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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 75 - (hide annotations) (download)
Sat Feb 24 21:40:37 2007 UTC (7 years, 5 months ago) by nigel
File size: 71737 byte(s)
Load pcre-5.0 into code/trunk.

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

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12