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

Removed from v.89  
changed lines
  Added in v.93

  ViewVC Help
Powered by ViewVC 1.1.5