/[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 107 by ph10, Wed Mar 7 11:02:28 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.1 05-Mar-07
5    ---------------------
6    
7     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
8        that is more "standard", making use of automake and other autotools. There
9        is some re-arrangement of the files and adjustment of comments consequent
10        on this.
11    
12     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
13        for recursive directory scanning broke on some systems because the files
14        are not scanned in any specific order and on different systems the order
15        was different. A call to "sort" has been inserted into RunGrepTest for the
16        approprate test as a short-term fix. In the longer term there may be an
17        alternative.
18    
19     3. I had an email from Eric Raymond about problems translating some of PCRE's
20        man pages to HTML (despite the fact that I distribute HTML pages, some
21        people do their own conversions for various reasons). The problems
22        concerned the use of low-level troff macros .br and .in. I have therefore
23        removed all such uses from the man pages (some were redundant, some could
24        be replaced by .nf/.fi pairs). The maintain/132html script that I use to
25        generate HTML has been updated to handle .nf/.fi and to complain if it
26        encounters .br or .in.
27    
28     4. Updated comments in configure.ac that get placed in config.h.in and also
29        arranged for config.h to be included in the distribution, for the benefit
30        of those who have to compile without Autotools (compare pcre.h).
31    
32     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
33        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
34        makevp.bat for the current PCRE, using the additional files !compile.txt,
35        !linklib.txt, and pcregexp.pas.
36    
37     6. A Windows user reported a minor discrepancy with test 2, which turned out
38        to be caused by a trailing space on an input line that had got lost in his
39        copy. The trailing space was an accident, so I've just removed it.
40    
41     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
42        that is needed.
43    
44     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
45        as "const" (a) because they are and (b) because it helps the PHP
46        maintainers who have recently made a script to detect big data structures
47        in the php code that should be moved to the .rodata section. I remembered
48        to update Builducptable as well, so it won't revert if ucptable.h is ever
49        re-created.
50    
51     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
52        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
53        order to be able to cut out the UTF-8 tables in the latter when UTF-8
54        support is not required. This saves 1.5-2K of code, which is important in
55        some applications.
56    
57    
58    Version 7.0 19-Dec-06
59    ---------------------
60    
61     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
62        moving to gcc 4.1.1.
63    
64     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
65        sys/time.h, which is documented as needed for this function. It doesn't
66        seem to matter on Linux, but it showed up on some releases of OS X.
67    
68     3. It seems that there are systems where bytes whose values are greater than
69        127 match isprint() in the "C" locale. The "C" locale should be the
70        default when a C program starts up. In most systems, only ASCII printing
71        characters match isprint(). This difference caused the output from pcretest
72        to vary, making some of the tests fail. I have changed pcretest so that:
73    
74        (a) When it is outputting text in the compiled version of a pattern, bytes
75            other than 32-126 are always shown as hex escapes.
76    
77        (b) When it is outputting text that is a matched part of a subject string,
78            it does the same, unless a different locale has been set for the match
79            (using the /L modifier). In this case, it uses isprint() to decide.
80    
81     4. Fixed a major bug that caused incorrect computation of the amount of memory
82        required for a compiled pattern when options that changed within the
83        pattern affected the logic of the preliminary scan that determines the
84        length. The relevant options are -x, and -i in UTF-8 mode. The result was
85        that the computed length was too small. The symptoms of this bug were
86        either the PCRE error "internal error: code overflow" from pcre_compile(),
87        or a glibc crash with a message such as "pcretest: free(): invalid next
88        size (fast)". Examples of patterns that provoked this bug (shown in
89        pcretest format) are:
90    
91          /(?-x: )/x
92          /(?x)(?-x: \s*#\s*)/
93          /((?i)[\x{c0}])/8
94          /(?i:[\x{c0}])/8
95    
96        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
97        is now done differently.
98    
99     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
100        wrapper classes; (b) implement a new function in the C++ scanner that is
101        more efficient than the old way of doing things because it avoids levels of
102        recursion in the regex matching; (c) add a paragraph to the documentation
103        for the FullMatch() function.
104    
105     6. The escape sequence \n was being treated as whatever was defined as
106        "newline". Not only was this contrary to the documentation, which states
107        that \n is character 10 (hex 0A), but it also went horribly wrong when
108        "newline" was defined as CRLF. This has been fixed.
109    
110     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
111        was being set to -1 for the "end of line" case (supposedly a value that no
112        character can have). Though this value is never used (the check for end of
113        line is "zero bytes in current character"), it caused compiler complaints.
114        I've changed it to 0xffffffff.
115    
116     8. In pcre_version.c, the version string was being built by a sequence of
117        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
118        string (as it is for production releases) called a macro with an empty
119        argument. The C standard says the result of this is undefined. The gcc
120        compiler treats it as an empty string (which was what was wanted) but it is
121        reported that Visual C gives an error. The source has been hacked around to
122        avoid this problem.
123    
124     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
125        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
126        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
127        of them did).
128    
129    10. Originally, pcretest opened its input and output without "b"; then I was
130        told that "b" was needed in some environments, so it was added for release
131        5.0 to both the input and output. (It makes no difference on Unix-like
132        systems.) Later I was told that it is wrong for the input on Windows. I've
133        now abstracted the modes into two macros, to make it easier to fiddle with
134        them, and removed "b" from the input mode under Windows.
135    
136    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
137    
138    12. Added -help and --help to pcretest as an official way of being reminded
139        of the options.
140    
141    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
142        and pcrecpp.cc because they annoy compilers at high warning levels.
143    
144    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
145    
146    15. Fixed an occurrence of == in configure.ac that should have been = (shell
147        scripts are not C programs :-) and which was not noticed because it works
148        on Linux.
149    
150    16. pcretest is supposed to handle any length of pattern and data line (as one
151        line or as a continued sequence of lines) by extending its input buffer if
152        necessary. This feature was broken for very long pattern lines, leading to
153        a string of junk being passed to pcre_compile() if the pattern was longer
154        than about 50K.
155    
156    17. I have done a major re-factoring of the way pcre_compile() computes the
157        amount of memory needed for a compiled pattern. Previously, there was code
158        that made a preliminary scan of the pattern in order to do this. That was
159        OK when PCRE was new, but as the facilities have expanded, it has become
160        harder and harder to keep it in step with the real compile phase, and there
161        have been a number of bugs (see for example, 4 above). I have now found a
162        cunning way of running the real compile function in a "fake" mode that
163        enables it to compute how much memory it would need, while actually only
164        ever using a few hundred bytes of working memory and without too many
165        tests of the mode. This should make future maintenance and development
166        easier. A side effect of this work is that the limit of 200 on the nesting
167        depth of parentheses has been removed (though this was never a serious
168        limitation, I suspect). However, there is a downside: pcre_compile() now
169        runs more slowly than before (30% or more, depending on the pattern). I
170        hope this isn't a big issue. There is no effect on runtime performance.
171    
172    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
173        newline (only possible for the last line of a file) and it was a
174        pattern that set a locale (followed by /Lsomething), pcretest crashed.
175    
176    19. Added additional timing features to pcretest. (1) The -tm option now times
177        matching only, not compiling. (2) Both -t and -tm can be followed, as a
178        separate command line item, by a number that specifies the number of
179        repeats to use when timing. The default is 50000; this gives better
180        precision, but takes uncomfortably long for very large patterns.
181    
182    20. Extended pcre_study() to be more clever in cases where a branch of a
183        subpattern has no definite first character. For example, (a*|b*)[cd] would
184        previously give no result from pcre_study(). Now it recognizes that the
185        first character must be a, b, c, or d.
186    
187    21. There was an incorrect error "recursive call could loop indefinitely" if
188        a subpattern (or the entire pattern) that was being tested for matching an
189        empty string contained only one non-empty item after a nested subpattern.
190        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
191        incorrectly, because the \d was being skipped in the check.
192    
193    22. The pcretest program now has a new pattern option /B and a command line
194        option -b, which is equivalent to adding /B to every pattern. This causes
195        it to show the compiled bytecode, without the additional information that
196        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
197        is the same as /B/I).
198    
199    23. A new optimization is now able automatically to treat some sequences such
200        as a*b as a*+b. More specifically, if something simple (such as a character
201        or a simple class like \d) has an unlimited quantifier, and is followed by
202        something that cannot possibly match the quantified thing, the quantifier
203        is automatically "possessified".
204    
205    24. A recursive reference to a subpattern whose number was greater than 39
206        went wrong under certain circumstances in UTF-8 mode. This bug could also
207        have affected the operation of pcre_study().
208    
209    25. Realized that a little bit of performance could be had by replacing
210        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
211    
212    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
213    
214    27. Possessive quantifiers such as a++ were previously implemented by turning
215        them into atomic groups such as ($>a+). Now they have their own opcodes,
216        which improves performance. This includes the automatically created ones
217        from 23 above.
218    
219    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
220        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
221        the first matched character to be a colon. This applied both to named and
222        numbered groups.
223    
224    29. The ucpinternal.h header file was missing its idempotency #ifdef.
225    
226    30. I was sent a "project" file called libpcre.a.dev which I understand makes
227        building PCRE on Windows easier, so I have included it in the distribution.
228    
229    31. There is now a check in pcretest against a ridiculously large number being
230        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
231        loop, the loop is abandoned.
232    
233    32. Forward references to subpatterns in conditions such as (?(2)...) where
234        subpattern 2 is defined later cause pcre_compile() to search forwards in
235        the pattern for the relevant set of parentheses. This search went wrong
236        when there were unescaped parentheses in a character class, parentheses
237        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
238    
239    33. "Subroutine" calls and backreferences were previously restricted to
240        referencing subpatterns earlier in the regex. This restriction has now
241        been removed.
242    
243    34. Added a number of extra features that are going to be in Perl 5.10. On the
244        whole, these are just syntactic alternatives for features that PCRE had
245        previously implemented using the Python syntax or my own invention. The
246        other formats are all retained for compatibility.
247    
248        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
249            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
250            also .NET compatible.
251    
252        (b) A recursion or subroutine call to a named group can now be defined as
253            (?&name) as well as (?P>name).
254    
255        (c) A backreference to a named group can now be defined as \k<name> or
256            \k'name' as well as (?P=name). The new forms, as well as being in Perl
257            5.10, are also .NET compatible.
258    
259        (d) A conditional reference to a named group can now use the syntax
260            (?(<name>) or (?('name') as well as (?(name).
261    
262        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
263            groups (named and numbered) that are never evaluated inline, but can be
264            called as "subroutines" from elsewhere. In effect, the DEFINE condition
265            is always false. There may be only one alternative in such a group.
266    
267        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
268            as the simple (?(R). The condition is true only if the most recent
269            recursion is that of the given number or name. It does not search out
270            through the entire recursion stack.
271    
272        (g) The escape \gN or \g{N} has been added, where N is a positive or
273            negative number, specifying an absolute or relative reference.
274    
275    35. Tidied to get rid of some further signed/unsigned compiler warnings and
276        some "unreachable code" warnings.
277    
278    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
279        things, this adds five new scripts.
280    
281    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
282        There were also incompatibilities regarding the handling of \Q..\E inside
283        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
284        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
285    
286    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
287        matches an empty string, and forcibly breaks the loop. There were bugs in
288        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
289        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
290        separate and independent bugs (that affected different cases) have been
291        fixed.
292    
293    39. Refactored the code to abolish the use of different opcodes for small
294        capturing bracket numbers. This is a tidy that I avoided doing when I
295        removed the limit on the number of capturing brackets for 3.5 back in 2001.
296        The new approach is not only tidier, it makes it possible to reduce the
297        memory needed to fix the previous bug (38).
298    
299    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
300        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
301        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
302        mode.
303    
304    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
305        report.
306    
307    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
308        copy construction and assignment in the C++ wrapper.
309    
310    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
311        couple of bugs that could have given wrong results in the "--newline=crlf"
312        case.
313    
314    44. Added a number of casts and did some reorganization of signed/unsigned int
315        variables following suggestions from Dair Grant. Also renamed the variable
316        "this" as "item" because it is a C++ keyword.
317    
318    45. Arranged for dftables to add
319    
320          #include "pcre_internal.h"
321    
322        to pcre_chartables.c because without it, gcc 4.x may remove the array
323        definition from the final binary if PCRE is built into a static library and
324        dead code stripping is activated.
325    
326    46. For an unanchored pattern, if a match attempt fails at the start of a
327        newline sequence, and the newline setting is CRLF or ANY, and the next two
328        characters are CRLF, advance by two characters instead of one.
329    
330    
331    Version 6.7 04-Jul-06
332    ---------------------
333    
334     1. In order to handle tests when input lines are enormously long, pcretest has
335        been re-factored so that it automatically extends its buffers when
336        necessary. The code is crude, but this _is_ just a test program. The
337        default size has been increased from 32K to 50K.
338    
339     2. The code in pcre_study() was using the value of the re argument before
340        testing it for NULL. (Of course, in any sensible call of the function, it
341        won't be NULL.)
342    
343     3. The memmove() emulation function in pcre_internal.h, which is used on
344        systems that lack both memmove() and bcopy() - that is, hardly ever -
345        was missing a "static" storage class specifier.
346    
347     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
348        containing an extended class (one that cannot be represented by a bitmap
349        because it contains high-valued characters or Unicode property items, e.g.
350        [\pZ]). Almost always one would set UTF-8 mode when processing such a
351        pattern, but PCRE should not loop if you do not (it no longer does).
352        [Detail: two cases were found: (a) a repeated subpattern containing an
353        extended class; (b) a recursive reference to a subpattern that followed a
354        previous extended class. It wasn't skipping over the extended class
355        correctly when UTF-8 mode was not set.]
356    
357     5. A negated single-character class was not being recognized as fixed-length
358        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
359        compile error "lookbehind assertion is not fixed length".
360    
361     6. The RunPerlTest auxiliary script was showing an unexpected difference
362        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
363        write a Perl script that can interpret lines of an input file either as
364        byte characters or as UTF-8, which is what "perltest" was being required to
365        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
366        can't do is switch easily at run time between having the "use utf8;" pragma
367        or not. In the end, I fudged it by using the RunPerlTest script to insert
368        "use utf8;" explicitly for the UTF-8 tests.
369    
370     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
371        the end of the subject string, contrary to the documentation and to what
372        Perl does. This was true of both matching functions. Now it matches only at
373        the start of the subject and immediately after *internal* newlines.
374    
375     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
376        a pointer to an int instead of a pointer to an unsigned long int. This
377        caused problems on 64-bit systems.
378    
379     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
380        instance of the 'standard' template library not being so standard".
381    
382    10. There was no check on the number of named subpatterns nor the maximum
383        length of a subpattern name. The product of these values is used to compute
384        the size of the memory block for a compiled pattern. By supplying a very
385        long subpattern name and a large number of named subpatterns, the size
386        computation could be caused to overflow. This is now prevented by limiting
387        the length of names to 32 characters, and the number of named subpatterns
388        to 10,000.
389    
390    11. Subpatterns that are repeated with specific counts have to be replicated in
391        the compiled pattern. The size of memory for this was computed from the
392        length of the subpattern and the repeat count. The latter is limited to
393        65535, but there was no limit on the former, meaning that integer overflow
394        could in principle occur. The compiled length of a repeated subpattern is
395        now limited to 30,000 bytes in order to prevent this.
396    
397    12. Added the optional facility to have named substrings with the same name.
398    
399    13. Added the ability to use a named substring as a condition, using the
400        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
401        are numbers (not recommended). Forward references are permitted.
402    
403    14. Added forward references in named backreferences (if you see what I mean).
404    
405    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
406        pattern could run off the end of the subject. For example, the pattern
407        "(?s)(.{1,5})"8 did this with the subject "ab".
408    
409    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
410        PCRE_CASELESS was set when matching characters that were quantified with ?
411        or *.
412    
413    17. A character class other than a single negated character that had a minimum
414        but no maximum quantifier - for example [ab]{6,} - was not handled
415        correctly by pce_dfa_exec(). It would match only one character.
416    
417    18. A valid (though odd) pattern that looked like a POSIX character
418        class but used an invalid character after [ (for example [[,abc,]]) caused
419        pcre_compile() to give the error "Failed: internal error: code overflow" or
420        in some cases to crash with a glibc free() error. This could even happen if
421        the pattern terminated after [[ but there just happened to be a sequence of
422        letters, a binary zero, and a closing ] in the memory that followed.
423    
424    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
425        over the years. Originally (before any Unicode support), just the bottom 8
426        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
427        output from "man perlunicode" includes this:
428    
429          The regular expression compiler produces polymorphic opcodes.  That
430          is, the pattern adapts to the data and automatically switches to
431          the Unicode character scheme when presented with Unicode data--or
432          instead uses a traditional byte scheme when presented with byte
433          data.
434    
435        Sadly, a wide octal escape does not cause a switch, and in a string with
436        no other multibyte characters, these octal escapes are treated as before.
437        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
438        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
439        Unicode string.
440    
441        I have not perpetrated such confusion in PCRE. Up till now, it took just
442        the bottom 8 bits, as in old Perl. I have now made octal escapes with
443        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
444        translate to the appropriate multibyte character.
445    
446    29. Applied some refactoring to reduce the number of warnings from Microsoft
447        and Borland compilers. This has included removing the fudge introduced
448        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
449        a warning about an unused variable.
450    
451    21. PCRE has not included VT (character 0x0b) in the set of whitespace
452        characters since release 4.0, because Perl (from release 5.004) does not.
453        [Or at least, is documented not to: some releases seem to be in conflict
454        with the documentation.] However, when a pattern was studied with
455        pcre_study() and all its branches started with \s, PCRE still included VT
456        as a possible starting character. Of course, this did no harm; it just
457        caused an unnecessary match attempt.
458    
459    22. Removed a now-redundant internal flag bit that recorded the fact that case
460        dependency changed within the pattern. This was once needed for "required
461        byte" processing, but is no longer used. This recovers a now-scarce options
462        bit. Also moved the least significant internal flag bit to the most-
463        significant bit of the word, which was not previously used (hangover from
464        the days when it was an int rather than a uint) to free up another bit for
465        the future.
466    
467    23. Added support for CRLF line endings as well as CR and LF. As well as the
468        default being selectable at build time, it can now be changed at runtime
469        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
470        specify that it is scanning data with non-default line endings.
471    
472    24. Changed the definition of CXXLINK to make it agree with the definition of
473        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
474    
475    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
476        recursions. This makes a big different to stack usage for some patterns.
477    
478    26. If a subpattern containing a named recursion or subroutine reference such
479        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
480        the space required for the compiled pattern went wrong and gave too small a
481        value. Depending on the environment, this could lead to "Failed: internal
482        error: code overflow at offset 49" or "glibc detected double free or
483        corruption" errors.
484    
485    27. Applied patches from Google (a) to support the new newline modes and (b) to
486        advance over multibyte UTF-8 characters in GlobalReplace.
487    
488    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
489        difference for some implementation of PCRE in some Windows version.
490    
491    29. Added some extra testing facilities to pcretest:
492    
493        \q<number>   in a data line sets the "match limit" value
494        \Q<number>   in a data line sets the "match recursion limt" value
495        -S <number>  sets the stack size, where <number> is in megabytes
496    
497        The -S option isn't available for Windows.
498    
499    
500  Version 6.6 06-Feb-06  Version 6.6 06-Feb-06
501  ---------------------  ---------------------
502    

Legend:
Removed from v.89  
changed lines
  Added in v.107

  ViewVC Help
Powered by ViewVC 1.1.5