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

Diff of /code/trunk/ChangeLog

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

revision 65 by nigel, Sat Feb 24 21:40:08 2007 UTC revision 263 by ph10, Mon Nov 12 16:53:25 2007 UTC
# Line 1  Line 1 
1  ChangeLog for PCRE  ChangeLog for PCRE
2  ------------------  ------------------
3    
4    Version 7.5 12-Nov-07
5    ---------------------
6    
7    1.  Applied a patch from Craig: "This patch makes it possible to 'ignore'
8        values in parens when parsing an RE using the C++ wrapper."
9    
10    
11    Version 7.4 21-Sep-07
12    ---------------------
13    
14    1.  Change 7.3/28 was implemented for classes by looking at the bitmap. This
15        means that a class such as [\s] counted as "explicit reference to CR or
16        LF". That isn't really right - the whole point of the change was to try to
17        help when there was an actual mention of one of the two characters. So now
18        the change happens only if \r or \n (or a literal CR or LF) character is
19        encountered.
20    
21    2.  The 32-bit options word was also used for 6 internal flags, but the numbers
22        of both had grown to the point where there were only 3 bits left.
23        Fortunately, there was spare space in the data structure, and so I have
24        moved the internal flags into a new 16-bit field to free up more option
25        bits.
26    
27    3.  The appearance of (?J) at the start of a pattern set the DUPNAMES option,
28        but did not set the internal JCHANGED flag - either of these is enough to
29        control the way the "get" function works - but the PCRE_INFO_JCHANGED
30        facility is supposed to tell if (?J) was ever used, so now (?J) at the
31        start sets both bits.
32    
33    4.  Added options (at build time, compile time, exec time) to change \R from
34        matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
35    
36    5.  doc/pcresyntax.html was missing from the distribution.
37    
38    6.  Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
39        compatibility, even though it is no longer used.
40    
41    7.  Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
42        strtoull to pcrecpp.cc to select the available functions in WIN32 when the
43        windows.h file is present (where different names are used). [This was
44        reversed later after testing - see 16 below.]
45    
46    8.  Changed all #include <config.h> to #include "config.h". There were also
47        some further <pcre.h> cases that I changed to "pcre.h".
48    
49    9.  When pcregrep was used with the --colour option, it missed the line ending
50        sequence off the lines that it output.
51    
52    10. It was pointed out to me that arrays of string pointers cause lots of
53        relocations when a shared library is dynamically loaded. A technique of
54        using a single long string with a table of offsets can drastically reduce
55        these. I have refactored PCRE in four places to do this. The result is
56        dramatic:
57    
58          Originally:                          290
59          After changing UCP table:            187
60          After changing error message table:   43
61          After changing table of "verbs"       36
62          After changing table of Posix names   22
63    
64        Thanks to the folks working on Gregex for glib for this insight.
65    
66    11. --disable-stack-for-recursion caused compiling to fail unless -enable-
67        unicode-properties was also set.
68    
69    12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
70    
71    13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
72        checked only for CRLF.
73    
74    14. Added casts to pcretest.c to avoid compiler warnings.
75    
76    15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
77    
78    16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
79        and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
80        entirely. This removes changes made in 7 above.
81    
82    17. The CMake files have been updated, and there is now more information about
83        building with CMake in the NON-UNIX-USE document.
84    
85    
86    Version 7.3 28-Aug-07
87    ---------------------
88    
89     1. In the rejigging of the build system that eventually resulted in 7.1, the
90        line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
91        brackets there is not right, since it causes compilers to look for an
92        installed pcre.h, not the version that is in the source that is being
93        compiled (which of course may be different). I have changed it back to:
94    
95          #include "pcre.h"
96    
97        I have a vague recollection that the change was concerned with compiling in
98        different directories, but in the new build system, that is taken care of
99        by the VPATH setting the Makefile.
100    
101     2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
102        when the subject happened to end in the byte 0x85 (e.g. if the last
103        character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
104        characters but of course it shouldn't be taken as a newline when it is part
105        of another character. The bug was that, for an unlimited repeat of . in
106        not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
107        characters when looking for a newline.
108    
109     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
110    
111     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
112        in debug output.
113    
114     5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
115        long printing in the pcrecpp unittest when running under MinGW.
116    
117     6. ESC_K was left out of the EBCDIC table.
118    
119     7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
120        parentheses; I made it 1000, which seemed large enough. Unfortunately, the
121        limit also applies to "virtual nesting" when a pattern is recursive, and in
122        this case 1000 isn't so big. I have been able to remove this limit at the
123        expense of backing off one optimization in certain circumstances. Normally,
124        when pcre_exec() would call its internal match() function recursively and
125        immediately return the result unconditionally, it uses a "tail recursion"
126        feature to save stack. However, when a subpattern that can match an empty
127        string has an unlimited repetition quantifier, it no longer makes this
128        optimization. That gives it a stack frame in which to save the data for
129        checking that an empty string has been matched. Previously this was taken
130        from the 1000-entry workspace that had been reserved. So now there is no
131        explicit limit, but more stack is used.
132    
133     8. Applied Daniel's patches to solve problems with the import/export magic
134        syntax that is required for Windows, and which was going wrong for the
135        pcreposix and pcrecpp parts of the library. These were overlooked when this
136        problem was solved for the main library.
137    
138     9. There were some crude static tests to avoid integer overflow when computing
139        the size of patterns that contain repeated groups with explicit upper
140        limits. As the maximum quantifier is 65535, the maximum group length was
141        set at 30,000 so that the product of these two numbers did not overflow a
142        32-bit integer. However, it turns out that people want to use groups that
143        are longer than 30,000 bytes (though not repeat them that many times).
144        Change 7.0/17 (the refactoring of the way the pattern size is computed) has
145        made it possible to implement the integer overflow checks in a much more
146        dynamic way, which I have now done. The artificial limitation on group
147        length has been removed - we now have only the limit on the total length of
148        the compiled pattern, which depends on the LINK_SIZE setting.
149    
150    10. Fixed a bug in the documentation for get/copy named substring when
151        duplicate names are permitted. If none of the named substrings are set, the
152        functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
153        empty string.
154    
155    11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
156        instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
157        because the ] is interpreted as the first data character and the
158        terminating ] is not found. PCRE has been made compatible with Perl in this
159        regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
160        cause memory overwriting.
161    
162    10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
163        string has been matched (to stop an infinite loop). It was not recognizing
164        a conditional subpattern that could match an empty string if that
165        subpattern was within another subpattern. For example, it looped when
166        trying to match  (((?(1)X|))*)  but it was OK with  ((?(1)X|)*)  where the
167        condition was not nested. This bug has been fixed.
168    
169    12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
170        past the start of the subject in the presence of bytes with the top bit
171        set, for example "\x8aBCD".
172    
173    13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
174        (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
175    
176    14. Optimized (?!) to (*FAIL).
177    
178    15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
179        This restricts code points to be within the range 0 to 0x10FFFF, excluding
180        the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
181        full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
182        does: it's just the validity check that is more restrictive.
183    
184    16. Inserted checks for integer overflows during escape sequence (backslash)
185        processing, and also fixed erroneous offset values for syntax errors during
186        backslash processing.
187    
188    17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
189        for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
190    
191    18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
192        caused an overrun.
193    
194    19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
195        something other than just ASCII characters) inside a group that had an
196        unlimited repeat caused a loop at compile time (while checking to see
197        whether the group could match an empty string).
198    
199    20. Debugging a pattern containing \p or \P could cause a crash. For example,
200        [\P{Any}] did so. (Error in the code for printing property names.)
201    
202    21. An orphan \E inside a character class could cause a crash.
203    
204    22. A repeated capturing bracket such as (A)? could cause a wild memory
205        reference during compilation.
206    
207    23. There are several functions in pcre_compile() that scan along a compiled
208        expression for various reasons (e.g. to see if it's fixed length for look
209        behind). There were bugs in these functions when a repeated \p or \P was
210        present in the pattern. These operators have additional parameters compared
211        with \d, etc, and these were not being taken into account when moving along
212        the compiled data. Specifically:
213    
214        (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
215            length.
216    
217        (b) An item such as \pL+ within a repeated group could cause crashes or
218            loops.
219    
220        (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
221            "reference to non-existent subpattern" error.
222    
223        (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
224    
225    24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
226        characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
227    
228    25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
229    
230    26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
231        character were causing crashes (broken optimization).
232    
233    27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
234        \p or \P) caused a compile-time loop.
235    
236    28. More problems have arisen in unanchored patterns when CRLF is a valid line
237        break. For example, the unstudied pattern [\r\n]A does not match the string
238        "\r\nA" because change 7.0/46 below moves the current point on by two
239        characters after failing to match at the start. However, the pattern \nA
240        *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
241        the same is true. There doesn't seem any very clean way out of this, but
242        what I have chosen to do makes the common cases work: PCRE now takes note
243        of whether there can be an explicit match for \r or \n anywhere in the
244        pattern, and if so, 7.0/46 no longer applies. As part of this change,
245        there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
246        pattern has explicit CR or LF references.
247    
248    29. Added (*CR) etc for changing newline setting at start of pattern.
249    
250    
251    Version 7.2 19-Jun-07
252    ---------------------
253    
254     1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
255        which is apparently normally available under Windows.
256    
257     2. Re-jig the pcregrep tests with different newline settings in an attempt
258        to make them independent of the local environment's newline setting.
259    
260     3. Add code to configure.ac to remove -g from the CFLAGS default settings.
261    
262     4. Some of the "internals" tests were previously cut out when the link size
263        was not 2, because the output contained actual offsets. The recent new
264        "Z" feature of pcretest means that these can be cut out, making the tests
265        usable with all link sizes.
266    
267     5. Implemented Stan Switzer's goto replacement for longjmp() when not using
268        stack recursion. This gives a massive performance boost under BSD, but just
269        a small improvement under Linux. However, it saves one field in the frame
270        in all cases.
271    
272     6. Added more features from the forthcoming Perl 5.10:
273    
274        (a) (?-n) (where n is a string of digits) is a relative subroutine or
275            recursion call. It refers to the nth most recently opened parentheses.
276    
277        (b) (?+n) is also a relative subroutine call; it refers to the nth next
278            to be opened parentheses.
279    
280        (c) Conditions that refer to capturing parentheses can be specified
281            relatively, for example, (?(-2)... or (?(+3)...
282    
283        (d) \K resets the start of the current match so that everything before
284            is not part of it.
285    
286        (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
287    
288        (f) \g{name} is another synonym - part of Perl 5.10's unification of
289            reference syntax.
290    
291        (g) (?| introduces a group in which the numbering of parentheses in each
292            alternative starts with the same number.
293    
294        (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
295    
296     7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
297        PCRE_INFO_JCHANGED.
298    
299     8. A pattern such as  (.*(.)?)*  caused pcre_exec() to fail by either not
300        terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
301        for detecting groups that can match an empty string.
302    
303     9. A pattern with a very large number of alternatives (more than several
304        hundred) was running out of internal workspace during the pre-compile
305        phase, where pcre_compile() figures out how much memory will be needed. A
306        bit of new cunning has reduced the workspace needed for groups with
307        alternatives. The 1000-alternative test pattern now uses 12 bytes of
308        workspace instead of running out of the 4096 that are available.
309    
310    10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
311    
312    11. Applied patch from Google to remove an optimization that didn't quite work.
313        The report of the bug said:
314    
315          pcrecpp::RE("a*").FullMatch("aaa") matches, while
316          pcrecpp::RE("a*?").FullMatch("aaa") does not, and
317          pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
318    
319    12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
320        it matched the wrong number of bytes.
321    
322    
323    Version 7.1 24-Apr-07
324    ---------------------
325    
326     1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
327        that is more "standard", making use of automake and other Autotools. There
328        is some re-arrangement of the files and adjustment of comments consequent
329        on this.
330    
331     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
332        for recursive directory scanning broke on some systems because the files
333        are not scanned in any specific order and on different systems the order
334        was different. A call to "sort" has been inserted into RunGrepTest for the
335        approprate test as a short-term fix. In the longer term there may be an
336        alternative.
337    
338     3. I had an email from Eric Raymond about problems translating some of PCRE's
339        man pages to HTML (despite the fact that I distribute HTML pages, some
340        people do their own conversions for various reasons). The problems
341        concerned the use of low-level troff macros .br and .in. I have therefore
342        removed all such uses from the man pages (some were redundant, some could
343        be replaced by .nf/.fi pairs). The 132html script that I use to generate
344        HTML has been updated to handle .nf/.fi and to complain if it encounters
345        .br or .in.
346    
347     4. Updated comments in configure.ac that get placed in config.h.in and also
348        arranged for config.h to be included in the distribution, with the name
349        config.h.generic, for the benefit of those who have to compile without
350        Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
351    
352     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
353        Weber: (1) pcre_internal.h was missing some function renames; (2) updated
354        makevp.bat for the current PCRE, using the additional files
355        makevp_c.txt, makevp_l.txt, and pcregexp.pas.
356    
357     6. A Windows user reported a minor discrepancy with test 2, which turned out
358        to be caused by a trailing space on an input line that had got lost in his
359        copy. The trailing space was an accident, so I've just removed it.
360    
361     7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
362        that is needed.
363    
364     8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
365        as "const" (a) because they are and (b) because it helps the PHP
366        maintainers who have recently made a script to detect big data structures
367        in the php code that should be moved to the .rodata section. I remembered
368        to update Builducptable as well, so it won't revert if ucptable.h is ever
369        re-created.
370    
371     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
372        pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
373        order to be able to cut out the UTF-8 tables in the latter when UTF-8
374        support is not required. This saves 1.5-2K of code, which is important in
375        some applications.
376    
377        Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
378        so as not to refer to the tables, even though these functions will never be
379        called when UTF-8 support is disabled. Otherwise there are problems with a
380        shared library.
381    
382    10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
383    
384        (a) It was defining its arguments as char * instead of void *.
385    
386        (b) It was assuming that all moves were upwards in memory; this was true
387            a long time ago when I wrote it, but is no longer the case.
388    
389        The emulated memove() is provided for those environments that have neither
390        memmove() nor bcopy(). I didn't think anyone used it these days, but that
391        is clearly not the case, as these two bugs were recently reported.
392    
393    11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
394        and Detrail to create the HTML documentation, the .txt form of the man
395        pages, and it removes trailing spaces from listed files. It also creates
396        pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
397        case, it wraps all the #defines with #ifndefs. This script should be run
398        before "make dist".
399    
400    12. Fixed two fairly obscure bugs concerned with quantified caseless matching
401        with Unicode property support.
402    
403        (a) For a maximizing quantifier, if the two different cases of the
404            character were of different lengths in their UTF-8 codings (there are
405            some cases like this - I found 11), and the matching function had to
406            back up over a mixture of the two cases, it incorrectly assumed they
407            were both the same length.
408    
409        (b) When PCRE was configured to use the heap rather than the stack for
410            recursion during matching, it was not correctly preserving the data for
411            the other case of a UTF-8 character when checking ahead for a match
412            while processing a minimizing repeat. If the check also involved
413            matching a wide character, but failed, corruption could cause an
414            erroneous result when trying to check for a repeat of the original
415            character.
416    
417    13. Some tidying changes to the testing mechanism:
418    
419        (a) The RunTest script now detects the internal link size and whether there
420            is UTF-8 and UCP support by running ./pcretest -C instead of relying on
421            values substituted by "configure". (The RunGrepTest script already did
422            this for UTF-8.) The configure.ac script no longer substitutes the
423            relevant variables.
424    
425        (b) The debugging options /B and /D in pcretest show the compiled bytecode
426            with length and offset values. This means that the output is different
427            for different internal link sizes. Test 2 is skipped for link sizes
428            other than 2 because of this, bypassing the problem. Unfortunately,
429            there was also a test in test 3 (the locale tests) that used /B and
430            failed for link sizes other than 2. Rather than cut the whole test out,
431            I have added a new /Z option to pcretest that replaces the length and
432            offset values with spaces. This is now used to make test 3 independent
433            of link size. (Test 2 will be tidied up later.)
434    
435    14. If erroroffset was passed as NULL to pcre_compile, it provoked a
436        segmentation fault instead of returning the appropriate error message.
437    
438    15. In multiline mode when the newline sequence was set to "any", the pattern
439        ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
440        This doesn't seem right; it now treats the CRLF combination as the line
441        ending, and so does not match in that case. It's only a pattern such as ^$
442        that would hit this one: something like ^ABC$ would have failed after \r
443        and then tried again after \r\n.
444    
445    16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
446        in an attempt to make files that differ only in their line terminators
447        compare equal. This works on Linux.
448    
449    17. Under certain error circumstances pcregrep might try to free random memory
450        as it exited. This is now fixed, thanks to valgrind.
451    
452    19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
453        "abc\r\n\r\n", it found an unwanted second match after the second \r. This
454        was because its rules for how to advance for /g after matching an empty
455        string at the end of a line did not allow for this case. They now check for
456        it specially.
457    
458    20. pcretest is supposed to handle patterns and data of any length, by
459        extending its buffers when necessary. It was getting this wrong when the
460        buffer for a data line had to be extended.
461    
462    21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
463        CRLF as a newline sequence.
464    
465    22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
466        out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
467        I have nevertheless tidied it up.
468    
469    23. Added some casts to kill warnings from HP-UX ia64 compiler.
470    
471    24. Added a man page for pcre-config.
472    
473    
474    Version 7.0 19-Dec-06
475    ---------------------
476    
477     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
478        moving to gcc 4.1.1.
479    
480     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
481        sys/time.h, which is documented as needed for this function. It doesn't
482        seem to matter on Linux, but it showed up on some releases of OS X.
483    
484     3. It seems that there are systems where bytes whose values are greater than
485        127 match isprint() in the "C" locale. The "C" locale should be the
486        default when a C program starts up. In most systems, only ASCII printing
487        characters match isprint(). This difference caused the output from pcretest
488        to vary, making some of the tests fail. I have changed pcretest so that:
489    
490        (a) When it is outputting text in the compiled version of a pattern, bytes
491            other than 32-126 are always shown as hex escapes.
492    
493        (b) When it is outputting text that is a matched part of a subject string,
494            it does the same, unless a different locale has been set for the match
495            (using the /L modifier). In this case, it uses isprint() to decide.
496    
497     4. Fixed a major bug that caused incorrect computation of the amount of memory
498        required for a compiled pattern when options that changed within the
499        pattern affected the logic of the preliminary scan that determines the
500        length. The relevant options are -x, and -i in UTF-8 mode. The result was
501        that the computed length was too small. The symptoms of this bug were
502        either the PCRE error "internal error: code overflow" from pcre_compile(),
503        or a glibc crash with a message such as "pcretest: free(): invalid next
504        size (fast)". Examples of patterns that provoked this bug (shown in
505        pcretest format) are:
506    
507          /(?-x: )/x
508          /(?x)(?-x: \s*#\s*)/
509          /((?i)[\x{c0}])/8
510          /(?i:[\x{c0}])/8
511    
512        HOWEVER: Change 17 below makes this fix obsolete as the memory computation
513        is now done differently.
514    
515     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
516        wrapper classes; (b) implement a new function in the C++ scanner that is
517        more efficient than the old way of doing things because it avoids levels of
518        recursion in the regex matching; (c) add a paragraph to the documentation
519        for the FullMatch() function.
520    
521     6. The escape sequence \n was being treated as whatever was defined as
522        "newline". Not only was this contrary to the documentation, which states
523        that \n is character 10 (hex 0A), but it also went horribly wrong when
524        "newline" was defined as CRLF. This has been fixed.
525    
526     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
527        was being set to -1 for the "end of line" case (supposedly a value that no
528        character can have). Though this value is never used (the check for end of
529        line is "zero bytes in current character"), it caused compiler complaints.
530        I've changed it to 0xffffffff.
531    
532     8. In pcre_version.c, the version string was being built by a sequence of
533        C macros that, in the event of PCRE_PRERELEASE being defined as an empty
534        string (as it is for production releases) called a macro with an empty
535        argument. The C standard says the result of this is undefined. The gcc
536        compiler treats it as an empty string (which was what was wanted) but it is
537        reported that Visual C gives an error. The source has been hacked around to
538        avoid this problem.
539    
540     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
541        builds of pcretest, and changed the call to _setmode() to use _O_BINARY
542        instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
543        of them did).
544    
545    10. Originally, pcretest opened its input and output without "b"; then I was
546        told that "b" was needed in some environments, so it was added for release
547        5.0 to both the input and output. (It makes no difference on Unix-like
548        systems.) Later I was told that it is wrong for the input on Windows. I've
549        now abstracted the modes into two macros, to make it easier to fiddle with
550        them, and removed "b" from the input mode under Windows.
551    
552    11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
553    
554    12. Added -help and --help to pcretest as an official way of being reminded
555        of the options.
556    
557    13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
558        and pcrecpp.cc because they annoy compilers at high warning levels.
559    
560    14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
561    
562    15. Fixed an occurrence of == in configure.ac that should have been = (shell
563        scripts are not C programs :-) and which was not noticed because it works
564        on Linux.
565    
566    16. pcretest is supposed to handle any length of pattern and data line (as one
567        line or as a continued sequence of lines) by extending its input buffer if
568        necessary. This feature was broken for very long pattern lines, leading to
569        a string of junk being passed to pcre_compile() if the pattern was longer
570        than about 50K.
571    
572    17. I have done a major re-factoring of the way pcre_compile() computes the
573        amount of memory needed for a compiled pattern. Previously, there was code
574        that made a preliminary scan of the pattern in order to do this. That was
575        OK when PCRE was new, but as the facilities have expanded, it has become
576        harder and harder to keep it in step with the real compile phase, and there
577        have been a number of bugs (see for example, 4 above). I have now found a
578        cunning way of running the real compile function in a "fake" mode that
579        enables it to compute how much memory it would need, while actually only
580        ever using a few hundred bytes of working memory and without too many
581        tests of the mode. This should make future maintenance and development
582        easier. A side effect of this work is that the limit of 200 on the nesting
583        depth of parentheses has been removed (though this was never a serious
584        limitation, I suspect). However, there is a downside: pcre_compile() now
585        runs more slowly than before (30% or more, depending on the pattern). I
586        hope this isn't a big issue. There is no effect on runtime performance.
587    
588    18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
589        newline (only possible for the last line of a file) and it was a
590        pattern that set a locale (followed by /Lsomething), pcretest crashed.
591    
592    19. Added additional timing features to pcretest. (1) The -tm option now times
593        matching only, not compiling. (2) Both -t and -tm can be followed, as a
594        separate command line item, by a number that specifies the number of
595        repeats to use when timing. The default is 50000; this gives better
596        precision, but takes uncomfortably long for very large patterns.
597    
598    20. Extended pcre_study() to be more clever in cases where a branch of a
599        subpattern has no definite first character. For example, (a*|b*)[cd] would
600        previously give no result from pcre_study(). Now it recognizes that the
601        first character must be a, b, c, or d.
602    
603    21. There was an incorrect error "recursive call could loop indefinitely" if
604        a subpattern (or the entire pattern) that was being tested for matching an
605        empty string contained only one non-empty item after a nested subpattern.
606        For example, the pattern (?>\x{100}*)\d(?R) provoked this error
607        incorrectly, because the \d was being skipped in the check.
608    
609    22. The pcretest program now has a new pattern option /B and a command line
610        option -b, which is equivalent to adding /B to every pattern. This causes
611        it to show the compiled bytecode, without the additional information that
612        -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
613        is the same as /B/I).
614    
615    23. A new optimization is now able automatically to treat some sequences such
616        as a*b as a*+b. More specifically, if something simple (such as a character
617        or a simple class like \d) has an unlimited quantifier, and is followed by
618        something that cannot possibly match the quantified thing, the quantifier
619        is automatically "possessified".
620    
621    24. A recursive reference to a subpattern whose number was greater than 39
622        went wrong under certain circumstances in UTF-8 mode. This bug could also
623        have affected the operation of pcre_study().
624    
625    25. Realized that a little bit of performance could be had by replacing
626        (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
627    
628    26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
629    
630    27. Possessive quantifiers such as a++ were previously implemented by turning
631        them into atomic groups such as ($>a+). Now they have their own opcodes,
632        which improves performance. This includes the automatically created ones
633        from 23 above.
634    
635    28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
636        lookahead was broken if it was not anchored. PCRE was mistakenly expecting
637        the first matched character to be a colon. This applied both to named and
638        numbered groups.
639    
640    29. The ucpinternal.h header file was missing its idempotency #ifdef.
641    
642    30. I was sent a "project" file called libpcre.a.dev which I understand makes
643        building PCRE on Windows easier, so I have included it in the distribution.
644    
645    31. There is now a check in pcretest against a ridiculously large number being
646        returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
647        loop, the loop is abandoned.
648    
649    32. Forward references to subpatterns in conditions such as (?(2)...) where
650        subpattern 2 is defined later cause pcre_compile() to search forwards in
651        the pattern for the relevant set of parentheses. This search went wrong
652        when there were unescaped parentheses in a character class, parentheses
653        escaped with \Q...\E, or parentheses in a #-comment in /x mode.
654    
655    33. "Subroutine" calls and backreferences were previously restricted to
656        referencing subpatterns earlier in the regex. This restriction has now
657        been removed.
658    
659    34. Added a number of extra features that are going to be in Perl 5.10. On the
660        whole, these are just syntactic alternatives for features that PCRE had
661        previously implemented using the Python syntax or my own invention. The
662        other formats are all retained for compatibility.
663    
664        (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
665            as (?P<name>...). The new forms, as well as being in Perl 5.10, are
666            also .NET compatible.
667    
668        (b) A recursion or subroutine call to a named group can now be defined as
669            (?&name) as well as (?P>name).
670    
671        (c) A backreference to a named group can now be defined as \k<name> or
672            \k'name' as well as (?P=name). The new forms, as well as being in Perl
673            5.10, are also .NET compatible.
674    
675        (d) A conditional reference to a named group can now use the syntax
676            (?(<name>) or (?('name') as well as (?(name).
677    
678        (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
679            groups (named and numbered) that are never evaluated inline, but can be
680            called as "subroutines" from elsewhere. In effect, the DEFINE condition
681            is always false. There may be only one alternative in such a group.
682    
683        (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
684            as the simple (?(R). The condition is true only if the most recent
685            recursion is that of the given number or name. It does not search out
686            through the entire recursion stack.
687    
688        (g) The escape \gN or \g{N} has been added, where N is a positive or
689            negative number, specifying an absolute or relative reference.
690    
691    35. Tidied to get rid of some further signed/unsigned compiler warnings and
692        some "unreachable code" warnings.
693    
694    36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
695        things, this adds five new scripts.
696    
697    37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
698        There were also incompatibilities regarding the handling of \Q..\E inside
699        character classes, for example with patterns like [\Qa\E-\Qz\E] where the
700        hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
701    
702    38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
703        matches an empty string, and forcibly breaks the loop. There were bugs in
704        this code in non-simple cases. For a pattern such as  ^(a()*)*  matched
705        against  aaaa  the result was just "a" rather than "aaaa", for example. Two
706        separate and independent bugs (that affected different cases) have been
707        fixed.
708    
709    39. Refactored the code to abolish the use of different opcodes for small
710        capturing bracket numbers. This is a tidy that I avoided doing when I
711        removed the limit on the number of capturing brackets for 3.5 back in 2001.
712        The new approach is not only tidier, it makes it possible to reduce the
713        memory needed to fix the previous bug (38).
714    
715    40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
716        sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
717        processing dot, circumflex, or dollar metacharacters, or #-comments in /x
718        mode.
719    
720    41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
721        report.
722    
723    42. Applied patch, originally from Ari Pollak, modified by Google, to allow
724        copy construction and assignment in the C++ wrapper.
725    
726    43. Updated pcregrep to support "--newline=any". In the process, I fixed a
727        couple of bugs that could have given wrong results in the "--newline=crlf"
728        case.
729    
730    44. Added a number of casts and did some reorganization of signed/unsigned int
731        variables following suggestions from Dair Grant. Also renamed the variable
732        "this" as "item" because it is a C++ keyword.
733    
734    45. Arranged for dftables to add
735    
736          #include "pcre_internal.h"
737    
738        to pcre_chartables.c because without it, gcc 4.x may remove the array
739        definition from the final binary if PCRE is built into a static library and
740        dead code stripping is activated.
741    
742    46. For an unanchored pattern, if a match attempt fails at the start of a
743        newline sequence, and the newline setting is CRLF or ANY, and the next two
744        characters are CRLF, advance by two characters instead of one.
745    
746    
747    Version 6.7 04-Jul-06
748    ---------------------
749    
750     1. In order to handle tests when input lines are enormously long, pcretest has
751        been re-factored so that it automatically extends its buffers when
752        necessary. The code is crude, but this _is_ just a test program. The
753        default size has been increased from 32K to 50K.
754    
755     2. The code in pcre_study() was using the value of the re argument before
756        testing it for NULL. (Of course, in any sensible call of the function, it
757        won't be NULL.)
758    
759     3. The memmove() emulation function in pcre_internal.h, which is used on
760        systems that lack both memmove() and bcopy() - that is, hardly ever -
761        was missing a "static" storage class specifier.
762    
763     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
764        containing an extended class (one that cannot be represented by a bitmap
765        because it contains high-valued characters or Unicode property items, e.g.
766        [\pZ]). Almost always one would set UTF-8 mode when processing such a
767        pattern, but PCRE should not loop if you do not (it no longer does).
768        [Detail: two cases were found: (a) a repeated subpattern containing an
769        extended class; (b) a recursive reference to a subpattern that followed a
770        previous extended class. It wasn't skipping over the extended class
771        correctly when UTF-8 mode was not set.]
772    
773     5. A negated single-character class was not being recognized as fixed-length
774        in lookbehind assertions such as (?<=[^f]), leading to an incorrect
775        compile error "lookbehind assertion is not fixed length".
776    
777     6. The RunPerlTest auxiliary script was showing an unexpected difference
778        between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
779        write a Perl script that can interpret lines of an input file either as
780        byte characters or as UTF-8, which is what "perltest" was being required to
781        do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
782        can't do is switch easily at run time between having the "use utf8;" pragma
783        or not. In the end, I fudged it by using the RunPerlTest script to insert
784        "use utf8;" explicitly for the UTF-8 tests.
785    
786     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
787        the end of the subject string, contrary to the documentation and to what
788        Perl does. This was true of both matching functions. Now it matches only at
789        the start of the subject and immediately after *internal* newlines.
790    
791     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
792        a pointer to an int instead of a pointer to an unsigned long int. This
793        caused problems on 64-bit systems.
794    
795     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
796        instance of the 'standard' template library not being so standard".
797    
798    10. There was no check on the number of named subpatterns nor the maximum
799        length of a subpattern name. The product of these values is used to compute
800        the size of the memory block for a compiled pattern. By supplying a very
801        long subpattern name and a large number of named subpatterns, the size
802        computation could be caused to overflow. This is now prevented by limiting
803        the length of names to 32 characters, and the number of named subpatterns
804        to 10,000.
805    
806    11. Subpatterns that are repeated with specific counts have to be replicated in
807        the compiled pattern. The size of memory for this was computed from the
808        length of the subpattern and the repeat count. The latter is limited to
809        65535, but there was no limit on the former, meaning that integer overflow
810        could in principle occur. The compiled length of a repeated subpattern is
811        now limited to 30,000 bytes in order to prevent this.
812    
813    12. Added the optional facility to have named substrings with the same name.
814    
815    13. Added the ability to use a named substring as a condition, using the
816        Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
817        are numbers (not recommended). Forward references are permitted.
818    
819    14. Added forward references in named backreferences (if you see what I mean).
820    
821    15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
822        pattern could run off the end of the subject. For example, the pattern
823        "(?s)(.{1,5})"8 did this with the subject "ab".
824    
825    16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
826        PCRE_CASELESS was set when matching characters that were quantified with ?
827        or *.
828    
829    17. A character class other than a single negated character that had a minimum
830        but no maximum quantifier - for example [ab]{6,} - was not handled
831        correctly by pce_dfa_exec(). It would match only one character.
832    
833    18. A valid (though odd) pattern that looked like a POSIX character
834        class but used an invalid character after [ (for example [[,abc,]]) caused
835        pcre_compile() to give the error "Failed: internal error: code overflow" or
836        in some cases to crash with a glibc free() error. This could even happen if
837        the pattern terminated after [[ but there just happened to be a sequence of
838        letters, a binary zero, and a closing ] in the memory that followed.
839    
840    19. Perl's treatment of octal escapes in the range \400 to \777 has changed
841        over the years. Originally (before any Unicode support), just the bottom 8
842        bits were taken. Thus, for example, \500 really meant \100. Nowadays the
843        output from "man perlunicode" includes this:
844    
845          The regular expression compiler produces polymorphic opcodes.  That
846          is, the pattern adapts to the data and automatically switches to
847          the Unicode character scheme when presented with Unicode data--or
848          instead uses a traditional byte scheme when presented with byte
849          data.
850    
851        Sadly, a wide octal escape does not cause a switch, and in a string with
852        no other multibyte characters, these octal escapes are treated as before.
853        Thus, in Perl, the pattern  /\500/ actually matches \100 but the pattern
854        /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
855        Unicode string.
856    
857        I have not perpetrated such confusion in PCRE. Up till now, it took just
858        the bottom 8 bits, as in old Perl. I have now made octal escapes with
859        values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
860        translate to the appropriate multibyte character.
861    
862    29. Applied some refactoring to reduce the number of warnings from Microsoft
863        and Borland compilers. This has included removing the fudge introduced
864        seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
865        a warning about an unused variable.
866    
867    21. PCRE has not included VT (character 0x0b) in the set of whitespace
868        characters since release 4.0, because Perl (from release 5.004) does not.
869        [Or at least, is documented not to: some releases seem to be in conflict
870        with the documentation.] However, when a pattern was studied with
871        pcre_study() and all its branches started with \s, PCRE still included VT
872        as a possible starting character. Of course, this did no harm; it just
873        caused an unnecessary match attempt.
874    
875    22. Removed a now-redundant internal flag bit that recorded the fact that case
876        dependency changed within the pattern. This was once needed for "required
877        byte" processing, but is no longer used. This recovers a now-scarce options
878        bit. Also moved the least significant internal flag bit to the most-
879        significant bit of the word, which was not previously used (hangover from
880        the days when it was an int rather than a uint) to free up another bit for
881        the future.
882    
883    23. Added support for CRLF line endings as well as CR and LF. As well as the
884        default being selectable at build time, it can now be changed at runtime
885        via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
886        specify that it is scanning data with non-default line endings.
887    
888    24. Changed the definition of CXXLINK to make it agree with the definition of
889        LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
890    
891    25. Applied Ian Taylor's patches to avoid using another stack frame for tail
892        recursions. This makes a big different to stack usage for some patterns.
893    
894    26. If a subpattern containing a named recursion or subroutine reference such
895        as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
896        the space required for the compiled pattern went wrong and gave too small a
897        value. Depending on the environment, this could lead to "Failed: internal
898        error: code overflow at offset 49" or "glibc detected double free or
899        corruption" errors.
900    
901    27. Applied patches from Google (a) to support the new newline modes and (b) to
902        advance over multibyte UTF-8 characters in GlobalReplace.
903    
904    28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
905        difference for some implementation of PCRE in some Windows version.
906    
907    29. Added some extra testing facilities to pcretest:
908    
909        \q<number>   in a data line sets the "match limit" value
910        \Q<number>   in a data line sets the "match recursion limt" value
911        -S <number>  sets the stack size, where <number> is in megabytes
912    
913        The -S option isn't available for Windows.
914    
915    
916    Version 6.6 06-Feb-06
917    ---------------------
918    
919     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
920        in pcreposix.h. I have copied the definition from pcre.h.
921    
922     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
923        because pcre.h is no longer a built file.
924    
925     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
926        not normally included in the compiled code.
927    
928    
929    Version 6.5 01-Feb-06
930    ---------------------
931    
932     1. When using the partial match feature with pcre_dfa_exec(), it was not
933        anchoring the second and subsequent partial matches at the new starting
934        point. This could lead to incorrect results. For example, with the pattern
935        /1234/, partially matching against "123" and then "a4" gave a match.
936    
937     2. Changes to pcregrep:
938    
939        (a) All non-match returns from pcre_exec() were being treated as failures
940            to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
941            error message is output. Some extra information is given for the
942            PCRE_ERROR_MATCHLIMIT and PCRE_ERROR_RECURSIONLIMIT errors, which are
943            probably the only errors that are likely to be caused by users (by
944            specifying a regex that has nested indefinite repeats, for instance).
945            If there are more than 20 of these errors, pcregrep is abandoned.
946    
947        (b) A binary zero was treated as data while matching, but terminated the
948            output line if it was written out. This has been fixed: binary zeroes
949            are now no different to any other data bytes.
950    
951        (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
952            used to set a locale for matching. The --locale=xxxx long option has
953            been added (no short equivalent) to specify a locale explicitly on the
954            pcregrep command, overriding the environment variables.
955    
956        (d) When -B was used with -n, some line numbers in the output were one less
957            than they should have been.
958    
959        (e) Added the -o (--only-matching) option.
960    
961        (f) If -A or -C was used with -c (count only), some lines of context were
962            accidentally printed for the final match.
963    
964        (g) Added the -H (--with-filename) option.
965    
966        (h) The combination of options -rh failed to suppress file names for files
967            that were found from directory arguments.
968    
969        (i) Added the -D (--devices) and -d (--directories) options.
970    
971        (j) Added the -F (--fixed-strings) option.
972    
973        (k) Allow "-" to be used as a file name for -f as well as for a data file.
974    
975        (l) Added the --colo(u)r option.
976    
977        (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
978            is not present by default.
979    
980     3. A nasty bug was discovered in the handling of recursive patterns, that is,
981        items such as (?R) or (?1), when the recursion could match a number of
982        alternatives. If it matched one of the alternatives, but subsequently,
983        outside the recursion, there was a failure, the code tried to back up into
984        the recursion. However, because of the way PCRE is implemented, this is not
985        possible, and the result was an incorrect result from the match.
986    
987        In order to prevent this happening, the specification of recursion has
988        been changed so that all such subpatterns are automatically treated as
989        atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
990    
991     4. I had overlooked the fact that, in some locales, there are characters for
992        which isalpha() is true but neither isupper() nor islower() are true. In
993        the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
994        and ordfeminine) are like this. This affected the treatment of \w and \W
995        when they appeared in character classes, but not when they appeared outside
996        a character class. The bit map for "word" characters is now created
997        separately from the results of isalnum() instead of just taking it from the
998        upper, lower, and digit maps. (Plus the underscore character, of course.)
999    
1000     5. The above bug also affected the handling of POSIX character classes such as
1001        [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1002        permanent tables. Instead, the bit maps for such a class were previously
1003        created as the appropriate unions of the upper, lower, and digit bitmaps.
1004        Now they are created by subtraction from the [[:word:]] class, which has
1005        its own bitmap.
1006    
1007     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1008        It is created by subtracting the vertical space characters (\x09, \x0a,
1009        \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1010        subtraction was done in the overall bitmap for a character class, meaning
1011        that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1012        be recognized. This bug has been fixed.
1013    
1014     7. Patches from the folks at Google:
1015    
1016          (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1017          real life, but is still worth protecting against".
1018    
1019          (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1020          regular expressions".
1021    
1022          (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1023          have it.
1024    
1025          (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1026          "configure" and the latter not, in order to fix a problem somebody had
1027          with compiling the Arg class on HP-UX.
1028    
1029          (e) Improve the error-handling of the C++ wrapper a little bit.
1030    
1031          (f) New tests for checking recursion limiting.
1032    
1033     8. The pcre_memmove() function, which is used only if the environment does not
1034        have a standard memmove() function (and is therefore rarely compiled),
1035        contained two bugs: (a) use of int instead of size_t, and (b) it was not
1036        returning a result (though PCRE never actually uses the result).
1037    
1038     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1039        large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1040        returned instead of calling malloc() with an overflowing number that would
1041        most likely cause subsequent chaos.
1042    
1043    10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1044    
1045    11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1046        with this option is matched, the nmatch and pmatch options of regexec() are
1047        ignored.
1048    
1049    12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1050        provided in case anyone wants to the the POSIX interface with UTF-8
1051        strings.
1052    
1053    13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1054        C++ linking (needed for some HP-UX environments).
1055    
1056    14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1057        (unused parameter) and in the pcre_printint() function (omitted "default"
1058        switch label when the default is to do nothing).
1059    
1060    15. Added some code to make it possible, when PCRE is compiled as a C++
1061        library, to replace subject pointers for pcre_exec() with a smart pointer
1062        class, thus making it possible to process discontinuous strings.
1063    
1064    16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1065        much the same function. They were added by different people who were trying
1066        to make PCRE easy to compile on non-Unix systems. It has been suggested
1067        that PCRE_EXPORT be abolished now that there is more automatic apparatus
1068        for compiling on Windows systems. I have therefore replaced it with
1069        PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1070        defaults to "extern" for C or "extern C" for C++, which works fine on
1071        Unix-like systems. It is now possible to override the value of PCRE_DATA_
1072        SCOPE with something explicit in config.h. In addition:
1073    
1074        (a) pcreposix.h still had just "extern" instead of either of these macros;
1075            I have replaced it with PCRE_DATA_SCOPE.
1076    
1077        (b) Functions such as _pcre_xclass(), which are internal to the library,
1078            but external in the C sense, all had PCRE_EXPORT in their definitions.
1079            This is apparently wrong for the Windows case, so I have removed it.
1080            (It makes no difference on Unix-like systems.)
1081    
1082    17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1083        of recursive calls to match(). This is different to MATCH_LIMIT because
1084        that limits the total number of calls to match(), not all of which increase
1085        the depth of recursion. Limiting the recursion depth limits the amount of
1086        stack (or heap if NO_RECURSE is set) that is used. The default can be set
1087        when PCRE is compiled, and changed at run time. A patch from Google adds
1088        this functionality to the C++ interface.
1089    
1090    18. Changes to the handling of Unicode character properties:
1091    
1092        (a) Updated the table to Unicode 4.1.0.
1093    
1094        (b) Recognize characters that are not in the table as "Cn" (undefined).
1095    
1096        (c) I revised the way the table is implemented to a much improved format
1097            which includes recognition of ranges. It now supports the ranges that
1098            are defined in UnicodeData.txt, and it also amalgamates other
1099            characters into ranges. This has reduced the number of entries in the
1100            table from around 16,000 to around 3,000, thus reducing its size
1101            considerably. I realized I did not need to use a tree structure after
1102            all - a binary chop search is just as efficient. Having reduced the
1103            number of entries, I extended their size from 6 bytes to 8 bytes to
1104            allow for more data.
1105    
1106        (d) Added support for Unicode script names via properties such as \p{Han}.
1107    
1108    19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1109        matching that character.
1110    
1111    20. When matching a repeated Unicode property with a minimum greater than zero,
1112        (for example \pL{2,}), PCRE could look past the end of the subject if it
1113        reached it while seeking the minimum number of characters. This could
1114        happen only if some of the characters were more than one byte long, because
1115        there is a check for at least the minimum number of bytes.
1116    
1117    21. Refactored the implementation of \p and \P so as to be more general, to
1118        allow for more different types of property in future. This has changed the
1119        compiled form incompatibly. Anybody with saved compiled patterns that use
1120        \p or \P will have to recompile them.
1121    
1122    22. Added "Any" and "L&" to the supported property types.
1123    
1124    23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1125        but give a compile time error if the value is greater than 0xff.
1126    
1127    24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1128        accidentally not being installed or uninstalled.
1129    
1130    25. The pcre.h file was built from pcre.h.in, but the only changes that were
1131        made were to insert the current release number. This seemed silly, because
1132        it made things harder for people building PCRE on systems that don't run
1133        "configure". I have turned pcre.h into a distributed file, no longer built
1134        by "configure", with the version identification directly included. There is
1135        no longer a pcre.h.in file.
1136    
1137        However, this change necessitated a change to the pcre-config script as
1138        well. It is built from pcre-config.in, and one of the substitutions was the
1139        release number. I have updated configure.ac so that ./configure now finds
1140        the release number by grepping pcre.h.
1141    
1142    26. Added the ability to run the tests under valgrind.
1143    
1144    
1145    Version 6.4 05-Sep-05
1146    ---------------------
1147    
1148     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1149        "--" to be printed when multiple files were scanned, even when none of the
1150        -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1151        consider it to be a bug, and have restored the previous behaviour.
1152    
1153     2. A couple of code tidies to get rid of compiler warnings.
1154    
1155     3. The pcretest program used to cheat by referring to symbols in the library
1156        whose names begin with _pcre_. These are internal symbols that are not
1157        really supposed to be visible externally, and in some environments it is
1158        possible to suppress them. The cheating is now confined to including
1159        certain files from the library's source, which is a bit cleaner.
1160    
1161     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1162        file's purpose clearer.
1163    
1164     5. Reorganized pcre_ucp_findchar().
1165    
1166    
1167    Version 6.3 15-Aug-05
1168    ---------------------
1169    
1170     1. The file libpcre.pc.in did not have general read permission in the tarball.
1171    
1172     2. There were some problems when building without C++ support:
1173    
1174        (a) If C++ support was not built, "make install" and "make test" still
1175            tried to test it.
1176    
1177        (b) There were problems when the value of CXX was explicitly set. Some
1178            changes have been made to try to fix these, and ...
1179    
1180        (c) --disable-cpp can now be used to explicitly disable C++ support.
1181    
1182        (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1183            backslash in a target when C++ was disabled. This confuses some
1184            versions of "make", apparently. Using an intermediate variable solves
1185            this. (Same for CPP_LOBJ.)
1186    
1187     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1188        (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1189        necessary on certain architectures.
1190    
1191     4. Added a setting of -export-symbols-regex to the link command to remove
1192        those symbols that are exported in the C sense, but actually are local
1193        within the library, and not documented. Their names all begin with
1194        "_pcre_". This is not a perfect job, because (a) we have to except some
1195        symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1196        available (and never for static libraries). I have made a note to try to
1197        find a way round (a) in the future.
1198    
1199    
1200    Version 6.2 01-Aug-05
1201    ---------------------
1202    
1203     1. There was no test for integer overflow of quantifier values. A construction
1204        such as {1111111111111111} would give undefined results. What is worse, if
1205        a minimum quantifier for a parenthesized subpattern overflowed and became
1206        negative, the calculation of the memory size went wrong. This could have
1207        led to memory overwriting.
1208    
1209     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1210    
1211     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1212        operating environments where this matters.
1213    
1214     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1215        PCRE options from within the C++ wrapper.
1216    
1217     5. Named capturing subpatterns were not being correctly counted when a pattern
1218        was compiled. This caused two problems: (a) If there were more than 100
1219        such subpatterns, the calculation of the memory needed for the whole
1220        compiled pattern went wrong, leading to an overflow error. (b) Numerical
1221        back references of the form \12, where the number was greater than 9, were
1222        not recognized as back references, even though there were sufficient
1223        previous subpatterns.
1224    
1225     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1226        versions of gcc, e.g. 2.95.4.
1227    
1228    
1229    Version 6.1 21-Jun-05
1230    ---------------------
1231    
1232     1. There was one reference to the variable "posix" in pcretest.c that was not
1233        surrounded by "#if !defined NOPOSIX".
1234    
1235     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1236        the cross-check on the old pcre_info() function, for the benefit of the
1237        cut-down version of PCRE that is currently imported into Exim.
1238    
1239     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1240        allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1241        patterns that start (?i) but I don't think that matters. The use of (?i) is
1242        just an example; this all applies to the other options as well.
1243    
1244     4. Since libtool seems to echo the compile commands it is issuing, the output
1245        from "make" can be reduced a bit by putting "@" in front of each libtool
1246        compile command.
1247    
1248     5. Patch from the folks at Google for configure.in to be a bit more thorough
1249        in checking for a suitable C++ installation before trying to compile the
1250        C++ stuff. This should fix a reported problem when a compiler was present,
1251        but no suitable headers.
1252    
1253     6. The man pages all had just "PCRE" as their title. I have changed them to
1254        be the relevant file name. I have also arranged that these names are
1255        retained in the file doc/pcre.txt, which is a concatenation in text format
1256        of all the man pages except the little individual ones for each function.
1257    
1258     7. The NON-UNIX-USE file had not been updated for the different set of source
1259        files that come with release 6. I also added a few comments about the C++
1260        wrapper.
1261    
1262    
1263    Version 6.0 07-Jun-05
1264    ---------------------
1265    
1266     1. Some minor internal re-organization to help with my DFA experiments.
1267    
1268     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1269        didn't matter for the library itself when fully configured, but did matter
1270        when compiling without UCP support, or within Exim, where the ucp files are
1271        not imported.
1272    
1273     3. Refactoring of the library code to split up the various functions into
1274        different source modules. The addition of the new DFA matching code (see
1275        below) to a single monolithic source would have made it really too
1276        unwieldy, quite apart from causing all the code to be include in a
1277        statically linked application, when only some functions are used. This is
1278        relevant even without the DFA addition now that patterns can be compiled in
1279        one application and matched in another.
1280    
1281        The downside of splitting up is that there have to be some external
1282        functions and data tables that are used internally in different modules of
1283        the library but which are not part of the API. These have all had their
1284        names changed to start with "_pcre_" so that they are unlikely to clash
1285        with other external names.
1286    
1287     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1288        a different (DFA) algorithm. Although it is slower than the original
1289        function, it does have some advantages for certain types of matching
1290        problem.
1291    
1292     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1293        including restarting after a partial match.
1294    
1295     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1296        defined when compiling for Windows was sent to me. I have put it into the
1297        code, though I have no means of testing or verifying it.
1298    
1299     7. Added the pcre_refcount() auxiliary function.
1300    
1301     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1302        match before or at the first newline in the subject string. In pcretest,
1303        the /f option on a pattern can be used to set this.
1304    
1305     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1306        would behave wrongly. This has been present in PCRE since release 4.0.
1307    
1308    10. A number of changes to the pcregrep command:
1309    
1310        (a) Refactored how -x works; insert ^(...)$ instead of setting
1311            PCRE_ANCHORED and checking the length, in preparation for adding
1312            something similar for -w.
1313    
1314        (b) Added the -w (match as a word) option.
1315    
1316        (c) Refactored the way lines are read and buffered so as to have more
1317            than one at a time available.
1318    
1319        (d) Implemented a pcregrep test script.
1320    
1321        (e) Added the -M (multiline match) option. This allows patterns to match
1322            over several lines of the subject. The buffering ensures that at least
1323            8K, or the rest of the document (whichever is the shorter) is available
1324            for matching (and similarly the previous 8K for lookbehind assertions).
1325    
1326        (f) Changed the --help output so that it now says
1327    
1328              -w, --word-regex(p)
1329    
1330            instead of two lines, one with "regex" and the other with "regexp"
1331            because that confused at least one person since the short forms are the
1332            same. (This required a bit of code, as the output is generated
1333            automatically from a table. It wasn't just a text change.)
1334    
1335        (g) -- can be used to terminate pcregrep options if the next thing isn't an
1336            option but starts with a hyphen. Could be a pattern or a path name
1337            starting with a hyphen, for instance.
1338    
1339        (h) "-" can be given as a file name to represent stdin.
1340    
1341        (i) When file names are being printed, "(standard input)" is used for
1342            the standard input, for compatibility with GNU grep. Previously
1343            "<stdin>" was used.
1344    
1345        (j) The option --label=xxx can be used to supply a name to be used for
1346            stdin when file names are being printed. There is no short form.
1347    
1348        (k) Re-factored the options decoding logic because we are going to add
1349            two more options that take data. Such options can now be given in four
1350            different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1351    
1352        (l) Added the -A, -B, and -C options for requesting that lines of context
1353            around matches be printed.
1354    
1355        (m) Added the -L option to print the names of files that do not contain
1356            any matching lines, that is, the complement of -l.
1357    
1358        (n) The return code is 2 if any file cannot be opened, but pcregrep does
1359            continue to scan other files.
1360    
1361        (o) The -s option was incorrectly implemented. For compatibility with other
1362            greps, it now suppresses the error message for a non-existent or non-
1363            accessible file (but not the return code). There is a new option called
1364            -q that suppresses the output of matching lines, which was what -s was
1365            previously doing.
1366    
1367        (p) Added --include and --exclude options to specify files for inclusion
1368            and exclusion when recursing.
1369    
1370    11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1371        Hopefully, it now does.
1372    
1373    12. Missing cast in pcre_study().
1374    
1375    13. Added an "uninstall" target to the makefile.
1376    
1377    14. Replaced "extern" in the function prototypes in Makefile.in with
1378        "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1379        world, but is set differently for Windows.
1380    
1381    15. Added a second compiling function called pcre_compile2(). The only
1382        difference is that it has an extra argument, which is a pointer to an
1383        integer error code. When there is a compile-time failure, this is set
1384        non-zero, in addition to the error test pointer being set to point to an
1385        error message. The new argument may be NULL if no error number is required
1386        (but then you may as well call pcre_compile(), which is now just a
1387        wrapper). This facility is provided because some applications need a
1388        numeric error indication, but it has also enabled me to tidy up the way
1389        compile-time errors are handled in the POSIX wrapper.
1390    
1391    16. Added VPATH=.libs to the makefile; this should help when building with one
1392        prefix path and installing with another. (Or so I'm told by someone who
1393        knows more about this stuff than I do.)
1394    
1395    17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1396        passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1397        match everything, including newlines. This is not POSIX-compatible, but
1398        somebody wanted the feature. From pcretest it can be activated by using
1399        both the P and the s flags.
1400    
1401    18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1402    
1403    19. libpcre.pc was being incorrectly installed as executable.
1404    
1405    20. A couple of places in pcretest check for end-of-line by looking for '\n';
1406        it now also looks for '\r' so that it will work unmodified on Windows.
1407    
1408    21. Added Google's contributed C++ wrapper to the distribution.
1409    
1410    22. Added some untidy missing memory free() calls in pcretest, to keep
1411        Electric Fence happy when testing.
1412    
1413    
1414    
1415    Version 5.0 13-Sep-04
1416    ---------------------
1417    
1418     1. Internal change: literal characters are no longer packed up into items
1419        containing multiple characters in a single byte-string. Each character
1420        is now matched using a separate opcode. However, there may be more than one
1421        byte in the character in UTF-8 mode.
1422    
1423     2. The pcre_callout_block structure has two new fields: pattern_position and
1424        next_item_length. These contain the offset in the pattern to the next match
1425        item, and its length, respectively.
1426    
1427     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1428        insertion of callouts before each pattern item. Added the /C option to
1429        pcretest to make use of this.
1430    
1431     4. On the advice of a Windows user, the lines
1432    
1433          #if defined(_WIN32) || defined(WIN32)
1434          _setmode( _fileno( stdout ), 0x8000 );
1435          #endif  /* defined(_WIN32) || defined(WIN32) */
1436    
1437        have been added to the source of pcretest. This apparently does useful
1438        magic in relation to line terminators.
1439    
1440     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1441        for the benefit of those environments where the "b" makes a difference.
1442    
1443     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1444        to know about it. I have put a hack into configure.in that adds in code
1445        to set GCC=yes if CC=icc. This seems to end up at a point in the
1446        generated configure script that is early enough to affect the setting of
1447        compiler options, which is what is needed, but I have no means of testing
1448        whether it really works. (The user who reported this had patched the
1449        generated configure script, which of course I cannot do.)
1450    
1451        LATER: After change 22 below (new libtool files), the configure script
1452        seems to know about icc (and also ecc). Therefore, I have commented out
1453        this hack in configure.in.
1454    
1455     7. Added support for pkg-config (2 patches were sent in).
1456    
1457     8. Negated POSIX character classes that used a combination of internal tables
1458        were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1459        [[:^ascii]]. Typically, they would match almost any characters. The other
1460        POSIX classes were not broken in this way.
1461    
1462     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1463        to find the match, as PCRE was deluded into thinking that the match had to
1464        start at the start point or following a newline. The same bug applied to
1465        patterns with negative forward assertions or any backward assertions
1466        preceding ".*" at the start, unless the pattern required a fixed first
1467        character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1468    
1469    10. In UTF-8 mode, when moving forwards in the subject after a failed match
1470        starting at the last subject character, bytes beyond the end of the subject
1471        string were read.
1472    
1473    11. Renamed the variable "class" as "classbits" to make life easier for C++
1474        users. (Previously there was a macro definition, but it apparently wasn't
1475        enough.)
1476    
1477    12. Added the new field "tables" to the extra data so that tables can be passed
1478        in at exec time, or the internal tables can be re-selected. This allows
1479        a compiled regex to be saved and re-used at a later time by a different
1480        program that might have everything at different addresses.
1481    
1482    13. Modified the pcre-config script so that, when run on Solaris, it shows a
1483        -R library as well as a -L library.
1484    
1485    14. The debugging options of pcretest (-d on the command line or D on a
1486        pattern) showed incorrect output for anything following an extended class
1487        that contained multibyte characters and which was followed by a quantifier.
1488    
1489    15. Added optional support for general category Unicode character properties
1490        via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1491        support. It adds about 90K to the size of the library. The meanings of the
1492        inbuilt class escapes such as \d and \s have NOT been changed.
1493    
1494    16. Updated pcredemo.c to include calls to free() to release the memory for the
1495        compiled pattern.
1496    
1497    17. The generated file chartables.c was being created in the source directory
1498        instead of in the building directory. This caused the build to fail if the
1499        source directory was different from the building directory, and was
1500        read-only.
1501    
1502    18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1503        file. No doubt somebody will tell me if they don't make sense... Also added
1504        Dan Mooney's comments about building on OpenVMS.
1505    
1506    19. Added support for partial matching via the PCRE_PARTIAL option for
1507        pcre_exec() and the \P data escape in pcretest.
1508    
1509    20. Extended pcretest with 3 new pattern features:
1510    
1511        (i)   A pattern option of the form ">rest-of-line" causes pcretest to
1512              write the compiled pattern to the file whose name is "rest-of-line".
1513              This is a straight binary dump of the data, with the saved pointer to
1514              the character tables forced to be NULL. The study data, if any, is
1515              written too. After writing, pcretest reads a new pattern.
1516    
1517        (ii)  If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1518              compiled pattern from the given file. There must not be any
1519              occurrences of "<" in the file name (pretty unlikely); if there are,
1520              pcretest will instead treat the initial "<" as a pattern delimiter.
1521              After reading in the pattern, pcretest goes on to read data lines as
1522              usual.
1523    
1524        (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1525              and 16-bit fields in a compiled pattern, to simulate a pattern that
1526              was compiled on a host of opposite endianness.
1527    
1528    21. The pcre-exec() function can now cope with patterns that were compiled on
1529        hosts of opposite endianness, with this restriction:
1530    
1531          As for any compiled expression that is saved and used later, the tables
1532          pointer field cannot be preserved; the extra_data field in the arguments
1533          to pcre_exec() should be used to pass in a tables address if a value
1534          other than the default internal tables were used at compile time.
1535    
1536    22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1537        now diagnosed as an error. Previously, most of the time, a negative number
1538        would have been treated as zero, but if in addition "ovector" was passed as
1539        NULL, a crash could occur.
1540    
1541    23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1542        new versions from the libtool 1.5 distribution (the last one is a copy of
1543        a file called libtool.m4). This seems to have fixed the need to patch
1544        "configure" to support Darwin 1.3 (which I used to do). However, I still
1545        had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1546        workstation).
1547    
1548    24. Changed the PCRE licence to be the more standard "BSD" licence.
1549    
1550    
1551    Version 4.5 01-Dec-03
1552    ---------------------
1553    
1554     1. There has been some re-arrangement of the code for the match() function so
1555        that it can be compiled in a version that does not call itself recursively.
1556        Instead, it keeps those local variables that need separate instances for
1557        each "recursion" in a frame on the heap, and gets/frees frames whenever it
1558        needs to "recurse". Keeping track of where control must go is done by means
1559        of setjmp/longjmp. The whole thing is implemented by a set of macros that
1560        hide most of the details from the main code, and operates only if
1561        NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1562        "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1563        operating.
1564    
1565        To make it easier for callers to provide specially tailored get/free
1566        functions for this usage, two new functions, pcre_stack_malloc, and
1567        pcre_stack_free, are used. They are always called in strict stacking order,
1568        and the size of block requested is always the same.
1569    
1570        The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1571        PCRE has been compiled to use the stack or the heap for recursion. The
1572        -C option of pcretest uses this to show which version is compiled.
1573    
1574        A new data escape \S, is added to pcretest; it causes the amounts of store
1575        obtained and freed by both kinds of malloc/free at match time to be added
1576        to the output.
1577    
1578     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1579        what's available on my current Linux desktop machine.
1580    
1581     3. When matching a UTF-8 string, the test for a valid string at the start has
1582        been extended. If start_offset is not zero, PCRE now checks that it points
1583        to a byte that is the start of a UTF-8 character. If not, it returns
1584        PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1585        this is necessary because there may be backward assertions in the pattern.
1586        When matching the same subject several times, it may save resources to use
1587        PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1588    
1589     4. The code for checking the validity of UTF-8 strings has been tightened so
1590        that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1591        containing "overlong sequences".
1592    
1593     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1594        I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1595        should have been "&", but it just so happened that all the cases this let
1596        through by mistake were picked up later in the function.
1597    
1598     6. I had used a variable called "isblank" - this is a C99 function, causing
1599        some compilers to warn. To avoid this, I renamed it (as "blankclass").
1600    
1601     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1602        prompting; (b) run "./pcretest /dev/null" at the start of the test script
1603        so the version is shown; (c) stop "make test" echoing "./RunTest".
1604    
1605     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1606    
1607     9. The prototype for memmove() for systems that don't have it was using
1608        size_t, but the inclusion of the header that defines size_t was later. I've
1609        moved the #includes for the C headers earlier to avoid this.
1610    
1611    10. Added some adjustments to the code to make it easier to compiler on certain
1612        special systems:
1613    
1614          (a) Some "const" qualifiers were missing.
1615          (b) Added the macro EXPORT before all exported functions; by default this
1616              is defined to be empty.
1617          (c) Changed the dftables auxiliary program (that builds chartables.c) so
1618              that it reads its output file name as an argument instead of writing
1619              to the standard output and assuming this can be redirected.
1620    
1621    11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1622        class containing characters with values greater than 255, PCRE compilation
1623        went into a loop.
1624    
1625    12. A recursive reference to a subpattern that was within another subpattern
1626        that had a minimum quantifier of zero caused PCRE to crash. For example,
1627        (x(y(?2))z)? provoked this bug with a subject that got as far as the
1628        recursion. If the recursively-called subpattern itself had a zero repeat,
1629        that was OK.
1630    
1631    13. In pcretest, the buffer for reading a data line was set at 30K, but the
1632        buffer into which it was copied (for escape processing) was still set at
1633        1024, so long lines caused crashes.
1634    
1635    14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1636        "internal error: code overflow...". This applied to any character class
1637        that was followed by a possessive quantifier.
1638    
1639    15. Modified the Makefile to add libpcre.la as a prerequisite for
1640        libpcreposix.la because I was told this is needed for a parallel build to
1641        work.
1642    
1643    16. If a pattern that contained .* following optional items at the start was
1644        studied, the wrong optimizing data was generated, leading to matching
1645        errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1646        matching string must start with a or b or c. The correct conclusion for
1647        this pattern is that a match can start with any character.
1648    
1649    
1650    Version 4.4 13-Aug-03
1651    ---------------------
1652    
1653     1. In UTF-8 mode, a character class containing characters with values between
1654        127 and 255 was not handled correctly if the compiled pattern was studied.
1655        In fixing this, I have also improved the studying algorithm for such
1656        classes (slightly).
1657    
1658     2. Three internal functions had redundant arguments passed to them. Removal
1659        might give a very teeny performance improvement.
1660    
1661     3. Documentation bug: the value of the capture_top field in a callout is *one
1662        more than* the number of the hightest numbered captured substring.
1663    
1664     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
1665        in incorrectly linking with a previously installed version. They now link
1666        explicitly with libpcre.la.
1667    
1668     5. configure.in no longer needs to recognize Cygwin specially.
1669    
1670     6. A problem in pcre.in for Windows platforms is fixed.
1671    
1672     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
1673        pcretest, it used to include the size of the study block as part of its
1674        output. Unfortunately, the structure contains a field that has a different
1675        size on different hardware architectures. This meant that the tests that
1676        showed this size failed. As the block is currently always of a fixed size,
1677        this information isn't actually particularly useful in pcretest output, so
1678        I have just removed it.
1679    
1680     8. Three pre-processor statements accidentally did not start in column 1.
1681        Sadly, there are *still* compilers around that complain, even though
1682        standard C has not required this for well over a decade. Sigh.
1683    
1684     9. In pcretest, the code for checking callouts passed small integers in the
1685        callout_data field, which is a void * field. However, some picky compilers
1686        complained about the casts involved for this on 64-bit systems. Now
1687        pcretest passes the address of the small integer instead, which should get
1688        rid of the warnings.
1689    
1690    10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
1691        both compile and run time, and gives an error if an invalid UTF-8 sequence
1692        is found. There is a option for disabling this check in cases where the
1693        string is known to be correct and/or the maximum performance is wanted.
1694    
1695    11. In response to a bug report, I changed one line in Makefile.in from
1696    
1697            -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
1698        to
1699            -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
1700    
1701        to look similar to other lines, but I have no way of telling whether this
1702        is the right thing to do, as I do not use Windows. No doubt I'll get told
1703        if it's wrong...
1704    
1705    
1706    Version 4.3 21-May-03
1707    ---------------------
1708    
1709    1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
1710       Makefile.
1711    
1712    2. Some refactoring to improve the quality of the code:
1713    
1714       (i)   The utf8_table... variables are now declared "const".
1715    
1716       (ii)  The code for \cx, which used the "case flipping" table to upper case
1717             lower case letters, now just substracts 32. This is ASCII-specific,
1718             but the whole concept of \cx is ASCII-specific, so it seems
1719             reasonable.
1720    
1721       (iii) PCRE was using its character types table to recognize decimal and
1722             hexadecimal digits in the pattern. This is silly, because it handles
1723             only 0-9, a-f, and A-F, but the character types table is locale-
1724             specific, which means strange things might happen. A private
1725             table is now used for this - though it costs 256 bytes, a table is
1726             much faster than multiple explicit tests. Of course, the standard
1727             character types table is still used for matching digits in subject
1728             strings against \d.
1729    
1730       (iv)  Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
1731             ending in _t are). So I've renamed it as ESC_tee.
1732    
1733    3. The first argument for regexec() in the POSIX wrapper should have been
1734       defined as "const".
1735    
1736    4. Changed pcretest to use malloc() for its buffers so that they can be
1737       Electric Fenced for debugging.
1738    
1739    5. There were several places in the code where, in UTF-8 mode, PCRE would try
1740       to read one or more bytes before the start of the subject string. Often this
1741       had no effect on PCRE's behaviour, but in some circumstances it could
1742       provoke a segmentation fault.
1743    
1744    6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
1745       to try to read one or more bytes before the start of the subject string.
1746    
1747    7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
1748       UTF-8 support could misbehave in various ways if the subject string
1749       contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
1750       area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
1751       back over UTF-8 characters.)
1752    
1753    
1754    Version 4.2 14-Apr-03
1755    ---------------------
1756    
1757    1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
1758    
1759    2. Changes to the building process, supplied by Ronald Landheer-Cieslak
1760         [ON_WINDOWS]: new variable, "#" on non-Windows platforms
1761         [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
1762         [WIN_PREFIX]: new variable, "cyg" for Cygwin
1763         * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
1764           and BUILD_EXEEXT
1765         Note: automatic setting of the BUILD variables is not yet working
1766         set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
1767           compile-time but not at link-time
1768         [LINK]: use for linking executables only
1769         make different versions for Windows and non-Windows
1770         [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
1771           libraries
1772         [LINK_FOR_BUILD]: new variable
1773         [OBJEXT]: use throughout
1774         [EXEEXT]: use throughout
1775         <winshared>: new target
1776         <wininstall>: new target
1777         <dftables.o>: use native compiler
1778         <dftables>: use native linker
1779         <install>: handle Windows platform correctly
1780         <clean>: ditto
1781         <check>: ditto
1782         copy DLL to top builddir before testing
1783    
1784       As part of these changes, -no-undefined was removed again. This was reported
1785       to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
1786       in any case.
1787    
1788    3. Some tidies to get rid of compiler warnings:
1789    
1790       . In the match_data structure, match_limit was an unsigned long int, whereas
1791         match_call_count was an int. I've made them both unsigned long ints.
1792    
1793       . In pcretest the fact that a const uschar * doesn't automatically cast to
1794         a void * provoked a warning.
1795    
1796       . Turning on some more compiler warnings threw up some "shadow" variables
1797         and a few more missing casts.
1798    
1799    4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
1800       option, a class that contained a single character with a value between 128
1801       and 255 (e.g. /[\xFF]/) caused PCRE to crash.
1802    
1803    5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
1804       option, a class that contained several characters, but with at least one
1805       whose value was between 128 and 255 caused PCRE to crash.
1806    
1807    
1808  Version 4.1 12-Mar-03  Version 4.1 12-Mar-03
1809  ---------------------  ---------------------
1810    

Legend:
Removed from v.65  
changed lines
  Added in v.263

  ViewVC Help
Powered by ViewVC 1.1.5