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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 79 - (hide annotations) (download)
Sat Feb 24 21:40:52 2007 UTC (7 years, 5 months ago) by nigel
File size: 80365 byte(s)
Load pcre-6.1 into code/trunk.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 nigel 79 Version 6.1 21-Jun-05
5     ---------------------
6    
7     1. There was one reference to the variable "posix" in pcretest.c that was not
8     surrounded by "#if !defined NOPOSIX".
9    
10     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
11     the cross-check on the old pcre_info() function, for the benefit of the
12     cut-down version of PCRE that is currently imported into Exim.
13    
14     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
15     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
16     patterns that start (?i) but I don't think that matters. The use of (?i) is
17     just an example; this all applies to the other options as well.
18    
19     4. Since libtool seems to echo the compile commands it is issuing, the output
20     from "make" can be reduced a bit by putting "@" in front of each libtool
21     compile command.
22    
23     5. Patch from the folks at Google for configure.in to be a bit more thorough
24     in checking for a suitable C++ installation before trying to compile the
25     C++ stuff. This should fix a reported problem when a compiler was present,
26     but no suitable headers.
27    
28     6. The man pages all had just "PCRE" as their title. I have changed them to
29     be the relevant file name. I have also arranged that these names are
30     retained in the file doc/pcre.txt, which is a concatenation in text format
31     of all the man pages except the little individual ones for each function.
32    
33     7. The NON-UNIX-USE file had not been updated for the different set of source
34     files that come with release 6. I also added a few comments about the C++
35     wrapper.
36    
37    
38 nigel 77 Version 6.0 07-Jun-05
39     ---------------------
40    
41     1. Some minor internal re-organization to help with my DFA experiments.
42    
43     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
44     didn't matter for the library itself when fully configured, but did matter
45     when compiling without UCP support, or within Exim, where the ucp files are
46     not imported.
47    
48     3. Refactoring of the library code to split up the various functions into
49     different source modules. The addition of the new DFA matching code (see
50     below) to a single monolithic source would have made it really too
51     unwieldy, quite apart from causing all the code to be include in a
52     statically linked application, when only some functions are used. This is
53     relevant even without the DFA addition now that patterns can be compiled in
54     one application and matched in another.
55    
56     The downside of splitting up is that there have to be some external
57     functions and data tables that are used internally in different modules of
58     the library but which are not part of the API. These have all had their
59     names changed to start with "_pcre_" so that they are unlikely to clash
60     with other external names.
61    
62     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
63     a different (DFA) algorithm. Although it is slower than the original
64     function, it does have some advantages for certain types of matching
65     problem.
66    
67     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
68     including restarting after a partial match.
69    
70     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
71     defined when compiling for Windows was sent to me. I have put it into the
72     code, though I have no means of testing or verifying it.
73    
74     7. Added the pcre_refcount() auxiliary function.
75    
76     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
77     match before or at the first newline in the subject string. In pcretest,
78     the /f option on a pattern can be used to set this.
79    
80     9. A repeated \w when used in UTF-8 mode with characters greater than 256
81     would behave wrongly. This has been present in PCRE since release 4.0.
82    
83     10. A number of changes to the pcregrep command:
84    
85     (a) Refactored how -x works; insert ^(...)$ instead of setting
86     PCRE_ANCHORED and checking the length, in preparation for adding
87     something similar for -w.
88    
89     (b) Added the -w (match as a word) option.
90    
91     (c) Refactored the way lines are read and buffered so as to have more
92     than one at a time available.
93    
94     (d) Implemented a pcregrep test script.
95    
96     (e) Added the -M (multiline match) option. This allows patterns to match
97     over several lines of the subject. The buffering ensures that at least
98     8K, or the rest of the document (whichever is the shorter) is available
99     for matching (and similarly the previous 8K for lookbehind assertions).
100    
101     (f) Changed the --help output so that it now says
102    
103     -w, --word-regex(p)
104    
105     instead of two lines, one with "regex" and the other with "regexp"
106     because that confused at least one person since the short forms are the
107     same. (This required a bit of code, as the output is generated
108     automatically from a table. It wasn't just a text change.)
109    
110     (g) -- can be used to terminate pcregrep options if the next thing isn't an
111     option but starts with a hyphen. Could be a pattern or a path name
112     starting with a hyphen, for instance.
113    
114     (h) "-" can be given as a file name to represent stdin.
115    
116     (i) When file names are being printed, "(standard input)" is used for
117     the standard input, for compatibility with GNU grep. Previously
118     "<stdin>" was used.
119    
120     (j) The option --label=xxx can be used to supply a name to be used for
121     stdin when file names are being printed. There is no short form.
122    
123     (k) Re-factored the options decoding logic because we are going to add
124     two more options that take data. Such options can now be given in four
125     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
126    
127     (l) Added the -A, -B, and -C options for requesting that lines of context
128     around matches be printed.
129    
130     (m) Added the -L option to print the names of files that do not contain
131     any matching lines, that is, the complement of -l.
132    
133     (n) The return code is 2 if any file cannot be opened, but pcregrep does
134     continue to scan other files.
135    
136     (o) The -s option was incorrectly implemented. For compatibility with other
137     greps, it now suppresses the error message for a non-existent or non-
138     accessible file (but not the return code). There is a new option called
139     -q that suppresses the output of matching lines, which was what -s was
140     previously doing.
141    
142     (p) Added --include and --exclude options to specify files for inclusion
143     and exclusion when recursing.
144    
145     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
146     Hopefully, it now does.
147    
148     12. Missing cast in pcre_study().
149    
150     13. Added an "uninstall" target to the makefile.
151    
152     14. Replaced "extern" in the function prototypes in Makefile.in with
153     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
154     world, but is set differently for Windows.
155    
156     15. Added a second compiling function called pcre_compile2(). The only
157     difference is that it has an extra argument, which is a pointer to an
158     integer error code. When there is a compile-time failure, this is set
159     non-zero, in addition to the error test pointer being set to point to an
160     error message. The new argument may be NULL if no error number is required
161     (but then you may as well call pcre_compile(), which is now just a
162     wrapper). This facility is provided because some applications need a
163     numeric error indication, but it has also enabled me to tidy up the way
164     compile-time errors are handled in the POSIX wrapper.
165    
166     16. Added VPATH=.libs to the makefile; this should help when building with one
167     prefix path and installing with another. (Or so I'm told by someone who
168     knows more about this stuff than I do.)
169    
170     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
171     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
172     match everything, including newlines. This is not POSIX-compatible, but
173     somebody wanted the feature. From pcretest it can be activated by using
174     both the P and the s flags.
175    
176     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
177    
178     19. libpcre.pc was being incorrectly installed as executable.
179    
180     20. A couple of places in pcretest check for end-of-line by looking for '\n';
181     it now also looks for '\r' so that it will work unmodified on Windows.
182    
183     21. Added Google's contributed C++ wrapper to the distribution.
184    
185     22. Added some untidy missing memory free() calls in pcretest, to keep
186     Electric Fence happy when testing.
187    
188    
189    
190 nigel 75 Version 5.0 13-Sep-04
191     ---------------------
192    
193     1. Internal change: literal characters are no longer packed up into items
194     containing multiple characters in a single byte-string. Each character
195     is now matched using a separate opcode. However, there may be more than one
196     byte in the character in UTF-8 mode.
197    
198     2. The pcre_callout_block structure has two new fields: pattern_position and
199     next_item_length. These contain the offset in the pattern to the next match
200     item, and its length, respectively.
201    
202     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
203     insertion of callouts before each pattern item. Added the /C option to
204     pcretest to make use of this.
205    
206     4. On the advice of a Windows user, the lines
207    
208     #if defined(_WIN32) || defined(WIN32)
209     _setmode( _fileno( stdout ), 0x8000 );
210     #endif /* defined(_WIN32) || defined(WIN32) */
211    
212     have been added to the source of pcretest. This apparently does useful
213     magic in relation to line terminators.
214    
215     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
216     for the benefit of those environments where the "b" makes a difference.
217    
218     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
219     to know about it. I have put a hack into configure.in that adds in code
220     to set GCC=yes if CC=icc. This seems to end up at a point in the
221     generated configure script that is early enough to affect the setting of
222     compiler options, which is what is needed, but I have no means of testing
223     whether it really works. (The user who reported this had patched the
224     generated configure script, which of course I cannot do.)
225    
226     LATER: After change 22 below (new libtool files), the configure script
227     seems to know about icc (and also ecc). Therefore, I have commented out
228     this hack in configure.in.
229    
230     7. Added support for pkg-config (2 patches were sent in).
231    
232     8. Negated POSIX character classes that used a combination of internal tables
233     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
234     [[:^ascii]]. Typically, they would match almost any characters. The other
235     POSIX classes were not broken in this way.
236    
237     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
238     to find the match, as PCRE was deluded into thinking that the match had to
239     start at the start point or following a newline. The same bug applied to
240     patterns with negative forward assertions or any backward assertions
241     preceding ".*" at the start, unless the pattern required a fixed first
242     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
243    
244     10. In UTF-8 mode, when moving forwards in the subject after a failed match
245     starting at the last subject character, bytes beyond the end of the subject
246     string were read.
247    
248     11. Renamed the variable "class" as "classbits" to make life easier for C++
249     users. (Previously there was a macro definition, but it apparently wasn't
250     enough.)
251    
252     12. Added the new field "tables" to the extra data so that tables can be passed
253     in at exec time, or the internal tables can be re-selected. This allows
254     a compiled regex to be saved and re-used at a later time by a different
255     program that might have everything at different addresses.
256    
257     13. Modified the pcre-config script so that, when run on Solaris, it shows a
258     -R library as well as a -L library.
259    
260     14. The debugging options of pcretest (-d on the command line or D on a
261     pattern) showed incorrect output for anything following an extended class
262     that contained multibyte characters and which was followed by a quantifier.
263    
264     15. Added optional support for general category Unicode character properties
265     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
266     support. It adds about 90K to the size of the library. The meanings of the
267     inbuilt class escapes such as \d and \s have NOT been changed.
268    
269     16. Updated pcredemo.c to include calls to free() to release the memory for the
270     compiled pattern.
271    
272     17. The generated file chartables.c was being created in the source directory
273     instead of in the building directory. This caused the build to fail if the
274     source directory was different from the building directory, and was
275     read-only.
276    
277     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
278     file. No doubt somebody will tell me if they don't make sense... Also added
279     Dan Mooney's comments about building on OpenVMS.
280    
281     19. Added support for partial matching via the PCRE_PARTIAL option for
282     pcre_exec() and the \P data escape in pcretest.
283    
284     20. Extended pcretest with 3 new pattern features:
285    
286     (i) A pattern option of the form ">rest-of-line" causes pcretest to
287     write the compiled pattern to the file whose name is "rest-of-line".
288     This is a straight binary dump of the data, with the saved pointer to
289     the character tables forced to be NULL. The study data, if any, is
290     written too. After writing, pcretest reads a new pattern.
291    
292     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
293     compiled pattern from the given file. There must not be any
294     occurrences of "<" in the file name (pretty unlikely); if there are,
295     pcretest will instead treat the initial "<" as a pattern delimiter.
296     After reading in the pattern, pcretest goes on to read data lines as
297     usual.
298    
299     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
300     and 16-bit fields in a compiled pattern, to simulate a pattern that
301     was compiled on a host of opposite endianness.
302    
303     21. The pcre-exec() function can now cope with patterns that were compiled on
304     hosts of opposite endianness, with this restriction:
305    
306     As for any compiled expression that is saved and used later, the tables
307     pointer field cannot be preserved; the extra_data field in the arguments
308     to pcre_exec() should be used to pass in a tables address if a value
309     other than the default internal tables were used at compile time.
310    
311     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
312     now diagnosed as an error. Previously, most of the time, a negative number
313     would have been treated as zero, but if in addition "ovector" was passed as
314     NULL, a crash could occur.
315    
316     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
317     new versions from the libtool 1.5 distribution (the last one is a copy of
318     a file called libtool.m4). This seems to have fixed the need to patch
319     "configure" to support Darwin 1.3 (which I used to do). However, I still
320     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
321     workstation).
322    
323     24. Changed the PCRE licence to be the more standard "BSD" licence.
324    
325    
326 nigel 73 Version 4.5 01-Dec-03
327     ---------------------
328    
329     1. There has been some re-arrangement of the code for the match() function so
330     that it can be compiled in a version that does not call itself recursively.
331     Instead, it keeps those local variables that need separate instances for
332     each "recursion" in a frame on the heap, and gets/frees frames whenever it
333     needs to "recurse". Keeping track of where control must go is done by means
334     of setjmp/longjmp. The whole thing is implemented by a set of macros that
335     hide most of the details from the main code, and operates only if
336     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
337     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
338     operating.
339    
340     To make it easier for callers to provide specially tailored get/free
341     functions for this usage, two new functions, pcre_stack_malloc, and
342     pcre_stack_free, are used. They are always called in strict stacking order,
343     and the size of block requested is always the same.
344    
345     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
346     PCRE has been compiled to use the stack or the heap for recursion. The
347     -C option of pcretest uses this to show which version is compiled.
348    
349     A new data escape \S, is added to pcretest; it causes the amounts of store
350     obtained and freed by both kinds of malloc/free at match time to be added
351     to the output.
352    
353     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
354     what's available on my current Linux desktop machine.
355    
356     3. When matching a UTF-8 string, the test for a valid string at the start has
357     been extended. If start_offset is not zero, PCRE now checks that it points
358     to a byte that is the start of a UTF-8 character. If not, it returns
359     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
360     this is necessary because there may be backward assertions in the pattern.
361     When matching the same subject several times, it may save resources to use
362     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
363    
364     4. The code for checking the validity of UTF-8 strings has been tightened so
365     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
366     containing "overlong sequences".
367    
368     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
369     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
370     should have been "&", but it just so happened that all the cases this let
371     through by mistake were picked up later in the function.
372    
373     6. I had used a variable called "isblank" - this is a C99 function, causing
374     some compilers to warn. To avoid this, I renamed it (as "blankclass").
375    
376     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
377     prompting; (b) run "./pcretest /dev/null" at the start of the test script
378     so the version is shown; (c) stop "make test" echoing "./RunTest".
379    
380     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
381    
382     9. The prototype for memmove() for systems that don't have it was using
383     size_t, but the inclusion of the header that defines size_t was later. I've
384     moved the #includes for the C headers earlier to avoid this.
385    
386     10. Added some adjustments to the code to make it easier to compiler on certain
387     special systems:
388    
389     (a) Some "const" qualifiers were missing.
390     (b) Added the macro EXPORT before all exported functions; by default this
391     is defined to be empty.
392     (c) Changed the dftables auxiliary program (that builds chartables.c) so
393     that it reads its output file name as an argument instead of writing
394     to the standard output and assuming this can be redirected.
395    
396     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
397     class containing characters with values greater than 255, PCRE compilation
398     went into a loop.
399    
400     12. A recursive reference to a subpattern that was within another subpattern
401     that had a minimum quantifier of zero caused PCRE to crash. For example,
402     (x(y(?2))z)? provoked this bug with a subject that got as far as the
403     recursion. If the recursively-called subpattern itself had a zero repeat,
404     that was OK.
405    
406     13. In pcretest, the buffer for reading a data line was set at 30K, but the
407     buffer into which it was copied (for escape processing) was still set at
408     1024, so long lines caused crashes.
409    
410     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
411     "internal error: code overflow...". This applied to any character class
412     that was followed by a possessive quantifier.
413    
414     15. Modified the Makefile to add libpcre.la as a prerequisite for
415     libpcreposix.la because I was told this is needed for a parallel build to
416     work.
417    
418     16. If a pattern that contained .* following optional items at the start was
419     studied, the wrong optimizing data was generated, leading to matching
420     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
421     matching string must start with a or b or c. The correct conclusion for
422     this pattern is that a match can start with any character.
423    
424    
425 nigel 71 Version 4.4 13-Aug-03
426     ---------------------
427    
428     1. In UTF-8 mode, a character class containing characters with values between
429     127 and 255 was not handled correctly if the compiled pattern was studied.
430     In fixing this, I have also improved the studying algorithm for such
431     classes (slightly).
432    
433     2. Three internal functions had redundant arguments passed to them. Removal
434     might give a very teeny performance improvement.
435    
436     3. Documentation bug: the value of the capture_top field in a callout is *one
437     more than* the number of the hightest numbered captured substring.
438    
439     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
440     in incorrectly linking with a previously installed version. They now link
441     explicitly with libpcre.la.
442    
443     5. configure.in no longer needs to recognize Cygwin specially.
444    
445     6. A problem in pcre.in for Windows platforms is fixed.
446    
447     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
448     pcretest, it used to include the size of the study block as part of its
449     output. Unfortunately, the structure contains a field that has a different
450     size on different hardware architectures. This meant that the tests that
451     showed this size failed. As the block is currently always of a fixed size,
452     this information isn't actually particularly useful in pcretest output, so
453     I have just removed it.
454    
455     8. Three pre-processor statements accidentally did not start in column 1.
456     Sadly, there are *still* compilers around that complain, even though
457     standard C has not required this for well over a decade. Sigh.
458    
459     9. In pcretest, the code for checking callouts passed small integers in the
460     callout_data field, which is a void * field. However, some picky compilers
461     complained about the casts involved for this on 64-bit systems. Now
462     pcretest passes the address of the small integer instead, which should get
463     rid of the warnings.
464    
465     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
466     both compile and run time, and gives an error if an invalid UTF-8 sequence
467     is found. There is a option for disabling this check in cases where the
468     string is known to be correct and/or the maximum performance is wanted.
469    
470     11. In response to a bug report, I changed one line in Makefile.in from
471    
472     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
473     to
474     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
475    
476     to look similar to other lines, but I have no way of telling whether this
477     is the right thing to do, as I do not use Windows. No doubt I'll get told
478     if it's wrong...
479    
480    
481 nigel 69 Version 4.3 21-May-03
482     ---------------------
483    
484     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
485     Makefile.
486    
487     2. Some refactoring to improve the quality of the code:
488    
489     (i) The utf8_table... variables are now declared "const".
490    
491     (ii) The code for \cx, which used the "case flipping" table to upper case
492     lower case letters, now just substracts 32. This is ASCII-specific,
493     but the whole concept of \cx is ASCII-specific, so it seems
494     reasonable.
495    
496     (iii) PCRE was using its character types table to recognize decimal and
497     hexadecimal digits in the pattern. This is silly, because it handles
498     only 0-9, a-f, and A-F, but the character types table is locale-
499     specific, which means strange things might happen. A private
500     table is now used for this - though it costs 256 bytes, a table is
501     much faster than multiple explicit tests. Of course, the standard
502     character types table is still used for matching digits in subject
503     strings against \d.
504    
505     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
506     ending in _t are). So I've renamed it as ESC_tee.
507    
508     3. The first argument for regexec() in the POSIX wrapper should have been
509     defined as "const".
510    
511     4. Changed pcretest to use malloc() for its buffers so that they can be
512     Electric Fenced for debugging.
513    
514     5. There were several places in the code where, in UTF-8 mode, PCRE would try
515     to read one or more bytes before the start of the subject string. Often this
516     had no effect on PCRE's behaviour, but in some circumstances it could
517     provoke a segmentation fault.
518    
519     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
520     to try to read one or more bytes before the start of the subject string.
521    
522     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
523     UTF-8 support could misbehave in various ways if the subject string
524     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
525     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
526     back over UTF-8 characters.)
527    
528    
529 nigel 67 Version 4.2 14-Apr-03
530     ---------------------
531    
532     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
533    
534     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
535     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
536     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
537     [WIN_PREFIX]: new variable, "cyg" for Cygwin
538     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
539     and BUILD_EXEEXT
540     Note: automatic setting of the BUILD variables is not yet working
541     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
542     compile-time but not at link-time
543     [LINK]: use for linking executables only
544     make different versions for Windows and non-Windows
545     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
546     libraries
547     [LINK_FOR_BUILD]: new variable
548     [OBJEXT]: use throughout
549     [EXEEXT]: use throughout
550     <winshared>: new target
551     <wininstall>: new target
552     <dftables.o>: use native compiler
553     <dftables>: use native linker
554     <install>: handle Windows platform correctly
555     <clean>: ditto
556     <check>: ditto
557     copy DLL to top builddir before testing
558    
559     As part of these changes, -no-undefined was removed again. This was reported
560     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
561     in any case.
562    
563     3. Some tidies to get rid of compiler warnings:
564    
565     . In the match_data structure, match_limit was an unsigned long int, whereas
566     match_call_count was an int. I've made them both unsigned long ints.
567    
568     . In pcretest the fact that a const uschar * doesn't automatically cast to
569     a void * provoked a warning.
570    
571     . Turning on some more compiler warnings threw up some "shadow" variables
572     and a few more missing casts.
573    
574     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
575     option, a class that contained a single character with a value between 128
576     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
577    
578     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
579     option, a class that contained several characters, but with at least one
580     whose value was between 128 and 255 caused PCRE to crash.
581    
582    
583 nigel 65 Version 4.1 12-Mar-03
584     ---------------------
585 nigel 63
586 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
587     needed, and a string in dftables.c that was longer than standard compilers are
588     required to support.
589    
590     2. Compiling with Sun's compiler found a few more places where the code could
591     be tidied up in order to avoid warnings.
592    
593     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
594     first of these names is deprecated in the latest Autoconf in favour of the name
595     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
596     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
597     analogy I have changed it to CFLAGS_FOR_BUILD.
598    
599     4. Added -no-undefined to the linking command in the Makefile, because this is
600     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
601     linking step for the pcreposix library.
602    
603     5. PCRE was failing to diagnose the case of two named groups with the same
604     name.
605    
606     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
607     literal character that is needed in the subject for a match, and scans along to
608     ensure that it is present before embarking on the full matching process. This
609     saves time in cases of nested unlimited repeats that are never going to match.
610     Problem: the scan can take a lot of time if the subject is very long (e.g.
611     megabytes), thus penalizing straightforward matches. It is now done only if the
612     amount of subject to be scanned is less than 1000 bytes.
613    
614     7. A lesser problem with the same optimization is that it was recording the
615     first character of an anchored pattern as "needed", thus provoking a search
616     right along the subject, even when the first match of the pattern was going to
617     fail. The "needed" character is now not set for anchored patterns, unless it
618     follows something in the pattern that is of non-fixed length. Thus, it still
619     fulfils its original purpose of finding quick non-matches in cases of nested
620     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
621    
622    
623     Version 4.0 17-Feb-03
624     ---------------------
625    
626 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
627     extended to the end of string, PCRE compiled incorrect data. This could lead to
628     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
629    
630     2. Moved to autoconf 2.53 and libtool 1.4.2.
631    
632     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
633     the special perltest8 script is no longer needed - all the tests can be run
634     from a single perltest script.
635    
636     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
637     by \s. It has now been removed in PCRE. This means it isn't recognized as
638     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
639     class [:space:] *does* include VT, thereby creating a mess.
640    
641     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
642     space and tab.
643    
644     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
645     its new features into the main test script, reducing the number of scripts.
646    
647     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
648     were backward compatible, and made the (?i) apply to the whole pattern, as if
649     /i were given. Now it behaves more logically, and applies the option setting
650     only to what follows. PCRE has been changed to follow suit. However, if it
651     finds options settings right at the start of the pattern, it extracts them into
652     the global options, as before. Thus, they show up in the info data.
653    
654     8. Added support for the \Q...\E escape sequence. Characters in between are
655     treated as literals. This is slightly different from Perl in that $ and @ are
656     also handled as literals inside the quotes. In Perl, they will cause variable
657     interpolation. Note the following examples:
658    
659     Pattern PCRE matches Perl matches
660    
661     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
662     \Qabc\$xyz\E abc\$xyz abc\$xyz
663     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
664    
665     For compatibility with Perl, \Q...\E sequences are recognized inside character
666     classes as well as outside them.
667    
668     9. Re-organized 3 code statements in pcretest to avoid "overflow in
669     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
670     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
671     signed/unsigned warnings.
672    
673     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
674     option for pcretest, so I've replaced it by a simple function that does just
675     that job.
676    
677     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
678     "pcregrep -".
679    
680     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
681     Java package. This provides some syntactic sugar for simple cases of what my
682     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
683     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
684     item, you can use this simplified notation. Note that only makes sense with
685     greedy quantifiers. Consequently, the use of the possessive quantifier forces
686     greediness, whatever the setting of the PCRE_UNGREEDY option.
687    
688     13. A change of greediness default within a pattern was not taking effect at
689     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
690     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
691     was abstracted outside.
692    
693     14. PCRE now supports the \G assertion. It is true when the current matching
694     position is at the start point of the match. This differs from \A when the
695     starting offset is non-zero. Used with the /g option of pcretest (or similar
696     code), it works in the same way as it does for Perl's /g option. If all
697     alternatives of a regex begin with \G, the expression is anchored to the start
698     match position, and the "anchored" flag is set in the compiled expression.
699    
700     15. Some bugs concerning the handling of certain option changes within patterns
701     have been fixed. These applied to options other than (?ims). For example,
702     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
703     been the other way round. Some of this was related to change 7 above.
704    
705     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
706     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
707     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
708     POSIX classes only within a class (e.g. /[[:alpha:]]/).
709    
710     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
711     mode. Unlike ".", it always matches newline, whatever the setting of
712     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
713     assertions. Perl allows it, but it doesn't (in general) work because it can't
714     calculate the length of the lookbehind. At least, that's the case for Perl
715     5.8.0 - I've been told they are going to document that it doesn't work in
716     future.
717    
718     18. Added an error diagnosis for escapes that PCRE does not support: these are
719     \L, \l, \N, \P, \p, \U, \u, and \X.
720    
721     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
722     reading past the end of the pattern in cases such as /[abcd/.
723    
724     20. PCRE was getting more memory than necessary for patterns with classes that
725     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
726    
727     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
728     compiling PCRE for use with Virtual Pascal.
729    
730     22. Small fix to the Makefile to make it work properly if the build is done
731     outside the source tree.
732    
733     23. Added a new extension: a condition to go with recursion. If a conditional
734     subpattern starts with (?(R) the "true" branch is used if recursion has
735     happened, whereas the "false" branch is used only at the top level.
736    
737     24. When there was a very long string of literal characters (over 255 bytes
738     without UTF support, over 250 bytes with UTF support), the computation of how
739     much memory was required could be incorrect, leading to segfaults or other
740     strange effects.
741    
742     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
743     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
744     there was a subsequent back reference to those brackets. This meant that, for
745     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
746     possible to check for precisely this case. All we can do is abandon the
747     optimization if .* occurs inside capturing brackets when there are any back
748     references whatsoever. (See below for a better fix that came later.)
749    
750     26. The handling of the optimization for finding the first character of a
751     non-anchored pattern, and for finding a character that is required later in the
752     match were failing in some cases. This didn't break the matching; it just
753     failed to optimize when it could. The way this is done has been re-implemented.
754    
755     27. Fixed typo in error message for invalid (?R item (it said "(?p").
756    
757     28. Added a new feature that provides some of the functionality that Perl
758     provides with (?{...}). The facility is termed a "callout". The way it is done
759     in PCRE is for the caller to provide an optional function, by setting
760     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
761     global variable. By default it is unset, which disables all calling out. To get
762     the function called, the regex must include (?C) at appropriate points. This
763     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
764     This provides a means of identifying different callout points. When PCRE
765     reaches such a point in the regex, if pcre_callout has been set, the external
766     function is called. It is provided with data in a structure called
767     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
768     matching continues; if it returns a non-zero value, the match at the current
769     point fails. However, backtracking will occur if possible. [This was changed
770     later and other features added - see item 49 below.]
771    
772     29. pcretest is upgraded to test the callout functionality. It provides a
773     callout function that displays information. By default, it shows the start of
774     the match and the current position in the text. There are some new data escapes
775     to vary what happens:
776    
777     \C+ in addition, show current contents of captured substrings
778     \C- do not supply a callout function
779     \C!n return 1 when callout number n is reached
780     \C!n!m return 1 when callout number n is reached for the mth time
781    
782     30. If pcregrep was called with the -l option and just a single file name, it
783     output "<stdin>" if a match was found, instead of the file name.
784    
785     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
786     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
787     pcre_exec(). This saves a malloc/free per call. The default value of
788     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
789     when configuring.
790    
791     32. The default maximum size of a compiled pattern is 64K. There have been a
792     few cases of people hitting this limit. The code now uses macros to handle the
793     storing of links as offsets within the compiled pattern. It defaults to 2-byte
794     links, but this can be changed to 3 or 4 bytes by --with-link-size when
795     configuring. Tests 2 and 5 work only with 2-byte links because they output
796     debugging information about compiled patterns.
797    
798     33. Internal code re-arrangements:
799    
800     (a) Moved the debugging function for printing out a compiled regex into
801     its own source file (printint.c) and used #include to pull it into
802     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
803     separate copies.
804    
805     (b) Defined the list of op-code names for debugging as a macro in
806     internal.h so that it is next to the definition of the opcodes.
807    
808     (c) Defined a table of op-code lengths for simpler skipping along compiled
809     code. This is again a macro in internal.h so that it is next to the
810     definition of the opcodes.
811    
812     34. Added support for recursive calls to individual subpatterns, along the
813     lines of Robin Houston's patch (but implemented somewhat differently).
814    
815     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
816     allow it to read and process whole directories in Win32. This code was
817     contributed by Lionel Fourquaux; it has not been tested by me.
818    
819     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
820     used to name a group. Names consist of alphanumerics and underscores, and must
821     be unique. Back references use the syntax (?P=name) and recursive calls use
822     (?P>name) which is a PCRE extension to the Python extension. Groups still have
823     numbers. The function pcre_fullinfo() can be used after compilation to extract
824     a name/number map. There are three relevant calls:
825    
826     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
827     PCRE_INFO_NAMECOUNT yields the number of entries
828     PCRE_INFO_NAMETABLE yields a pointer to the map.
829    
830     The map is a vector of fixed-size entries. The size of each entry depends on
831     the length of the longest name used. The first two bytes of each entry are the
832     group number, most significant byte first. There follows the corresponding
833     name, zero terminated. The names are in alphabetical order.
834    
835     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
836     case instead of 255. Making it the same both with and without UTF-8 support
837     means that the same test output works with both.
838    
839     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
840     calling malloc() with a zero argument.
841    
842     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
843     optimization. I've improved things by keeping a bitmap of backreferences with
844     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
845     fact referenced, the optimization can be applied. It is unlikely that a
846     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
847     the match to follow \n) will appear inside brackets with a number greater than
848     31, but if it does, any back reference > 31 suppresses the optimization.
849    
850     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
851     of disabling numbered capturing parentheses. Any opening parenthesis that is
852     not followed by ? behaves as if it were followed by ?: but named parentheses
853     can still be used for capturing (and they will acquire numbers in the usual
854     way).
855    
856     41. Redesigned the return codes from the match() function into yes/no/error so
857     that errors can be passed back from deep inside the nested calls. A malloc
858     failure while inside a recursive subpattern call now causes the
859     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
860    
861     42. It is now possible to set a limit on the number of times the match()
862     function is called in a call to pcre_exec(). This facility makes it possible to
863     limit the amount of recursion and backtracking, though not in a directly
864     obvious way, because the match() function is used in a number of different
865     circumstances. The count starts from zero for each position in the subject
866     string (for non-anchored patterns). The default limit is, for compatibility, a
867     large number, namely 10 000 000. You can change this in two ways:
868    
869     (a) When configuring PCRE before making, you can use --with-match-limit=n
870     to set a default value for the compiled library.
871    
872     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
873     a different value is set. See 45 below.
874    
875     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
876    
877     43. Added a new function pcre_config(int, void *) to enable run-time extraction
878     of things that can be changed at compile time. The first argument specifies
879     what is wanted and the second points to where the information is to be placed.
880     The current list of available information is:
881    
882     PCRE_CONFIG_UTF8
883    
884     The output is an integer that is set to one if UTF-8 support is available;
885     otherwise it is set to zero.
886    
887     PCRE_CONFIG_NEWLINE
888    
889     The output is an integer that it set to the value of the code that is used for
890     newline. It is either LF (10) or CR (13).
891    
892     PCRE_CONFIG_LINK_SIZE
893    
894     The output is an integer that contains the number of bytes used for internal
895     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
896    
897     PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
898    
899     The output is an integer that contains the threshold above which the POSIX
900     interface uses malloc() for output vectors. See item 31 above.
901    
902     PCRE_CONFIG_MATCH_LIMIT
903    
904     The output is an unsigned integer that contains the default limit of the number
905     of match() calls in a pcre_exec() execution. See 42 above.
906    
907     44. pcretest has been upgraded by the addition of the -C option. This causes it
908     to extract all the available output from the new pcre_config() function, and to
909     output it. The program then exits immediately.
910    
911     45. A need has arisen to pass over additional data with calls to pcre_exec() in
912     order to support additional features. One way would have been to define
913     pcre_exec2() (for example) with extra arguments, but this would not have been
914     extensible, and would also have required all calls to the original function to
915     be mapped to the new one. Instead, I have chosen to extend the mechanism that
916     is used for passing in "extra" data from pcre_study().
917    
918     The pcre_extra structure is now exposed and defined in pcre.h. It currently
919     contains the following fields:
920    
921     flags a bitmap indicating which of the following fields are set
922     study_data opaque data from pcre_study()
923     match_limit a way of specifying a limit on match() calls for a specific
924     call to pcre_exec()
925     callout_data data for callouts (see 49 below)
926    
927     The flag bits are also defined in pcre.h, and are
928    
929     PCRE_EXTRA_STUDY_DATA
930     PCRE_EXTRA_MATCH_LIMIT
931     PCRE_EXTRA_CALLOUT_DATA
932    
933     The pcre_study() function now returns one of these new pcre_extra blocks, with
934     the actual study data pointed to by the study_data field, and the
935     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
936     before. That is, this change is entirely upwards-compatible and requires no
937     change to existing code.
938    
939     If you want to pass in additional data to pcre_exec(), you can either place it
940     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
941     block.
942    
943     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
944     data string contains the escape sequence \M, pcretest calls pcre_exec() several
945     times with different match limits, until it finds the minimum value needed for
946     pcre_exec() to complete. The value is then output. This can be instructive; for
947     most simple matches the number is quite small, but for pathological cases it
948     gets very large very quickly.
949    
950     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
951     returns the size of the data block pointed to by the study_data field in a
952     pcre_extra block, that is, the value that was passed as the argument to
953     pcre_malloc() when PCRE was getting memory in which to place the information
954     created by pcre_study(). The fourth argument should point to a size_t variable.
955     pcretest has been extended so that this information is shown after a successful
956     pcre_study() call when information about the compiled regex is being displayed.
957    
958     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
959     because what follows is always an absolute path. (Later: it turns out that this
960     is more than cosmetic for MinGW, because it doesn't like empty path
961     components.)
962    
963     49. Some changes have been made to the callout feature (see 28 above):
964    
965     (i) A callout function now has three choices for what it returns:
966    
967     0 => success, carry on matching
968     > 0 => failure at this point, but backtrack if possible
969     < 0 => serious error, return this value from pcre_exec()
970    
971     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
972     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
973     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
974     use by callout functions. It will never be used by PCRE itself.
975    
976     (ii) The pcre_extra structure (see 45 above) has a void * field called
977     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
978     pcre_callout_block structure has a field of the same name. The contents of
979     the field passed in the pcre_extra structure are passed to the callout
980     function in the corresponding field in the callout block. This makes it
981     easier to use the same callout-containing regex from multiple threads. For
982     testing, the pcretest program has a new data escape
983    
984     \C*n pass the number n (may be negative) as callout_data
985    
986     If the callout function in pcretest receives a non-zero value as
987     callout_data, it returns that value.
988    
989     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
990     there were some redundant $(CFLAGS) in commands that are now specified as
991     $(LINK), which already includes $(CFLAGS).
992    
993     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
994     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
995     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
996     one-byte characters throughout. Note that case-insensitive matching applies
997     only to characters whose values are less than 256. PCRE doesn't support the
998     notion of cases for higher-valued characters.
999    
1000     (i) A character class whose characters are all within 0-255 is handled as
1001     a bit map, and the map is inverted for negative classes. Previously, a
1002     character > 255 always failed to match such a class; however it should
1003     match if the class was a negative one (e.g. [^ab]). This has been fixed.
1004    
1005     (ii) A negated character class with a single character < 255 is coded as
1006     "not this character" (OP_NOT). This wasn't working properly when the test
1007     character was multibyte, either singly or repeated.
1008    
1009     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
1010     mode, for example: \x{100}{2,3}.
1011    
1012     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
1013     singly or repeated) now correctly test multibyte characters. However,
1014     PCRE doesn't recognize any characters with values greater than 255 as
1015     digits, spaces, or word characters. Such characters always match \D, \S,
1016     and \W, and never match \d, \s, or \w.
1017    
1018     (v) Classes may now contain characters and character ranges with values
1019     greater than 255. For example: [ab\x{100}-\x{400}].
1020    
1021     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
1022     PCRE in UTF-8 mode.
1023    
1024     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
1025     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
1026     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
1027     value.)
1028    
1029     53. The single man page has become too large. I have therefore split it up into
1030     a number of separate man pages. These also give rise to individual HTML pages;
1031     these are now put in a separate directory, and there is an index.html page that
1032     lists them all. Some hyperlinking between the pages has been installed.
1033    
1034     54. Added convenience functions for handling named capturing parentheses.
1035    
1036     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
1037     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
1038     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
1039     are faulted.
1040    
1041     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
1042     calling configure. These values are used when compiling the dftables.c program
1043     which is run to generate the source of the default character tables. They
1044     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
1045     you will need to set these values.
1046    
1047     57. Updated the building process for Windows DLL, as provided by Fred Cox.
1048    
1049    
1050     Version 3.9 02-Jan-02
1051 nigel 61 ---------------------
1052    
1053     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
1054    
1055     2. If --disable-static was given, the building process failed when trying to
1056     build pcretest and pcregrep. (For some reason it was using libtool to compile
1057     them, which is not right, as they aren't part of the library.)
1058    
1059    
1060 nigel 59 Version 3.8 18-Dec-01
1061     ---------------------
1062    
1063     1. The experimental UTF-8 code was completely screwed up. It was packing the
1064     bytes in the wrong order. How dumb can you get?
1065    
1066    
1067 nigel 57 Version 3.7 29-Oct-01
1068     ---------------------
1069    
1070     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
1071     This caused pcretest, when used on the test data, to segfault. Unfortunately,
1072     this didn't happen under Solaris 8, where I normally test things.
1073    
1074 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
1075     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
1076     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
1077     this fix an hour or so after the initial 3.7 release.)
1078 nigel 57
1079 nigel 59
1080 nigel 55 Version 3.6 23-Oct-01
1081     ---------------------
1082    
1083     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
1084     offsets passed as NULL with zero offset count.
1085    
1086     2. The config.guess and config.sub files had not been updated when I moved to
1087     the latest autoconf.
1088    
1089    
1090 nigel 53 Version 3.5 15-Aug-01
1091     ---------------------
1092 nigel 5
1093 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
1094     had been forgotten.
1095    
1096     2. By using declared but undefined structures, we can avoid using "void"
1097     definitions in pcre.h while keeping the internal definitions of the structures
1098     private.
1099    
1100     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
1101     user point of view, this means that both static and shared libraries are built
1102     by default, but this can be individually controlled. More of the work of
1103     handling this static/shared cases is now inside libtool instead of PCRE's make
1104     file.
1105    
1106     4. The pcretest utility is now installed along with pcregrep because it is
1107     useful for users (to test regexs) and by doing this, it automatically gets
1108     relinked by libtool. The documentation has been turned into a man page, so
1109     there are now .1, .txt, and .html versions in /doc.
1110    
1111     5. Upgrades to pcregrep:
1112     (i) Added long-form option names like gnu grep.
1113     (ii) Added --help to list all options with an explanatory phrase.
1114     (iii) Added -r, --recursive to recurse into sub-directories.
1115     (iv) Added -f, --file to read patterns from a file.
1116    
1117     6. pcre_exec() was referring to its "code" argument before testing that
1118     argument for NULL (and giving an error if it was NULL).
1119    
1120     7. Upgraded Makefile.in to allow for compiling in a different directory from
1121     the source directory.
1122    
1123     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
1124     options bits, the pointer it was passed was to an int instead of to an unsigned
1125     long int. This mattered only on 64-bit systems.
1126    
1127     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
1128     generated) instead of pcre.in, which it its source. Also made the same change
1129     in several of the .c files.
1130    
1131     10. A new release of gcc defines printf() as a macro, which broke pcretest
1132     because it had an ifdef in the middle of a string argument for printf(). Fixed
1133     by using separate calls to printf().
1134    
1135     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
1136     script, to force use of CR or LF instead of \n in the source. On non-Unix
1137     systems, the value can be set in config.h.
1138    
1139     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
1140     absolute limit. Changed the text of the error message to make this clear, and
1141     likewise updated the man page.
1142    
1143     13. The limit of 99 on the number of capturing subpatterns has been removed.
1144     The new limit is 65535, which I hope will not be a "real" limit.
1145    
1146    
1147 nigel 51 Version 3.4 22-Aug-00
1148     ---------------------
1149    
1150     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
1151    
1152     2. Diagnose condition (?(0) as an error instead of crashing on matching.
1153    
1154    
1155 nigel 49 Version 3.3 01-Aug-00
1156     ---------------------
1157    
1158     1. If an octal character was given, but the value was greater than \377, it
1159     was not getting masked to the least significant bits, as documented. This could
1160     lead to crashes in some systems.
1161    
1162     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
1163     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
1164    
1165     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
1166     These just pass their arguments on to (pcre_free)(), but they are provided
1167     because some uses of PCRE bind it to non-C systems that can call its functions,
1168     but cannot call free() or pcre_free() directly.
1169    
1170     4. Add "make test" as a synonym for "make check". Corrected some comments in
1171     the Makefile.
1172    
1173     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
1174     Makefile.
1175    
1176     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
1177     command called pgrep for grepping around the active processes.
1178    
1179     7. Added the beginnings of support for UTF-8 character strings.
1180    
1181     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
1182     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
1183     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
1184     out for the ar command.)
1185    
1186    
1187 nigel 47 Version 3.2 12-May-00
1188     ---------------------
1189    
1190     This is purely a bug fixing release.
1191    
1192     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
1193     of ZA. This was just one example of several cases that could provoke this bug,
1194     which was introduced by change 9 of version 2.00. The code for breaking
1195     infinite loops after an iteration that matches an empty string was't working
1196     correctly.
1197    
1198     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
1199     when matched against abbab (for example). After matching an empty string, it
1200     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
1201     caused it to match further down the string than it should.
1202    
1203     3. The code contained an inclusion of sys/types.h. It isn't clear why this
1204     was there because it doesn't seem to be needed, and it causes trouble on some
1205     systems, as it is not a Standard C header. It has been removed.
1206    
1207     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
1208     were reported on the Macintosh. The changes were from
1209    
1210     while ((c = *(++ptr)) != 0 && c != '\n');
1211     to
1212     while ((c = *(++ptr)) != 0 && c != '\n') ;
1213    
1214     Totally extraordinary, but if that's what it takes...
1215    
1216     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
1217     available. Added HAVE_BCOPY and an autoconf test for it; if neither
1218     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
1219     assumes the way PCRE uses memmove() (always moving upwards).
1220    
1221     6. PCRE is being used in one environment where strchr() is not available. There
1222     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
1223     faster code anyway.
1224    
1225    
1226 nigel 45 Version 3.1 09-Feb-00
1227     ---------------------
1228    
1229     The only change in this release is the fixing of some bugs in Makefile.in for
1230     the "install" target:
1231    
1232     (1) It was failing to install pcreposix.h.
1233    
1234     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
1235    
1236    
1237 nigel 43 Version 3.0 01-Feb-00
1238     ---------------------
1239 nigel 41
1240     1. Add support for the /+ modifier to perltest (to output $` like it does in
1241     pcretest).
1242    
1243     2. Add support for the /g modifier to perltest.
1244    
1245     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
1246     matches null strings.
1247    
1248     4. Fix perltest so that it doesn't do unwanted things when fed an empty
1249     pattern. Perl treats empty patterns specially - it reuses the most recent
1250     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
1251     effect.
1252    
1253     5. The POSIX interface was broken in that it was just handing over the POSIX
1254     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
1255     required a bigger vector, with some working space on the end. This means that
1256     the POSIX wrapper now has to get and free some memory, and copy the results.
1257    
1258 nigel 43 6. Added some simple autoconf support, placing the test data and the
1259     documentation in separate directories, re-organizing some of the
1260     information files, and making it build pcre-config (a GNU standard). Also added
1261     libtool support for building PCRE as a shared library, which is now the
1262     default.
1263 nigel 41
1264 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
1265     09 are not valid octal constants. Single digits will be used for minor values
1266     less than 10.
1267    
1268     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
1269     existing programs that set these in the POSIX interface can use PCRE without
1270     modification.
1271    
1272     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
1273     return all that pcre_info() returns, plus additional data. The pcre_info()
1274     function is retained for compatibility, but is considered to be obsolete.
1275    
1276     10. Added experimental recursion feature (?R) to handle one common case that
1277     Perl 5.6 will be able to do with (?p{...}).
1278    
1279     11. Added support for POSIX character classes like [:alpha:], which Perl is
1280     adopting.
1281    
1282    
1283 nigel 39 Version 2.08 31-Aug-99
1284     ----------------------
1285    
1286     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
1287     trying to match at the startoffset position, but instead was moving forward to
1288     the next newline as if a previous match had failed.
1289    
1290     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
1291     and could get into a loop if a null string was matched other than at the start
1292     of the subject.
1293    
1294     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
1295     be distinguished at compile time, and for completeness also added PCRE_DATE.
1296    
1297     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
1298     in GnuWin32 environments.
1299    
1300    
1301 nigel 37 Version 2.07 29-Jul-99
1302     ----------------------
1303    
1304     1. The documentation is now supplied in plain text form and HTML as well as in
1305     the form of man page sources.
1306    
1307     2. C++ compilers don't like assigning (void *) values to other pointer types.
1308     In particular this affects malloc(). Although there is no problem in Standard
1309     C, I've put in casts to keep C++ compilers happy.
1310    
1311     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
1312     should be (const char *).
1313    
1314     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
1315     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
1316     However, I haven't made this a standard facility. The documentation doesn't
1317     mention it, and the Makefile doesn't support it.
1318    
1319     5. The Makefile now contains an "install" target, with editable destinations at
1320     the top of the file. The pcretest program is not installed.
1321    
1322     6. pgrep -V now gives the PCRE version number and date.
1323    
1324     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
1325     causing the entire string to be ignored, instead of just the last character.
1326    
1327     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
1328     non-matching string, it can take a very, very long time, even for strings of
1329     quite modest length, because of the nested recursion. PCRE now does better in
1330     some of these cases. It does this by remembering the last required literal
1331     character in the pattern, and pre-searching the subject to ensure it is present
1332     before running the real match. In other words, it applies a heuristic to detect
1333     some types of certain failure quickly, and in the above example, if presented
1334     with a string that has no trailing " it gives "no match" very quickly.
1335    
1336     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
1337     other alternatives are tried instead.
1338    
1339    
1340 nigel 35 Version 2.06 09-Jun-99
1341     ----------------------
1342    
1343     1. Change pcretest's output for amount of store used to show just the code
1344     space, because the remainder (the data block) varies in size between 32-bit and
1345     64-bit systems.
1346    
1347     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
1348     start matching at. This allows lookbehinds to work when searching for multiple
1349     occurrences in a string.
1350    
1351     3. Added additional options to pcretest for testing multiple occurrences:
1352    
1353     /+ outputs the rest of the string that follows a match
1354     /g loops for multiple occurrences, using the new startoffset argument
1355     /G loops for multiple occurrences by passing an incremented pointer
1356    
1357     4. PCRE wasn't doing the "first character" optimization for patterns starting
1358     with \b or \B, though it was doing it for other lookbehind assertions. That is,
1359     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
1360     the letter 'x'. On long subject strings, this gives a significant speed-up.
1361    
1362    
1363 nigel 33 Version 2.05 21-Apr-99
1364     ----------------------
1365    
1366     1. Changed the type of magic_number from int to long int so that it works
1367     properly on 16-bit systems.
1368    
1369     2. Fixed a bug which caused patterns starting with .* not to work correctly
1370     when the subject string contained newline characters. PCRE was assuming
1371     anchoring for such patterns in all cases, which is not correct because .* will
1372     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
1373     DOTALL is set at top level; otherwise it knows that patterns starting with .*
1374     must be retried after every newline in the subject.
1375    
1376    
1377 nigel 31 Version 2.04 18-Feb-99
1378     ----------------------
1379    
1380     1. For parenthesized subpatterns with repeats whose minimum was zero, the
1381     computation of the store needed to hold the pattern was incorrect (too large).
1382     If such patterns were nested a few deep, this could multiply and become a real
1383     problem.
1384    
1385     2. Added /M option to pcretest to show the memory requirement of a specific
1386     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
1387    
1388     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
1389     compiled in such a way that the backtracking after subsequent failure was
1390     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
1391     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
1392    
1393    
1394 nigel 29 Version 2.03 02-Feb-99
1395     ----------------------
1396    
1397     1. Fixed typo and small mistake in man page.
1398    
1399 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
1400     LICENCE file containing the conditions.
1401 nigel 29
1402     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
1403     Perl, that is the internal \ allows the delimiter to be included in the
1404     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
1405     the final delimiter, add \ to the end of the pattern (to test the error).
1406    
1407     4. Added the convenience functions for extracting substrings after a successful
1408     match. Updated pcretest to make it able to test these functions.
1409    
1410    
1411 nigel 27 Version 2.02 14-Jan-99
1412     ----------------------
1413    
1414     1. Initialized the working variables associated with each extraction so that
1415     their saving and restoring doesn't refer to uninitialized store.
1416    
1417     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
1418     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
1419     fix the problem.
1420    
1421     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
1422     calls, and wasn't printing the correct value for compiling calls. Increased the
1423     default value of LOOPREPEAT, and the number of significant figures in the
1424     times.
1425    
1426     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
1427    
1428     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
1429     a building problem on Windows NT with a FAT file system.
1430    
1431    
1432 nigel 25 Version 2.01 21-Oct-98
1433     ----------------------
1434    
1435     1. Changed the API for pcre_compile() to allow for the provision of a pointer
1436     to character tables built by pcre_maketables() in the current locale. If NULL
1437     is passed, the default tables are used.
1438    
1439    
1440 nigel 23 Version 2.00 24-Sep-98
1441 nigel 21 ----------------------
1442    
1443 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
1444     it any more.
1445    
1446     2. Allow quantification of (?>) groups, and make it work correctly.
1447    
1448     3. The first character computation wasn't working for (?>) groups.
1449    
1450     4. Correct the implementation of \Z (it is permitted to match on the \n at the
1451     end of the subject) and add 5.005's \z, which really does match only at the
1452     very end of the subject.
1453    
1454     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
1455    
1456     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
1457     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
1458     localized options. All options to pcre_study() were also removed.
1459    
1460     7. Add other new features from 5.005:
1461    
1462     $(?<= positive lookbehind
1463     $(?<! negative lookbehind
1464     (?imsx-imsx) added the unsetting capability
1465     such a setting is global if at outer level; local otherwise
1466     (?imsx-imsx:) non-capturing groups with option setting
1467     (?(cond)re|re) conditional pattern matching
1468    
1469     A backreference to itself in a repeated group matches the previous
1470     captured string.
1471    
1472     8. General tidying up of studying (both automatic and via "study")
1473     consequential on the addition of new assertions.
1474    
1475     9. As in 5.005, unlimited repeated groups that could match an empty substring
1476     are no longer faulted at compile time. Instead, the loop is forcibly broken at
1477     runtime if any iteration does actually match an empty substring.
1478    
1479     10. Include the RunTest script in the distribution.
1480    
1481     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
1482     discrepancies, some of which were old and were also with respect to 5.004. They
1483     have now been fixed.
1484    
1485    
1486     Version 1.09 28-Apr-98
1487     ----------------------
1488    
1489 nigel 21 1. A negated single character class followed by a quantifier with a minimum
1490     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
1491     program crashes, or just wrong answers. This did not apply to negated classes
1492     containing more than one character, or to minima other than one.
1493    
1494    
1495 nigel 19 Version 1.08 27-Mar-98
1496     ----------------------
1497    
1498     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
1499    
1500     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
1501     latter must appear before anything that relies on it in the pattern.
1502    
1503    
1504 nigel 17 Version 1.07 16-Feb-98
1505     ----------------------
1506    
1507     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
1508     repeat of a potentially empty string).
1509    
1510    
1511 nigel 15 Version 1.06 23-Jan-98
1512     ----------------------
1513    
1514     1. Added Markus Oberhumer's little patches for C++.
1515    
1516     2. Literal strings longer than 255 characters were broken.
1517    
1518    
1519 nigel 13 Version 1.05 23-Dec-97
1520     ----------------------
1521    
1522     1. Negated character classes containing more than one character were failing if
1523     PCRE_CASELESS was set at run time.
1524    
1525    
1526 nigel 11 Version 1.04 19-Dec-97
1527     ----------------------
1528    
1529     1. Corrected the man page, where some "const" qualifiers had been omitted.
1530    
1531     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
1532     input syntax.
1533    
1534     3. Fixed memory leak which occurred when a regex with back references was
1535     matched with an offsets vector that wasn't big enough. The temporary memory
1536     that is used in this case wasn't being freed if the match failed.
1537    
1538     4. Tidied pcretest to ensure it frees memory that it gets.
1539    
1540     5. Temporary memory was being obtained in the case where the passed offsets
1541     vector was exactly big enough.
1542    
1543     6. Corrected definition of offsetof() from change 5 below.
1544    
1545     7. I had screwed up change 6 below and broken the rules for the use of
1546     setjmp(). Now fixed.
1547    
1548    
1549 nigel 9 Version 1.03 18-Dec-97
1550     ----------------------
1551    
1552     1. A erroneous regex with a missing opening parenthesis was correctly
1553     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
1554     on some systems.
1555    
1556     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
1557     it was reported that one broken compiler failed on the former because "code" is
1558     also an independent variable.
1559    
1560     3. The erroneous regex a[]b caused an array overrun reference.
1561    
1562     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
1563     fail on data ending with that character. (It was going on too far, and checking
1564     the next character, typically a binary zero.) This was specific to the
1565     optimized code for single-character negative classes.
1566    
1567     5. Added a contributed patch from the TIN world which does the following:
1568    
1569     + Add an undef for memmove, in case the the system defines a macro for it.
1570    
1571     + Add a definition of offsetof(), in case there isn't one. (I don't know
1572     the reason behind this - offsetof() is part of the ANSI standard - but
1573     it does no harm).
1574    
1575     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
1576     most of the places where whitespace preceded '#'. I have given up and
1577     allowed the remaining 2 cases to be at the margin.
1578    
1579     + Rename some variables in pcre to eliminate shadowing. This seems very
1580     pedantic, but does no harm, of course.
1581    
1582     6. Moved the call to setjmp() into its own function, to get rid of warnings
1583     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
1584    
1585     7. Constructs such as \d{8,} were compiling into the equivalent of
1586 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
1587 nigel 9 outcome, but in this particular case used more store than had been allocated,
1588     which caused the bug to be discovered because it threw up an internal error.
1589    
1590     8. The debugging code in both pcre and pcretest for outputting the compiled
1591     form of a regex was going wrong in the case of back references followed by
1592     curly-bracketed repeats.
1593    
1594    
1595 nigel 7 Version 1.02 12-Dec-97
1596     ----------------------
1597    
1598     1. Typos in pcre.3 and comments in the source fixed.
1599    
1600     2. Applied a contributed patch to get rid of places where it used to remove
1601     'const' from variables, and fixed some signed/unsigned and uninitialized
1602     variable warnings.
1603    
1604     3. Added the "runtest" target to Makefile.
1605    
1606     4. Set default compiler flag to -O2 rather than just -O.
1607    
1608    
1609 nigel 5 Version 1.01 19-Nov-97
1610     ----------------------
1611    
1612     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
1613     like /([ab]*)*/, that is, for classes with more than one character in them.
1614    
1615     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
1616     as /((?>a*))*/ (a PCRE_EXTRA facility).
1617    
1618    
1619     Version 1.00 18-Nov-97
1620     ----------------------
1621    
1622     1. Added compile-time macros to support systems such as SunOS4 which don't have
1623     memmove() or strerror() but have other things that can be used instead.
1624    
1625     2. Arranged that "make clean" removes the executables.
1626    
1627    
1628 nigel 3 Version 0.99 27-Oct-97
1629     ----------------------
1630    
1631     1. Fixed bug in code for optimizing classes with only one character. It was
1632     initializing a 32-byte map regardless, which could cause it to run off the end
1633     of the memory it had got.
1634    
1635     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
1636    
1637    
1638     Version 0.98 22-Oct-97
1639     ----------------------
1640    
1641     1. Fixed bug in code for handling temporary memory usage when there are more
1642     back references than supplied space in the ovector. This could cause segfaults.
1643    
1644    
1645     Version 0.97 21-Oct-97
1646     ----------------------
1647    
1648     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
1649    
1650     2. Optimized negated single characters not to use a bit map.
1651    
1652     3. Brought error texts together as macro definitions; clarified some of them;
1653     fixed one that was wrong - it said "range out of order" when it meant "invalid
1654     escape sequence".
1655    
1656     4. Changed some char * arguments to const char *.
1657    
1658     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
1659    
1660     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
1661     pcretest.
1662    
1663    
1664     Version 0.96 16-Oct-97
1665     ----------------------
1666    
1667     1. Added a simple "pgrep" utility to the distribution.
1668    
1669     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
1670     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
1671     where "ddd" means "one or more decimal digits".
1672    
1673     3. Fixed serious bug. If a pattern had a back reference, but the call to
1674     pcre_exec() didn't supply a large enough ovector to record the related
1675     identifying subpattern, the match always failed. PCRE now remembers the number
1676     of the largest back reference, and gets some temporary memory in which to save
1677     the offsets during matching if necessary, in order to ensure that
1678     backreferences always work.
1679    
1680     4. Increased the compatibility with Perl in a number of ways:
1681    
1682     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
1683     to request this handling. The option can be set at compile or exec time.
1684    
1685     (b) $ matches before a terminating newline by default; an option
1686     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
1687     mode). The option can be set at compile or exec time.
1688    
1689     (c) The handling of \ followed by a digit other than 0 is now supposed to be
1690     the same as Perl's. If the decimal number it represents is less than 10
1691     or there aren't that many previous left capturing parentheses, an octal
1692     escape is read. Inside a character class, it's always an octal escape,
1693     even if it is a single digit.
1694    
1695     (d) An escaped but undefined alphabetic character is taken as a literal,
1696     unless PCRE_EXTRA is set. Currently this just reserves the remaining
1697     escapes.
1698    
1699     (e) {0} is now permitted. (The previous item is removed from the compiled
1700     pattern).
1701    
1702     5. Changed all the names of code files so that the basic parts are no longer
1703     than 10 characters, and abolished the teeny "globals.c" file.
1704    
1705     6. Changed the handling of character classes; they are now done with a 32-byte
1706     bit map always.
1707    
1708     7. Added the -d and /D options to pcretest to make it possible to look at the
1709     internals of compilation without having to recompile pcre.
1710    
1711    
1712     Version 0.95 23-Sep-97
1713     ----------------------
1714    
1715     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
1716     \x20 at the start of a run of normal characters. These were being treated as
1717     real characters, instead of the source characters being re-checked.
1718    
1719    
1720     Version 0.94 18-Sep-97
1721     ----------------------
1722    
1723     1. The functions are now thread-safe, with the caveat that the global variables
1724     containing pointers to malloc() and free() or alternative functions are the
1725     same for all threads.
1726    
1727     2. Get pcre_study() to generate a bitmap of initial characters for non-
1728     anchored patterns when this is possible, and use it if passed to pcre_exec().
1729    
1730    
1731     Version 0.93 15-Sep-97
1732     ----------------------
1733    
1734     1. /(b)|(:+)/ was computing an incorrect first character.
1735    
1736     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
1737     but not actually doing anything yet.
1738    
1739     3. Treat "-" characters in classes that cannot be part of ranges as literals,
1740     as Perl does (e.g. [-az] or [az-]).
1741    
1742     4. Set the anchored flag if a branch starts with .* or .*? because that tests
1743     all possible positions.
1744    
1745     5. Split up into different modules to avoid including unneeded functions in a
1746     compiled binary. However, compile and exec are still in one module. The "study"
1747     function is split off.
1748    
1749     6. The character tables are now in a separate module whose source is generated
1750     by an auxiliary program - but can then be edited by hand if required. There are
1751     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
1752     toupper() in the code.
1753    
1754     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
1755     make them global. Abolish the function for setting them, as the caller can now
1756     set them directly.
1757    
1758    
1759     Version 0.92 11-Sep-97
1760     ----------------------
1761    
1762     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
1763     (e.g. /a{1,3}/) was broken (I mis-optimized it).
1764    
1765     2. Caseless matching was not working in character classes if the characters in
1766     the pattern were in upper case.
1767    
1768     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
1769    
1770     4. Make PCRE_ANCHORED public and accept as a compile option.
1771    
1772     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
1773     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
1774     pass them.
1775    
1776     6. Give an error if bad option bits passed at compile or run time.
1777    
1778     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
1779     pcretest to cause it to pass that flag.
1780    
1781     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
1782     options, and the first character, if set.
1783    
1784     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
1785    
1786    
1787     Version 0.91 10-Sep-97
1788     ----------------------
1789    
1790     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
1791     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
1792    
1793     2. PCRE was looping on encountering an indefinitely repeated back reference to
1794     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
1795     Perl does - treats the match as successful.
1796    
1797     ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12