ViewVC logotype

Diff of /code/trunk/ChangeLog

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

revision 51 by nigel, Sat Feb 24 21:39:37 2007 UTC revision 332 by ph10, Sat Apr 5 16:11:05 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
4    Version 7.7 05-Mar-08
5    ---------------------
7    1.  Applied Craig's patch to sort out a long long problem: "If we can't convert
8        a string to a long long, pretend we don't even have a long long." This is
9        done by checking for the strtoq, strtoll, and _strtoi64 functions.
11    2.  Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
12        pre-7.6 versions, which defined a global no_arg variable instead of putting
13        it in the RE class. (See also #8 below.)
15    3.  Remove a line of dead code, identified by coverity and reported by Nuno
16        Lopes.
18    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
20        (1) The include/exclude patterns were being applied to the whole pathnames
21            of files, instead of just to the final components.
23        (2) If there was more than one level of directory, the subdirectories were
24            skipped unless they satisfied the include/exclude conditions. This is
25            inconsistent with GNU grep (and could even be seen as contrary to the
26            pcregrep specification - which I improved to make it absolutely clear).
27            The action now is always to scan all levels of directory, and just
28            apply the include/exclude patterns to regular files.
30    5.  Added the --include_dir and --exclude_dir patterns to pcregrep, and used
31        --exclude_dir in the tests to avoid scanning .svn directories.
33    6.  Applied Craig's patch to the QuoteMeta function so that it escapes the
34        NUL character as backslash + 0 rather than backslash + NUL, because PCRE
35        doesn't support NULs in patterns.
37    7.  Added some missing "const"s to declarations of static tables in
38        pcre_compile.c and pcre_dfa_exec.c.
40    8.  Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
41        caused by fix #2  above. (Subsequently also a second patch to fix the
42        first patch. And a third patch - this was a messy problem.)
44    9.  Applied Craig's patch to remove the use of push_back().
46    10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
47        matching function regexec().
50    Version 7.6 28-Jan-08
51    ---------------------
53    1.  A character class containing a very large number of characters with
54        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
55        overflow.
57    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
58        HAVE_LONG_LONG is not defined.
60    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
61        bring them up to date and include new features. This patch includes:
63        - Fixed PH's badly added libz and libbz2 support.
64        - Fixed a problem with static linking.
65        - Added pcredemo. [But later removed - see 7 below.]
66        - Fixed dftables problem and added an option.
67        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
68            HAVE_LONG_LONG.
69        - Added readline support for pcretest.
70        - Added an listing of the option settings after cmake has run.
72    4.  A user submitted a patch to Makefile that makes it easy to create
73        "pcre.dll" under mingw when using Configure/Make. I added stuff to
74        Makefile.am that cause it to include this special target, without
75        affecting anything else. Note that the same mingw target plus all
76        the other distribution libraries and programs are now supported
77        when configuring with CMake (see 6 below) instead of with
78        Configure/Make.
80    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
81        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
82        exported in the Windows port". It has not yet been confirmed that the patch
83        solves the problem, but it does no harm.
85    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
86        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
87        with CMake, and also correct the comment about stack recursion.
89    7.  Remove the automatic building of pcredemo from the ./configure system and
90        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
91        of a program that users should build themselves after PCRE is installed, so
92        building it automatically is not really right. What is more, it gave
93        trouble in some build environments.
95    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
98    Version 7.5 10-Jan-08
99    ---------------------
101    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
102        values in parens when parsing an RE using the C++ wrapper."
104    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
105        Characters greater than 255 were excluded from the class instead of being
106        included.
108    3.  The same bug as (2) above applied to negated POSIX classes such as
109        [:^space:].
111    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
112        defined or documented. It seems to have been a typo for PCRE_STATIC, so
113        I have changed it.
115    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
116        first named subpattern) and a construct such as (?&a) would reference the
117        first named subpattern whose name started with "a" (in other words, the
118        length check was missing). Both these problems are fixed. "Subpattern name
119        expected" is now given for (?&) (a zero-length name), and this patch also
120        makes it give the same error for \k'' (previously it complained that that
121        was a reference to a non-existent subpattern).
123    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
124        this is right because (?- can be followed by option settings as well as by
125        digits. I have, however, made the messages clearer.
127    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
128        than the number used in the conditional) now cause a compile-time error.
129        This is actually not compatible with Perl, which accepts such patterns, but
130        treats the conditional as always being FALSE (as PCRE used to), but it
131        seems to me that giving a diagnostic is better.
133    8.  Change "alphameric" to the more common word "alphanumeric" in comments
134        and messages.
136    9.  Fix two occurrences of "backslash" in comments that should have been
137        "backspace".
139    10. Remove two redundant lines of code that can never be obeyed (their function
140        was moved elsewhere).
142    11. The program that makes PCRE's Unicode character property table had a bug
143        which caused it to generate incorrect table entries for sequences of
144        characters that have the same character type, but are in different scripts.
145        It amalgamated them into a single range, with the script of the first of
146        them. In other words, some characters were in the wrong script. There were
147        thirteen such cases, affecting characters in the following ranges:
149          U+002b0 - U+002c1
150          U+0060c - U+0060d
151          U+0061e - U+00612
152          U+0064b - U+0065e
153          U+0074d - U+0076d
154          U+01800 - U+01805
155          U+01d00 - U+01d77
156          U+01d9b - U+01dbf
157          U+0200b - U+0200f
158          U+030fc - U+030fe
159          U+03260 - U+0327f
160          U+0fb46 - U+0fbb1
161          U+10450 - U+1049d
163    12. The -o option (show only the matching part of a line) for pcregrep was not
164        compatible with GNU grep in that, if there was more than one match in a
165        line, it showed only the first of them. It now behaves in the same way as
166        GNU grep.
168    13. If the -o and -v options were combined for pcregrep, it printed a blank
169        line for every non-matching line. GNU grep prints nothing, and pcregrep now
170        does the same. The return code can be used to tell if there were any
171        non-matching lines.
173    14. Added --file-offsets and --line-offsets to pcregrep.
175    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
176        infinitely looping recursion. The bug was that positive lookaheads were not
177        being skipped when checking for a possible empty match (negative lookaheads
178        and both kinds of lookbehind were skipped).
180    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
181        inclusion of <windows.h> to before rather than after the definition of
182        INVALID_FILE_ATTRIBUTES (patch from David Byron).
184    17. Specifying a possessive quantifier with a specific limit for a Unicode
185        character property caused pcre_compile() to compile bad code, which led at
186        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
187        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
188        caused the error; without that there was no problem.
190    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
192    19. Added --enable-pcretest-libreadline.
194    20. In pcrecpp.cc, the variable 'count' was incremented twice in
195        RE::GlobalReplace(). As a result, the number of replacements returned was
196        double what it should be. I removed one of the increments, but Craig sent a
197        later patch that removed the other one (the right fix) and added unit tests
198        that check the return values (which was not done before).
200    21. Several CMake things:
202        (1) Arranged that, when cmake is used on Unix, the libraries end up with
203            the names libpcre and libpcreposix, not just pcre and pcreposix.
205        (2) The above change means that pcretest and pcregrep are now correctly
206            linked with the newly-built libraries, not previously installed ones.
210    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
211        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
212        UTF-8 newline character). The key issue is that the pattern starts .*;
213        this means that the match must be either at the beginning, or after a
214        newline. The bug was in the code for advancing after a failed match and
215        checking that the new position followed a newline. It was not taking
216        account of UTF-8 characters correctly.
218    23. PCRE was behaving differently from Perl in the way it recognized POSIX
219        character classes. PCRE was not treating the sequence [:...:] as a
220        character class unless the ... were all letters. Perl, however, seems to
221        allow any characters between [: and :], though of course it rejects as
222        unknown any "names" that contain non-letters, because all the known class
223        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
224        for example, whereas PCRE did not - it did not recognize a POSIX character
225        class. This seemed a bit dangerous, so the code has been changed to be
226        closer to Perl. The behaviour is not identical to Perl, because PCRE will
227        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
228        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
229        Perl does, and where it didn't before.
231    24. Rewrite so as to remove the single use of %n from pcregrep because in some
232        Windows environments %n is disabled by default.
235    Version 7.4 21-Sep-07
236    ---------------------
238    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
239        means that a class such as [\s] counted as "explicit reference to CR or
240        LF". That isn't really right - the whole point of the change was to try to
241        help when there was an actual mention of one of the two characters. So now
242        the change happens only if \r or \n (or a literal CR or LF) character is
243        encountered.
245    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
246        of both had grown to the point where there were only 3 bits left.
247        Fortunately, there was spare space in the data structure, and so I have
248        moved the internal flags into a new 16-bit field to free up more option
249        bits.
251    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
252        but did not set the internal JCHANGED flag - either of these is enough to
253        control the way the "get" function works - but the PCRE_INFO_JCHANGED
254        facility is supposed to tell if (?J) was ever used, so now (?J) at the
255        start sets both bits.
257    4.  Added options (at build time, compile time, exec time) to change \R from
258        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
260    5.  doc/pcresyntax.html was missing from the distribution.
262    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
263        compatibility, even though it is no longer used.
265    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
266        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
267        windows.h file is present (where different names are used). [This was
268        reversed later after testing - see 16 below.]
270    8.  Changed all #include <config.h> to #include "config.h". There were also
271        some further <pcre.h> cases that I changed to "pcre.h".
273    9.  When pcregrep was used with the --colour option, it missed the line ending
274        sequence off the lines that it output.
276    10. It was pointed out to me that arrays of string pointers cause lots of
277        relocations when a shared library is dynamically loaded. A technique of
278        using a single long string with a table of offsets can drastically reduce
279        these. I have refactored PCRE in four places to do this. The result is
280        dramatic:
282          Originally:                          290
283          After changing UCP table:            187
284          After changing error message table:   43
285          After changing table of "verbs"       36
286          After changing table of Posix names   22
288        Thanks to the folks working on Gregex for glib for this insight.
290    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
291        unicode-properties was also set.
293    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
295    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
296        checked only for CRLF.
298    14. Added casts to pcretest.c to avoid compiler warnings.
300    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
302    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
303        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
304        entirely. This removes changes made in 7 above.
306    17. The CMake files have been updated, and there is now more information about
307        building with CMake in the NON-UNIX-USE document.
310    Version 7.3 28-Aug-07
311    ---------------------
313     1. In the rejigging of the build system that eventually resulted in 7.1, the
314        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
315        brackets there is not right, since it causes compilers to look for an
316        installed pcre.h, not the version that is in the source that is being
317        compiled (which of course may be different). I have changed it back to:
319          #include "pcre.h"
321        I have a vague recollection that the change was concerned with compiling in
322        different directories, but in the new build system, that is taken care of
323        by the VPATH setting the Makefile.
325     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
326        when the subject happened to end in the byte 0x85 (e.g. if the last
327        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
328        characters but of course it shouldn't be taken as a newline when it is part
329        of another character. The bug was that, for an unlimited repeat of . in
330        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
331        characters when looking for a newline.
333     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
335     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
336        in debug output.
338     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
339        long printing in the pcrecpp unittest when running under MinGW.
341     6. ESC_K was left out of the EBCDIC table.
343     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
344        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
345        limit also applies to "virtual nesting" when a pattern is recursive, and in
346        this case 1000 isn't so big. I have been able to remove this limit at the
347        expense of backing off one optimization in certain circumstances. Normally,
348        when pcre_exec() would call its internal match() function recursively and
349        immediately return the result unconditionally, it uses a "tail recursion"
350        feature to save stack. However, when a subpattern that can match an empty
351        string has an unlimited repetition quantifier, it no longer makes this
352        optimization. That gives it a stack frame in which to save the data for
353        checking that an empty string has been matched. Previously this was taken
354        from the 1000-entry workspace that had been reserved. So now there is no
355        explicit limit, but more stack is used.
357     8. Applied Daniel's patches to solve problems with the import/export magic
358        syntax that is required for Windows, and which was going wrong for the
359        pcreposix and pcrecpp parts of the library. These were overlooked when this
360        problem was solved for the main library.
362     9. There were some crude static tests to avoid integer overflow when computing
363        the size of patterns that contain repeated groups with explicit upper
364        limits. As the maximum quantifier is 65535, the maximum group length was
365        set at 30,000 so that the product of these two numbers did not overflow a
366        32-bit integer. However, it turns out that people want to use groups that
367        are longer than 30,000 bytes (though not repeat them that many times).
368        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
369        made it possible to implement the integer overflow checks in a much more
370        dynamic way, which I have now done. The artificial limitation on group
371        length has been removed - we now have only the limit on the total length of
372        the compiled pattern, which depends on the LINK_SIZE setting.
374    10. Fixed a bug in the documentation for get/copy named substring when
375        duplicate names are permitted. If none of the named substrings are set, the
376        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
377        empty string.
379    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
380        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
381        because the ] is interpreted as the first data character and the
382        terminating ] is not found. PCRE has been made compatible with Perl in this
383        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
384        cause memory overwriting.
386    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
387        string has been matched (to stop an infinite loop). It was not recognizing
388        a conditional subpattern that could match an empty string if that
389        subpattern was within another subpattern. For example, it looped when
390        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
391        condition was not nested. This bug has been fixed.
393    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
394        past the start of the subject in the presence of bytes with the top bit
395        set, for example "\x8aBCD".
397    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
398        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
400    14. Optimized (?!) to (*FAIL).
402    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
403        This restricts code points to be within the range 0 to 0x10FFFF, excluding
404        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
405        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
406        does: it's just the validity check that is more restrictive.
408    16. Inserted checks for integer overflows during escape sequence (backslash)
409        processing, and also fixed erroneous offset values for syntax errors during
410        backslash processing.
412    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
413        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
415    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
416        caused an overrun.
418    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
419        something other than just ASCII characters) inside a group that had an
420        unlimited repeat caused a loop at compile time (while checking to see
421        whether the group could match an empty string).
423    20. Debugging a pattern containing \p or \P could cause a crash. For example,
424        [\P{Any}] did so. (Error in the code for printing property names.)
426    21. An orphan \E inside a character class could cause a crash.
428    22. A repeated capturing bracket such as (A)? could cause a wild memory
429        reference during compilation.
431    23. There are several functions in pcre_compile() that scan along a compiled
432        expression for various reasons (e.g. to see if it's fixed length for look
433        behind). There were bugs in these functions when a repeated \p or \P was
434        present in the pattern. These operators have additional parameters compared
435        with \d, etc, and these were not being taken into account when moving along
436        the compiled data. Specifically:
438        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
439            length.
441        (b) An item such as \pL+ within a repeated group could cause crashes or
442            loops.
444        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
445            "reference to non-existent subpattern" error.
447        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
449    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
450        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
452    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
454    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
455        character were causing crashes (broken optimization).
457    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
458        \p or \P) caused a compile-time loop.
460    28. More problems have arisen in unanchored patterns when CRLF is a valid line
461        break. For example, the unstudied pattern [\r\n]A does not match the string
462        "\r\nA" because change 7.0/46 below moves the current point on by two
463        characters after failing to match at the start. However, the pattern \nA
464        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
465        the same is true. There doesn't seem any very clean way out of this, but
466        what I have chosen to do makes the common cases work: PCRE now takes note
467        of whether there can be an explicit match for \r or \n anywhere in the
468        pattern, and if so, 7.0/46 no longer applies. As part of this change,
469        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
470        pattern has explicit CR or LF references.
472    29. Added (*CR) etc for changing newline setting at start of pattern.
475    Version 7.2 19-Jun-07
476    ---------------------
478     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
479        which is apparently normally available under Windows.
481     2. Re-jig the pcregrep tests with different newline settings in an attempt
482        to make them independent of the local environment's newline setting.
484     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
486     4. Some of the "internals" tests were previously cut out when the link size
487        was not 2, because the output contained actual offsets. The recent new
488        "Z" feature of pcretest means that these can be cut out, making the tests
489        usable with all link sizes.
491     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
492        stack recursion. This gives a massive performance boost under BSD, but just
493        a small improvement under Linux. However, it saves one field in the frame
494        in all cases.
496     6. Added more features from the forthcoming Perl 5.10:
498        (a) (?-n) (where n is a string of digits) is a relative subroutine or
499            recursion call. It refers to the nth most recently opened parentheses.
501        (b) (?+n) is also a relative subroutine call; it refers to the nth next
502            to be opened parentheses.
504        (c) Conditions that refer to capturing parentheses can be specified
505            relatively, for example, (?(-2)... or (?(+3)...
507        (d) \K resets the start of the current match so that everything before
508            is not part of it.
510        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
512        (f) \g{name} is another synonym - part of Perl 5.10's unification of
513            reference syntax.
515        (g) (?| introduces a group in which the numbering of parentheses in each
516            alternative starts with the same number.
518        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
520     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
523     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
524        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
525        for detecting groups that can match an empty string.
527     9. A pattern with a very large number of alternatives (more than several
528        hundred) was running out of internal workspace during the pre-compile
529        phase, where pcre_compile() figures out how much memory will be needed. A
530        bit of new cunning has reduced the workspace needed for groups with
531        alternatives. The 1000-alternative test pattern now uses 12 bytes of
532        workspace instead of running out of the 4096 that are available.
534    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
536    11. Applied patch from Google to remove an optimization that didn't quite work.
537        The report of the bug said:
539          pcrecpp::RE("a*").FullMatch("aaa") matches, while
540          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
541          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
543    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
544        it matched the wrong number of bytes.
547    Version 7.1 24-Apr-07
548    ---------------------
550     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
551        that is more "standard", making use of automake and other Autotools. There
552        is some re-arrangement of the files and adjustment of comments consequent
553        on this.
555     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
556        for recursive directory scanning broke on some systems because the files
557        are not scanned in any specific order and on different systems the order
558        was different. A call to "sort" has been inserted into RunGrepTest for the
559        approprate test as a short-term fix. In the longer term there may be an
560        alternative.
562     3. I had an email from Eric Raymond about problems translating some of PCRE's
563        man pages to HTML (despite the fact that I distribute HTML pages, some
564        people do their own conversions for various reasons). The problems
565        concerned the use of low-level troff macros .br and .in. I have therefore
566        removed all such uses from the man pages (some were redundant, some could
567        be replaced by .nf/.fi pairs). The 132html script that I use to generate
568        HTML has been updated to handle .nf/.fi and to complain if it encounters
569        .br or .in.
571     4. Updated comments in configure.ac that get placed in config.h.in and also
572        arranged for config.h to be included in the distribution, with the name
573        config.h.generic, for the benefit of those who have to compile without
574        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
576     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
577        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
578        makevp.bat for the current PCRE, using the additional files
579        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
581     6. A Windows user reported a minor discrepancy with test 2, which turned out
582        to be caused by a trailing space on an input line that had got lost in his
583        copy. The trailing space was an accident, so I've just removed it.
585     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
586        that is needed.
588     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
589        as "const" (a) because they are and (b) because it helps the PHP
590        maintainers who have recently made a script to detect big data structures
591        in the php code that should be moved to the .rodata section. I remembered
592        to update Builducptable as well, so it won't revert if ucptable.h is ever
593        re-created.
595     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
596        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
597        order to be able to cut out the UTF-8 tables in the latter when UTF-8
598        support is not required. This saves 1.5-2K of code, which is important in
599        some applications.
601        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
602        so as not to refer to the tables, even though these functions will never be
603        called when UTF-8 support is disabled. Otherwise there are problems with a
604        shared library.
606    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
608        (a) It was defining its arguments as char * instead of void *.
610        (b) It was assuming that all moves were upwards in memory; this was true
611            a long time ago when I wrote it, but is no longer the case.
613        The emulated memove() is provided for those environments that have neither
614        memmove() nor bcopy(). I didn't think anyone used it these days, but that
615        is clearly not the case, as these two bugs were recently reported.
617    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
618        and Detrail to create the HTML documentation, the .txt form of the man
619        pages, and it removes trailing spaces from listed files. It also creates
620        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
621        case, it wraps all the #defines with #ifndefs. This script should be run
622        before "make dist".
624    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
625        with Unicode property support.
627        (a) For a maximizing quantifier, if the two different cases of the
628            character were of different lengths in their UTF-8 codings (there are
629            some cases like this - I found 11), and the matching function had to
630            back up over a mixture of the two cases, it incorrectly assumed they
631            were both the same length.
633        (b) When PCRE was configured to use the heap rather than the stack for
634            recursion during matching, it was not correctly preserving the data for
635            the other case of a UTF-8 character when checking ahead for a match
636            while processing a minimizing repeat. If the check also involved
637            matching a wide character, but failed, corruption could cause an
638            erroneous result when trying to check for a repeat of the original
639            character.
641    13. Some tidying changes to the testing mechanism:
643        (a) The RunTest script now detects the internal link size and whether there
644            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
645            values substituted by "configure". (The RunGrepTest script already did
646            this for UTF-8.) The configure.ac script no longer substitutes the
647            relevant variables.
649        (b) The debugging options /B and /D in pcretest show the compiled bytecode
650            with length and offset values. This means that the output is different
651            for different internal link sizes. Test 2 is skipped for link sizes
652            other than 2 because of this, bypassing the problem. Unfortunately,
653            there was also a test in test 3 (the locale tests) that used /B and
654            failed for link sizes other than 2. Rather than cut the whole test out,
655            I have added a new /Z option to pcretest that replaces the length and
656            offset values with spaces. This is now used to make test 3 independent
657            of link size. (Test 2 will be tidied up later.)
659    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
660        segmentation fault instead of returning the appropriate error message.
662    15. In multiline mode when the newline sequence was set to "any", the pattern
663        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
664        This doesn't seem right; it now treats the CRLF combination as the line
665        ending, and so does not match in that case. It's only a pattern such as ^$
666        that would hit this one: something like ^ABC$ would have failed after \r
667        and then tried again after \r\n.
669    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
670        in an attempt to make files that differ only in their line terminators
671        compare equal. This works on Linux.
673    17. Under certain error circumstances pcregrep might try to free random memory
674        as it exited. This is now fixed, thanks to valgrind.
676    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
677        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
678        was because its rules for how to advance for /g after matching an empty
679        string at the end of a line did not allow for this case. They now check for
680        it specially.
682    20. pcretest is supposed to handle patterns and data of any length, by
683        extending its buffers when necessary. It was getting this wrong when the
684        buffer for a data line had to be extended.
686    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
687        CRLF as a newline sequence.
689    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
690        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
691        I have nevertheless tidied it up.
693    23. Added some casts to kill warnings from HP-UX ia64 compiler.
695    24. Added a man page for pcre-config.
698    Version 7.0 19-Dec-06
699    ---------------------
701     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
702        moving to gcc 4.1.1.
704     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
705        sys/time.h, which is documented as needed for this function. It doesn't
706        seem to matter on Linux, but it showed up on some releases of OS X.
708     3. It seems that there are systems where bytes whose values are greater than
709        127 match isprint() in the "C" locale. The "C" locale should be the
710        default when a C program starts up. In most systems, only ASCII printing
711        characters match isprint(). This difference caused the output from pcretest
712        to vary, making some of the tests fail. I have changed pcretest so that:
714        (a) When it is outputting text in the compiled version of a pattern, bytes
715            other than 32-126 are always shown as hex escapes.
717        (b) When it is outputting text that is a matched part of a subject string,
718            it does the same, unless a different locale has been set for the match
719            (using the /L modifier). In this case, it uses isprint() to decide.
721     4. Fixed a major bug that caused incorrect computation of the amount of memory
722        required for a compiled pattern when options that changed within the
723        pattern affected the logic of the preliminary scan that determines the
724        length. The relevant options are -x, and -i in UTF-8 mode. The result was
725        that the computed length was too small. The symptoms of this bug were
726        either the PCRE error "internal error: code overflow" from pcre_compile(),
727        or a glibc crash with a message such as "pcretest: free(): invalid next
728        size (fast)". Examples of patterns that provoked this bug (shown in
729        pcretest format) are:
731          /(?-x: )/x
732          /(?x)(?-x: \s*#\s*)/
733          /((?i)[\x{c0}])/8
734          /(?i:[\x{c0}])/8
736        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
737        is now done differently.
739     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
740        wrapper classes; (b) implement a new function in the C++ scanner that is
741        more efficient than the old way of doing things because it avoids levels of
742        recursion in the regex matching; (c) add a paragraph to the documentation
743        for the FullMatch() function.
745     6. The escape sequence \n was being treated as whatever was defined as
746        "newline". Not only was this contrary to the documentation, which states
747        that \n is character 10 (hex 0A), but it also went horribly wrong when
748        "newline" was defined as CRLF. This has been fixed.
750     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
751        was being set to -1 for the "end of line" case (supposedly a value that no
752        character can have). Though this value is never used (the check for end of
753        line is "zero bytes in current character"), it caused compiler complaints.
754        I've changed it to 0xffffffff.
756     8. In pcre_version.c, the version string was being built by a sequence of
757        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
758        string (as it is for production releases) called a macro with an empty
759        argument. The C standard says the result of this is undefined. The gcc
760        compiler treats it as an empty string (which was what was wanted) but it is
761        reported that Visual C gives an error. The source has been hacked around to
762        avoid this problem.
764     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
765        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
766        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
767        of them did).
769    10. Originally, pcretest opened its input and output without "b"; then I was
770        told that "b" was needed in some environments, so it was added for release
771        5.0 to both the input and output. (It makes no difference on Unix-like
772        systems.) Later I was told that it is wrong for the input on Windows. I've
773        now abstracted the modes into two macros, to make it easier to fiddle with
774        them, and removed "b" from the input mode under Windows.
776    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
778    12. Added -help and --help to pcretest as an official way of being reminded
779        of the options.
781    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
782        and pcrecpp.cc because they annoy compilers at high warning levels.
784    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
786    15. Fixed an occurrence of == in configure.ac that should have been = (shell
787        scripts are not C programs :-) and which was not noticed because it works
788        on Linux.
790    16. pcretest is supposed to handle any length of pattern and data line (as one
791        line or as a continued sequence of lines) by extending its input buffer if
792        necessary. This feature was broken for very long pattern lines, leading to
793        a string of junk being passed to pcre_compile() if the pattern was longer
794        than about 50K.
796    17. I have done a major re-factoring of the way pcre_compile() computes the
797        amount of memory needed for a compiled pattern. Previously, there was code
798        that made a preliminary scan of the pattern in order to do this. That was
799        OK when PCRE was new, but as the facilities have expanded, it has become
800        harder and harder to keep it in step with the real compile phase, and there
801        have been a number of bugs (see for example, 4 above). I have now found a
802        cunning way of running the real compile function in a "fake" mode that
803        enables it to compute how much memory it would need, while actually only
804        ever using a few hundred bytes of working memory and without too many
805        tests of the mode. This should make future maintenance and development
806        easier. A side effect of this work is that the limit of 200 on the nesting
807        depth of parentheses has been removed (though this was never a serious
808        limitation, I suspect). However, there is a downside: pcre_compile() now
809        runs more slowly than before (30% or more, depending on the pattern). I
810        hope this isn't a big issue. There is no effect on runtime performance.
812    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
813        newline (only possible for the last line of a file) and it was a
814        pattern that set a locale (followed by /Lsomething), pcretest crashed.
816    19. Added additional timing features to pcretest. (1) The -tm option now times
817        matching only, not compiling. (2) Both -t and -tm can be followed, as a
818        separate command line item, by a number that specifies the number of
819        repeats to use when timing. The default is 50000; this gives better
820        precision, but takes uncomfortably long for very large patterns.
822    20. Extended pcre_study() to be more clever in cases where a branch of a
823        subpattern has no definite first character. For example, (a*|b*)[cd] would
824        previously give no result from pcre_study(). Now it recognizes that the
825        first character must be a, b, c, or d.
827    21. There was an incorrect error "recursive call could loop indefinitely" if
828        a subpattern (or the entire pattern) that was being tested for matching an
829        empty string contained only one non-empty item after a nested subpattern.
830        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
831        incorrectly, because the \d was being skipped in the check.
833    22. The pcretest program now has a new pattern option /B and a command line
834        option -b, which is equivalent to adding /B to every pattern. This causes
835        it to show the compiled bytecode, without the additional information that
836        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
837        is the same as /B/I).
839    23. A new optimization is now able automatically to treat some sequences such
840        as a*b as a*+b. More specifically, if something simple (such as a character
841        or a simple class like \d) has an unlimited quantifier, and is followed by
842        something that cannot possibly match the quantified thing, the quantifier
843        is automatically "possessified".
845    24. A recursive reference to a subpattern whose number was greater than 39
846        went wrong under certain circumstances in UTF-8 mode. This bug could also
847        have affected the operation of pcre_study().
849    25. Realized that a little bit of performance could be had by replacing
850        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
852    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
854    27. Possessive quantifiers such as a++ were previously implemented by turning
855        them into atomic groups such as ($>a+). Now they have their own opcodes,
856        which improves performance. This includes the automatically created ones
857        from 23 above.
859    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
860        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
861        the first matched character to be a colon. This applied both to named and
862        numbered groups.
864    29. The ucpinternal.h header file was missing its idempotency #ifdef.
866    30. I was sent a "project" file called libpcre.a.dev which I understand makes
867        building PCRE on Windows easier, so I have included it in the distribution.
869    31. There is now a check in pcretest against a ridiculously large number being
870        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
871        loop, the loop is abandoned.
873    32. Forward references to subpatterns in conditions such as (?(2)...) where
874        subpattern 2 is defined later cause pcre_compile() to search forwards in
875        the pattern for the relevant set of parentheses. This search went wrong
876        when there were unescaped parentheses in a character class, parentheses
877        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
879    33. "Subroutine" calls and backreferences were previously restricted to
880        referencing subpatterns earlier in the regex. This restriction has now
881        been removed.
883    34. Added a number of extra features that are going to be in Perl 5.10. On the
884        whole, these are just syntactic alternatives for features that PCRE had
885        previously implemented using the Python syntax or my own invention. The
886        other formats are all retained for compatibility.
888        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
889            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
890            also .NET compatible.
892        (b) A recursion or subroutine call to a named group can now be defined as
893            (?&name) as well as (?P>name).
895        (c) A backreference to a named group can now be defined as \k<name> or
896            \k'name' as well as (?P=name). The new forms, as well as being in Perl
897            5.10, are also .NET compatible.
899        (d) A conditional reference to a named group can now use the syntax
900            (?(<name>) or (?('name') as well as (?(name).
902        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
903            groups (named and numbered) that are never evaluated inline, but can be
904            called as "subroutines" from elsewhere. In effect, the DEFINE condition
905            is always false. There may be only one alternative in such a group.
907        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
908            as the simple (?(R). The condition is true only if the most recent
909            recursion is that of the given number or name. It does not search out
910            through the entire recursion stack.
912        (g) The escape \gN or \g{N} has been added, where N is a positive or
913            negative number, specifying an absolute or relative reference.
915    35. Tidied to get rid of some further signed/unsigned compiler warnings and
916        some "unreachable code" warnings.
918    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
919        things, this adds five new scripts.
921    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
922        There were also incompatibilities regarding the handling of \Q..\E inside
923        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
924        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
926    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
927        matches an empty string, and forcibly breaks the loop. There were bugs in
928        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
929        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
930        separate and independent bugs (that affected different cases) have been
931        fixed.
933    39. Refactored the code to abolish the use of different opcodes for small
934        capturing bracket numbers. This is a tidy that I avoided doing when I
935        removed the limit on the number of capturing brackets for 3.5 back in 2001.
936        The new approach is not only tidier, it makes it possible to reduce the
937        memory needed to fix the previous bug (38).
939    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
940        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
941        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
942        mode.
944    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
945        report.
947    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
948        copy construction and assignment in the C++ wrapper.
950    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
951        couple of bugs that could have given wrong results in the "--newline=crlf"
952        case.
954    44. Added a number of casts and did some reorganization of signed/unsigned int
955        variables following suggestions from Dair Grant. Also renamed the variable
956        "this" as "item" because it is a C++ keyword.
958    45. Arranged for dftables to add
960          #include "pcre_internal.h"
962        to pcre_chartables.c because without it, gcc 4.x may remove the array
963        definition from the final binary if PCRE is built into a static library and
964        dead code stripping is activated.
966    46. For an unanchored pattern, if a match attempt fails at the start of a
967        newline sequence, and the newline setting is CRLF or ANY, and the next two
968        characters are CRLF, advance by two characters instead of one.
971    Version 6.7 04-Jul-06
972    ---------------------
974     1. In order to handle tests when input lines are enormously long, pcretest has
975        been re-factored so that it automatically extends its buffers when
976        necessary. The code is crude, but this _is_ just a test program. The
977        default size has been increased from 32K to 50K.
979     2. The code in pcre_study() was using the value of the re argument before
980        testing it for NULL. (Of course, in any sensible call of the function, it
981        won't be NULL.)
983     3. The memmove() emulation function in pcre_internal.h, which is used on
984        systems that lack both memmove() and bcopy() - that is, hardly ever -
985        was missing a "static" storage class specifier.
987     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
988        containing an extended class (one that cannot be represented by a bitmap
989        because it contains high-valued characters or Unicode property items, e.g.
990        [\pZ]). Almost always one would set UTF-8 mode when processing such a
991        pattern, but PCRE should not loop if you do not (it no longer does).
992        [Detail: two cases were found: (a) a repeated subpattern containing an
993        extended class; (b) a recursive reference to a subpattern that followed a
994        previous extended class. It wasn't skipping over the extended class
995        correctly when UTF-8 mode was not set.]
997     5. A negated single-character class was not being recognized as fixed-length
998        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
999        compile error "lookbehind assertion is not fixed length".
1001     6. The RunPerlTest auxiliary script was showing an unexpected difference
1002        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1003        write a Perl script that can interpret lines of an input file either as
1004        byte characters or as UTF-8, which is what "perltest" was being required to
1005        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1006        can't do is switch easily at run time between having the "use utf8;" pragma
1007        or not. In the end, I fudged it by using the RunPerlTest script to insert
1008        "use utf8;" explicitly for the UTF-8 tests.
1010     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1011        the end of the subject string, contrary to the documentation and to what
1012        Perl does. This was true of both matching functions. Now it matches only at
1013        the start of the subject and immediately after *internal* newlines.
1015     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1016        a pointer to an int instead of a pointer to an unsigned long int. This
1017        caused problems on 64-bit systems.
1019     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1020        instance of the 'standard' template library not being so standard".
1022    10. There was no check on the number of named subpatterns nor the maximum
1023        length of a subpattern name. The product of these values is used to compute
1024        the size of the memory block for a compiled pattern. By supplying a very
1025        long subpattern name and a large number of named subpatterns, the size
1026        computation could be caused to overflow. This is now prevented by limiting
1027        the length of names to 32 characters, and the number of named subpatterns
1028        to 10,000.
1030    11. Subpatterns that are repeated with specific counts have to be replicated in
1031        the compiled pattern. The size of memory for this was computed from the
1032        length of the subpattern and the repeat count. The latter is limited to
1033        65535, but there was no limit on the former, meaning that integer overflow
1034        could in principle occur. The compiled length of a repeated subpattern is
1035        now limited to 30,000 bytes in order to prevent this.
1037    12. Added the optional facility to have named substrings with the same name.
1039    13. Added the ability to use a named substring as a condition, using the
1040        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1041        are numbers (not recommended). Forward references are permitted.
1043    14. Added forward references in named backreferences (if you see what I mean).
1045    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1046        pattern could run off the end of the subject. For example, the pattern
1047        "(?s)(.{1,5})"8 did this with the subject "ab".
1049    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1050        PCRE_CASELESS was set when matching characters that were quantified with ?
1051        or *.
1053    17. A character class other than a single negated character that had a minimum
1054        but no maximum quantifier - for example [ab]{6,} - was not handled
1055        correctly by pce_dfa_exec(). It would match only one character.
1057    18. A valid (though odd) pattern that looked like a POSIX character
1058        class but used an invalid character after [ (for example [[,abc,]]) caused
1059        pcre_compile() to give the error "Failed: internal error: code overflow" or
1060        in some cases to crash with a glibc free() error. This could even happen if
1061        the pattern terminated after [[ but there just happened to be a sequence of
1062        letters, a binary zero, and a closing ] in the memory that followed.
1064    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1065        over the years. Originally (before any Unicode support), just the bottom 8
1066        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1067        output from "man perlunicode" includes this:
1069          The regular expression compiler produces polymorphic opcodes.  That
1070          is, the pattern adapts to the data and automatically switches to
1071          the Unicode character scheme when presented with Unicode data--or
1072          instead uses a traditional byte scheme when presented with byte
1073          data.
1075        Sadly, a wide octal escape does not cause a switch, and in a string with
1076        no other multibyte characters, these octal escapes are treated as before.
1077        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1078        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1079        Unicode string.
1081        I have not perpetrated such confusion in PCRE. Up till now, it took just
1082        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1083        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1084        translate to the appropriate multibyte character.
1086    29. Applied some refactoring to reduce the number of warnings from Microsoft
1087        and Borland compilers. This has included removing the fudge introduced
1088        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1089        a warning about an unused variable.
1091    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1092        characters since release 4.0, because Perl (from release 5.004) does not.
1093        [Or at least, is documented not to: some releases seem to be in conflict
1094        with the documentation.] However, when a pattern was studied with
1095        pcre_study() and all its branches started with \s, PCRE still included VT
1096        as a possible starting character. Of course, this did no harm; it just
1097        caused an unnecessary match attempt.
1099    22. Removed a now-redundant internal flag bit that recorded the fact that case
1100        dependency changed within the pattern. This was once needed for "required
1101        byte" processing, but is no longer used. This recovers a now-scarce options
1102        bit. Also moved the least significant internal flag bit to the most-
1103        significant bit of the word, which was not previously used (hangover from
1104        the days when it was an int rather than a uint) to free up another bit for
1105        the future.
1107    23. Added support for CRLF line endings as well as CR and LF. As well as the
1108        default being selectable at build time, it can now be changed at runtime
1109        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1110        specify that it is scanning data with non-default line endings.
1112    24. Changed the definition of CXXLINK to make it agree with the definition of
1113        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1115    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1116        recursions. This makes a big different to stack usage for some patterns.
1118    26. If a subpattern containing a named recursion or subroutine reference such
1119        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1120        the space required for the compiled pattern went wrong and gave too small a
1121        value. Depending on the environment, this could lead to "Failed: internal
1122        error: code overflow at offset 49" or "glibc detected double free or
1123        corruption" errors.
1125    27. Applied patches from Google (a) to support the new newline modes and (b) to
1126        advance over multibyte UTF-8 characters in GlobalReplace.
1128    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1129        difference for some implementation of PCRE in some Windows version.
1131    29. Added some extra testing facilities to pcretest:
1133        \q<number>   in a data line sets the "match limit" value
1134        \Q<number>   in a data line sets the "match recursion limt" value
1135        -S <number>  sets the stack size, where <number> is in megabytes
1137        The -S option isn't available for Windows.
1140    Version 6.6 06-Feb-06
1141    ---------------------
1143     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1144        in pcreposix.h. I have copied the definition from pcre.h.
1146     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1147        because pcre.h is no longer a built file.
1149     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1150        not normally included in the compiled code.
1153    Version 6.5 01-Feb-06
1154    ---------------------
1156     1. When using the partial match feature with pcre_dfa_exec(), it was not
1157        anchoring the second and subsequent partial matches at the new starting
1158        point. This could lead to incorrect results. For example, with the pattern
1159        /1234/, partially matching against "123" and then "a4" gave a match.
1161     2. Changes to pcregrep:
1163        (a) All non-match returns from pcre_exec() were being treated as failures
1164            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1165            error message is output. Some extra information is given for the
1166            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1167            probably the only errors that are likely to be caused by users (by
1168            specifying a regex that has nested indefinite repeats, for instance).
1169            If there are more than 20 of these errors, pcregrep is abandoned.
1171        (b) A binary zero was treated as data while matching, but terminated the
1172            output line if it was written out. This has been fixed: binary zeroes
1173            are now no different to any other data bytes.
1175        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1176            used to set a locale for matching. The --locale=xxxx long option has
1177            been added (no short equivalent) to specify a locale explicitly on the
1178            pcregrep command, overriding the environment variables.
1180        (d) When -B was used with -n, some line numbers in the output were one less
1181            than they should have been.
1183        (e) Added the -o (--only-matching) option.
1185        (f) If -A or -C was used with -c (count only), some lines of context were
1186            accidentally printed for the final match.
1188        (g) Added the -H (--with-filename) option.
1190        (h) The combination of options -rh failed to suppress file names for files
1191            that were found from directory arguments.
1193        (i) Added the -D (--devices) and -d (--directories) options.
1195        (j) Added the -F (--fixed-strings) option.
1197        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1199        (l) Added the --colo(u)r option.
1201        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1202            is not present by default.
1204     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1205        items such as (?R) or (?1), when the recursion could match a number of
1206        alternatives. If it matched one of the alternatives, but subsequently,
1207        outside the recursion, there was a failure, the code tried to back up into
1208        the recursion. However, because of the way PCRE is implemented, this is not
1209        possible, and the result was an incorrect result from the match.
1211        In order to prevent this happening, the specification of recursion has
1212        been changed so that all such subpatterns are automatically treated as
1213        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1215     4. I had overlooked the fact that, in some locales, there are characters for
1216        which isalpha() is true but neither isupper() nor islower() are true. In
1217        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1218        and ordfeminine) are like this. This affected the treatment of \w and \W
1219        when they appeared in character classes, but not when they appeared outside
1220        a character class. The bit map for "word" characters is now created
1221        separately from the results of isalnum() instead of just taking it from the
1222        upper, lower, and digit maps. (Plus the underscore character, of course.)
1224     5. The above bug also affected the handling of POSIX character classes such as
1225        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1226        permanent tables. Instead, the bit maps for such a class were previously
1227        created as the appropriate unions of the upper, lower, and digit bitmaps.
1228        Now they are created by subtraction from the [[:word:]] class, which has
1229        its own bitmap.
1231     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1232        It is created by subtracting the vertical space characters (\x09, \x0a,
1233        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1234        subtraction was done in the overall bitmap for a character class, meaning
1235        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1236        be recognized. This bug has been fixed.
1238     7. Patches from the folks at Google:
1240          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1241          real life, but is still worth protecting against".
1243          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1244          regular expressions".
1246          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1247          have it.
1249          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1250          "configure" and the latter not, in order to fix a problem somebody had
1251          with compiling the Arg class on HP-UX.
1253          (e) Improve the error-handling of the C++ wrapper a little bit.
1255          (f) New tests for checking recursion limiting.
1257     8. The pcre_memmove() function, which is used only if the environment does not
1258        have a standard memmove() function (and is therefore rarely compiled),
1259        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1260        returning a result (though PCRE never actually uses the result).
1262     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1263        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1264        returned instead of calling malloc() with an overflowing number that would
1265        most likely cause subsequent chaos.
1267    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1269    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1270        with this option is matched, the nmatch and pmatch options of regexec() are
1271        ignored.
1273    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1274        provided in case anyone wants to the the POSIX interface with UTF-8
1275        strings.
1277    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1278        C++ linking (needed for some HP-UX environments).
1280    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1281        (unused parameter) and in the pcre_printint() function (omitted "default"
1282        switch label when the default is to do nothing).
1284    15. Added some code to make it possible, when PCRE is compiled as a C++
1285        library, to replace subject pointers for pcre_exec() with a smart pointer
1286        class, thus making it possible to process discontinuous strings.
1288    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1289        much the same function. They were added by different people who were trying
1290        to make PCRE easy to compile on non-Unix systems. It has been suggested
1291        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1292        for compiling on Windows systems. I have therefore replaced it with
1293        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1294        defaults to "extern" for C or "extern C" for C++, which works fine on
1295        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1296        SCOPE with something explicit in config.h. In addition:
1298        (a) pcreposix.h still had just "extern" instead of either of these macros;
1299            I have replaced it with PCRE_DATA_SCOPE.
1301        (b) Functions such as _pcre_xclass(), which are internal to the library,
1302            but external in the C sense, all had PCRE_EXPORT in their definitions.
1303            This is apparently wrong for the Windows case, so I have removed it.
1304            (It makes no difference on Unix-like systems.)
1306    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1307        of recursive calls to match(). This is different to MATCH_LIMIT because
1308        that limits the total number of calls to match(), not all of which increase
1309        the depth of recursion. Limiting the recursion depth limits the amount of
1310        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1311        when PCRE is compiled, and changed at run time. A patch from Google adds
1312        this functionality to the C++ interface.
1314    18. Changes to the handling of Unicode character properties:
1316        (a) Updated the table to Unicode 4.1.0.
1318        (b) Recognize characters that are not in the table as "Cn" (undefined).
1320        (c) I revised the way the table is implemented to a much improved format
1321            which includes recognition of ranges. It now supports the ranges that
1322            are defined in UnicodeData.txt, and it also amalgamates other
1323            characters into ranges. This has reduced the number of entries in the
1324            table from around 16,000 to around 3,000, thus reducing its size
1325            considerably. I realized I did not need to use a tree structure after
1326            all - a binary chop search is just as efficient. Having reduced the
1327            number of entries, I extended their size from 6 bytes to 8 bytes to
1328            allow for more data.
1330        (d) Added support for Unicode script names via properties such as \p{Han}.
1332    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1333        matching that character.
1335    20. When matching a repeated Unicode property with a minimum greater than zero,
1336        (for example \pL{2,}), PCRE could look past the end of the subject if it
1337        reached it while seeking the minimum number of characters. This could
1338        happen only if some of the characters were more than one byte long, because
1339        there is a check for at least the minimum number of bytes.
1341    21. Refactored the implementation of \p and \P so as to be more general, to
1342        allow for more different types of property in future. This has changed the
1343        compiled form incompatibly. Anybody with saved compiled patterns that use
1344        \p or \P will have to recompile them.
1346    22. Added "Any" and "L&" to the supported property types.
1348    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1349        but give a compile time error if the value is greater than 0xff.
1351    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1352        accidentally not being installed or uninstalled.
1354    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1355        made were to insert the current release number. This seemed silly, because
1356        it made things harder for people building PCRE on systems that don't run
1357        "configure". I have turned pcre.h into a distributed file, no longer built
1358        by "configure", with the version identification directly included. There is
1359        no longer a pcre.h.in file.
1361        However, this change necessitated a change to the pcre-config script as
1362        well. It is built from pcre-config.in, and one of the substitutions was the
1363        release number. I have updated configure.ac so that ./configure now finds
1364        the release number by grepping pcre.h.
1366    26. Added the ability to run the tests under valgrind.
1369    Version 6.4 05-Sep-05
1370    ---------------------
1372     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1373        "--" to be printed when multiple files were scanned, even when none of the
1374        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1375        consider it to be a bug, and have restored the previous behaviour.
1377     2. A couple of code tidies to get rid of compiler warnings.
1379     3. The pcretest program used to cheat by referring to symbols in the library
1380        whose names begin with _pcre_. These are internal symbols that are not
1381        really supposed to be visible externally, and in some environments it is
1382        possible to suppress them. The cheating is now confined to including
1383        certain files from the library's source, which is a bit cleaner.
1385     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1386        file's purpose clearer.
1388     5. Reorganized pcre_ucp_findchar().
1391    Version 6.3 15-Aug-05
1392    ---------------------
1394     1. The file libpcre.pc.in did not have general read permission in the tarball.
1396     2. There were some problems when building without C++ support:
1398        (a) If C++ support was not built, "make install" and "make test" still
1399            tried to test it.
1401        (b) There were problems when the value of CXX was explicitly set. Some
1402            changes have been made to try to fix these, and ...
1404        (c) --disable-cpp can now be used to explicitly disable C++ support.
1406        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1407            backslash in a target when C++ was disabled. This confuses some
1408            versions of "make", apparently. Using an intermediate variable solves
1409            this. (Same for CPP_LOBJ.)
1411     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1412        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1413        necessary on certain architectures.
1415     4. Added a setting of -export-symbols-regex to the link command to remove
1416        those symbols that are exported in the C sense, but actually are local
1417        within the library, and not documented. Their names all begin with
1418        "_pcre_". This is not a perfect job, because (a) we have to except some
1419        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1420        available (and never for static libraries). I have made a note to try to
1421        find a way round (a) in the future.
1424    Version 6.2 01-Aug-05
1425    ---------------------
1427     1. There was no test for integer overflow of quantifier values. A construction
1428        such as {1111111111111111} would give undefined results. What is worse, if
1429        a minimum quantifier for a parenthesized subpattern overflowed and became
1430        negative, the calculation of the memory size went wrong. This could have
1431        led to memory overwriting.
1433     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1435     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1436        operating environments where this matters.
1438     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1439        PCRE options from within the C++ wrapper.
1441     5. Named capturing subpatterns were not being correctly counted when a pattern
1442        was compiled. This caused two problems: (a) If there were more than 100
1443        such subpatterns, the calculation of the memory needed for the whole
1444        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1445        back references of the form \12, where the number was greater than 9, were
1446        not recognized as back references, even though there were sufficient
1447        previous subpatterns.
1449     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1450        versions of gcc, e.g. 2.95.4.
1453    Version 6.1 21-Jun-05
1454    ---------------------
1456     1. There was one reference to the variable "posix" in pcretest.c that was not
1457        surrounded by "#if !defined NOPOSIX".
1459     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1460        the cross-check on the old pcre_info() function, for the benefit of the
1461        cut-down version of PCRE that is currently imported into Exim.
1463     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1464        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1465        patterns that start (?i) but I don't think that matters. The use of (?i) is
1466        just an example; this all applies to the other options as well.
1468     4. Since libtool seems to echo the compile commands it is issuing, the output
1469        from "make" can be reduced a bit by putting "@" in front of each libtool
1470        compile command.
1472     5. Patch from the folks at Google for configure.in to be a bit more thorough
1473        in checking for a suitable C++ installation before trying to compile the
1474        C++ stuff. This should fix a reported problem when a compiler was present,
1475        but no suitable headers.
1477     6. The man pages all had just "PCRE" as their title. I have changed them to
1478        be the relevant file name. I have also arranged that these names are
1479        retained in the file doc/pcre.txt, which is a concatenation in text format
1480        of all the man pages except the little individual ones for each function.
1482     7. The NON-UNIX-USE file had not been updated for the different set of source
1483        files that come with release 6. I also added a few comments about the C++
1484        wrapper.
1487    Version 6.0 07-Jun-05
1488    ---------------------
1490     1. Some minor internal re-organization to help with my DFA experiments.
1492     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1493        didn't matter for the library itself when fully configured, but did matter
1494        when compiling without UCP support, or within Exim, where the ucp files are
1495        not imported.
1497     3. Refactoring of the library code to split up the various functions into
1498        different source modules. The addition of the new DFA matching code (see
1499        below) to a single monolithic source would have made it really too
1500        unwieldy, quite apart from causing all the code to be include in a
1501        statically linked application, when only some functions are used. This is
1502        relevant even without the DFA addition now that patterns can be compiled in
1503        one application and matched in another.
1505        The downside of splitting up is that there have to be some external
1506        functions and data tables that are used internally in different modules of
1507        the library but which are not part of the API. These have all had their
1508        names changed to start with "_pcre_" so that they are unlikely to clash
1509        with other external names.
1511     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1512        a different (DFA) algorithm. Although it is slower than the original
1513        function, it does have some advantages for certain types of matching
1514        problem.
1516     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1517        including restarting after a partial match.
1519     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1520        defined when compiling for Windows was sent to me. I have put it into the
1521        code, though I have no means of testing or verifying it.
1523     7. Added the pcre_refcount() auxiliary function.
1525     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1526        match before or at the first newline in the subject string. In pcretest,
1527        the /f option on a pattern can be used to set this.
1529     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1530        would behave wrongly. This has been present in PCRE since release 4.0.
1532    10. A number of changes to the pcregrep command:
1534        (a) Refactored how -x works; insert ^(...)$ instead of setting
1535            PCRE_ANCHORED and checking the length, in preparation for adding
1536            something similar for -w.
1538        (b) Added the -w (match as a word) option.
1540        (c) Refactored the way lines are read and buffered so as to have more
1541            than one at a time available.
1543        (d) Implemented a pcregrep test script.
1545        (e) Added the -M (multiline match) option. This allows patterns to match
1546            over several lines of the subject. The buffering ensures that at least
1547            8K, or the rest of the document (whichever is the shorter) is available
1548            for matching (and similarly the previous 8K for lookbehind assertions).
1550        (f) Changed the --help output so that it now says
1552              -w, --word-regex(p)
1554            instead of two lines, one with "regex" and the other with "regexp"
1555            because that confused at least one person since the short forms are the
1556            same. (This required a bit of code, as the output is generated
1557            automatically from a table. It wasn't just a text change.)
1559        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1560            option but starts with a hyphen. Could be a pattern or a path name
1561            starting with a hyphen, for instance.
1563        (h) "-" can be given as a file name to represent stdin.
1565        (i) When file names are being printed, "(standard input)" is used for
1566            the standard input, for compatibility with GNU grep. Previously
1567            "<stdin>" was used.
1569        (j) The option --label=xxx can be used to supply a name to be used for
1570            stdin when file names are being printed. There is no short form.
1572        (k) Re-factored the options decoding logic because we are going to add
1573            two more options that take data. Such options can now be given in four
1574            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1576        (l) Added the -A, -B, and -C options for requesting that lines of context
1577            around matches be printed.
1579        (m) Added the -L option to print the names of files that do not contain
1580            any matching lines, that is, the complement of -l.
1582        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1583            continue to scan other files.
1585        (o) The -s option was incorrectly implemented. For compatibility with other
1586            greps, it now suppresses the error message for a non-existent or non-
1587            accessible file (but not the return code). There is a new option called
1588            -q that suppresses the output of matching lines, which was what -s was
1589            previously doing.
1591        (p) Added --include and --exclude options to specify files for inclusion
1592            and exclusion when recursing.
1594    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1595        Hopefully, it now does.
1597    12. Missing cast in pcre_study().
1599    13. Added an "uninstall" target to the makefile.
1601    14. Replaced "extern" in the function prototypes in Makefile.in with
1602        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1603        world, but is set differently for Windows.
1605    15. Added a second compiling function called pcre_compile2(). The only
1606        difference is that it has an extra argument, which is a pointer to an
1607        integer error code. When there is a compile-time failure, this is set
1608        non-zero, in addition to the error test pointer being set to point to an
1609        error message. The new argument may be NULL if no error number is required
1610        (but then you may as well call pcre_compile(), which is now just a
1611        wrapper). This facility is provided because some applications need a
1612        numeric error indication, but it has also enabled me to tidy up the way
1613        compile-time errors are handled in the POSIX wrapper.
1615    16. Added VPATH=.libs to the makefile; this should help when building with one
1616        prefix path and installing with another. (Or so I'm told by someone who
1617        knows more about this stuff than I do.)
1619    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1620        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1621        match everything, including newlines. This is not POSIX-compatible, but
1622        somebody wanted the feature. From pcretest it can be activated by using
1623        both the P and the s flags.
1625    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1627    19. libpcre.pc was being incorrectly installed as executable.
1629    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1630        it now also looks for '\r' so that it will work unmodified on Windows.
1632    21. Added Google's contributed C++ wrapper to the distribution.
1634    22. Added some untidy missing memory free() calls in pcretest, to keep
1635        Electric Fence happy when testing.
1639    Version 5.0 13-Sep-04
1640    ---------------------
1642     1. Internal change: literal characters are no longer packed up into items
1643        containing multiple characters in a single byte-string. Each character
1644        is now matched using a separate opcode. However, there may be more than one
1645        byte in the character in UTF-8 mode.
1647     2. The pcre_callout_block structure has two new fields: pattern_position and
1648        next_item_length. These contain the offset in the pattern to the next match
1649        item, and its length, respectively.
1651     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1652        insertion of callouts before each pattern item. Added the /C option to
1653        pcretest to make use of this.
1655     4. On the advice of a Windows user, the lines
1657          #if defined(_WIN32) || defined(WIN32)
1658          _setmode( _fileno( stdout ), 0x8000 );
1659          #endif  /* defined(_WIN32) || defined(WIN32) */
1661        have been added to the source of pcretest. This apparently does useful
1662        magic in relation to line terminators.
1664     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1665        for the benefit of those environments where the "b" makes a difference.
1667     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1668        to know about it. I have put a hack into configure.in that adds in code
1669        to set GCC=yes if CC=icc. This seems to end up at a point in the
1670        generated configure script that is early enough to affect the setting of
1671        compiler options, which is what is needed, but I have no means of testing
1672        whether it really works. (The user who reported this had patched the
1673        generated configure script, which of course I cannot do.)
1675        LATER: After change 22 below (new libtool files), the configure script
1676        seems to know about icc (and also ecc). Therefore, I have commented out
1677        this hack in configure.in.
1679     7. Added support for pkg-config (2 patches were sent in).
1681     8. Negated POSIX character classes that used a combination of internal tables
1682        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1683        [[:^ascii]]. Typically, they would match almost any characters. The other
1684        POSIX classes were not broken in this way.
1686     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1687        to find the match, as PCRE was deluded into thinking that the match had to
1688        start at the start point or following a newline. The same bug applied to
1689        patterns with negative forward assertions or any backward assertions
1690        preceding ".*" at the start, unless the pattern required a fixed first
1691        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1693    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1694        starting at the last subject character, bytes beyond the end of the subject
1695        string were read.
1697    11. Renamed the variable "class" as "classbits" to make life easier for C++
1698        users. (Previously there was a macro definition, but it apparently wasn't
1699        enough.)
1701    12. Added the new field "tables" to the extra data so that tables can be passed
1702        in at exec time, or the internal tables can be re-selected. This allows
1703        a compiled regex to be saved and re-used at a later time by a different
1704        program that might have everything at different addresses.
1706    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1707        -R library as well as a -L library.
1709    14. The debugging options of pcretest (-d on the command line or D on a
1710        pattern) showed incorrect output for anything following an extended class
1711        that contained multibyte characters and which was followed by a quantifier.
1713    15. Added optional support for general category Unicode character properties
1714        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1715        support. It adds about 90K to the size of the library. The meanings of the
1716        inbuilt class escapes such as \d and \s have NOT been changed.
1718    16. Updated pcredemo.c to include calls to free() to release the memory for the
1719        compiled pattern.
1721    17. The generated file chartables.c was being created in the source directory
1722        instead of in the building directory. This caused the build to fail if the
1723        source directory was different from the building directory, and was
1724        read-only.
1726    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1727        file. No doubt somebody will tell me if they don't make sense... Also added
1728        Dan Mooney's comments about building on OpenVMS.
1730    19. Added support for partial matching via the PCRE_PARTIAL option for
1731        pcre_exec() and the \P data escape in pcretest.
1733    20. Extended pcretest with 3 new pattern features:
1735        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1736              write the compiled pattern to the file whose name is "rest-of-line".
1737              This is a straight binary dump of the data, with the saved pointer to
1738              the character tables forced to be NULL. The study data, if any, is
1739              written too. After writing, pcretest reads a new pattern.
1741        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1742              compiled pattern from the given file. There must not be any
1743              occurrences of "<" in the file name (pretty unlikely); if there are,
1744              pcretest will instead treat the initial "<" as a pattern delimiter.
1745              After reading in the pattern, pcretest goes on to read data lines as
1746              usual.
1748        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1749              and 16-bit fields in a compiled pattern, to simulate a pattern that
1750              was compiled on a host of opposite endianness.
1752    21. The pcre-exec() function can now cope with patterns that were compiled on
1753        hosts of opposite endianness, with this restriction:
1755          As for any compiled expression that is saved and used later, the tables
1756          pointer field cannot be preserved; the extra_data field in the arguments
1757          to pcre_exec() should be used to pass in a tables address if a value
1758          other than the default internal tables were used at compile time.
1760    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1761        now diagnosed as an error. Previously, most of the time, a negative number
1762        would have been treated as zero, but if in addition "ovector" was passed as
1763        NULL, a crash could occur.
1765    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1766        new versions from the libtool 1.5 distribution (the last one is a copy of
1767        a file called libtool.m4). This seems to have fixed the need to patch
1768        "configure" to support Darwin 1.3 (which I used to do). However, I still
1769        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1770        workstation).
1772    24. Changed the PCRE licence to be the more standard "BSD" licence.
1775    Version 4.5 01-Dec-03
1776    ---------------------
1778     1. There has been some re-arrangement of the code for the match() function so
1779        that it can be compiled in a version that does not call itself recursively.
1780        Instead, it keeps those local variables that need separate instances for
1781        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1782        needs to "recurse". Keeping track of where control must go is done by means
1783        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1784        hide most of the details from the main code, and operates only if
1785        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1786        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1787        operating.
1789        To make it easier for callers to provide specially tailored get/free
1790        functions for this usage, two new functions, pcre_stack_malloc, and
1791        pcre_stack_free, are used. They are always called in strict stacking order,
1792        and the size of block requested is always the same.
1794        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1795        PCRE has been compiled to use the stack or the heap for recursion. The
1796        -C option of pcretest uses this to show which version is compiled.
1798        A new data escape \S, is added to pcretest; it causes the amounts of store
1799        obtained and freed by both kinds of malloc/free at match time to be added
1800        to the output.
1802     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1803        what's available on my current Linux desktop machine.
1805     3. When matching a UTF-8 string, the test for a valid string at the start has
1806        been extended. If start_offset is not zero, PCRE now checks that it points
1807        to a byte that is the start of a UTF-8 character. If not, it returns
1808        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1809        this is necessary because there may be backward assertions in the pattern.
1810        When matching the same subject several times, it may save resources to use
1811        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1813     4. The code for checking the validity of UTF-8 strings has been tightened so
1814        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1815        containing "overlong sequences".
1817     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1818        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1819        should have been "&", but it just so happened that all the cases this let
1820        through by mistake were picked up later in the function.
1822     6. I had used a variable called "isblank" - this is a C99 function, causing
1823        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1825     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1826        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1827        so the version is shown; (c) stop "make test" echoing "./RunTest".
1829     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1831     9. The prototype for memmove() for systems that don't have it was using
1832        size_t, but the inclusion of the header that defines size_t was later. I've
1833        moved the #includes for the C headers earlier to avoid this.
1835    10. Added some adjustments to the code to make it easier to compiler on certain
1836        special systems:
1838          (a) Some "const" qualifiers were missing.
1839          (b) Added the macro EXPORT before all exported functions; by default this
1840              is defined to be empty.
1841          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1842              that it reads its output file name as an argument instead of writing
1843              to the standard output and assuming this can be redirected.
1845    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1846        class containing characters with values greater than 255, PCRE compilation
1847        went into a loop.
1849    12. A recursive reference to a subpattern that was within another subpattern
1850        that had a minimum quantifier of zero caused PCRE to crash. For example,
1851        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1852        recursion. If the recursively-called subpattern itself had a zero repeat,
1853        that was OK.
1855    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1856        buffer into which it was copied (for escape processing) was still set at
1857        1024, so long lines caused crashes.
1859    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1860        "internal error: code overflow...". This applied to any character class
1861        that was followed by a possessive quantifier.
1863    15. Modified the Makefile to add libpcre.la as a prerequisite for
1864        libpcreposix.la because I was told this is needed for a parallel build to
1865        work.
1867    16. If a pattern that contained .* following optional items at the start was
1868        studied, the wrong optimizing data was generated, leading to matching
1869        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1870        matching string must start with a or b or c. The correct conclusion for
1871        this pattern is that a match can start with any character.
1874    Version 4.4 13-Aug-03
1875    ---------------------
1877     1. In UTF-8 mode, a character class containing characters with values between
1878        127 and 255 was not handled correctly if the compiled pattern was studied.
1879        In fixing this, I have also improved the studying algorithm for such
1880        classes (slightly).
1882     2. Three internal functions had redundant arguments passed to them. Removal
1883        might give a very teeny performance improvement.
1885     3. Documentation bug: the value of the capture_top field in a callout is *one
1886        more than* the number of the hightest numbered captured substring.
1888     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1889        in incorrectly linking with a previously installed version. They now link
1890        explicitly with libpcre.la.
1892     5. configure.in no longer needs to recognize Cygwin specially.
1894     6. A problem in pcre.in for Windows platforms is fixed.
1896     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1897        pcretest, it used to include the size of the study block as part of its
1898        output. Unfortunately, the structure contains a field that has a different
1899        size on different hardware architectures. This meant that the tests that
1900        showed this size failed. As the block is currently always of a fixed size,
1901        this information isn't actually particularly useful in pcretest output, so
1902        I have just removed it.
1904     8. Three pre-processor statements accidentally did not start in column 1.
1905        Sadly, there are *still* compilers around that complain, even though
1906        standard C has not required this for well over a decade. Sigh.
1908     9. In pcretest, the code for checking callouts passed small integers in the
1909        callout_data field, which is a void * field. However, some picky compilers
1910        complained about the casts involved for this on 64-bit systems. Now
1911        pcretest passes the address of the small integer instead, which should get
1912        rid of the warnings.
1914    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1915        both compile and run time, and gives an error if an invalid UTF-8 sequence
1916        is found. There is a option for disabling this check in cases where the
1917        string is known to be correct and/or the maximum performance is wanted.
1919    11. In response to a bug report, I changed one line in Makefile.in from
1921            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1922        to
1923            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1925        to look similar to other lines, but I have no way of telling whether this
1926        is the right thing to do, as I do not use Windows. No doubt I'll get told
1927        if it's wrong...
1930    Version 4.3 21-May-03
1931    ---------------------
1933    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1934       Makefile.
1936    2. Some refactoring to improve the quality of the code:
1938       (i)   The utf8_table... variables are now declared "const".
1940       (ii)  The code for \cx, which used the "case flipping" table to upper case
1941             lower case letters, now just substracts 32. This is ASCII-specific,
1942             but the whole concept of \cx is ASCII-specific, so it seems
1943             reasonable.
1945       (iii) PCRE was using its character types table to recognize decimal and
1946             hexadecimal digits in the pattern. This is silly, because it handles
1947             only 0-9, a-f, and A-F, but the character types table is locale-
1948             specific, which means strange things might happen. A private
1949             table is now used for this - though it costs 256 bytes, a table is
1950             much faster than multiple explicit tests. Of course, the standard
1951             character types table is still used for matching digits in subject
1952             strings against \d.
1954       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1955             ending in _t are). So I've renamed it as ESC_tee.
1957    3. The first argument for regexec() in the POSIX wrapper should have been
1958       defined as "const".
1960    4. Changed pcretest to use malloc() for its buffers so that they can be
1961       Electric Fenced for debugging.
1963    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1964       to read one or more bytes before the start of the subject string. Often this
1965       had no effect on PCRE's behaviour, but in some circumstances it could
1966       provoke a segmentation fault.
1968    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1969       to try to read one or more bytes before the start of the subject string.
1971    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1972       UTF-8 support could misbehave in various ways if the subject string
1973       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1974       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1975       back over UTF-8 characters.)
1978    Version 4.2 14-Apr-03
1979    ---------------------
1981    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1983    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1984         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1985         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1986         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1987         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1988           and BUILD_EXEEXT
1989         Note: automatic setting of the BUILD variables is not yet working
1990         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1991           compile-time but not at link-time
1992         [LINK]: use for linking executables only
1993         make different versions for Windows and non-Windows
1994         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1995           libraries
1996         [LINK_FOR_BUILD]: new variable
1997         [OBJEXT]: use throughout
1998         [EXEEXT]: use throughout
1999         <winshared>: new target
2000         <wininstall>: new target
2001         <dftables.o>: use native compiler
2002         <dftables>: use native linker
2003         <install>: handle Windows platform correctly
2004         <clean>: ditto
2005         <check>: ditto
2006         copy DLL to top builddir before testing
2008       As part of these changes, -no-undefined was removed again. This was reported
2009       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2010       in any case.
2012    3. Some tidies to get rid of compiler warnings:
2014       . In the match_data structure, match_limit was an unsigned long int, whereas
2015         match_call_count was an int. I've made them both unsigned long ints.
2017       . In pcretest the fact that a const uschar * doesn't automatically cast to
2018         a void * provoked a warning.
2020       . Turning on some more compiler warnings threw up some "shadow" variables
2021         and a few more missing casts.
2023    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2024       option, a class that contained a single character with a value between 128
2025       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2027    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2028       option, a class that contained several characters, but with at least one
2029       whose value was between 128 and 255 caused PCRE to crash.
2032    Version 4.1 12-Mar-03
2033    ---------------------
2035    1. Compiling with gcc -pedantic found a couple of places where casts were
2036    needed, and a string in dftables.c that was longer than standard compilers are
2037    required to support.
2039    2. Compiling with Sun's compiler found a few more places where the code could
2040    be tidied up in order to avoid warnings.
2042    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2043    first of these names is deprecated in the latest Autoconf in favour of the name
2044    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2045    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2046    analogy I have changed it to CFLAGS_FOR_BUILD.
2048    4. Added -no-undefined to the linking command in the Makefile, because this is
2049    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2050    linking step for the pcreposix library.
2052    5. PCRE was failing to diagnose the case of two named groups with the same
2053    name.
2055    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2056    literal character that is needed in the subject for a match, and scans along to
2057    ensure that it is present before embarking on the full matching process. This
2058    saves time in cases of nested unlimited repeats that are never going to match.
2059    Problem: the scan can take a lot of time if the subject is very long (e.g.
2060    megabytes), thus penalizing straightforward matches. It is now done only if the
2061    amount of subject to be scanned is less than 1000 bytes.
2063    7. A lesser problem with the same optimization is that it was recording the
2064    first character of an anchored pattern as "needed", thus provoking a search
2065    right along the subject, even when the first match of the pattern was going to
2066    fail. The "needed" character is now not set for anchored patterns, unless it
2067    follows something in the pattern that is of non-fixed length. Thus, it still
2068    fulfils its original purpose of finding quick non-matches in cases of nested
2069    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2072    Version 4.0 17-Feb-03
2073    ---------------------
2075    1. If a comment in an extended regex that started immediately after a meta-item
2076    extended to the end of string, PCRE compiled incorrect data. This could lead to
2077    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2079    2. Moved to autoconf 2.53 and libtool 1.4.2.
2081    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2082    the special perltest8 script is no longer needed - all the tests can be run
2083    from a single perltest script.
2085    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2086    by \s. It has now been removed in PCRE. This means it isn't recognized as
2087    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2088    class [:space:] *does* include VT, thereby creating a mess.
2090    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2091    space and tab.
2093    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2094    its new features into the main test script, reducing the number of scripts.
2096    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2097    were backward compatible, and made the (?i) apply to the whole pattern, as if
2098    /i were given. Now it behaves more logically, and applies the option setting
2099    only to what follows. PCRE has been changed to follow suit. However, if it
2100    finds options settings right at the start of the pattern, it extracts them into
2101    the global options, as before. Thus, they show up in the info data.
2103    8. Added support for the \Q...\E escape sequence. Characters in between are
2104    treated as literals. This is slightly different from Perl in that $ and @ are
2105    also handled as literals inside the quotes. In Perl, they will cause variable
2106    interpolation. Note the following examples:
2108        Pattern            PCRE matches      Perl matches
2110        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2111        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2112        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2114    For compatibility with Perl, \Q...\E sequences are recognized inside character
2115    classes as well as outside them.
2117    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2118    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2119    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2120    signed/unsigned warnings.
2122    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2123    option for pcretest, so I've replaced it by a simple function that does just
2124    that job.
2126    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2127    "pcregrep -".
2129    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2130    Java package. This provides some syntactic sugar for simple cases of what my
2131    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2132    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2133    item, you can use this simplified notation. Note that only makes sense with
2134    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2135    greediness, whatever the setting of the PCRE_UNGREEDY option.
2137    13. A change of greediness default within a pattern was not taking effect at
2138    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2139    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2140    was abstracted outside.
2142    14. PCRE now supports the \G assertion. It is true when the current matching
2143    position is at the start point of the match. This differs from \A when the
2144    starting offset is non-zero. Used with the /g option of pcretest (or similar
2145    code), it works in the same way as it does for Perl's /g option. If all
2146    alternatives of a regex begin with \G, the expression is anchored to the start
2147    match position, and the "anchored" flag is set in the compiled expression.
2149    15. Some bugs concerning the handling of certain option changes within patterns
2150    have been fixed. These applied to options other than (?ims). For example,
2151    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2152    been the other way round. Some of this was related to change 7 above.
2154    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2155    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2156    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2157    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2159    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2160    mode. Unlike ".", it always matches newline, whatever the setting of
2161    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2162    assertions. Perl allows it, but it doesn't (in general) work because it can't
2163    calculate the length of the lookbehind. At least, that's the case for Perl
2164    5.8.0 - I've been told they are going to document that it doesn't work in
2165    future.
2167    18. Added an error diagnosis for escapes that PCRE does not support: these are
2168    \L, \l, \N, \P, \p, \U, \u, and \X.
2170    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2171    reading past the end of the pattern in cases such as /[abcd/.
2173    20. PCRE was getting more memory than necessary for patterns with classes that
2174    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2176    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2177    compiling PCRE for use with Virtual Pascal.
2179    22. Small fix to the Makefile to make it work properly if the build is done
2180    outside the source tree.
2182    23. Added a new extension: a condition to go with recursion. If a conditional
2183    subpattern starts with (?(R) the "true" branch is used if recursion has
2184    happened, whereas the "false" branch is used only at the top level.
2186    24. When there was a very long string of literal characters (over 255 bytes
2187    without UTF support, over 250 bytes with UTF support), the computation of how
2188    much memory was required could be incorrect, leading to segfaults or other
2189    strange effects.
2191    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2192    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2193    there was a subsequent back reference to those brackets. This meant that, for
2194    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2195    possible to check for precisely this case. All we can do is abandon the
2196    optimization if .* occurs inside capturing brackets when there are any back
2197    references whatsoever. (See below for a better fix that came later.)
2199    26. The handling of the optimization for finding the first character of a
2200    non-anchored pattern, and for finding a character that is required later in the
2201    match were failing in some cases. This didn't break the matching; it just
2202    failed to optimize when it could. The way this is done has been re-implemented.
2204    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2206    28. Added a new feature that provides some of the functionality that Perl
2207    provides with (?{...}). The facility is termed a "callout". The way it is done
2208    in PCRE is for the caller to provide an optional function, by setting
2209    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2210    global variable. By default it is unset, which disables all calling out. To get
2211    the function called, the regex must include (?C) at appropriate points. This
2212    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2213    This provides a means of identifying different callout points. When PCRE
2214    reaches such a point in the regex, if pcre_callout has been set, the external
2215    function is called. It is provided with data in a structure called
2216    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2217    matching continues; if it returns a non-zero value, the match at the current
2218    point fails. However, backtracking will occur if possible. [This was changed
2219    later and other features added - see item 49 below.]
2221    29. pcretest is upgraded to test the callout functionality. It provides a
2222    callout function that displays information. By default, it shows the start of
2223    the match and the current position in the text. There are some new data escapes
2224    to vary what happens:
2226        \C+         in addition, show current contents of captured substrings
2227        \C-         do not supply a callout function
2228        \C!n        return 1 when callout number n is reached
2229        \C!n!m      return 1 when callout number n is reached for the mth time
2231    30. If pcregrep was called with the -l option and just a single file name, it
2232    output "<stdin>" if a match was found, instead of the file name.
2234    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2235    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2236    pcre_exec(). This saves a malloc/free per call. The default value of
2237    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2238    when configuring.
2240    32. The default maximum size of a compiled pattern is 64K. There have been a
2241    few cases of people hitting this limit. The code now uses macros to handle the
2242    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2243    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2244    configuring. Tests 2 and 5 work only with 2-byte links because they output
2245    debugging information about compiled patterns.
2247    33. Internal code re-arrangements:
2249    (a) Moved the debugging function for printing out a compiled regex into
2250        its own source file (printint.c) and used #include to pull it into
2251        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2252        separate copies.
2254    (b) Defined the list of op-code names for debugging as a macro in
2255        internal.h so that it is next to the definition of the opcodes.
2257    (c) Defined a table of op-code lengths for simpler skipping along compiled
2258        code. This is again a macro in internal.h so that it is next to the
2259        definition of the opcodes.
2261    34. Added support for recursive calls to individual subpatterns, along the
2262    lines of Robin Houston's patch (but implemented somewhat differently).
2264    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2265    allow it to read and process whole directories in Win32. This code was
2266    contributed by Lionel Fourquaux; it has not been tested by me.
2268    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2269    used to name a group. Names consist of alphanumerics and underscores, and must
2270    be unique. Back references use the syntax (?P=name) and recursive calls use
2271    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2272    numbers. The function pcre_fullinfo() can be used after compilation to extract
2273    a name/number map. There are three relevant calls:
2275      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2276      PCRE_INFO_NAMECOUNT            yields the number of entries
2277      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2279    The map is a vector of fixed-size entries. The size of each entry depends on
2280    the length of the longest name used. The first two bytes of each entry are the
2281    group number, most significant byte first. There follows the corresponding
2282    name, zero terminated. The names are in alphabetical order.
2284    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2285    case instead of 255. Making it the same both with and without UTF-8 support
2286    means that the same test output works with both.
2288    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2289    calling malloc() with a zero argument.
2291    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2292    optimization. I've improved things by keeping a bitmap of backreferences with
2293    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2294    fact referenced, the optimization can be applied. It is unlikely that a
2295    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2296    the match to follow \n) will appear inside brackets with a number greater than
2297    31, but if it does, any back reference > 31 suppresses the optimization.
2299    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2300    of disabling numbered capturing parentheses. Any opening parenthesis that is
2301    not followed by ? behaves as if it were followed by ?: but named parentheses
2302    can still be used for capturing (and they will acquire numbers in the usual
2303    way).
2305    41. Redesigned the return codes from the match() function into yes/no/error so
2306    that errors can be passed back from deep inside the nested calls. A malloc
2307    failure while inside a recursive subpattern call now causes the
2308    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2310    42. It is now possible to set a limit on the number of times the match()
2311    function is called in a call to pcre_exec(). This facility makes it possible to
2312    limit the amount of recursion and backtracking, though not in a directly
2313    obvious way, because the match() function is used in a number of different
2314    circumstances. The count starts from zero for each position in the subject
2315    string (for non-anchored patterns). The default limit is, for compatibility, a
2316    large number, namely 10 000 000. You can change this in two ways:
2318    (a) When configuring PCRE before making, you can use --with-match-limit=n
2319        to set a default value for the compiled library.
2321    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2322        a different value is set. See 45 below.
2324    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2326    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2327    of things that can be changed at compile time. The first argument specifies
2328    what is wanted and the second points to where the information is to be placed.
2329    The current list of available information is:
2331      PCRE_CONFIG_UTF8
2333    The output is an integer that is set to one if UTF-8 support is available;
2334    otherwise it is set to zero.
2338    The output is an integer that it set to the value of the code that is used for
2339    newline. It is either LF (10) or CR (13).
2343    The output is an integer that contains the number of bytes used for internal
2344    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2348    The output is an integer that contains the threshold above which the POSIX
2349    interface uses malloc() for output vectors. See item 31 above.
2353    The output is an unsigned integer that contains the default limit of the number
2354    of match() calls in a pcre_exec() execution. See 42 above.
2356    44. pcretest has been upgraded by the addition of the -C option. This causes it
2357    to extract all the available output from the new pcre_config() function, and to
2358    output it. The program then exits immediately.
2360    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2361    order to support additional features. One way would have been to define
2362    pcre_exec2() (for example) with extra arguments, but this would not have been
2363    extensible, and would also have required all calls to the original function to
2364    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2365    is used for passing in "extra" data from pcre_study().
2367    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2368    contains the following fields:
2370      flags         a bitmap indicating which of the following fields are set
2371      study_data    opaque data from pcre_study()
2372      match_limit   a way of specifying a limit on match() calls for a specific
2373                      call to pcre_exec()
2374      callout_data  data for callouts (see 49 below)
2376    The flag bits are also defined in pcre.h, and are
2382    The pcre_study() function now returns one of these new pcre_extra blocks, with
2383    the actual study data pointed to by the study_data field, and the
2384    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2385    before. That is, this change is entirely upwards-compatible and requires no
2386    change to existing code.
2388    If you want to pass in additional data to pcre_exec(), you can either place it
2389    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2390    block.
2392    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2393    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2394    times with different match limits, until it finds the minimum value needed for
2395    pcre_exec() to complete. The value is then output. This can be instructive; for
2396    most simple matches the number is quite small, but for pathological cases it
2397    gets very large very quickly.
2399    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2400    returns the size of the data block pointed to by the study_data field in a
2401    pcre_extra block, that is, the value that was passed as the argument to
2402    pcre_malloc() when PCRE was getting memory in which to place the information
2403    created by pcre_study(). The fourth argument should point to a size_t variable.
2404    pcretest has been extended so that this information is shown after a successful
2405    pcre_study() call when information about the compiled regex is being displayed.
2407    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2408    because what follows is always an absolute path. (Later: it turns out that this
2409    is more than cosmetic for MinGW, because it doesn't like empty path
2410    components.)
2412    49. Some changes have been made to the callout feature (see 28 above):
2414    (i)  A callout function now has three choices for what it returns:
2416           0  =>  success, carry on matching
2417         > 0  =>  failure at this point, but backtrack if possible
2418         < 0  =>  serious error, return this value from pcre_exec()
2420         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2421         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2422         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2423         use by callout functions. It will never be used by PCRE itself.
2425    (ii) The pcre_extra structure (see 45 above) has a void * field called
2426         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2427         pcre_callout_block structure has a field of the same name. The contents of
2428         the field passed in the pcre_extra structure are passed to the callout
2429         function in the corresponding field in the callout block. This makes it
2430         easier to use the same callout-containing regex from multiple threads. For
2431         testing, the pcretest program has a new data escape
2433           \C*n        pass the number n (may be negative) as callout_data
2435         If the callout function in pcretest receives a non-zero value as
2436         callout_data, it returns that value.
2438    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2439    there were some redundant $(CFLAGS) in commands that are now specified as
2440    $(LINK), which already includes $(CFLAGS).
2442    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2443    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2444    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2445    one-byte characters throughout. Note that case-insensitive matching applies
2446    only to characters whose values are less than 256. PCRE doesn't support the
2447    notion of cases for higher-valued characters.
2449    (i)   A character class whose characters are all within 0-255 is handled as
2450          a bit map, and the map is inverted for negative classes. Previously, a
2451          character > 255 always failed to match such a class; however it should
2452          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2454    (ii)  A negated character class with a single character < 255 is coded as
2455          "not this character" (OP_NOT). This wasn't working properly when the test
2456          character was multibyte, either singly or repeated.
2458    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2459          mode, for example: \x{100}{2,3}.
2461    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2462          singly or repeated) now correctly test multibyte characters. However,
2463          PCRE doesn't recognize any characters with values greater than 255 as
2464          digits, spaces, or word characters. Such characters always match \D, \S,
2465          and \W, and never match \d, \s, or \w.
2467    (v)   Classes may now contain characters and character ranges with values
2468          greater than 255. For example: [ab\x{100}-\x{400}].
2470    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2471          PCRE in UTF-8 mode.
2473    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2474    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2475    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2476    value.)
2478    53. The single man page has become too large. I have therefore split it up into
2479    a number of separate man pages. These also give rise to individual HTML pages;
2480    these are now put in a separate directory, and there is an index.html page that
2481    lists them all. Some hyperlinking between the pages has been installed.
2483    54. Added convenience functions for handling named capturing parentheses.
2485    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2486    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2487    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2488    are faulted.
2490    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2491    calling configure. These values are used when compiling the dftables.c program
2492    which is run to generate the source of the default character tables. They
2493    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2494    you will need to set these values.
2496    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2499    Version 3.9 02-Jan-02
2500    ---------------------
2502    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2504    2. If --disable-static was given, the building process failed when trying to
2505    build pcretest and pcregrep. (For some reason it was using libtool to compile
2506    them, which is not right, as they aren't part of the library.)
2509    Version 3.8 18-Dec-01
2510    ---------------------
2512    1. The experimental UTF-8 code was completely screwed up. It was packing the
2513    bytes in the wrong order. How dumb can you get?
2516    Version 3.7 29-Oct-01
2517    ---------------------
2519    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2520    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2521    this didn't happen under Solaris 8, where I normally test things.
2523    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2524    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2525    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2526    this fix an hour or so after the initial 3.7 release.)
2529    Version 3.6 23-Oct-01
2530    ---------------------
2532    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2533    offsets passed as NULL with zero offset count.
2535    2. The config.guess and config.sub files had not been updated when I moved to
2536    the latest autoconf.
2539    Version 3.5 15-Aug-01
2540    ---------------------
2542    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2543    had been forgotten.
2545    2. By using declared but undefined structures, we can avoid using "void"
2546    definitions in pcre.h while keeping the internal definitions of the structures
2547    private.
2549    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2550    user point of view, this means that both static and shared libraries are built
2551    by default, but this can be individually controlled. More of the work of
2552    handling this static/shared cases is now inside libtool instead of PCRE's make
2553    file.
2555    4. The pcretest utility is now installed along with pcregrep because it is
2556    useful for users (to test regexs) and by doing this, it automatically gets
2557    relinked by libtool. The documentation has been turned into a man page, so
2558    there are now .1, .txt, and .html versions in /doc.
2560    5. Upgrades to pcregrep:
2561       (i)   Added long-form option names like gnu grep.
2562       (ii)  Added --help to list all options with an explanatory phrase.
2563       (iii) Added -r, --recursive to recurse into sub-directories.
2564       (iv)  Added -f, --file to read patterns from a file.
2566    6. pcre_exec() was referring to its "code" argument before testing that
2567    argument for NULL (and giving an error if it was NULL).
2569    7. Upgraded Makefile.in to allow for compiling in a different directory from
2570    the source directory.
2572    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2573    options bits, the pointer it was passed was to an int instead of to an unsigned
2574    long int. This mattered only on 64-bit systems.
2576    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2577    generated) instead of pcre.in, which it its source. Also made the same change
2578    in several of the .c files.
2580    10. A new release of gcc defines printf() as a macro, which broke pcretest
2581    because it had an ifdef in the middle of a string argument for printf(). Fixed
2582    by using separate calls to printf().
2584    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2585    script, to force use of CR or LF instead of \n in the source. On non-Unix
2586    systems, the value can be set in config.h.
2588    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2589    absolute limit. Changed the text of the error message to make this clear, and
2590    likewise updated the man page.
2592    13. The limit of 99 on the number of capturing subpatterns has been removed.
2593    The new limit is 65535, which I hope will not be a "real" limit.
2596  Version 3.4 22-Aug-00  Version 3.4 22-Aug-00
2597  ---------------------  ---------------------

Removed from v.51  
changed lines
  Added in v.332

  ViewVC Help
Powered by ViewVC 1.1.5