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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.89  
changed lines
  Added in v.349

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12