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

Diff of /code/trunk/ChangeLog

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

revision 63 by nigel, Sat Feb 24 21:40:03 2007 UTC revision 101 by ph10, Tue Mar 6 15:19:44 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4  Version 4.00 17-Feb-03  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    
38    Version 7.0 19-Dec-06
39    ---------------------
40    
41     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
42        moving to gcc 4.1.1.
43    
44     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
45        sys/time.h, which is documented as needed for this function. It doesn't
46        seem to matter on Linux, but it showed up on some releases of OS X.
47    
48     3. It seems that there are systems where bytes whose values are greater than
49        127 match isprint() in the "C" locale. The "C" locale should be the
50        default when a C program starts up. In most systems, only ASCII printing
51        characters match isprint(). This difference caused the output from pcretest
52        to vary, making some of the tests fail. I have changed pcretest so that:
53    
54        (a) When it is outputting text in the compiled version of a pattern, bytes
55            other than 32-126 are always shown as hex escapes.
56    
57        (b) When it is outputting text that is a matched part of a subject string,
58            it does the same, unless a different locale has been set for the match
59            (using the /L modifier). In this case, it uses isprint() to decide.
60    
61     4. Fixed a major bug that caused incorrect computation of the amount of memory
62        required for a compiled pattern when options that changed within the
63        pattern affected the logic of the preliminary scan that determines the
64        length. The relevant options are -x, and -i in UTF-8 mode. The result was
65        that the computed length was too small. The symptoms of this bug were
66        either the PCRE error "internal error: code overflow" from pcre_compile(),
67        or a glibc crash with a message such as "pcretest: free(): invalid next
68        size (fast)". Examples of patterns that provoked this bug (shown in
69        pcretest format) are:
70    
71          /(?-x: )/x
72          /(?x)(?-x: \s*#\s*)/
73          /((?i)[\x{c0}])/8
74          /(?i:[\x{c0}])/8
75    
76        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
77        is now done differently.
78    
79     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
80        wrapper classes; (b) implement a new function in the C++ scanner that is
81        more efficient than the old way of doing things because it avoids levels of
82        recursion in the regex matching; (c) add a paragraph to the documentation
83        for the FullMatch() function.
84    
85     6. The escape sequence \n was being treated as whatever was defined as
86        "newline". Not only was this contrary to the documentation, which states
87        that \n is character 10 (hex 0A), but it also went horribly wrong when
88        "newline" was defined as CRLF. This has been fixed.
89    
90     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
91        was being set to -1 for the "end of line" case (supposedly a value that no
92        character can have). Though this value is never used (the check for end of
93        line is "zero bytes in current character"), it caused compiler complaints.
94        I've changed it to 0xffffffff.
95    
96     8. In pcre_version.c, the version string was being built by a sequence of
97        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
98        string (as it is for production releases) called a macro with an empty
99        argument. The C standard says the result of this is undefined. The gcc
100        compiler treats it as an empty string (which was what was wanted) but it is
101        reported that Visual C gives an error. The source has been hacked around to
102        avoid this problem.
103    
104     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
105        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
106        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
107        of them did).
108    
109    10. Originally, pcretest opened its input and output without "b"; then I was
110        told that "b" was needed in some environments, so it was added for release
111        5.0 to both the input and output. (It makes no difference on Unix-like
112        systems.) Later I was told that it is wrong for the input on Windows. I've
113        now abstracted the modes into two macros, to make it easier to fiddle with
114        them, and removed "b" from the input mode under Windows.
115    
116    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
117    
118    12. Added -help and --help to pcretest as an official way of being reminded
119        of the options.
120    
121    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
122        and pcrecpp.cc because they annoy compilers at high warning levels.
123    
124    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
125    
126    15. Fixed an occurrence of == in configure.ac that should have been = (shell
127        scripts are not C programs :-) and which was not noticed because it works
128        on Linux.
129    
130    16. pcretest is supposed to handle any length of pattern and data line (as one
131        line or as a continued sequence of lines) by extending its input buffer if
132        necessary. This feature was broken for very long pattern lines, leading to
133        a string of junk being passed to pcre_compile() if the pattern was longer
134        than about 50K.
135    
136    17. I have done a major re-factoring of the way pcre_compile() computes the
137        amount of memory needed for a compiled pattern. Previously, there was code
138        that made a preliminary scan of the pattern in order to do this. That was
139        OK when PCRE was new, but as the facilities have expanded, it has become
140        harder and harder to keep it in step with the real compile phase, and there
141        have been a number of bugs (see for example, 4 above). I have now found a
142        cunning way of running the real compile function in a "fake" mode that
143        enables it to compute how much memory it would need, while actually only
144        ever using a few hundred bytes of working memory and without too many
145        tests of the mode. This should make future maintenance and development
146        easier. A side effect of this work is that the limit of 200 on the nesting
147        depth of parentheses has been removed (though this was never a serious
148        limitation, I suspect). However, there is a downside: pcre_compile() now
149        runs more slowly than before (30% or more, depending on the pattern). I
150        hope this isn't a big issue. There is no effect on runtime performance.
151    
152    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
153        newline (only possible for the last line of a file) and it was a
154        pattern that set a locale (followed by /Lsomething), pcretest crashed.
155    
156    19. Added additional timing features to pcretest. (1) The -tm option now times
157        matching only, not compiling. (2) Both -t and -tm can be followed, as a
158        separate command line item, by a number that specifies the number of
159        repeats to use when timing. The default is 50000; this gives better
160        precision, but takes uncomfortably long for very large patterns.
161    
162    20. Extended pcre_study() to be more clever in cases where a branch of a
163        subpattern has no definite first character. For example, (a*|b*)[cd] would
164        previously give no result from pcre_study(). Now it recognizes that the
165        first character must be a, b, c, or d.
166    
167    21. There was an incorrect error "recursive call could loop indefinitely" if
168        a subpattern (or the entire pattern) that was being tested for matching an
169        empty string contained only one non-empty item after a nested subpattern.
170        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
171        incorrectly, because the \d was being skipped in the check.
172    
173    22. The pcretest program now has a new pattern option /B and a command line
174        option -b, which is equivalent to adding /B to every pattern. This causes
175        it to show the compiled bytecode, without the additional information that
176        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
177        is the same as /B/I).
178    
179    23. A new optimization is now able automatically to treat some sequences such
180        as a*b as a*+b. More specifically, if something simple (such as a character
181        or a simple class like \d) has an unlimited quantifier, and is followed by
182        something that cannot possibly match the quantified thing, the quantifier
183        is automatically "possessified".
184    
185    24. A recursive reference to a subpattern whose number was greater than 39
186        went wrong under certain circumstances in UTF-8 mode. This bug could also
187        have affected the operation of pcre_study().
188    
189    25. Realized that a little bit of performance could be had by replacing
190        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
191    
192    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
193    
194    27. Possessive quantifiers such as a++ were previously implemented by turning
195        them into atomic groups such as ($>a+). Now they have their own opcodes,
196        which improves performance. This includes the automatically created ones
197        from 23 above.
198    
199    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
200        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
201        the first matched character to be a colon. This applied both to named and
202        numbered groups.
203    
204    29. The ucpinternal.h header file was missing its idempotency #ifdef.
205    
206    30. I was sent a "project" file called libpcre.a.dev which I understand makes
207        building PCRE on Windows easier, so I have included it in the distribution.
208    
209    31. There is now a check in pcretest against a ridiculously large number being
210        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
211        loop, the loop is abandoned.
212    
213    32. Forward references to subpatterns in conditions such as (?(2)...) where
214        subpattern 2 is defined later cause pcre_compile() to search forwards in
215        the pattern for the relevant set of parentheses. This search went wrong
216        when there were unescaped parentheses in a character class, parentheses
217        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
218    
219    33. "Subroutine" calls and backreferences were previously restricted to
220        referencing subpatterns earlier in the regex. This restriction has now
221        been removed.
222    
223    34. Added a number of extra features that are going to be in Perl 5.10. On the
224        whole, these are just syntactic alternatives for features that PCRE had
225        previously implemented using the Python syntax or my own invention. The
226        other formats are all retained for compatibility.
227    
228        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
229            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
230            also .NET compatible.
231    
232        (b) A recursion or subroutine call to a named group can now be defined as
233            (?&name) as well as (?P>name).
234    
235        (c) A backreference to a named group can now be defined as \k<name> or
236            \k'name' as well as (?P=name). The new forms, as well as being in Perl
237            5.10, are also .NET compatible.
238    
239        (d) A conditional reference to a named group can now use the syntax
240            (?(<name>) or (?('name') as well as (?(name).
241    
242        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
243            groups (named and numbered) that are never evaluated inline, but can be
244            called as "subroutines" from elsewhere. In effect, the DEFINE condition
245            is always false. There may be only one alternative in such a group.
246    
247        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
248            as the simple (?(R). The condition is true only if the most recent
249            recursion is that of the given number or name. It does not search out
250            through the entire recursion stack.
251    
252        (g) The escape \gN or \g{N} has been added, where N is a positive or
253            negative number, specifying an absolute or relative reference.
254    
255    35. Tidied to get rid of some further signed/unsigned compiler warnings and
256        some "unreachable code" warnings.
257    
258    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
259        things, this adds five new scripts.
260    
261    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
262        There were also incompatibilities regarding the handling of \Q..\E inside
263        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
264        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
265    
266    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
267        matches an empty string, and forcibly breaks the loop. There were bugs in
268        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
269        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
270        separate and independent bugs (that affected different cases) have been
271        fixed.
272    
273    39. Refactored the code to abolish the use of different opcodes for small
274        capturing bracket numbers. This is a tidy that I avoided doing when I
275        removed the limit on the number of capturing brackets for 3.5 back in 2001.
276        The new approach is not only tidier, it makes it possible to reduce the
277        memory needed to fix the previous bug (38).
278    
279    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
280        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
281        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
282        mode.
283    
284    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
285        report.
286    
287    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
288        copy construction and assignment in the C++ wrapper.
289    
290    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
291        couple of bugs that could have given wrong results in the "--newline=crlf"
292        case.
293    
294    44. Added a number of casts and did some reorganization of signed/unsigned int
295        variables following suggestions from Dair Grant. Also renamed the variable
296        "this" as "item" because it is a C++ keyword.
297    
298    45. Arranged for dftables to add
299    
300          #include "pcre_internal.h"
301    
302        to pcre_chartables.c because without it, gcc 4.x may remove the array
303        definition from the final binary if PCRE is built into a static library and
304        dead code stripping is activated.
305    
306    46. For an unanchored pattern, if a match attempt fails at the start of a
307        newline sequence, and the newline setting is CRLF or ANY, and the next two
308        characters are CRLF, advance by two characters instead of one.
309    
310    
311    Version 6.7 04-Jul-06
312    ---------------------
313    
314     1. In order to handle tests when input lines are enormously long, pcretest has
315        been re-factored so that it automatically extends its buffers when
316        necessary. The code is crude, but this _is_ just a test program. The
317        default size has been increased from 32K to 50K.
318    
319     2. The code in pcre_study() was using the value of the re argument before
320        testing it for NULL. (Of course, in any sensible call of the function, it
321        won't be NULL.)
322    
323     3. The memmove() emulation function in pcre_internal.h, which is used on
324        systems that lack both memmove() and bcopy() - that is, hardly ever -
325        was missing a "static" storage class specifier.
326    
327     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
328        containing an extended class (one that cannot be represented by a bitmap
329        because it contains high-valued characters or Unicode property items, e.g.
330        [\pZ]). Almost always one would set UTF-8 mode when processing such a
331        pattern, but PCRE should not loop if you do not (it no longer does).
332        [Detail: two cases were found: (a) a repeated subpattern containing an
333        extended class; (b) a recursive reference to a subpattern that followed a
334        previous extended class. It wasn't skipping over the extended class
335        correctly when UTF-8 mode was not set.]
336    
337     5. A negated single-character class was not being recognized as fixed-length
338        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
339        compile error "lookbehind assertion is not fixed length".
340    
341     6. The RunPerlTest auxiliary script was showing an unexpected difference
342        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
343        write a Perl script that can interpret lines of an input file either as
344        byte characters or as UTF-8, which is what "perltest" was being required to
345        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
346        can't do is switch easily at run time between having the "use utf8;" pragma
347        or not. In the end, I fudged it by using the RunPerlTest script to insert
348        "use utf8;" explicitly for the UTF-8 tests.
349    
350     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
351        the end of the subject string, contrary to the documentation and to what
352        Perl does. This was true of both matching functions. Now it matches only at
353        the start of the subject and immediately after *internal* newlines.
354    
355     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
356        a pointer to an int instead of a pointer to an unsigned long int. This
357        caused problems on 64-bit systems.
358    
359     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
360        instance of the 'standard' template library not being so standard".
361    
362    10. There was no check on the number of named subpatterns nor the maximum
363        length of a subpattern name. The product of these values is used to compute
364        the size of the memory block for a compiled pattern. By supplying a very
365        long subpattern name and a large number of named subpatterns, the size
366        computation could be caused to overflow. This is now prevented by limiting
367        the length of names to 32 characters, and the number of named subpatterns
368        to 10,000.
369    
370    11. Subpatterns that are repeated with specific counts have to be replicated in
371        the compiled pattern. The size of memory for this was computed from the
372        length of the subpattern and the repeat count. The latter is limited to
373        65535, but there was no limit on the former, meaning that integer overflow
374        could in principle occur. The compiled length of a repeated subpattern is
375        now limited to 30,000 bytes in order to prevent this.
376    
377    12. Added the optional facility to have named substrings with the same name.
378    
379    13. Added the ability to use a named substring as a condition, using the
380        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
381        are numbers (not recommended). Forward references are permitted.
382    
383    14. Added forward references in named backreferences (if you see what I mean).
384    
385    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
386        pattern could run off the end of the subject. For example, the pattern
387        "(?s)(.{1,5})"8 did this with the subject "ab".
388    
389    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
390        PCRE_CASELESS was set when matching characters that were quantified with ?
391        or *.
392    
393    17. A character class other than a single negated character that had a minimum
394        but no maximum quantifier - for example [ab]{6,} - was not handled
395        correctly by pce_dfa_exec(). It would match only one character.
396    
397    18. A valid (though odd) pattern that looked like a POSIX character
398        class but used an invalid character after [ (for example [[,abc,]]) caused
399        pcre_compile() to give the error "Failed: internal error: code overflow" or
400        in some cases to crash with a glibc free() error. This could even happen if
401        the pattern terminated after [[ but there just happened to be a sequence of
402        letters, a binary zero, and a closing ] in the memory that followed.
403    
404    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
405        over the years. Originally (before any Unicode support), just the bottom 8
406        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
407        output from "man perlunicode" includes this:
408    
409          The regular expression compiler produces polymorphic opcodes.  That
410          is, the pattern adapts to the data and automatically switches to
411          the Unicode character scheme when presented with Unicode data--or
412          instead uses a traditional byte scheme when presented with byte
413          data.
414    
415        Sadly, a wide octal escape does not cause a switch, and in a string with
416        no other multibyte characters, these octal escapes are treated as before.
417        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
418        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
419        Unicode string.
420    
421        I have not perpetrated such confusion in PCRE. Up till now, it took just
422        the bottom 8 bits, as in old Perl. I have now made octal escapes with
423        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
424        translate to the appropriate multibyte character.
425    
426    29. Applied some refactoring to reduce the number of warnings from Microsoft
427        and Borland compilers. This has included removing the fudge introduced
428        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
429        a warning about an unused variable.
430    
431    21. PCRE has not included VT (character 0x0b) in the set of whitespace
432        characters since release 4.0, because Perl (from release 5.004) does not.
433        [Or at least, is documented not to: some releases seem to be in conflict
434        with the documentation.] However, when a pattern was studied with
435        pcre_study() and all its branches started with \s, PCRE still included VT
436        as a possible starting character. Of course, this did no harm; it just
437        caused an unnecessary match attempt.
438    
439    22. Removed a now-redundant internal flag bit that recorded the fact that case
440        dependency changed within the pattern. This was once needed for "required
441        byte" processing, but is no longer used. This recovers a now-scarce options
442        bit. Also moved the least significant internal flag bit to the most-
443        significant bit of the word, which was not previously used (hangover from
444        the days when it was an int rather than a uint) to free up another bit for
445        the future.
446    
447    23. Added support for CRLF line endings as well as CR and LF. As well as the
448        default being selectable at build time, it can now be changed at runtime
449        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
450        specify that it is scanning data with non-default line endings.
451    
452    24. Changed the definition of CXXLINK to make it agree with the definition of
453        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
454    
455    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
456        recursions. This makes a big different to stack usage for some patterns.
457    
458    26. If a subpattern containing a named recursion or subroutine reference such
459        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
460        the space required for the compiled pattern went wrong and gave too small a
461        value. Depending on the environment, this could lead to "Failed: internal
462        error: code overflow at offset 49" or "glibc detected double free or
463        corruption" errors.
464    
465    27. Applied patches from Google (a) to support the new newline modes and (b) to
466        advance over multibyte UTF-8 characters in GlobalReplace.
467    
468    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
469        difference for some implementation of PCRE in some Windows version.
470    
471    29. Added some extra testing facilities to pcretest:
472    
473        \q<number>   in a data line sets the "match limit" value
474        \Q<number>   in a data line sets the "match recursion limt" value
475        -S <number>  sets the stack size, where <number> is in megabytes
476    
477        The -S option isn't available for Windows.
478    
479    
480    Version 6.6 06-Feb-06
481    ---------------------
482    
483     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
484        in pcreposix.h. I have copied the definition from pcre.h.
485    
486     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
487        because pcre.h is no longer a built file.
488    
489     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
490        not normally included in the compiled code.
491    
492    
493    Version 6.5 01-Feb-06
494    ---------------------
495    
496     1. When using the partial match feature with pcre_dfa_exec(), it was not
497        anchoring the second and subsequent partial matches at the new starting
498        point. This could lead to incorrect results. For example, with the pattern
499        /1234/, partially matching against "123" and then "a4" gave a match.
500    
501     2. Changes to pcregrep:
502    
503        (a) All non-match returns from pcre_exec() were being treated as failures
504            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
505            error message is output. Some extra information is given for the
506            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
507            probably the only errors that are likely to be caused by users (by
508            specifying a regex that has nested indefinite repeats, for instance).
509            If there are more than 20 of these errors, pcregrep is abandoned.
510    
511        (b) A binary zero was treated as data while matching, but terminated the
512            output line if it was written out. This has been fixed: binary zeroes
513            are now no different to any other data bytes.
514    
515        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
516            used to set a locale for matching. The --locale=xxxx long option has
517            been added (no short equivalent) to specify a locale explicitly on the
518            pcregrep command, overriding the environment variables.
519    
520        (d) When -B was used with -n, some line numbers in the output were one less
521            than they should have been.
522    
523        (e) Added the -o (--only-matching) option.
524    
525        (f) If -A or -C was used with -c (count only), some lines of context were
526            accidentally printed for the final match.
527    
528        (g) Added the -H (--with-filename) option.
529    
530        (h) The combination of options -rh failed to suppress file names for files
531            that were found from directory arguments.
532    
533        (i) Added the -D (--devices) and -d (--directories) options.
534    
535        (j) Added the -F (--fixed-strings) option.
536    
537        (k) Allow "-" to be used as a file name for -f as well as for a data file.
538    
539        (l) Added the --colo(u)r option.
540    
541        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
542            is not present by default.
543    
544     3. A nasty bug was discovered in the handling of recursive patterns, that is,
545        items such as (?R) or (?1), when the recursion could match a number of
546        alternatives. If it matched one of the alternatives, but subsequently,
547        outside the recursion, there was a failure, the code tried to back up into
548        the recursion. However, because of the way PCRE is implemented, this is not
549        possible, and the result was an incorrect result from the match.
550    
551        In order to prevent this happening, the specification of recursion has
552        been changed so that all such subpatterns are automatically treated as
553        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
554    
555     4. I had overlooked the fact that, in some locales, there are characters for
556        which isalpha() is true but neither isupper() nor islower() are true. In
557        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
558        and ordfeminine) are like this. This affected the treatment of \w and \W
559        when they appeared in character classes, but not when they appeared outside
560        a character class. The bit map for "word" characters is now created
561        separately from the results of isalnum() instead of just taking it from the
562        upper, lower, and digit maps. (Plus the underscore character, of course.)
563    
564     5. The above bug also affected the handling of POSIX character classes such as
565        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
566        permanent tables. Instead, the bit maps for such a class were previously
567        created as the appropriate unions of the upper, lower, and digit bitmaps.
568        Now they are created by subtraction from the [[:word:]] class, which has
569        its own bitmap.
570    
571     6. The [[:blank:]] character class matches horizontal, but not vertical space.
572        It is created by subtracting the vertical space characters (\x09, \x0a,
573        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
574        subtraction was done in the overall bitmap for a character class, meaning
575        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
576        be recognized. This bug has been fixed.
577    
578     7. Patches from the folks at Google:
579    
580          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
581          real life, but is still worth protecting against".
582    
583          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
584          regular expressions".
585    
586          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
587          have it.
588    
589          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
590          "configure" and the latter not, in order to fix a problem somebody had
591          with compiling the Arg class on HP-UX.
592    
593          (e) Improve the error-handling of the C++ wrapper a little bit.
594    
595          (f) New tests for checking recursion limiting.
596    
597     8. The pcre_memmove() function, which is used only if the environment does not
598        have a standard memmove() function (and is therefore rarely compiled),
599        contained two bugs: (a) use of int instead of size_t, and (b) it was not
600        returning a result (though PCRE never actually uses the result).
601    
602     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
603        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
604        returned instead of calling malloc() with an overflowing number that would
605        most likely cause subsequent chaos.
606    
607    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
608    
609    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
610        with this option is matched, the nmatch and pmatch options of regexec() are
611        ignored.
612    
613    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
614        provided in case anyone wants to the the POSIX interface with UTF-8
615        strings.
616    
617    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
618        C++ linking (needed for some HP-UX environments).
619    
620    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
621        (unused parameter) and in the pcre_printint() function (omitted "default"
622        switch label when the default is to do nothing).
623    
624    15. Added some code to make it possible, when PCRE is compiled as a C++
625        library, to replace subject pointers for pcre_exec() with a smart pointer
626        class, thus making it possible to process discontinuous strings.
627    
628    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
629        much the same function. They were added by different people who were trying
630        to make PCRE easy to compile on non-Unix systems. It has been suggested
631        that PCRE_EXPORT be abolished now that there is more automatic apparatus
632        for compiling on Windows systems. I have therefore replaced it with
633        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
634        defaults to "extern" for C or "extern C" for C++, which works fine on
635        Unix-like systems. It is now possible to override the value of PCRE_DATA_
636        SCOPE with something explicit in config.h. In addition:
637    
638        (a) pcreposix.h still had just "extern" instead of either of these macros;
639            I have replaced it with PCRE_DATA_SCOPE.
640    
641        (b) Functions such as _pcre_xclass(), which are internal to the library,
642            but external in the C sense, all had PCRE_EXPORT in their definitions.
643            This is apparently wrong for the Windows case, so I have removed it.
644            (It makes no difference on Unix-like systems.)
645    
646    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
647        of recursive calls to match(). This is different to MATCH_LIMIT because
648        that limits the total number of calls to match(), not all of which increase
649        the depth of recursion. Limiting the recursion depth limits the amount of
650        stack (or heap if NO_RECURSE is set) that is used. The default can be set
651        when PCRE is compiled, and changed at run time. A patch from Google adds
652        this functionality to the C++ interface.
653    
654    18. Changes to the handling of Unicode character properties:
655    
656        (a) Updated the table to Unicode 4.1.0.
657    
658        (b) Recognize characters that are not in the table as "Cn" (undefined).
659    
660        (c) I revised the way the table is implemented to a much improved format
661            which includes recognition of ranges. It now supports the ranges that
662            are defined in UnicodeData.txt, and it also amalgamates other
663            characters into ranges. This has reduced the number of entries in the
664            table from around 16,000 to around 3,000, thus reducing its size
665            considerably. I realized I did not need to use a tree structure after
666            all - a binary chop search is just as efficient. Having reduced the
667            number of entries, I extended their size from 6 bytes to 8 bytes to
668            allow for more data.
669    
670        (d) Added support for Unicode script names via properties such as \p{Han}.
671    
672    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
673        matching that character.
674    
675    20. When matching a repeated Unicode property with a minimum greater than zero,
676        (for example \pL{2,}), PCRE could look past the end of the subject if it
677        reached it while seeking the minimum number of characters. This could
678        happen only if some of the characters were more than one byte long, because
679        there is a check for at least the minimum number of bytes.
680    
681    21. Refactored the implementation of \p and \P so as to be more general, to
682        allow for more different types of property in future. This has changed the
683        compiled form incompatibly. Anybody with saved compiled patterns that use
684        \p or \P will have to recompile them.
685    
686    22. Added "Any" and "L&" to the supported property types.
687    
688    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
689        but give a compile time error if the value is greater than 0xff.
690    
691    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
692        accidentally not being installed or uninstalled.
693    
694    25. The pcre.h file was built from pcre.h.in, but the only changes that were
695        made were to insert the current release number. This seemed silly, because
696        it made things harder for people building PCRE on systems that don't run
697        "configure". I have turned pcre.h into a distributed file, no longer built
698        by "configure", with the version identification directly included. There is
699        no longer a pcre.h.in file.
700    
701        However, this change necessitated a change to the pcre-config script as
702        well. It is built from pcre-config.in, and one of the substitutions was the
703        release number. I have updated configure.ac so that ./configure now finds
704        the release number by grepping pcre.h.
705    
706    26. Added the ability to run the tests under valgrind.
707    
708    
709    Version 6.4 05-Sep-05
710    ---------------------
711    
712     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
713        "--" to be printed when multiple files were scanned, even when none of the
714        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
715        consider it to be a bug, and have restored the previous behaviour.
716    
717     2. A couple of code tidies to get rid of compiler warnings.
718    
719     3. The pcretest program used to cheat by referring to symbols in the library
720        whose names begin with _pcre_. These are internal symbols that are not
721        really supposed to be visible externally, and in some environments it is
722        possible to suppress them. The cheating is now confined to including
723        certain files from the library's source, which is a bit cleaner.
724    
725     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
726        file's purpose clearer.
727    
728     5. Reorganized pcre_ucp_findchar().
729    
730    
731    Version 6.3 15-Aug-05
732    ---------------------
733    
734     1. The file libpcre.pc.in did not have general read permission in the tarball.
735    
736     2. There were some problems when building without C++ support:
737    
738        (a) If C++ support was not built, "make install" and "make test" still
739            tried to test it.
740    
741        (b) There were problems when the value of CXX was explicitly set. Some
742            changes have been made to try to fix these, and ...
743    
744        (c) --disable-cpp can now be used to explicitly disable C++ support.
745    
746        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
747            backslash in a target when C++ was disabled. This confuses some
748            versions of "make", apparently. Using an intermediate variable solves
749            this. (Same for CPP_LOBJ.)
750    
751     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
752        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
753        necessary on certain architectures.
754    
755     4. Added a setting of -export-symbols-regex to the link command to remove
756        those symbols that are exported in the C sense, but actually are local
757        within the library, and not documented. Their names all begin with
758        "_pcre_". This is not a perfect job, because (a) we have to except some
759        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
760        available (and never for static libraries). I have made a note to try to
761        find a way round (a) in the future.
762    
763    
764    Version 6.2 01-Aug-05
765    ---------------------
766    
767     1. There was no test for integer overflow of quantifier values. A construction
768        such as {1111111111111111} would give undefined results. What is worse, if
769        a minimum quantifier for a parenthesized subpattern overflowed and became
770        negative, the calculation of the memory size went wrong. This could have
771        led to memory overwriting.
772    
773     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
774    
775     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
776        operating environments where this matters.
777    
778     4. Applied Giuseppe Maxia's patch to add additional features for controlling
779        PCRE options from within the C++ wrapper.
780    
781     5. Named capturing subpatterns were not being correctly counted when a pattern
782        was compiled. This caused two problems: (a) If there were more than 100
783        such subpatterns, the calculation of the memory needed for the whole
784        compiled pattern went wrong, leading to an overflow error. (b) Numerical
785        back references of the form \12, where the number was greater than 9, were
786        not recognized as back references, even though there were sufficient
787        previous subpatterns.
788    
789     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
790        versions of gcc, e.g. 2.95.4.
791    
792    
793    Version 6.1 21-Jun-05
794    ---------------------
795    
796     1. There was one reference to the variable "posix" in pcretest.c that was not
797        surrounded by "#if !defined NOPOSIX".
798    
799     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
800        the cross-check on the old pcre_info() function, for the benefit of the
801        cut-down version of PCRE that is currently imported into Exim.
802    
803     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
804        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
805        patterns that start (?i) but I don't think that matters. The use of (?i) is
806        just an example; this all applies to the other options as well.
807    
808     4. Since libtool seems to echo the compile commands it is issuing, the output
809        from "make" can be reduced a bit by putting "@" in front of each libtool
810        compile command.
811    
812     5. Patch from the folks at Google for configure.in to be a bit more thorough
813        in checking for a suitable C++ installation before trying to compile the
814        C++ stuff. This should fix a reported problem when a compiler was present,
815        but no suitable headers.
816    
817     6. The man pages all had just "PCRE" as their title. I have changed them to
818        be the relevant file name. I have also arranged that these names are
819        retained in the file doc/pcre.txt, which is a concatenation in text format
820        of all the man pages except the little individual ones for each function.
821    
822     7. The NON-UNIX-USE file had not been updated for the different set of source
823        files that come with release 6. I also added a few comments about the C++
824        wrapper.
825    
826    
827    Version 6.0 07-Jun-05
828    ---------------------
829    
830     1. Some minor internal re-organization to help with my DFA experiments.
831    
832     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
833        didn't matter for the library itself when fully configured, but did matter
834        when compiling without UCP support, or within Exim, where the ucp files are
835        not imported.
836    
837     3. Refactoring of the library code to split up the various functions into
838        different source modules. The addition of the new DFA matching code (see
839        below) to a single monolithic source would have made it really too
840        unwieldy, quite apart from causing all the code to be include in a
841        statically linked application, when only some functions are used. This is
842        relevant even without the DFA addition now that patterns can be compiled in
843        one application and matched in another.
844    
845        The downside of splitting up is that there have to be some external
846        functions and data tables that are used internally in different modules of
847        the library but which are not part of the API. These have all had their
848        names changed to start with "_pcre_" so that they are unlikely to clash
849        with other external names.
850    
851     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
852        a different (DFA) algorithm. Although it is slower than the original
853        function, it does have some advantages for certain types of matching
854        problem.
855    
856     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
857        including restarting after a partial match.
858    
859     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
860        defined when compiling for Windows was sent to me. I have put it into the
861        code, though I have no means of testing or verifying it.
862    
863     7. Added the pcre_refcount() auxiliary function.
864    
865     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
866        match before or at the first newline in the subject string. In pcretest,
867        the /f option on a pattern can be used to set this.
868    
869     9. A repeated \w when used in UTF-8 mode with characters greater than 256
870        would behave wrongly. This has been present in PCRE since release 4.0.
871    
872    10. A number of changes to the pcregrep command:
873    
874        (a) Refactored how -x works; insert ^(...)$ instead of setting
875            PCRE_ANCHORED and checking the length, in preparation for adding
876            something similar for -w.
877    
878        (b) Added the -w (match as a word) option.
879    
880        (c) Refactored the way lines are read and buffered so as to have more
881            than one at a time available.
882    
883        (d) Implemented a pcregrep test script.
884    
885        (e) Added the -M (multiline match) option. This allows patterns to match
886            over several lines of the subject. The buffering ensures that at least
887            8K, or the rest of the document (whichever is the shorter) is available
888            for matching (and similarly the previous 8K for lookbehind assertions).
889    
890        (f) Changed the --help output so that it now says
891    
892              -w, --word-regex(p)
893    
894            instead of two lines, one with "regex" and the other with "regexp"
895            because that confused at least one person since the short forms are the
896            same. (This required a bit of code, as the output is generated
897            automatically from a table. It wasn't just a text change.)
898    
899        (g) -- can be used to terminate pcregrep options if the next thing isn't an
900            option but starts with a hyphen. Could be a pattern or a path name
901            starting with a hyphen, for instance.
902    
903        (h) "-" can be given as a file name to represent stdin.
904    
905        (i) When file names are being printed, "(standard input)" is used for
906            the standard input, for compatibility with GNU grep. Previously
907            "<stdin>" was used.
908    
909        (j) The option --label=xxx can be used to supply a name to be used for
910            stdin when file names are being printed. There is no short form.
911    
912        (k) Re-factored the options decoding logic because we are going to add
913            two more options that take data. Such options can now be given in four
914            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
915    
916        (l) Added the -A, -B, and -C options for requesting that lines of context
917            around matches be printed.
918    
919        (m) Added the -L option to print the names of files that do not contain
920            any matching lines, that is, the complement of -l.
921    
922        (n) The return code is 2 if any file cannot be opened, but pcregrep does
923            continue to scan other files.
924    
925        (o) The -s option was incorrectly implemented. For compatibility with other
926            greps, it now suppresses the error message for a non-existent or non-
927            accessible file (but not the return code). There is a new option called
928            -q that suppresses the output of matching lines, which was what -s was
929            previously doing.
930    
931        (p) Added --include and --exclude options to specify files for inclusion
932            and exclusion when recursing.
933    
934    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
935        Hopefully, it now does.
936    
937    12. Missing cast in pcre_study().
938    
939    13. Added an "uninstall" target to the makefile.
940    
941    14. Replaced "extern" in the function prototypes in Makefile.in with
942        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
943        world, but is set differently for Windows.
944    
945    15. Added a second compiling function called pcre_compile2(). The only
946        difference is that it has an extra argument, which is a pointer to an
947        integer error code. When there is a compile-time failure, this is set
948        non-zero, in addition to the error test pointer being set to point to an
949        error message. The new argument may be NULL if no error number is required
950        (but then you may as well call pcre_compile(), which is now just a
951        wrapper). This facility is provided because some applications need a
952        numeric error indication, but it has also enabled me to tidy up the way
953        compile-time errors are handled in the POSIX wrapper.
954    
955    16. Added VPATH=.libs to the makefile; this should help when building with one
956        prefix path and installing with another. (Or so I'm told by someone who
957        knows more about this stuff than I do.)
958    
959    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
960        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
961        match everything, including newlines. This is not POSIX-compatible, but
962        somebody wanted the feature. From pcretest it can be activated by using
963        both the P and the s flags.
964    
965    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
966    
967    19. libpcre.pc was being incorrectly installed as executable.
968    
969    20. A couple of places in pcretest check for end-of-line by looking for '\n';
970        it now also looks for '\r' so that it will work unmodified on Windows.
971    
972    21. Added Google's contributed C++ wrapper to the distribution.
973    
974    22. Added some untidy missing memory free() calls in pcretest, to keep
975        Electric Fence happy when testing.
976    
977    
978    
979    Version 5.0 13-Sep-04
980    ---------------------
981    
982     1. Internal change: literal characters are no longer packed up into items
983        containing multiple characters in a single byte-string. Each character
984        is now matched using a separate opcode. However, there may be more than one
985        byte in the character in UTF-8 mode.
986    
987     2. The pcre_callout_block structure has two new fields: pattern_position and
988        next_item_length. These contain the offset in the pattern to the next match
989        item, and its length, respectively.
990    
991     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
992        insertion of callouts before each pattern item. Added the /C option to
993        pcretest to make use of this.
994    
995     4. On the advice of a Windows user, the lines
996    
997          #if defined(_WIN32) || defined(WIN32)
998          _setmode( _fileno( stdout ), 0x8000 );
999          #endif  /* defined(_WIN32) || defined(WIN32) */
1000    
1001        have been added to the source of pcretest. This apparently does useful
1002        magic in relation to line terminators.
1003    
1004     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1005        for the benefit of those environments where the "b" makes a difference.
1006    
1007     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1008        to know about it. I have put a hack into configure.in that adds in code
1009        to set GCC=yes if CC=icc. This seems to end up at a point in the
1010        generated configure script that is early enough to affect the setting of
1011        compiler options, which is what is needed, but I have no means of testing
1012        whether it really works. (The user who reported this had patched the
1013        generated configure script, which of course I cannot do.)
1014    
1015        LATER: After change 22 below (new libtool files), the configure script
1016        seems to know about icc (and also ecc). Therefore, I have commented out
1017        this hack in configure.in.
1018    
1019     7. Added support for pkg-config (2 patches were sent in).
1020    
1021     8. Negated POSIX character classes that used a combination of internal tables
1022        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1023        [[:^ascii]]. Typically, they would match almost any characters. The other
1024        POSIX classes were not broken in this way.
1025    
1026     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1027        to find the match, as PCRE was deluded into thinking that the match had to
1028        start at the start point or following a newline. The same bug applied to
1029        patterns with negative forward assertions or any backward assertions
1030        preceding ".*" at the start, unless the pattern required a fixed first
1031        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1032    
1033    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1034        starting at the last subject character, bytes beyond the end of the subject
1035        string were read.
1036    
1037    11. Renamed the variable "class" as "classbits" to make life easier for C++
1038        users. (Previously there was a macro definition, but it apparently wasn't
1039        enough.)
1040    
1041    12. Added the new field "tables" to the extra data so that tables can be passed
1042        in at exec time, or the internal tables can be re-selected. This allows
1043        a compiled regex to be saved and re-used at a later time by a different
1044        program that might have everything at different addresses.
1045    
1046    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1047        -R library as well as a -L library.
1048    
1049    14. The debugging options of pcretest (-d on the command line or D on a
1050        pattern) showed incorrect output for anything following an extended class
1051        that contained multibyte characters and which was followed by a quantifier.
1052    
1053    15. Added optional support for general category Unicode character properties
1054        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1055        support. It adds about 90K to the size of the library. The meanings of the
1056        inbuilt class escapes such as \d and \s have NOT been changed.
1057    
1058    16. Updated pcredemo.c to include calls to free() to release the memory for the
1059        compiled pattern.
1060    
1061    17. The generated file chartables.c was being created in the source directory
1062        instead of in the building directory. This caused the build to fail if the
1063        source directory was different from the building directory, and was
1064        read-only.
1065    
1066    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1067        file. No doubt somebody will tell me if they don't make sense... Also added
1068        Dan Mooney's comments about building on OpenVMS.
1069    
1070    19. Added support for partial matching via the PCRE_PARTIAL option for
1071        pcre_exec() and the \P data escape in pcretest.
1072    
1073    20. Extended pcretest with 3 new pattern features:
1074    
1075        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1076              write the compiled pattern to the file whose name is "rest-of-line".
1077              This is a straight binary dump of the data, with the saved pointer to
1078              the character tables forced to be NULL. The study data, if any, is
1079              written too. After writing, pcretest reads a new pattern.
1080    
1081        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1082              compiled pattern from the given file. There must not be any
1083              occurrences of "<" in the file name (pretty unlikely); if there are,
1084              pcretest will instead treat the initial "<" as a pattern delimiter.
1085              After reading in the pattern, pcretest goes on to read data lines as
1086              usual.
1087    
1088        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1089              and 16-bit fields in a compiled pattern, to simulate a pattern that
1090              was compiled on a host of opposite endianness.
1091    
1092    21. The pcre-exec() function can now cope with patterns that were compiled on
1093        hosts of opposite endianness, with this restriction:
1094    
1095          As for any compiled expression that is saved and used later, the tables
1096          pointer field cannot be preserved; the extra_data field in the arguments
1097          to pcre_exec() should be used to pass in a tables address if a value
1098          other than the default internal tables were used at compile time.
1099    
1100    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1101        now diagnosed as an error. Previously, most of the time, a negative number
1102        would have been treated as zero, but if in addition "ovector" was passed as
1103        NULL, a crash could occur.
1104    
1105    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1106        new versions from the libtool 1.5 distribution (the last one is a copy of
1107        a file called libtool.m4). This seems to have fixed the need to patch
1108        "configure" to support Darwin 1.3 (which I used to do). However, I still
1109        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1110        workstation).
1111    
1112    24. Changed the PCRE licence to be the more standard "BSD" licence.
1113    
1114    
1115    Version 4.5 01-Dec-03
1116    ---------------------
1117    
1118     1. There has been some re-arrangement of the code for the match() function so
1119        that it can be compiled in a version that does not call itself recursively.
1120        Instead, it keeps those local variables that need separate instances for
1121        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1122        needs to "recurse". Keeping track of where control must go is done by means
1123        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1124        hide most of the details from the main code, and operates only if
1125        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1126        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1127        operating.
1128    
1129        To make it easier for callers to provide specially tailored get/free
1130        functions for this usage, two new functions, pcre_stack_malloc, and
1131        pcre_stack_free, are used. They are always called in strict stacking order,
1132        and the size of block requested is always the same.
1133    
1134        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1135        PCRE has been compiled to use the stack or the heap for recursion. The
1136        -C option of pcretest uses this to show which version is compiled.
1137    
1138        A new data escape \S, is added to pcretest; it causes the amounts of store
1139        obtained and freed by both kinds of malloc/free at match time to be added
1140        to the output.
1141    
1142     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1143        what's available on my current Linux desktop machine.
1144    
1145     3. When matching a UTF-8 string, the test for a valid string at the start has
1146        been extended. If start_offset is not zero, PCRE now checks that it points
1147        to a byte that is the start of a UTF-8 character. If not, it returns
1148        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1149        this is necessary because there may be backward assertions in the pattern.
1150        When matching the same subject several times, it may save resources to use
1151        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1152    
1153     4. The code for checking the validity of UTF-8 strings has been tightened so
1154        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1155        containing "overlong sequences".
1156    
1157     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1158        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1159        should have been "&", but it just so happened that all the cases this let
1160        through by mistake were picked up later in the function.
1161    
1162     6. I had used a variable called "isblank" - this is a C99 function, causing
1163        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1164    
1165     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1166        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1167        so the version is shown; (c) stop "make test" echoing "./RunTest".
1168    
1169     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1170    
1171     9. The prototype for memmove() for systems that don't have it was using
1172        size_t, but the inclusion of the header that defines size_t was later. I've
1173        moved the #includes for the C headers earlier to avoid this.
1174    
1175    10. Added some adjustments to the code to make it easier to compiler on certain
1176        special systems:
1177    
1178          (a) Some "const" qualifiers were missing.
1179          (b) Added the macro EXPORT before all exported functions; by default this
1180              is defined to be empty.
1181          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1182              that it reads its output file name as an argument instead of writing
1183              to the standard output and assuming this can be redirected.
1184    
1185    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1186        class containing characters with values greater than 255, PCRE compilation
1187        went into a loop.
1188    
1189    12. A recursive reference to a subpattern that was within another subpattern
1190        that had a minimum quantifier of zero caused PCRE to crash. For example,
1191        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1192        recursion. If the recursively-called subpattern itself had a zero repeat,
1193        that was OK.
1194    
1195    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1196        buffer into which it was copied (for escape processing) was still set at
1197        1024, so long lines caused crashes.
1198    
1199    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1200        "internal error: code overflow...". This applied to any character class
1201        that was followed by a possessive quantifier.
1202    
1203    15. Modified the Makefile to add libpcre.la as a prerequisite for
1204        libpcreposix.la because I was told this is needed for a parallel build to
1205        work.
1206    
1207    16. If a pattern that contained .* following optional items at the start was
1208        studied, the wrong optimizing data was generated, leading to matching
1209        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1210        matching string must start with a or b or c. The correct conclusion for
1211        this pattern is that a match can start with any character.
1212    
1213    
1214    Version 4.4 13-Aug-03
1215    ---------------------
1216    
1217     1. In UTF-8 mode, a character class containing characters with values between
1218        127 and 255 was not handled correctly if the compiled pattern was studied.
1219        In fixing this, I have also improved the studying algorithm for such
1220        classes (slightly).
1221    
1222     2. Three internal functions had redundant arguments passed to them. Removal
1223        might give a very teeny performance improvement.
1224    
1225     3. Documentation bug: the value of the capture_top field in a callout is *one
1226        more than* the number of the hightest numbered captured substring.
1227    
1228     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1229        in incorrectly linking with a previously installed version. They now link
1230        explicitly with libpcre.la.
1231    
1232     5. configure.in no longer needs to recognize Cygwin specially.
1233    
1234     6. A problem in pcre.in for Windows platforms is fixed.
1235    
1236     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1237        pcretest, it used to include the size of the study block as part of its
1238        output. Unfortunately, the structure contains a field that has a different
1239        size on different hardware architectures. This meant that the tests that
1240        showed this size failed. As the block is currently always of a fixed size,
1241        this information isn't actually particularly useful in pcretest output, so
1242        I have just removed it.
1243    
1244     8. Three pre-processor statements accidentally did not start in column 1.
1245        Sadly, there are *still* compilers around that complain, even though
1246        standard C has not required this for well over a decade. Sigh.
1247    
1248     9. In pcretest, the code for checking callouts passed small integers in the
1249        callout_data field, which is a void * field. However, some picky compilers
1250        complained about the casts involved for this on 64-bit systems. Now
1251        pcretest passes the address of the small integer instead, which should get
1252        rid of the warnings.
1253    
1254    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1255        both compile and run time, and gives an error if an invalid UTF-8 sequence
1256        is found. There is a option for disabling this check in cases where the
1257        string is known to be correct and/or the maximum performance is wanted.
1258    
1259    11. In response to a bug report, I changed one line in Makefile.in from
1260    
1261            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1262        to
1263            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1264    
1265        to look similar to other lines, but I have no way of telling whether this
1266        is the right thing to do, as I do not use Windows. No doubt I'll get told
1267        if it's wrong...
1268    
1269    
1270    Version 4.3 21-May-03
1271    ---------------------
1272    
1273    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1274       Makefile.
1275    
1276    2. Some refactoring to improve the quality of the code:
1277    
1278       (i)   The utf8_table... variables are now declared "const".
1279    
1280       (ii)  The code for \cx, which used the "case flipping" table to upper case
1281             lower case letters, now just substracts 32. This is ASCII-specific,
1282             but the whole concept of \cx is ASCII-specific, so it seems
1283             reasonable.
1284    
1285       (iii) PCRE was using its character types table to recognize decimal and
1286             hexadecimal digits in the pattern. This is silly, because it handles
1287             only 0-9, a-f, and A-F, but the character types table is locale-
1288             specific, which means strange things might happen. A private
1289             table is now used for this - though it costs 256 bytes, a table is
1290             much faster than multiple explicit tests. Of course, the standard
1291             character types table is still used for matching digits in subject
1292             strings against \d.
1293    
1294       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1295             ending in _t are). So I've renamed it as ESC_tee.
1296    
1297    3. The first argument for regexec() in the POSIX wrapper should have been
1298       defined as "const".
1299    
1300    4. Changed pcretest to use malloc() for its buffers so that they can be
1301       Electric Fenced for debugging.
1302    
1303    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1304       to read one or more bytes before the start of the subject string. Often this
1305       had no effect on PCRE's behaviour, but in some circumstances it could
1306       provoke a segmentation fault.
1307    
1308    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1309       to try to read one or more bytes before the start of the subject string.
1310    
1311    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1312       UTF-8 support could misbehave in various ways if the subject string
1313       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1314       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1315       back over UTF-8 characters.)
1316    
1317    
1318    Version 4.2 14-Apr-03
1319    ---------------------
1320    
1321    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1322    
1323    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1324         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1325         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1326         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1327         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1328           and BUILD_EXEEXT
1329         Note: automatic setting of the BUILD variables is not yet working
1330         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1331           compile-time but not at link-time
1332         [LINK]: use for linking executables only
1333         make different versions for Windows and non-Windows
1334         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1335           libraries
1336         [LINK_FOR_BUILD]: new variable
1337         [OBJEXT]: use throughout
1338         [EXEEXT]: use throughout
1339         <winshared>: new target
1340         <wininstall>: new target
1341         <dftables.o>: use native compiler
1342         <dftables>: use native linker
1343         <install>: handle Windows platform correctly
1344         <clean>: ditto
1345         <check>: ditto
1346         copy DLL to top builddir before testing
1347    
1348       As part of these changes, -no-undefined was removed again. This was reported
1349       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1350       in any case.
1351    
1352    3. Some tidies to get rid of compiler warnings:
1353    
1354       . In the match_data structure, match_limit was an unsigned long int, whereas
1355         match_call_count was an int. I've made them both unsigned long ints.
1356    
1357       . In pcretest the fact that a const uschar * doesn't automatically cast to
1358         a void * provoked a warning.
1359    
1360       . Turning on some more compiler warnings threw up some "shadow" variables
1361         and a few more missing casts.
1362    
1363    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1364       option, a class that contained a single character with a value between 128
1365       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1366    
1367    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1368       option, a class that contained several characters, but with at least one
1369       whose value was between 128 and 255 caused PCRE to crash.
1370    
1371    
1372    Version 4.1 12-Mar-03
1373    ---------------------
1374    
1375    1. Compiling with gcc -pedantic found a couple of places where casts were
1376    needed, and a string in dftables.c that was longer than standard compilers are
1377    required to support.
1378    
1379    2. Compiling with Sun's compiler found a few more places where the code could
1380    be tidied up in order to avoid warnings.
1381    
1382    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1383    first of these names is deprecated in the latest Autoconf in favour of the name
1384    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1385    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1386    analogy I have changed it to CFLAGS_FOR_BUILD.
1387    
1388    4. Added -no-undefined to the linking command in the Makefile, because this is
1389    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1390    linking step for the pcreposix library.
1391    
1392    5. PCRE was failing to diagnose the case of two named groups with the same
1393    name.
1394    
1395    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1396    literal character that is needed in the subject for a match, and scans along to
1397    ensure that it is present before embarking on the full matching process. This
1398    saves time in cases of nested unlimited repeats that are never going to match.
1399    Problem: the scan can take a lot of time if the subject is very long (e.g.
1400    megabytes), thus penalizing straightforward matches. It is now done only if the
1401    amount of subject to be scanned is less than 1000 bytes.
1402    
1403    7. A lesser problem with the same optimization is that it was recording the
1404    first character of an anchored pattern as "needed", thus provoking a search
1405    right along the subject, even when the first match of the pattern was going to
1406    fail. The "needed" character is now not set for anchored patterns, unless it
1407    follows something in the pattern that is of non-fixed length. Thus, it still
1408    fulfils its original purpose of finding quick non-matches in cases of nested
1409    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1410    
1411    
1412    Version 4.0 17-Feb-03
1413    ---------------------
1414    
1415  1. If a comment in an extended regex that started immediately after a meta-item  1. If a comment in an extended regex that started immediately after a meta-item
1416  extended to the end of string, PCRE compiled incorrect data. This could lead to  extended to the end of string, PCRE compiled incorrect data. This could lead to

Legend:
Removed from v.63  
changed lines
  Added in v.101

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12