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

Diff of /code/trunk/ChangeLog

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

revision 49 by nigel, Sat Feb 24 21:39:33 2007 UTC revision 325 by ph10, Sat Mar 8 17:13:02 2008 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.7 05-Mar-08
5    ---------------------
6    
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.
10    
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.
14    
15    3.  Remove a line of dead code, identified by coverity and reported by Nuno
16        Lopes.
17    
18    4.  Fixed two related pcregrep bugs involving -r with --include or --exclude:
19    
20        (1) The include/exclude patterns were being applied to the whole pathnames
21            of files, instead of just to the final components.
22    
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.
29    
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.
32    
33    
34    Version 7.6 28-Jan-08
35    ---------------------
36    
37    1.  A character class containing a very large number of characters with
38        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
39        overflow.
40    
41    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
42        HAVE_LONG_LONG is not defined.
43    
44    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
45        bring them up to date and include new features. This patch includes:
46    
47        - Fixed PH's badly added libz and libbz2 support.
48        - Fixed a problem with static linking.
49        - Added pcredemo. [But later removed - see 7 below.]
50        - Fixed dftables problem and added an option.
51        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
52            HAVE_LONG_LONG.
53        - Added readline support for pcretest.
54        - Added an listing of the option settings after cmake has run.
55    
56    4.  A user submitted a patch to Makefile that makes it easy to create
57        "pcre.dll" under mingw when using Configure/Make. I added stuff to
58        Makefile.am that cause it to include this special target, without
59        affecting anything else. Note that the same mingw target plus all
60        the other distribution libraries and programs are now supported
61        when configuring with CMake (see 6 below) instead of with
62        Configure/Make.
63    
64    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
65        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
66        exported in the Windows port". It has not yet been confirmed that the patch
67        solves the problem, but it does no harm.
68    
69    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
70        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
71        with CMake, and also correct the comment about stack recursion.
72    
73    7.  Remove the automatic building of pcredemo from the ./configure system and
74        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
75        of a program that users should build themselves after PCRE is installed, so
76        building it automatically is not really right. What is more, it gave
77        trouble in some build environments.
78    
79    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
80    
81    
82    Version 7.5 10-Jan-08
83    ---------------------
84    
85    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
86        values in parens when parsing an RE using the C++ wrapper."
87    
88    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
89        Characters greater than 255 were excluded from the class instead of being
90        included.
91    
92    3.  The same bug as (2) above applied to negated POSIX classes such as
93        [:^space:].
94    
95    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
96        defined or documented. It seems to have been a typo for PCRE_STATIC, so
97        I have changed it.
98    
99    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
100        first named subpattern) and a construct such as (?&a) would reference the
101        first named subpattern whose name started with "a" (in other words, the
102        length check was missing). Both these problems are fixed. "Subpattern name
103        expected" is now given for (?&) (a zero-length name), and this patch also
104        makes it give the same error for \k'' (previously it complained that that
105        was a reference to a non-existent subpattern).
106    
107    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
108        this is right because (?- can be followed by option settings as well as by
109        digits. I have, however, made the messages clearer.
110    
111    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
112        than the number used in the conditional) now cause a compile-time error.
113        This is actually not compatible with Perl, which accepts such patterns, but
114        treats the conditional as always being FALSE (as PCRE used to), but it
115        seems to me that giving a diagnostic is better.
116    
117    8.  Change "alphameric" to the more common word "alphanumeric" in comments
118        and messages.
119    
120    9.  Fix two occurrences of "backslash" in comments that should have been
121        "backspace".
122    
123    10. Remove two redundant lines of code that can never be obeyed (their function
124        was moved elsewhere).
125    
126    11. The program that makes PCRE's Unicode character property table had a bug
127        which caused it to generate incorrect table entries for sequences of
128        characters that have the same character type, but are in different scripts.
129        It amalgamated them into a single range, with the script of the first of
130        them. In other words, some characters were in the wrong script. There were
131        thirteen such cases, affecting characters in the following ranges:
132    
133          U+002b0 - U+002c1
134          U+0060c - U+0060d
135          U+0061e - U+00612
136          U+0064b - U+0065e
137          U+0074d - U+0076d
138          U+01800 - U+01805
139          U+01d00 - U+01d77
140          U+01d9b - U+01dbf
141          U+0200b - U+0200f
142          U+030fc - U+030fe
143          U+03260 - U+0327f
144          U+0fb46 - U+0fbb1
145          U+10450 - U+1049d
146    
147    12. The -o option (show only the matching part of a line) for pcregrep was not
148        compatible with GNU grep in that, if there was more than one match in a
149        line, it showed only the first of them. It now behaves in the same way as
150        GNU grep.
151    
152    13. If the -o and -v options were combined for pcregrep, it printed a blank
153        line for every non-matching line. GNU grep prints nothing, and pcregrep now
154        does the same. The return code can be used to tell if there were any
155        non-matching lines.
156    
157    14. Added --file-offsets and --line-offsets to pcregrep.
158    
159    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
160        infinitely looping recursion. The bug was that positive lookaheads were not
161        being skipped when checking for a possible empty match (negative lookaheads
162        and both kinds of lookbehind were skipped).
163    
164    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
165        inclusion of <windows.h> to before rather than after the definition of
166        INVALID_FILE_ATTRIBUTES (patch from David Byron).
167    
168    17. Specifying a possessive quantifier with a specific limit for a Unicode
169        character property caused pcre_compile() to compile bad code, which led at
170        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
171        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
172        caused the error; without that there was no problem.
173    
174    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
175    
176    19. Added --enable-pcretest-libreadline.
177    
178    20. In pcrecpp.cc, the variable 'count' was incremented twice in
179        RE::GlobalReplace(). As a result, the number of replacements returned was
180        double what it should be. I removed one of the increments, but Craig sent a
181        later patch that removed the other one (the right fix) and added unit tests
182        that check the return values (which was not done before).
183    
184    21. Several CMake things:
185    
186        (1) Arranged that, when cmake is used on Unix, the libraries end up with
187            the names libpcre and libpcreposix, not just pcre and pcreposix.
188    
189        (2) The above change means that pcretest and pcregrep are now correctly
190            linked with the newly-built libraries, not previously installed ones.
191    
192        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
193    
194    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
195        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
196        UTF-8 newline character). The key issue is that the pattern starts .*;
197        this means that the match must be either at the beginning, or after a
198        newline. The bug was in the code for advancing after a failed match and
199        checking that the new position followed a newline. It was not taking
200        account of UTF-8 characters correctly.
201    
202    23. PCRE was behaving differently from Perl in the way it recognized POSIX
203        character classes. PCRE was not treating the sequence [:...:] as a
204        character class unless the ... were all letters. Perl, however, seems to
205        allow any characters between [: and :], though of course it rejects as
206        unknown any "names" that contain non-letters, because all the known class
207        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
208        for example, whereas PCRE did not - it did not recognize a POSIX character
209        class. This seemed a bit dangerous, so the code has been changed to be
210        closer to Perl. The behaviour is not identical to Perl, because PCRE will
211        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
212        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
213        Perl does, and where it didn't before.
214    
215    24. Rewrite so as to remove the single use of %n from pcregrep because in some
216        Windows environments %n is disabled by default.
217    
218    
219    Version 7.4 21-Sep-07
220    ---------------------
221    
222    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
223        means that a class such as [\s] counted as "explicit reference to CR or
224        LF". That isn't really right - the whole point of the change was to try to
225        help when there was an actual mention of one of the two characters. So now
226        the change happens only if \r or \n (or a literal CR or LF) character is
227        encountered.
228    
229    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
230        of both had grown to the point where there were only 3 bits left.
231        Fortunately, there was spare space in the data structure, and so I have
232        moved the internal flags into a new 16-bit field to free up more option
233        bits.
234    
235    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
236        but did not set the internal JCHANGED flag - either of these is enough to
237        control the way the "get" function works - but the PCRE_INFO_JCHANGED
238        facility is supposed to tell if (?J) was ever used, so now (?J) at the
239        start sets both bits.
240    
241    4.  Added options (at build time, compile time, exec time) to change \R from
242        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
243    
244    5.  doc/pcresyntax.html was missing from the distribution.
245    
246    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
247        compatibility, even though it is no longer used.
248    
249    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
250        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
251        windows.h file is present (where different names are used). [This was
252        reversed later after testing - see 16 below.]
253    
254    8.  Changed all #include <config.h> to #include "config.h". There were also
255        some further <pcre.h> cases that I changed to "pcre.h".
256    
257    9.  When pcregrep was used with the --colour option, it missed the line ending
258        sequence off the lines that it output.
259    
260    10. It was pointed out to me that arrays of string pointers cause lots of
261        relocations when a shared library is dynamically loaded. A technique of
262        using a single long string with a table of offsets can drastically reduce
263        these. I have refactored PCRE in four places to do this. The result is
264        dramatic:
265    
266          Originally:                          290
267          After changing UCP table:            187
268          After changing error message table:   43
269          After changing table of "verbs"       36
270          After changing table of Posix names   22
271    
272        Thanks to the folks working on Gregex for glib for this insight.
273    
274    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
275        unicode-properties was also set.
276    
277    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
278    
279    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
280        checked only for CRLF.
281    
282    14. Added casts to pcretest.c to avoid compiler warnings.
283    
284    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
285    
286    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
287        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
288        entirely. This removes changes made in 7 above.
289    
290    17. The CMake files have been updated, and there is now more information about
291        building with CMake in the NON-UNIX-USE document.
292    
293    
294    Version 7.3 28-Aug-07
295    ---------------------
296    
297     1. In the rejigging of the build system that eventually resulted in 7.1, the
298        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
299        brackets there is not right, since it causes compilers to look for an
300        installed pcre.h, not the version that is in the source that is being
301        compiled (which of course may be different). I have changed it back to:
302    
303          #include "pcre.h"
304    
305        I have a vague recollection that the change was concerned with compiling in
306        different directories, but in the new build system, that is taken care of
307        by the VPATH setting the Makefile.
308    
309     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
310        when the subject happened to end in the byte 0x85 (e.g. if the last
311        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
312        characters but of course it shouldn't be taken as a newline when it is part
313        of another character. The bug was that, for an unlimited repeat of . in
314        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
315        characters when looking for a newline.
316    
317     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
318    
319     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
320        in debug output.
321    
322     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
323        long printing in the pcrecpp unittest when running under MinGW.
324    
325     6. ESC_K was left out of the EBCDIC table.
326    
327     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
328        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
329        limit also applies to "virtual nesting" when a pattern is recursive, and in
330        this case 1000 isn't so big. I have been able to remove this limit at the
331        expense of backing off one optimization in certain circumstances. Normally,
332        when pcre_exec() would call its internal match() function recursively and
333        immediately return the result unconditionally, it uses a "tail recursion"
334        feature to save stack. However, when a subpattern that can match an empty
335        string has an unlimited repetition quantifier, it no longer makes this
336        optimization. That gives it a stack frame in which to save the data for
337        checking that an empty string has been matched. Previously this was taken
338        from the 1000-entry workspace that had been reserved. So now there is no
339        explicit limit, but more stack is used.
340    
341     8. Applied Daniel's patches to solve problems with the import/export magic
342        syntax that is required for Windows, and which was going wrong for the
343        pcreposix and pcrecpp parts of the library. These were overlooked when this
344        problem was solved for the main library.
345    
346     9. There were some crude static tests to avoid integer overflow when computing
347        the size of patterns that contain repeated groups with explicit upper
348        limits. As the maximum quantifier is 65535, the maximum group length was
349        set at 30,000 so that the product of these two numbers did not overflow a
350        32-bit integer. However, it turns out that people want to use groups that
351        are longer than 30,000 bytes (though not repeat them that many times).
352        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
353        made it possible to implement the integer overflow checks in a much more
354        dynamic way, which I have now done. The artificial limitation on group
355        length has been removed - we now have only the limit on the total length of
356        the compiled pattern, which depends on the LINK_SIZE setting.
357    
358    10. Fixed a bug in the documentation for get/copy named substring when
359        duplicate names are permitted. If none of the named substrings are set, the
360        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
361        empty string.
362    
363    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
364        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
365        because the ] is interpreted as the first data character and the
366        terminating ] is not found. PCRE has been made compatible with Perl in this
367        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
368        cause memory overwriting.
369    
370    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
371        string has been matched (to stop an infinite loop). It was not recognizing
372        a conditional subpattern that could match an empty string if that
373        subpattern was within another subpattern. For example, it looped when
374        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
375        condition was not nested. This bug has been fixed.
376    
377    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
378        past the start of the subject in the presence of bytes with the top bit
379        set, for example "\x8aBCD".
380    
381    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
382        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
383    
384    14. Optimized (?!) to (*FAIL).
385    
386    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
387        This restricts code points to be within the range 0 to 0x10FFFF, excluding
388        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
389        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
390        does: it's just the validity check that is more restrictive.
391    
392    16. Inserted checks for integer overflows during escape sequence (backslash)
393        processing, and also fixed erroneous offset values for syntax errors during
394        backslash processing.
395    
396    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
397        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
398    
399    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
400        caused an overrun.
401    
402    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
403        something other than just ASCII characters) inside a group that had an
404        unlimited repeat caused a loop at compile time (while checking to see
405        whether the group could match an empty string).
406    
407    20. Debugging a pattern containing \p or \P could cause a crash. For example,
408        [\P{Any}] did so. (Error in the code for printing property names.)
409    
410    21. An orphan \E inside a character class could cause a crash.
411    
412    22. A repeated capturing bracket such as (A)? could cause a wild memory
413        reference during compilation.
414    
415    23. There are several functions in pcre_compile() that scan along a compiled
416        expression for various reasons (e.g. to see if it's fixed length for look
417        behind). There were bugs in these functions when a repeated \p or \P was
418        present in the pattern. These operators have additional parameters compared
419        with \d, etc, and these were not being taken into account when moving along
420        the compiled data. Specifically:
421    
422        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
423            length.
424    
425        (b) An item such as \pL+ within a repeated group could cause crashes or
426            loops.
427    
428        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
429            "reference to non-existent subpattern" error.
430    
431        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
432    
433    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
434        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
435    
436    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
437    
438    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
439        character were causing crashes (broken optimization).
440    
441    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
442        \p or \P) caused a compile-time loop.
443    
444    28. More problems have arisen in unanchored patterns when CRLF is a valid line
445        break. For example, the unstudied pattern [\r\n]A does not match the string
446        "\r\nA" because change 7.0/46 below moves the current point on by two
447        characters after failing to match at the start. However, the pattern \nA
448        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
449        the same is true. There doesn't seem any very clean way out of this, but
450        what I have chosen to do makes the common cases work: PCRE now takes note
451        of whether there can be an explicit match for \r or \n anywhere in the
452        pattern, and if so, 7.0/46 no longer applies. As part of this change,
453        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
454        pattern has explicit CR or LF references.
455    
456    29. Added (*CR) etc for changing newline setting at start of pattern.
457    
458    
459    Version 7.2 19-Jun-07
460    ---------------------
461    
462     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
463        which is apparently normally available under Windows.
464    
465     2. Re-jig the pcregrep tests with different newline settings in an attempt
466        to make them independent of the local environment's newline setting.
467    
468     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
469    
470     4. Some of the "internals" tests were previously cut out when the link size
471        was not 2, because the output contained actual offsets. The recent new
472        "Z" feature of pcretest means that these can be cut out, making the tests
473        usable with all link sizes.
474    
475     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
476        stack recursion. This gives a massive performance boost under BSD, but just
477        a small improvement under Linux. However, it saves one field in the frame
478        in all cases.
479    
480     6. Added more features from the forthcoming Perl 5.10:
481    
482        (a) (?-n) (where n is a string of digits) is a relative subroutine or
483            recursion call. It refers to the nth most recently opened parentheses.
484    
485        (b) (?+n) is also a relative subroutine call; it refers to the nth next
486            to be opened parentheses.
487    
488        (c) Conditions that refer to capturing parentheses can be specified
489            relatively, for example, (?(-2)... or (?(+3)...
490    
491        (d) \K resets the start of the current match so that everything before
492            is not part of it.
493    
494        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
495    
496        (f) \g{name} is another synonym - part of Perl 5.10's unification of
497            reference syntax.
498    
499        (g) (?| introduces a group in which the numbering of parentheses in each
500            alternative starts with the same number.
501    
502        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
503    
504     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
505        PCRE_INFO_JCHANGED.
506    
507     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
508        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
509        for detecting groups that can match an empty string.
510    
511     9. A pattern with a very large number of alternatives (more than several
512        hundred) was running out of internal workspace during the pre-compile
513        phase, where pcre_compile() figures out how much memory will be needed. A
514        bit of new cunning has reduced the workspace needed for groups with
515        alternatives. The 1000-alternative test pattern now uses 12 bytes of
516        workspace instead of running out of the 4096 that are available.
517    
518    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
519    
520    11. Applied patch from Google to remove an optimization that didn't quite work.
521        The report of the bug said:
522    
523          pcrecpp::RE("a*").FullMatch("aaa") matches, while
524          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
525          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
526    
527    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
528        it matched the wrong number of bytes.
529    
530    
531    Version 7.1 24-Apr-07
532    ---------------------
533    
534     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
535        that is more "standard", making use of automake and other Autotools. There
536        is some re-arrangement of the files and adjustment of comments consequent
537        on this.
538    
539     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
540        for recursive directory scanning broke on some systems because the files
541        are not scanned in any specific order and on different systems the order
542        was different. A call to "sort" has been inserted into RunGrepTest for the
543        approprate test as a short-term fix. In the longer term there may be an
544        alternative.
545    
546     3. I had an email from Eric Raymond about problems translating some of PCRE's
547        man pages to HTML (despite the fact that I distribute HTML pages, some
548        people do their own conversions for various reasons). The problems
549        concerned the use of low-level troff macros .br and .in. I have therefore
550        removed all such uses from the man pages (some were redundant, some could
551        be replaced by .nf/.fi pairs). The 132html script that I use to generate
552        HTML has been updated to handle .nf/.fi and to complain if it encounters
553        .br or .in.
554    
555     4. Updated comments in configure.ac that get placed in config.h.in and also
556        arranged for config.h to be included in the distribution, with the name
557        config.h.generic, for the benefit of those who have to compile without
558        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
559    
560     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
561        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
562        makevp.bat for the current PCRE, using the additional files
563        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
564    
565     6. A Windows user reported a minor discrepancy with test 2, which turned out
566        to be caused by a trailing space on an input line that had got lost in his
567        copy. The trailing space was an accident, so I've just removed it.
568    
569     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
570        that is needed.
571    
572     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
573        as "const" (a) because they are and (b) because it helps the PHP
574        maintainers who have recently made a script to detect big data structures
575        in the php code that should be moved to the .rodata section. I remembered
576        to update Builducptable as well, so it won't revert if ucptable.h is ever
577        re-created.
578    
579     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
580        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
581        order to be able to cut out the UTF-8 tables in the latter when UTF-8
582        support is not required. This saves 1.5-2K of code, which is important in
583        some applications.
584    
585        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
586        so as not to refer to the tables, even though these functions will never be
587        called when UTF-8 support is disabled. Otherwise there are problems with a
588        shared library.
589    
590    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
591    
592        (a) It was defining its arguments as char * instead of void *.
593    
594        (b) It was assuming that all moves were upwards in memory; this was true
595            a long time ago when I wrote it, but is no longer the case.
596    
597        The emulated memove() is provided for those environments that have neither
598        memmove() nor bcopy(). I didn't think anyone used it these days, but that
599        is clearly not the case, as these two bugs were recently reported.
600    
601    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
602        and Detrail to create the HTML documentation, the .txt form of the man
603        pages, and it removes trailing spaces from listed files. It also creates
604        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
605        case, it wraps all the #defines with #ifndefs. This script should be run
606        before "make dist".
607    
608    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
609        with Unicode property support.
610    
611        (a) For a maximizing quantifier, if the two different cases of the
612            character were of different lengths in their UTF-8 codings (there are
613            some cases like this - I found 11), and the matching function had to
614            back up over a mixture of the two cases, it incorrectly assumed they
615            were both the same length.
616    
617        (b) When PCRE was configured to use the heap rather than the stack for
618            recursion during matching, it was not correctly preserving the data for
619            the other case of a UTF-8 character when checking ahead for a match
620            while processing a minimizing repeat. If the check also involved
621            matching a wide character, but failed, corruption could cause an
622            erroneous result when trying to check for a repeat of the original
623            character.
624    
625    13. Some tidying changes to the testing mechanism:
626    
627        (a) The RunTest script now detects the internal link size and whether there
628            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
629            values substituted by "configure". (The RunGrepTest script already did
630            this for UTF-8.) The configure.ac script no longer substitutes the
631            relevant variables.
632    
633        (b) The debugging options /B and /D in pcretest show the compiled bytecode
634            with length and offset values. This means that the output is different
635            for different internal link sizes. Test 2 is skipped for link sizes
636            other than 2 because of this, bypassing the problem. Unfortunately,
637            there was also a test in test 3 (the locale tests) that used /B and
638            failed for link sizes other than 2. Rather than cut the whole test out,
639            I have added a new /Z option to pcretest that replaces the length and
640            offset values with spaces. This is now used to make test 3 independent
641            of link size. (Test 2 will be tidied up later.)
642    
643    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
644        segmentation fault instead of returning the appropriate error message.
645    
646    15. In multiline mode when the newline sequence was set to "any", the pattern
647        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
648        This doesn't seem right; it now treats the CRLF combination as the line
649        ending, and so does not match in that case. It's only a pattern such as ^$
650        that would hit this one: something like ^ABC$ would have failed after \r
651        and then tried again after \r\n.
652    
653    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
654        in an attempt to make files that differ only in their line terminators
655        compare equal. This works on Linux.
656    
657    17. Under certain error circumstances pcregrep might try to free random memory
658        as it exited. This is now fixed, thanks to valgrind.
659    
660    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
661        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
662        was because its rules for how to advance for /g after matching an empty
663        string at the end of a line did not allow for this case. They now check for
664        it specially.
665    
666    20. pcretest is supposed to handle patterns and data of any length, by
667        extending its buffers when necessary. It was getting this wrong when the
668        buffer for a data line had to be extended.
669    
670    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
671        CRLF as a newline sequence.
672    
673    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
674        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
675        I have nevertheless tidied it up.
676    
677    23. Added some casts to kill warnings from HP-UX ia64 compiler.
678    
679    24. Added a man page for pcre-config.
680    
681    
682    Version 7.0 19-Dec-06
683    ---------------------
684    
685     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
686        moving to gcc 4.1.1.
687    
688     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
689        sys/time.h, which is documented as needed for this function. It doesn't
690        seem to matter on Linux, but it showed up on some releases of OS X.
691    
692     3. It seems that there are systems where bytes whose values are greater than
693        127 match isprint() in the "C" locale. The "C" locale should be the
694        default when a C program starts up. In most systems, only ASCII printing
695        characters match isprint(). This difference caused the output from pcretest
696        to vary, making some of the tests fail. I have changed pcretest so that:
697    
698        (a) When it is outputting text in the compiled version of a pattern, bytes
699            other than 32-126 are always shown as hex escapes.
700    
701        (b) When it is outputting text that is a matched part of a subject string,
702            it does the same, unless a different locale has been set for the match
703            (using the /L modifier). In this case, it uses isprint() to decide.
704    
705     4. Fixed a major bug that caused incorrect computation of the amount of memory
706        required for a compiled pattern when options that changed within the
707        pattern affected the logic of the preliminary scan that determines the
708        length. The relevant options are -x, and -i in UTF-8 mode. The result was
709        that the computed length was too small. The symptoms of this bug were
710        either the PCRE error "internal error: code overflow" from pcre_compile(),
711        or a glibc crash with a message such as "pcretest: free(): invalid next
712        size (fast)". Examples of patterns that provoked this bug (shown in
713        pcretest format) are:
714    
715          /(?-x: )/x
716          /(?x)(?-x: \s*#\s*)/
717          /((?i)[\x{c0}])/8
718          /(?i:[\x{c0}])/8
719    
720        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
721        is now done differently.
722    
723     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
724        wrapper classes; (b) implement a new function in the C++ scanner that is
725        more efficient than the old way of doing things because it avoids levels of
726        recursion in the regex matching; (c) add a paragraph to the documentation
727        for the FullMatch() function.
728    
729     6. The escape sequence \n was being treated as whatever was defined as
730        "newline". Not only was this contrary to the documentation, which states
731        that \n is character 10 (hex 0A), but it also went horribly wrong when
732        "newline" was defined as CRLF. This has been fixed.
733    
734     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
735        was being set to -1 for the "end of line" case (supposedly a value that no
736        character can have). Though this value is never used (the check for end of
737        line is "zero bytes in current character"), it caused compiler complaints.
738        I've changed it to 0xffffffff.
739    
740     8. In pcre_version.c, the version string was being built by a sequence of
741        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
742        string (as it is for production releases) called a macro with an empty
743        argument. The C standard says the result of this is undefined. The gcc
744        compiler treats it as an empty string (which was what was wanted) but it is
745        reported that Visual C gives an error. The source has been hacked around to
746        avoid this problem.
747    
748     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
749        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
750        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
751        of them did).
752    
753    10. Originally, pcretest opened its input and output without "b"; then I was
754        told that "b" was needed in some environments, so it was added for release
755        5.0 to both the input and output. (It makes no difference on Unix-like
756        systems.) Later I was told that it is wrong for the input on Windows. I've
757        now abstracted the modes into two macros, to make it easier to fiddle with
758        them, and removed "b" from the input mode under Windows.
759    
760    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
761    
762    12. Added -help and --help to pcretest as an official way of being reminded
763        of the options.
764    
765    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
766        and pcrecpp.cc because they annoy compilers at high warning levels.
767    
768    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
769    
770    15. Fixed an occurrence of == in configure.ac that should have been = (shell
771        scripts are not C programs :-) and which was not noticed because it works
772        on Linux.
773    
774    16. pcretest is supposed to handle any length of pattern and data line (as one
775        line or as a continued sequence of lines) by extending its input buffer if
776        necessary. This feature was broken for very long pattern lines, leading to
777        a string of junk being passed to pcre_compile() if the pattern was longer
778        than about 50K.
779    
780    17. I have done a major re-factoring of the way pcre_compile() computes the
781        amount of memory needed for a compiled pattern. Previously, there was code
782        that made a preliminary scan of the pattern in order to do this. That was
783        OK when PCRE was new, but as the facilities have expanded, it has become
784        harder and harder to keep it in step with the real compile phase, and there
785        have been a number of bugs (see for example, 4 above). I have now found a
786        cunning way of running the real compile function in a "fake" mode that
787        enables it to compute how much memory it would need, while actually only
788        ever using a few hundred bytes of working memory and without too many
789        tests of the mode. This should make future maintenance and development
790        easier. A side effect of this work is that the limit of 200 on the nesting
791        depth of parentheses has been removed (though this was never a serious
792        limitation, I suspect). However, there is a downside: pcre_compile() now
793        runs more slowly than before (30% or more, depending on the pattern). I
794        hope this isn't a big issue. There is no effect on runtime performance.
795    
796    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
797        newline (only possible for the last line of a file) and it was a
798        pattern that set a locale (followed by /Lsomething), pcretest crashed.
799    
800    19. Added additional timing features to pcretest. (1) The -tm option now times
801        matching only, not compiling. (2) Both -t and -tm can be followed, as a
802        separate command line item, by a number that specifies the number of
803        repeats to use when timing. The default is 50000; this gives better
804        precision, but takes uncomfortably long for very large patterns.
805    
806    20. Extended pcre_study() to be more clever in cases where a branch of a
807        subpattern has no definite first character. For example, (a*|b*)[cd] would
808        previously give no result from pcre_study(). Now it recognizes that the
809        first character must be a, b, c, or d.
810    
811    21. There was an incorrect error "recursive call could loop indefinitely" if
812        a subpattern (or the entire pattern) that was being tested for matching an
813        empty string contained only one non-empty item after a nested subpattern.
814        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
815        incorrectly, because the \d was being skipped in the check.
816    
817    22. The pcretest program now has a new pattern option /B and a command line
818        option -b, which is equivalent to adding /B to every pattern. This causes
819        it to show the compiled bytecode, without the additional information that
820        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
821        is the same as /B/I).
822    
823    23. A new optimization is now able automatically to treat some sequences such
824        as a*b as a*+b. More specifically, if something simple (such as a character
825        or a simple class like \d) has an unlimited quantifier, and is followed by
826        something that cannot possibly match the quantified thing, the quantifier
827        is automatically "possessified".
828    
829    24. A recursive reference to a subpattern whose number was greater than 39
830        went wrong under certain circumstances in UTF-8 mode. This bug could also
831        have affected the operation of pcre_study().
832    
833    25. Realized that a little bit of performance could be had by replacing
834        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
835    
836    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
837    
838    27. Possessive quantifiers such as a++ were previously implemented by turning
839        them into atomic groups such as ($>a+). Now they have their own opcodes,
840        which improves performance. This includes the automatically created ones
841        from 23 above.
842    
843    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
844        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
845        the first matched character to be a colon. This applied both to named and
846        numbered groups.
847    
848    29. The ucpinternal.h header file was missing its idempotency #ifdef.
849    
850    30. I was sent a "project" file called libpcre.a.dev which I understand makes
851        building PCRE on Windows easier, so I have included it in the distribution.
852    
853    31. There is now a check in pcretest against a ridiculously large number being
854        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
855        loop, the loop is abandoned.
856    
857    32. Forward references to subpatterns in conditions such as (?(2)...) where
858        subpattern 2 is defined later cause pcre_compile() to search forwards in
859        the pattern for the relevant set of parentheses. This search went wrong
860        when there were unescaped parentheses in a character class, parentheses
861        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
862    
863    33. "Subroutine" calls and backreferences were previously restricted to
864        referencing subpatterns earlier in the regex. This restriction has now
865        been removed.
866    
867    34. Added a number of extra features that are going to be in Perl 5.10. On the
868        whole, these are just syntactic alternatives for features that PCRE had
869        previously implemented using the Python syntax or my own invention. The
870        other formats are all retained for compatibility.
871    
872        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
873            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
874            also .NET compatible.
875    
876        (b) A recursion or subroutine call to a named group can now be defined as
877            (?&name) as well as (?P>name).
878    
879        (c) A backreference to a named group can now be defined as \k<name> or
880            \k'name' as well as (?P=name). The new forms, as well as being in Perl
881            5.10, are also .NET compatible.
882    
883        (d) A conditional reference to a named group can now use the syntax
884            (?(<name>) or (?('name') as well as (?(name).
885    
886        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
887            groups (named and numbered) that are never evaluated inline, but can be
888            called as "subroutines" from elsewhere. In effect, the DEFINE condition
889            is always false. There may be only one alternative in such a group.
890    
891        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
892            as the simple (?(R). The condition is true only if the most recent
893            recursion is that of the given number or name. It does not search out
894            through the entire recursion stack.
895    
896        (g) The escape \gN or \g{N} has been added, where N is a positive or
897            negative number, specifying an absolute or relative reference.
898    
899    35. Tidied to get rid of some further signed/unsigned compiler warnings and
900        some "unreachable code" warnings.
901    
902    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
903        things, this adds five new scripts.
904    
905    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
906        There were also incompatibilities regarding the handling of \Q..\E inside
907        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
908        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
909    
910    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
911        matches an empty string, and forcibly breaks the loop. There were bugs in
912        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
913        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
914        separate and independent bugs (that affected different cases) have been
915        fixed.
916    
917    39. Refactored the code to abolish the use of different opcodes for small
918        capturing bracket numbers. This is a tidy that I avoided doing when I
919        removed the limit on the number of capturing brackets for 3.5 back in 2001.
920        The new approach is not only tidier, it makes it possible to reduce the
921        memory needed to fix the previous bug (38).
922    
923    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
924        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
925        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
926        mode.
927    
928    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
929        report.
930    
931    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
932        copy construction and assignment in the C++ wrapper.
933    
934    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
935        couple of bugs that could have given wrong results in the "--newline=crlf"
936        case.
937    
938    44. Added a number of casts and did some reorganization of signed/unsigned int
939        variables following suggestions from Dair Grant. Also renamed the variable
940        "this" as "item" because it is a C++ keyword.
941    
942    45. Arranged for dftables to add
943    
944          #include "pcre_internal.h"
945    
946        to pcre_chartables.c because without it, gcc 4.x may remove the array
947        definition from the final binary if PCRE is built into a static library and
948        dead code stripping is activated.
949    
950    46. For an unanchored pattern, if a match attempt fails at the start of a
951        newline sequence, and the newline setting is CRLF or ANY, and the next two
952        characters are CRLF, advance by two characters instead of one.
953    
954    
955    Version 6.7 04-Jul-06
956    ---------------------
957    
958     1. In order to handle tests when input lines are enormously long, pcretest has
959        been re-factored so that it automatically extends its buffers when
960        necessary. The code is crude, but this _is_ just a test program. The
961        default size has been increased from 32K to 50K.
962    
963     2. The code in pcre_study() was using the value of the re argument before
964        testing it for NULL. (Of course, in any sensible call of the function, it
965        won't be NULL.)
966    
967     3. The memmove() emulation function in pcre_internal.h, which is used on
968        systems that lack both memmove() and bcopy() - that is, hardly ever -
969        was missing a "static" storage class specifier.
970    
971     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
972        containing an extended class (one that cannot be represented by a bitmap
973        because it contains high-valued characters or Unicode property items, e.g.
974        [\pZ]). Almost always one would set UTF-8 mode when processing such a
975        pattern, but PCRE should not loop if you do not (it no longer does).
976        [Detail: two cases were found: (a) a repeated subpattern containing an
977        extended class; (b) a recursive reference to a subpattern that followed a
978        previous extended class. It wasn't skipping over the extended class
979        correctly when UTF-8 mode was not set.]
980    
981     5. A negated single-character class was not being recognized as fixed-length
982        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
983        compile error "lookbehind assertion is not fixed length".
984    
985     6. The RunPerlTest auxiliary script was showing an unexpected difference
986        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
987        write a Perl script that can interpret lines of an input file either as
988        byte characters or as UTF-8, which is what "perltest" was being required to
989        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
990        can't do is switch easily at run time between having the "use utf8;" pragma
991        or not. In the end, I fudged it by using the RunPerlTest script to insert
992        "use utf8;" explicitly for the UTF-8 tests.
993    
994     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
995        the end of the subject string, contrary to the documentation and to what
996        Perl does. This was true of both matching functions. Now it matches only at
997        the start of the subject and immediately after *internal* newlines.
998    
999     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1000        a pointer to an int instead of a pointer to an unsigned long int. This
1001        caused problems on 64-bit systems.
1002    
1003     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1004        instance of the 'standard' template library not being so standard".
1005    
1006    10. There was no check on the number of named subpatterns nor the maximum
1007        length of a subpattern name. The product of these values is used to compute
1008        the size of the memory block for a compiled pattern. By supplying a very
1009        long subpattern name and a large number of named subpatterns, the size
1010        computation could be caused to overflow. This is now prevented by limiting
1011        the length of names to 32 characters, and the number of named subpatterns
1012        to 10,000.
1013    
1014    11. Subpatterns that are repeated with specific counts have to be replicated in
1015        the compiled pattern. The size of memory for this was computed from the
1016        length of the subpattern and the repeat count. The latter is limited to
1017        65535, but there was no limit on the former, meaning that integer overflow
1018        could in principle occur. The compiled length of a repeated subpattern is
1019        now limited to 30,000 bytes in order to prevent this.
1020    
1021    12. Added the optional facility to have named substrings with the same name.
1022    
1023    13. Added the ability to use a named substring as a condition, using the
1024        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1025        are numbers (not recommended). Forward references are permitted.
1026    
1027    14. Added forward references in named backreferences (if you see what I mean).
1028    
1029    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1030        pattern could run off the end of the subject. For example, the pattern
1031        "(?s)(.{1,5})"8 did this with the subject "ab".
1032    
1033    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1034        PCRE_CASELESS was set when matching characters that were quantified with ?
1035        or *.
1036    
1037    17. A character class other than a single negated character that had a minimum
1038        but no maximum quantifier - for example [ab]{6,} - was not handled
1039        correctly by pce_dfa_exec(). It would match only one character.
1040    
1041    18. A valid (though odd) pattern that looked like a POSIX character
1042        class but used an invalid character after [ (for example [[,abc,]]) caused
1043        pcre_compile() to give the error "Failed: internal error: code overflow" or
1044        in some cases to crash with a glibc free() error. This could even happen if
1045        the pattern terminated after [[ but there just happened to be a sequence of
1046        letters, a binary zero, and a closing ] in the memory that followed.
1047    
1048    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1049        over the years. Originally (before any Unicode support), just the bottom 8
1050        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1051        output from "man perlunicode" includes this:
1052    
1053          The regular expression compiler produces polymorphic opcodes.  That
1054          is, the pattern adapts to the data and automatically switches to
1055          the Unicode character scheme when presented with Unicode data--or
1056          instead uses a traditional byte scheme when presented with byte
1057          data.
1058    
1059        Sadly, a wide octal escape does not cause a switch, and in a string with
1060        no other multibyte characters, these octal escapes are treated as before.
1061        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1062        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1063        Unicode string.
1064    
1065        I have not perpetrated such confusion in PCRE. Up till now, it took just
1066        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1067        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1068        translate to the appropriate multibyte character.
1069    
1070    29. Applied some refactoring to reduce the number of warnings from Microsoft
1071        and Borland compilers. This has included removing the fudge introduced
1072        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1073        a warning about an unused variable.
1074    
1075    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1076        characters since release 4.0, because Perl (from release 5.004) does not.
1077        [Or at least, is documented not to: some releases seem to be in conflict
1078        with the documentation.] However, when a pattern was studied with
1079        pcre_study() and all its branches started with \s, PCRE still included VT
1080        as a possible starting character. Of course, this did no harm; it just
1081        caused an unnecessary match attempt.
1082    
1083    22. Removed a now-redundant internal flag bit that recorded the fact that case
1084        dependency changed within the pattern. This was once needed for "required
1085        byte" processing, but is no longer used. This recovers a now-scarce options
1086        bit. Also moved the least significant internal flag bit to the most-
1087        significant bit of the word, which was not previously used (hangover from
1088        the days when it was an int rather than a uint) to free up another bit for
1089        the future.
1090    
1091    23. Added support for CRLF line endings as well as CR and LF. As well as the
1092        default being selectable at build time, it can now be changed at runtime
1093        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1094        specify that it is scanning data with non-default line endings.
1095    
1096    24. Changed the definition of CXXLINK to make it agree with the definition of
1097        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1098    
1099    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1100        recursions. This makes a big different to stack usage for some patterns.
1101    
1102    26. If a subpattern containing a named recursion or subroutine reference such
1103        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1104        the space required for the compiled pattern went wrong and gave too small a
1105        value. Depending on the environment, this could lead to "Failed: internal
1106        error: code overflow at offset 49" or "glibc detected double free or
1107        corruption" errors.
1108    
1109    27. Applied patches from Google (a) to support the new newline modes and (b) to
1110        advance over multibyte UTF-8 characters in GlobalReplace.
1111    
1112    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1113        difference for some implementation of PCRE in some Windows version.
1114    
1115    29. Added some extra testing facilities to pcretest:
1116    
1117        \q<number>   in a data line sets the "match limit" value
1118        \Q<number>   in a data line sets the "match recursion limt" value
1119        -S <number>  sets the stack size, where <number> is in megabytes
1120    
1121        The -S option isn't available for Windows.
1122    
1123    
1124    Version 6.6 06-Feb-06
1125    ---------------------
1126    
1127     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1128        in pcreposix.h. I have copied the definition from pcre.h.
1129    
1130     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1131        because pcre.h is no longer a built file.
1132    
1133     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1134        not normally included in the compiled code.
1135    
1136    
1137    Version 6.5 01-Feb-06
1138    ---------------------
1139    
1140     1. When using the partial match feature with pcre_dfa_exec(), it was not
1141        anchoring the second and subsequent partial matches at the new starting
1142        point. This could lead to incorrect results. For example, with the pattern
1143        /1234/, partially matching against "123" and then "a4" gave a match.
1144    
1145     2. Changes to pcregrep:
1146    
1147        (a) All non-match returns from pcre_exec() were being treated as failures
1148            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1149            error message is output. Some extra information is given for the
1150            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1151            probably the only errors that are likely to be caused by users (by
1152            specifying a regex that has nested indefinite repeats, for instance).
1153            If there are more than 20 of these errors, pcregrep is abandoned.
1154    
1155        (b) A binary zero was treated as data while matching, but terminated the
1156            output line if it was written out. This has been fixed: binary zeroes
1157            are now no different to any other data bytes.
1158    
1159        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1160            used to set a locale for matching. The --locale=xxxx long option has
1161            been added (no short equivalent) to specify a locale explicitly on the
1162            pcregrep command, overriding the environment variables.
1163    
1164        (d) When -B was used with -n, some line numbers in the output were one less
1165            than they should have been.
1166    
1167        (e) Added the -o (--only-matching) option.
1168    
1169        (f) If -A or -C was used with -c (count only), some lines of context were
1170            accidentally printed for the final match.
1171    
1172        (g) Added the -H (--with-filename) option.
1173    
1174        (h) The combination of options -rh failed to suppress file names for files
1175            that were found from directory arguments.
1176    
1177        (i) Added the -D (--devices) and -d (--directories) options.
1178    
1179        (j) Added the -F (--fixed-strings) option.
1180    
1181        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1182    
1183        (l) Added the --colo(u)r option.
1184    
1185        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1186            is not present by default.
1187    
1188     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1189        items such as (?R) or (?1), when the recursion could match a number of
1190        alternatives. If it matched one of the alternatives, but subsequently,
1191        outside the recursion, there was a failure, the code tried to back up into
1192        the recursion. However, because of the way PCRE is implemented, this is not
1193        possible, and the result was an incorrect result from the match.
1194    
1195        In order to prevent this happening, the specification of recursion has
1196        been changed so that all such subpatterns are automatically treated as
1197        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1198    
1199     4. I had overlooked the fact that, in some locales, there are characters for
1200        which isalpha() is true but neither isupper() nor islower() are true. In
1201        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1202        and ordfeminine) are like this. This affected the treatment of \w and \W
1203        when they appeared in character classes, but not when they appeared outside
1204        a character class. The bit map for "word" characters is now created
1205        separately from the results of isalnum() instead of just taking it from the
1206        upper, lower, and digit maps. (Plus the underscore character, of course.)
1207    
1208     5. The above bug also affected the handling of POSIX character classes such as
1209        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1210        permanent tables. Instead, the bit maps for such a class were previously
1211        created as the appropriate unions of the upper, lower, and digit bitmaps.
1212        Now they are created by subtraction from the [[:word:]] class, which has
1213        its own bitmap.
1214    
1215     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1216        It is created by subtracting the vertical space characters (\x09, \x0a,
1217        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1218        subtraction was done in the overall bitmap for a character class, meaning
1219        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1220        be recognized. This bug has been fixed.
1221    
1222     7. Patches from the folks at Google:
1223    
1224          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1225          real life, but is still worth protecting against".
1226    
1227          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1228          regular expressions".
1229    
1230          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1231          have it.
1232    
1233          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1234          "configure" and the latter not, in order to fix a problem somebody had
1235          with compiling the Arg class on HP-UX.
1236    
1237          (e) Improve the error-handling of the C++ wrapper a little bit.
1238    
1239          (f) New tests for checking recursion limiting.
1240    
1241     8. The pcre_memmove() function, which is used only if the environment does not
1242        have a standard memmove() function (and is therefore rarely compiled),
1243        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1244        returning a result (though PCRE never actually uses the result).
1245    
1246     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1247        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1248        returned instead of calling malloc() with an overflowing number that would
1249        most likely cause subsequent chaos.
1250    
1251    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1252    
1253    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1254        with this option is matched, the nmatch and pmatch options of regexec() are
1255        ignored.
1256    
1257    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1258        provided in case anyone wants to the the POSIX interface with UTF-8
1259        strings.
1260    
1261    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1262        C++ linking (needed for some HP-UX environments).
1263    
1264    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1265        (unused parameter) and in the pcre_printint() function (omitted "default"
1266        switch label when the default is to do nothing).
1267    
1268    15. Added some code to make it possible, when PCRE is compiled as a C++
1269        library, to replace subject pointers for pcre_exec() with a smart pointer
1270        class, thus making it possible to process discontinuous strings.
1271    
1272    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1273        much the same function. They were added by different people who were trying
1274        to make PCRE easy to compile on non-Unix systems. It has been suggested
1275        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1276        for compiling on Windows systems. I have therefore replaced it with
1277        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1278        defaults to "extern" for C or "extern C" for C++, which works fine on
1279        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1280        SCOPE with something explicit in config.h. In addition:
1281    
1282        (a) pcreposix.h still had just "extern" instead of either of these macros;
1283            I have replaced it with PCRE_DATA_SCOPE.
1284    
1285        (b) Functions such as _pcre_xclass(), which are internal to the library,
1286            but external in the C sense, all had PCRE_EXPORT in their definitions.
1287            This is apparently wrong for the Windows case, so I have removed it.
1288            (It makes no difference on Unix-like systems.)
1289    
1290    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1291        of recursive calls to match(). This is different to MATCH_LIMIT because
1292        that limits the total number of calls to match(), not all of which increase
1293        the depth of recursion. Limiting the recursion depth limits the amount of
1294        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1295        when PCRE is compiled, and changed at run time. A patch from Google adds
1296        this functionality to the C++ interface.
1297    
1298    18. Changes to the handling of Unicode character properties:
1299    
1300        (a) Updated the table to Unicode 4.1.0.
1301    
1302        (b) Recognize characters that are not in the table as "Cn" (undefined).
1303    
1304        (c) I revised the way the table is implemented to a much improved format
1305            which includes recognition of ranges. It now supports the ranges that
1306            are defined in UnicodeData.txt, and it also amalgamates other
1307            characters into ranges. This has reduced the number of entries in the
1308            table from around 16,000 to around 3,000, thus reducing its size
1309            considerably. I realized I did not need to use a tree structure after
1310            all - a binary chop search is just as efficient. Having reduced the
1311            number of entries, I extended their size from 6 bytes to 8 bytes to
1312            allow for more data.
1313    
1314        (d) Added support for Unicode script names via properties such as \p{Han}.
1315    
1316    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1317        matching that character.
1318    
1319    20. When matching a repeated Unicode property with a minimum greater than zero,
1320        (for example \pL{2,}), PCRE could look past the end of the subject if it
1321        reached it while seeking the minimum number of characters. This could
1322        happen only if some of the characters were more than one byte long, because
1323        there is a check for at least the minimum number of bytes.
1324    
1325    21. Refactored the implementation of \p and \P so as to be more general, to
1326        allow for more different types of property in future. This has changed the
1327        compiled form incompatibly. Anybody with saved compiled patterns that use
1328        \p or \P will have to recompile them.
1329    
1330    22. Added "Any" and "L&" to the supported property types.
1331    
1332    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1333        but give a compile time error if the value is greater than 0xff.
1334    
1335    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1336        accidentally not being installed or uninstalled.
1337    
1338    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1339        made were to insert the current release number. This seemed silly, because
1340        it made things harder for people building PCRE on systems that don't run
1341        "configure". I have turned pcre.h into a distributed file, no longer built
1342        by "configure", with the version identification directly included. There is
1343        no longer a pcre.h.in file.
1344    
1345        However, this change necessitated a change to the pcre-config script as
1346        well. It is built from pcre-config.in, and one of the substitutions was the
1347        release number. I have updated configure.ac so that ./configure now finds
1348        the release number by grepping pcre.h.
1349    
1350    26. Added the ability to run the tests under valgrind.
1351    
1352    
1353    Version 6.4 05-Sep-05
1354    ---------------------
1355    
1356     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1357        "--" to be printed when multiple files were scanned, even when none of the
1358        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1359        consider it to be a bug, and have restored the previous behaviour.
1360    
1361     2. A couple of code tidies to get rid of compiler warnings.
1362    
1363     3. The pcretest program used to cheat by referring to symbols in the library
1364        whose names begin with _pcre_. These are internal symbols that are not
1365        really supposed to be visible externally, and in some environments it is
1366        possible to suppress them. The cheating is now confined to including
1367        certain files from the library's source, which is a bit cleaner.
1368    
1369     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1370        file's purpose clearer.
1371    
1372     5. Reorganized pcre_ucp_findchar().
1373    
1374    
1375    Version 6.3 15-Aug-05
1376    ---------------------
1377    
1378     1. The file libpcre.pc.in did not have general read permission in the tarball.
1379    
1380     2. There were some problems when building without C++ support:
1381    
1382        (a) If C++ support was not built, "make install" and "make test" still
1383            tried to test it.
1384    
1385        (b) There were problems when the value of CXX was explicitly set. Some
1386            changes have been made to try to fix these, and ...
1387    
1388        (c) --disable-cpp can now be used to explicitly disable C++ support.
1389    
1390        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1391            backslash in a target when C++ was disabled. This confuses some
1392            versions of "make", apparently. Using an intermediate variable solves
1393            this. (Same for CPP_LOBJ.)
1394    
1395     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1396        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1397        necessary on certain architectures.
1398    
1399     4. Added a setting of -export-symbols-regex to the link command to remove
1400        those symbols that are exported in the C sense, but actually are local
1401        within the library, and not documented. Their names all begin with
1402        "_pcre_". This is not a perfect job, because (a) we have to except some
1403        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1404        available (and never for static libraries). I have made a note to try to
1405        find a way round (a) in the future.
1406    
1407    
1408    Version 6.2 01-Aug-05
1409    ---------------------
1410    
1411     1. There was no test for integer overflow of quantifier values. A construction
1412        such as {1111111111111111} would give undefined results. What is worse, if
1413        a minimum quantifier for a parenthesized subpattern overflowed and became
1414        negative, the calculation of the memory size went wrong. This could have
1415        led to memory overwriting.
1416    
1417     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1418    
1419     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1420        operating environments where this matters.
1421    
1422     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1423        PCRE options from within the C++ wrapper.
1424    
1425     5. Named capturing subpatterns were not being correctly counted when a pattern
1426        was compiled. This caused two problems: (a) If there were more than 100
1427        such subpatterns, the calculation of the memory needed for the whole
1428        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1429        back references of the form \12, where the number was greater than 9, were
1430        not recognized as back references, even though there were sufficient
1431        previous subpatterns.
1432    
1433     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1434        versions of gcc, e.g. 2.95.4.
1435    
1436    
1437    Version 6.1 21-Jun-05
1438    ---------------------
1439    
1440     1. There was one reference to the variable "posix" in pcretest.c that was not
1441        surrounded by "#if !defined NOPOSIX".
1442    
1443     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1444        the cross-check on the old pcre_info() function, for the benefit of the
1445        cut-down version of PCRE that is currently imported into Exim.
1446    
1447     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1448        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1449        patterns that start (?i) but I don't think that matters. The use of (?i) is
1450        just an example; this all applies to the other options as well.
1451    
1452     4. Since libtool seems to echo the compile commands it is issuing, the output
1453        from "make" can be reduced a bit by putting "@" in front of each libtool
1454        compile command.
1455    
1456     5. Patch from the folks at Google for configure.in to be a bit more thorough
1457        in checking for a suitable C++ installation before trying to compile the
1458        C++ stuff. This should fix a reported problem when a compiler was present,
1459        but no suitable headers.
1460    
1461     6. The man pages all had just "PCRE" as their title. I have changed them to
1462        be the relevant file name. I have also arranged that these names are
1463        retained in the file doc/pcre.txt, which is a concatenation in text format
1464        of all the man pages except the little individual ones for each function.
1465    
1466     7. The NON-UNIX-USE file had not been updated for the different set of source
1467        files that come with release 6. I also added a few comments about the C++
1468        wrapper.
1469    
1470    
1471    Version 6.0 07-Jun-05
1472    ---------------------
1473    
1474     1. Some minor internal re-organization to help with my DFA experiments.
1475    
1476     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1477        didn't matter for the library itself when fully configured, but did matter
1478        when compiling without UCP support, or within Exim, where the ucp files are
1479        not imported.
1480    
1481     3. Refactoring of the library code to split up the various functions into
1482        different source modules. The addition of the new DFA matching code (see
1483        below) to a single monolithic source would have made it really too
1484        unwieldy, quite apart from causing all the code to be include in a
1485        statically linked application, when only some functions are used. This is
1486        relevant even without the DFA addition now that patterns can be compiled in
1487        one application and matched in another.
1488    
1489        The downside of splitting up is that there have to be some external
1490        functions and data tables that are used internally in different modules of
1491        the library but which are not part of the API. These have all had their
1492        names changed to start with "_pcre_" so that they are unlikely to clash
1493        with other external names.
1494    
1495     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1496        a different (DFA) algorithm. Although it is slower than the original
1497        function, it does have some advantages for certain types of matching
1498        problem.
1499    
1500     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1501        including restarting after a partial match.
1502    
1503     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1504        defined when compiling for Windows was sent to me. I have put it into the
1505        code, though I have no means of testing or verifying it.
1506    
1507     7. Added the pcre_refcount() auxiliary function.
1508    
1509     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1510        match before or at the first newline in the subject string. In pcretest,
1511        the /f option on a pattern can be used to set this.
1512    
1513     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1514        would behave wrongly. This has been present in PCRE since release 4.0.
1515    
1516    10. A number of changes to the pcregrep command:
1517    
1518        (a) Refactored how -x works; insert ^(...)$ instead of setting
1519            PCRE_ANCHORED and checking the length, in preparation for adding
1520            something similar for -w.
1521    
1522        (b) Added the -w (match as a word) option.
1523    
1524        (c) Refactored the way lines are read and buffered so as to have more
1525            than one at a time available.
1526    
1527        (d) Implemented a pcregrep test script.
1528    
1529        (e) Added the -M (multiline match) option. This allows patterns to match
1530            over several lines of the subject. The buffering ensures that at least
1531            8K, or the rest of the document (whichever is the shorter) is available
1532            for matching (and similarly the previous 8K for lookbehind assertions).
1533    
1534        (f) Changed the --help output so that it now says
1535    
1536              -w, --word-regex(p)
1537    
1538            instead of two lines, one with "regex" and the other with "regexp"
1539            because that confused at least one person since the short forms are the
1540            same. (This required a bit of code, as the output is generated
1541            automatically from a table. It wasn't just a text change.)
1542    
1543        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1544            option but starts with a hyphen. Could be a pattern or a path name
1545            starting with a hyphen, for instance.
1546    
1547        (h) "-" can be given as a file name to represent stdin.
1548    
1549        (i) When file names are being printed, "(standard input)" is used for
1550            the standard input, for compatibility with GNU grep. Previously
1551            "<stdin>" was used.
1552    
1553        (j) The option --label=xxx can be used to supply a name to be used for
1554            stdin when file names are being printed. There is no short form.
1555    
1556        (k) Re-factored the options decoding logic because we are going to add
1557            two more options that take data. Such options can now be given in four
1558            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1559    
1560        (l) Added the -A, -B, and -C options for requesting that lines of context
1561            around matches be printed.
1562    
1563        (m) Added the -L option to print the names of files that do not contain
1564            any matching lines, that is, the complement of -l.
1565    
1566        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1567            continue to scan other files.
1568    
1569        (o) The -s option was incorrectly implemented. For compatibility with other
1570            greps, it now suppresses the error message for a non-existent or non-
1571            accessible file (but not the return code). There is a new option called
1572            -q that suppresses the output of matching lines, which was what -s was
1573            previously doing.
1574    
1575        (p) Added --include and --exclude options to specify files for inclusion
1576            and exclusion when recursing.
1577    
1578    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1579        Hopefully, it now does.
1580    
1581    12. Missing cast in pcre_study().
1582    
1583    13. Added an "uninstall" target to the makefile.
1584    
1585    14. Replaced "extern" in the function prototypes in Makefile.in with
1586        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1587        world, but is set differently for Windows.
1588    
1589    15. Added a second compiling function called pcre_compile2(). The only
1590        difference is that it has an extra argument, which is a pointer to an
1591        integer error code. When there is a compile-time failure, this is set
1592        non-zero, in addition to the error test pointer being set to point to an
1593        error message. The new argument may be NULL if no error number is required
1594        (but then you may as well call pcre_compile(), which is now just a
1595        wrapper). This facility is provided because some applications need a
1596        numeric error indication, but it has also enabled me to tidy up the way
1597        compile-time errors are handled in the POSIX wrapper.
1598    
1599    16. Added VPATH=.libs to the makefile; this should help when building with one
1600        prefix path and installing with another. (Or so I'm told by someone who
1601        knows more about this stuff than I do.)
1602    
1603    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1604        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1605        match everything, including newlines. This is not POSIX-compatible, but
1606        somebody wanted the feature. From pcretest it can be activated by using
1607        both the P and the s flags.
1608    
1609    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1610    
1611    19. libpcre.pc was being incorrectly installed as executable.
1612    
1613    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1614        it now also looks for '\r' so that it will work unmodified on Windows.
1615    
1616    21. Added Google's contributed C++ wrapper to the distribution.
1617    
1618    22. Added some untidy missing memory free() calls in pcretest, to keep
1619        Electric Fence happy when testing.
1620    
1621    
1622    
1623    Version 5.0 13-Sep-04
1624    ---------------------
1625    
1626     1. Internal change: literal characters are no longer packed up into items
1627        containing multiple characters in a single byte-string. Each character
1628        is now matched using a separate opcode. However, there may be more than one
1629        byte in the character in UTF-8 mode.
1630    
1631     2. The pcre_callout_block structure has two new fields: pattern_position and
1632        next_item_length. These contain the offset in the pattern to the next match
1633        item, and its length, respectively.
1634    
1635     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1636        insertion of callouts before each pattern item. Added the /C option to
1637        pcretest to make use of this.
1638    
1639     4. On the advice of a Windows user, the lines
1640    
1641          #if defined(_WIN32) || defined(WIN32)
1642          _setmode( _fileno( stdout ), 0x8000 );
1643          #endif  /* defined(_WIN32) || defined(WIN32) */
1644    
1645        have been added to the source of pcretest. This apparently does useful
1646        magic in relation to line terminators.
1647    
1648     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1649        for the benefit of those environments where the "b" makes a difference.
1650    
1651     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1652        to know about it. I have put a hack into configure.in that adds in code
1653        to set GCC=yes if CC=icc. This seems to end up at a point in the
1654        generated configure script that is early enough to affect the setting of
1655        compiler options, which is what is needed, but I have no means of testing
1656        whether it really works. (The user who reported this had patched the
1657        generated configure script, which of course I cannot do.)
1658    
1659        LATER: After change 22 below (new libtool files), the configure script
1660        seems to know about icc (and also ecc). Therefore, I have commented out
1661        this hack in configure.in.
1662    
1663     7. Added support for pkg-config (2 patches were sent in).
1664    
1665     8. Negated POSIX character classes that used a combination of internal tables
1666        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1667        [[:^ascii]]. Typically, they would match almost any characters. The other
1668        POSIX classes were not broken in this way.
1669    
1670     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1671        to find the match, as PCRE was deluded into thinking that the match had to
1672        start at the start point or following a newline. The same bug applied to
1673        patterns with negative forward assertions or any backward assertions
1674        preceding ".*" at the start, unless the pattern required a fixed first
1675        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1676    
1677    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1678        starting at the last subject character, bytes beyond the end of the subject
1679        string were read.
1680    
1681    11. Renamed the variable "class" as "classbits" to make life easier for C++
1682        users. (Previously there was a macro definition, but it apparently wasn't
1683        enough.)
1684    
1685    12. Added the new field "tables" to the extra data so that tables can be passed
1686        in at exec time, or the internal tables can be re-selected. This allows
1687        a compiled regex to be saved and re-used at a later time by a different
1688        program that might have everything at different addresses.
1689    
1690    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1691        -R library as well as a -L library.
1692    
1693    14. The debugging options of pcretest (-d on the command line or D on a
1694        pattern) showed incorrect output for anything following an extended class
1695        that contained multibyte characters and which was followed by a quantifier.
1696    
1697    15. Added optional support for general category Unicode character properties
1698        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1699        support. It adds about 90K to the size of the library. The meanings of the
1700        inbuilt class escapes such as \d and \s have NOT been changed.
1701    
1702    16. Updated pcredemo.c to include calls to free() to release the memory for the
1703        compiled pattern.
1704    
1705    17. The generated file chartables.c was being created in the source directory
1706        instead of in the building directory. This caused the build to fail if the
1707        source directory was different from the building directory, and was
1708        read-only.
1709    
1710    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1711        file. No doubt somebody will tell me if they don't make sense... Also added
1712        Dan Mooney's comments about building on OpenVMS.
1713    
1714    19. Added support for partial matching via the PCRE_PARTIAL option for
1715        pcre_exec() and the \P data escape in pcretest.
1716    
1717    20. Extended pcretest with 3 new pattern features:
1718    
1719        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1720              write the compiled pattern to the file whose name is "rest-of-line".
1721              This is a straight binary dump of the data, with the saved pointer to
1722              the character tables forced to be NULL. The study data, if any, is
1723              written too. After writing, pcretest reads a new pattern.
1724    
1725        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1726              compiled pattern from the given file. There must not be any
1727              occurrences of "<" in the file name (pretty unlikely); if there are,
1728              pcretest will instead treat the initial "<" as a pattern delimiter.
1729              After reading in the pattern, pcretest goes on to read data lines as
1730              usual.
1731    
1732        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1733              and 16-bit fields in a compiled pattern, to simulate a pattern that
1734              was compiled on a host of opposite endianness.
1735    
1736    21. The pcre-exec() function can now cope with patterns that were compiled on
1737        hosts of opposite endianness, with this restriction:
1738    
1739          As for any compiled expression that is saved and used later, the tables
1740          pointer field cannot be preserved; the extra_data field in the arguments
1741          to pcre_exec() should be used to pass in a tables address if a value
1742          other than the default internal tables were used at compile time.
1743    
1744    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1745        now diagnosed as an error. Previously, most of the time, a negative number
1746        would have been treated as zero, but if in addition "ovector" was passed as
1747        NULL, a crash could occur.
1748    
1749    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1750        new versions from the libtool 1.5 distribution (the last one is a copy of
1751        a file called libtool.m4). This seems to have fixed the need to patch
1752        "configure" to support Darwin 1.3 (which I used to do). However, I still
1753        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1754        workstation).
1755    
1756    24. Changed the PCRE licence to be the more standard "BSD" licence.
1757    
1758    
1759    Version 4.5 01-Dec-03
1760    ---------------------
1761    
1762     1. There has been some re-arrangement of the code for the match() function so
1763        that it can be compiled in a version that does not call itself recursively.
1764        Instead, it keeps those local variables that need separate instances for
1765        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1766        needs to "recurse". Keeping track of where control must go is done by means
1767        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1768        hide most of the details from the main code, and operates only if
1769        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1770        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1771        operating.
1772    
1773        To make it easier for callers to provide specially tailored get/free
1774        functions for this usage, two new functions, pcre_stack_malloc, and
1775        pcre_stack_free, are used. They are always called in strict stacking order,
1776        and the size of block requested is always the same.
1777    
1778        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1779        PCRE has been compiled to use the stack or the heap for recursion. The
1780        -C option of pcretest uses this to show which version is compiled.
1781    
1782        A new data escape \S, is added to pcretest; it causes the amounts of store
1783        obtained and freed by both kinds of malloc/free at match time to be added
1784        to the output.
1785    
1786     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1787        what's available on my current Linux desktop machine.
1788    
1789     3. When matching a UTF-8 string, the test for a valid string at the start has
1790        been extended. If start_offset is not zero, PCRE now checks that it points
1791        to a byte that is the start of a UTF-8 character. If not, it returns
1792        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1793        this is necessary because there may be backward assertions in the pattern.
1794        When matching the same subject several times, it may save resources to use
1795        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1796    
1797     4. The code for checking the validity of UTF-8 strings has been tightened so
1798        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1799        containing "overlong sequences".
1800    
1801     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1802        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1803        should have been "&", but it just so happened that all the cases this let
1804        through by mistake were picked up later in the function.
1805    
1806     6. I had used a variable called "isblank" - this is a C99 function, causing
1807        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1808    
1809     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1810        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1811        so the version is shown; (c) stop "make test" echoing "./RunTest".
1812    
1813     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1814    
1815     9. The prototype for memmove() for systems that don't have it was using
1816        size_t, but the inclusion of the header that defines size_t was later. I've
1817        moved the #includes for the C headers earlier to avoid this.
1818    
1819    10. Added some adjustments to the code to make it easier to compiler on certain
1820        special systems:
1821    
1822          (a) Some "const" qualifiers were missing.
1823          (b) Added the macro EXPORT before all exported functions; by default this
1824              is defined to be empty.
1825          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1826              that it reads its output file name as an argument instead of writing
1827              to the standard output and assuming this can be redirected.
1828    
1829    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1830        class containing characters with values greater than 255, PCRE compilation
1831        went into a loop.
1832    
1833    12. A recursive reference to a subpattern that was within another subpattern
1834        that had a minimum quantifier of zero caused PCRE to crash. For example,
1835        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1836        recursion. If the recursively-called subpattern itself had a zero repeat,
1837        that was OK.
1838    
1839    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1840        buffer into which it was copied (for escape processing) was still set at
1841        1024, so long lines caused crashes.
1842    
1843    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1844        "internal error: code overflow...". This applied to any character class
1845        that was followed by a possessive quantifier.
1846    
1847    15. Modified the Makefile to add libpcre.la as a prerequisite for
1848        libpcreposix.la because I was told this is needed for a parallel build to
1849        work.
1850    
1851    16. If a pattern that contained .* following optional items at the start was
1852        studied, the wrong optimizing data was generated, leading to matching
1853        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1854        matching string must start with a or b or c. The correct conclusion for
1855        this pattern is that a match can start with any character.
1856    
1857    
1858    Version 4.4 13-Aug-03
1859    ---------------------
1860    
1861     1. In UTF-8 mode, a character class containing characters with values between
1862        127 and 255 was not handled correctly if the compiled pattern was studied.
1863        In fixing this, I have also improved the studying algorithm for such
1864        classes (slightly).
1865    
1866     2. Three internal functions had redundant arguments passed to them. Removal
1867        might give a very teeny performance improvement.
1868    
1869     3. Documentation bug: the value of the capture_top field in a callout is *one
1870        more than* the number of the hightest numbered captured substring.
1871    
1872     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1873        in incorrectly linking with a previously installed version. They now link
1874        explicitly with libpcre.la.
1875    
1876     5. configure.in no longer needs to recognize Cygwin specially.
1877    
1878     6. A problem in pcre.in for Windows platforms is fixed.
1879    
1880     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1881        pcretest, it used to include the size of the study block as part of its
1882        output. Unfortunately, the structure contains a field that has a different
1883        size on different hardware architectures. This meant that the tests that
1884        showed this size failed. As the block is currently always of a fixed size,
1885        this information isn't actually particularly useful in pcretest output, so
1886        I have just removed it.
1887    
1888     8. Three pre-processor statements accidentally did not start in column 1.
1889        Sadly, there are *still* compilers around that complain, even though
1890        standard C has not required this for well over a decade. Sigh.
1891    
1892     9. In pcretest, the code for checking callouts passed small integers in the
1893        callout_data field, which is a void * field. However, some picky compilers
1894        complained about the casts involved for this on 64-bit systems. Now
1895        pcretest passes the address of the small integer instead, which should get
1896        rid of the warnings.
1897    
1898    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1899        both compile and run time, and gives an error if an invalid UTF-8 sequence
1900        is found. There is a option for disabling this check in cases where the
1901        string is known to be correct and/or the maximum performance is wanted.
1902    
1903    11. In response to a bug report, I changed one line in Makefile.in from
1904    
1905            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1906        to
1907            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1908    
1909        to look similar to other lines, but I have no way of telling whether this
1910        is the right thing to do, as I do not use Windows. No doubt I'll get told
1911        if it's wrong...
1912    
1913    
1914    Version 4.3 21-May-03
1915    ---------------------
1916    
1917    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1918       Makefile.
1919    
1920    2. Some refactoring to improve the quality of the code:
1921    
1922       (i)   The utf8_table... variables are now declared "const".
1923    
1924       (ii)  The code for \cx, which used the "case flipping" table to upper case
1925             lower case letters, now just substracts 32. This is ASCII-specific,
1926             but the whole concept of \cx is ASCII-specific, so it seems
1927             reasonable.
1928    
1929       (iii) PCRE was using its character types table to recognize decimal and
1930             hexadecimal digits in the pattern. This is silly, because it handles
1931             only 0-9, a-f, and A-F, but the character types table is locale-
1932             specific, which means strange things might happen. A private
1933             table is now used for this - though it costs 256 bytes, a table is
1934             much faster than multiple explicit tests. Of course, the standard
1935             character types table is still used for matching digits in subject
1936             strings against \d.
1937    
1938       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1939             ending in _t are). So I've renamed it as ESC_tee.
1940    
1941    3. The first argument for regexec() in the POSIX wrapper should have been
1942       defined as "const".
1943    
1944    4. Changed pcretest to use malloc() for its buffers so that they can be
1945       Electric Fenced for debugging.
1946    
1947    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1948       to read one or more bytes before the start of the subject string. Often this
1949       had no effect on PCRE's behaviour, but in some circumstances it could
1950       provoke a segmentation fault.
1951    
1952    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1953       to try to read one or more bytes before the start of the subject string.
1954    
1955    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1956       UTF-8 support could misbehave in various ways if the subject string
1957       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1958       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1959       back over UTF-8 characters.)
1960    
1961    
1962    Version 4.2 14-Apr-03
1963    ---------------------
1964    
1965    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1966    
1967    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1968         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1969         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1970         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1971         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1972           and BUILD_EXEEXT
1973         Note: automatic setting of the BUILD variables is not yet working
1974         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1975           compile-time but not at link-time
1976         [LINK]: use for linking executables only
1977         make different versions for Windows and non-Windows
1978         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1979           libraries
1980         [LINK_FOR_BUILD]: new variable
1981         [OBJEXT]: use throughout
1982         [EXEEXT]: use throughout
1983         <winshared>: new target
1984         <wininstall>: new target
1985         <dftables.o>: use native compiler
1986         <dftables>: use native linker
1987         <install>: handle Windows platform correctly
1988         <clean>: ditto
1989         <check>: ditto
1990         copy DLL to top builddir before testing
1991    
1992       As part of these changes, -no-undefined was removed again. This was reported
1993       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1994       in any case.
1995    
1996    3. Some tidies to get rid of compiler warnings:
1997    
1998       . In the match_data structure, match_limit was an unsigned long int, whereas
1999         match_call_count was an int. I've made them both unsigned long ints.
2000    
2001       . In pcretest the fact that a const uschar * doesn't automatically cast to
2002         a void * provoked a warning.
2003    
2004       . Turning on some more compiler warnings threw up some "shadow" variables
2005         and a few more missing casts.
2006    
2007    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2008       option, a class that contained a single character with a value between 128
2009       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2010    
2011    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2012       option, a class that contained several characters, but with at least one
2013       whose value was between 128 and 255 caused PCRE to crash.
2014    
2015    
2016    Version 4.1 12-Mar-03
2017    ---------------------
2018    
2019    1. Compiling with gcc -pedantic found a couple of places where casts were
2020    needed, and a string in dftables.c that was longer than standard compilers are
2021    required to support.
2022    
2023    2. Compiling with Sun's compiler found a few more places where the code could
2024    be tidied up in order to avoid warnings.
2025    
2026    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2027    first of these names is deprecated in the latest Autoconf in favour of the name
2028    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2029    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2030    analogy I have changed it to CFLAGS_FOR_BUILD.
2031    
2032    4. Added -no-undefined to the linking command in the Makefile, because this is
2033    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2034    linking step for the pcreposix library.
2035    
2036    5. PCRE was failing to diagnose the case of two named groups with the same
2037    name.
2038    
2039    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2040    literal character that is needed in the subject for a match, and scans along to
2041    ensure that it is present before embarking on the full matching process. This
2042    saves time in cases of nested unlimited repeats that are never going to match.
2043    Problem: the scan can take a lot of time if the subject is very long (e.g.
2044    megabytes), thus penalizing straightforward matches. It is now done only if the
2045    amount of subject to be scanned is less than 1000 bytes.
2046    
2047    7. A lesser problem with the same optimization is that it was recording the
2048    first character of an anchored pattern as "needed", thus provoking a search
2049    right along the subject, even when the first match of the pattern was going to
2050    fail. The "needed" character is now not set for anchored patterns, unless it
2051    follows something in the pattern that is of non-fixed length. Thus, it still
2052    fulfils its original purpose of finding quick non-matches in cases of nested
2053    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2054    
2055    
2056    Version 4.0 17-Feb-03
2057    ---------------------
2058    
2059    1. If a comment in an extended regex that started immediately after a meta-item
2060    extended to the end of string, PCRE compiled incorrect data. This could lead to
2061    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2062    
2063    2. Moved to autoconf 2.53 and libtool 1.4.2.
2064    
2065    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2066    the special perltest8 script is no longer needed - all the tests can be run
2067    from a single perltest script.
2068    
2069    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2070    by \s. It has now been removed in PCRE. This means it isn't recognized as
2071    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2072    class [:space:] *does* include VT, thereby creating a mess.
2073    
2074    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2075    space and tab.
2076    
2077    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2078    its new features into the main test script, reducing the number of scripts.
2079    
2080    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2081    were backward compatible, and made the (?i) apply to the whole pattern, as if
2082    /i were given. Now it behaves more logically, and applies the option setting
2083    only to what follows. PCRE has been changed to follow suit. However, if it
2084    finds options settings right at the start of the pattern, it extracts them into
2085    the global options, as before. Thus, they show up in the info data.
2086    
2087    8. Added support for the \Q...\E escape sequence. Characters in between are
2088    treated as literals. This is slightly different from Perl in that $ and @ are
2089    also handled as literals inside the quotes. In Perl, they will cause variable
2090    interpolation. Note the following examples:
2091    
2092        Pattern            PCRE matches      Perl matches
2093    
2094        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2095        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2096        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2097    
2098    For compatibility with Perl, \Q...\E sequences are recognized inside character
2099    classes as well as outside them.
2100    
2101    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2102    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2103    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2104    signed/unsigned warnings.
2105    
2106    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2107    option for pcretest, so I've replaced it by a simple function that does just
2108    that job.
2109    
2110    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2111    "pcregrep -".
2112    
2113    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2114    Java package. This provides some syntactic sugar for simple cases of what my
2115    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2116    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2117    item, you can use this simplified notation. Note that only makes sense with
2118    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2119    greediness, whatever the setting of the PCRE_UNGREEDY option.
2120    
2121    13. A change of greediness default within a pattern was not taking effect at
2122    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2123    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2124    was abstracted outside.
2125    
2126    14. PCRE now supports the \G assertion. It is true when the current matching
2127    position is at the start point of the match. This differs from \A when the
2128    starting offset is non-zero. Used with the /g option of pcretest (or similar
2129    code), it works in the same way as it does for Perl's /g option. If all
2130    alternatives of a regex begin with \G, the expression is anchored to the start
2131    match position, and the "anchored" flag is set in the compiled expression.
2132    
2133    15. Some bugs concerning the handling of certain option changes within patterns
2134    have been fixed. These applied to options other than (?ims). For example,
2135    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2136    been the other way round. Some of this was related to change 7 above.
2137    
2138    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2139    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2140    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2141    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2142    
2143    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2144    mode. Unlike ".", it always matches newline, whatever the setting of
2145    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2146    assertions. Perl allows it, but it doesn't (in general) work because it can't
2147    calculate the length of the lookbehind. At least, that's the case for Perl
2148    5.8.0 - I've been told they are going to document that it doesn't work in
2149    future.
2150    
2151    18. Added an error diagnosis for escapes that PCRE does not support: these are
2152    \L, \l, \N, \P, \p, \U, \u, and \X.
2153    
2154    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2155    reading past the end of the pattern in cases such as /[abcd/.
2156    
2157    20. PCRE was getting more memory than necessary for patterns with classes that
2158    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2159    
2160    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2161    compiling PCRE for use with Virtual Pascal.
2162    
2163    22. Small fix to the Makefile to make it work properly if the build is done
2164    outside the source tree.
2165    
2166    23. Added a new extension: a condition to go with recursion. If a conditional
2167    subpattern starts with (?(R) the "true" branch is used if recursion has
2168    happened, whereas the "false" branch is used only at the top level.
2169    
2170    24. When there was a very long string of literal characters (over 255 bytes
2171    without UTF support, over 250 bytes with UTF support), the computation of how
2172    much memory was required could be incorrect, leading to segfaults or other
2173    strange effects.
2174    
2175    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2176    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2177    there was a subsequent back reference to those brackets. This meant that, for
2178    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2179    possible to check for precisely this case. All we can do is abandon the
2180    optimization if .* occurs inside capturing brackets when there are any back
2181    references whatsoever. (See below for a better fix that came later.)
2182    
2183    26. The handling of the optimization for finding the first character of a
2184    non-anchored pattern, and for finding a character that is required later in the
2185    match were failing in some cases. This didn't break the matching; it just
2186    failed to optimize when it could. The way this is done has been re-implemented.
2187    
2188    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2189    
2190    28. Added a new feature that provides some of the functionality that Perl
2191    provides with (?{...}). The facility is termed a "callout". The way it is done
2192    in PCRE is for the caller to provide an optional function, by setting
2193    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2194    global variable. By default it is unset, which disables all calling out. To get
2195    the function called, the regex must include (?C) at appropriate points. This
2196    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2197    This provides a means of identifying different callout points. When PCRE
2198    reaches such a point in the regex, if pcre_callout has been set, the external
2199    function is called. It is provided with data in a structure called
2200    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2201    matching continues; if it returns a non-zero value, the match at the current
2202    point fails. However, backtracking will occur if possible. [This was changed
2203    later and other features added - see item 49 below.]
2204    
2205    29. pcretest is upgraded to test the callout functionality. It provides a
2206    callout function that displays information. By default, it shows the start of
2207    the match and the current position in the text. There are some new data escapes
2208    to vary what happens:
2209    
2210        \C+         in addition, show current contents of captured substrings
2211        \C-         do not supply a callout function
2212        \C!n        return 1 when callout number n is reached
2213        \C!n!m      return 1 when callout number n is reached for the mth time
2214    
2215    30. If pcregrep was called with the -l option and just a single file name, it
2216    output "<stdin>" if a match was found, instead of the file name.
2217    
2218    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2219    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2220    pcre_exec(). This saves a malloc/free per call. The default value of
2221    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2222    when configuring.
2223    
2224    32. The default maximum size of a compiled pattern is 64K. There have been a
2225    few cases of people hitting this limit. The code now uses macros to handle the
2226    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2227    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2228    configuring. Tests 2 and 5 work only with 2-byte links because they output
2229    debugging information about compiled patterns.
2230    
2231    33. Internal code re-arrangements:
2232    
2233    (a) Moved the debugging function for printing out a compiled regex into
2234        its own source file (printint.c) and used #include to pull it into
2235        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2236        separate copies.
2237    
2238    (b) Defined the list of op-code names for debugging as a macro in
2239        internal.h so that it is next to the definition of the opcodes.
2240    
2241    (c) Defined a table of op-code lengths for simpler skipping along compiled
2242        code. This is again a macro in internal.h so that it is next to the
2243        definition of the opcodes.
2244    
2245    34. Added support for recursive calls to individual subpatterns, along the
2246    lines of Robin Houston's patch (but implemented somewhat differently).
2247    
2248    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2249    allow it to read and process whole directories in Win32. This code was
2250    contributed by Lionel Fourquaux; it has not been tested by me.
2251    
2252    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2253    used to name a group. Names consist of alphanumerics and underscores, and must
2254    be unique. Back references use the syntax (?P=name) and recursive calls use
2255    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2256    numbers. The function pcre_fullinfo() can be used after compilation to extract
2257    a name/number map. There are three relevant calls:
2258    
2259      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2260      PCRE_INFO_NAMECOUNT            yields the number of entries
2261      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2262    
2263    The map is a vector of fixed-size entries. The size of each entry depends on
2264    the length of the longest name used. The first two bytes of each entry are the
2265    group number, most significant byte first. There follows the corresponding
2266    name, zero terminated. The names are in alphabetical order.
2267    
2268    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2269    case instead of 255. Making it the same both with and without UTF-8 support
2270    means that the same test output works with both.
2271    
2272    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2273    calling malloc() with a zero argument.
2274    
2275    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2276    optimization. I've improved things by keeping a bitmap of backreferences with
2277    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2278    fact referenced, the optimization can be applied. It is unlikely that a
2279    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2280    the match to follow \n) will appear inside brackets with a number greater than
2281    31, but if it does, any back reference > 31 suppresses the optimization.
2282    
2283    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2284    of disabling numbered capturing parentheses. Any opening parenthesis that is
2285    not followed by ? behaves as if it were followed by ?: but named parentheses
2286    can still be used for capturing (and they will acquire numbers in the usual
2287    way).
2288    
2289    41. Redesigned the return codes from the match() function into yes/no/error so
2290    that errors can be passed back from deep inside the nested calls. A malloc
2291    failure while inside a recursive subpattern call now causes the
2292    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2293    
2294    42. It is now possible to set a limit on the number of times the match()
2295    function is called in a call to pcre_exec(). This facility makes it possible to
2296    limit the amount of recursion and backtracking, though not in a directly
2297    obvious way, because the match() function is used in a number of different
2298    circumstances. The count starts from zero for each position in the subject
2299    string (for non-anchored patterns). The default limit is, for compatibility, a
2300    large number, namely 10 000 000. You can change this in two ways:
2301    
2302    (a) When configuring PCRE before making, you can use --with-match-limit=n
2303        to set a default value for the compiled library.
2304    
2305    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2306        a different value is set. See 45 below.
2307    
2308    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2309    
2310    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2311    of things that can be changed at compile time. The first argument specifies
2312    what is wanted and the second points to where the information is to be placed.
2313    The current list of available information is:
2314    
2315      PCRE_CONFIG_UTF8
2316    
2317    The output is an integer that is set to one if UTF-8 support is available;
2318    otherwise it is set to zero.
2319    
2320      PCRE_CONFIG_NEWLINE
2321    
2322    The output is an integer that it set to the value of the code that is used for
2323    newline. It is either LF (10) or CR (13).
2324    
2325      PCRE_CONFIG_LINK_SIZE
2326    
2327    The output is an integer that contains the number of bytes used for internal
2328    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2329    
2330      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2331    
2332    The output is an integer that contains the threshold above which the POSIX
2333    interface uses malloc() for output vectors. See item 31 above.
2334    
2335      PCRE_CONFIG_MATCH_LIMIT
2336    
2337    The output is an unsigned integer that contains the default limit of the number
2338    of match() calls in a pcre_exec() execution. See 42 above.
2339    
2340    44. pcretest has been upgraded by the addition of the -C option. This causes it
2341    to extract all the available output from the new pcre_config() function, and to
2342    output it. The program then exits immediately.
2343    
2344    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2345    order to support additional features. One way would have been to define
2346    pcre_exec2() (for example) with extra arguments, but this would not have been
2347    extensible, and would also have required all calls to the original function to
2348    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2349    is used for passing in "extra" data from pcre_study().
2350    
2351    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2352    contains the following fields:
2353    
2354      flags         a bitmap indicating which of the following fields are set
2355      study_data    opaque data from pcre_study()
2356      match_limit   a way of specifying a limit on match() calls for a specific
2357                      call to pcre_exec()
2358      callout_data  data for callouts (see 49 below)
2359    
2360    The flag bits are also defined in pcre.h, and are
2361    
2362      PCRE_EXTRA_STUDY_DATA
2363      PCRE_EXTRA_MATCH_LIMIT
2364      PCRE_EXTRA_CALLOUT_DATA
2365    
2366    The pcre_study() function now returns one of these new pcre_extra blocks, with
2367    the actual study data pointed to by the study_data field, and the
2368    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2369    before. That is, this change is entirely upwards-compatible and requires no
2370    change to existing code.
2371    
2372    If you want to pass in additional data to pcre_exec(), you can either place it
2373    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2374    block.
2375    
2376    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2377    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2378    times with different match limits, until it finds the minimum value needed for
2379    pcre_exec() to complete. The value is then output. This can be instructive; for
2380    most simple matches the number is quite small, but for pathological cases it
2381    gets very large very quickly.
2382    
2383    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2384    returns the size of the data block pointed to by the study_data field in a
2385    pcre_extra block, that is, the value that was passed as the argument to
2386    pcre_malloc() when PCRE was getting memory in which to place the information
2387    created by pcre_study(). The fourth argument should point to a size_t variable.
2388    pcretest has been extended so that this information is shown after a successful
2389    pcre_study() call when information about the compiled regex is being displayed.
2390    
2391    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2392    because what follows is always an absolute path. (Later: it turns out that this
2393    is more than cosmetic for MinGW, because it doesn't like empty path
2394    components.)
2395    
2396    49. Some changes have been made to the callout feature (see 28 above):
2397    
2398    (i)  A callout function now has three choices for what it returns:
2399    
2400           0  =>  success, carry on matching
2401         > 0  =>  failure at this point, but backtrack if possible
2402         < 0  =>  serious error, return this value from pcre_exec()
2403    
2404         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2405         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2406         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2407         use by callout functions. It will never be used by PCRE itself.
2408    
2409    (ii) The pcre_extra structure (see 45 above) has a void * field called
2410         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2411         pcre_callout_block structure has a field of the same name. The contents of
2412         the field passed in the pcre_extra structure are passed to the callout
2413         function in the corresponding field in the callout block. This makes it
2414         easier to use the same callout-containing regex from multiple threads. For
2415         testing, the pcretest program has a new data escape
2416    
2417           \C*n        pass the number n (may be negative) as callout_data
2418    
2419         If the callout function in pcretest receives a non-zero value as
2420         callout_data, it returns that value.
2421    
2422    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2423    there were some redundant $(CFLAGS) in commands that are now specified as
2424    $(LINK), which already includes $(CFLAGS).
2425    
2426    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2427    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2428    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2429    one-byte characters throughout. Note that case-insensitive matching applies
2430    only to characters whose values are less than 256. PCRE doesn't support the
2431    notion of cases for higher-valued characters.
2432    
2433    (i)   A character class whose characters are all within 0-255 is handled as
2434          a bit map, and the map is inverted for negative classes. Previously, a
2435          character > 255 always failed to match such a class; however it should
2436          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2437    
2438    (ii)  A negated character class with a single character < 255 is coded as
2439          "not this character" (OP_NOT). This wasn't working properly when the test
2440          character was multibyte, either singly or repeated.
2441    
2442    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2443          mode, for example: \x{100}{2,3}.
2444    
2445    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2446          singly or repeated) now correctly test multibyte characters. However,
2447          PCRE doesn't recognize any characters with values greater than 255 as
2448          digits, spaces, or word characters. Such characters always match \D, \S,
2449          and \W, and never match \d, \s, or \w.
2450    
2451    (v)   Classes may now contain characters and character ranges with values
2452          greater than 255. For example: [ab\x{100}-\x{400}].
2453    
2454    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2455          PCRE in UTF-8 mode.
2456    
2457    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2458    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2459    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2460    value.)
2461    
2462    53. The single man page has become too large. I have therefore split it up into
2463    a number of separate man pages. These also give rise to individual HTML pages;
2464    these are now put in a separate directory, and there is an index.html page that
2465    lists them all. Some hyperlinking between the pages has been installed.
2466    
2467    54. Added convenience functions for handling named capturing parentheses.
2468    
2469    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2470    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2471    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2472    are faulted.
2473    
2474    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2475    calling configure. These values are used when compiling the dftables.c program
2476    which is run to generate the source of the default character tables. They
2477    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2478    you will need to set these values.
2479    
2480    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2481    
2482    
2483    Version 3.9 02-Jan-02
2484    ---------------------
2485    
2486    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2487    
2488    2. If --disable-static was given, the building process failed when trying to
2489    build pcretest and pcregrep. (For some reason it was using libtool to compile
2490    them, which is not right, as they aren't part of the library.)
2491    
2492    
2493    Version 3.8 18-Dec-01
2494    ---------------------
2495    
2496    1. The experimental UTF-8 code was completely screwed up. It was packing the
2497    bytes in the wrong order. How dumb can you get?
2498    
2499    
2500    Version 3.7 29-Oct-01
2501    ---------------------
2502    
2503    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2504    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2505    this didn't happen under Solaris 8, where I normally test things.
2506    
2507    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2508    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2509    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2510    this fix an hour or so after the initial 3.7 release.)
2511    
2512    
2513    Version 3.6 23-Oct-01
2514    ---------------------
2515    
2516    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2517    offsets passed as NULL with zero offset count.
2518    
2519    2. The config.guess and config.sub files had not been updated when I moved to
2520    the latest autoconf.
2521    
2522    
2523    Version 3.5 15-Aug-01
2524    ---------------------
2525    
2526    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2527    had been forgotten.
2528    
2529    2. By using declared but undefined structures, we can avoid using "void"
2530    definitions in pcre.h while keeping the internal definitions of the structures
2531    private.
2532    
2533    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2534    user point of view, this means that both static and shared libraries are built
2535    by default, but this can be individually controlled. More of the work of
2536    handling this static/shared cases is now inside libtool instead of PCRE's make
2537    file.
2538    
2539    4. The pcretest utility is now installed along with pcregrep because it is
2540    useful for users (to test regexs) and by doing this, it automatically gets
2541    relinked by libtool. The documentation has been turned into a man page, so
2542    there are now .1, .txt, and .html versions in /doc.
2543    
2544    5. Upgrades to pcregrep:
2545       (i)   Added long-form option names like gnu grep.
2546       (ii)  Added --help to list all options with an explanatory phrase.
2547       (iii) Added -r, --recursive to recurse into sub-directories.
2548       (iv)  Added -f, --file to read patterns from a file.
2549    
2550    6. pcre_exec() was referring to its "code" argument before testing that
2551    argument for NULL (and giving an error if it was NULL).
2552    
2553    7. Upgraded Makefile.in to allow for compiling in a different directory from
2554    the source directory.
2555    
2556    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2557    options bits, the pointer it was passed was to an int instead of to an unsigned
2558    long int. This mattered only on 64-bit systems.
2559    
2560    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2561    generated) instead of pcre.in, which it its source. Also made the same change
2562    in several of the .c files.
2563    
2564    10. A new release of gcc defines printf() as a macro, which broke pcretest
2565    because it had an ifdef in the middle of a string argument for printf(). Fixed
2566    by using separate calls to printf().
2567    
2568    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2569    script, to force use of CR or LF instead of \n in the source. On non-Unix
2570    systems, the value can be set in config.h.
2571    
2572    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2573    absolute limit. Changed the text of the error message to make this clear, and
2574    likewise updated the man page.
2575    
2576    13. The limit of 99 on the number of capturing subpatterns has been removed.
2577    The new limit is 65535, which I hope will not be a "real" limit.
2578    
2579    
2580    Version 3.4 22-Aug-00
2581    ---------------------
2582    
2583    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2584    
2585    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2586    
2587    
2588  Version 3.3 01-Aug-00  Version 3.3 01-Aug-00
2589  ---------------------  ---------------------

Legend:
Removed from v.49  
changed lines
  Added in v.325

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12