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

Diff of /code/trunk/ChangeLog

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

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

Legend:
Removed from v.53  
changed lines
  Added in v.327

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12