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

Diff of /code/trunk/ChangeLog

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

revision 41 by nigel, Sat Feb 24 21:39:17 2007 UTC revision 102 by ph10, Tue Mar 6 15:50:45 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.1 05-Mar-07
5    ---------------------
6    
7     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
8        that is more "standard", making use of automake and other autotools. There
9        is some re-arrangement of the files and adjustment of comments consequent
10        on this.
11    
12     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
13        for recursive directory scanning broke on some systems because the files
14        are not scanned in any specific order and on different systems the order
15        was different. A call to "sort" has been inserted into RunGrepTest for the
16        approprate test as a short-term fix. In the longer term there may be an
17        alternative.
18    
19     3. I had an email from Eric Raymond about problems translating some of PCRE's
20        man pages to HTML (despite the fact that I distribute HTML pages, some
21        people do their own conversions for various reasons). The problems
22        concerned the use of low-level troff macros .br and .in. I have therefore
23        removed all such uses from the man pages (some were redundant, some could
24        be replaced by .nf/.fi pairs). The maintain/132html script that I use to
25        generate HTML has been updated to handle .nf/.fi and to complain if it
26        encounters .br or .in.
27    
28     4. Updated comments in configure.ac that get placed in config.h.in and also
29        arranged for config.h to be included in the distribution, for the benefit
30        of those who have to compile without Autotools (compare pcre.h).
31    
32     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
33        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
34        makevp.bat for the current PCRE, using the additional files !compile.txt,
35        !linklib.txt, and pcregexp.pas.
36    
37     6. A Windows user reported a minor discrepancy with test 2, which turned out
38        to be caused by a trailing space on an input line that had got lost in his
39        copy. The trailing space was an accident, so I've just removed it.
40    
41    
42    Version 7.0 19-Dec-06
43    ---------------------
44    
45     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
46        moving to gcc 4.1.1.
47    
48     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
49        sys/time.h, which is documented as needed for this function. It doesn't
50        seem to matter on Linux, but it showed up on some releases of OS X.
51    
52     3. It seems that there are systems where bytes whose values are greater than
53        127 match isprint() in the "C" locale. The "C" locale should be the
54        default when a C program starts up. In most systems, only ASCII printing
55        characters match isprint(). This difference caused the output from pcretest
56        to vary, making some of the tests fail. I have changed pcretest so that:
57    
58        (a) When it is outputting text in the compiled version of a pattern, bytes
59            other than 32-126 are always shown as hex escapes.
60    
61        (b) When it is outputting text that is a matched part of a subject string,
62            it does the same, unless a different locale has been set for the match
63            (using the /L modifier). In this case, it uses isprint() to decide.
64    
65     4. Fixed a major bug that caused incorrect computation of the amount of memory
66        required for a compiled pattern when options that changed within the
67        pattern affected the logic of the preliminary scan that determines the
68        length. The relevant options are -x, and -i in UTF-8 mode. The result was
69        that the computed length was too small. The symptoms of this bug were
70        either the PCRE error "internal error: code overflow" from pcre_compile(),
71        or a glibc crash with a message such as "pcretest: free(): invalid next
72        size (fast)". Examples of patterns that provoked this bug (shown in
73        pcretest format) are:
74    
75          /(?-x: )/x
76          /(?x)(?-x: \s*#\s*)/
77          /((?i)[\x{c0}])/8
78          /(?i:[\x{c0}])/8
79    
80        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
81        is now done differently.
82    
83     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
84        wrapper classes; (b) implement a new function in the C++ scanner that is
85        more efficient than the old way of doing things because it avoids levels of
86        recursion in the regex matching; (c) add a paragraph to the documentation
87        for the FullMatch() function.
88    
89     6. The escape sequence \n was being treated as whatever was defined as
90        "newline". Not only was this contrary to the documentation, which states
91        that \n is character 10 (hex 0A), but it also went horribly wrong when
92        "newline" was defined as CRLF. This has been fixed.
93    
94     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
95        was being set to -1 for the "end of line" case (supposedly a value that no
96        character can have). Though this value is never used (the check for end of
97        line is "zero bytes in current character"), it caused compiler complaints.
98        I've changed it to 0xffffffff.
99    
100     8. In pcre_version.c, the version string was being built by a sequence of
101        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
102        string (as it is for production releases) called a macro with an empty
103        argument. The C standard says the result of this is undefined. The gcc
104        compiler treats it as an empty string (which was what was wanted) but it is
105        reported that Visual C gives an error. The source has been hacked around to
106        avoid this problem.
107    
108     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
109        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
110        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
111        of them did).
112    
113    10. Originally, pcretest opened its input and output without "b"; then I was
114        told that "b" was needed in some environments, so it was added for release
115        5.0 to both the input and output. (It makes no difference on Unix-like
116        systems.) Later I was told that it is wrong for the input on Windows. I've
117        now abstracted the modes into two macros, to make it easier to fiddle with
118        them, and removed "b" from the input mode under Windows.
119    
120    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
121    
122    12. Added -help and --help to pcretest as an official way of being reminded
123        of the options.
124    
125    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
126        and pcrecpp.cc because they annoy compilers at high warning levels.
127    
128    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
129    
130    15. Fixed an occurrence of == in configure.ac that should have been = (shell
131        scripts are not C programs :-) and which was not noticed because it works
132        on Linux.
133    
134    16. pcretest is supposed to handle any length of pattern and data line (as one
135        line or as a continued sequence of lines) by extending its input buffer if
136        necessary. This feature was broken for very long pattern lines, leading to
137        a string of junk being passed to pcre_compile() if the pattern was longer
138        than about 50K.
139    
140    17. I have done a major re-factoring of the way pcre_compile() computes the
141        amount of memory needed for a compiled pattern. Previously, there was code
142        that made a preliminary scan of the pattern in order to do this. That was
143        OK when PCRE was new, but as the facilities have expanded, it has become
144        harder and harder to keep it in step with the real compile phase, and there
145        have been a number of bugs (see for example, 4 above). I have now found a
146        cunning way of running the real compile function in a "fake" mode that
147        enables it to compute how much memory it would need, while actually only
148        ever using a few hundred bytes of working memory and without too many
149        tests of the mode. This should make future maintenance and development
150        easier. A side effect of this work is that the limit of 200 on the nesting
151        depth of parentheses has been removed (though this was never a serious
152        limitation, I suspect). However, there is a downside: pcre_compile() now
153        runs more slowly than before (30% or more, depending on the pattern). I
154        hope this isn't a big issue. There is no effect on runtime performance.
155    
156    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
157        newline (only possible for the last line of a file) and it was a
158        pattern that set a locale (followed by /Lsomething), pcretest crashed.
159    
160    19. Added additional timing features to pcretest. (1) The -tm option now times
161        matching only, not compiling. (2) Both -t and -tm can be followed, as a
162        separate command line item, by a number that specifies the number of
163        repeats to use when timing. The default is 50000; this gives better
164        precision, but takes uncomfortably long for very large patterns.
165    
166    20. Extended pcre_study() to be more clever in cases where a branch of a
167        subpattern has no definite first character. For example, (a*|b*)[cd] would
168        previously give no result from pcre_study(). Now it recognizes that the
169        first character must be a, b, c, or d.
170    
171    21. There was an incorrect error "recursive call could loop indefinitely" if
172        a subpattern (or the entire pattern) that was being tested for matching an
173        empty string contained only one non-empty item after a nested subpattern.
174        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
175        incorrectly, because the \d was being skipped in the check.
176    
177    22. The pcretest program now has a new pattern option /B and a command line
178        option -b, which is equivalent to adding /B to every pattern. This causes
179        it to show the compiled bytecode, without the additional information that
180        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
181        is the same as /B/I).
182    
183    23. A new optimization is now able automatically to treat some sequences such
184        as a*b as a*+b. More specifically, if something simple (such as a character
185        or a simple class like \d) has an unlimited quantifier, and is followed by
186        something that cannot possibly match the quantified thing, the quantifier
187        is automatically "possessified".
188    
189    24. A recursive reference to a subpattern whose number was greater than 39
190        went wrong under certain circumstances in UTF-8 mode. This bug could also
191        have affected the operation of pcre_study().
192    
193    25. Realized that a little bit of performance could be had by replacing
194        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
195    
196    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
197    
198    27. Possessive quantifiers such as a++ were previously implemented by turning
199        them into atomic groups such as ($>a+). Now they have their own opcodes,
200        which improves performance. This includes the automatically created ones
201        from 23 above.
202    
203    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
204        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
205        the first matched character to be a colon. This applied both to named and
206        numbered groups.
207    
208    29. The ucpinternal.h header file was missing its idempotency #ifdef.
209    
210    30. I was sent a "project" file called libpcre.a.dev which I understand makes
211        building PCRE on Windows easier, so I have included it in the distribution.
212    
213    31. There is now a check in pcretest against a ridiculously large number being
214        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
215        loop, the loop is abandoned.
216    
217    32. Forward references to subpatterns in conditions such as (?(2)...) where
218        subpattern 2 is defined later cause pcre_compile() to search forwards in
219        the pattern for the relevant set of parentheses. This search went wrong
220        when there were unescaped parentheses in a character class, parentheses
221        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
222    
223    33. "Subroutine" calls and backreferences were previously restricted to
224        referencing subpatterns earlier in the regex. This restriction has now
225        been removed.
226    
227    34. Added a number of extra features that are going to be in Perl 5.10. On the
228        whole, these are just syntactic alternatives for features that PCRE had
229        previously implemented using the Python syntax or my own invention. The
230        other formats are all retained for compatibility.
231    
232        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
233            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
234            also .NET compatible.
235    
236        (b) A recursion or subroutine call to a named group can now be defined as
237            (?&name) as well as (?P>name).
238    
239        (c) A backreference to a named group can now be defined as \k<name> or
240            \k'name' as well as (?P=name). The new forms, as well as being in Perl
241            5.10, are also .NET compatible.
242    
243        (d) A conditional reference to a named group can now use the syntax
244            (?(<name>) or (?('name') as well as (?(name).
245    
246        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
247            groups (named and numbered) that are never evaluated inline, but can be
248            called as "subroutines" from elsewhere. In effect, the DEFINE condition
249            is always false. There may be only one alternative in such a group.
250    
251        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
252            as the simple (?(R). The condition is true only if the most recent
253            recursion is that of the given number or name. It does not search out
254            through the entire recursion stack.
255    
256        (g) The escape \gN or \g{N} has been added, where N is a positive or
257            negative number, specifying an absolute or relative reference.
258    
259    35. Tidied to get rid of some further signed/unsigned compiler warnings and
260        some "unreachable code" warnings.
261    
262    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
263        things, this adds five new scripts.
264    
265    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
266        There were also incompatibilities regarding the handling of \Q..\E inside
267        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
268        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
269    
270    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
271        matches an empty string, and forcibly breaks the loop. There were bugs in
272        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
273        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
274        separate and independent bugs (that affected different cases) have been
275        fixed.
276    
277    39. Refactored the code to abolish the use of different opcodes for small
278        capturing bracket numbers. This is a tidy that I avoided doing when I
279        removed the limit on the number of capturing brackets for 3.5 back in 2001.
280        The new approach is not only tidier, it makes it possible to reduce the
281        memory needed to fix the previous bug (38).
282    
283    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
284        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
285        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
286        mode.
287    
288    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
289        report.
290    
291    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
292        copy construction and assignment in the C++ wrapper.
293    
294    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
295        couple of bugs that could have given wrong results in the "--newline=crlf"
296        case.
297    
298    44. Added a number of casts and did some reorganization of signed/unsigned int
299        variables following suggestions from Dair Grant. Also renamed the variable
300        "this" as "item" because it is a C++ keyword.
301    
302    45. Arranged for dftables to add
303    
304          #include "pcre_internal.h"
305    
306        to pcre_chartables.c because without it, gcc 4.x may remove the array
307        definition from the final binary if PCRE is built into a static library and
308        dead code stripping is activated.
309    
310    46. For an unanchored pattern, if a match attempt fails at the start of a
311        newline sequence, and the newline setting is CRLF or ANY, and the next two
312        characters are CRLF, advance by two characters instead of one.
313    
314    
315    Version 6.7 04-Jul-06
316    ---------------------
317    
318     1. In order to handle tests when input lines are enormously long, pcretest has
319        been re-factored so that it automatically extends its buffers when
320        necessary. The code is crude, but this _is_ just a test program. The
321        default size has been increased from 32K to 50K.
322    
323     2. The code in pcre_study() was using the value of the re argument before
324        testing it for NULL. (Of course, in any sensible call of the function, it
325        won't be NULL.)
326    
327     3. The memmove() emulation function in pcre_internal.h, which is used on
328        systems that lack both memmove() and bcopy() - that is, hardly ever -
329        was missing a "static" storage class specifier.
330    
331     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
332        containing an extended class (one that cannot be represented by a bitmap
333        because it contains high-valued characters or Unicode property items, e.g.
334        [\pZ]). Almost always one would set UTF-8 mode when processing such a
335        pattern, but PCRE should not loop if you do not (it no longer does).
336        [Detail: two cases were found: (a) a repeated subpattern containing an
337        extended class; (b) a recursive reference to a subpattern that followed a
338        previous extended class. It wasn't skipping over the extended class
339        correctly when UTF-8 mode was not set.]
340    
341     5. A negated single-character class was not being recognized as fixed-length
342        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
343        compile error "lookbehind assertion is not fixed length".
344    
345     6. The RunPerlTest auxiliary script was showing an unexpected difference
346        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
347        write a Perl script that can interpret lines of an input file either as
348        byte characters or as UTF-8, which is what "perltest" was being required to
349        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
350        can't do is switch easily at run time between having the "use utf8;" pragma
351        or not. In the end, I fudged it by using the RunPerlTest script to insert
352        "use utf8;" explicitly for the UTF-8 tests.
353    
354     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
355        the end of the subject string, contrary to the documentation and to what
356        Perl does. This was true of both matching functions. Now it matches only at
357        the start of the subject and immediately after *internal* newlines.
358    
359     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
360        a pointer to an int instead of a pointer to an unsigned long int. This
361        caused problems on 64-bit systems.
362    
363     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
364        instance of the 'standard' template library not being so standard".
365    
366    10. There was no check on the number of named subpatterns nor the maximum
367        length of a subpattern name. The product of these values is used to compute
368        the size of the memory block for a compiled pattern. By supplying a very
369        long subpattern name and a large number of named subpatterns, the size
370        computation could be caused to overflow. This is now prevented by limiting
371        the length of names to 32 characters, and the number of named subpatterns
372        to 10,000.
373    
374    11. Subpatterns that are repeated with specific counts have to be replicated in
375        the compiled pattern. The size of memory for this was computed from the
376        length of the subpattern and the repeat count. The latter is limited to
377        65535, but there was no limit on the former, meaning that integer overflow
378        could in principle occur. The compiled length of a repeated subpattern is
379        now limited to 30,000 bytes in order to prevent this.
380    
381    12. Added the optional facility to have named substrings with the same name.
382    
383    13. Added the ability to use a named substring as a condition, using the
384        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
385        are numbers (not recommended). Forward references are permitted.
386    
387    14. Added forward references in named backreferences (if you see what I mean).
388    
389    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
390        pattern could run off the end of the subject. For example, the pattern
391        "(?s)(.{1,5})"8 did this with the subject "ab".
392    
393    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
394        PCRE_CASELESS was set when matching characters that were quantified with ?
395        or *.
396    
397    17. A character class other than a single negated character that had a minimum
398        but no maximum quantifier - for example [ab]{6,} - was not handled
399        correctly by pce_dfa_exec(). It would match only one character.
400    
401    18. A valid (though odd) pattern that looked like a POSIX character
402        class but used an invalid character after [ (for example [[,abc,]]) caused
403        pcre_compile() to give the error "Failed: internal error: code overflow" or
404        in some cases to crash with a glibc free() error. This could even happen if
405        the pattern terminated after [[ but there just happened to be a sequence of
406        letters, a binary zero, and a closing ] in the memory that followed.
407    
408    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
409        over the years. Originally (before any Unicode support), just the bottom 8
410        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
411        output from "man perlunicode" includes this:
412    
413          The regular expression compiler produces polymorphic opcodes.  That
414          is, the pattern adapts to the data and automatically switches to
415          the Unicode character scheme when presented with Unicode data--or
416          instead uses a traditional byte scheme when presented with byte
417          data.
418    
419        Sadly, a wide octal escape does not cause a switch, and in a string with
420        no other multibyte characters, these octal escapes are treated as before.
421        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
422        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
423        Unicode string.
424    
425        I have not perpetrated such confusion in PCRE. Up till now, it took just
426        the bottom 8 bits, as in old Perl. I have now made octal escapes with
427        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
428        translate to the appropriate multibyte character.
429    
430    29. Applied some refactoring to reduce the number of warnings from Microsoft
431        and Borland compilers. This has included removing the fudge introduced
432        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
433        a warning about an unused variable.
434    
435    21. PCRE has not included VT (character 0x0b) in the set of whitespace
436        characters since release 4.0, because Perl (from release 5.004) does not.
437        [Or at least, is documented not to: some releases seem to be in conflict
438        with the documentation.] However, when a pattern was studied with
439        pcre_study() and all its branches started with \s, PCRE still included VT
440        as a possible starting character. Of course, this did no harm; it just
441        caused an unnecessary match attempt.
442    
443    22. Removed a now-redundant internal flag bit that recorded the fact that case
444        dependency changed within the pattern. This was once needed for "required
445        byte" processing, but is no longer used. This recovers a now-scarce options
446        bit. Also moved the least significant internal flag bit to the most-
447        significant bit of the word, which was not previously used (hangover from
448        the days when it was an int rather than a uint) to free up another bit for
449        the future.
450    
451    23. Added support for CRLF line endings as well as CR and LF. As well as the
452        default being selectable at build time, it can now be changed at runtime
453        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
454        specify that it is scanning data with non-default line endings.
455    
456    24. Changed the definition of CXXLINK to make it agree with the definition of
457        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
458    
459    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
460        recursions. This makes a big different to stack usage for some patterns.
461    
462    26. If a subpattern containing a named recursion or subroutine reference such
463        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
464        the space required for the compiled pattern went wrong and gave too small a
465        value. Depending on the environment, this could lead to "Failed: internal
466        error: code overflow at offset 49" or "glibc detected double free or
467        corruption" errors.
468    
469    27. Applied patches from Google (a) to support the new newline modes and (b) to
470        advance over multibyte UTF-8 characters in GlobalReplace.
471    
472    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
473        difference for some implementation of PCRE in some Windows version.
474    
475    29. Added some extra testing facilities to pcretest:
476    
477        \q<number>   in a data line sets the "match limit" value
478        \Q<number>   in a data line sets the "match recursion limt" value
479        -S <number>  sets the stack size, where <number> is in megabytes
480    
481        The -S option isn't available for Windows.
482    
483    
484    Version 6.6 06-Feb-06
485    ---------------------
486    
487     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
488        in pcreposix.h. I have copied the definition from pcre.h.
489    
490     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
491        because pcre.h is no longer a built file.
492    
493     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
494        not normally included in the compiled code.
495    
496    
497    Version 6.5 01-Feb-06
498    ---------------------
499    
500     1. When using the partial match feature with pcre_dfa_exec(), it was not
501        anchoring the second and subsequent partial matches at the new starting
502        point. This could lead to incorrect results. For example, with the pattern
503        /1234/, partially matching against "123" and then "a4" gave a match.
504    
505     2. Changes to pcregrep:
506    
507        (a) All non-match returns from pcre_exec() were being treated as failures
508            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
509            error message is output. Some extra information is given for the
510            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
511            probably the only errors that are likely to be caused by users (by
512            specifying a regex that has nested indefinite repeats, for instance).
513            If there are more than 20 of these errors, pcregrep is abandoned.
514    
515        (b) A binary zero was treated as data while matching, but terminated the
516            output line if it was written out. This has been fixed: binary zeroes
517            are now no different to any other data bytes.
518    
519        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
520            used to set a locale for matching. The --locale=xxxx long option has
521            been added (no short equivalent) to specify a locale explicitly on the
522            pcregrep command, overriding the environment variables.
523    
524        (d) When -B was used with -n, some line numbers in the output were one less
525            than they should have been.
526    
527        (e) Added the -o (--only-matching) option.
528    
529        (f) If -A or -C was used with -c (count only), some lines of context were
530            accidentally printed for the final match.
531    
532        (g) Added the -H (--with-filename) option.
533    
534        (h) The combination of options -rh failed to suppress file names for files
535            that were found from directory arguments.
536    
537        (i) Added the -D (--devices) and -d (--directories) options.
538    
539        (j) Added the -F (--fixed-strings) option.
540    
541        (k) Allow "-" to be used as a file name for -f as well as for a data file.
542    
543        (l) Added the --colo(u)r option.
544    
545        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
546            is not present by default.
547    
548     3. A nasty bug was discovered in the handling of recursive patterns, that is,
549        items such as (?R) or (?1), when the recursion could match a number of
550        alternatives. If it matched one of the alternatives, but subsequently,
551        outside the recursion, there was a failure, the code tried to back up into
552        the recursion. However, because of the way PCRE is implemented, this is not
553        possible, and the result was an incorrect result from the match.
554    
555        In order to prevent this happening, the specification of recursion has
556        been changed so that all such subpatterns are automatically treated as
557        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
558    
559     4. I had overlooked the fact that, in some locales, there are characters for
560        which isalpha() is true but neither isupper() nor islower() are true. In
561        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
562        and ordfeminine) are like this. This affected the treatment of \w and \W
563        when they appeared in character classes, but not when they appeared outside
564        a character class. The bit map for "word" characters is now created
565        separately from the results of isalnum() instead of just taking it from the
566        upper, lower, and digit maps. (Plus the underscore character, of course.)
567    
568     5. The above bug also affected the handling of POSIX character classes such as
569        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
570        permanent tables. Instead, the bit maps for such a class were previously
571        created as the appropriate unions of the upper, lower, and digit bitmaps.
572        Now they are created by subtraction from the [[:word:]] class, which has
573        its own bitmap.
574    
575     6. The [[:blank:]] character class matches horizontal, but not vertical space.
576        It is created by subtracting the vertical space characters (\x09, \x0a,
577        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
578        subtraction was done in the overall bitmap for a character class, meaning
579        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
580        be recognized. This bug has been fixed.
581    
582     7. Patches from the folks at Google:
583    
584          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
585          real life, but is still worth protecting against".
586    
587          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
588          regular expressions".
589    
590          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
591          have it.
592    
593          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
594          "configure" and the latter not, in order to fix a problem somebody had
595          with compiling the Arg class on HP-UX.
596    
597          (e) Improve the error-handling of the C++ wrapper a little bit.
598    
599          (f) New tests for checking recursion limiting.
600    
601     8. The pcre_memmove() function, which is used only if the environment does not
602        have a standard memmove() function (and is therefore rarely compiled),
603        contained two bugs: (a) use of int instead of size_t, and (b) it was not
604        returning a result (though PCRE never actually uses the result).
605    
606     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
607        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
608        returned instead of calling malloc() with an overflowing number that would
609        most likely cause subsequent chaos.
610    
611    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
612    
613    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
614        with this option is matched, the nmatch and pmatch options of regexec() are
615        ignored.
616    
617    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
618        provided in case anyone wants to the the POSIX interface with UTF-8
619        strings.
620    
621    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
622        C++ linking (needed for some HP-UX environments).
623    
624    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
625        (unused parameter) and in the pcre_printint() function (omitted "default"
626        switch label when the default is to do nothing).
627    
628    15. Added some code to make it possible, when PCRE is compiled as a C++
629        library, to replace subject pointers for pcre_exec() with a smart pointer
630        class, thus making it possible to process discontinuous strings.
631    
632    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
633        much the same function. They were added by different people who were trying
634        to make PCRE easy to compile on non-Unix systems. It has been suggested
635        that PCRE_EXPORT be abolished now that there is more automatic apparatus
636        for compiling on Windows systems. I have therefore replaced it with
637        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
638        defaults to "extern" for C or "extern C" for C++, which works fine on
639        Unix-like systems. It is now possible to override the value of PCRE_DATA_
640        SCOPE with something explicit in config.h. In addition:
641    
642        (a) pcreposix.h still had just "extern" instead of either of these macros;
643            I have replaced it with PCRE_DATA_SCOPE.
644    
645        (b) Functions such as _pcre_xclass(), which are internal to the library,
646            but external in the C sense, all had PCRE_EXPORT in their definitions.
647            This is apparently wrong for the Windows case, so I have removed it.
648            (It makes no difference on Unix-like systems.)
649    
650    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
651        of recursive calls to match(). This is different to MATCH_LIMIT because
652        that limits the total number of calls to match(), not all of which increase
653        the depth of recursion. Limiting the recursion depth limits the amount of
654        stack (or heap if NO_RECURSE is set) that is used. The default can be set
655        when PCRE is compiled, and changed at run time. A patch from Google adds
656        this functionality to the C++ interface.
657    
658    18. Changes to the handling of Unicode character properties:
659    
660        (a) Updated the table to Unicode 4.1.0.
661    
662        (b) Recognize characters that are not in the table as "Cn" (undefined).
663    
664        (c) I revised the way the table is implemented to a much improved format
665            which includes recognition of ranges. It now supports the ranges that
666            are defined in UnicodeData.txt, and it also amalgamates other
667            characters into ranges. This has reduced the number of entries in the
668            table from around 16,000 to around 3,000, thus reducing its size
669            considerably. I realized I did not need to use a tree structure after
670            all - a binary chop search is just as efficient. Having reduced the
671            number of entries, I extended their size from 6 bytes to 8 bytes to
672            allow for more data.
673    
674        (d) Added support for Unicode script names via properties such as \p{Han}.
675    
676    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
677        matching that character.
678    
679    20. When matching a repeated Unicode property with a minimum greater than zero,
680        (for example \pL{2,}), PCRE could look past the end of the subject if it
681        reached it while seeking the minimum number of characters. This could
682        happen only if some of the characters were more than one byte long, because
683        there is a check for at least the minimum number of bytes.
684    
685    21. Refactored the implementation of \p and \P so as to be more general, to
686        allow for more different types of property in future. This has changed the
687        compiled form incompatibly. Anybody with saved compiled patterns that use
688        \p or \P will have to recompile them.
689    
690    22. Added "Any" and "L&" to the supported property types.
691    
692    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
693        but give a compile time error if the value is greater than 0xff.
694    
695    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
696        accidentally not being installed or uninstalled.
697    
698    25. The pcre.h file was built from pcre.h.in, but the only changes that were
699        made were to insert the current release number. This seemed silly, because
700        it made things harder for people building PCRE on systems that don't run
701        "configure". I have turned pcre.h into a distributed file, no longer built
702        by "configure", with the version identification directly included. There is
703        no longer a pcre.h.in file.
704    
705        However, this change necessitated a change to the pcre-config script as
706        well. It is built from pcre-config.in, and one of the substitutions was the
707        release number. I have updated configure.ac so that ./configure now finds
708        the release number by grepping pcre.h.
709    
710    26. Added the ability to run the tests under valgrind.
711    
712    
713    Version 6.4 05-Sep-05
714    ---------------------
715    
716     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
717        "--" to be printed when multiple files were scanned, even when none of the
718        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
719        consider it to be a bug, and have restored the previous behaviour.
720    
721     2. A couple of code tidies to get rid of compiler warnings.
722    
723     3. The pcretest program used to cheat by referring to symbols in the library
724        whose names begin with _pcre_. These are internal symbols that are not
725        really supposed to be visible externally, and in some environments it is
726        possible to suppress them. The cheating is now confined to including
727        certain files from the library's source, which is a bit cleaner.
728    
729     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
730        file's purpose clearer.
731    
732     5. Reorganized pcre_ucp_findchar().
733    
734    
735    Version 6.3 15-Aug-05
736    ---------------------
737    
738     1. The file libpcre.pc.in did not have general read permission in the tarball.
739    
740     2. There were some problems when building without C++ support:
741    
742        (a) If C++ support was not built, "make install" and "make test" still
743            tried to test it.
744    
745        (b) There were problems when the value of CXX was explicitly set. Some
746            changes have been made to try to fix these, and ...
747    
748        (c) --disable-cpp can now be used to explicitly disable C++ support.
749    
750        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
751            backslash in a target when C++ was disabled. This confuses some
752            versions of "make", apparently. Using an intermediate variable solves
753            this. (Same for CPP_LOBJ.)
754    
755     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
756        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
757        necessary on certain architectures.
758    
759     4. Added a setting of -export-symbols-regex to the link command to remove
760        those symbols that are exported in the C sense, but actually are local
761        within the library, and not documented. Their names all begin with
762        "_pcre_". This is not a perfect job, because (a) we have to except some
763        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
764        available (and never for static libraries). I have made a note to try to
765        find a way round (a) in the future.
766    
767    
768    Version 6.2 01-Aug-05
769    ---------------------
770    
771     1. There was no test for integer overflow of quantifier values. A construction
772        such as {1111111111111111} would give undefined results. What is worse, if
773        a minimum quantifier for a parenthesized subpattern overflowed and became
774        negative, the calculation of the memory size went wrong. This could have
775        led to memory overwriting.
776    
777     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
778    
779     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
780        operating environments where this matters.
781    
782     4. Applied Giuseppe Maxia's patch to add additional features for controlling
783        PCRE options from within the C++ wrapper.
784    
785     5. Named capturing subpatterns were not being correctly counted when a pattern
786        was compiled. This caused two problems: (a) If there were more than 100
787        such subpatterns, the calculation of the memory needed for the whole
788        compiled pattern went wrong, leading to an overflow error. (b) Numerical
789        back references of the form \12, where the number was greater than 9, were
790        not recognized as back references, even though there were sufficient
791        previous subpatterns.
792    
793     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
794        versions of gcc, e.g. 2.95.4.
795    
796    
797    Version 6.1 21-Jun-05
798    ---------------------
799    
800     1. There was one reference to the variable "posix" in pcretest.c that was not
801        surrounded by "#if !defined NOPOSIX".
802    
803     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
804        the cross-check on the old pcre_info() function, for the benefit of the
805        cut-down version of PCRE that is currently imported into Exim.
806    
807     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
808        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
809        patterns that start (?i) but I don't think that matters. The use of (?i) is
810        just an example; this all applies to the other options as well.
811    
812     4. Since libtool seems to echo the compile commands it is issuing, the output
813        from "make" can be reduced a bit by putting "@" in front of each libtool
814        compile command.
815    
816     5. Patch from the folks at Google for configure.in to be a bit more thorough
817        in checking for a suitable C++ installation before trying to compile the
818        C++ stuff. This should fix a reported problem when a compiler was present,
819        but no suitable headers.
820    
821     6. The man pages all had just "PCRE" as their title. I have changed them to
822        be the relevant file name. I have also arranged that these names are
823        retained in the file doc/pcre.txt, which is a concatenation in text format
824        of all the man pages except the little individual ones for each function.
825    
826     7. The NON-UNIX-USE file had not been updated for the different set of source
827        files that come with release 6. I also added a few comments about the C++
828        wrapper.
829    
830    
831    Version 6.0 07-Jun-05
832    ---------------------
833    
834     1. Some minor internal re-organization to help with my DFA experiments.
835    
836     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
837        didn't matter for the library itself when fully configured, but did matter
838        when compiling without UCP support, or within Exim, where the ucp files are
839        not imported.
840    
841     3. Refactoring of the library code to split up the various functions into
842        different source modules. The addition of the new DFA matching code (see
843        below) to a single monolithic source would have made it really too
844        unwieldy, quite apart from causing all the code to be include in a
845        statically linked application, when only some functions are used. This is
846        relevant even without the DFA addition now that patterns can be compiled in
847        one application and matched in another.
848    
849        The downside of splitting up is that there have to be some external
850        functions and data tables that are used internally in different modules of
851        the library but which are not part of the API. These have all had their
852        names changed to start with "_pcre_" so that they are unlikely to clash
853        with other external names.
854    
855     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
856        a different (DFA) algorithm. Although it is slower than the original
857        function, it does have some advantages for certain types of matching
858        problem.
859    
860     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
861        including restarting after a partial match.
862    
863     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
864        defined when compiling for Windows was sent to me. I have put it into the
865        code, though I have no means of testing or verifying it.
866    
867     7. Added the pcre_refcount() auxiliary function.
868    
869     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
870        match before or at the first newline in the subject string. In pcretest,
871        the /f option on a pattern can be used to set this.
872    
873     9. A repeated \w when used in UTF-8 mode with characters greater than 256
874        would behave wrongly. This has been present in PCRE since release 4.0.
875    
876    10. A number of changes to the pcregrep command:
877    
878        (a) Refactored how -x works; insert ^(...)$ instead of setting
879            PCRE_ANCHORED and checking the length, in preparation for adding
880            something similar for -w.
881    
882        (b) Added the -w (match as a word) option.
883    
884        (c) Refactored the way lines are read and buffered so as to have more
885            than one at a time available.
886    
887        (d) Implemented a pcregrep test script.
888    
889        (e) Added the -M (multiline match) option. This allows patterns to match
890            over several lines of the subject. The buffering ensures that at least
891            8K, or the rest of the document (whichever is the shorter) is available
892            for matching (and similarly the previous 8K for lookbehind assertions).
893    
894        (f) Changed the --help output so that it now says
895    
896              -w, --word-regex(p)
897    
898            instead of two lines, one with "regex" and the other with "regexp"
899            because that confused at least one person since the short forms are the
900            same. (This required a bit of code, as the output is generated
901            automatically from a table. It wasn't just a text change.)
902    
903        (g) -- can be used to terminate pcregrep options if the next thing isn't an
904            option but starts with a hyphen. Could be a pattern or a path name
905            starting with a hyphen, for instance.
906    
907        (h) "-" can be given as a file name to represent stdin.
908    
909        (i) When file names are being printed, "(standard input)" is used for
910            the standard input, for compatibility with GNU grep. Previously
911            "<stdin>" was used.
912    
913        (j) The option --label=xxx can be used to supply a name to be used for
914            stdin when file names are being printed. There is no short form.
915    
916        (k) Re-factored the options decoding logic because we are going to add
917            two more options that take data. Such options can now be given in four
918            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
919    
920        (l) Added the -A, -B, and -C options for requesting that lines of context
921            around matches be printed.
922    
923        (m) Added the -L option to print the names of files that do not contain
924            any matching lines, that is, the complement of -l.
925    
926        (n) The return code is 2 if any file cannot be opened, but pcregrep does
927            continue to scan other files.
928    
929        (o) The -s option was incorrectly implemented. For compatibility with other
930            greps, it now suppresses the error message for a non-existent or non-
931            accessible file (but not the return code). There is a new option called
932            -q that suppresses the output of matching lines, which was what -s was
933            previously doing.
934    
935        (p) Added --include and --exclude options to specify files for inclusion
936            and exclusion when recursing.
937    
938    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
939        Hopefully, it now does.
940    
941    12. Missing cast in pcre_study().
942    
943    13. Added an "uninstall" target to the makefile.
944    
945    14. Replaced "extern" in the function prototypes in Makefile.in with
946        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
947        world, but is set differently for Windows.
948    
949    15. Added a second compiling function called pcre_compile2(). The only
950        difference is that it has an extra argument, which is a pointer to an
951        integer error code. When there is a compile-time failure, this is set
952        non-zero, in addition to the error test pointer being set to point to an
953        error message. The new argument may be NULL if no error number is required
954        (but then you may as well call pcre_compile(), which is now just a
955        wrapper). This facility is provided because some applications need a
956        numeric error indication, but it has also enabled me to tidy up the way
957        compile-time errors are handled in the POSIX wrapper.
958    
959    16. Added VPATH=.libs to the makefile; this should help when building with one
960        prefix path and installing with another. (Or so I'm told by someone who
961        knows more about this stuff than I do.)
962    
963    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
964        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
965        match everything, including newlines. This is not POSIX-compatible, but
966        somebody wanted the feature. From pcretest it can be activated by using
967        both the P and the s flags.
968    
969    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
970    
971    19. libpcre.pc was being incorrectly installed as executable.
972    
973    20. A couple of places in pcretest check for end-of-line by looking for '\n';
974        it now also looks for '\r' so that it will work unmodified on Windows.
975    
976    21. Added Google's contributed C++ wrapper to the distribution.
977    
978    22. Added some untidy missing memory free() calls in pcretest, to keep
979        Electric Fence happy when testing.
980    
981    
982    
983    Version 5.0 13-Sep-04
984    ---------------------
985    
986     1. Internal change: literal characters are no longer packed up into items
987        containing multiple characters in a single byte-string. Each character
988        is now matched using a separate opcode. However, there may be more than one
989        byte in the character in UTF-8 mode.
990    
991     2. The pcre_callout_block structure has two new fields: pattern_position and
992        next_item_length. These contain the offset in the pattern to the next match
993        item, and its length, respectively.
994    
995     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
996        insertion of callouts before each pattern item. Added the /C option to
997        pcretest to make use of this.
998    
999     4. On the advice of a Windows user, the lines
1000    
1001          #if defined(_WIN32) || defined(WIN32)
1002          _setmode( _fileno( stdout ), 0x8000 );
1003          #endif  /* defined(_WIN32) || defined(WIN32) */
1004    
1005        have been added to the source of pcretest. This apparently does useful
1006        magic in relation to line terminators.
1007    
1008     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1009        for the benefit of those environments where the "b" makes a difference.
1010    
1011     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1012        to know about it. I have put a hack into configure.in that adds in code
1013        to set GCC=yes if CC=icc. This seems to end up at a point in the
1014        generated configure script that is early enough to affect the setting of
1015        compiler options, which is what is needed, but I have no means of testing
1016        whether it really works. (The user who reported this had patched the
1017        generated configure script, which of course I cannot do.)
1018    
1019        LATER: After change 22 below (new libtool files), the configure script
1020        seems to know about icc (and also ecc). Therefore, I have commented out
1021        this hack in configure.in.
1022    
1023     7. Added support for pkg-config (2 patches were sent in).
1024    
1025     8. Negated POSIX character classes that used a combination of internal tables
1026        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1027        [[:^ascii]]. Typically, they would match almost any characters. The other
1028        POSIX classes were not broken in this way.
1029    
1030     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1031        to find the match, as PCRE was deluded into thinking that the match had to
1032        start at the start point or following a newline. The same bug applied to
1033        patterns with negative forward assertions or any backward assertions
1034        preceding ".*" at the start, unless the pattern required a fixed first
1035        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1036    
1037    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1038        starting at the last subject character, bytes beyond the end of the subject
1039        string were read.
1040    
1041    11. Renamed the variable "class" as "classbits" to make life easier for C++
1042        users. (Previously there was a macro definition, but it apparently wasn't
1043        enough.)
1044    
1045    12. Added the new field "tables" to the extra data so that tables can be passed
1046        in at exec time, or the internal tables can be re-selected. This allows
1047        a compiled regex to be saved and re-used at a later time by a different
1048        program that might have everything at different addresses.
1049    
1050    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1051        -R library as well as a -L library.
1052    
1053    14. The debugging options of pcretest (-d on the command line or D on a
1054        pattern) showed incorrect output for anything following an extended class
1055        that contained multibyte characters and which was followed by a quantifier.
1056    
1057    15. Added optional support for general category Unicode character properties
1058        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1059        support. It adds about 90K to the size of the library. The meanings of the
1060        inbuilt class escapes such as \d and \s have NOT been changed.
1061    
1062    16. Updated pcredemo.c to include calls to free() to release the memory for the
1063        compiled pattern.
1064    
1065    17. The generated file chartables.c was being created in the source directory
1066        instead of in the building directory. This caused the build to fail if the
1067        source directory was different from the building directory, and was
1068        read-only.
1069    
1070    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1071        file. No doubt somebody will tell me if they don't make sense... Also added
1072        Dan Mooney's comments about building on OpenVMS.
1073    
1074    19. Added support for partial matching via the PCRE_PARTIAL option for
1075        pcre_exec() and the \P data escape in pcretest.
1076    
1077    20. Extended pcretest with 3 new pattern features:
1078    
1079        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1080              write the compiled pattern to the file whose name is "rest-of-line".
1081              This is a straight binary dump of the data, with the saved pointer to
1082              the character tables forced to be NULL. The study data, if any, is
1083              written too. After writing, pcretest reads a new pattern.
1084    
1085        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1086              compiled pattern from the given file. There must not be any
1087              occurrences of "<" in the file name (pretty unlikely); if there are,
1088              pcretest will instead treat the initial "<" as a pattern delimiter.
1089              After reading in the pattern, pcretest goes on to read data lines as
1090              usual.
1091    
1092        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1093              and 16-bit fields in a compiled pattern, to simulate a pattern that
1094              was compiled on a host of opposite endianness.
1095    
1096    21. The pcre-exec() function can now cope with patterns that were compiled on
1097        hosts of opposite endianness, with this restriction:
1098    
1099          As for any compiled expression that is saved and used later, the tables
1100          pointer field cannot be preserved; the extra_data field in the arguments
1101          to pcre_exec() should be used to pass in a tables address if a value
1102          other than the default internal tables were used at compile time.
1103    
1104    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1105        now diagnosed as an error. Previously, most of the time, a negative number
1106        would have been treated as zero, but if in addition "ovector" was passed as
1107        NULL, a crash could occur.
1108    
1109    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1110        new versions from the libtool 1.5 distribution (the last one is a copy of
1111        a file called libtool.m4). This seems to have fixed the need to patch
1112        "configure" to support Darwin 1.3 (which I used to do). However, I still
1113        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1114        workstation).
1115    
1116    24. Changed the PCRE licence to be the more standard "BSD" licence.
1117    
1118    
1119    Version 4.5 01-Dec-03
1120    ---------------------
1121    
1122     1. There has been some re-arrangement of the code for the match() function so
1123        that it can be compiled in a version that does not call itself recursively.
1124        Instead, it keeps those local variables that need separate instances for
1125        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1126        needs to "recurse". Keeping track of where control must go is done by means
1127        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1128        hide most of the details from the main code, and operates only if
1129        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1130        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1131        operating.
1132    
1133        To make it easier for callers to provide specially tailored get/free
1134        functions for this usage, two new functions, pcre_stack_malloc, and
1135        pcre_stack_free, are used. They are always called in strict stacking order,
1136        and the size of block requested is always the same.
1137    
1138        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1139        PCRE has been compiled to use the stack or the heap for recursion. The
1140        -C option of pcretest uses this to show which version is compiled.
1141    
1142        A new data escape \S, is added to pcretest; it causes the amounts of store
1143        obtained and freed by both kinds of malloc/free at match time to be added
1144        to the output.
1145    
1146     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1147        what's available on my current Linux desktop machine.
1148    
1149     3. When matching a UTF-8 string, the test for a valid string at the start has
1150        been extended. If start_offset is not zero, PCRE now checks that it points
1151        to a byte that is the start of a UTF-8 character. If not, it returns
1152        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1153        this is necessary because there may be backward assertions in the pattern.
1154        When matching the same subject several times, it may save resources to use
1155        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1156    
1157     4. The code for checking the validity of UTF-8 strings has been tightened so
1158        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1159        containing "overlong sequences".
1160    
1161     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1162        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1163        should have been "&", but it just so happened that all the cases this let
1164        through by mistake were picked up later in the function.
1165    
1166     6. I had used a variable called "isblank" - this is a C99 function, causing
1167        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1168    
1169     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1170        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1171        so the version is shown; (c) stop "make test" echoing "./RunTest".
1172    
1173     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1174    
1175     9. The prototype for memmove() for systems that don't have it was using
1176        size_t, but the inclusion of the header that defines size_t was later. I've
1177        moved the #includes for the C headers earlier to avoid this.
1178    
1179    10. Added some adjustments to the code to make it easier to compiler on certain
1180        special systems:
1181    
1182          (a) Some "const" qualifiers were missing.
1183          (b) Added the macro EXPORT before all exported functions; by default this
1184              is defined to be empty.
1185          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1186              that it reads its output file name as an argument instead of writing
1187              to the standard output and assuming this can be redirected.
1188    
1189    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1190        class containing characters with values greater than 255, PCRE compilation
1191        went into a loop.
1192    
1193    12. A recursive reference to a subpattern that was within another subpattern
1194        that had a minimum quantifier of zero caused PCRE to crash. For example,
1195        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1196        recursion. If the recursively-called subpattern itself had a zero repeat,
1197        that was OK.
1198    
1199    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1200        buffer into which it was copied (for escape processing) was still set at
1201        1024, so long lines caused crashes.
1202    
1203    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1204        "internal error: code overflow...". This applied to any character class
1205        that was followed by a possessive quantifier.
1206    
1207    15. Modified the Makefile to add libpcre.la as a prerequisite for
1208        libpcreposix.la because I was told this is needed for a parallel build to
1209        work.
1210    
1211    16. If a pattern that contained .* following optional items at the start was
1212        studied, the wrong optimizing data was generated, leading to matching
1213        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1214        matching string must start with a or b or c. The correct conclusion for
1215        this pattern is that a match can start with any character.
1216    
1217    
1218    Version 4.4 13-Aug-03
1219    ---------------------
1220    
1221     1. In UTF-8 mode, a character class containing characters with values between
1222        127 and 255 was not handled correctly if the compiled pattern was studied.
1223        In fixing this, I have also improved the studying algorithm for such
1224        classes (slightly).
1225    
1226     2. Three internal functions had redundant arguments passed to them. Removal
1227        might give a very teeny performance improvement.
1228    
1229     3. Documentation bug: the value of the capture_top field in a callout is *one
1230        more than* the number of the hightest numbered captured substring.
1231    
1232     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1233        in incorrectly linking with a previously installed version. They now link
1234        explicitly with libpcre.la.
1235    
1236     5. configure.in no longer needs to recognize Cygwin specially.
1237    
1238     6. A problem in pcre.in for Windows platforms is fixed.
1239    
1240     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1241        pcretest, it used to include the size of the study block as part of its
1242        output. Unfortunately, the structure contains a field that has a different
1243        size on different hardware architectures. This meant that the tests that
1244        showed this size failed. As the block is currently always of a fixed size,
1245        this information isn't actually particularly useful in pcretest output, so
1246        I have just removed it.
1247    
1248     8. Three pre-processor statements accidentally did not start in column 1.
1249        Sadly, there are *still* compilers around that complain, even though
1250        standard C has not required this for well over a decade. Sigh.
1251    
1252     9. In pcretest, the code for checking callouts passed small integers in the
1253        callout_data field, which is a void * field. However, some picky compilers
1254        complained about the casts involved for this on 64-bit systems. Now
1255        pcretest passes the address of the small integer instead, which should get
1256        rid of the warnings.
1257    
1258    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1259        both compile and run time, and gives an error if an invalid UTF-8 sequence
1260        is found. There is a option for disabling this check in cases where the
1261        string is known to be correct and/or the maximum performance is wanted.
1262    
1263    11. In response to a bug report, I changed one line in Makefile.in from
1264    
1265            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1266        to
1267            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1268    
1269        to look similar to other lines, but I have no way of telling whether this
1270        is the right thing to do, as I do not use Windows. No doubt I'll get told
1271        if it's wrong...
1272    
1273    
1274    Version 4.3 21-May-03
1275    ---------------------
1276    
1277    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1278       Makefile.
1279    
1280    2. Some refactoring to improve the quality of the code:
1281    
1282       (i)   The utf8_table... variables are now declared "const".
1283    
1284       (ii)  The code for \cx, which used the "case flipping" table to upper case
1285             lower case letters, now just substracts 32. This is ASCII-specific,
1286             but the whole concept of \cx is ASCII-specific, so it seems
1287             reasonable.
1288    
1289       (iii) PCRE was using its character types table to recognize decimal and
1290             hexadecimal digits in the pattern. This is silly, because it handles
1291             only 0-9, a-f, and A-F, but the character types table is locale-
1292             specific, which means strange things might happen. A private
1293             table is now used for this - though it costs 256 bytes, a table is
1294             much faster than multiple explicit tests. Of course, the standard
1295             character types table is still used for matching digits in subject
1296             strings against \d.
1297    
1298       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1299             ending in _t are). So I've renamed it as ESC_tee.
1300    
1301    3. The first argument for regexec() in the POSIX wrapper should have been
1302       defined as "const".
1303    
1304    4. Changed pcretest to use malloc() for its buffers so that they can be
1305       Electric Fenced for debugging.
1306    
1307    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1308       to read one or more bytes before the start of the subject string. Often this
1309       had no effect on PCRE's behaviour, but in some circumstances it could
1310       provoke a segmentation fault.
1311    
1312    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1313       to try to read one or more bytes before the start of the subject string.
1314    
1315    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1316       UTF-8 support could misbehave in various ways if the subject string
1317       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1318       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1319       back over UTF-8 characters.)
1320    
1321    
1322    Version 4.2 14-Apr-03
1323    ---------------------
1324    
1325    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1326    
1327    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1328         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1329         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1330         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1331         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1332           and BUILD_EXEEXT
1333         Note: automatic setting of the BUILD variables is not yet working
1334         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1335           compile-time but not at link-time
1336         [LINK]: use for linking executables only
1337         make different versions for Windows and non-Windows
1338         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1339           libraries
1340         [LINK_FOR_BUILD]: new variable
1341         [OBJEXT]: use throughout
1342         [EXEEXT]: use throughout
1343         <winshared>: new target
1344         <wininstall>: new target
1345         <dftables.o>: use native compiler
1346         <dftables>: use native linker
1347         <install>: handle Windows platform correctly
1348         <clean>: ditto
1349         <check>: ditto
1350         copy DLL to top builddir before testing
1351    
1352       As part of these changes, -no-undefined was removed again. This was reported
1353       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1354       in any case.
1355    
1356    3. Some tidies to get rid of compiler warnings:
1357    
1358       . In the match_data structure, match_limit was an unsigned long int, whereas
1359         match_call_count was an int. I've made them both unsigned long ints.
1360    
1361       . In pcretest the fact that a const uschar * doesn't automatically cast to
1362         a void * provoked a warning.
1363    
1364       . Turning on some more compiler warnings threw up some "shadow" variables
1365         and a few more missing casts.
1366    
1367    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1368       option, a class that contained a single character with a value between 128
1369       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1370    
1371    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1372       option, a class that contained several characters, but with at least one
1373       whose value was between 128 and 255 caused PCRE to crash.
1374    
1375    
1376    Version 4.1 12-Mar-03
1377    ---------------------
1378    
1379    1. Compiling with gcc -pedantic found a couple of places where casts were
1380    needed, and a string in dftables.c that was longer than standard compilers are
1381    required to support.
1382    
1383    2. Compiling with Sun's compiler found a few more places where the code could
1384    be tidied up in order to avoid warnings.
1385    
1386    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
1387    first of these names is deprecated in the latest Autoconf in favour of the name
1388    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
1389    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
1390    analogy I have changed it to CFLAGS_FOR_BUILD.
1391    
1392    4. Added -no-undefined to the linking command in the Makefile, because this is
1393    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
1394    linking step for the pcreposix library.
1395    
1396    5. PCRE was failing to diagnose the case of two named groups with the same
1397    name.
1398    
1399    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
1400    literal character that is needed in the subject for a match, and scans along to
1401    ensure that it is present before embarking on the full matching process. This
1402    saves time in cases of nested unlimited repeats that are never going to match.
1403    Problem: the scan can take a lot of time if the subject is very long (e.g.
1404    megabytes), thus penalizing straightforward matches. It is now done only if the
1405    amount of subject to be scanned is less than 1000 bytes.
1406    
1407    7. A lesser problem with the same optimization is that it was recording the
1408    first character of an anchored pattern as "needed", thus provoking a search
1409    right along the subject, even when the first match of the pattern was going to
1410    fail. The "needed" character is now not set for anchored patterns, unless it
1411    follows something in the pattern that is of non-fixed length. Thus, it still
1412    fulfils its original purpose of finding quick non-matches in cases of nested
1413    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
1414    
1415    
1416    Version 4.0 17-Feb-03
1417    ---------------------
1418    
1419    1. If a comment in an extended regex that started immediately after a meta-item
1420    extended to the end of string, PCRE compiled incorrect data. This could lead to
1421    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
1422    
1423    2. Moved to autoconf 2.53 and libtool 1.4.2.
1424    
1425    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
1426    the special perltest8 script is no longer needed - all the tests can be run
1427    from a single perltest script.
1428    
1429    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
1430    by \s. It has now been removed in PCRE. This means it isn't recognized as
1431    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
1432    class [:space:] *does* include VT, thereby creating a mess.
1433    
1434    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
1435    space and tab.
1436    
1437    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
1438    its new features into the main test script, reducing the number of scripts.
1439    
1440    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
1441    were backward compatible, and made the (?i) apply to the whole pattern, as if
1442    /i were given. Now it behaves more logically, and applies the option setting
1443    only to what follows. PCRE has been changed to follow suit. However, if it
1444    finds options settings right at the start of the pattern, it extracts them into
1445    the global options, as before. Thus, they show up in the info data.
1446    
1447    8. Added support for the \Q...\E escape sequence. Characters in between are
1448    treated as literals. This is slightly different from Perl in that $ and @ are
1449    also handled as literals inside the quotes. In Perl, they will cause variable
1450    interpolation. Note the following examples:
1451    
1452        Pattern            PCRE matches      Perl matches
1453    
1454        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
1455        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
1456        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
1457    
1458    For compatibility with Perl, \Q...\E sequences are recognized inside character
1459    classes as well as outside them.
1460    
1461    9. Re-organized 3 code statements in pcretest to avoid "overflow in
1462    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
1463    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
1464    signed/unsigned warnings.
1465    
1466    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
1467    option for pcretest, so I've replaced it by a simple function that does just
1468    that job.
1469    
1470    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
1471    "pcregrep -".
1472    
1473    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
1474    Java package. This provides some syntactic sugar for simple cases of what my
1475    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
1476    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
1477    item, you can use this simplified notation. Note that only makes sense with
1478    greedy quantifiers. Consequently, the use of the possessive quantifier forces
1479    greediness, whatever the setting of the PCRE_UNGREEDY option.
1480    
1481    13. A change of greediness default within a pattern was not taking effect at
1482    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
1483    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
1484    was abstracted outside.
1485    
1486    14. PCRE now supports the \G assertion. It is true when the current matching
1487    position is at the start point of the match. This differs from \A when the
1488    starting offset is non-zero. Used with the /g option of pcretest (or similar
1489    code), it works in the same way as it does for Perl's /g option. If all
1490    alternatives of a regex begin with \G, the expression is anchored to the start
1491    match position, and the "anchored" flag is set in the compiled expression.
1492    
1493    15. Some bugs concerning the handling of certain option changes within patterns
1494    have been fixed. These applied to options other than (?ims). For example,
1495    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
1496    been the other way round. Some of this was related to change 7 above.
1497    
1498    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
1499    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
1500    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
1501    POSIX classes only within a class (e.g. /[[:alpha:]]/).
1502    
1503    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
1504    mode. Unlike ".", it always matches newline, whatever the setting of
1505    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
1506    assertions. Perl allows it, but it doesn't (in general) work because it can't
1507    calculate the length of the lookbehind. At least, that's the case for Perl
1508    5.8.0 - I've been told they are going to document that it doesn't work in
1509    future.
1510    
1511    18. Added an error diagnosis for escapes that PCRE does not support: these are
1512    \L, \l, \N, \P, \p, \U, \u, and \X.
1513    
1514    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
1515    reading past the end of the pattern in cases such as /[abcd/.
1516    
1517    20. PCRE was getting more memory than necessary for patterns with classes that
1518    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
1519    
1520    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
1521    compiling PCRE for use with Virtual Pascal.
1522    
1523    22. Small fix to the Makefile to make it work properly if the build is done
1524    outside the source tree.
1525    
1526    23. Added a new extension: a condition to go with recursion. If a conditional
1527    subpattern starts with (?(R) the "true" branch is used if recursion has
1528    happened, whereas the "false" branch is used only at the top level.
1529    
1530    24. When there was a very long string of literal characters (over 255 bytes
1531    without UTF support, over 250 bytes with UTF support), the computation of how
1532    much memory was required could be incorrect, leading to segfaults or other
1533    strange effects.
1534    
1535    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
1536    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
1537    there was a subsequent back reference to those brackets. This meant that, for
1538    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
1539    possible to check for precisely this case. All we can do is abandon the
1540    optimization if .* occurs inside capturing brackets when there are any back
1541    references whatsoever. (See below for a better fix that came later.)
1542    
1543    26. The handling of the optimization for finding the first character of a
1544    non-anchored pattern, and for finding a character that is required later in the
1545    match were failing in some cases. This didn't break the matching; it just
1546    failed to optimize when it could. The way this is done has been re-implemented.
1547    
1548    27. Fixed typo in error message for invalid (?R item (it said "(?p").
1549    
1550    28. Added a new feature that provides some of the functionality that Perl
1551    provides with (?{...}). The facility is termed a "callout". The way it is done
1552    in PCRE is for the caller to provide an optional function, by setting
1553    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
1554    global variable. By default it is unset, which disables all calling out. To get
1555    the function called, the regex must include (?C) at appropriate points. This
1556    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
1557    This provides a means of identifying different callout points. When PCRE
1558    reaches such a point in the regex, if pcre_callout has been set, the external
1559    function is called. It is provided with data in a structure called
1560    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
1561    matching continues; if it returns a non-zero value, the match at the current
1562    point fails. However, backtracking will occur if possible. [This was changed
1563    later and other features added - see item 49 below.]
1564    
1565    29. pcretest is upgraded to test the callout functionality. It provides a
1566    callout function that displays information. By default, it shows the start of
1567    the match and the current position in the text. There are some new data escapes
1568    to vary what happens:
1569    
1570        \C+         in addition, show current contents of captured substrings
1571        \C-         do not supply a callout function
1572        \C!n        return 1 when callout number n is reached
1573        \C!n!m      return 1 when callout number n is reached for the mth time
1574    
1575    30. If pcregrep was called with the -l option and just a single file name, it
1576    output "<stdin>" if a match was found, instead of the file name.
1577    
1578    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
1579    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
1580    pcre_exec(). This saves a malloc/free per call. The default value of
1581    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
1582    when configuring.
1583    
1584    32. The default maximum size of a compiled pattern is 64K. There have been a
1585    few cases of people hitting this limit. The code now uses macros to handle the
1586    storing of links as offsets within the compiled pattern. It defaults to 2-byte
1587    links, but this can be changed to 3 or 4 bytes by --with-link-size when
1588    configuring. Tests 2 and 5 work only with 2-byte links because they output
1589    debugging information about compiled patterns.
1590    
1591    33. Internal code re-arrangements:
1592    
1593    (a) Moved the debugging function for printing out a compiled regex into
1594        its own source file (printint.c) and used #include to pull it into
1595        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
1596        separate copies.
1597    
1598    (b) Defined the list of op-code names for debugging as a macro in
1599        internal.h so that it is next to the definition of the opcodes.
1600    
1601    (c) Defined a table of op-code lengths for simpler skipping along compiled
1602        code. This is again a macro in internal.h so that it is next to the
1603        definition of the opcodes.
1604    
1605    34. Added support for recursive calls to individual subpatterns, along the
1606    lines of Robin Houston's patch (but implemented somewhat differently).
1607    
1608    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
1609    allow it to read and process whole directories in Win32. This code was
1610    contributed by Lionel Fourquaux; it has not been tested by me.
1611    
1612    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
1613    used to name a group. Names consist of alphanumerics and underscores, and must
1614    be unique. Back references use the syntax (?P=name) and recursive calls use
1615    (?P>name) which is a PCRE extension to the Python extension. Groups still have
1616    numbers. The function pcre_fullinfo() can be used after compilation to extract
1617    a name/number map. There are three relevant calls:
1618    
1619      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
1620      PCRE_INFO_NAMECOUNT            yields the number of entries
1621      PCRE_INFO_NAMETABLE            yields a pointer to the map.
1622    
1623    The map is a vector of fixed-size entries. The size of each entry depends on
1624    the length of the longest name used. The first two bytes of each entry are the
1625    group number, most significant byte first. There follows the corresponding
1626    name, zero terminated. The names are in alphabetical order.
1627    
1628    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
1629    case instead of 255. Making it the same both with and without UTF-8 support
1630    means that the same test output works with both.
1631    
1632    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
1633    calling malloc() with a zero argument.
1634    
1635    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
1636    optimization. I've improved things by keeping a bitmap of backreferences with
1637    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
1638    fact referenced, the optimization can be applied. It is unlikely that a
1639    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
1640    the match to follow \n) will appear inside brackets with a number greater than
1641    31, but if it does, any back reference > 31 suppresses the optimization.
1642    
1643    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
1644    of disabling numbered capturing parentheses. Any opening parenthesis that is
1645    not followed by ? behaves as if it were followed by ?: but named parentheses
1646    can still be used for capturing (and they will acquire numbers in the usual
1647    way).
1648    
1649    41. Redesigned the return codes from the match() function into yes/no/error so
1650    that errors can be passed back from deep inside the nested calls. A malloc
1651    failure while inside a recursive subpattern call now causes the
1652    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
1653    
1654    42. It is now possible to set a limit on the number of times the match()
1655    function is called in a call to pcre_exec(). This facility makes it possible to
1656    limit the amount of recursion and backtracking, though not in a directly
1657    obvious way, because the match() function is used in a number of different
1658    circumstances. The count starts from zero for each position in the subject
1659    string (for non-anchored patterns). The default limit is, for compatibility, a
1660    large number, namely 10 000 000. You can change this in two ways:
1661    
1662    (a) When configuring PCRE before making, you can use --with-match-limit=n
1663        to set a default value for the compiled library.
1664    
1665    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
1666        a different value is set. See 45 below.
1667    
1668    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1669    
1670    43. Added a new function pcre_config(int, void *) to enable run-time extraction
1671    of things that can be changed at compile time. The first argument specifies
1672    what is wanted and the second points to where the information is to be placed.
1673    The current list of available information is:
1674    
1675      PCRE_CONFIG_UTF8
1676    
1677    The output is an integer that is set to one if UTF-8 support is available;
1678    otherwise it is set to zero.
1679    
1680      PCRE_CONFIG_NEWLINE
1681    
1682    The output is an integer that it set to the value of the code that is used for
1683    newline. It is either LF (10) or CR (13).
1684    
1685      PCRE_CONFIG_LINK_SIZE
1686    
1687    The output is an integer that contains the number of bytes used for internal
1688    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
1689    
1690      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1691    
1692    The output is an integer that contains the threshold above which the POSIX
1693    interface uses malloc() for output vectors. See item 31 above.
1694    
1695      PCRE_CONFIG_MATCH_LIMIT
1696    
1697    The output is an unsigned integer that contains the default limit of the number
1698    of match() calls in a pcre_exec() execution. See 42 above.
1699    
1700    44. pcretest has been upgraded by the addition of the -C option. This causes it
1701    to extract all the available output from the new pcre_config() function, and to
1702    output it. The program then exits immediately.
1703    
1704    45. A need has arisen to pass over additional data with calls to pcre_exec() in
1705    order to support additional features. One way would have been to define
1706    pcre_exec2() (for example) with extra arguments, but this would not have been
1707    extensible, and would also have required all calls to the original function to
1708    be mapped to the new one. Instead, I have chosen to extend the mechanism that
1709    is used for passing in "extra" data from pcre_study().
1710    
1711    The pcre_extra structure is now exposed and defined in pcre.h. It currently
1712    contains the following fields:
1713    
1714      flags         a bitmap indicating which of the following fields are set
1715      study_data    opaque data from pcre_study()
1716      match_limit   a way of specifying a limit on match() calls for a specific
1717                      call to pcre_exec()
1718      callout_data  data for callouts (see 49 below)
1719    
1720    The flag bits are also defined in pcre.h, and are
1721    
1722      PCRE_EXTRA_STUDY_DATA
1723      PCRE_EXTRA_MATCH_LIMIT
1724      PCRE_EXTRA_CALLOUT_DATA
1725    
1726    The pcre_study() function now returns one of these new pcre_extra blocks, with
1727    the actual study data pointed to by the study_data field, and the
1728    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
1729    before. That is, this change is entirely upwards-compatible and requires no
1730    change to existing code.
1731    
1732    If you want to pass in additional data to pcre_exec(), you can either place it
1733    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
1734    block.
1735    
1736    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
1737    data string contains the escape sequence \M, pcretest calls pcre_exec() several
1738    times with different match limits, until it finds the minimum value needed for
1739    pcre_exec() to complete. The value is then output. This can be instructive; for
1740    most simple matches the number is quite small, but for pathological cases it
1741    gets very large very quickly.
1742    
1743    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
1744    returns the size of the data block pointed to by the study_data field in a
1745    pcre_extra block, that is, the value that was passed as the argument to
1746    pcre_malloc() when PCRE was getting memory in which to place the information
1747    created by pcre_study(). The fourth argument should point to a size_t variable.
1748    pcretest has been extended so that this information is shown after a successful
1749    pcre_study() call when information about the compiled regex is being displayed.
1750    
1751    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
1752    because what follows is always an absolute path. (Later: it turns out that this
1753    is more than cosmetic for MinGW, because it doesn't like empty path
1754    components.)
1755    
1756    49. Some changes have been made to the callout feature (see 28 above):
1757    
1758    (i)  A callout function now has three choices for what it returns:
1759    
1760           0  =>  success, carry on matching
1761         > 0  =>  failure at this point, but backtrack if possible
1762         < 0  =>  serious error, return this value from pcre_exec()
1763    
1764         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
1765         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
1766         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
1767         use by callout functions. It will never be used by PCRE itself.
1768    
1769    (ii) The pcre_extra structure (see 45 above) has a void * field called
1770         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
1771         pcre_callout_block structure has a field of the same name. The contents of
1772         the field passed in the pcre_extra structure are passed to the callout
1773         function in the corresponding field in the callout block. This makes it
1774         easier to use the same callout-containing regex from multiple threads. For
1775         testing, the pcretest program has a new data escape
1776    
1777           \C*n        pass the number n (may be negative) as callout_data
1778    
1779         If the callout function in pcretest receives a non-zero value as
1780         callout_data, it returns that value.
1781    
1782    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
1783    there were some redundant $(CFLAGS) in commands that are now specified as
1784    $(LINK), which already includes $(CFLAGS).
1785    
1786    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
1787    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
1788    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
1789    one-byte characters throughout. Note that case-insensitive matching applies
1790    only to characters whose values are less than 256. PCRE doesn't support the
1791    notion of cases for higher-valued characters.
1792    
1793    (i)   A character class whose characters are all within 0-255 is handled as
1794          a bit map, and the map is inverted for negative classes. Previously, a
1795          character > 255 always failed to match such a class; however it should
1796          match if the class was a negative one (e.g. [^ab]). This has been fixed.
1797    
1798    (ii)  A negated character class with a single character < 255 is coded as
1799          "not this character" (OP_NOT). This wasn't working properly when the test
1800          character was multibyte, either singly or repeated.
1801    
1802    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1803          mode, for example: \x{100}{2,3}.
1804    
1805    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1806          singly or repeated) now correctly test multibyte characters. However,
1807          PCRE doesn't recognize any characters with values greater than 255 as
1808          digits, spaces, or word characters. Such characters always match \D, \S,
1809          and \W, and never match \d, \s, or \w.
1810    
1811    (v)   Classes may now contain characters and character ranges with values
1812          greater than 255. For example: [ab\x{100}-\x{400}].
1813    
1814    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
1815          PCRE in UTF-8 mode.
1816    
1817    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1818    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1819    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1820    value.)
1821    
1822    53. The single man page has become too large. I have therefore split it up into
1823    a number of separate man pages. These also give rise to individual HTML pages;
1824    these are now put in a separate directory, and there is an index.html page that
1825    lists them all. Some hyperlinking between the pages has been installed.
1826    
1827    54. Added convenience functions for handling named capturing parentheses.
1828    
1829    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1830    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1831    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1832    are faulted.
1833    
1834    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1835    calling configure. These values are used when compiling the dftables.c program
1836    which is run to generate the source of the default character tables. They
1837    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1838    you will need to set these values.
1839    
1840    57. Updated the building process for Windows DLL, as provided by Fred Cox.
1841    
1842    
1843    Version 3.9 02-Jan-02
1844    ---------------------
1845    
1846    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1847    
1848    2. If --disable-static was given, the building process failed when trying to
1849    build pcretest and pcregrep. (For some reason it was using libtool to compile
1850    them, which is not right, as they aren't part of the library.)
1851    
1852    
1853    Version 3.8 18-Dec-01
1854    ---------------------
1855    
1856    1. The experimental UTF-8 code was completely screwed up. It was packing the
1857    bytes in the wrong order. How dumb can you get?
1858    
1859    
1860    Version 3.7 29-Oct-01
1861    ---------------------
1862    
1863    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1864    This caused pcretest, when used on the test data, to segfault. Unfortunately,
1865    this didn't happen under Solaris 8, where I normally test things.
1866    
1867    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1868    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1869    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1870    this fix an hour or so after the initial 3.7 release.)
1871    
1872    
1873    Version 3.6 23-Oct-01
1874    ---------------------
1875    
1876    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1877    offsets passed as NULL with zero offset count.
1878    
1879    2. The config.guess and config.sub files had not been updated when I moved to
1880    the latest autoconf.
1881    
1882    
1883    Version 3.5 15-Aug-01
1884    ---------------------
1885    
1886    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1887    had been forgotten.
1888    
1889    2. By using declared but undefined structures, we can avoid using "void"
1890    definitions in pcre.h while keeping the internal definitions of the structures
1891    private.
1892    
1893    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1894    user point of view, this means that both static and shared libraries are built
1895    by default, but this can be individually controlled. More of the work of
1896    handling this static/shared cases is now inside libtool instead of PCRE's make
1897    file.
1898    
1899    4. The pcretest utility is now installed along with pcregrep because it is
1900    useful for users (to test regexs) and by doing this, it automatically gets
1901    relinked by libtool. The documentation has been turned into a man page, so
1902    there are now .1, .txt, and .html versions in /doc.
1903    
1904    5. Upgrades to pcregrep:
1905       (i)   Added long-form option names like gnu grep.
1906       (ii)  Added --help to list all options with an explanatory phrase.
1907       (iii) Added -r, --recursive to recurse into sub-directories.
1908       (iv)  Added -f, --file to read patterns from a file.
1909    
1910    6. pcre_exec() was referring to its "code" argument before testing that
1911    argument for NULL (and giving an error if it was NULL).
1912    
1913    7. Upgraded Makefile.in to allow for compiling in a different directory from
1914    the source directory.
1915    
1916    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
1917    options bits, the pointer it was passed was to an int instead of to an unsigned
1918    long int. This mattered only on 64-bit systems.
1919    
1920    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
1921    generated) instead of pcre.in, which it its source. Also made the same change
1922    in several of the .c files.
1923    
1924    10. A new release of gcc defines printf() as a macro, which broke pcretest
1925    because it had an ifdef in the middle of a string argument for printf(). Fixed
1926    by using separate calls to printf().
1927    
1928    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
1929    script, to force use of CR or LF instead of \n in the source. On non-Unix
1930    systems, the value can be set in config.h.
1931    
1932    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
1933    absolute limit. Changed the text of the error message to make this clear, and
1934    likewise updated the man page.
1935    
1936    13. The limit of 99 on the number of capturing subpatterns has been removed.
1937    The new limit is 65535, which I hope will not be a "real" limit.
1938    
1939    
1940    Version 3.4 22-Aug-00
1941    ---------------------
1942    
1943    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
1944    
1945    2. Diagnose condition (?(0) as an error instead of crashing on matching.
1946    
1947    
1948    Version 3.3 01-Aug-00
1949    ---------------------
1950    
1951    1. If an octal character was given, but the value was greater than \377, it
1952    was not getting masked to the least significant bits, as documented. This could
1953    lead to crashes in some systems.
1954    
1955    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
1956    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
1957    
1958    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
1959    These just pass their arguments on to (pcre_free)(), but they are provided
1960    because some uses of PCRE bind it to non-C systems that can call its functions,
1961    but cannot call free() or pcre_free() directly.
1962    
1963    4. Add "make test" as a synonym for "make check". Corrected some comments in
1964    the Makefile.
1965    
1966    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
1967    Makefile.
1968    
1969    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
1970    command called pgrep for grepping around the active processes.
1971    
1972    7. Added the beginnings of support for UTF-8 character strings.
1973    
1974    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
1975    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
1976    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
1977    out for the ar command.)
1978    
1979    
1980    Version 3.2 12-May-00
1981    ---------------------
1982    
1983    This is purely a bug fixing release.
1984    
1985    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
1986    of ZA. This was just one example of several cases that could provoke this bug,
1987    which was introduced by change 9 of version 2.00. The code for breaking
1988    infinite loops after an iteration that matches an empty string was't working
1989    correctly.
1990    
1991    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
1992    when matched against abbab (for example). After matching an empty string, it
1993    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
1994    caused it to match further down the string than it should.
1995    
1996    3. The code contained an inclusion of sys/types.h. It isn't clear why this
1997    was there because it doesn't seem to be needed, and it causes trouble on some
1998    systems, as it is not a Standard C header. It has been removed.
1999    
2000    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2001    were reported on the Macintosh. The changes were from
2002    
2003      while ((c = *(++ptr)) != 0 && c != '\n');
2004    to
2005      while ((c = *(++ptr)) != 0 && c != '\n') ;
2006    
2007    Totally extraordinary, but if that's what it takes...
2008    
2009    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2010    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2011    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2012    assumes the way PCRE uses memmove() (always moving upwards).
2013    
2014    6. PCRE is being used in one environment where strchr() is not available. There
2015    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2016    faster code anyway.
2017    
2018    
2019    Version 3.1 09-Feb-00
2020    ---------------------
2021    
2022    The only change in this release is the fixing of some bugs in Makefile.in for
2023    the "install" target:
2024    
2025    (1) It was failing to install pcreposix.h.
2026    
2027    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2028    
2029  Version 2.09 14-Sep-99  
2030  ----------------------  Version 3.0 01-Feb-00
2031    ---------------------
2032    
2033  1. Add support for the /+ modifier to perltest (to output $` like it does in  1. Add support for the /+ modifier to perltest (to output $` like it does in
2034  pcretest).  pcretest).
# Line 23  captured string vector to pcre_exec(), b Line 2048  captured string vector to pcre_exec(), b
2048  required a bigger vector, with some working space on the end. This means that  required a bigger vector, with some working space on the end. This means that
2049  the POSIX wrapper now has to get and free some memory, and copy the results.  the POSIX wrapper now has to get and free some memory, and copy the results.
2050    
2051    6. Added some simple autoconf support, placing the test data and the
2052    documentation in separate directories, re-organizing some of the
2053    information files, and making it build pcre-config (a GNU standard). Also added
2054    libtool support for building PCRE as a shared library, which is now the
2055    default.
2056    
2057    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2058    09 are not valid octal constants. Single digits will be used for minor values
2059    less than 10.
2060    
2061    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2062    existing programs that set these in the POSIX interface can use PCRE without
2063    modification.
2064    
2065    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2066    return all that pcre_info() returns, plus additional data. The pcre_info()
2067    function is retained for compatibility, but is considered to be obsolete.
2068    
2069    10. Added experimental recursion feature (?R) to handle one common case that
2070    Perl 5.6 will be able to do with (?p{...}).
2071    
2072    11. Added support for POSIX character classes like [:alpha:], which Perl is
2073    adopting.
2074    
2075    
2076  Version 2.08 31-Aug-99  Version 2.08 31-Aug-99
2077  ----------------------  ----------------------

Legend:
Removed from v.41  
changed lines
  Added in v.102

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12