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

Diff of /code/trunk/ChangeLog

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

revision 25 by nigel, Sat Feb 24 21:38:45 2007 UTC revision 322 by ph10, Wed Mar 5 17:14:08 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    
16    Version 7.6 28-Jan-08
17    ---------------------
18    
19    1.  A character class containing a very large number of characters with
20        codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
21        overflow.
22    
23    2.  Patch to cut out the "long long" test in pcrecpp_unittest when
24        HAVE_LONG_LONG is not defined.
25    
26    3.  Applied Christian Ehrlicher's patch to update the CMake build files to
27        bring them up to date and include new features. This patch includes:
28    
29        - Fixed PH's badly added libz and libbz2 support.
30        - Fixed a problem with static linking.
31        - Added pcredemo. [But later removed - see 7 below.]
32        - Fixed dftables problem and added an option.
33        - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
34            HAVE_LONG_LONG.
35        - Added readline support for pcretest.
36        - Added an listing of the option settings after cmake has run.
37    
38    4.  A user submitted a patch to Makefile that makes it easy to create
39        "pcre.dll" under mingw when using Configure/Make. I added stuff to
40        Makefile.am that cause it to include this special target, without
41        affecting anything else. Note that the same mingw target plus all
42        the other distribution libraries and programs are now supported
43        when configuring with CMake (see 6 below) instead of with
44        Configure/Make.
45    
46    5.  Applied Craig's patch that moves no_arg into the RE class in the C++ code.
47        This is an attempt to solve the reported problem "pcrecpp::no_arg is not
48        exported in the Windows port". It has not yet been confirmed that the patch
49        solves the problem, but it does no harm.
50    
51    6.  Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
52        NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
53        with CMake, and also correct the comment about stack recursion.
54    
55    7.  Remove the automatic building of pcredemo from the ./configure system and
56        from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
57        of a program that users should build themselves after PCRE is installed, so
58        building it automatically is not really right. What is more, it gave
59        trouble in some build environments.
60    
61    8.  Further tidies to CMakeLists.txt from Sheri and Christian.
62    
63    
64    Version 7.5 10-Jan-08
65    ---------------------
66    
67    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
68        values in parens when parsing an RE using the C++ wrapper."
69    
70    2.  Negative specials like \S did not work in character classes in UTF-8 mode.
71        Characters greater than 255 were excluded from the class instead of being
72        included.
73    
74    3.  The same bug as (2) above applied to negated POSIX classes such as
75        [:^space:].
76    
77    4.  PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
78        defined or documented. It seems to have been a typo for PCRE_STATIC, so
79        I have changed it.
80    
81    5.  The construct (?&) was not diagnosed as a syntax error (it referenced the
82        first named subpattern) and a construct such as (?&a) would reference the
83        first named subpattern whose name started with "a" (in other words, the
84        length check was missing). Both these problems are fixed. "Subpattern name
85        expected" is now given for (?&) (a zero-length name), and this patch also
86        makes it give the same error for \k'' (previously it complained that that
87        was a reference to a non-existent subpattern).
88    
89    6.  The erroneous patterns (?+-a) and (?-+a) give different error messages;
90        this is right because (?- can be followed by option settings as well as by
91        digits. I have, however, made the messages clearer.
92    
93    7.  Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
94        than the number used in the conditional) now cause a compile-time error.
95        This is actually not compatible with Perl, which accepts such patterns, but
96        treats the conditional as always being FALSE (as PCRE used to), but it
97        seems to me that giving a diagnostic is better.
98    
99    8.  Change "alphameric" to the more common word "alphanumeric" in comments
100        and messages.
101    
102    9.  Fix two occurrences of "backslash" in comments that should have been
103        "backspace".
104    
105    10. Remove two redundant lines of code that can never be obeyed (their function
106        was moved elsewhere).
107    
108    11. The program that makes PCRE's Unicode character property table had a bug
109        which caused it to generate incorrect table entries for sequences of
110        characters that have the same character type, but are in different scripts.
111        It amalgamated them into a single range, with the script of the first of
112        them. In other words, some characters were in the wrong script. There were
113        thirteen such cases, affecting characters in the following ranges:
114    
115          U+002b0 - U+002c1
116          U+0060c - U+0060d
117          U+0061e - U+00612
118          U+0064b - U+0065e
119          U+0074d - U+0076d
120          U+01800 - U+01805
121          U+01d00 - U+01d77
122          U+01d9b - U+01dbf
123          U+0200b - U+0200f
124          U+030fc - U+030fe
125          U+03260 - U+0327f
126          U+0fb46 - U+0fbb1
127          U+10450 - U+1049d
128    
129    12. The -o option (show only the matching part of a line) for pcregrep was not
130        compatible with GNU grep in that, if there was more than one match in a
131        line, it showed only the first of them. It now behaves in the same way as
132        GNU grep.
133    
134    13. If the -o and -v options were combined for pcregrep, it printed a blank
135        line for every non-matching line. GNU grep prints nothing, and pcregrep now
136        does the same. The return code can be used to tell if there were any
137        non-matching lines.
138    
139    14. Added --file-offsets and --line-offsets to pcregrep.
140    
141    15. The pattern (?=something)(?R) was not being diagnosed as a potentially
142        infinitely looping recursion. The bug was that positive lookaheads were not
143        being skipped when checking for a possible empty match (negative lookaheads
144        and both kinds of lookbehind were skipped).
145    
146    16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
147        inclusion of <windows.h> to before rather than after the definition of
148        INVALID_FILE_ATTRIBUTES (patch from David Byron).
149    
150    17. Specifying a possessive quantifier with a specific limit for a Unicode
151        character property caused pcre_compile() to compile bad code, which led at
152        runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
153        are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
154        caused the error; without that there was no problem.
155    
156    18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
157    
158    19. Added --enable-pcretest-libreadline.
159    
160    20. In pcrecpp.cc, the variable 'count' was incremented twice in
161        RE::GlobalReplace(). As a result, the number of replacements returned was
162        double what it should be. I removed one of the increments, but Craig sent a
163        later patch that removed the other one (the right fix) and added unit tests
164        that check the return values (which was not done before).
165    
166    21. Several CMake things:
167    
168        (1) Arranged that, when cmake is used on Unix, the libraries end up with
169            the names libpcre and libpcreposix, not just pcre and pcreposix.
170    
171        (2) The above change means that pcretest and pcregrep are now correctly
172            linked with the newly-built libraries, not previously installed ones.
173    
174        (3) Added PCRE_SUPPORT_LIBREADLINE, PCRE_SUPPORT_LIBZ, PCRE_SUPPORT_LIBBZ2.
175    
176    22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
177        crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
178        UTF-8 newline character). The key issue is that the pattern starts .*;
179        this means that the match must be either at the beginning, or after a
180        newline. The bug was in the code for advancing after a failed match and
181        checking that the new position followed a newline. It was not taking
182        account of UTF-8 characters correctly.
183    
184    23. PCRE was behaving differently from Perl in the way it recognized POSIX
185        character classes. PCRE was not treating the sequence [:...:] as a
186        character class unless the ... were all letters. Perl, however, seems to
187        allow any characters between [: and :], though of course it rejects as
188        unknown any "names" that contain non-letters, because all the known class
189        names consist only of letters. Thus, Perl gives an error for [[:1234:]],
190        for example, whereas PCRE did not - it did not recognize a POSIX character
191        class. This seemed a bit dangerous, so the code has been changed to be
192        closer to Perl. The behaviour is not identical to Perl, because PCRE will
193        diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
194        treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
195        Perl does, and where it didn't before.
196    
197    24. Rewrite so as to remove the single use of %n from pcregrep because in some
198        Windows environments %n is disabled by default.
199    
200    
201    Version 7.4 21-Sep-07
202    ---------------------
203    
204    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
205        means that a class such as [\s] counted as "explicit reference to CR or
206        LF". That isn't really right - the whole point of the change was to try to
207        help when there was an actual mention of one of the two characters. So now
208        the change happens only if \r or \n (or a literal CR or LF) character is
209        encountered.
210    
211    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
212        of both had grown to the point where there were only 3 bits left.
213        Fortunately, there was spare space in the data structure, and so I have
214        moved the internal flags into a new 16-bit field to free up more option
215        bits.
216    
217    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
218        but did not set the internal JCHANGED flag - either of these is enough to
219        control the way the "get" function works - but the PCRE_INFO_JCHANGED
220        facility is supposed to tell if (?J) was ever used, so now (?J) at the
221        start sets both bits.
222    
223    4.  Added options (at build time, compile time, exec time) to change \R from
224        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
225    
226    5.  doc/pcresyntax.html was missing from the distribution.
227    
228    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
229        compatibility, even though it is no longer used.
230    
231    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
232        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
233        windows.h file is present (where different names are used). [This was
234        reversed later after testing - see 16 below.]
235    
236    8.  Changed all #include <config.h> to #include "config.h". There were also
237        some further <pcre.h> cases that I changed to "pcre.h".
238    
239    9.  When pcregrep was used with the --colour option, it missed the line ending
240        sequence off the lines that it output.
241    
242    10. It was pointed out to me that arrays of string pointers cause lots of
243        relocations when a shared library is dynamically loaded. A technique of
244        using a single long string with a table of offsets can drastically reduce
245        these. I have refactored PCRE in four places to do this. The result is
246        dramatic:
247    
248          Originally:                          290
249          After changing UCP table:            187
250          After changing error message table:   43
251          After changing table of "verbs"       36
252          After changing table of Posix names   22
253    
254        Thanks to the folks working on Gregex for glib for this insight.
255    
256    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
257        unicode-properties was also set.
258    
259    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
260    
261    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
262        checked only for CRLF.
263    
264    14. Added casts to pcretest.c to avoid compiler warnings.
265    
266    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
267    
268    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
269        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
270        entirely. This removes changes made in 7 above.
271    
272    17. The CMake files have been updated, and there is now more information about
273        building with CMake in the NON-UNIX-USE document.
274    
275    
276    Version 7.3 28-Aug-07
277    ---------------------
278    
279     1. In the rejigging of the build system that eventually resulted in 7.1, the
280        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
281        brackets there is not right, since it causes compilers to look for an
282        installed pcre.h, not the version that is in the source that is being
283        compiled (which of course may be different). I have changed it back to:
284    
285          #include "pcre.h"
286    
287        I have a vague recollection that the change was concerned with compiling in
288        different directories, but in the new build system, that is taken care of
289        by the VPATH setting the Makefile.
290    
291     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
292        when the subject happened to end in the byte 0x85 (e.g. if the last
293        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
294        characters but of course it shouldn't be taken as a newline when it is part
295        of another character. The bug was that, for an unlimited repeat of . in
296        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
297        characters when looking for a newline.
298    
299     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
300    
301     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
302        in debug output.
303    
304     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
305        long printing in the pcrecpp unittest when running under MinGW.
306    
307     6. ESC_K was left out of the EBCDIC table.
308    
309     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
310        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
311        limit also applies to "virtual nesting" when a pattern is recursive, and in
312        this case 1000 isn't so big. I have been able to remove this limit at the
313        expense of backing off one optimization in certain circumstances. Normally,
314        when pcre_exec() would call its internal match() function recursively and
315        immediately return the result unconditionally, it uses a "tail recursion"
316        feature to save stack. However, when a subpattern that can match an empty
317        string has an unlimited repetition quantifier, it no longer makes this
318        optimization. That gives it a stack frame in which to save the data for
319        checking that an empty string has been matched. Previously this was taken
320        from the 1000-entry workspace that had been reserved. So now there is no
321        explicit limit, but more stack is used.
322    
323     8. Applied Daniel's patches to solve problems with the import/export magic
324        syntax that is required for Windows, and which was going wrong for the
325        pcreposix and pcrecpp parts of the library. These were overlooked when this
326        problem was solved for the main library.
327    
328     9. There were some crude static tests to avoid integer overflow when computing
329        the size of patterns that contain repeated groups with explicit upper
330        limits. As the maximum quantifier is 65535, the maximum group length was
331        set at 30,000 so that the product of these two numbers did not overflow a
332        32-bit integer. However, it turns out that people want to use groups that
333        are longer than 30,000 bytes (though not repeat them that many times).
334        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
335        made it possible to implement the integer overflow checks in a much more
336        dynamic way, which I have now done. The artificial limitation on group
337        length has been removed - we now have only the limit on the total length of
338        the compiled pattern, which depends on the LINK_SIZE setting.
339    
340    10. Fixed a bug in the documentation for get/copy named substring when
341        duplicate names are permitted. If none of the named substrings are set, the
342        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
343        empty string.
344    
345    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
346        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
347        because the ] is interpreted as the first data character and the
348        terminating ] is not found. PCRE has been made compatible with Perl in this
349        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
350        cause memory overwriting.
351    
352    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
353        string has been matched (to stop an infinite loop). It was not recognizing
354        a conditional subpattern that could match an empty string if that
355        subpattern was within another subpattern. For example, it looped when
356        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
357        condition was not nested. This bug has been fixed.
358    
359    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
360        past the start of the subject in the presence of bytes with the top bit
361        set, for example "\x8aBCD".
362    
363    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
364        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
365    
366    14. Optimized (?!) to (*FAIL).
367    
368    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
369        This restricts code points to be within the range 0 to 0x10FFFF, excluding
370        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
371        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
372        does: it's just the validity check that is more restrictive.
373    
374    16. Inserted checks for integer overflows during escape sequence (backslash)
375        processing, and also fixed erroneous offset values for syntax errors during
376        backslash processing.
377    
378    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
379        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
380    
381    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
382        caused an overrun.
383    
384    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
385        something other than just ASCII characters) inside a group that had an
386        unlimited repeat caused a loop at compile time (while checking to see
387        whether the group could match an empty string).
388    
389    20. Debugging a pattern containing \p or \P could cause a crash. For example,
390        [\P{Any}] did so. (Error in the code for printing property names.)
391    
392    21. An orphan \E inside a character class could cause a crash.
393    
394    22. A repeated capturing bracket such as (A)? could cause a wild memory
395        reference during compilation.
396    
397    23. There are several functions in pcre_compile() that scan along a compiled
398        expression for various reasons (e.g. to see if it's fixed length for look
399        behind). There were bugs in these functions when a repeated \p or \P was
400        present in the pattern. These operators have additional parameters compared
401        with \d, etc, and these were not being taken into account when moving along
402        the compiled data. Specifically:
403    
404        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
405            length.
406    
407        (b) An item such as \pL+ within a repeated group could cause crashes or
408            loops.
409    
410        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
411            "reference to non-existent subpattern" error.
412    
413        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
414    
415    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
416        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
417    
418    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
419    
420    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
421        character were causing crashes (broken optimization).
422    
423    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
424        \p or \P) caused a compile-time loop.
425    
426    28. More problems have arisen in unanchored patterns when CRLF is a valid line
427        break. For example, the unstudied pattern [\r\n]A does not match the string
428        "\r\nA" because change 7.0/46 below moves the current point on by two
429        characters after failing to match at the start. However, the pattern \nA
430        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
431        the same is true. There doesn't seem any very clean way out of this, but
432        what I have chosen to do makes the common cases work: PCRE now takes note
433        of whether there can be an explicit match for \r or \n anywhere in the
434        pattern, and if so, 7.0/46 no longer applies. As part of this change,
435        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
436        pattern has explicit CR or LF references.
437    
438    29. Added (*CR) etc for changing newline setting at start of pattern.
439    
440    
441    Version 7.2 19-Jun-07
442    ---------------------
443    
444     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
445        which is apparently normally available under Windows.
446    
447     2. Re-jig the pcregrep tests with different newline settings in an attempt
448        to make them independent of the local environment's newline setting.
449    
450     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
451    
452     4. Some of the "internals" tests were previously cut out when the link size
453        was not 2, because the output contained actual offsets. The recent new
454        "Z" feature of pcretest means that these can be cut out, making the tests
455        usable with all link sizes.
456    
457     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
458        stack recursion. This gives a massive performance boost under BSD, but just
459        a small improvement under Linux. However, it saves one field in the frame
460        in all cases.
461    
462     6. Added more features from the forthcoming Perl 5.10:
463    
464        (a) (?-n) (where n is a string of digits) is a relative subroutine or
465            recursion call. It refers to the nth most recently opened parentheses.
466    
467        (b) (?+n) is also a relative subroutine call; it refers to the nth next
468            to be opened parentheses.
469    
470        (c) Conditions that refer to capturing parentheses can be specified
471            relatively, for example, (?(-2)... or (?(+3)...
472    
473        (d) \K resets the start of the current match so that everything before
474            is not part of it.
475    
476        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
477    
478        (f) \g{name} is another synonym - part of Perl 5.10's unification of
479            reference syntax.
480    
481        (g) (?| introduces a group in which the numbering of parentheses in each
482            alternative starts with the same number.
483    
484        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
485    
486     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
487        PCRE_INFO_JCHANGED.
488    
489     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
490        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
491        for detecting groups that can match an empty string.
492    
493     9. A pattern with a very large number of alternatives (more than several
494        hundred) was running out of internal workspace during the pre-compile
495        phase, where pcre_compile() figures out how much memory will be needed. A
496        bit of new cunning has reduced the workspace needed for groups with
497        alternatives. The 1000-alternative test pattern now uses 12 bytes of
498        workspace instead of running out of the 4096 that are available.
499    
500    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
501    
502    11. Applied patch from Google to remove an optimization that didn't quite work.
503        The report of the bug said:
504    
505          pcrecpp::RE("a*").FullMatch("aaa") matches, while
506          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
507          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
508    
509    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
510        it matched the wrong number of bytes.
511    
512    
513    Version 7.1 24-Apr-07
514    ---------------------
515    
516     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
517        that is more "standard", making use of automake and other Autotools. There
518        is some re-arrangement of the files and adjustment of comments consequent
519        on this.
520    
521     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
522        for recursive directory scanning broke on some systems because the files
523        are not scanned in any specific order and on different systems the order
524        was different. A call to "sort" has been inserted into RunGrepTest for the
525        approprate test as a short-term fix. In the longer term there may be an
526        alternative.
527    
528     3. I had an email from Eric Raymond about problems translating some of PCRE's
529        man pages to HTML (despite the fact that I distribute HTML pages, some
530        people do their own conversions for various reasons). The problems
531        concerned the use of low-level troff macros .br and .in. I have therefore
532        removed all such uses from the man pages (some were redundant, some could
533        be replaced by .nf/.fi pairs). The 132html script that I use to generate
534        HTML has been updated to handle .nf/.fi and to complain if it encounters
535        .br or .in.
536    
537     4. Updated comments in configure.ac that get placed in config.h.in and also
538        arranged for config.h to be included in the distribution, with the name
539        config.h.generic, for the benefit of those who have to compile without
540        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
541    
542     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
543        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
544        makevp.bat for the current PCRE, using the additional files
545        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
546    
547     6. A Windows user reported a minor discrepancy with test 2, which turned out
548        to be caused by a trailing space on an input line that had got lost in his
549        copy. The trailing space was an accident, so I've just removed it.
550    
551     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
552        that is needed.
553    
554     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
555        as "const" (a) because they are and (b) because it helps the PHP
556        maintainers who have recently made a script to detect big data structures
557        in the php code that should be moved to the .rodata section. I remembered
558        to update Builducptable as well, so it won't revert if ucptable.h is ever
559        re-created.
560    
561     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
562        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
563        order to be able to cut out the UTF-8 tables in the latter when UTF-8
564        support is not required. This saves 1.5-2K of code, which is important in
565        some applications.
566    
567        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
568        so as not to refer to the tables, even though these functions will never be
569        called when UTF-8 support is disabled. Otherwise there are problems with a
570        shared library.
571    
572    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
573    
574        (a) It was defining its arguments as char * instead of void *.
575    
576        (b) It was assuming that all moves were upwards in memory; this was true
577            a long time ago when I wrote it, but is no longer the case.
578    
579        The emulated memove() is provided for those environments that have neither
580        memmove() nor bcopy(). I didn't think anyone used it these days, but that
581        is clearly not the case, as these two bugs were recently reported.
582    
583    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
584        and Detrail to create the HTML documentation, the .txt form of the man
585        pages, and it removes trailing spaces from listed files. It also creates
586        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
587        case, it wraps all the #defines with #ifndefs. This script should be run
588        before "make dist".
589    
590    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
591        with Unicode property support.
592    
593        (a) For a maximizing quantifier, if the two different cases of the
594            character were of different lengths in their UTF-8 codings (there are
595            some cases like this - I found 11), and the matching function had to
596            back up over a mixture of the two cases, it incorrectly assumed they
597            were both the same length.
598    
599        (b) When PCRE was configured to use the heap rather than the stack for
600            recursion during matching, it was not correctly preserving the data for
601            the other case of a UTF-8 character when checking ahead for a match
602            while processing a minimizing repeat. If the check also involved
603            matching a wide character, but failed, corruption could cause an
604            erroneous result when trying to check for a repeat of the original
605            character.
606    
607    13. Some tidying changes to the testing mechanism:
608    
609        (a) The RunTest script now detects the internal link size and whether there
610            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
611            values substituted by "configure". (The RunGrepTest script already did
612            this for UTF-8.) The configure.ac script no longer substitutes the
613            relevant variables.
614    
615        (b) The debugging options /B and /D in pcretest show the compiled bytecode
616            with length and offset values. This means that the output is different
617            for different internal link sizes. Test 2 is skipped for link sizes
618            other than 2 because of this, bypassing the problem. Unfortunately,
619            there was also a test in test 3 (the locale tests) that used /B and
620            failed for link sizes other than 2. Rather than cut the whole test out,
621            I have added a new /Z option to pcretest that replaces the length and
622            offset values with spaces. This is now used to make test 3 independent
623            of link size. (Test 2 will be tidied up later.)
624    
625    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
626        segmentation fault instead of returning the appropriate error message.
627    
628    15. In multiline mode when the newline sequence was set to "any", the pattern
629        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
630        This doesn't seem right; it now treats the CRLF combination as the line
631        ending, and so does not match in that case. It's only a pattern such as ^$
632        that would hit this one: something like ^ABC$ would have failed after \r
633        and then tried again after \r\n.
634    
635    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
636        in an attempt to make files that differ only in their line terminators
637        compare equal. This works on Linux.
638    
639    17. Under certain error circumstances pcregrep might try to free random memory
640        as it exited. This is now fixed, thanks to valgrind.
641    
642    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
643        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
644        was because its rules for how to advance for /g after matching an empty
645        string at the end of a line did not allow for this case. They now check for
646        it specially.
647    
648    20. pcretest is supposed to handle patterns and data of any length, by
649        extending its buffers when necessary. It was getting this wrong when the
650        buffer for a data line had to be extended.
651    
652    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
653        CRLF as a newline sequence.
654    
655    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
656        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
657        I have nevertheless tidied it up.
658    
659    23. Added some casts to kill warnings from HP-UX ia64 compiler.
660    
661    24. Added a man page for pcre-config.
662    
663    
664    Version 7.0 19-Dec-06
665    ---------------------
666    
667     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
668        moving to gcc 4.1.1.
669    
670     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
671        sys/time.h, which is documented as needed for this function. It doesn't
672        seem to matter on Linux, but it showed up on some releases of OS X.
673    
674     3. It seems that there are systems where bytes whose values are greater than
675        127 match isprint() in the "C" locale. The "C" locale should be the
676        default when a C program starts up. In most systems, only ASCII printing
677        characters match isprint(). This difference caused the output from pcretest
678        to vary, making some of the tests fail. I have changed pcretest so that:
679    
680        (a) When it is outputting text in the compiled version of a pattern, bytes
681            other than 32-126 are always shown as hex escapes.
682    
683        (b) When it is outputting text that is a matched part of a subject string,
684            it does the same, unless a different locale has been set for the match
685            (using the /L modifier). In this case, it uses isprint() to decide.
686    
687     4. Fixed a major bug that caused incorrect computation of the amount of memory
688        required for a compiled pattern when options that changed within the
689        pattern affected the logic of the preliminary scan that determines the
690        length. The relevant options are -x, and -i in UTF-8 mode. The result was
691        that the computed length was too small. The symptoms of this bug were
692        either the PCRE error "internal error: code overflow" from pcre_compile(),
693        or a glibc crash with a message such as "pcretest: free(): invalid next
694        size (fast)". Examples of patterns that provoked this bug (shown in
695        pcretest format) are:
696    
697          /(?-x: )/x
698          /(?x)(?-x: \s*#\s*)/
699          /((?i)[\x{c0}])/8
700          /(?i:[\x{c0}])/8
701    
702        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
703        is now done differently.
704    
705     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
706        wrapper classes; (b) implement a new function in the C++ scanner that is
707        more efficient than the old way of doing things because it avoids levels of
708        recursion in the regex matching; (c) add a paragraph to the documentation
709        for the FullMatch() function.
710    
711     6. The escape sequence \n was being treated as whatever was defined as
712        "newline". Not only was this contrary to the documentation, which states
713        that \n is character 10 (hex 0A), but it also went horribly wrong when
714        "newline" was defined as CRLF. This has been fixed.
715    
716     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
717        was being set to -1 for the "end of line" case (supposedly a value that no
718        character can have). Though this value is never used (the check for end of
719        line is "zero bytes in current character"), it caused compiler complaints.
720        I've changed it to 0xffffffff.
721    
722     8. In pcre_version.c, the version string was being built by a sequence of
723        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
724        string (as it is for production releases) called a macro with an empty
725        argument. The C standard says the result of this is undefined. The gcc
726        compiler treats it as an empty string (which was what was wanted) but it is
727        reported that Visual C gives an error. The source has been hacked around to
728        avoid this problem.
729    
730     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
731        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
732        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
733        of them did).
734    
735    10. Originally, pcretest opened its input and output without "b"; then I was
736        told that "b" was needed in some environments, so it was added for release
737        5.0 to both the input and output. (It makes no difference on Unix-like
738        systems.) Later I was told that it is wrong for the input on Windows. I've
739        now abstracted the modes into two macros, to make it easier to fiddle with
740        them, and removed "b" from the input mode under Windows.
741    
742    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
743    
744    12. Added -help and --help to pcretest as an official way of being reminded
745        of the options.
746    
747    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
748        and pcrecpp.cc because they annoy compilers at high warning levels.
749    
750    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
751    
752    15. Fixed an occurrence of == in configure.ac that should have been = (shell
753        scripts are not C programs :-) and which was not noticed because it works
754        on Linux.
755    
756    16. pcretest is supposed to handle any length of pattern and data line (as one
757        line or as a continued sequence of lines) by extending its input buffer if
758        necessary. This feature was broken for very long pattern lines, leading to
759        a string of junk being passed to pcre_compile() if the pattern was longer
760        than about 50K.
761    
762    17. I have done a major re-factoring of the way pcre_compile() computes the
763        amount of memory needed for a compiled pattern. Previously, there was code
764        that made a preliminary scan of the pattern in order to do this. That was
765        OK when PCRE was new, but as the facilities have expanded, it has become
766        harder and harder to keep it in step with the real compile phase, and there
767        have been a number of bugs (see for example, 4 above). I have now found a
768        cunning way of running the real compile function in a "fake" mode that
769        enables it to compute how much memory it would need, while actually only
770        ever using a few hundred bytes of working memory and without too many
771        tests of the mode. This should make future maintenance and development
772        easier. A side effect of this work is that the limit of 200 on the nesting
773        depth of parentheses has been removed (though this was never a serious
774        limitation, I suspect). However, there is a downside: pcre_compile() now
775        runs more slowly than before (30% or more, depending on the pattern). I
776        hope this isn't a big issue. There is no effect on runtime performance.
777    
778    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
779        newline (only possible for the last line of a file) and it was a
780        pattern that set a locale (followed by /Lsomething), pcretest crashed.
781    
782    19. Added additional timing features to pcretest. (1) The -tm option now times
783        matching only, not compiling. (2) Both -t and -tm can be followed, as a
784        separate command line item, by a number that specifies the number of
785        repeats to use when timing. The default is 50000; this gives better
786        precision, but takes uncomfortably long for very large patterns.
787    
788    20. Extended pcre_study() to be more clever in cases where a branch of a
789        subpattern has no definite first character. For example, (a*|b*)[cd] would
790        previously give no result from pcre_study(). Now it recognizes that the
791        first character must be a, b, c, or d.
792    
793    21. There was an incorrect error "recursive call could loop indefinitely" if
794        a subpattern (or the entire pattern) that was being tested for matching an
795        empty string contained only one non-empty item after a nested subpattern.
796        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
797        incorrectly, because the \d was being skipped in the check.
798    
799    22. The pcretest program now has a new pattern option /B and a command line
800        option -b, which is equivalent to adding /B to every pattern. This causes
801        it to show the compiled bytecode, without the additional information that
802        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
803        is the same as /B/I).
804    
805    23. A new optimization is now able automatically to treat some sequences such
806        as a*b as a*+b. More specifically, if something simple (such as a character
807        or a simple class like \d) has an unlimited quantifier, and is followed by
808        something that cannot possibly match the quantified thing, the quantifier
809        is automatically "possessified".
810    
811    24. A recursive reference to a subpattern whose number was greater than 39
812        went wrong under certain circumstances in UTF-8 mode. This bug could also
813        have affected the operation of pcre_study().
814    
815    25. Realized that a little bit of performance could be had by replacing
816        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
817    
818    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
819    
820    27. Possessive quantifiers such as a++ were previously implemented by turning
821        them into atomic groups such as ($>a+). Now they have their own opcodes,
822        which improves performance. This includes the automatically created ones
823        from 23 above.
824    
825    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
826        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
827        the first matched character to be a colon. This applied both to named and
828        numbered groups.
829    
830    29. The ucpinternal.h header file was missing its idempotency #ifdef.
831    
832    30. I was sent a "project" file called libpcre.a.dev which I understand makes
833        building PCRE on Windows easier, so I have included it in the distribution.
834    
835    31. There is now a check in pcretest against a ridiculously large number being
836        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
837        loop, the loop is abandoned.
838    
839    32. Forward references to subpatterns in conditions such as (?(2)...) where
840        subpattern 2 is defined later cause pcre_compile() to search forwards in
841        the pattern for the relevant set of parentheses. This search went wrong
842        when there were unescaped parentheses in a character class, parentheses
843        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
844    
845    33. "Subroutine" calls and backreferences were previously restricted to
846        referencing subpatterns earlier in the regex. This restriction has now
847        been removed.
848    
849    34. Added a number of extra features that are going to be in Perl 5.10. On the
850        whole, these are just syntactic alternatives for features that PCRE had
851        previously implemented using the Python syntax or my own invention. The
852        other formats are all retained for compatibility.
853    
854        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
855            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
856            also .NET compatible.
857    
858        (b) A recursion or subroutine call to a named group can now be defined as
859            (?&name) as well as (?P>name).
860    
861        (c) A backreference to a named group can now be defined as \k<name> or
862            \k'name' as well as (?P=name). The new forms, as well as being in Perl
863            5.10, are also .NET compatible.
864    
865        (d) A conditional reference to a named group can now use the syntax
866            (?(<name>) or (?('name') as well as (?(name).
867    
868        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
869            groups (named and numbered) that are never evaluated inline, but can be
870            called as "subroutines" from elsewhere. In effect, the DEFINE condition
871            is always false. There may be only one alternative in such a group.
872    
873        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
874            as the simple (?(R). The condition is true only if the most recent
875            recursion is that of the given number or name. It does not search out
876            through the entire recursion stack.
877    
878        (g) The escape \gN or \g{N} has been added, where N is a positive or
879            negative number, specifying an absolute or relative reference.
880    
881    35. Tidied to get rid of some further signed/unsigned compiler warnings and
882        some "unreachable code" warnings.
883    
884    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
885        things, this adds five new scripts.
886    
887    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
888        There were also incompatibilities regarding the handling of \Q..\E inside
889        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
890        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
891    
892    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
893        matches an empty string, and forcibly breaks the loop. There were bugs in
894        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
895        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
896        separate and independent bugs (that affected different cases) have been
897        fixed.
898    
899    39. Refactored the code to abolish the use of different opcodes for small
900        capturing bracket numbers. This is a tidy that I avoided doing when I
901        removed the limit on the number of capturing brackets for 3.5 back in 2001.
902        The new approach is not only tidier, it makes it possible to reduce the
903        memory needed to fix the previous bug (38).
904    
905    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
906        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
907        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
908        mode.
909    
910    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
911        report.
912    
913    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
914        copy construction and assignment in the C++ wrapper.
915    
916    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
917        couple of bugs that could have given wrong results in the "--newline=crlf"
918        case.
919    
920    44. Added a number of casts and did some reorganization of signed/unsigned int
921        variables following suggestions from Dair Grant. Also renamed the variable
922        "this" as "item" because it is a C++ keyword.
923    
924    45. Arranged for dftables to add
925    
926          #include "pcre_internal.h"
927    
928        to pcre_chartables.c because without it, gcc 4.x may remove the array
929        definition from the final binary if PCRE is built into a static library and
930        dead code stripping is activated.
931    
932    46. For an unanchored pattern, if a match attempt fails at the start of a
933        newline sequence, and the newline setting is CRLF or ANY, and the next two
934        characters are CRLF, advance by two characters instead of one.
935    
936    
937    Version 6.7 04-Jul-06
938    ---------------------
939    
940     1. In order to handle tests when input lines are enormously long, pcretest has
941        been re-factored so that it automatically extends its buffers when
942        necessary. The code is crude, but this _is_ just a test program. The
943        default size has been increased from 32K to 50K.
944    
945     2. The code in pcre_study() was using the value of the re argument before
946        testing it for NULL. (Of course, in any sensible call of the function, it
947        won't be NULL.)
948    
949     3. The memmove() emulation function in pcre_internal.h, which is used on
950        systems that lack both memmove() and bcopy() - that is, hardly ever -
951        was missing a "static" storage class specifier.
952    
953     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
954        containing an extended class (one that cannot be represented by a bitmap
955        because it contains high-valued characters or Unicode property items, e.g.
956        [\pZ]). Almost always one would set UTF-8 mode when processing such a
957        pattern, but PCRE should not loop if you do not (it no longer does).
958        [Detail: two cases were found: (a) a repeated subpattern containing an
959        extended class; (b) a recursive reference to a subpattern that followed a
960        previous extended class. It wasn't skipping over the extended class
961        correctly when UTF-8 mode was not set.]
962    
963     5. A negated single-character class was not being recognized as fixed-length
964        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
965        compile error "lookbehind assertion is not fixed length".
966    
967     6. The RunPerlTest auxiliary script was showing an unexpected difference
968        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
969        write a Perl script that can interpret lines of an input file either as
970        byte characters or as UTF-8, which is what "perltest" was being required to
971        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
972        can't do is switch easily at run time between having the "use utf8;" pragma
973        or not. In the end, I fudged it by using the RunPerlTest script to insert
974        "use utf8;" explicitly for the UTF-8 tests.
975    
976     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
977        the end of the subject string, contrary to the documentation and to what
978        Perl does. This was true of both matching functions. Now it matches only at
979        the start of the subject and immediately after *internal* newlines.
980    
981     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
982        a pointer to an int instead of a pointer to an unsigned long int. This
983        caused problems on 64-bit systems.
984    
985     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
986        instance of the 'standard' template library not being so standard".
987    
988    10. There was no check on the number of named subpatterns nor the maximum
989        length of a subpattern name. The product of these values is used to compute
990        the size of the memory block for a compiled pattern. By supplying a very
991        long subpattern name and a large number of named subpatterns, the size
992        computation could be caused to overflow. This is now prevented by limiting
993        the length of names to 32 characters, and the number of named subpatterns
994        to 10,000.
995    
996    11. Subpatterns that are repeated with specific counts have to be replicated in
997        the compiled pattern. The size of memory for this was computed from the
998        length of the subpattern and the repeat count. The latter is limited to
999        65535, but there was no limit on the former, meaning that integer overflow
1000        could in principle occur. The compiled length of a repeated subpattern is
1001        now limited to 30,000 bytes in order to prevent this.
1002    
1003    12. Added the optional facility to have named substrings with the same name.
1004    
1005    13. Added the ability to use a named substring as a condition, using the
1006        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1007        are numbers (not recommended). Forward references are permitted.
1008    
1009    14. Added forward references in named backreferences (if you see what I mean).
1010    
1011    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1012        pattern could run off the end of the subject. For example, the pattern
1013        "(?s)(.{1,5})"8 did this with the subject "ab".
1014    
1015    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1016        PCRE_CASELESS was set when matching characters that were quantified with ?
1017        or *.
1018    
1019    17. A character class other than a single negated character that had a minimum
1020        but no maximum quantifier - for example [ab]{6,} - was not handled
1021        correctly by pce_dfa_exec(). It would match only one character.
1022    
1023    18. A valid (though odd) pattern that looked like a POSIX character
1024        class but used an invalid character after [ (for example [[,abc,]]) caused
1025        pcre_compile() to give the error "Failed: internal error: code overflow" or
1026        in some cases to crash with a glibc free() error. This could even happen if
1027        the pattern terminated after [[ but there just happened to be a sequence of
1028        letters, a binary zero, and a closing ] in the memory that followed.
1029    
1030    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1031        over the years. Originally (before any Unicode support), just the bottom 8
1032        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1033        output from "man perlunicode" includes this:
1034    
1035          The regular expression compiler produces polymorphic opcodes.  That
1036          is, the pattern adapts to the data and automatically switches to
1037          the Unicode character scheme when presented with Unicode data--or
1038          instead uses a traditional byte scheme when presented with byte
1039          data.
1040    
1041        Sadly, a wide octal escape does not cause a switch, and in a string with
1042        no other multibyte characters, these octal escapes are treated as before.
1043        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
1044        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1045        Unicode string.
1046    
1047        I have not perpetrated such confusion in PCRE. Up till now, it took just
1048        the bottom 8 bits, as in old Perl. I have now made octal escapes with
1049        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1050        translate to the appropriate multibyte character.
1051    
1052    29. Applied some refactoring to reduce the number of warnings from Microsoft
1053        and Borland compilers. This has included removing the fudge introduced
1054        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1055        a warning about an unused variable.
1056    
1057    21. PCRE has not included VT (character 0x0b) in the set of whitespace
1058        characters since release 4.0, because Perl (from release 5.004) does not.
1059        [Or at least, is documented not to: some releases seem to be in conflict
1060        with the documentation.] However, when a pattern was studied with
1061        pcre_study() and all its branches started with \s, PCRE still included VT
1062        as a possible starting character. Of course, this did no harm; it just
1063        caused an unnecessary match attempt.
1064    
1065    22. Removed a now-redundant internal flag bit that recorded the fact that case
1066        dependency changed within the pattern. This was once needed for "required
1067        byte" processing, but is no longer used. This recovers a now-scarce options
1068        bit. Also moved the least significant internal flag bit to the most-
1069        significant bit of the word, which was not previously used (hangover from
1070        the days when it was an int rather than a uint) to free up another bit for
1071        the future.
1072    
1073    23. Added support for CRLF line endings as well as CR and LF. As well as the
1074        default being selectable at build time, it can now be changed at runtime
1075        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1076        specify that it is scanning data with non-default line endings.
1077    
1078    24. Changed the definition of CXXLINK to make it agree with the definition of
1079        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1080    
1081    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1082        recursions. This makes a big different to stack usage for some patterns.
1083    
1084    26. If a subpattern containing a named recursion or subroutine reference such
1085        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1086        the space required for the compiled pattern went wrong and gave too small a
1087        value. Depending on the environment, this could lead to "Failed: internal
1088        error: code overflow at offset 49" or "glibc detected double free or
1089        corruption" errors.
1090    
1091    27. Applied patches from Google (a) to support the new newline modes and (b) to
1092        advance over multibyte UTF-8 characters in GlobalReplace.
1093    
1094    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1095        difference for some implementation of PCRE in some Windows version.
1096    
1097    29. Added some extra testing facilities to pcretest:
1098    
1099        \q<number>   in a data line sets the "match limit" value
1100        \Q<number>   in a data line sets the "match recursion limt" value
1101        -S <number>  sets the stack size, where <number> is in megabytes
1102    
1103        The -S option isn't available for Windows.
1104    
1105    
1106    Version 6.6 06-Feb-06
1107    ---------------------
1108    
1109     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1110        in pcreposix.h. I have copied the definition from pcre.h.
1111    
1112     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1113        because pcre.h is no longer a built file.
1114    
1115     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1116        not normally included in the compiled code.
1117    
1118    
1119    Version 6.5 01-Feb-06
1120    ---------------------
1121    
1122     1. When using the partial match feature with pcre_dfa_exec(), it was not
1123        anchoring the second and subsequent partial matches at the new starting
1124        point. This could lead to incorrect results. For example, with the pattern
1125        /1234/, partially matching against "123" and then "a4" gave a match.
1126    
1127     2. Changes to pcregrep:
1128    
1129        (a) All non-match returns from pcre_exec() were being treated as failures
1130            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1131            error message is output. Some extra information is given for the
1132            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
1133            probably the only errors that are likely to be caused by users (by
1134            specifying a regex that has nested indefinite repeats, for instance).
1135            If there are more than 20 of these errors, pcregrep is abandoned.
1136    
1137        (b) A binary zero was treated as data while matching, but terminated the
1138            output line if it was written out. This has been fixed: binary zeroes
1139            are now no different to any other data bytes.
1140    
1141        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1142            used to set a locale for matching. The --locale=xxxx long option has
1143            been added (no short equivalent) to specify a locale explicitly on the
1144            pcregrep command, overriding the environment variables.
1145    
1146        (d) When -B was used with -n, some line numbers in the output were one less
1147            than they should have been.
1148    
1149        (e) Added the -o (--only-matching) option.
1150    
1151        (f) If -A or -C was used with -c (count only), some lines of context were
1152            accidentally printed for the final match.
1153    
1154        (g) Added the -H (--with-filename) option.
1155    
1156        (h) The combination of options -rh failed to suppress file names for files
1157            that were found from directory arguments.
1158    
1159        (i) Added the -D (--devices) and -d (--directories) options.
1160    
1161        (j) Added the -F (--fixed-strings) option.
1162    
1163        (k) Allow "-" to be used as a file name for -f as well as for a data file.
1164    
1165        (l) Added the --colo(u)r option.
1166    
1167        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1168            is not present by default.
1169    
1170     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1171        items such as (?R) or (?1), when the recursion could match a number of
1172        alternatives. If it matched one of the alternatives, but subsequently,
1173        outside the recursion, there was a failure, the code tried to back up into
1174        the recursion. However, because of the way PCRE is implemented, this is not
1175        possible, and the result was an incorrect result from the match.
1176    
1177        In order to prevent this happening, the specification of recursion has
1178        been changed so that all such subpatterns are automatically treated as
1179        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1180    
1181     4. I had overlooked the fact that, in some locales, there are characters for
1182        which isalpha() is true but neither isupper() nor islower() are true. In
1183        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1184        and ordfeminine) are like this. This affected the treatment of \w and \W
1185        when they appeared in character classes, but not when they appeared outside
1186        a character class. The bit map for "word" characters is now created
1187        separately from the results of isalnum() instead of just taking it from the
1188        upper, lower, and digit maps. (Plus the underscore character, of course.)
1189    
1190     5. The above bug also affected the handling of POSIX character classes such as
1191        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1192        permanent tables. Instead, the bit maps for such a class were previously
1193        created as the appropriate unions of the upper, lower, and digit bitmaps.
1194        Now they are created by subtraction from the [[:word:]] class, which has
1195        its own bitmap.
1196    
1197     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1198        It is created by subtracting the vertical space characters (\x09, \x0a,
1199        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1200        subtraction was done in the overall bitmap for a character class, meaning
1201        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1202        be recognized. This bug has been fixed.
1203    
1204     7. Patches from the folks at Google:
1205    
1206          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1207          real life, but is still worth protecting against".
1208    
1209          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1210          regular expressions".
1211    
1212          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1213          have it.
1214    
1215          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1216          "configure" and the latter not, in order to fix a problem somebody had
1217          with compiling the Arg class on HP-UX.
1218    
1219          (e) Improve the error-handling of the C++ wrapper a little bit.
1220    
1221          (f) New tests for checking recursion limiting.
1222    
1223     8. The pcre_memmove() function, which is used only if the environment does not
1224        have a standard memmove() function (and is therefore rarely compiled),
1225        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1226        returning a result (though PCRE never actually uses the result).
1227    
1228     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1229        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1230        returned instead of calling malloc() with an overflowing number that would
1231        most likely cause subsequent chaos.
1232    
1233    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1234    
1235    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1236        with this option is matched, the nmatch and pmatch options of regexec() are
1237        ignored.
1238    
1239    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1240        provided in case anyone wants to the the POSIX interface with UTF-8
1241        strings.
1242    
1243    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1244        C++ linking (needed for some HP-UX environments).
1245    
1246    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1247        (unused parameter) and in the pcre_printint() function (omitted "default"
1248        switch label when the default is to do nothing).
1249    
1250    15. Added some code to make it possible, when PCRE is compiled as a C++
1251        library, to replace subject pointers for pcre_exec() with a smart pointer
1252        class, thus making it possible to process discontinuous strings.
1253    
1254    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1255        much the same function. They were added by different people who were trying
1256        to make PCRE easy to compile on non-Unix systems. It has been suggested
1257        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1258        for compiling on Windows systems. I have therefore replaced it with
1259        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1260        defaults to "extern" for C or "extern C" for C++, which works fine on
1261        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1262        SCOPE with something explicit in config.h. In addition:
1263    
1264        (a) pcreposix.h still had just "extern" instead of either of these macros;
1265            I have replaced it with PCRE_DATA_SCOPE.
1266    
1267        (b) Functions such as _pcre_xclass(), which are internal to the library,
1268            but external in the C sense, all had PCRE_EXPORT in their definitions.
1269            This is apparently wrong for the Windows case, so I have removed it.
1270            (It makes no difference on Unix-like systems.)
1271    
1272    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1273        of recursive calls to match(). This is different to MATCH_LIMIT because
1274        that limits the total number of calls to match(), not all of which increase
1275        the depth of recursion. Limiting the recursion depth limits the amount of
1276        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1277        when PCRE is compiled, and changed at run time. A patch from Google adds
1278        this functionality to the C++ interface.
1279    
1280    18. Changes to the handling of Unicode character properties:
1281    
1282        (a) Updated the table to Unicode 4.1.0.
1283    
1284        (b) Recognize characters that are not in the table as "Cn" (undefined).
1285    
1286        (c) I revised the way the table is implemented to a much improved format
1287            which includes recognition of ranges. It now supports the ranges that
1288            are defined in UnicodeData.txt, and it also amalgamates other
1289            characters into ranges. This has reduced the number of entries in the
1290            table from around 16,000 to around 3,000, thus reducing its size
1291            considerably. I realized I did not need to use a tree structure after
1292            all - a binary chop search is just as efficient. Having reduced the
1293            number of entries, I extended their size from 6 bytes to 8 bytes to
1294            allow for more data.
1295    
1296        (d) Added support for Unicode script names via properties such as \p{Han}.
1297    
1298    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1299        matching that character.
1300    
1301    20. When matching a repeated Unicode property with a minimum greater than zero,
1302        (for example \pL{2,}), PCRE could look past the end of the subject if it
1303        reached it while seeking the minimum number of characters. This could
1304        happen only if some of the characters were more than one byte long, because
1305        there is a check for at least the minimum number of bytes.
1306    
1307    21. Refactored the implementation of \p and \P so as to be more general, to
1308        allow for more different types of property in future. This has changed the
1309        compiled form incompatibly. Anybody with saved compiled patterns that use
1310        \p or \P will have to recompile them.
1311    
1312    22. Added "Any" and "L&" to the supported property types.
1313    
1314    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1315        but give a compile time error if the value is greater than 0xff.
1316    
1317    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1318        accidentally not being installed or uninstalled.
1319    
1320    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1321        made were to insert the current release number. This seemed silly, because
1322        it made things harder for people building PCRE on systems that don't run
1323        "configure". I have turned pcre.h into a distributed file, no longer built
1324        by "configure", with the version identification directly included. There is
1325        no longer a pcre.h.in file.
1326    
1327        However, this change necessitated a change to the pcre-config script as
1328        well. It is built from pcre-config.in, and one of the substitutions was the
1329        release number. I have updated configure.ac so that ./configure now finds
1330        the release number by grepping pcre.h.
1331    
1332    26. Added the ability to run the tests under valgrind.
1333    
1334    
1335    Version 6.4 05-Sep-05
1336    ---------------------
1337    
1338     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1339        "--" to be printed when multiple files were scanned, even when none of the
1340        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1341        consider it to be a bug, and have restored the previous behaviour.
1342    
1343     2. A couple of code tidies to get rid of compiler warnings.
1344    
1345     3. The pcretest program used to cheat by referring to symbols in the library
1346        whose names begin with _pcre_. These are internal symbols that are not
1347        really supposed to be visible externally, and in some environments it is
1348        possible to suppress them. The cheating is now confined to including
1349        certain files from the library's source, which is a bit cleaner.
1350    
1351     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1352        file's purpose clearer.
1353    
1354     5. Reorganized pcre_ucp_findchar().
1355    
1356    
1357    Version 6.3 15-Aug-05
1358    ---------------------
1359    
1360     1. The file libpcre.pc.in did not have general read permission in the tarball.
1361    
1362     2. There were some problems when building without C++ support:
1363    
1364        (a) If C++ support was not built, "make install" and "make test" still
1365            tried to test it.
1366    
1367        (b) There were problems when the value of CXX was explicitly set. Some
1368            changes have been made to try to fix these, and ...
1369    
1370        (c) --disable-cpp can now be used to explicitly disable C++ support.
1371    
1372        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1373            backslash in a target when C++ was disabled. This confuses some
1374            versions of "make", apparently. Using an intermediate variable solves
1375            this. (Same for CPP_LOBJ.)
1376    
1377     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1378        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1379        necessary on certain architectures.
1380    
1381     4. Added a setting of -export-symbols-regex to the link command to remove
1382        those symbols that are exported in the C sense, but actually are local
1383        within the library, and not documented. Their names all begin with
1384        "_pcre_". This is not a perfect job, because (a) we have to except some
1385        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1386        available (and never for static libraries). I have made a note to try to
1387        find a way round (a) in the future.
1388    
1389    
1390    Version 6.2 01-Aug-05
1391    ---------------------
1392    
1393     1. There was no test for integer overflow of quantifier values. A construction
1394        such as {1111111111111111} would give undefined results. What is worse, if
1395        a minimum quantifier for a parenthesized subpattern overflowed and became
1396        negative, the calculation of the memory size went wrong. This could have
1397        led to memory overwriting.
1398    
1399     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1400    
1401     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1402        operating environments where this matters.
1403    
1404     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1405        PCRE options from within the C++ wrapper.
1406    
1407     5. Named capturing subpatterns were not being correctly counted when a pattern
1408        was compiled. This caused two problems: (a) If there were more than 100
1409        such subpatterns, the calculation of the memory needed for the whole
1410        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1411        back references of the form \12, where the number was greater than 9, were
1412        not recognized as back references, even though there were sufficient
1413        previous subpatterns.
1414    
1415     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1416        versions of gcc, e.g. 2.95.4.
1417    
1418    
1419    Version 6.1 21-Jun-05
1420    ---------------------
1421    
1422     1. There was one reference to the variable "posix" in pcretest.c that was not
1423        surrounded by "#if !defined NOPOSIX".
1424    
1425     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1426        the cross-check on the old pcre_info() function, for the benefit of the
1427        cut-down version of PCRE that is currently imported into Exim.
1428    
1429     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1430        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1431        patterns that start (?i) but I don't think that matters. The use of (?i) is
1432        just an example; this all applies to the other options as well.
1433    
1434     4. Since libtool seems to echo the compile commands it is issuing, the output
1435        from "make" can be reduced a bit by putting "@" in front of each libtool
1436        compile command.
1437    
1438     5. Patch from the folks at Google for configure.in to be a bit more thorough
1439        in checking for a suitable C++ installation before trying to compile the
1440        C++ stuff. This should fix a reported problem when a compiler was present,
1441        but no suitable headers.
1442    
1443     6. The man pages all had just "PCRE" as their title. I have changed them to
1444        be the relevant file name. I have also arranged that these names are
1445        retained in the file doc/pcre.txt, which is a concatenation in text format
1446        of all the man pages except the little individual ones for each function.
1447    
1448     7. The NON-UNIX-USE file had not been updated for the different set of source
1449        files that come with release 6. I also added a few comments about the C++
1450        wrapper.
1451    
1452    
1453    Version 6.0 07-Jun-05
1454    ---------------------
1455    
1456     1. Some minor internal re-organization to help with my DFA experiments.
1457    
1458     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1459        didn't matter for the library itself when fully configured, but did matter
1460        when compiling without UCP support, or within Exim, where the ucp files are
1461        not imported.
1462    
1463     3. Refactoring of the library code to split up the various functions into
1464        different source modules. The addition of the new DFA matching code (see
1465        below) to a single monolithic source would have made it really too
1466        unwieldy, quite apart from causing all the code to be include in a
1467        statically linked application, when only some functions are used. This is
1468        relevant even without the DFA addition now that patterns can be compiled in
1469        one application and matched in another.
1470    
1471        The downside of splitting up is that there have to be some external
1472        functions and data tables that are used internally in different modules of
1473        the library but which are not part of the API. These have all had their
1474        names changed to start with "_pcre_" so that they are unlikely to clash
1475        with other external names.
1476    
1477     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1478        a different (DFA) algorithm. Although it is slower than the original
1479        function, it does have some advantages for certain types of matching
1480        problem.
1481    
1482     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1483        including restarting after a partial match.
1484    
1485     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1486        defined when compiling for Windows was sent to me. I have put it into the
1487        code, though I have no means of testing or verifying it.
1488    
1489     7. Added the pcre_refcount() auxiliary function.
1490    
1491     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1492        match before or at the first newline in the subject string. In pcretest,
1493        the /f option on a pattern can be used to set this.
1494    
1495     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1496        would behave wrongly. This has been present in PCRE since release 4.0.
1497    
1498    10. A number of changes to the pcregrep command:
1499    
1500        (a) Refactored how -x works; insert ^(...)$ instead of setting
1501            PCRE_ANCHORED and checking the length, in preparation for adding
1502            something similar for -w.
1503    
1504        (b) Added the -w (match as a word) option.
1505    
1506        (c) Refactored the way lines are read and buffered so as to have more
1507            than one at a time available.
1508    
1509        (d) Implemented a pcregrep test script.
1510    
1511        (e) Added the -M (multiline match) option. This allows patterns to match
1512            over several lines of the subject. The buffering ensures that at least
1513            8K, or the rest of the document (whichever is the shorter) is available
1514            for matching (and similarly the previous 8K for lookbehind assertions).
1515    
1516        (f) Changed the --help output so that it now says
1517    
1518              -w, --word-regex(p)
1519    
1520            instead of two lines, one with "regex" and the other with "regexp"
1521            because that confused at least one person since the short forms are the
1522            same. (This required a bit of code, as the output is generated
1523            automatically from a table. It wasn't just a text change.)
1524    
1525        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1526            option but starts with a hyphen. Could be a pattern or a path name
1527            starting with a hyphen, for instance.
1528    
1529        (h) "-" can be given as a file name to represent stdin.
1530    
1531        (i) When file names are being printed, "(standard input)" is used for
1532            the standard input, for compatibility with GNU grep. Previously
1533            "<stdin>" was used.
1534    
1535        (j) The option --label=xxx can be used to supply a name to be used for
1536            stdin when file names are being printed. There is no short form.
1537    
1538        (k) Re-factored the options decoding logic because we are going to add
1539            two more options that take data. Such options can now be given in four
1540            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1541    
1542        (l) Added the -A, -B, and -C options for requesting that lines of context
1543            around matches be printed.
1544    
1545        (m) Added the -L option to print the names of files that do not contain
1546            any matching lines, that is, the complement of -l.
1547    
1548        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1549            continue to scan other files.
1550    
1551        (o) The -s option was incorrectly implemented. For compatibility with other
1552            greps, it now suppresses the error message for a non-existent or non-
1553            accessible file (but not the return code). There is a new option called
1554            -q that suppresses the output of matching lines, which was what -s was
1555            previously doing.
1556    
1557        (p) Added --include and --exclude options to specify files for inclusion
1558            and exclusion when recursing.
1559    
1560    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1561        Hopefully, it now does.
1562    
1563    12. Missing cast in pcre_study().
1564    
1565    13. Added an "uninstall" target to the makefile.
1566    
1567    14. Replaced "extern" in the function prototypes in Makefile.in with
1568        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1569        world, but is set differently for Windows.
1570    
1571    15. Added a second compiling function called pcre_compile2(). The only
1572        difference is that it has an extra argument, which is a pointer to an
1573        integer error code. When there is a compile-time failure, this is set
1574        non-zero, in addition to the error test pointer being set to point to an
1575        error message. The new argument may be NULL if no error number is required
1576        (but then you may as well call pcre_compile(), which is now just a
1577        wrapper). This facility is provided because some applications need a
1578        numeric error indication, but it has also enabled me to tidy up the way
1579        compile-time errors are handled in the POSIX wrapper.
1580    
1581    16. Added VPATH=.libs to the makefile; this should help when building with one
1582        prefix path and installing with another. (Or so I'm told by someone who
1583        knows more about this stuff than I do.)
1584    
1585    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1586        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1587        match everything, including newlines. This is not POSIX-compatible, but
1588        somebody wanted the feature. From pcretest it can be activated by using
1589        both the P and the s flags.
1590    
1591    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1592    
1593    19. libpcre.pc was being incorrectly installed as executable.
1594    
1595    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1596        it now also looks for '\r' so that it will work unmodified on Windows.
1597    
1598    21. Added Google's contributed C++ wrapper to the distribution.
1599    
1600    22. Added some untidy missing memory free() calls in pcretest, to keep
1601        Electric Fence happy when testing.
1602    
1603    
1604    
1605    Version 5.0 13-Sep-04
1606    ---------------------
1607    
1608     1. Internal change: literal characters are no longer packed up into items
1609        containing multiple characters in a single byte-string. Each character
1610        is now matched using a separate opcode. However, there may be more than one
1611        byte in the character in UTF-8 mode.
1612    
1613     2. The pcre_callout_block structure has two new fields: pattern_position and
1614        next_item_length. These contain the offset in the pattern to the next match
1615        item, and its length, respectively.
1616    
1617     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1618        insertion of callouts before each pattern item. Added the /C option to
1619        pcretest to make use of this.
1620    
1621     4. On the advice of a Windows user, the lines
1622    
1623          #if defined(_WIN32) || defined(WIN32)
1624          _setmode( _fileno( stdout ), 0x8000 );
1625          #endif  /* defined(_WIN32) || defined(WIN32) */
1626    
1627        have been added to the source of pcretest. This apparently does useful
1628        magic in relation to line terminators.
1629    
1630     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1631        for the benefit of those environments where the "b" makes a difference.
1632    
1633     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1634        to know about it. I have put a hack into configure.in that adds in code
1635        to set GCC=yes if CC=icc. This seems to end up at a point in the
1636        generated configure script that is early enough to affect the setting of
1637        compiler options, which is what is needed, but I have no means of testing
1638        whether it really works. (The user who reported this had patched the
1639        generated configure script, which of course I cannot do.)
1640    
1641        LATER: After change 22 below (new libtool files), the configure script
1642        seems to know about icc (and also ecc). Therefore, I have commented out
1643        this hack in configure.in.
1644    
1645     7. Added support for pkg-config (2 patches were sent in).
1646    
1647     8. Negated POSIX character classes that used a combination of internal tables
1648        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1649        [[:^ascii]]. Typically, they would match almost any characters. The other
1650        POSIX classes were not broken in this way.
1651    
1652     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1653        to find the match, as PCRE was deluded into thinking that the match had to
1654        start at the start point or following a newline. The same bug applied to
1655        patterns with negative forward assertions or any backward assertions
1656        preceding ".*" at the start, unless the pattern required a fixed first
1657        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1658    
1659    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1660        starting at the last subject character, bytes beyond the end of the subject
1661        string were read.
1662    
1663    11. Renamed the variable "class" as "classbits" to make life easier for C++
1664        users. (Previously there was a macro definition, but it apparently wasn't
1665        enough.)
1666    
1667    12. Added the new field "tables" to the extra data so that tables can be passed
1668        in at exec time, or the internal tables can be re-selected. This allows
1669        a compiled regex to be saved and re-used at a later time by a different
1670        program that might have everything at different addresses.
1671    
1672    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1673        -R library as well as a -L library.
1674    
1675    14. The debugging options of pcretest (-d on the command line or D on a
1676        pattern) showed incorrect output for anything following an extended class
1677        that contained multibyte characters and which was followed by a quantifier.
1678    
1679    15. Added optional support for general category Unicode character properties
1680        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1681        support. It adds about 90K to the size of the library. The meanings of the
1682        inbuilt class escapes such as \d and \s have NOT been changed.
1683    
1684    16. Updated pcredemo.c to include calls to free() to release the memory for the
1685        compiled pattern.
1686    
1687    17. The generated file chartables.c was being created in the source directory
1688        instead of in the building directory. This caused the build to fail if the
1689        source directory was different from the building directory, and was
1690        read-only.
1691    
1692    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1693        file. No doubt somebody will tell me if they don't make sense... Also added
1694        Dan Mooney's comments about building on OpenVMS.
1695    
1696    19. Added support for partial matching via the PCRE_PARTIAL option for
1697        pcre_exec() and the \P data escape in pcretest.
1698    
1699    20. Extended pcretest with 3 new pattern features:
1700    
1701        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1702              write the compiled pattern to the file whose name is "rest-of-line".
1703              This is a straight binary dump of the data, with the saved pointer to
1704              the character tables forced to be NULL. The study data, if any, is
1705              written too. After writing, pcretest reads a new pattern.
1706    
1707        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1708              compiled pattern from the given file. There must not be any
1709              occurrences of "<" in the file name (pretty unlikely); if there are,
1710              pcretest will instead treat the initial "<" as a pattern delimiter.
1711              After reading in the pattern, pcretest goes on to read data lines as
1712              usual.
1713    
1714        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1715              and 16-bit fields in a compiled pattern, to simulate a pattern that
1716              was compiled on a host of opposite endianness.
1717    
1718    21. The pcre-exec() function can now cope with patterns that were compiled on
1719        hosts of opposite endianness, with this restriction:
1720    
1721          As for any compiled expression that is saved and used later, the tables
1722          pointer field cannot be preserved; the extra_data field in the arguments
1723          to pcre_exec() should be used to pass in a tables address if a value
1724          other than the default internal tables were used at compile time.
1725    
1726    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1727        now diagnosed as an error. Previously, most of the time, a negative number
1728        would have been treated as zero, but if in addition "ovector" was passed as
1729        NULL, a crash could occur.
1730    
1731    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1732        new versions from the libtool 1.5 distribution (the last one is a copy of
1733        a file called libtool.m4). This seems to have fixed the need to patch
1734        "configure" to support Darwin 1.3 (which I used to do). However, I still
1735        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1736        workstation).
1737    
1738    24. Changed the PCRE licence to be the more standard "BSD" licence.
1739    
1740    
1741    Version 4.5 01-Dec-03
1742    ---------------------
1743    
1744     1. There has been some re-arrangement of the code for the match() function so
1745        that it can be compiled in a version that does not call itself recursively.
1746        Instead, it keeps those local variables that need separate instances for
1747        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1748        needs to "recurse". Keeping track of where control must go is done by means
1749        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1750        hide most of the details from the main code, and operates only if
1751        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1752        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1753        operating.
1754    
1755        To make it easier for callers to provide specially tailored get/free
1756        functions for this usage, two new functions, pcre_stack_malloc, and
1757        pcre_stack_free, are used. They are always called in strict stacking order,
1758        and the size of block requested is always the same.
1759    
1760        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1761        PCRE has been compiled to use the stack or the heap for recursion. The
1762        -C option of pcretest uses this to show which version is compiled.
1763    
1764        A new data escape \S, is added to pcretest; it causes the amounts of store
1765        obtained and freed by both kinds of malloc/free at match time to be added
1766        to the output.
1767    
1768     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1769        what's available on my current Linux desktop machine.
1770    
1771     3. When matching a UTF-8 string, the test for a valid string at the start has
1772        been extended. If start_offset is not zero, PCRE now checks that it points
1773        to a byte that is the start of a UTF-8 character. If not, it returns
1774        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1775        this is necessary because there may be backward assertions in the pattern.
1776        When matching the same subject several times, it may save resources to use
1777        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1778    
1779     4. The code for checking the validity of UTF-8 strings has been tightened so
1780        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1781        containing "overlong sequences".
1782    
1783     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1784        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1785        should have been "&", but it just so happened that all the cases this let
1786        through by mistake were picked up later in the function.
1787    
1788     6. I had used a variable called "isblank" - this is a C99 function, causing
1789        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1790    
1791     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1792        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1793        so the version is shown; (c) stop "make test" echoing "./RunTest".
1794    
1795     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1796    
1797     9. The prototype for memmove() for systems that don't have it was using
1798        size_t, but the inclusion of the header that defines size_t was later. I've
1799        moved the #includes for the C headers earlier to avoid this.
1800    
1801    10. Added some adjustments to the code to make it easier to compiler on certain
1802        special systems:
1803    
1804          (a) Some "const" qualifiers were missing.
1805          (b) Added the macro EXPORT before all exported functions; by default this
1806              is defined to be empty.
1807          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1808              that it reads its output file name as an argument instead of writing
1809              to the standard output and assuming this can be redirected.
1810    
1811    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1812        class containing characters with values greater than 255, PCRE compilation
1813        went into a loop.
1814    
1815    12. A recursive reference to a subpattern that was within another subpattern
1816        that had a minimum quantifier of zero caused PCRE to crash. For example,
1817        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1818        recursion. If the recursively-called subpattern itself had a zero repeat,
1819        that was OK.
1820    
1821    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1822        buffer into which it was copied (for escape processing) was still set at
1823        1024, so long lines caused crashes.
1824    
1825    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1826        "internal error: code overflow...". This applied to any character class
1827        that was followed by a possessive quantifier.
1828    
1829    15. Modified the Makefile to add libpcre.la as a prerequisite for
1830        libpcreposix.la because I was told this is needed for a parallel build to
1831        work.
1832    
1833    16. If a pattern that contained .* following optional items at the start was
1834        studied, the wrong optimizing data was generated, leading to matching
1835        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1836        matching string must start with a or b or c. The correct conclusion for
1837        this pattern is that a match can start with any character.
1838    
1839    
1840    Version 4.4 13-Aug-03
1841    ---------------------
1842    
1843     1. In UTF-8 mode, a character class containing characters with values between
1844        127 and 255 was not handled correctly if the compiled pattern was studied.
1845        In fixing this, I have also improved the studying algorithm for such
1846        classes (slightly).
1847    
1848     2. Three internal functions had redundant arguments passed to them. Removal
1849        might give a very teeny performance improvement.
1850    
1851     3. Documentation bug: the value of the capture_top field in a callout is *one
1852        more than* the number of the hightest numbered captured substring.
1853    
1854     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1855        in incorrectly linking with a previously installed version. They now link
1856        explicitly with libpcre.la.
1857    
1858     5. configure.in no longer needs to recognize Cygwin specially.
1859    
1860     6. A problem in pcre.in for Windows platforms is fixed.
1861    
1862     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1863        pcretest, it used to include the size of the study block as part of its
1864        output. Unfortunately, the structure contains a field that has a different
1865        size on different hardware architectures. This meant that the tests that
1866        showed this size failed. As the block is currently always of a fixed size,
1867        this information isn't actually particularly useful in pcretest output, so
1868        I have just removed it.
1869    
1870     8. Three pre-processor statements accidentally did not start in column 1.
1871        Sadly, there are *still* compilers around that complain, even though
1872        standard C has not required this for well over a decade. Sigh.
1873    
1874     9. In pcretest, the code for checking callouts passed small integers in the
1875        callout_data field, which is a void * field. However, some picky compilers
1876        complained about the casts involved for this on 64-bit systems. Now
1877        pcretest passes the address of the small integer instead, which should get
1878        rid of the warnings.
1879    
1880    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1881        both compile and run time, and gives an error if an invalid UTF-8 sequence
1882        is found. There is a option for disabling this check in cases where the
1883        string is known to be correct and/or the maximum performance is wanted.
1884    
1885    11. In response to a bug report, I changed one line in Makefile.in from
1886    
1887            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1888        to
1889            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1890    
1891        to look similar to other lines, but I have no way of telling whether this
1892        is the right thing to do, as I do not use Windows. No doubt I'll get told
1893        if it's wrong...
1894    
1895    
1896    Version 4.3 21-May-03
1897    ---------------------
1898    
1899    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1900       Makefile.
1901    
1902    2. Some refactoring to improve the quality of the code:
1903    
1904       (i)   The utf8_table... variables are now declared "const".
1905    
1906       (ii)  The code for \cx, which used the "case flipping" table to upper case
1907             lower case letters, now just substracts 32. This is ASCII-specific,
1908             but the whole concept of \cx is ASCII-specific, so it seems
1909             reasonable.
1910    
1911       (iii) PCRE was using its character types table to recognize decimal and
1912             hexadecimal digits in the pattern. This is silly, because it handles
1913             only 0-9, a-f, and A-F, but the character types table is locale-
1914             specific, which means strange things might happen. A private
1915             table is now used for this - though it costs 256 bytes, a table is
1916             much faster than multiple explicit tests. Of course, the standard
1917             character types table is still used for matching digits in subject
1918             strings against \d.
1919    
1920       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1921             ending in _t are). So I've renamed it as ESC_tee.
1922    
1923    3. The first argument for regexec() in the POSIX wrapper should have been
1924       defined as "const".
1925    
1926    4. Changed pcretest to use malloc() for its buffers so that they can be
1927       Electric Fenced for debugging.
1928    
1929    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1930       to read one or more bytes before the start of the subject string. Often this
1931       had no effect on PCRE's behaviour, but in some circumstances it could
1932       provoke a segmentation fault.
1933    
1934    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1935       to try to read one or more bytes before the start of the subject string.
1936    
1937    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1938       UTF-8 support could misbehave in various ways if the subject string
1939       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1940       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1941       back over UTF-8 characters.)
1942    
1943    
1944    Version 4.2 14-Apr-03
1945    ---------------------
1946    
1947    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1948    
1949    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1950         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1951         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1952         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1953         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1954           and BUILD_EXEEXT
1955         Note: automatic setting of the BUILD variables is not yet working
1956         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1957           compile-time but not at link-time
1958         [LINK]: use for linking executables only
1959         make different versions for Windows and non-Windows
1960         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1961           libraries
1962         [LINK_FOR_BUILD]: new variable
1963         [OBJEXT]: use throughout
1964         [EXEEXT]: use throughout
1965         <winshared>: new target
1966         <wininstall>: new target
1967         <dftables.o>: use native compiler
1968         <dftables>: use native linker
1969         <install>: handle Windows platform correctly
1970         <clean>: ditto
1971         <check>: ditto
1972         copy DLL to top builddir before testing
1973    
1974       As part of these changes, -no-undefined was removed again. This was reported
1975       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1976       in any case.
1977    
1978    3. Some tidies to get rid of compiler warnings:
1979    
1980       . In the match_data structure, match_limit was an unsigned long int, whereas
1981         match_call_count was an int. I've made them both unsigned long ints.
1982    
1983       . In pcretest the fact that a const uschar * doesn't automatically cast to
1984         a void * provoked a warning.
1985    
1986       . Turning on some more compiler warnings threw up some "shadow" variables
1987         and a few more missing casts.
1988    
1989    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1990       option, a class that contained a single character with a value between 128
1991       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1992    
1993    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1994       option, a class that contained several characters, but with at least one
1995       whose value was between 128 and 255 caused PCRE to crash.
1996    
1997    
1998    Version 4.1 12-Mar-03
1999    ---------------------
2000    
2001    1. Compiling with gcc -pedantic found a couple of places where casts were
2002    needed, and a string in dftables.c that was longer than standard compilers are
2003    required to support.
2004    
2005    2. Compiling with Sun's compiler found a few more places where the code could
2006    be tidied up in order to avoid warnings.
2007    
2008    3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2009    first of these names is deprecated in the latest Autoconf in favour of the name
2010    CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2011    compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2012    analogy I have changed it to CFLAGS_FOR_BUILD.
2013    
2014    4. Added -no-undefined to the linking command in the Makefile, because this is
2015    apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2016    linking step for the pcreposix library.
2017    
2018    5. PCRE was failing to diagnose the case of two named groups with the same
2019    name.
2020    
2021    6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2022    literal character that is needed in the subject for a match, and scans along to
2023    ensure that it is present before embarking on the full matching process. This
2024    saves time in cases of nested unlimited repeats that are never going to match.
2025    Problem: the scan can take a lot of time if the subject is very long (e.g.
2026    megabytes), thus penalizing straightforward matches. It is now done only if the
2027    amount of subject to be scanned is less than 1000 bytes.
2028    
2029    7. A lesser problem with the same optimization is that it was recording the
2030    first character of an anchored pattern as "needed", thus provoking a search
2031    right along the subject, even when the first match of the pattern was going to
2032    fail. The "needed" character is now not set for anchored patterns, unless it
2033    follows something in the pattern that is of non-fixed length. Thus, it still
2034    fulfils its original purpose of finding quick non-matches in cases of nested
2035    unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2036    
2037    
2038    Version 4.0 17-Feb-03
2039    ---------------------
2040    
2041    1. If a comment in an extended regex that started immediately after a meta-item
2042    extended to the end of string, PCRE compiled incorrect data. This could lead to
2043    all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2044    
2045    2. Moved to autoconf 2.53 and libtool 1.4.2.
2046    
2047    3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2048    the special perltest8 script is no longer needed - all the tests can be run
2049    from a single perltest script.
2050    
2051    4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2052    by \s. It has now been removed in PCRE. This means it isn't recognized as
2053    whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2054    class [:space:] *does* include VT, thereby creating a mess.
2055    
2056    5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2057    space and tab.
2058    
2059    6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2060    its new features into the main test script, reducing the number of scripts.
2061    
2062    7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2063    were backward compatible, and made the (?i) apply to the whole pattern, as if
2064    /i were given. Now it behaves more logically, and applies the option setting
2065    only to what follows. PCRE has been changed to follow suit. However, if it
2066    finds options settings right at the start of the pattern, it extracts them into
2067    the global options, as before. Thus, they show up in the info data.
2068    
2069    8. Added support for the \Q...\E escape sequence. Characters in between are
2070    treated as literals. This is slightly different from Perl in that $ and @ are
2071    also handled as literals inside the quotes. In Perl, they will cause variable
2072    interpolation. Note the following examples:
2073    
2074        Pattern            PCRE matches      Perl matches
2075    
2076        \Qabc$xyz\E        abc$xyz           abc followed by the contents of $xyz
2077        \Qabc\$xyz\E       abc\$xyz          abc\$xyz
2078        \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
2079    
2080    For compatibility with Perl, \Q...\E sequences are recognized inside character
2081    classes as well as outside them.
2082    
2083    9. Re-organized 3 code statements in pcretest to avoid "overflow in
2084    floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2085    (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2086    signed/unsigned warnings.
2087    
2088    10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2089    option for pcretest, so I've replaced it by a simple function that does just
2090    that job.
2091    
2092    11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2093    "pcregrep -".
2094    
2095    12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2096    Java package. This provides some syntactic sugar for simple cases of what my
2097    documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2098    as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2099    item, you can use this simplified notation. Note that only makes sense with
2100    greedy quantifiers. Consequently, the use of the possessive quantifier forces
2101    greediness, whatever the setting of the PCRE_UNGREEDY option.
2102    
2103    13. A change of greediness default within a pattern was not taking effect at
2104    the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2105    subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2106    was abstracted outside.
2107    
2108    14. PCRE now supports the \G assertion. It is true when the current matching
2109    position is at the start point of the match. This differs from \A when the
2110    starting offset is non-zero. Used with the /g option of pcretest (or similar
2111    code), it works in the same way as it does for Perl's /g option. If all
2112    alternatives of a regex begin with \G, the expression is anchored to the start
2113    match position, and the "anchored" flag is set in the compiled expression.
2114    
2115    15. Some bugs concerning the handling of certain option changes within patterns
2116    have been fixed. These applied to options other than (?ims). For example,
2117    "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2118    been the other way round. Some of this was related to change 7 above.
2119    
2120    16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2121    features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2122    and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2123    POSIX classes only within a class (e.g. /[[:alpha:]]/).
2124    
2125    17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2126    mode. Unlike ".", it always matches newline, whatever the setting of
2127    PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2128    assertions. Perl allows it, but it doesn't (in general) work because it can't
2129    calculate the length of the lookbehind. At least, that's the case for Perl
2130    5.8.0 - I've been told they are going to document that it doesn't work in
2131    future.
2132    
2133    18. Added an error diagnosis for escapes that PCRE does not support: these are
2134    \L, \l, \N, \P, \p, \U, \u, and \X.
2135    
2136    19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2137    reading past the end of the pattern in cases such as /[abcd/.
2138    
2139    20. PCRE was getting more memory than necessary for patterns with classes that
2140    contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2141    
2142    21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2143    compiling PCRE for use with Virtual Pascal.
2144    
2145    22. Small fix to the Makefile to make it work properly if the build is done
2146    outside the source tree.
2147    
2148    23. Added a new extension: a condition to go with recursion. If a conditional
2149    subpattern starts with (?(R) the "true" branch is used if recursion has
2150    happened, whereas the "false" branch is used only at the top level.
2151    
2152    24. When there was a very long string of literal characters (over 255 bytes
2153    without UTF support, over 250 bytes with UTF support), the computation of how
2154    much memory was required could be incorrect, leading to segfaults or other
2155    strange effects.
2156    
2157    25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2158    start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2159    there was a subsequent back reference to those brackets. This meant that, for
2160    example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2161    possible to check for precisely this case. All we can do is abandon the
2162    optimization if .* occurs inside capturing brackets when there are any back
2163    references whatsoever. (See below for a better fix that came later.)
2164    
2165    26. The handling of the optimization for finding the first character of a
2166    non-anchored pattern, and for finding a character that is required later in the
2167    match were failing in some cases. This didn't break the matching; it just
2168    failed to optimize when it could. The way this is done has been re-implemented.
2169    
2170    27. Fixed typo in error message for invalid (?R item (it said "(?p").
2171    
2172    28. Added a new feature that provides some of the functionality that Perl
2173    provides with (?{...}). The facility is termed a "callout". The way it is done
2174    in PCRE is for the caller to provide an optional function, by setting
2175    pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2176    global variable. By default it is unset, which disables all calling out. To get
2177    the function called, the regex must include (?C) at appropriate points. This
2178    is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2179    This provides a means of identifying different callout points. When PCRE
2180    reaches such a point in the regex, if pcre_callout has been set, the external
2181    function is called. It is provided with data in a structure called
2182    pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2183    matching continues; if it returns a non-zero value, the match at the current
2184    point fails. However, backtracking will occur if possible. [This was changed
2185    later and other features added - see item 49 below.]
2186    
2187    29. pcretest is upgraded to test the callout functionality. It provides a
2188    callout function that displays information. By default, it shows the start of
2189    the match and the current position in the text. There are some new data escapes
2190    to vary what happens:
2191    
2192        \C+         in addition, show current contents of captured substrings
2193        \C-         do not supply a callout function
2194        \C!n        return 1 when callout number n is reached
2195        \C!n!m      return 1 when callout number n is reached for the mth time
2196    
2197    30. If pcregrep was called with the -l option and just a single file name, it
2198    output "<stdin>" if a match was found, instead of the file name.
2199    
2200    31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2201    slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2202    pcre_exec(). This saves a malloc/free per call. The default value of
2203    POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2204    when configuring.
2205    
2206    32. The default maximum size of a compiled pattern is 64K. There have been a
2207    few cases of people hitting this limit. The code now uses macros to handle the
2208    storing of links as offsets within the compiled pattern. It defaults to 2-byte
2209    links, but this can be changed to 3 or 4 bytes by --with-link-size when
2210    configuring. Tests 2 and 5 work only with 2-byte links because they output
2211    debugging information about compiled patterns.
2212    
2213    33. Internal code re-arrangements:
2214    
2215    (a) Moved the debugging function for printing out a compiled regex into
2216        its own source file (printint.c) and used #include to pull it into
2217        pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2218        separate copies.
2219    
2220    (b) Defined the list of op-code names for debugging as a macro in
2221        internal.h so that it is next to the definition of the opcodes.
2222    
2223    (c) Defined a table of op-code lengths for simpler skipping along compiled
2224        code. This is again a macro in internal.h so that it is next to the
2225        definition of the opcodes.
2226    
2227    34. Added support for recursive calls to individual subpatterns, along the
2228    lines of Robin Houston's patch (but implemented somewhat differently).
2229    
2230    35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2231    allow it to read and process whole directories in Win32. This code was
2232    contributed by Lionel Fourquaux; it has not been tested by me.
2233    
2234    36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2235    used to name a group. Names consist of alphanumerics and underscores, and must
2236    be unique. Back references use the syntax (?P=name) and recursive calls use
2237    (?P>name) which is a PCRE extension to the Python extension. Groups still have
2238    numbers. The function pcre_fullinfo() can be used after compilation to extract
2239    a name/number map. There are three relevant calls:
2240    
2241      PCRE_INFO_NAMEENTRYSIZE        yields the size of each entry in the map
2242      PCRE_INFO_NAMECOUNT            yields the number of entries
2243      PCRE_INFO_NAMETABLE            yields a pointer to the map.
2244    
2245    The map is a vector of fixed-size entries. The size of each entry depends on
2246    the length of the longest name used. The first two bytes of each entry are the
2247    group number, most significant byte first. There follows the corresponding
2248    name, zero terminated. The names are in alphabetical order.
2249    
2250    37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2251    case instead of 255. Making it the same both with and without UTF-8 support
2252    means that the same test output works with both.
2253    
2254    38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2255    calling malloc() with a zero argument.
2256    
2257    39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2258    optimization. I've improved things by keeping a bitmap of backreferences with
2259    numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2260    fact referenced, the optimization can be applied. It is unlikely that a
2261    relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2262    the match to follow \n) will appear inside brackets with a number greater than
2263    31, but if it does, any back reference > 31 suppresses the optimization.
2264    
2265    40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2266    of disabling numbered capturing parentheses. Any opening parenthesis that is
2267    not followed by ? behaves as if it were followed by ?: but named parentheses
2268    can still be used for capturing (and they will acquire numbers in the usual
2269    way).
2270    
2271    41. Redesigned the return codes from the match() function into yes/no/error so
2272    that errors can be passed back from deep inside the nested calls. A malloc
2273    failure while inside a recursive subpattern call now causes the
2274    PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2275    
2276    42. It is now possible to set a limit on the number of times the match()
2277    function is called in a call to pcre_exec(). This facility makes it possible to
2278    limit the amount of recursion and backtracking, though not in a directly
2279    obvious way, because the match() function is used in a number of different
2280    circumstances. The count starts from zero for each position in the subject
2281    string (for non-anchored patterns). The default limit is, for compatibility, a
2282    large number, namely 10 000 000. You can change this in two ways:
2283    
2284    (a) When configuring PCRE before making, you can use --with-match-limit=n
2285        to set a default value for the compiled library.
2286    
2287    (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2288        a different value is set. See 45 below.
2289    
2290    If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2291    
2292    43. Added a new function pcre_config(int, void *) to enable run-time extraction
2293    of things that can be changed at compile time. The first argument specifies
2294    what is wanted and the second points to where the information is to be placed.
2295    The current list of available information is:
2296    
2297      PCRE_CONFIG_UTF8
2298    
2299    The output is an integer that is set to one if UTF-8 support is available;
2300    otherwise it is set to zero.
2301    
2302      PCRE_CONFIG_NEWLINE
2303    
2304    The output is an integer that it set to the value of the code that is used for
2305    newline. It is either LF (10) or CR (13).
2306    
2307      PCRE_CONFIG_LINK_SIZE
2308    
2309    The output is an integer that contains the number of bytes used for internal
2310    linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2311    
2312      PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
2313    
2314    The output is an integer that contains the threshold above which the POSIX
2315    interface uses malloc() for output vectors. See item 31 above.
2316    
2317      PCRE_CONFIG_MATCH_LIMIT
2318    
2319    The output is an unsigned integer that contains the default limit of the number
2320    of match() calls in a pcre_exec() execution. See 42 above.
2321    
2322    44. pcretest has been upgraded by the addition of the -C option. This causes it
2323    to extract all the available output from the new pcre_config() function, and to
2324    output it. The program then exits immediately.
2325    
2326    45. A need has arisen to pass over additional data with calls to pcre_exec() in
2327    order to support additional features. One way would have been to define
2328    pcre_exec2() (for example) with extra arguments, but this would not have been
2329    extensible, and would also have required all calls to the original function to
2330    be mapped to the new one. Instead, I have chosen to extend the mechanism that
2331    is used for passing in "extra" data from pcre_study().
2332    
2333    The pcre_extra structure is now exposed and defined in pcre.h. It currently
2334    contains the following fields:
2335    
2336      flags         a bitmap indicating which of the following fields are set
2337      study_data    opaque data from pcre_study()
2338      match_limit   a way of specifying a limit on match() calls for a specific
2339                      call to pcre_exec()
2340      callout_data  data for callouts (see 49 below)
2341    
2342    The flag bits are also defined in pcre.h, and are
2343    
2344      PCRE_EXTRA_STUDY_DATA
2345      PCRE_EXTRA_MATCH_LIMIT
2346      PCRE_EXTRA_CALLOUT_DATA
2347    
2348    The pcre_study() function now returns one of these new pcre_extra blocks, with
2349    the actual study data pointed to by the study_data field, and the
2350    PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2351    before. That is, this change is entirely upwards-compatible and requires no
2352    change to existing code.
2353    
2354    If you want to pass in additional data to pcre_exec(), you can either place it
2355    in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2356    block.
2357    
2358    46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2359    data string contains the escape sequence \M, pcretest calls pcre_exec() several
2360    times with different match limits, until it finds the minimum value needed for
2361    pcre_exec() to complete. The value is then output. This can be instructive; for
2362    most simple matches the number is quite small, but for pathological cases it
2363    gets very large very quickly.
2364    
2365    47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2366    returns the size of the data block pointed to by the study_data field in a
2367    pcre_extra block, that is, the value that was passed as the argument to
2368    pcre_malloc() when PCRE was getting memory in which to place the information
2369    created by pcre_study(). The fourth argument should point to a size_t variable.
2370    pcretest has been extended so that this information is shown after a successful
2371    pcre_study() call when information about the compiled regex is being displayed.
2372    
2373    48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2374    because what follows is always an absolute path. (Later: it turns out that this
2375    is more than cosmetic for MinGW, because it doesn't like empty path
2376    components.)
2377    
2378    49. Some changes have been made to the callout feature (see 28 above):
2379    
2380    (i)  A callout function now has three choices for what it returns:
2381    
2382           0  =>  success, carry on matching
2383         > 0  =>  failure at this point, but backtrack if possible
2384         < 0  =>  serious error, return this value from pcre_exec()
2385    
2386         Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2387         values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2388         "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2389         use by callout functions. It will never be used by PCRE itself.
2390    
2391    (ii) The pcre_extra structure (see 45 above) has a void * field called
2392         callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2393         pcre_callout_block structure has a field of the same name. The contents of
2394         the field passed in the pcre_extra structure are passed to the callout
2395         function in the corresponding field in the callout block. This makes it
2396         easier to use the same callout-containing regex from multiple threads. For
2397         testing, the pcretest program has a new data escape
2398    
2399           \C*n        pass the number n (may be negative) as callout_data
2400    
2401         If the callout function in pcretest receives a non-zero value as
2402         callout_data, it returns that value.
2403    
2404    50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2405    there were some redundant $(CFLAGS) in commands that are now specified as
2406    $(LINK), which already includes $(CFLAGS).
2407    
2408    51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2409    has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2410    with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2411    one-byte characters throughout. Note that case-insensitive matching applies
2412    only to characters whose values are less than 256. PCRE doesn't support the
2413    notion of cases for higher-valued characters.
2414    
2415    (i)   A character class whose characters are all within 0-255 is handled as
2416          a bit map, and the map is inverted for negative classes. Previously, a
2417          character > 255 always failed to match such a class; however it should
2418          match if the class was a negative one (e.g. [^ab]). This has been fixed.
2419    
2420    (ii)  A negated character class with a single character < 255 is coded as
2421          "not this character" (OP_NOT). This wasn't working properly when the test
2422          character was multibyte, either singly or repeated.
2423    
2424    (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2425          mode, for example: \x{100}{2,3}.
2426    
2427    (iv)  The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2428          singly or repeated) now correctly test multibyte characters. However,
2429          PCRE doesn't recognize any characters with values greater than 255 as
2430          digits, spaces, or word characters. Such characters always match \D, \S,
2431          and \W, and never match \d, \s, or \w.
2432    
2433    (v)   Classes may now contain characters and character ranges with values
2434          greater than 255. For example: [ab\x{100}-\x{400}].
2435    
2436    (vi)  pcregrep now has a --utf-8 option (synonym -u) which makes it call
2437          PCRE in UTF-8 mode.
2438    
2439    52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2440    PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2441    retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2442    value.)
2443    
2444    53. The single man page has become too large. I have therefore split it up into
2445    a number of separate man pages. These also give rise to individual HTML pages;
2446    these are now put in a separate directory, and there is an index.html page that
2447    lists them all. Some hyperlinking between the pages has been installed.
2448    
2449    54. Added convenience functions for handling named capturing parentheses.
2450    
2451    55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2452    aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2453    true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2454    are faulted.
2455    
2456    56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2457    calling configure. These values are used when compiling the dftables.c program
2458    which is run to generate the source of the default character tables. They
2459    default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2460    you will need to set these values.
2461    
2462    57. Updated the building process for Windows DLL, as provided by Fred Cox.
2463    
2464    
2465    Version 3.9 02-Jan-02
2466    ---------------------
2467    
2468    1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2469    
2470    2. If --disable-static was given, the building process failed when trying to
2471    build pcretest and pcregrep. (For some reason it was using libtool to compile
2472    them, which is not right, as they aren't part of the library.)
2473    
2474    
2475    Version 3.8 18-Dec-01
2476    ---------------------
2477    
2478    1. The experimental UTF-8 code was completely screwed up. It was packing the
2479    bytes in the wrong order. How dumb can you get?
2480    
2481    
2482    Version 3.7 29-Oct-01
2483    ---------------------
2484    
2485    1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2486    This caused pcretest, when used on the test data, to segfault. Unfortunately,
2487    this didn't happen under Solaris 8, where I normally test things.
2488    
2489    2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2490    doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2491    isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2492    this fix an hour or so after the initial 3.7 release.)
2493    
2494    
2495    Version 3.6 23-Oct-01
2496    ---------------------
2497    
2498    1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2499    offsets passed as NULL with zero offset count.
2500    
2501    2. The config.guess and config.sub files had not been updated when I moved to
2502    the latest autoconf.
2503    
2504    
2505    Version 3.5 15-Aug-01
2506    ---------------------
2507    
2508    1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2509    had been forgotten.
2510    
2511    2. By using declared but undefined structures, we can avoid using "void"
2512    definitions in pcre.h while keeping the internal definitions of the structures
2513    private.
2514    
2515    3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2516    user point of view, this means that both static and shared libraries are built
2517    by default, but this can be individually controlled. More of the work of
2518    handling this static/shared cases is now inside libtool instead of PCRE's make
2519    file.
2520    
2521    4. The pcretest utility is now installed along with pcregrep because it is
2522    useful for users (to test regexs) and by doing this, it automatically gets
2523    relinked by libtool. The documentation has been turned into a man page, so
2524    there are now .1, .txt, and .html versions in /doc.
2525    
2526    5. Upgrades to pcregrep:
2527       (i)   Added long-form option names like gnu grep.
2528       (ii)  Added --help to list all options with an explanatory phrase.
2529       (iii) Added -r, --recursive to recurse into sub-directories.
2530       (iv)  Added -f, --file to read patterns from a file.
2531    
2532    6. pcre_exec() was referring to its "code" argument before testing that
2533    argument for NULL (and giving an error if it was NULL).
2534    
2535    7. Upgraded Makefile.in to allow for compiling in a different directory from
2536    the source directory.
2537    
2538    8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2539    options bits, the pointer it was passed was to an int instead of to an unsigned
2540    long int. This mattered only on 64-bit systems.
2541    
2542    9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2543    generated) instead of pcre.in, which it its source. Also made the same change
2544    in several of the .c files.
2545    
2546    10. A new release of gcc defines printf() as a macro, which broke pcretest
2547    because it had an ifdef in the middle of a string argument for printf(). Fixed
2548    by using separate calls to printf().
2549    
2550    11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2551    script, to force use of CR or LF instead of \n in the source. On non-Unix
2552    systems, the value can be set in config.h.
2553    
2554    12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2555    absolute limit. Changed the text of the error message to make this clear, and
2556    likewise updated the man page.
2557    
2558    13. The limit of 99 on the number of capturing subpatterns has been removed.
2559    The new limit is 65535, which I hope will not be a "real" limit.
2560    
2561    
2562    Version 3.4 22-Aug-00
2563    ---------------------
2564    
2565    1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2566    
2567    2. Diagnose condition (?(0) as an error instead of crashing on matching.
2568    
2569    
2570    Version 3.3 01-Aug-00
2571    ---------------------
2572    
2573    1. If an octal character was given, but the value was greater than \377, it
2574    was not getting masked to the least significant bits, as documented. This could
2575    lead to crashes in some systems.
2576    
2577    2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2578    the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2579    
2580    3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2581    These just pass their arguments on to (pcre_free)(), but they are provided
2582    because some uses of PCRE bind it to non-C systems that can call its functions,
2583    but cannot call free() or pcre_free() directly.
2584    
2585    4. Add "make test" as a synonym for "make check". Corrected some comments in
2586    the Makefile.
2587    
2588    5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2589    Makefile.
2590    
2591    6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2592    command called pgrep for grepping around the active processes.
2593    
2594    7. Added the beginnings of support for UTF-8 character strings.
2595    
2596    8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2597    RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2598    the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2599    out for the ar command.)
2600    
2601    
2602    Version 3.2 12-May-00
2603    ---------------------
2604    
2605    This is purely a bug fixing release.
2606    
2607    1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2608    of ZA. This was just one example of several cases that could provoke this bug,
2609    which was introduced by change 9 of version 2.00. The code for breaking
2610    infinite loops after an iteration that matches an empty string was't working
2611    correctly.
2612    
2613    2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2614    when matched against abbab (for example). After matching an empty string, it
2615    wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2616    caused it to match further down the string than it should.
2617    
2618    3. The code contained an inclusion of sys/types.h. It isn't clear why this
2619    was there because it doesn't seem to be needed, and it causes trouble on some
2620    systems, as it is not a Standard C header. It has been removed.
2621    
2622    4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2623    were reported on the Macintosh. The changes were from
2624    
2625      while ((c = *(++ptr)) != 0 && c != '\n');
2626    to
2627      while ((c = *(++ptr)) != 0 && c != '\n') ;
2628    
2629    Totally extraordinary, but if that's what it takes...
2630    
2631    5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2632    available. Added HAVE_BCOPY and an autoconf test for it; if neither
2633    HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2634    assumes the way PCRE uses memmove() (always moving upwards).
2635    
2636    6. PCRE is being used in one environment where strchr() is not available. There
2637    was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2638    faster code anyway.
2639    
2640    
2641    Version 3.1 09-Feb-00
2642    ---------------------
2643    
2644    The only change in this release is the fixing of some bugs in Makefile.in for
2645    the "install" target:
2646    
2647    (1) It was failing to install pcreposix.h.
2648    
2649    (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2650    
2651    
2652    Version 3.0 01-Feb-00
2653    ---------------------
2654    
2655    1. Add support for the /+ modifier to perltest (to output $` like it does in
2656    pcretest).
2657    
2658    2. Add support for the /g modifier to perltest.
2659    
2660    3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2661    matches null strings.
2662    
2663    4. Fix perltest so that it doesn't do unwanted things when fed an empty
2664    pattern. Perl treats empty patterns specially - it reuses the most recent
2665    pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2666    effect.
2667    
2668    5. The POSIX interface was broken in that it was just handing over the POSIX
2669    captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2670    required a bigger vector, with some working space on the end. This means that
2671    the POSIX wrapper now has to get and free some memory, and copy the results.
2672    
2673    6. Added some simple autoconf support, placing the test data and the
2674    documentation in separate directories, re-organizing some of the
2675    information files, and making it build pcre-config (a GNU standard). Also added
2676    libtool support for building PCRE as a shared library, which is now the
2677    default.
2678    
2679    7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2680    09 are not valid octal constants. Single digits will be used for minor values
2681    less than 10.
2682    
2683    8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2684    existing programs that set these in the POSIX interface can use PCRE without
2685    modification.
2686    
2687    9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2688    return all that pcre_info() returns, plus additional data. The pcre_info()
2689    function is retained for compatibility, but is considered to be obsolete.
2690    
2691    10. Added experimental recursion feature (?R) to handle one common case that
2692    Perl 5.6 will be able to do with (?p{...}).
2693    
2694    11. Added support for POSIX character classes like [:alpha:], which Perl is
2695    adopting.
2696    
2697    
2698    Version 2.08 31-Aug-99
2699    ----------------------
2700    
2701    1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2702    trying to match at the startoffset position, but instead was moving forward to
2703    the next newline as if a previous match had failed.
2704    
2705    2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2706    and could get into a loop if a null string was matched other than at the start
2707    of the subject.
2708    
2709    3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2710    be distinguished at compile time, and for completeness also added PCRE_DATE.
2711    
2712    5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2713    in GnuWin32 environments.
2714    
2715    
2716    Version 2.07 29-Jul-99
2717    ----------------------
2718    
2719    1. The documentation is now supplied in plain text form and HTML as well as in
2720    the form of man page sources.
2721    
2722    2. C++ compilers don't like assigning (void *) values to other pointer types.
2723    In particular this affects malloc(). Although there is no problem in Standard
2724    C, I've put in casts to keep C++ compilers happy.
2725    
2726    3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2727    should be (const char *).
2728    
2729    4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2730    be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2731    However, I haven't made this a standard facility. The documentation doesn't
2732    mention it, and the Makefile doesn't support it.
2733    
2734    5. The Makefile now contains an "install" target, with editable destinations at
2735    the top of the file. The pcretest program is not installed.
2736    
2737    6. pgrep -V now gives the PCRE version number and date.
2738    
2739    7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2740    causing the entire string to be ignored, instead of just the last character.
2741    
2742    8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2743    non-matching string, it can take a very, very long time, even for strings of
2744    quite modest length, because of the nested recursion. PCRE now does better in
2745    some of these cases. It does this by remembering the last required literal
2746    character in the pattern, and pre-searching the subject to ensure it is present
2747    before running the real match. In other words, it applies a heuristic to detect
2748    some types of certain failure quickly, and in the above example, if presented
2749    with a string that has no trailing " it gives "no match" very quickly.
2750    
2751    9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2752    other alternatives are tried instead.
2753    
2754    
2755    Version 2.06 09-Jun-99
2756    ----------------------
2757    
2758    1. Change pcretest's output for amount of store used to show just the code
2759    space, because the remainder (the data block) varies in size between 32-bit and
2760    64-bit systems.
2761    
2762    2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2763    start matching at. This allows lookbehinds to work when searching for multiple
2764    occurrences in a string.
2765    
2766    3. Added additional options to pcretest for testing multiple occurrences:
2767    
2768       /+   outputs the rest of the string that follows a match
2769       /g   loops for multiple occurrences, using the new startoffset argument
2770       /G   loops for multiple occurrences by passing an incremented pointer
2771    
2772    4. PCRE wasn't doing the "first character" optimization for patterns starting
2773    with \b or \B, though it was doing it for other lookbehind assertions. That is,
2774    it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2775    the letter 'x'. On long subject strings, this gives a significant speed-up.
2776    
2777    
2778    Version 2.05 21-Apr-99
2779    ----------------------
2780    
2781    1. Changed the type of magic_number from int to long int so that it works
2782    properly on 16-bit systems.
2783    
2784    2. Fixed a bug which caused patterns starting with .* not to work correctly
2785    when the subject string contained newline characters. PCRE was assuming
2786    anchoring for such patterns in all cases, which is not correct because .* will
2787    not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2788    DOTALL is set at top level; otherwise it knows that patterns starting with .*
2789    must be retried after every newline in the subject.
2790    
2791    
2792    Version 2.04 18-Feb-99
2793    ----------------------
2794    
2795    1. For parenthesized subpatterns with repeats whose minimum was zero, the
2796    computation of the store needed to hold the pattern was incorrect (too large).
2797    If such patterns were nested a few deep, this could multiply and become a real
2798    problem.
2799    
2800    2. Added /M option to pcretest to show the memory requirement of a specific
2801    pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2802    
2803    3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2804    compiled in such a way that the backtracking after subsequent failure was
2805    pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2806    ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2807    
2808    
2809    Version 2.03 02-Feb-99
2810    ----------------------
2811    
2812    1. Fixed typo and small mistake in man page.
2813    
2814    2. Added 4th condition (GPL supersedes if conflict) and created separate
2815    LICENCE file containing the conditions.
2816    
2817    3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2818    Perl, that is the internal \ allows the delimiter to be included in the
2819    pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2820    the final delimiter, add \ to the end of the pattern (to test the error).
2821    
2822    4. Added the convenience functions for extracting substrings after a successful
2823    match. Updated pcretest to make it able to test these functions.
2824    
2825    
2826    Version 2.02 14-Jan-99
2827    ----------------------
2828    
2829    1. Initialized the working variables associated with each extraction so that
2830    their saving and restoring doesn't refer to uninitialized store.
2831    
2832    2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2833    compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2834    fix the problem.
2835    
2836    3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2837    calls, and wasn't printing the correct value for compiling calls. Increased the
2838    default value of LOOPREPEAT, and the number of significant figures in the
2839    times.
2840    
2841    4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2842    
2843    5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2844    a building problem on Windows NT with a FAT file system.
2845    
2846    
2847  Version 2.01 21-Oct-98  Version 2.01 21-Oct-98
2848  ----------------------  ----------------------

Legend:
Removed from v.25  
changed lines
  Added in v.322

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12