ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 391 - (hide annotations) (download)
Tue Mar 17 21:16:01 2009 UTC (6 years, 1 month ago) by ph10
File size: 162708 byte(s)
Add support for UTF-8 in EBCDIC environments.

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 376 Version 7.9 xx-xxx-09
5     ---------------------
7     1. When building with support for bzlib/zlib (pcregrep) and/or readline
8     (pcretest), all targets were linked against these libraries. This included
9     libpcre, libpcreposix, and libpcrecpp, even though they do not use these
10     libraries. This caused unwanted dependencies to be created. This problem
11     has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12     pcretest is linked with readline.
14     2. The "typedef int BOOL" in pcre_internal.h that was included inside the
15     "#ifndef FALSE" condition by an earlier change (probably 7.8/18) has been
16     moved outside it again, because FALSE and TRUE are already defined in AIX,
17     but BOOL is not.
19     3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20     PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
22     4. The pcregrep documentation said spaces were inserted as well as colons (or
23     hyphens) following file names and line numbers when outputting matching
24     lines. This is not true; no spaces are inserted. I have also clarified the
25     wording for the --colour (or --color) option.
26 ph10 377
27 ph10 378 5. In pcregrep, when --colour was used with -o, the list of matching strings
28     was not coloured; this is different to GNU grep, so I have changed it to be
29     the same.
31     6. When --colo(u)r was used in pcregrep, only the first matching substring in
32     each matching line was coloured. Now it goes on to look for further matches
33     of any of the test patterns, which is the same behaviour as GNU grep.
34 ph10 379
35     7. A pattern that could match an empty string could cause pcregrep to loop; it
36     doesn't make sense to accept an empty string match in pcregrep, so I have
37     locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38     seems to be how GNU grep behaves.
39 ph10 380
40     8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41     start or after a newline", because the conditional assertion was not being
42 ph10 381 correctly handled. The rule now is that both the assertion and what follows
43     in the first alternative must satisfy the test.
45     9. If auto-callout was enabled in a pattern with a conditional group, PCRE
46     could crash during matching.
47 ph10 383
48     10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
49     used for matching.
50 ph10 384
51     11. Unicode property support in character classes was not working for
52 ph10 385 characters (bytes) greater than 127 when not in UTF-8 mode.
53 ph10 386
54     12. Added the -M command line option to pcretest.
55 ph10 388
56     14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
57 ph10 389
58     15. Added the PCRE_NO_START_OPTIMIZE match-time option.
59 ph10 390
60     16. Added comments and documentation about mis-use of no_arg in the C++
61     wrapper.
62 ph10 391
63     17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
64     from Martin Jerabek that uses macro names for all relevant character and
65     string constants.
66 ph10 376
68 ph10 374 Version 7.8 05-Sep-08
69 ph10 349 ---------------------
71     1. Replaced UCP searching code with optimized version as implemented for Ad
72     Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
73 ph10 351 stage table and inline lookup instead of a function, giving speed ups of 2
74     to 5 times on some simple patterns that I tested. Permission was given to
75     distribute the MultiStage2.py script that generates the tables (it's not in
76     the tarball, but is in the Subversion repository).
77 ph10 349
78 ph10 351 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
79     scripts.
80 ph10 358
81 ph10 353 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
82     a group with a zero qualifier. The result of the study could be incorrect,
83 ph10 358 or the function might crash, depending on the pattern.
85     4. Caseless matching was not working for non-ASCII characters in back
86 ph10 354 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
87 ph10 358 It now works when Unicode Property Support is available.
89 ph10 355 5. In pcretest, an escape such as \x{de} in the data was always generating
90     a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
91     non-UTF-8 mode. If the value is greater than 255, it gives a warning about
92 ph10 358 truncation.
93 ph10 349
94 ph10 358 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
96 ph10 357 7. Added two (int) casts to pcregrep when printing the difference of two
97     pointers, in case they are 64-bit values.
98 ph10 371
99     8. Added comments about Mac OS X stack usage to the pcrestack man page and to
100     test 2 if it fails.
102 ph10 359 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
103 ph10 371 and a #define of that name to empty if it is not externally set. This is to
104     allow users of MSVC to set it if necessary.
106     10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
107 ph10 359 the convenience functions in the pcre_get.c source file.
108 ph10 371
109 ph10 360 11. An option change at the start of a pattern that had top-level alternatives
110     could cause overwriting and/or a crash. This command provoked a crash in
111 ph10 371 some environments:
113     printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
115 ph10 360 This potential security problem was recorded as CVE-2008-2371.
116 ph10 371
117 ph10 361 12. For a pattern where the match had to start at the beginning or immediately
118     after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
119 ph10 371 pcre_dfa_exec() could read past the end of the passed subject if there was
120 ph10 361 no match. To help with detecting such bugs (e.g. with valgrind), I modified
121     pcretest so that it places the subject at the end of its malloc-ed buffer.
122 ph10 371
123 ph10 364 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
124 ph10 371 exec() might read past the end of the data buffer in UTF-8 mode.
126 ph10 364 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
127 ph10 371 the data contained the byte 0x85 as part of a UTF-8 character within its
128     first line. This applied both to normal and DFA matching.
130 ph10 366 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
131 ph10 371 /^[^d]*?$/8 failed to match "abc".
132 ph10 351
133 ph10 371 16. Added a missing copyright notice to pcrecpp_internal.h.
135     17. Make it more clear in the documentation that values returned from
136 ph10 368 pcre_exec() in ovector are byte offsets, not character counts.
137 ph10 357
138 ph10 371 18. Tidied a few places to stop certain compilers from issuing warnings.
139 ph10 368
140 ph10 373 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
141     supplied by Stefan Weber. I made a further small update for 7.8 because
142 ph10 374 there is a change of source arrangements: the pcre_searchfuncs.c module is
143     replaced by pcre_ucd.c.
144 ph10 371
145 ph10 373
146 ph10 347 Version 7.7 07-May-08
147 ph10 321 ---------------------
149     1. Applied Craig's patch to sort out a long long problem: "If we can't convert
150 ph10 345 a string to a long long, pretend we don't even have a long long." This is
151 ph10 321 done by checking for the strtoq, strtoll, and _strtoi64 functions.
152 ph10 345
153 ph10 322 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
154     pre-7.6 versions, which defined a global no_arg variable instead of putting
155 ph10 332 it in the RE class. (See also #8 below.)
156 ph10 345
157     3. Remove a line of dead code, identified by coverity and reported by Nuno
158     Lopes.
160 ph10 324 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
161 ph10 321
162 ph10 324 (1) The include/exclude patterns were being applied to the whole pathnames
163 ph10 345 of files, instead of just to the final components.
165 ph10 324 (2) If there was more than one level of directory, the subdirectories were
166     skipped unless they satisfied the include/exclude conditions. This is
167     inconsistent with GNU grep (and could even be seen as contrary to the
168     pcregrep specification - which I improved to make it absolutely clear).
169     The action now is always to scan all levels of directory, and just
170     apply the include/exclude patterns to regular files.
171 ph10 345
172 ph10 325 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
173 ph10 345 --exclude_dir in the tests to avoid scanning .svn directories.
175 ph10 326 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
176 ph10 345 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
177 ph10 326 doesn't support NULs in patterns.
178 ph10 345
179     7. Added some missing "const"s to declarations of static tables in
180     pcre_compile.c and pcre_dfa_exec.c.
182 ph10 328 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
183 ph10 329 caused by fix #2 above. (Subsequently also a second patch to fix the
184 ph10 345 first patch. And a third patch - this was a messy problem.)
185 ph10 321
186 ph10 345 9. Applied Craig's patch to remove the use of push_back().
188 ph10 332 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
189     matching function regexec().
190 ph10 345
191     11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
192     which, however, unlike Perl's \g{...}, are subroutine calls, not back
193 ph10 333 references. PCRE supports relative numbers with this syntax (I don't think
194     Oniguruma does).
195 ph10 345
196     12. Previously, a group with a zero repeat such as (...){0} was completely
197 ph10 335 omitted from the compiled regex. However, this means that if the group
198     was called as a subroutine from elsewhere in the pattern, things went wrong
199 ph10 345 (an internal error was given). Such groups are now left in the compiled
200     pattern, with a new opcode that causes them to be skipped at execution
201 ph10 335 time.
202 ph10 345
203 ph10 341 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
204     to the way PCRE behaves:
205 ph10 345
206 ph10 336 (a) A lone ] character is dis-allowed (Perl treats it as data).
207 ph10 345
208     (b) A back reference to an unmatched subpattern matches an empty string
209 ph10 336 (Perl fails the current match path).
210 ph10 345
211 ph10 341 (c) A data ] in a character class must be notated as \] because if the
212 ph10 345 first data character in a class is ], it defines an empty class. (In
213 ph10 341 Perl it is not possible to have an empty class.) The empty class []
214     never matches; it forces failure and is equivalent to (*FAIL) or (?!).
215     The negative empty class [^] matches any one character, independently
216     of the DOTALL setting.
217 ph10 345
218     14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
219 ph10 340 non-existent subpattern following a character class starting with ']' and
220     containing () gave an internal compiling error instead of "reference to
221     non-existent subpattern". Fortunately, when the pattern did exist, the
222 ph10 345 compiled code was correct. (When scanning forwards to check for the
223 ph10 340 existencd of the subpattern, it was treating the data ']' as terminating
224 ph10 345 the class, so got the count wrong. When actually compiling, the reference
225 ph10 340 was subsequently set up correctly.)
226 ph10 345
227 ph10 341 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
228 ph10 345 it was being rejected as not supported by pcre_dfa_exec(), even though
229     other assertions are supported. I have made pcre_dfa_exec() support
230     (*FAIL).
232 ph10 342 16. The implementation of 13c above involved the invention of a new opcode,
233 ph10 345 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
234     cannot be changed at match time, I realized I could make a small
235     improvement to matching performance by compiling OP_ALLANY instead of
236     OP_ANY for "." when DOTALL was set, and then removing the runtime tests
237     on the OP_ANY path.
239     17. Compiling pcretest on Windows with readline support failed without the
240 ph10 343 following two fixes: (1) Make the unistd.h include conditional on
241 ph10 345 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
243 ph10 344 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
244 ph10 345 ncurses library to be included for pcretest when ReadLine support is
245     requested, but also to allow for it to be overridden. This patch came from
246     Daniel Bergström.
247 ph10 347
248 ph10 346 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
249 ph10 347 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
250     any errors with the current Unicode tables. Thanks to Peter Kankowski for
251     spotting this.
252 ph10 324
253 ph10 332
254 ph10 319 Version 7.6 28-Jan-08
255 ph10 300 ---------------------
257 ph10 302 1. A character class containing a very large number of characters with
258     codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
259     overflow.
260 ph10 309
261     2. Patch to cut out the "long long" test in pcrecpp_unittest when
262     HAVE_LONG_LONG is not defined.
264 ph10 303 3. Applied Christian Ehrlicher's patch to update the CMake build files to
265 ph10 304 bring them up to date and include new features. This patch includes:
266 ph10 309
267 ph10 304 - Fixed PH's badly added libz and libbz2 support.
268     - Fixed a problem with static linking.
269 ph10 312 - Added pcredemo. [But later removed - see 7 below.]
270 ph10 304 - Fixed dftables problem and added an option.
271     - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
273     - Added readline support for pcretest.
274 ph10 309 - Added an listing of the option settings after cmake has run.
276 ph10 314 4. A user submitted a patch to Makefile that makes it easy to create
277     "pcre.dll" under mingw when using Configure/Make. I added stuff to
278     Makefile.am that cause it to include this special target, without
279     affecting anything else. Note that the same mingw target plus all
280     the other distribution libraries and programs are now supported
281     when configuring with CMake (see 6 below) instead of with
282     Configure/Make.
283 ph10 309
284 ph10 308 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
285     This is an attempt to solve the reported problem "pcrecpp::no_arg is not
286 ph10 309 exported in the Windows port". It has not yet been confirmed that the patch
287     solves the problem, but it does no harm.
288 ph10 313
289 ph10 311 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
290 ph10 319 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
291 ph10 314 with CMake, and also correct the comment about stack recursion.
292 ph10 313
293 ph10 312 7. Remove the automatic building of pcredemo from the ./configure system and
294 ph10 313 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
295     of a program that users should build themselves after PCRE is installed, so
296     building it automatically is not really right. What is more, it gave
297 ph10 312 trouble in some build environments.
298 ph10 300
299 ph10 319 8. Further tidies to CMakeLists.txt from Sheri and Christian.
300 ph10 308
301 ph10 319
302 ph10 298 Version 7.5 10-Jan-08
303 ph10 263 ---------------------
305     1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
306     values in parens when parsing an RE using the C++ wrapper."
307 ph10 286
308 ph10 264 2. Negative specials like \S did not work in character classes in UTF-8 mode.
309     Characters greater than 255 were excluded from the class instead of being
310     included.
311 ph10 286
312     3. The same bug as (2) above applied to negated POSIX classes such as
313 ph10 265 [:^space:].
314 ph10 286
315 ph10 267 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
316     defined or documented. It seems to have been a typo for PCRE_STATIC, so
317 ph10 286 I have changed it.
319 ph10 268 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
320 ph10 286 first named subpattern) and a construct such as (?&a) would reference the
321     first named subpattern whose name started with "a" (in other words, the
322 ph10 272 length check was missing). Both these problems are fixed. "Subpattern name
323     expected" is now given for (?&) (a zero-length name), and this patch also
324     makes it give the same error for \k'' (previously it complained that that
325     was a reference to a non-existent subpattern).
326 ph10 286
327 ph10 269 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
328 ph10 286 this is right because (?- can be followed by option settings as well as by
329 ph10 269 digits. I have, however, made the messages clearer.
330 ph10 286
331 ph10 270 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
332     than the number used in the conditional) now cause a compile-time error.
333     This is actually not compatible with Perl, which accepts such patterns, but
334     treats the conditional as always being FALSE (as PCRE used to), but it
335     seems to me that giving a diagnostic is better.
336 ph10 286
337 ph10 274 8. Change "alphameric" to the more common word "alphanumeric" in comments
338 ph10 275 and messages.
339 ph10 286
340     9. Fix two occurrences of "backslash" in comments that should have been
341     "backspace".
343     10. Remove two redundant lines of code that can never be obeyed (their function
344     was moved elsewhere).
346     11. The program that makes PCRE's Unicode character property table had a bug
347     which caused it to generate incorrect table entries for sequences of
348 ph10 277 characters that have the same character type, but are in different scripts.
349 ph10 286 It amalgamated them into a single range, with the script of the first of
350 ph10 277 them. In other words, some characters were in the wrong script. There were
351     thirteen such cases, affecting characters in the following ranges:
352 ph10 286
353 ph10 277 U+002b0 - U+002c1
354     U+0060c - U+0060d
355 ph10 286 U+0061e - U+00612
356 ph10 277 U+0064b - U+0065e
357     U+0074d - U+0076d
358     U+01800 - U+01805
359     U+01d00 - U+01d77
360     U+01d9b - U+01dbf
361     U+0200b - U+0200f
362     U+030fc - U+030fe
363     U+03260 - U+0327f
364     U+0fb46 - U+0fbb1
365     U+10450 - U+1049d
366 ph10 286
367 ph10 279 12. The -o option (show only the matching part of a line) for pcregrep was not
368     compatible with GNU grep in that, if there was more than one match in a
369     line, it showed only the first of them. It now behaves in the same way as
370     GNU grep.
371 ph10 286
372     13. If the -o and -v options were combined for pcregrep, it printed a blank
373     line for every non-matching line. GNU grep prints nothing, and pcregrep now
374     does the same. The return code can be used to tell if there were any
375     non-matching lines.
377 ph10 289 14. Added --file-offsets and --line-offsets to pcregrep.
379     15. The pattern (?=something)(?R) was not being diagnosed as a potentially
380 ph10 282 infinitely looping recursion. The bug was that positive lookaheads were not
381 ph10 286 being skipped when checking for a possible empty match (negative lookaheads
382     and both kinds of lookbehind were skipped).
384 ph10 289 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
385 ph10 284 inclusion of <windows.h> to before rather than after the definition of
386     INVALID_FILE_ATTRIBUTES (patch from David Byron).
387 ph10 263
388 ph10 289 17. Specifying a possessive quantifier with a specific limit for a Unicode
389 ph10 286 character property caused pcre_compile() to compile bad code, which led at
390     runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
391     are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
392     caused the error; without that there was no problem.
393 ph10 263
394 ph10 289 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
395 ph10 286
396 ph10 289 19. Added --enable-pcretest-libreadline.
397 ph10 286
398 ph10 289 20. In pcrecpp.cc, the variable 'count' was incremented twice in
399 ph10 288 RE::GlobalReplace(). As a result, the number of replacements returned was
400 ph10 298 double what it should be. I removed one of the increments, but Craig sent a
401     later patch that removed the other one (the right fix) and added unit tests
402     that check the return values (which was not done before).
403 ph10 292
404 ph10 291 21. Several CMake things:
405 ph10 287
406 ph10 291 (1) Arranged that, when cmake is used on Unix, the libraries end up with
407 ph10 292 the names libpcre and libpcreposix, not just pcre and pcreposix.
408 ph10 288
409 ph10 292 (2) The above change means that pcretest and pcregrep are now correctly
410     linked with the newly-built libraries, not previously installed ones.
411 ph10 291
413 ph10 298
414     22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
415     crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
416     UTF-8 newline character). The key issue is that the pattern starts .*;
417     this means that the match must be either at the beginning, or after a
418     newline. The bug was in the code for advancing after a failed match and
419     checking that the new position followed a newline. It was not taking
420 ph10 294 account of UTF-8 characters correctly.
421 ph10 298
422     23. PCRE was behaving differently from Perl in the way it recognized POSIX
423     character classes. PCRE was not treating the sequence [:...:] as a
424     character class unless the ... were all letters. Perl, however, seems to
425     allow any characters between [: and :], though of course it rejects as
426     unknown any "names" that contain non-letters, because all the known class
427     names consist only of letters. Thus, Perl gives an error for [[:1234:]],
428     for example, whereas PCRE did not - it did not recognize a POSIX character
429     class. This seemed a bit dangerous, so the code has been changed to be
430     closer to Perl. The behaviour is not identical to Perl, because PCRE will
431     diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
432     treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
433     Perl does, and where it didn't before.
435 ph10 296 24. Rewrite so as to remove the single use of %n from pcregrep because in some
436     Windows environments %n is disabled by default.
437 ph10 292
439 ph10 260 Version 7.4 21-Sep-07
440 ph10 230 ---------------------
442     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
443 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
444     LF". That isn't really right - the whole point of the change was to try to
445     help when there was an actual mention of one of the two characters. So now
446     the change happens only if \r or \n (or a literal CR or LF) character is
447 ph10 230 encountered.
448 ph10 231
449     2. The 32-bit options word was also used for 6 internal flags, but the numbers
450     of both had grown to the point where there were only 3 bits left.
451     Fortunately, there was spare space in the data structure, and so I have
452     moved the internal flags into a new 16-bit field to free up more option
453 ph10 230 bits.
454 ph10 231
455     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
456     but did not set the internal JCHANGED flag - either of these is enough to
457     control the way the "get" function works - but the PCRE_INFO_JCHANGED
458     facility is supposed to tell if (?J) was ever used, so now (?J) at the
459 ph10 230 start sets both bits.
461 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
462     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
463 ph10 230
464 ph10 243 5. doc/pcresyntax.html was missing from the distribution.
466     6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
467 ph10 233 compatibility, even though it is no longer used.
468 ph10 243
469     7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
470 ph10 254 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
471 ph10 259 windows.h file is present (where different names are used). [This was
472 ph10 257 reversed later after testing - see 16 below.]
473 ph10 243
474     8. Changed all #include <config.h> to #include "config.h". There were also
475 ph10 236 some further <pcre.h> cases that I changed to "pcre.h".
476 ph10 243
477 ph10 239 9. When pcregrep was used with the --colour option, it missed the line ending
478 ph10 243 sequence off the lines that it output.
480     10. It was pointed out to me that arrays of string pointers cause lots of
481     relocations when a shared library is dynamically loaded. A technique of
482     using a single long string with a table of offsets can drastically reduce
483     these. I have refactored PCRE in four places to do this. The result is
484 ph10 240 dramatic:
485 ph10 243
486 ph10 240 Originally: 290
487     After changing UCP table: 187
488 ph10 243 After changing error message table: 43
489 ph10 240 After changing table of "verbs" 36
490     After changing table of Posix names 22
491 ph10 243
492 ph10 240 Thanks to the folks working on Gregex for glib for this insight.
493 ph10 247
494 ph10 244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
495 ph10 247 unicode-properties was also set.
497 ph10 248 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
498 ph10 231
499 ph10 253 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
500     checked only for CRLF.
501 ph10 233
502 ph10 259 14. Added casts to pcretest.c to avoid compiler warnings.
504 ph10 256 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
505 ph10 253
506 ph10 257 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
507 ph10 259 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
508     entirely. This removes changes made in 7 above.
509 ph10 256
510 ph10 261 17. The CMake files have been updated, and there is now more information about
511     building with CMake in the NON-UNIX-USE document.
512 ph10 257
513 ph10 261
514 ph10 228 Version 7.3 28-Aug-07
515 ph10 157 ---------------------
517 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
518     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
519     brackets there is not right, since it causes compilers to look for an
520     installed pcre.h, not the version that is in the source that is being
521     compiled (which of course may be different). I have changed it back to:
522 ph10 197
523 ph10 189 #include "pcre.h"
524 ph10 197
525     I have a vague recollection that the change was concerned with compiling in
526     different directories, but in the new build system, that is taken care of
527     by the VPATH setting the Makefile.
529 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
530     when the subject happened to end in the byte 0x85 (e.g. if the last
531     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
532     characters but of course it shouldn't be taken as a newline when it is part
533     of another character. The bug was that, for an unlimited repeat of . in
534     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
535     characters when looking for a newline.
536 ph10 197
537     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
539     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
540     in debug output.
542 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
543     long printing in the pcrecpp unittest when running under MinGW.
544 ph10 189
545 ph10 197 6. ESC_K was left out of the EBCDIC table.
546 ph10 189
547 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
548     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
549     limit also applies to "virtual nesting" when a pattern is recursive, and in
550     this case 1000 isn't so big. I have been able to remove this limit at the
551     expense of backing off one optimization in certain circumstances. Normally,
552     when pcre_exec() would call its internal match() function recursively and
553     immediately return the result unconditionally, it uses a "tail recursion"
554     feature to save stack. However, when a subpattern that can match an empty
555     string has an unlimited repetition quantifier, it no longer makes this
556     optimization. That gives it a stack frame in which to save the data for
557     checking that an empty string has been matched. Previously this was taken
558     from the 1000-entry workspace that had been reserved. So now there is no
559     explicit limit, but more stack is used.
560 ph10 189
561 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
562     syntax that is required for Windows, and which was going wrong for the
563     pcreposix and pcrecpp parts of the library. These were overlooked when this
564     problem was solved for the main library.
565 ph10 197
566 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
567     the size of patterns that contain repeated groups with explicit upper
568     limits. As the maximum quantifier is 65535, the maximum group length was
569     set at 30,000 so that the product of these two numbers did not overflow a
570     32-bit integer. However, it turns out that people want to use groups that
571     are longer than 30,000 bytes (though not repeat them that many times).
572     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
573     made it possible to implement the integer overflow checks in a much more
574     dynamic way, which I have now done. The artificial limitation on group
575     length has been removed - we now have only the limit on the total length of
576     the compiled pattern, which depends on the LINK_SIZE setting.
577 ph10 208
578     10. Fixed a bug in the documentation for get/copy named substring when
579     duplicate names are permitted. If none of the named substrings are set, the
580     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
581     empty string.
583     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
584     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
585     because the ] is interpreted as the first data character and the
586     terminating ] is not found. PCRE has been made compatible with Perl in this
587     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
588     cause memory overwriting.
590 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
591     string has been matched (to stop an infinite loop). It was not recognizing
592 ph10 208 a conditional subpattern that could match an empty string if that
593 ph10 206 subpattern was within another subpattern. For example, it looped when
594 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
595 ph10 206 condition was not nested. This bug has been fixed.
596 ph10 208
597 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
598     past the start of the subject in the presence of bytes with the top bit
599     set, for example "\x8aBCD".
600 ph10 211
601 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
602     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
603 ph10 200
604 ph10 211 14. Optimized (?!) to (*FAIL).
605 ph10 202
606 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
607     This restricts code points to be within the range 0 to 0x10FFFF, excluding
608     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
609     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
610     does: it's just the validity check that is more restrictive.
611 ph10 220
612     16. Inserted checks for integer overflows during escape sequence (backslash)
613     processing, and also fixed erroneous offset values for syntax errors during
614     backslash processing.
616 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
617 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
619 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
620     caused an overrun.
621 ph10 220
622     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
623     something other than just ASCII characters) inside a group that had an
624     unlimited repeat caused a loop at compile time (while checking to see
625     whether the group could match an empty string).
627 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
628     [\P{Any}] did so. (Error in the code for printing property names.)
629 ph10 210
630 ph10 220 21. An orphan \E inside a character class could cause a crash.
632     22. A repeated capturing bracket such as (A)? could cause a wild memory
633 ph10 218 reference during compilation.
634 ph10 220
635     23. There are several functions in pcre_compile() that scan along a compiled
636     expression for various reasons (e.g. to see if it's fixed length for look
637 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
638 ph10 220 present in the pattern. These operators have additional parameters compared
639     with \d, etc, and these were not being taken into account when moving along
640 ph10 218 the compiled data. Specifically:
641 ph10 220
642     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
643     length.
645     (b) An item such as \pL+ within a repeated group could cause crashes or
646 ph10 218 loops.
647 ph10 220
648 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
649 ph10 220 "reference to non-existent subpattern" error.
651 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
653 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
654 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
655 ph10 211
656 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
657 ph10 218
658 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
659 ph10 223 character were causing crashes (broken optimization).
660 ph10 227
661 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
662 ph10 227 \p or \P) caused a compile-time loop.
664 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
665     break. For example, the unstudied pattern [\r\n]A does not match the string
666     "\r\nA" because change 7.0/46 below moves the current point on by two
667     characters after failing to match at the start. However, the pattern \nA
668 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
669     the same is true. There doesn't seem any very clean way out of this, but
670     what I have chosen to do makes the common cases work: PCRE now takes note
671     of whether there can be an explicit match for \r or \n anywhere in the
672     pattern, and if so, 7.0/46 no longer applies. As part of this change,
673     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
674     pattern has explicit CR or LF references.
675 ph10 222
676 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
677 ph10 223
678 ph10 227
679 ph10 189 Version 7.2 19-Jun-07
680     ---------------------
682 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
683     which is apparently normally available under Windows.
684 ph10 159
685     2. Re-jig the pcregrep tests with different newline settings in an attempt
686     to make them independent of the local environment's newline setting.
687 ph10 157
688 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
689 ph10 165
690 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
691     was not 2, because the output contained actual offsets. The recent new
692 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
693     usable with all link sizes.
695 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
696     stack recursion. This gives a massive performance boost under BSD, but just
697 ph10 165 a small improvement under Linux. However, it saves one field in the frame
698 ph10 164 in all cases.
699 ph10 172
700 ph10 166 6. Added more features from the forthcoming Perl 5.10:
701 ph10 172
702 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
703     recursion call. It refers to the nth most recently opened parentheses.
704 ph10 172
705 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
706 ph10 172 to be opened parentheses.
708     (c) Conditions that refer to capturing parentheses can be specified
709 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
710 ph10 172
711 ph10 168 (d) \K resets the start of the current match so that everything before
712 ph10 172 is not part of it.
714 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
715 ph10 172
716 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
717 ph10 172 reference syntax.
718 ph10 182
719 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
720 ph10 182 alternative starts with the same number.
721 ph10 172
722 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
724 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
727     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
728     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
729 ph10 170 for detecting groups that can match an empty string.
730 ph10 159
731 ph10 172 9. A pattern with a very large number of alternatives (more than several
732     hundred) was running out of internal workspace during the pre-compile
733     phase, where pcre_compile() figures out how much memory will be needed. A
734     bit of new cunning has reduced the workspace needed for groups with
735     alternatives. The 1000-alternative test pattern now uses 12 bytes of
736     workspace instead of running out of the 4096 that are available.
737 ph10 182
738 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
739 ph10 172
740 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
741     The report of the bug said:
742 ph10 182
743 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
744     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
745 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
746 ph10 185
747 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
748 ph10 185 it matched the wrong number of bytes.
749 ph10 172
750 ph10 179
751 ph10 155 Version 7.1 24-Apr-07
752 ph10 98 ---------------------
754 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
755 ph10 122 that is more "standard", making use of automake and other Autotools. There
756 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
757     on this.
758 ph10 111
759     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
760     for recursive directory scanning broke on some systems because the files
761     are not scanned in any specific order and on different systems the order
762     was different. A call to "sort" has been inserted into RunGrepTest for the
763     approprate test as a short-term fix. In the longer term there may be an
764 ph10 100 alternative.
765 ph10 111
766 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
767 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
768     people do their own conversions for various reasons). The problems
769     concerned the use of low-level troff macros .br and .in. I have therefore
770     removed all such uses from the man pages (some were redundant, some could
771 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
772     HTML has been updated to handle .nf/.fi and to complain if it encounters
773     .br or .in.
774 ph10 111
775 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
776 ph10 123 arranged for config.h to be included in the distribution, with the name
777 ph10 111 config.h.generic, for the benefit of those who have to compile without
778     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
780     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
781     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
782 ph10 127 makevp.bat for the current PCRE, using the additional files
783 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
784 ph10 111
785     6. A Windows user reported a minor discrepancy with test 2, which turned out
786     to be caused by a trailing space on an input line that had got lost in his
787 ph10 102 copy. The trailing space was an accident, so I've just removed it.
788 ph10 111
789 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
790 ph10 111 that is needed.
792 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
793     as "const" (a) because they are and (b) because it helps the PHP
794     maintainers who have recently made a script to detect big data structures
795 ph10 111 in the php code that should be moved to the .rodata section. I remembered
796     to update Builducptable as well, so it won't revert if ucptable.h is ever
797 ph10 105 re-created.
798 ph10 111
799     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
800 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
801 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
802     support is not required. This saves 1.5-2K of code, which is important in
803 ph10 107 some applications.
804 ph10 111
805 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
806 ph10 111 so as not to refer to the tables, even though these functions will never be
807     called when UTF-8 support is disabled. Otherwise there are problems with a
808     shared library.
810 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
812     (a) It was defining its arguments as char * instead of void *.
814 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
815     a long time ago when I wrote it, but is no longer the case.
817 ph10 118 The emulated memove() is provided for those environments that have neither
818 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
819 ph10 118 is clearly not the case, as these two bugs were recently reported.
820 ph10 123
821 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
822 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
823     pages, and it removes trailing spaces from listed files. It also creates
824     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
825     case, it wraps all the #defines with #ifndefs. This script should be run
826 ph10 111 before "make dist".
827 ph10 123
828 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
829     with Unicode property support.
830 ph10 123
831     (a) For a maximizing quantifier, if the two different cases of the
832     character were of different lengths in their UTF-8 codings (there are
833     some cases like this - I found 11), and the matching function had to
834 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
835     were both the same length.
836 ph10 123
837     (b) When PCRE was configured to use the heap rather than the stack for
838     recursion during matching, it was not correctly preserving the data for
839     the other case of a UTF-8 character when checking ahead for a match
840     while processing a minimizing repeat. If the check also involved
841 ph10 115 matching a wide character, but failed, corruption could cause an
842     erroneous result when trying to check for a repeat of the original
843     character.
844 ph10 123
845 ph10 116 13. Some tidying changes to the testing mechanism:
846 ph10 98
847 ph10 116 (a) The RunTest script now detects the internal link size and whether there
848     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
849 ph10 123 values substituted by "configure". (The RunGrepTest script already did
850     this for UTF-8.) The configure.ac script no longer substitutes the
851     relevant variables.
853 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
854     with length and offset values. This means that the output is different
855     for different internal link sizes. Test 2 is skipped for link sizes
856     other than 2 because of this, bypassing the problem. Unfortunately,
857     there was also a test in test 3 (the locale tests) that used /B and
858 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
859     I have added a new /Z option to pcretest that replaces the length and
860     offset values with spaces. This is now used to make test 3 independent
861 ph10 122 of link size. (Test 2 will be tidied up later.)
862 ph10 123
863     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
864 ph10 122 segmentation fault instead of returning the appropriate error message.
865 ph10 134
866 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
867 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
868     This doesn't seem right; it now treats the CRLF combination as the line
869     ending, and so does not match in that case. It's only a pattern such as ^$
870     that would hit this one: something like ^ABC$ would have failed after \r
871     and then tried again after \r\n.
873 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
874 ph10 134 in an attempt to make files that differ only in their line terminators
875     compare equal. This works on Linux.
876 ph10 142
877 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
878     as it exited. This is now fixed, thanks to valgrind.
879 ph10 142
880 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
881 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
882     was because its rules for how to advance for /g after matching an empty
883 ph10 143 string at the end of a line did not allow for this case. They now check for
884     it specially.
885 ph10 150
886     20. pcretest is supposed to handle patterns and data of any length, by
887     extending its buffers when necessary. It was getting this wrong when the
888 ph10 147 buffer for a data line had to be extended.
889 ph10 150
890 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
891 ph10 150 CRLF as a newline sequence.
892 ph10 152
893 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
894 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
895     I have nevertheless tidied it up.
896 ph10 111
897 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
898 ph10 123
899 ph10 153 24. Added a man page for pcre-config.
900 ph10 152
901 ph10 153
902 nigel 93 Version 7.0 19-Dec-06
903     ---------------------
905     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
906     moving to gcc 4.1.1.
908     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
909     sys/time.h, which is documented as needed for this function. It doesn't
910     seem to matter on Linux, but it showed up on some releases of OS X.
912     3. It seems that there are systems where bytes whose values are greater than
913     127 match isprint() in the "C" locale. The "C" locale should be the
914     default when a C program starts up. In most systems, only ASCII printing
915     characters match isprint(). This difference caused the output from pcretest
916     to vary, making some of the tests fail. I have changed pcretest so that:
918     (a) When it is outputting text in the compiled version of a pattern, bytes
919     other than 32-126 are always shown as hex escapes.
921     (b) When it is outputting text that is a matched part of a subject string,
922     it does the same, unless a different locale has been set for the match
923     (using the /L modifier). In this case, it uses isprint() to decide.
925     4. Fixed a major bug that caused incorrect computation of the amount of memory
926     required for a compiled pattern when options that changed within the
927     pattern affected the logic of the preliminary scan that determines the
928     length. The relevant options are -x, and -i in UTF-8 mode. The result was
929     that the computed length was too small. The symptoms of this bug were
930     either the PCRE error "internal error: code overflow" from pcre_compile(),
931     or a glibc crash with a message such as "pcretest: free(): invalid next
932     size (fast)". Examples of patterns that provoked this bug (shown in
933     pcretest format) are:
935     /(?-x: )/x
936     /(?x)(?-x: \s*#\s*)/
937     /((?i)[\x{c0}])/8
938     /(?i:[\x{c0}])/8
940     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
941     is now done differently.
943     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
944     wrapper classes; (b) implement a new function in the C++ scanner that is
945     more efficient than the old way of doing things because it avoids levels of
946     recursion in the regex matching; (c) add a paragraph to the documentation
947     for the FullMatch() function.
949     6. The escape sequence \n was being treated as whatever was defined as
950     "newline". Not only was this contrary to the documentation, which states
951     that \n is character 10 (hex 0A), but it also went horribly wrong when
952     "newline" was defined as CRLF. This has been fixed.
954     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
955     was being set to -1 for the "end of line" case (supposedly a value that no
956     character can have). Though this value is never used (the check for end of
957     line is "zero bytes in current character"), it caused compiler complaints.
958     I've changed it to 0xffffffff.
960     8. In pcre_version.c, the version string was being built by a sequence of
961     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
962     string (as it is for production releases) called a macro with an empty
963     argument. The C standard says the result of this is undefined. The gcc
964     compiler treats it as an empty string (which was what was wanted) but it is
965     reported that Visual C gives an error. The source has been hacked around to
966     avoid this problem.
968     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
969     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
970     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
971     of them did).
973     10. Originally, pcretest opened its input and output without "b"; then I was
974     told that "b" was needed in some environments, so it was added for release
975     5.0 to both the input and output. (It makes no difference on Unix-like
976     systems.) Later I was told that it is wrong for the input on Windows. I've
977     now abstracted the modes into two macros, to make it easier to fiddle with
978     them, and removed "b" from the input mode under Windows.
980     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
982     12. Added -help and --help to pcretest as an official way of being reminded
983     of the options.
985     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
986     and pcrecpp.cc because they annoy compilers at high warning levels.
988     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
990     15. Fixed an occurrence of == in configure.ac that should have been = (shell
991     scripts are not C programs :-) and which was not noticed because it works
992     on Linux.
994     16. pcretest is supposed to handle any length of pattern and data line (as one
995     line or as a continued sequence of lines) by extending its input buffer if
996     necessary. This feature was broken for very long pattern lines, leading to
997     a string of junk being passed to pcre_compile() if the pattern was longer
998     than about 50K.
1000     17. I have done a major re-factoring of the way pcre_compile() computes the
1001     amount of memory needed for a compiled pattern. Previously, there was code
1002     that made a preliminary scan of the pattern in order to do this. That was
1003     OK when PCRE was new, but as the facilities have expanded, it has become
1004     harder and harder to keep it in step with the real compile phase, and there
1005     have been a number of bugs (see for example, 4 above). I have now found a
1006     cunning way of running the real compile function in a "fake" mode that
1007     enables it to compute how much memory it would need, while actually only
1008     ever using a few hundred bytes of working memory and without too many
1009     tests of the mode. This should make future maintenance and development
1010     easier. A side effect of this work is that the limit of 200 on the nesting
1011     depth of parentheses has been removed (though this was never a serious
1012     limitation, I suspect). However, there is a downside: pcre_compile() now
1013     runs more slowly than before (30% or more, depending on the pattern). I
1014     hope this isn't a big issue. There is no effect on runtime performance.
1016     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1017     newline (only possible for the last line of a file) and it was a
1018     pattern that set a locale (followed by /Lsomething), pcretest crashed.
1020     19. Added additional timing features to pcretest. (1) The -tm option now times
1021     matching only, not compiling. (2) Both -t and -tm can be followed, as a
1022     separate command line item, by a number that specifies the number of
1023     repeats to use when timing. The default is 50000; this gives better
1024     precision, but takes uncomfortably long for very large patterns.
1026     20. Extended pcre_study() to be more clever in cases where a branch of a
1027     subpattern has no definite first character. For example, (a*|b*)[cd] would
1028     previously give no result from pcre_study(). Now it recognizes that the
1029     first character must be a, b, c, or d.
1031     21. There was an incorrect error "recursive call could loop indefinitely" if
1032     a subpattern (or the entire pattern) that was being tested for matching an
1033     empty string contained only one non-empty item after a nested subpattern.
1034     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1035     incorrectly, because the \d was being skipped in the check.
1037     22. The pcretest program now has a new pattern option /B and a command line
1038     option -b, which is equivalent to adding /B to every pattern. This causes
1039     it to show the compiled bytecode, without the additional information that
1040     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1041     is the same as /B/I).
1043     23. A new optimization is now able automatically to treat some sequences such
1044     as a*b as a*+b. More specifically, if something simple (such as a character
1045     or a simple class like \d) has an unlimited quantifier, and is followed by
1046     something that cannot possibly match the quantified thing, the quantifier
1047     is automatically "possessified".
1049     24. A recursive reference to a subpattern whose number was greater than 39
1050     went wrong under certain circumstances in UTF-8 mode. This bug could also
1051     have affected the operation of pcre_study().
1053     25. Realized that a little bit of performance could be had by replacing
1054     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1056     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1058     27. Possessive quantifiers such as a++ were previously implemented by turning
1059     them into atomic groups such as ($>a+). Now they have their own opcodes,
1060     which improves performance. This includes the automatically created ones
1061     from 23 above.
1063     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1064     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1065     the first matched character to be a colon. This applied both to named and
1066     numbered groups.
1068     29. The ucpinternal.h header file was missing its idempotency #ifdef.
1070     30. I was sent a "project" file called libpcre.a.dev which I understand makes
1071     building PCRE on Windows easier, so I have included it in the distribution.
1073     31. There is now a check in pcretest against a ridiculously large number being
1074     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1075     loop, the loop is abandoned.
1077     32. Forward references to subpatterns in conditions such as (?(2)...) where
1078     subpattern 2 is defined later cause pcre_compile() to search forwards in
1079     the pattern for the relevant set of parentheses. This search went wrong
1080     when there were unescaped parentheses in a character class, parentheses
1081     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1083     33. "Subroutine" calls and backreferences were previously restricted to
1084     referencing subpatterns earlier in the regex. This restriction has now
1085     been removed.
1087     34. Added a number of extra features that are going to be in Perl 5.10. On the
1088     whole, these are just syntactic alternatives for features that PCRE had
1089     previously implemented using the Python syntax or my own invention. The
1090     other formats are all retained for compatibility.
1092     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1093     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1094     also .NET compatible.
1096     (b) A recursion or subroutine call to a named group can now be defined as
1097     (?&name) as well as (?P>name).
1099     (c) A backreference to a named group can now be defined as \k<name> or
1100     \k'name' as well as (?P=name). The new forms, as well as being in Perl
1101     5.10, are also .NET compatible.
1103     (d) A conditional reference to a named group can now use the syntax
1104     (?(<name>) or (?('name') as well as (?(name).
1106     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1107     groups (named and numbered) that are never evaluated inline, but can be
1108     called as "subroutines" from elsewhere. In effect, the DEFINE condition
1109     is always false. There may be only one alternative in such a group.
1111     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1112     as the simple (?(R). The condition is true only if the most recent
1113     recursion is that of the given number or name. It does not search out
1114     through the entire recursion stack.
1116     (g) The escape \gN or \g{N} has been added, where N is a positive or
1117     negative number, specifying an absolute or relative reference.
1119     35. Tidied to get rid of some further signed/unsigned compiler warnings and
1120     some "unreachable code" warnings.
1122     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1123     things, this adds five new scripts.
1125     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1126     There were also incompatibilities regarding the handling of \Q..\E inside
1127     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1128     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1130     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1131     matches an empty string, and forcibly breaks the loop. There were bugs in
1132     this code in non-simple cases. For a pattern such as ^(a()*)* matched
1133     against aaaa the result was just "a" rather than "aaaa", for example. Two
1134     separate and independent bugs (that affected different cases) have been
1135     fixed.
1137     39. Refactored the code to abolish the use of different opcodes for small
1138     capturing bracket numbers. This is a tidy that I avoided doing when I
1139     removed the limit on the number of capturing brackets for 3.5 back in 2001.
1140     The new approach is not only tidier, it makes it possible to reduce the
1141     memory needed to fix the previous bug (38).
1143     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1144     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1145     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1146     mode.
1148     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1149     report.
1151     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1152     copy construction and assignment in the C++ wrapper.
1154     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1155     couple of bugs that could have given wrong results in the "--newline=crlf"
1156     case.
1158     44. Added a number of casts and did some reorganization of signed/unsigned int
1159     variables following suggestions from Dair Grant. Also renamed the variable
1160     "this" as "item" because it is a C++ keyword.
1162     45. Arranged for dftables to add
1164     #include "pcre_internal.h"
1166     to pcre_chartables.c because without it, gcc 4.x may remove the array
1167     definition from the final binary if PCRE is built into a static library and
1168     dead code stripping is activated.
1170     46. For an unanchored pattern, if a match attempt fails at the start of a
1171     newline sequence, and the newline setting is CRLF or ANY, and the next two
1172     characters are CRLF, advance by two characters instead of one.
1175 nigel 91 Version 6.7 04-Jul-06
1176     ---------------------
1178     1. In order to handle tests when input lines are enormously long, pcretest has
1179     been re-factored so that it automatically extends its buffers when
1180     necessary. The code is crude, but this _is_ just a test program. The
1181     default size has been increased from 32K to 50K.
1183     2. The code in pcre_study() was using the value of the re argument before
1184     testing it for NULL. (Of course, in any sensible call of the function, it
1185     won't be NULL.)
1187     3. The memmove() emulation function in pcre_internal.h, which is used on
1188     systems that lack both memmove() and bcopy() - that is, hardly ever -
1189     was missing a "static" storage class specifier.
1191     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1192     containing an extended class (one that cannot be represented by a bitmap
1193     because it contains high-valued characters or Unicode property items, e.g.
1194     [\pZ]). Almost always one would set UTF-8 mode when processing such a
1195     pattern, but PCRE should not loop if you do not (it no longer does).
1196     [Detail: two cases were found: (a) a repeated subpattern containing an
1197     extended class; (b) a recursive reference to a subpattern that followed a
1198     previous extended class. It wasn't skipping over the extended class
1199     correctly when UTF-8 mode was not set.]
1201     5. A negated single-character class was not being recognized as fixed-length
1202     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1203     compile error "lookbehind assertion is not fixed length".
1205     6. The RunPerlTest auxiliary script was showing an unexpected difference
1206     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1207     write a Perl script that can interpret lines of an input file either as
1208     byte characters or as UTF-8, which is what "perltest" was being required to
1209     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1210     can't do is switch easily at run time between having the "use utf8;" pragma
1211     or not. In the end, I fudged it by using the RunPerlTest script to insert
1212     "use utf8;" explicitly for the UTF-8 tests.
1214     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1215     the end of the subject string, contrary to the documentation and to what
1216     Perl does. This was true of both matching functions. Now it matches only at
1217     the start of the subject and immediately after *internal* newlines.
1219     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1220     a pointer to an int instead of a pointer to an unsigned long int. This
1221     caused problems on 64-bit systems.
1223     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1224     instance of the 'standard' template library not being so standard".
1226     10. There was no check on the number of named subpatterns nor the maximum
1227     length of a subpattern name. The product of these values is used to compute
1228     the size of the memory block for a compiled pattern. By supplying a very
1229     long subpattern name and a large number of named subpatterns, the size
1230     computation could be caused to overflow. This is now prevented by limiting
1231     the length of names to 32 characters, and the number of named subpatterns
1232     to 10,000.
1234     11. Subpatterns that are repeated with specific counts have to be replicated in
1235     the compiled pattern. The size of memory for this was computed from the
1236     length of the subpattern and the repeat count. The latter is limited to
1237     65535, but there was no limit on the former, meaning that integer overflow
1238     could in principle occur. The compiled length of a repeated subpattern is
1239     now limited to 30,000 bytes in order to prevent this.
1241     12. Added the optional facility to have named substrings with the same name.
1243     13. Added the ability to use a named substring as a condition, using the
1244     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1245     are numbers (not recommended). Forward references are permitted.
1247     14. Added forward references in named backreferences (if you see what I mean).
1249     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1250     pattern could run off the end of the subject. For example, the pattern
1251     "(?s)(.{1,5})"8 did this with the subject "ab".
1253     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1254     PCRE_CASELESS was set when matching characters that were quantified with ?
1255     or *.
1257     17. A character class other than a single negated character that had a minimum
1258     but no maximum quantifier - for example [ab]{6,} - was not handled
1259     correctly by pce_dfa_exec(). It would match only one character.
1261     18. A valid (though odd) pattern that looked like a POSIX character
1262     class but used an invalid character after [ (for example [[,abc,]]) caused
1263     pcre_compile() to give the error "Failed: internal error: code overflow" or
1264     in some cases to crash with a glibc free() error. This could even happen if
1265     the pattern terminated after [[ but there just happened to be a sequence of
1266     letters, a binary zero, and a closing ] in the memory that followed.
1268     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1269     over the years. Originally (before any Unicode support), just the bottom 8
1270     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1271     output from "man perlunicode" includes this:
1273     The regular expression compiler produces polymorphic opcodes. That
1274     is, the pattern adapts to the data and automatically switches to
1275     the Unicode character scheme when presented with Unicode data--or
1276     instead uses a traditional byte scheme when presented with byte
1277     data.
1279     Sadly, a wide octal escape does not cause a switch, and in a string with
1280     no other multibyte characters, these octal escapes are treated as before.
1281     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1282     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1283     Unicode string.
1285     I have not perpetrated such confusion in PCRE. Up till now, it took just
1286     the bottom 8 bits, as in old Perl. I have now made octal escapes with
1287     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1288     translate to the appropriate multibyte character.
1290     29. Applied some refactoring to reduce the number of warnings from Microsoft
1291     and Borland compilers. This has included removing the fudge introduced
1292     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1293     a warning about an unused variable.
1295     21. PCRE has not included VT (character 0x0b) in the set of whitespace
1296     characters since release 4.0, because Perl (from release 5.004) does not.
1297     [Or at least, is documented not to: some releases seem to be in conflict
1298     with the documentation.] However, when a pattern was studied with
1299     pcre_study() and all its branches started with \s, PCRE still included VT
1300     as a possible starting character. Of course, this did no harm; it just
1301     caused an unnecessary match attempt.
1303     22. Removed a now-redundant internal flag bit that recorded the fact that case
1304     dependency changed within the pattern. This was once needed for "required
1305     byte" processing, but is no longer used. This recovers a now-scarce options
1306     bit. Also moved the least significant internal flag bit to the most-
1307     significant bit of the word, which was not previously used (hangover from
1308     the days when it was an int rather than a uint) to free up another bit for
1309     the future.
1311     23. Added support for CRLF line endings as well as CR and LF. As well as the
1312     default being selectable at build time, it can now be changed at runtime
1313     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1314     specify that it is scanning data with non-default line endings.
1316     24. Changed the definition of CXXLINK to make it agree with the definition of
1317     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1319     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1320     recursions. This makes a big different to stack usage for some patterns.
1322     26. If a subpattern containing a named recursion or subroutine reference such
1323     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1324     the space required for the compiled pattern went wrong and gave too small a
1325     value. Depending on the environment, this could lead to "Failed: internal
1326     error: code overflow at offset 49" or "glibc detected double free or
1327     corruption" errors.
1329     27. Applied patches from Google (a) to support the new newline modes and (b) to
1330     advance over multibyte UTF-8 characters in GlobalReplace.
1332     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1333     difference for some implementation of PCRE in some Windows version.
1335     29. Added some extra testing facilities to pcretest:
1337     \q<number> in a data line sets the "match limit" value
1338     \Q<number> in a data line sets the "match recursion limt" value
1339     -S <number> sets the stack size, where <number> is in megabytes
1341     The -S option isn't available for Windows.
1344 nigel 89 Version 6.6 06-Feb-06
1345     ---------------------
1347     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1348     in pcreposix.h. I have copied the definition from pcre.h.
1350     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1351     because pcre.h is no longer a built file.
1353     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1354     not normally included in the compiled code.
1357 nigel 87 Version 6.5 01-Feb-06
1358     ---------------------
1360     1. When using the partial match feature with pcre_dfa_exec(), it was not
1361     anchoring the second and subsequent partial matches at the new starting
1362     point. This could lead to incorrect results. For example, with the pattern
1363     /1234/, partially matching against "123" and then "a4" gave a match.
1365     2. Changes to pcregrep:
1367     (a) All non-match returns from pcre_exec() were being treated as failures
1368     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1369     error message is output. Some extra information is given for the
1371     probably the only errors that are likely to be caused by users (by
1372     specifying a regex that has nested indefinite repeats, for instance).
1373     If there are more than 20 of these errors, pcregrep is abandoned.
1375     (b) A binary zero was treated as data while matching, but terminated the
1376     output line if it was written out. This has been fixed: binary zeroes
1377     are now no different to any other data bytes.
1379     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1380     used to set a locale for matching. The --locale=xxxx long option has
1381     been added (no short equivalent) to specify a locale explicitly on the
1382     pcregrep command, overriding the environment variables.
1384     (d) When -B was used with -n, some line numbers in the output were one less
1385     than they should have been.
1387     (e) Added the -o (--only-matching) option.
1389     (f) If -A or -C was used with -c (count only), some lines of context were
1390     accidentally printed for the final match.
1392     (g) Added the -H (--with-filename) option.
1394     (h) The combination of options -rh failed to suppress file names for files
1395     that were found from directory arguments.
1397     (i) Added the -D (--devices) and -d (--directories) options.
1399     (j) Added the -F (--fixed-strings) option.
1401     (k) Allow "-" to be used as a file name for -f as well as for a data file.
1403     (l) Added the --colo(u)r option.
1405     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1406     is not present by default.
1408     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1409     items such as (?R) or (?1), when the recursion could match a number of
1410     alternatives. If it matched one of the alternatives, but subsequently,
1411     outside the recursion, there was a failure, the code tried to back up into
1412     the recursion. However, because of the way PCRE is implemented, this is not
1413     possible, and the result was an incorrect result from the match.
1415     In order to prevent this happening, the specification of recursion has
1416     been changed so that all such subpatterns are automatically treated as
1417     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1419     4. I had overlooked the fact that, in some locales, there are characters for
1420     which isalpha() is true but neither isupper() nor islower() are true. In
1421     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1422     and ordfeminine) are like this. This affected the treatment of \w and \W
1423     when they appeared in character classes, but not when they appeared outside
1424     a character class. The bit map for "word" characters is now created
1425     separately from the results of isalnum() instead of just taking it from the
1426     upper, lower, and digit maps. (Plus the underscore character, of course.)
1428     5. The above bug also affected the handling of POSIX character classes such as
1429     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1430     permanent tables. Instead, the bit maps for such a class were previously
1431     created as the appropriate unions of the upper, lower, and digit bitmaps.
1432     Now they are created by subtraction from the [[:word:]] class, which has
1433     its own bitmap.
1435     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1436     It is created by subtracting the vertical space characters (\x09, \x0a,
1437     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1438     subtraction was done in the overall bitmap for a character class, meaning
1439     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1440     be recognized. This bug has been fixed.
1442     7. Patches from the folks at Google:
1444     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1445     real life, but is still worth protecting against".
1447     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1448     regular expressions".
1450     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1451     have it.
1453     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1454     "configure" and the latter not, in order to fix a problem somebody had
1455     with compiling the Arg class on HP-UX.
1457     (e) Improve the error-handling of the C++ wrapper a little bit.
1459     (f) New tests for checking recursion limiting.
1461     8. The pcre_memmove() function, which is used only if the environment does not
1462     have a standard memmove() function (and is therefore rarely compiled),
1463     contained two bugs: (a) use of int instead of size_t, and (b) it was not
1464     returning a result (though PCRE never actually uses the result).
1466     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1467     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1468     returned instead of calling malloc() with an overflowing number that would
1469     most likely cause subsequent chaos.
1471     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1473     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1474     with this option is matched, the nmatch and pmatch options of regexec() are
1475     ignored.
1477     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1478     provided in case anyone wants to the the POSIX interface with UTF-8
1479     strings.
1481     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1482     C++ linking (needed for some HP-UX environments).
1484     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1485     (unused parameter) and in the pcre_printint() function (omitted "default"
1486     switch label when the default is to do nothing).
1488     15. Added some code to make it possible, when PCRE is compiled as a C++
1489     library, to replace subject pointers for pcre_exec() with a smart pointer
1490     class, thus making it possible to process discontinuous strings.
1492     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1493     much the same function. They were added by different people who were trying
1494     to make PCRE easy to compile on non-Unix systems. It has been suggested
1495     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1496     for compiling on Windows systems. I have therefore replaced it with
1497     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1498     defaults to "extern" for C or "extern C" for C++, which works fine on
1499     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1500     SCOPE with something explicit in config.h. In addition:
1502     (a) pcreposix.h still had just "extern" instead of either of these macros;
1503     I have replaced it with PCRE_DATA_SCOPE.
1505     (b) Functions such as _pcre_xclass(), which are internal to the library,
1506     but external in the C sense, all had PCRE_EXPORT in their definitions.
1507     This is apparently wrong for the Windows case, so I have removed it.
1508     (It makes no difference on Unix-like systems.)
1510     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1511     of recursive calls to match(). This is different to MATCH_LIMIT because
1512     that limits the total number of calls to match(), not all of which increase
1513     the depth of recursion. Limiting the recursion depth limits the amount of
1514     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1515     when PCRE is compiled, and changed at run time. A patch from Google adds
1516     this functionality to the C++ interface.
1518     18. Changes to the handling of Unicode character properties:
1520     (a) Updated the table to Unicode 4.1.0.
1522     (b) Recognize characters that are not in the table as "Cn" (undefined).
1524     (c) I revised the way the table is implemented to a much improved format
1525     which includes recognition of ranges. It now supports the ranges that
1526     are defined in UnicodeData.txt, and it also amalgamates other
1527     characters into ranges. This has reduced the number of entries in the
1528     table from around 16,000 to around 3,000, thus reducing its size
1529     considerably. I realized I did not need to use a tree structure after
1530     all - a binary chop search is just as efficient. Having reduced the
1531     number of entries, I extended their size from 6 bytes to 8 bytes to
1532     allow for more data.
1534     (d) Added support for Unicode script names via properties such as \p{Han}.
1536     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1537     matching that character.
1539     20. When matching a repeated Unicode property with a minimum greater than zero,
1540     (for example \pL{2,}), PCRE could look past the end of the subject if it
1541     reached it while seeking the minimum number of characters. This could
1542     happen only if some of the characters were more than one byte long, because
1543     there is a check for at least the minimum number of bytes.
1545     21. Refactored the implementation of \p and \P so as to be more general, to
1546     allow for more different types of property in future. This has changed the
1547     compiled form incompatibly. Anybody with saved compiled patterns that use
1548     \p or \P will have to recompile them.
1550     22. Added "Any" and "L&" to the supported property types.
1552     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1553     but give a compile time error if the value is greater than 0xff.
1555     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1556     accidentally not being installed or uninstalled.
1558     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1559     made were to insert the current release number. This seemed silly, because
1560     it made things harder for people building PCRE on systems that don't run
1561     "configure". I have turned pcre.h into a distributed file, no longer built
1562     by "configure", with the version identification directly included. There is
1563     no longer a pcre.h.in file.
1565     However, this change necessitated a change to the pcre-config script as
1566     well. It is built from pcre-config.in, and one of the substitutions was the
1567     release number. I have updated configure.ac so that ./configure now finds
1568     the release number by grepping pcre.h.
1570     26. Added the ability to run the tests under valgrind.
1573 nigel 85 Version 6.4 05-Sep-05
1574     ---------------------
1576     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1577     "--" to be printed when multiple files were scanned, even when none of the
1578     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1579     consider it to be a bug, and have restored the previous behaviour.
1581     2. A couple of code tidies to get rid of compiler warnings.
1583     3. The pcretest program used to cheat by referring to symbols in the library
1584     whose names begin with _pcre_. These are internal symbols that are not
1585     really supposed to be visible externally, and in some environments it is
1586     possible to suppress them. The cheating is now confined to including
1587     certain files from the library's source, which is a bit cleaner.
1589     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1590     file's purpose clearer.
1592     5. Reorganized pcre_ucp_findchar().
1595 nigel 83 Version 6.3 15-Aug-05
1596     ---------------------
1598     1. The file libpcre.pc.in did not have general read permission in the tarball.
1600     2. There were some problems when building without C++ support:
1602     (a) If C++ support was not built, "make install" and "make test" still
1603     tried to test it.
1605     (b) There were problems when the value of CXX was explicitly set. Some
1606     changes have been made to try to fix these, and ...
1608     (c) --disable-cpp can now be used to explicitly disable C++ support.
1610     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1611     backslash in a target when C++ was disabled. This confuses some
1612     versions of "make", apparently. Using an intermediate variable solves
1613     this. (Same for CPP_LOBJ.)
1615     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1616     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1617     necessary on certain architectures.
1619     4. Added a setting of -export-symbols-regex to the link command to remove
1620     those symbols that are exported in the C sense, but actually are local
1621     within the library, and not documented. Their names all begin with
1622     "_pcre_". This is not a perfect job, because (a) we have to except some
1623     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1624     available (and never for static libraries). I have made a note to try to
1625     find a way round (a) in the future.
1628 nigel 81 Version 6.2 01-Aug-05
1629     ---------------------
1631     1. There was no test for integer overflow of quantifier values. A construction
1632     such as {1111111111111111} would give undefined results. What is worse, if
1633     a minimum quantifier for a parenthesized subpattern overflowed and became
1634     negative, the calculation of the memory size went wrong. This could have
1635     led to memory overwriting.
1637     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1639     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1640     operating environments where this matters.
1642     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1643     PCRE options from within the C++ wrapper.
1645     5. Named capturing subpatterns were not being correctly counted when a pattern
1646     was compiled. This caused two problems: (a) If there were more than 100
1647     such subpatterns, the calculation of the memory needed for the whole
1648     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1649     back references of the form \12, where the number was greater than 9, were
1650     not recognized as back references, even though there were sufficient
1651     previous subpatterns.
1653     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1654     versions of gcc, e.g. 2.95.4.
1657 nigel 79 Version 6.1 21-Jun-05
1658     ---------------------
1660     1. There was one reference to the variable "posix" in pcretest.c that was not
1661     surrounded by "#if !defined NOPOSIX".
1663     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1664     the cross-check on the old pcre_info() function, for the benefit of the
1665     cut-down version of PCRE that is currently imported into Exim.
1667     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1668     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1669     patterns that start (?i) but I don't think that matters. The use of (?i) is
1670     just an example; this all applies to the other options as well.
1672     4. Since libtool seems to echo the compile commands it is issuing, the output
1673     from "make" can be reduced a bit by putting "@" in front of each libtool
1674     compile command.
1676     5. Patch from the folks at Google for configure.in to be a bit more thorough
1677     in checking for a suitable C++ installation before trying to compile the
1678     C++ stuff. This should fix a reported problem when a compiler was present,
1679     but no suitable headers.
1681     6. The man pages all had just "PCRE" as their title. I have changed them to
1682     be the relevant file name. I have also arranged that these names are
1683     retained in the file doc/pcre.txt, which is a concatenation in text format
1684     of all the man pages except the little individual ones for each function.
1686     7. The NON-UNIX-USE file had not been updated for the different set of source
1687     files that come with release 6. I also added a few comments about the C++
1688     wrapper.
1691 nigel 77 Version 6.0 07-Jun-05
1692     ---------------------
1694     1. Some minor internal re-organization to help with my DFA experiments.
1696     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1697     didn't matter for the library itself when fully configured, but did matter
1698     when compiling without UCP support, or within Exim, where the ucp files are
1699     not imported.
1701     3. Refactoring of the library code to split up the various functions into
1702     different source modules. The addition of the new DFA matching code (see
1703     below) to a single monolithic source would have made it really too
1704     unwieldy, quite apart from causing all the code to be include in a
1705     statically linked application, when only some functions are used. This is
1706     relevant even without the DFA addition now that patterns can be compiled in
1707     one application and matched in another.
1709     The downside of splitting up is that there have to be some external
1710     functions and data tables that are used internally in different modules of
1711     the library but which are not part of the API. These have all had their
1712     names changed to start with "_pcre_" so that they are unlikely to clash
1713     with other external names.
1715     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1716     a different (DFA) algorithm. Although it is slower than the original
1717     function, it does have some advantages for certain types of matching
1718     problem.
1720     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1721     including restarting after a partial match.
1723     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1724     defined when compiling for Windows was sent to me. I have put it into the
1725     code, though I have no means of testing or verifying it.
1727     7. Added the pcre_refcount() auxiliary function.
1729     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1730     match before or at the first newline in the subject string. In pcretest,
1731     the /f option on a pattern can be used to set this.
1733     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1734     would behave wrongly. This has been present in PCRE since release 4.0.
1736     10. A number of changes to the pcregrep command:
1738     (a) Refactored how -x works; insert ^(...)$ instead of setting
1739     PCRE_ANCHORED and checking the length, in preparation for adding
1740     something similar for -w.
1742     (b) Added the -w (match as a word) option.
1744     (c) Refactored the way lines are read and buffered so as to have more
1745     than one at a time available.
1747     (d) Implemented a pcregrep test script.
1749     (e) Added the -M (multiline match) option. This allows patterns to match
1750     over several lines of the subject. The buffering ensures that at least
1751     8K, or the rest of the document (whichever is the shorter) is available
1752     for matching (and similarly the previous 8K for lookbehind assertions).
1754     (f) Changed the --help output so that it now says
1756     -w, --word-regex(p)
1758     instead of two lines, one with "regex" and the other with "regexp"
1759     because that confused at least one person since the short forms are the
1760     same. (This required a bit of code, as the output is generated
1761     automatically from a table. It wasn't just a text change.)
1763     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1764     option but starts with a hyphen. Could be a pattern or a path name
1765     starting with a hyphen, for instance.
1767     (h) "-" can be given as a file name to represent stdin.
1769     (i) When file names are being printed, "(standard input)" is used for
1770     the standard input, for compatibility with GNU grep. Previously
1771     "<stdin>" was used.
1773     (j) The option --label=xxx can be used to supply a name to be used for
1774     stdin when file names are being printed. There is no short form.
1776     (k) Re-factored the options decoding logic because we are going to add
1777     two more options that take data. Such options can now be given in four
1778     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1780     (l) Added the -A, -B, and -C options for requesting that lines of context
1781     around matches be printed.
1783     (m) Added the -L option to print the names of files that do not contain
1784     any matching lines, that is, the complement of -l.
1786     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1787     continue to scan other files.
1789     (o) The -s option was incorrectly implemented. For compatibility with other
1790     greps, it now suppresses the error message for a non-existent or non-
1791     accessible file (but not the return code). There is a new option called
1792     -q that suppresses the output of matching lines, which was what -s was
1793     previously doing.
1795     (p) Added --include and --exclude options to specify files for inclusion
1796     and exclusion when recursing.
1798     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1799     Hopefully, it now does.
1801     12. Missing cast in pcre_study().
1803     13. Added an "uninstall" target to the makefile.
1805     14. Replaced "extern" in the function prototypes in Makefile.in with
1806     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1807     world, but is set differently for Windows.
1809     15. Added a second compiling function called pcre_compile2(). The only
1810     difference is that it has an extra argument, which is a pointer to an
1811     integer error code. When there is a compile-time failure, this is set
1812     non-zero, in addition to the error test pointer being set to point to an
1813     error message. The new argument may be NULL if no error number is required
1814     (but then you may as well call pcre_compile(), which is now just a
1815     wrapper). This facility is provided because some applications need a
1816     numeric error indication, but it has also enabled me to tidy up the way
1817     compile-time errors are handled in the POSIX wrapper.
1819     16. Added VPATH=.libs to the makefile; this should help when building with one
1820     prefix path and installing with another. (Or so I'm told by someone who
1821     knows more about this stuff than I do.)
1823     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1824     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1825     match everything, including newlines. This is not POSIX-compatible, but
1826     somebody wanted the feature. From pcretest it can be activated by using
1827     both the P and the s flags.
1829     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1831     19. libpcre.pc was being incorrectly installed as executable.
1833     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1834     it now also looks for '\r' so that it will work unmodified on Windows.
1836     21. Added Google's contributed C++ wrapper to the distribution.
1838     22. Added some untidy missing memory free() calls in pcretest, to keep
1839     Electric Fence happy when testing.
1843 nigel 75 Version 5.0 13-Sep-04
1844     ---------------------
1846     1. Internal change: literal characters are no longer packed up into items
1847     containing multiple characters in a single byte-string. Each character
1848     is now matched using a separate opcode. However, there may be more than one
1849     byte in the character in UTF-8 mode.
1851     2. The pcre_callout_block structure has two new fields: pattern_position and
1852     next_item_length. These contain the offset in the pattern to the next match
1853     item, and its length, respectively.
1855     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1856     insertion of callouts before each pattern item. Added the /C option to
1857     pcretest to make use of this.
1859     4. On the advice of a Windows user, the lines
1861     #if defined(_WIN32) || defined(WIN32)
1862     _setmode( _fileno( stdout ), 0x8000 );
1863     #endif /* defined(_WIN32) || defined(WIN32) */
1865     have been added to the source of pcretest. This apparently does useful
1866     magic in relation to line terminators.
1868     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1869     for the benefit of those environments where the "b" makes a difference.
1871     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1872     to know about it. I have put a hack into configure.in that adds in code
1873     to set GCC=yes if CC=icc. This seems to end up at a point in the
1874     generated configure script that is early enough to affect the setting of
1875     compiler options, which is what is needed, but I have no means of testing
1876     whether it really works. (The user who reported this had patched the
1877     generated configure script, which of course I cannot do.)
1879     LATER: After change 22 below (new libtool files), the configure script
1880     seems to know about icc (and also ecc). Therefore, I have commented out
1881     this hack in configure.in.
1883     7. Added support for pkg-config (2 patches were sent in).
1885     8. Negated POSIX character classes that used a combination of internal tables
1886     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1887     [[:^ascii]]. Typically, they would match almost any characters. The other
1888     POSIX classes were not broken in this way.
1890     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1891     to find the match, as PCRE was deluded into thinking that the match had to
1892     start at the start point or following a newline. The same bug applied to
1893     patterns with negative forward assertions or any backward assertions
1894     preceding ".*" at the start, unless the pattern required a fixed first
1895     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1897     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1898     starting at the last subject character, bytes beyond the end of the subject
1899     string were read.
1901     11. Renamed the variable "class" as "classbits" to make life easier for C++
1902     users. (Previously there was a macro definition, but it apparently wasn't
1903     enough.)
1905     12. Added the new field "tables" to the extra data so that tables can be passed
1906     in at exec time, or the internal tables can be re-selected. This allows
1907     a compiled regex to be saved and re-used at a later time by a different
1908     program that might have everything at different addresses.
1910     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1911     -R library as well as a -L library.
1913     14. The debugging options of pcretest (-d on the command line or D on a
1914     pattern) showed incorrect output for anything following an extended class
1915     that contained multibyte characters and which was followed by a quantifier.
1917     15. Added optional support for general category Unicode character properties
1918     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1919     support. It adds about 90K to the size of the library. The meanings of the
1920     inbuilt class escapes such as \d and \s have NOT been changed.
1922     16. Updated pcredemo.c to include calls to free() to release the memory for the
1923     compiled pattern.
1925     17. The generated file chartables.c was being created in the source directory
1926     instead of in the building directory. This caused the build to fail if the
1927     source directory was different from the building directory, and was
1928     read-only.
1930     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1931     file. No doubt somebody will tell me if they don't make sense... Also added
1932     Dan Mooney's comments about building on OpenVMS.
1934     19. Added support for partial matching via the PCRE_PARTIAL option for
1935     pcre_exec() and the \P data escape in pcretest.
1937     20. Extended pcretest with 3 new pattern features:
1939     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1940     write the compiled pattern to the file whose name is "rest-of-line".
1941     This is a straight binary dump of the data, with the saved pointer to
1942     the character tables forced to be NULL. The study data, if any, is
1943     written too. After writing, pcretest reads a new pattern.
1945     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1946     compiled pattern from the given file. There must not be any
1947     occurrences of "<" in the file name (pretty unlikely); if there are,
1948     pcretest will instead treat the initial "<" as a pattern delimiter.
1949     After reading in the pattern, pcretest goes on to read data lines as
1950     usual.
1952     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1953     and 16-bit fields in a compiled pattern, to simulate a pattern that
1954     was compiled on a host of opposite endianness.
1956     21. The pcre-exec() function can now cope with patterns that were compiled on
1957     hosts of opposite endianness, with this restriction:
1959     As for any compiled expression that is saved and used later, the tables
1960     pointer field cannot be preserved; the extra_data field in the arguments
1961     to pcre_exec() should be used to pass in a tables address if a value
1962     other than the default internal tables were used at compile time.
1964     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1965     now diagnosed as an error. Previously, most of the time, a negative number
1966     would have been treated as zero, but if in addition "ovector" was passed as
1967     NULL, a crash could occur.
1969     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1970     new versions from the libtool 1.5 distribution (the last one is a copy of
1971     a file called libtool.m4). This seems to have fixed the need to patch
1972     "configure" to support Darwin 1.3 (which I used to do). However, I still
1973     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1974     workstation).
1976     24. Changed the PCRE licence to be the more standard "BSD" licence.
1979 nigel 73 Version 4.5 01-Dec-03
1980     ---------------------
1982     1. There has been some re-arrangement of the code for the match() function so
1983     that it can be compiled in a version that does not call itself recursively.
1984     Instead, it keeps those local variables that need separate instances for
1985     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1986     needs to "recurse". Keeping track of where control must go is done by means
1987     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1988     hide most of the details from the main code, and operates only if
1989     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1990     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1991     operating.
1993     To make it easier for callers to provide specially tailored get/free
1994     functions for this usage, two new functions, pcre_stack_malloc, and
1995     pcre_stack_free, are used. They are always called in strict stacking order,
1996     and the size of block requested is always the same.
1998     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1999     PCRE has been compiled to use the stack or the heap for recursion. The
2000     -C option of pcretest uses this to show which version is compiled.
2002     A new data escape \S, is added to pcretest; it causes the amounts of store
2003     obtained and freed by both kinds of malloc/free at match time to be added
2004     to the output.
2006     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2007     what's available on my current Linux desktop machine.
2009     3. When matching a UTF-8 string, the test for a valid string at the start has
2010     been extended. If start_offset is not zero, PCRE now checks that it points
2011     to a byte that is the start of a UTF-8 character. If not, it returns
2012     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2013     this is necessary because there may be backward assertions in the pattern.
2014     When matching the same subject several times, it may save resources to use
2015     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2017     4. The code for checking the validity of UTF-8 strings has been tightened so
2018     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2019     containing "overlong sequences".
2021     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2022     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2023     should have been "&", but it just so happened that all the cases this let
2024     through by mistake were picked up later in the function.
2026     6. I had used a variable called "isblank" - this is a C99 function, causing
2027     some compilers to warn. To avoid this, I renamed it (as "blankclass").
2029     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2030     prompting; (b) run "./pcretest /dev/null" at the start of the test script
2031     so the version is shown; (c) stop "make test" echoing "./RunTest".
2033     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2035     9. The prototype for memmove() for systems that don't have it was using
2036     size_t, but the inclusion of the header that defines size_t was later. I've
2037     moved the #includes for the C headers earlier to avoid this.
2039     10. Added some adjustments to the code to make it easier to compiler on certain
2040     special systems:
2042     (a) Some "const" qualifiers were missing.
2043     (b) Added the macro EXPORT before all exported functions; by default this
2044     is defined to be empty.
2045     (c) Changed the dftables auxiliary program (that builds chartables.c) so
2046     that it reads its output file name as an argument instead of writing
2047     to the standard output and assuming this can be redirected.
2049     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2050     class containing characters with values greater than 255, PCRE compilation
2051     went into a loop.
2053     12. A recursive reference to a subpattern that was within another subpattern
2054     that had a minimum quantifier of zero caused PCRE to crash. For example,
2055     (x(y(?2))z)? provoked this bug with a subject that got as far as the
2056     recursion. If the recursively-called subpattern itself had a zero repeat,
2057     that was OK.
2059     13. In pcretest, the buffer for reading a data line was set at 30K, but the
2060     buffer into which it was copied (for escape processing) was still set at
2061     1024, so long lines caused crashes.
2063     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2064     "internal error: code overflow...". This applied to any character class
2065     that was followed by a possessive quantifier.
2067     15. Modified the Makefile to add libpcre.la as a prerequisite for
2068     libpcreposix.la because I was told this is needed for a parallel build to
2069     work.
2071     16. If a pattern that contained .* following optional items at the start was
2072     studied, the wrong optimizing data was generated, leading to matching
2073     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2074     matching string must start with a or b or c. The correct conclusion for
2075     this pattern is that a match can start with any character.
2078 nigel 71 Version 4.4 13-Aug-03
2079     ---------------------
2081     1. In UTF-8 mode, a character class containing characters with values between
2082     127 and 255 was not handled correctly if the compiled pattern was studied.
2083     In fixing this, I have also improved the studying algorithm for such
2084     classes (slightly).
2086     2. Three internal functions had redundant arguments passed to them. Removal
2087     might give a very teeny performance improvement.
2089     3. Documentation bug: the value of the capture_top field in a callout is *one
2090     more than* the number of the hightest numbered captured substring.
2092     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2093     in incorrectly linking with a previously installed version. They now link
2094     explicitly with libpcre.la.
2096     5. configure.in no longer needs to recognize Cygwin specially.
2098     6. A problem in pcre.in for Windows platforms is fixed.
2100     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2101     pcretest, it used to include the size of the study block as part of its
2102     output. Unfortunately, the structure contains a field that has a different
2103     size on different hardware architectures. This meant that the tests that
2104     showed this size failed. As the block is currently always of a fixed size,
2105     this information isn't actually particularly useful in pcretest output, so
2106     I have just removed it.
2108     8. Three pre-processor statements accidentally did not start in column 1.
2109     Sadly, there are *still* compilers around that complain, even though
2110     standard C has not required this for well over a decade. Sigh.
2112     9. In pcretest, the code for checking callouts passed small integers in the
2113     callout_data field, which is a void * field. However, some picky compilers
2114     complained about the casts involved for this on 64-bit systems. Now
2115     pcretest passes the address of the small integer instead, which should get
2116     rid of the warnings.
2118     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2119     both compile and run time, and gives an error if an invalid UTF-8 sequence
2120     is found. There is a option for disabling this check in cases where the
2121     string is known to be correct and/or the maximum performance is wanted.
2123     11. In response to a bug report, I changed one line in Makefile.in from
2125     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2126     to
2127     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2129     to look similar to other lines, but I have no way of telling whether this
2130     is the right thing to do, as I do not use Windows. No doubt I'll get told
2131     if it's wrong...
2134 nigel 69 Version 4.3 21-May-03
2135     ---------------------
2137     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2138     Makefile.
2140     2. Some refactoring to improve the quality of the code:
2142     (i) The utf8_table... variables are now declared "const".
2144     (ii) The code for \cx, which used the "case flipping" table to upper case
2145     lower case letters, now just substracts 32. This is ASCII-specific,
2146     but the whole concept of \cx is ASCII-specific, so it seems
2147     reasonable.
2149     (iii) PCRE was using its character types table to recognize decimal and
2150     hexadecimal digits in the pattern. This is silly, because it handles
2151     only 0-9, a-f, and A-F, but the character types table is locale-
2152     specific, which means strange things might happen. A private
2153     table is now used for this - though it costs 256 bytes, a table is
2154     much faster than multiple explicit tests. Of course, the standard
2155     character types table is still used for matching digits in subject
2156     strings against \d.
2158     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2159     ending in _t are). So I've renamed it as ESC_tee.
2161     3. The first argument for regexec() in the POSIX wrapper should have been
2162     defined as "const".
2164     4. Changed pcretest to use malloc() for its buffers so that they can be
2165     Electric Fenced for debugging.
2167     5. There were several places in the code where, in UTF-8 mode, PCRE would try
2168     to read one or more bytes before the start of the subject string. Often this
2169     had no effect on PCRE's behaviour, but in some circumstances it could
2170     provoke a segmentation fault.
2172     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2173     to try to read one or more bytes before the start of the subject string.
2175     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2176     UTF-8 support could misbehave in various ways if the subject string
2177     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2178     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2179     back over UTF-8 characters.)
2182 nigel 67 Version 4.2 14-Apr-03
2183     ---------------------
2185     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2187     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2188     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2189     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2190     [WIN_PREFIX]: new variable, "cyg" for Cygwin
2191     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2192     and BUILD_EXEEXT
2193     Note: automatic setting of the BUILD variables is not yet working
2194     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2195     compile-time but not at link-time
2196     [LINK]: use for linking executables only
2197     make different versions for Windows and non-Windows
2198     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2199     libraries
2200     [LINK_FOR_BUILD]: new variable
2201     [OBJEXT]: use throughout
2202     [EXEEXT]: use throughout
2203     <winshared>: new target
2204     <wininstall>: new target
2205     <dftables.o>: use native compiler
2206     <dftables>: use native linker
2207     <install>: handle Windows platform correctly
2208     <clean>: ditto
2209     <check>: ditto
2210     copy DLL to top builddir before testing
2212     As part of these changes, -no-undefined was removed again. This was reported
2213     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2214     in any case.
2216     3. Some tidies to get rid of compiler warnings:
2218     . In the match_data structure, match_limit was an unsigned long int, whereas
2219     match_call_count was an int. I've made them both unsigned long ints.
2221     . In pcretest the fact that a const uschar * doesn't automatically cast to
2222     a void * provoked a warning.
2224     . Turning on some more compiler warnings threw up some "shadow" variables
2225     and a few more missing casts.
2227     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2228     option, a class that contained a single character with a value between 128
2229     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2231     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2232     option, a class that contained several characters, but with at least one
2233     whose value was between 128 and 255 caused PCRE to crash.
2236 nigel 65 Version 4.1 12-Mar-03
2237     ---------------------
2238 nigel 63
2239 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
2240     needed, and a string in dftables.c that was longer than standard compilers are
2241     required to support.
2243     2. Compiling with Sun's compiler found a few more places where the code could
2244     be tidied up in order to avoid warnings.
2246     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2247     first of these names is deprecated in the latest Autoconf in favour of the name
2248     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2249     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2250     analogy I have changed it to CFLAGS_FOR_BUILD.
2252     4. Added -no-undefined to the linking command in the Makefile, because this is
2253     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2254     linking step for the pcreposix library.
2256     5. PCRE was failing to diagnose the case of two named groups with the same
2257     name.
2259     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2260     literal character that is needed in the subject for a match, and scans along to
2261     ensure that it is present before embarking on the full matching process. This
2262     saves time in cases of nested unlimited repeats that are never going to match.
2263     Problem: the scan can take a lot of time if the subject is very long (e.g.
2264     megabytes), thus penalizing straightforward matches. It is now done only if the
2265     amount of subject to be scanned is less than 1000 bytes.
2267     7. A lesser problem with the same optimization is that it was recording the
2268     first character of an anchored pattern as "needed", thus provoking a search
2269     right along the subject, even when the first match of the pattern was going to
2270     fail. The "needed" character is now not set for anchored patterns, unless it
2271     follows something in the pattern that is of non-fixed length. Thus, it still
2272     fulfils its original purpose of finding quick non-matches in cases of nested
2273     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2276     Version 4.0 17-Feb-03
2277     ---------------------
2279 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
2280     extended to the end of string, PCRE compiled incorrect data. This could lead to
2281     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2283     2. Moved to autoconf 2.53 and libtool 1.4.2.
2285     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2286     the special perltest8 script is no longer needed - all the tests can be run
2287     from a single perltest script.
2289     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2290     by \s. It has now been removed in PCRE. This means it isn't recognized as
2291     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2292     class [:space:] *does* include VT, thereby creating a mess.
2294     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2295     space and tab.
2297     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2298     its new features into the main test script, reducing the number of scripts.
2300     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2301     were backward compatible, and made the (?i) apply to the whole pattern, as if
2302     /i were given. Now it behaves more logically, and applies the option setting
2303     only to what follows. PCRE has been changed to follow suit. However, if it
2304     finds options settings right at the start of the pattern, it extracts them into
2305     the global options, as before. Thus, they show up in the info data.
2307     8. Added support for the \Q...\E escape sequence. Characters in between are
2308     treated as literals. This is slightly different from Perl in that $ and @ are
2309     also handled as literals inside the quotes. In Perl, they will cause variable
2310     interpolation. Note the following examples:
2312     Pattern PCRE matches Perl matches
2314     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2315     \Qabc\$xyz\E abc\$xyz abc\$xyz
2316     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2318     For compatibility with Perl, \Q...\E sequences are recognized inside character
2319     classes as well as outside them.
2321     9. Re-organized 3 code statements in pcretest to avoid "overflow in
2322     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2323     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2324     signed/unsigned warnings.
2326     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2327     option for pcretest, so I've replaced it by a simple function that does just
2328     that job.
2330     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2331     "pcregrep -".
2333     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2334     Java package. This provides some syntactic sugar for simple cases of what my
2335     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2336     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2337     item, you can use this simplified notation. Note that only makes sense with
2338     greedy quantifiers. Consequently, the use of the possessive quantifier forces
2339     greediness, whatever the setting of the PCRE_UNGREEDY option.
2341     13. A change of greediness default within a pattern was not taking effect at
2342     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2343     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2344     was abstracted outside.
2346     14. PCRE now supports the \G assertion. It is true when the current matching
2347     position is at the start point of the match. This differs from \A when the
2348     starting offset is non-zero. Used with the /g option of pcretest (or similar
2349     code), it works in the same way as it does for Perl's /g option. If all
2350     alternatives of a regex begin with \G, the expression is anchored to the start
2351     match position, and the "anchored" flag is set in the compiled expression.
2353     15. Some bugs concerning the handling of certain option changes within patterns
2354     have been fixed. These applied to options other than (?ims). For example,
2355     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2356     been the other way round. Some of this was related to change 7 above.
2358     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2359     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2360     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2361     POSIX classes only within a class (e.g. /[[:alpha:]]/).
2363     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2364     mode. Unlike ".", it always matches newline, whatever the setting of
2365     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2366     assertions. Perl allows it, but it doesn't (in general) work because it can't
2367     calculate the length of the lookbehind. At least, that's the case for Perl
2368     5.8.0 - I've been told they are going to document that it doesn't work in
2369     future.
2371     18. Added an error diagnosis for escapes that PCRE does not support: these are
2372     \L, \l, \N, \P, \p, \U, \u, and \X.
2374     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2375     reading past the end of the pattern in cases such as /[abcd/.
2377     20. PCRE was getting more memory than necessary for patterns with classes that
2378     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2380     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2381     compiling PCRE for use with Virtual Pascal.
2383     22. Small fix to the Makefile to make it work properly if the build is done
2384     outside the source tree.
2386     23. Added a new extension: a condition to go with recursion. If a conditional
2387     subpattern starts with (?(R) the "true" branch is used if recursion has
2388     happened, whereas the "false" branch is used only at the top level.
2390     24. When there was a very long string of literal characters (over 255 bytes
2391     without UTF support, over 250 bytes with UTF support), the computation of how
2392     much memory was required could be incorrect, leading to segfaults or other
2393     strange effects.
2395     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2396     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2397     there was a subsequent back reference to those brackets. This meant that, for
2398     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2399     possible to check for precisely this case. All we can do is abandon the
2400     optimization if .* occurs inside capturing brackets when there are any back
2401     references whatsoever. (See below for a better fix that came later.)
2403     26. The handling of the optimization for finding the first character of a
2404     non-anchored pattern, and for finding a character that is required later in the
2405     match were failing in some cases. This didn't break the matching; it just
2406     failed to optimize when it could. The way this is done has been re-implemented.
2408     27. Fixed typo in error message for invalid (?R item (it said "(?p").
2410     28. Added a new feature that provides some of the functionality that Perl
2411     provides with (?{...}). The facility is termed a "callout". The way it is done
2412     in PCRE is for the caller to provide an optional function, by setting
2413     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2414     global variable. By default it is unset, which disables all calling out. To get
2415     the function called, the regex must include (?C) at appropriate points. This
2416     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2417     This provides a means of identifying different callout points. When PCRE
2418     reaches such a point in the regex, if pcre_callout has been set, the external
2419     function is called. It is provided with data in a structure called
2420     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2421     matching continues; if it returns a non-zero value, the match at the current
2422     point fails. However, backtracking will occur if possible. [This was changed
2423     later and other features added - see item 49 below.]
2425     29. pcretest is upgraded to test the callout functionality. It provides a
2426     callout function that displays information. By default, it shows the start of
2427     the match and the current position in the text. There are some new data escapes
2428     to vary what happens:
2430     \C+ in addition, show current contents of captured substrings
2431     \C- do not supply a callout function
2432     \C!n return 1 when callout number n is reached
2433     \C!n!m return 1 when callout number n is reached for the mth time
2435     30. If pcregrep was called with the -l option and just a single file name, it
2436     output "<stdin>" if a match was found, instead of the file name.
2438     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2439     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2440     pcre_exec(). This saves a malloc/free per call. The default value of
2441     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2442     when configuring.
2444     32. The default maximum size of a compiled pattern is 64K. There have been a
2445     few cases of people hitting this limit. The code now uses macros to handle the
2446     storing of links as offsets within the compiled pattern. It defaults to 2-byte
2447     links, but this can be changed to 3 or 4 bytes by --with-link-size when
2448     configuring. Tests 2 and 5 work only with 2-byte links because they output
2449     debugging information about compiled patterns.
2451     33. Internal code re-arrangements:
2453     (a) Moved the debugging function for printing out a compiled regex into
2454     its own source file (printint.c) and used #include to pull it into
2455     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2456     separate copies.
2458     (b) Defined the list of op-code names for debugging as a macro in
2459     internal.h so that it is next to the definition of the opcodes.
2461     (c) Defined a table of op-code lengths for simpler skipping along compiled
2462     code. This is again a macro in internal.h so that it is next to the
2463     definition of the opcodes.
2465     34. Added support for recursive calls to individual subpatterns, along the
2466     lines of Robin Houston's patch (but implemented somewhat differently).
2468     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2469     allow it to read and process whole directories in Win32. This code was
2470     contributed by Lionel Fourquaux; it has not been tested by me.
2472     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2473     used to name a group. Names consist of alphanumerics and underscores, and must
2474     be unique. Back references use the syntax (?P=name) and recursive calls use
2475     (?P>name) which is a PCRE extension to the Python extension. Groups still have
2476     numbers. The function pcre_fullinfo() can be used after compilation to extract
2477     a name/number map. There are three relevant calls:
2479     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2480     PCRE_INFO_NAMECOUNT yields the number of entries
2481     PCRE_INFO_NAMETABLE yields a pointer to the map.
2483     The map is a vector of fixed-size entries. The size of each entry depends on
2484     the length of the longest name used. The first two bytes of each entry are the
2485     group number, most significant byte first. There follows the corresponding
2486     name, zero terminated. The names are in alphabetical order.
2488     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2489     case instead of 255. Making it the same both with and without UTF-8 support
2490     means that the same test output works with both.
2492     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2493     calling malloc() with a zero argument.
2495     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2496     optimization. I've improved things by keeping a bitmap of backreferences with
2497     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2498     fact referenced, the optimization can be applied. It is unlikely that a
2499     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2500     the match to follow \n) will appear inside brackets with a number greater than
2501     31, but if it does, any back reference > 31 suppresses the optimization.
2503     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2504     of disabling numbered capturing parentheses. Any opening parenthesis that is
2505     not followed by ? behaves as if it were followed by ?: but named parentheses
2506     can still be used for capturing (and they will acquire numbers in the usual
2507     way).
2509     41. Redesigned the return codes from the match() function into yes/no/error so
2510     that errors can be passed back from deep inside the nested calls. A malloc
2511     failure while inside a recursive subpattern call now causes the
2512     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2514     42. It is now possible to set a limit on the number of times the match()
2515     function is called in a call to pcre_exec(). This facility makes it possible to
2516     limit the amount of recursion and backtracking, though not in a directly
2517     obvious way, because the match() function is used in a number of different
2518     circumstances. The count starts from zero for each position in the subject
2519     string (for non-anchored patterns). The default limit is, for compatibility, a
2520     large number, namely 10 000 000. You can change this in two ways:
2522     (a) When configuring PCRE before making, you can use --with-match-limit=n
2523     to set a default value for the compiled library.
2525     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2526     a different value is set. See 45 below.
2528     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2530     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2531     of things that can be changed at compile time. The first argument specifies
2532     what is wanted and the second points to where the information is to be placed.
2533     The current list of available information is:
2537     The output is an integer that is set to one if UTF-8 support is available;
2538     otherwise it is set to zero.
2542     The output is an integer that it set to the value of the code that is used for
2543     newline. It is either LF (10) or CR (13).
2547     The output is an integer that contains the number of bytes used for internal
2548     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2552     The output is an integer that contains the threshold above which the POSIX
2553     interface uses malloc() for output vectors. See item 31 above.
2557     The output is an unsigned integer that contains the default limit of the number
2558     of match() calls in a pcre_exec() execution. See 42 above.
2560     44. pcretest has been upgraded by the addition of the -C option. This causes it
2561     to extract all the available output from the new pcre_config() function, and to
2562     output it. The program then exits immediately.
2564     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2565     order to support additional features. One way would have been to define
2566     pcre_exec2() (for example) with extra arguments, but this would not have been
2567     extensible, and would also have required all calls to the original function to
2568     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2569     is used for passing in "extra" data from pcre_study().
2571     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2572     contains the following fields:
2574     flags a bitmap indicating which of the following fields are set
2575     study_data opaque data from pcre_study()
2576     match_limit a way of specifying a limit on match() calls for a specific
2577     call to pcre_exec()
2578     callout_data data for callouts (see 49 below)
2580     The flag bits are also defined in pcre.h, and are
2586     The pcre_study() function now returns one of these new pcre_extra blocks, with
2587     the actual study data pointed to by the study_data field, and the
2588     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2589     before. That is, this change is entirely upwards-compatible and requires no
2590     change to existing code.
2592     If you want to pass in additional data to pcre_exec(), you can either place it
2593     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2594     block.
2596     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2597     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2598     times with different match limits, until it finds the minimum value needed for
2599     pcre_exec() to complete. The value is then output. This can be instructive; for
2600     most simple matches the number is quite small, but for pathological cases it
2601     gets very large very quickly.
2603     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2604     returns the size of the data block pointed to by the study_data field in a
2605     pcre_extra block, that is, the value that was passed as the argument to
2606     pcre_malloc() when PCRE was getting memory in which to place the information
2607     created by pcre_study(). The fourth argument should point to a size_t variable.
2608     pcretest has been extended so that this information is shown after a successful
2609     pcre_study() call when information about the compiled regex is being displayed.
2611     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2612     because what follows is always an absolute path. (Later: it turns out that this
2613     is more than cosmetic for MinGW, because it doesn't like empty path
2614     components.)
2616     49. Some changes have been made to the callout feature (see 28 above):
2618     (i) A callout function now has three choices for what it returns:
2620     0 => success, carry on matching
2621     > 0 => failure at this point, but backtrack if possible
2622     < 0 => serious error, return this value from pcre_exec()
2624     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2625     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2626     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2627     use by callout functions. It will never be used by PCRE itself.
2629     (ii) The pcre_extra structure (see 45 above) has a void * field called
2630     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2631     pcre_callout_block structure has a field of the same name. The contents of
2632     the field passed in the pcre_extra structure are passed to the callout
2633     function in the corresponding field in the callout block. This makes it
2634     easier to use the same callout-containing regex from multiple threads. For
2635     testing, the pcretest program has a new data escape
2637     \C*n pass the number n (may be negative) as callout_data
2639     If the callout function in pcretest receives a non-zero value as
2640     callout_data, it returns that value.
2642     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2643     there were some redundant $(CFLAGS) in commands that are now specified as
2644     $(LINK), which already includes $(CFLAGS).
2646     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2647     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2648     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2649     one-byte characters throughout. Note that case-insensitive matching applies
2650     only to characters whose values are less than 256. PCRE doesn't support the
2651     notion of cases for higher-valued characters.
2653     (i) A character class whose characters are all within 0-255 is handled as
2654     a bit map, and the map is inverted for negative classes. Previously, a
2655     character > 255 always failed to match such a class; however it should
2656     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2658     (ii) A negated character class with a single character < 255 is coded as
2659     "not this character" (OP_NOT). This wasn't working properly when the test
2660     character was multibyte, either singly or repeated.
2662     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2663     mode, for example: \x{100}{2,3}.
2665     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2666     singly or repeated) now correctly test multibyte characters. However,
2667     PCRE doesn't recognize any characters with values greater than 255 as
2668     digits, spaces, or word characters. Such characters always match \D, \S,
2669     and \W, and never match \d, \s, or \w.
2671     (v) Classes may now contain characters and character ranges with values
2672     greater than 255. For example: [ab\x{100}-\x{400}].
2674     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2675     PCRE in UTF-8 mode.
2677     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2678     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2679     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2680     value.)
2682     53. The single man page has become too large. I have therefore split it up into
2683     a number of separate man pages. These also give rise to individual HTML pages;
2684     these are now put in a separate directory, and there is an index.html page that
2685     lists them all. Some hyperlinking between the pages has been installed.
2687     54. Added convenience functions for handling named capturing parentheses.
2689     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2690     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2691     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2692     are faulted.
2694     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2695     calling configure. These values are used when compiling the dftables.c program
2696     which is run to generate the source of the default character tables. They
2697     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2698     you will need to set these values.
2700     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2703     Version 3.9 02-Jan-02
2704 nigel 61 ---------------------
2706     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2708     2. If --disable-static was given, the building process failed when trying to
2709     build pcretest and pcregrep. (For some reason it was using libtool to compile
2710     them, which is not right, as they aren't part of the library.)
2713 nigel 59 Version 3.8 18-Dec-01
2714     ---------------------
2716     1. The experimental UTF-8 code was completely screwed up. It was packing the
2717     bytes in the wrong order. How dumb can you get?
2720 nigel 57 Version 3.7 29-Oct-01
2721     ---------------------
2723     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2724     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2725     this didn't happen under Solaris 8, where I normally test things.
2727 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2728     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2729     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2730     this fix an hour or so after the initial 3.7 release.)
2731 nigel 57
2732 nigel 59
2733 nigel 55 Version 3.6 23-Oct-01
2734     ---------------------
2736     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2737     offsets passed as NULL with zero offset count.
2739     2. The config.guess and config.sub files had not been updated when I moved to
2740     the latest autoconf.
2743 nigel 53 Version 3.5 15-Aug-01
2744     ---------------------
2745 nigel 5
2746 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2747     had been forgotten.
2749     2. By using declared but undefined structures, we can avoid using "void"
2750     definitions in pcre.h while keeping the internal definitions of the structures
2751     private.
2753     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2754     user point of view, this means that both static and shared libraries are built
2755     by default, but this can be individually controlled. More of the work of
2756     handling this static/shared cases is now inside libtool instead of PCRE's make
2757     file.
2759     4. The pcretest utility is now installed along with pcregrep because it is
2760     useful for users (to test regexs) and by doing this, it automatically gets
2761     relinked by libtool. The documentation has been turned into a man page, so
2762     there are now .1, .txt, and .html versions in /doc.
2764     5. Upgrades to pcregrep:
2765     (i) Added long-form option names like gnu grep.
2766     (ii) Added --help to list all options with an explanatory phrase.
2767     (iii) Added -r, --recursive to recurse into sub-directories.
2768     (iv) Added -f, --file to read patterns from a file.
2770     6. pcre_exec() was referring to its "code" argument before testing that
2771     argument for NULL (and giving an error if it was NULL).
2773     7. Upgraded Makefile.in to allow for compiling in a different directory from
2774     the source directory.
2776     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2777     options bits, the pointer it was passed was to an int instead of to an unsigned
2778     long int. This mattered only on 64-bit systems.
2780     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2781     generated) instead of pcre.in, which it its source. Also made the same change
2782     in several of the .c files.
2784     10. A new release of gcc defines printf() as a macro, which broke pcretest
2785     because it had an ifdef in the middle of a string argument for printf(). Fixed
2786     by using separate calls to printf().
2788     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2789     script, to force use of CR or LF instead of \n in the source. On non-Unix
2790     systems, the value can be set in config.h.
2792     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2793     absolute limit. Changed the text of the error message to make this clear, and
2794     likewise updated the man page.
2796     13. The limit of 99 on the number of capturing subpatterns has been removed.
2797     The new limit is 65535, which I hope will not be a "real" limit.
2800 nigel 51 Version 3.4 22-Aug-00
2801     ---------------------
2803     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2805     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2808 nigel 49 Version 3.3 01-Aug-00
2809     ---------------------
2811     1. If an octal character was given, but the value was greater than \377, it
2812     was not getting masked to the least significant bits, as documented. This could
2813     lead to crashes in some systems.
2815     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2816     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2818     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2819     These just pass their arguments on to (pcre_free)(), but they are provided
2820     because some uses of PCRE bind it to non-C systems that can call its functions,
2821     but cannot call free() or pcre_free() directly.
2823     4. Add "make test" as a synonym for "make check". Corrected some comments in
2824     the Makefile.
2826     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2827     Makefile.
2829     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2830     command called pgrep for grepping around the active processes.
2832     7. Added the beginnings of support for UTF-8 character strings.
2834     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2835     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2836     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2837     out for the ar command.)
2840 nigel 47 Version 3.2 12-May-00
2841     ---------------------
2843     This is purely a bug fixing release.
2845     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2846     of ZA. This was just one example of several cases that could provoke this bug,
2847     which was introduced by change 9 of version 2.00. The code for breaking
2848     infinite loops after an iteration that matches an empty string was't working
2849     correctly.
2851     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2852     when matched against abbab (for example). After matching an empty string, it
2853     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2854     caused it to match further down the string than it should.
2856     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2857     was there because it doesn't seem to be needed, and it causes trouble on some
2858     systems, as it is not a Standard C header. It has been removed.
2860     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2861     were reported on the Macintosh. The changes were from
2863     while ((c = *(++ptr)) != 0 && c != '\n');
2864     to
2865     while ((c = *(++ptr)) != 0 && c != '\n') ;
2867     Totally extraordinary, but if that's what it takes...
2869     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2870     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2871     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2872     assumes the way PCRE uses memmove() (always moving upwards).
2874     6. PCRE is being used in one environment where strchr() is not available. There
2875     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2876     faster code anyway.
2879 nigel 45 Version 3.1 09-Feb-00
2880     ---------------------
2882     The only change in this release is the fixing of some bugs in Makefile.in for
2883     the "install" target:
2885     (1) It was failing to install pcreposix.h.
2887     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2890 nigel 43 Version 3.0 01-Feb-00
2891     ---------------------
2892 nigel 41
2893     1. Add support for the /+ modifier to perltest (to output $` like it does in
2894     pcretest).
2896     2. Add support for the /g modifier to perltest.
2898     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2899     matches null strings.
2901     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2902     pattern. Perl treats empty patterns specially - it reuses the most recent
2903     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2904     effect.
2906     5. The POSIX interface was broken in that it was just handing over the POSIX
2907     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2908     required a bigger vector, with some working space on the end. This means that
2909     the POSIX wrapper now has to get and free some memory, and copy the results.
2911 nigel 43 6. Added some simple autoconf support, placing the test data and the
2912     documentation in separate directories, re-organizing some of the
2913     information files, and making it build pcre-config (a GNU standard). Also added
2914     libtool support for building PCRE as a shared library, which is now the
2915     default.
2916 nigel 41
2917 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2918     09 are not valid octal constants. Single digits will be used for minor values
2919     less than 10.
2921     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2922     existing programs that set these in the POSIX interface can use PCRE without
2923     modification.
2925     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2926     return all that pcre_info() returns, plus additional data. The pcre_info()
2927     function is retained for compatibility, but is considered to be obsolete.
2929     10. Added experimental recursion feature (?R) to handle one common case that
2930     Perl 5.6 will be able to do with (?p{...}).
2932     11. Added support for POSIX character classes like [:alpha:], which Perl is
2933     adopting.
2936 nigel 39 Version 2.08 31-Aug-99
2937     ----------------------
2939     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2940     trying to match at the startoffset position, but instead was moving forward to
2941     the next newline as if a previous match had failed.
2943     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2944     and could get into a loop if a null string was matched other than at the start
2945     of the subject.
2947     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2948     be distinguished at compile time, and for completeness also added PCRE_DATE.
2950     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2951     in GnuWin32 environments.
2954 nigel 37 Version 2.07 29-Jul-99
2955     ----------------------
2957     1. The documentation is now supplied in plain text form and HTML as well as in
2958     the form of man page sources.
2960     2. C++ compilers don't like assigning (void *) values to other pointer types.
2961     In particular this affects malloc(). Although there is no problem in Standard
2962     C, I've put in casts to keep C++ compilers happy.
2964     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2965     should be (const char *).
2967     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2968     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2969     However, I haven't made this a standard facility. The documentation doesn't
2970     mention it, and the Makefile doesn't support it.
2972     5. The Makefile now contains an "install" target, with editable destinations at
2973     the top of the file. The pcretest program is not installed.
2975     6. pgrep -V now gives the PCRE version number and date.
2977     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2978     causing the entire string to be ignored, instead of just the last character.
2980     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2981     non-matching string, it can take a very, very long time, even for strings of
2982     quite modest length, because of the nested recursion. PCRE now does better in
2983     some of these cases. It does this by remembering the last required literal
2984     character in the pattern, and pre-searching the subject to ensure it is present
2985     before running the real match. In other words, it applies a heuristic to detect
2986     some types of certain failure quickly, and in the above example, if presented
2987     with a string that has no trailing " it gives "no match" very quickly.
2989     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2990     other alternatives are tried instead.
2993 nigel 35 Version 2.06 09-Jun-99
2994     ----------------------
2996     1. Change pcretest's output for amount of store used to show just the code
2997     space, because the remainder (the data block) varies in size between 32-bit and
2998     64-bit systems.
3000     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3001     start matching at. This allows lookbehinds to work when searching for multiple
3002     occurrences in a string.
3004     3. Added additional options to pcretest for testing multiple occurrences:
3006     /+ outputs the rest of the string that follows a match
3007     /g loops for multiple occurrences, using the new startoffset argument
3008     /G loops for multiple occurrences by passing an incremented pointer
3010     4. PCRE wasn't doing the "first character" optimization for patterns starting
3011     with \b or \B, though it was doing it for other lookbehind assertions. That is,
3012     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3013     the letter 'x'. On long subject strings, this gives a significant speed-up.
3016 nigel 33 Version 2.05 21-Apr-99
3017     ----------------------
3019     1. Changed the type of magic_number from int to long int so that it works
3020     properly on 16-bit systems.
3022     2. Fixed a bug which caused patterns starting with .* not to work correctly
3023     when the subject string contained newline characters. PCRE was assuming
3024     anchoring for such patterns in all cases, which is not correct because .* will
3025     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3026     DOTALL is set at top level; otherwise it knows that patterns starting with .*
3027     must be retried after every newline in the subject.
3030 nigel 31 Version 2.04 18-Feb-99
3031     ----------------------
3033     1. For parenthesized subpatterns with repeats whose minimum was zero, the
3034     computation of the store needed to hold the pattern was incorrect (too large).
3035     If such patterns were nested a few deep, this could multiply and become a real
3036     problem.
3038     2. Added /M option to pcretest to show the memory requirement of a specific
3039     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3041     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3042     compiled in such a way that the backtracking after subsequent failure was
3043     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3044     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3047 nigel 29 Version 2.03 02-Feb-99
3048     ----------------------
3050     1. Fixed typo and small mistake in man page.
3052 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
3053     LICENCE file containing the conditions.
3054 nigel 29
3055     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3056     Perl, that is the internal \ allows the delimiter to be included in the
3057     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3058     the final delimiter, add \ to the end of the pattern (to test the error).
3060     4. Added the convenience functions for extracting substrings after a successful
3061     match. Updated pcretest to make it able to test these functions.
3064 nigel 27 Version 2.02 14-Jan-99
3065     ----------------------
3067     1. Initialized the working variables associated with each extraction so that
3068     their saving and restoring doesn't refer to uninitialized store.
3070     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3071     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3072     fix the problem.
3074     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3075     calls, and wasn't printing the correct value for compiling calls. Increased the
3076     default value of LOOPREPEAT, and the number of significant figures in the
3077     times.
3079     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3081     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3082     a building problem on Windows NT with a FAT file system.
3085 nigel 25 Version 2.01 21-Oct-98
3086     ----------------------
3088     1. Changed the API for pcre_compile() to allow for the provision of a pointer
3089     to character tables built by pcre_maketables() in the current locale. If NULL
3090     is passed, the default tables are used.
3093 nigel 23 Version 2.00 24-Sep-98
3094 nigel 21 ----------------------
3096 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3097     it any more.
3099     2. Allow quantification of (?>) groups, and make it work correctly.
3101     3. The first character computation wasn't working for (?>) groups.
3103     4. Correct the implementation of \Z (it is permitted to match on the \n at the
3104     end of the subject) and add 5.005's \z, which really does match only at the
3105     very end of the subject.
3107     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3109     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3110     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3111     localized options. All options to pcre_study() were also removed.
3113     7. Add other new features from 5.005:
3115     $(?<= positive lookbehind
3116     $(?<! negative lookbehind
3117     (?imsx-imsx) added the unsetting capability
3118     such a setting is global if at outer level; local otherwise
3119     (?imsx-imsx:) non-capturing groups with option setting
3120     (?(cond)re|re) conditional pattern matching
3122     A backreference to itself in a repeated group matches the previous
3123     captured string.
3125     8. General tidying up of studying (both automatic and via "study")
3126     consequential on the addition of new assertions.
3128     9. As in 5.005, unlimited repeated groups that could match an empty substring
3129     are no longer faulted at compile time. Instead, the loop is forcibly broken at
3130     runtime if any iteration does actually match an empty substring.
3132     10. Include the RunTest script in the distribution.
3134     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3135     discrepancies, some of which were old and were also with respect to 5.004. They
3136     have now been fixed.
3139     Version 1.09 28-Apr-98
3140     ----------------------
3142 nigel 21 1. A negated single character class followed by a quantifier with a minimum
3143     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3144     program crashes, or just wrong answers. This did not apply to negated classes
3145     containing more than one character, or to minima other than one.
3148 nigel 19 Version 1.08 27-Mar-98
3149     ----------------------
3151     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3153     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3154     latter must appear before anything that relies on it in the pattern.
3157 nigel 17 Version 1.07 16-Feb-98
3158     ----------------------
3160     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3161     repeat of a potentially empty string).
3164 nigel 15 Version 1.06 23-Jan-98
3165     ----------------------
3167     1. Added Markus Oberhumer's little patches for C++.
3169     2. Literal strings longer than 255 characters were broken.
3172 nigel 13 Version 1.05 23-Dec-97
3173     ----------------------
3175     1. Negated character classes containing more than one character were failing if
3176     PCRE_CASELESS was set at run time.
3179 nigel 11 Version 1.04 19-Dec-97
3180     ----------------------
3182     1. Corrected the man page, where some "const" qualifiers had been omitted.
3184     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3185     input syntax.
3187     3. Fixed memory leak which occurred when a regex with back references was
3188     matched with an offsets vector that wasn't big enough. The temporary memory
3189     that is used in this case wasn't being freed if the match failed.
3191     4. Tidied pcretest to ensure it frees memory that it gets.
3193     5. Temporary memory was being obtained in the case where the passed offsets
3194     vector was exactly big enough.
3196     6. Corrected definition of offsetof() from change 5 below.
3198     7. I had screwed up change 6 below and broken the rules for the use of
3199     setjmp(). Now fixed.
3202 nigel 9 Version 1.03 18-Dec-97
3203     ----------------------
3205     1. A erroneous regex with a missing opening parenthesis was correctly
3206     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3207     on some systems.
3209     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3210     it was reported that one broken compiler failed on the former because "code" is
3211     also an independent variable.
3213     3. The erroneous regex a[]b caused an array overrun reference.
3215     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3216     fail on data ending with that character. (It was going on too far, and checking
3217     the next character, typically a binary zero.) This was specific to the
3218     optimized code for single-character negative classes.
3220     5. Added a contributed patch from the TIN world which does the following:
3222     + Add an undef for memmove, in case the the system defines a macro for it.
3224     + Add a definition of offsetof(), in case there isn't one. (I don't know
3225     the reason behind this - offsetof() is part of the ANSI standard - but
3226     it does no harm).
3228     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3229     most of the places where whitespace preceded '#'. I have given up and
3230     allowed the remaining 2 cases to be at the margin.
3232     + Rename some variables in pcre to eliminate shadowing. This seems very
3233     pedantic, but does no harm, of course.
3235     6. Moved the call to setjmp() into its own function, to get rid of warnings
3236     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3238     7. Constructs such as \d{8,} were compiling into the equivalent of
3239 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3240 nigel 9 outcome, but in this particular case used more store than had been allocated,
3241     which caused the bug to be discovered because it threw up an internal error.
3243     8. The debugging code in both pcre and pcretest for outputting the compiled
3244     form of a regex was going wrong in the case of back references followed by
3245     curly-bracketed repeats.
3248 nigel 7 Version 1.02 12-Dec-97
3249     ----------------------
3251     1. Typos in pcre.3 and comments in the source fixed.
3253     2. Applied a contributed patch to get rid of places where it used to remove
3254     'const' from variables, and fixed some signed/unsigned and uninitialized
3255     variable warnings.
3257     3. Added the "runtest" target to Makefile.
3259     4. Set default compiler flag to -O2 rather than just -O.
3262 nigel 5 Version 1.01 19-Nov-97
3263     ----------------------
3265     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3266     like /([ab]*)*/, that is, for classes with more than one character in them.
3268     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3269     as /((?>a*))*/ (a PCRE_EXTRA facility).
3272     Version 1.00 18-Nov-97
3273     ----------------------
3275     1. Added compile-time macros to support systems such as SunOS4 which don't have
3276     memmove() or strerror() but have other things that can be used instead.
3278     2. Arranged that "make clean" removes the executables.
3281 nigel 3 Version 0.99 27-Oct-97
3282     ----------------------
3284     1. Fixed bug in code for optimizing classes with only one character. It was
3285     initializing a 32-byte map regardless, which could cause it to run off the end
3286     of the memory it had got.
3288     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3291     Version 0.98 22-Oct-97
3292     ----------------------
3294     1. Fixed bug in code for handling temporary memory usage when there are more
3295     back references than supplied space in the ovector. This could cause segfaults.
3298     Version 0.97 21-Oct-97
3299     ----------------------
3301     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3303     2. Optimized negated single characters not to use a bit map.
3305     3. Brought error texts together as macro definitions; clarified some of them;
3306     fixed one that was wrong - it said "range out of order" when it meant "invalid
3307     escape sequence".
3309     4. Changed some char * arguments to const char *.
3311     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3313     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3314     pcretest.
3317     Version 0.96 16-Oct-97
3318     ----------------------
3320     1. Added a simple "pgrep" utility to the distribution.
3322     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3323     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3324     where "ddd" means "one or more decimal digits".
3326     3. Fixed serious bug. If a pattern had a back reference, but the call to
3327     pcre_exec() didn't supply a large enough ovector to record the related
3328     identifying subpattern, the match always failed. PCRE now remembers the number
3329     of the largest back reference, and gets some temporary memory in which to save
3330     the offsets during matching if necessary, in order to ensure that
3331     backreferences always work.
3333     4. Increased the compatibility with Perl in a number of ways:
3335     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3336     to request this handling. The option can be set at compile or exec time.
3338     (b) $ matches before a terminating newline by default; an option
3339     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3340     mode). The option can be set at compile or exec time.
3342     (c) The handling of \ followed by a digit other than 0 is now supposed to be
3343     the same as Perl's. If the decimal number it represents is less than 10
3344     or there aren't that many previous left capturing parentheses, an octal
3345     escape is read. Inside a character class, it's always an octal escape,
3346     even if it is a single digit.
3348     (d) An escaped but undefined alphabetic character is taken as a literal,
3349     unless PCRE_EXTRA is set. Currently this just reserves the remaining
3350     escapes.
3352     (e) {0} is now permitted. (The previous item is removed from the compiled
3353     pattern).
3355     5. Changed all the names of code files so that the basic parts are no longer
3356     than 10 characters, and abolished the teeny "globals.c" file.
3358     6. Changed the handling of character classes; they are now done with a 32-byte
3359     bit map always.
3361     7. Added the -d and /D options to pcretest to make it possible to look at the
3362     internals of compilation without having to recompile pcre.
3365     Version 0.95 23-Sep-97
3366     ----------------------
3368     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3369     \x20 at the start of a run of normal characters. These were being treated as
3370     real characters, instead of the source characters being re-checked.
3373     Version 0.94 18-Sep-97
3374     ----------------------
3376     1. The functions are now thread-safe, with the caveat that the global variables
3377     containing pointers to malloc() and free() or alternative functions are the
3378     same for all threads.
3380     2. Get pcre_study() to generate a bitmap of initial characters for non-
3381     anchored patterns when this is possible, and use it if passed to pcre_exec().
3384     Version 0.93 15-Sep-97
3385     ----------------------
3387     1. /(b)|(:+)/ was computing an incorrect first character.
3389     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3390     but not actually doing anything yet.
3392     3. Treat "-" characters in classes that cannot be part of ranges as literals,
3393     as Perl does (e.g. [-az] or [az-]).
3395     4. Set the anchored flag if a branch starts with .* or .*? because that tests
3396     all possible positions.
3398     5. Split up into different modules to avoid including unneeded functions in a
3399     compiled binary. However, compile and exec are still in one module. The "study"
3400     function is split off.
3402     6. The character tables are now in a separate module whose source is generated
3403     by an auxiliary program - but can then be edited by hand if required. There are
3404     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3405     toupper() in the code.
3407     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3408     make them global. Abolish the function for setting them, as the caller can now
3409     set them directly.
3412     Version 0.92 11-Sep-97
3413     ----------------------
3415     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3416     (e.g. /a{1,3}/) was broken (I mis-optimized it).
3418     2. Caseless matching was not working in character classes if the characters in
3419     the pattern were in upper case.
3421     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3423     4. Make PCRE_ANCHORED public and accept as a compile option.
3425     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3426     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3427     pass them.
3429     6. Give an error if bad option bits passed at compile or run time.
3431     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3432     pcretest to cause it to pass that flag.
3434     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3435     options, and the first character, if set.
3437     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3440     Version 0.91 10-Sep-97
3441     ----------------------
3443     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3444     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3446     2. PCRE was looping on encountering an indefinitely repeated back reference to
3447     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3448     Perl does - treats the match as successful.
3450     ****


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

ViewVC Help
Powered by ViewVC 1.1.12