of ranges. It now supports the ranges that
1402            are defined in UnicodeData.txt, and it also amalgamates other
1403            characters into ranges. This has reduced the number of entries in the
1404            table from around 16,000 to around 3,000, thus reducing its size
1405            considerably. I realized I did not need to use a tree structure after
1406            all - a binary chop search is just as efficient. Having reduced the
1407            number of entries, I extended their size from 6 bytes to 8 bytes to
1408            allow for more data.
1409    
1410        (d) Added support for Unicode script names via properties such as \p{Han}.
1411    
1412    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1413        matching that character.
1414    
1415    20. When matching a repeated Unicode property with a minimum greater than zero,
1416        (for example \pL{2,}), PCRE could look past the end of the subject if it
1417        reached it while seeking the minimum number of characters. This could
1418        happen only if some of the characters were more than one byte long, because
1419        there is a check for at least the minimum number of bytes.
1420    
1421    21. Refactored the implementation of \p and \P so as to be more general, to
1422        allow for more different types of property in future. This has changed the
1423        compiled form incompatibly. Anybody with saved compiled patterns that use
1424        \p or \P will have to recompile them.
1425    
1426    22. Added "Any" and "L&" to the supported property types.
1427    
1428    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1429        but give a compile time error if the value is greater than 0xff.
1430    
1431    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1432        accidentally not being installed or uninstalled.
1433    
1434    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1435        made were to insert the current release number. This seemed silly, because
1436        it made things harder for people building PCRE on systems that don't run
1437        "configure". I have turned pcre.h into a distributed file, no longer built
1438        by "configure", with the version identification directly included. There is
1439        no longer a pcre.h.in file.
1440    
1441        However, this change necessitated a change to the pcre-config script as
1442        well. It is built from pcre-config.in, and one of the substitutions was the
1443        release number. I have updated configure.ac so that ./configure now finds
1444        the release number by grepping pcre.h.
1445    
1446    26. Added the ability to run the tests under valgrind.
1447    
1448    
1449    Version 6.4 05-Sep-05
1450    ---------------------
1451    
1452     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1453        "--" to be printed when multiple files were scanned, even when none of the
1454        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1455        consider it to be a bug, and have restored the previous behaviour.
1456    
1457     2. A couple of code tidies to get rid of compiler warnings.
1458    
1459     3. The pcretest program used to cheat by referring to symbols in the library
1460        whose names begin with _pcre_. These are internal symbols that are not
1461        really supposed to be visible externally, and in some environments it is
1462        possible to suppress them. The cheating is now confined to including
1463        certain files from the library's source, which is a bit cleaner.
1464    
1465     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1466        file's purpose clearer.
1467    
1468     5. Reorganized pcre_ucp_findchar().
1469    
1470    
1471    Version 6.3 15-Aug-05
1472    ---------------------
1473    
1474     1. The file libpcre.pc.in did not have general read permission in the tarball.
1475    
1476     2. There were some problems when building without C++ support:
1477    
1478        (a) If C++ support was not built, "make install" and "make test" still
1479            tried to test it.
1480    
1481        (b) There were problems when the value of CXX was explicitly set. Some
1482            changes have been made to try to fix these, and ...
1483    
1484        (c) --disable-cpp can now be used to explicitly disable C++ support.
1485    
1486        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1487            backslash in a target when C++ was disabled. This confuses some
1488            versions of "make", apparently. Using an intermediate variable solves
1489            this. (Same for CPP_LOBJ.)
1490    
1491     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1492        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1493        necessary on certain architectures.
1494    
1495     4. Added a setting of -export-symbols-regex to the link command to remove
1496        those symbols that are exported in the C sense, but actually are local
1497        within the library, and not documented. Their names all begin with
1498        "_pcre_". This is not a perfect job, because (a) we have to except some
1499        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1500        available (and never for static libraries). I have made a note to try to
1501        find a way round (a) in the future.
1502    
1503    
1504    Version 6.2 01-Aug-05
1505    ---------------------
1506    
1507     1. There was no test for integer overflow of quantifier values. A construction
1508        such as {1111111111111111} would give undefined results. What is worse, if
1509        a minimum quantifier for a parenthesized subpattern overflowed and became
1510        negative, the calculation of the memory size went wrong. This could have
1511        led to memory overwriting.
1512    
1513     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1514    
1515     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1516        operating environments where this matters.
1517    
1518     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1519        PCRE options from within the C++ wrapper.
1520    
1521     5. Named capturing subpatterns were not being correctly counted when a pattern
1522        was compiled. This caused two problems: (a) If there were more than 100
1523        such subpatterns, the calculation of the memory needed for the whole
1524        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1525        back references of the form \12, where the number was greater than 9, were
1526        not recognized as back references, even though there were sufficient
1527        previous subpatterns.
1528    
1529     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1530        versions of gcc, e.g. 2.95.4.
1531    
1532    
1533  Version 6.1 21-Jun-05  Version 6.1 21-Jun-05
1534  ---------------------  ---------------------
1535    

Legend:
Removed from v.79  
changed lines
  Added in v.353

  ViewVC Help
Powered by ViewVC 1.1.5