ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 405 - (hide annotations) (download)
Mon Mar 23 12:05:20 2009 UTC (6 years, 1 month ago) by ph10
File size: 164351 byte(s)
Patch to make CMake work when PCRE is included within another project.

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 376 Version 7.9 xx-xxx-09
5     ---------------------
7 ph10 392 1. When building with support for bzlib/zlib (pcregrep) and/or readline
8 ph10 376 (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 ph10 392 has been fixed, and now only pcregrep is linked with bzlib/zlib and only
12 ph10 376 pcretest is linked with readline.
13 ph10 392
14 ph10 376 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.
18 ph10 392
19     3. The pcre_config() function was treating the PCRE_MATCH_LIMIT and
20 ph10 376 PCRE_MATCH_LIMIT_RETURSION values as ints, when they should be long ints.
21 ph10 392
22 ph10 376 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 ph10 392 lines. This is not true; no spaces are inserted. I have also clarified the
25 ph10 376 wording for the --colour (or --color) option.
26 ph10 392
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.
30 ph10 392
31 ph10 378 6. When --colo(u)r was used in pcregrep, only the first matching substring in
32 ph10 392 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.
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 ph10 379 locked it out (using PCRE's PCRE_NOTEMPTY option). By experiment, this
38     seems to be how GNU grep behaves.
39 ph10 392
40 ph10 380 8. The pattern (?(?=.*b)b|^) was incorrectly compiled as "match must be at
41 ph10 392 start or after a newline", because the conditional assertion was not being
42     correctly handled. The rule now is that both the assertion and what follows
43     in the first alternative must satisfy the test.
45 ph10 405 9. If auto-callout was enabled in a pattern with a conditional group whose
46 ph10 399 condition was an assertion, PCRE could crash during matching, both with
47     pcre_exec() and pcre_dfa_exec().
48 ph10 392
49     10. The PCRE_DOLLAR_ENDONLY option was not working when pcre_dfa_exec() was
50     used for matching.
52     11. Unicode property support in character classes was not working for
53 ph10 385 characters (bytes) greater than 127 when not in UTF-8 mode.
54 ph10 388
55 ph10 392 12. Added the -M command line option to pcretest.
57 ph10 388 14. Added the non-standard REG_NOTEMPTY option to the POSIX interface.
58 ph10 389
59     15. Added the PCRE_NO_START_OPTIMIZE match-time option.
60 ph10 390
61 ph10 392 16. Added comments and documentation about mis-use of no_arg in the C++
62 ph10 390 wrapper.
63 ph10 392
64 ph10 391 17. Implemented support for UTF-8 encoding in EBCDIC environments, a patch
65 ph10 392 from Martin Jerabek that uses macro names for all relevant character and
66 ph10 391 string constants.
67 ph10 405
68 ph10 393 18. Added to pcre_internal.h two configuration checks: (a) If both EBCDIC and
69     SUPPORT_UTF8 are set, give an error; (b) If SUPPORT_UCP is set without
70 ph10 405 SUPPORT_UTF8, define SUPPORT_UTF8. The "configure" script handles both of
71     these, but not everybody uses configure.
73     19. A conditional group that had only one branch was not being correctly
74     recognized as an item that could match an empty string. This meant that an
75     enclosing group might also not be so recognized, causing infinite looping
76     (and probably a segfault) for patterns such as ^"((?(?=[a])[^"])|b)*"$
77     with the subject "ab", where knowledge that the repeated group can match
78 ph10 396 nothing is needed in order to break the loop.
79 ph10 405
80 ph10 397 20. If a pattern that was compiled with callouts was matched using pcre_dfa_
81 ph10 405 exec(), but without supplying a callout function, matching went wrong.
83     21. If PCRE_ERROR_MATCHLIMIT occurred during a recursion, there was a memory
84     leak if the size of the offset vector was greater than 30. When the vector
85     is smaller, the saved offsets during recursion go onto a local stack
86     vector, but for larger vectors malloc() is used. It was failing to free
87     when the recursion yielded PCRE_ERROR_MATCH_LIMIT (or any other "abnormal"
88 ph10 400 error, in fact).
89 ph10 405
90 ph10 402 22. There was a missing #ifdef SUPPORT_UTF8 round one of the variables in the
91 ph10 405 heapframe that is used only when UTF-8 support is enabled. This caused no
92     problem, but was untidy.
93 ph10 376
94 ph10 405 23. Steven Van Ingelgem's patch to CMakeLists.txt to change the name
95     CMAKE_BINARY_DIR to PROJECT_BINARY_DIR so that it works when PCRE is
96     included within another project.
97 ph10 392
98 ph10 405
100 ph10 374 Version 7.8 05-Sep-08
101 ph10 349 ---------------------
103     1. Replaced UCP searching code with optimized version as implemented for Ad
104     Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
105 ph10 351 stage table and inline lookup instead of a function, giving speed ups of 2
106     to 5 times on some simple patterns that I tested. Permission was given to
107     distribute the MultiStage2.py script that generates the tables (it's not in
108     the tarball, but is in the Subversion repository).
109 ph10 349
110 ph10 351 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
111     scripts.
112 ph10 358
113 ph10 353 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
114     a group with a zero qualifier. The result of the study could be incorrect,
115 ph10 358 or the function might crash, depending on the pattern.
117     4. Caseless matching was not working for non-ASCII characters in back
118 ph10 354 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
119 ph10 358 It now works when Unicode Property Support is available.
121 ph10 355 5. In pcretest, an escape such as \x{de} in the data was always generating
122     a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
123     non-UTF-8 mode. If the value is greater than 255, it gives a warning about
124 ph10 358 truncation.
125 ph10 349
126 ph10 358 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
128 ph10 357 7. Added two (int) casts to pcregrep when printing the difference of two
129     pointers, in case they are 64-bit values.
130 ph10 371
131     8. Added comments about Mac OS X stack usage to the pcrestack man page and to
132     test 2 if it fails.
134 ph10 359 9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
135 ph10 371 and a #define of that name to empty if it is not externally set. This is to
136     allow users of MSVC to set it if necessary.
138     10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
139 ph10 359 the convenience functions in the pcre_get.c source file.
140 ph10 371
141 ph10 360 11. An option change at the start of a pattern that had top-level alternatives
142     could cause overwriting and/or a crash. This command provoked a crash in
143 ph10 371 some environments:
145     printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
147 ph10 360 This potential security problem was recorded as CVE-2008-2371.
148 ph10 371
149 ph10 361 12. For a pattern where the match had to start at the beginning or immediately
150     after a newline (e.g /.*anything/ without the DOTALL flag), pcre_exec() and
151 ph10 371 pcre_dfa_exec() could read past the end of the passed subject if there was
152 ph10 361 no match. To help with detecting such bugs (e.g. with valgrind), I modified
153     pcretest so that it places the subject at the end of its malloc-ed buffer.
154 ph10 371
155 ph10 364 13. The change to pcretest in 12 above threw up a couple more cases when pcre_
156 ph10 371 exec() might read past the end of the data buffer in UTF-8 mode.
158 ph10 364 14. A similar bug to 7.3/2 existed when the PCRE_FIRSTLINE option was set and
159 ph10 371 the data contained the byte 0x85 as part of a UTF-8 character within its
160     first line. This applied both to normal and DFA matching.
162 ph10 366 15. Lazy qualifiers were not working in some cases in UTF-8 mode. For example,
163 ph10 371 /^[^d]*?$/8 failed to match "abc".
164 ph10 351
165 ph10 371 16. Added a missing copyright notice to pcrecpp_internal.h.
167     17. Make it more clear in the documentation that values returned from
168 ph10 368 pcre_exec() in ovector are byte offsets, not character counts.
169 ph10 357
170 ph10 371 18. Tidied a few places to stop certain compilers from issuing warnings.
171 ph10 368
172 ph10 373 19. Updated the Virtual Pascal + BCC files to compile the latest v7.7, as
173     supplied by Stefan Weber. I made a further small update for 7.8 because
174 ph10 374 there is a change of source arrangements: the pcre_searchfuncs.c module is
175     replaced by pcre_ucd.c.
176 ph10 371
177 ph10 373
178 ph10 347 Version 7.7 07-May-08
179 ph10 321 ---------------------
181     1. Applied Craig's patch to sort out a long long problem: "If we can't convert
182 ph10 345 a string to a long long, pretend we don't even have a long long." This is
183 ph10 321 done by checking for the strtoq, strtoll, and _strtoi64 functions.
184 ph10 345
185 ph10 322 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
186     pre-7.6 versions, which defined a global no_arg variable instead of putting
187 ph10 332 it in the RE class. (See also #8 below.)
188 ph10 345
189     3. Remove a line of dead code, identified by coverity and reported by Nuno
190     Lopes.
192 ph10 324 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
193 ph10 321
194 ph10 324 (1) The include/exclude patterns were being applied to the whole pathnames
195 ph10 345 of files, instead of just to the final components.
197 ph10 324 (2) If there was more than one level of directory, the subdirectories were
198     skipped unless they satisfied the include/exclude conditions. This is
199     inconsistent with GNU grep (and could even be seen as contrary to the
200     pcregrep specification - which I improved to make it absolutely clear).
201     The action now is always to scan all levels of directory, and just
202     apply the include/exclude patterns to regular files.
203 ph10 345
204 ph10 325 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
205 ph10 345 --exclude_dir in the tests to avoid scanning .svn directories.
207 ph10 326 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
208 ph10 345 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
209 ph10 326 doesn't support NULs in patterns.
210 ph10 345
211     7. Added some missing "const"s to declarations of static tables in
212     pcre_compile.c and pcre_dfa_exec.c.
214 ph10 328 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
215 ph10 329 caused by fix #2 above. (Subsequently also a second patch to fix the
216 ph10 345 first patch. And a third patch - this was a messy problem.)
217 ph10 321
218 ph10 345 9. Applied Craig's patch to remove the use of push_back().
220 ph10 332 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
221     matching function regexec().
222 ph10 345
223     11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
224     which, however, unlike Perl's \g{...}, are subroutine calls, not back
225 ph10 333 references. PCRE supports relative numbers with this syntax (I don't think
226     Oniguruma does).
227 ph10 345
228     12. Previously, a group with a zero repeat such as (...){0} was completely
229 ph10 335 omitted from the compiled regex. However, this means that if the group
230     was called as a subroutine from elsewhere in the pattern, things went wrong
231 ph10 345 (an internal error was given). Such groups are now left in the compiled
232     pattern, with a new opcode that causes them to be skipped at execution
233 ph10 335 time.
234 ph10 345
235 ph10 341 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
236     to the way PCRE behaves:
237 ph10 345
238 ph10 336 (a) A lone ] character is dis-allowed (Perl treats it as data).
239 ph10 345
240     (b) A back reference to an unmatched subpattern matches an empty string
241 ph10 336 (Perl fails the current match path).
242 ph10 345
243 ph10 341 (c) A data ] in a character class must be notated as \] because if the
244 ph10 345 first data character in a class is ], it defines an empty class. (In
245 ph10 341 Perl it is not possible to have an empty class.) The empty class []
246     never matches; it forces failure and is equivalent to (*FAIL) or (?!).
247     The negative empty class [^] matches any one character, independently
248     of the DOTALL setting.
249 ph10 345
250     14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
251 ph10 340 non-existent subpattern following a character class starting with ']' and
252     containing () gave an internal compiling error instead of "reference to
253     non-existent subpattern". Fortunately, when the pattern did exist, the
254 ph10 345 compiled code was correct. (When scanning forwards to check for the
255 ph10 340 existencd of the subpattern, it was treating the data ']' as terminating
256 ph10 345 the class, so got the count wrong. When actually compiling, the reference
257 ph10 340 was subsequently set up correctly.)
258 ph10 345
259 ph10 341 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
260 ph10 345 it was being rejected as not supported by pcre_dfa_exec(), even though
261     other assertions are supported. I have made pcre_dfa_exec() support
262     (*FAIL).
264 ph10 342 16. The implementation of 13c above involved the invention of a new opcode,
265 ph10 345 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
266     cannot be changed at match time, I realized I could make a small
267     improvement to matching performance by compiling OP_ALLANY instead of
268     OP_ANY for "." when DOTALL was set, and then removing the runtime tests
269     on the OP_ANY path.
271     17. Compiling pcretest on Windows with readline support failed without the
272 ph10 343 following two fixes: (1) Make the unistd.h include conditional on
273 ph10 345 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
275 ph10 344 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
276 ph10 345 ncurses library to be included for pcretest when ReadLine support is
277     requested, but also to allow for it to be overridden. This patch came from
278     Daniel Bergström.
279 ph10 347
280 ph10 346 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
281 ph10 347 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
282     any errors with the current Unicode tables. Thanks to Peter Kankowski for
283     spotting this.
284 ph10 324
285 ph10 332
286 ph10 319 Version 7.6 28-Jan-08
287 ph10 300 ---------------------
289 ph10 302 1. A character class containing a very large number of characters with
290     codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
291     overflow.
292 ph10 309
293     2. Patch to cut out the "long long" test in pcrecpp_unittest when
294     HAVE_LONG_LONG is not defined.
296 ph10 303 3. Applied Christian Ehrlicher's patch to update the CMake build files to
297 ph10 304 bring them up to date and include new features. This patch includes:
298 ph10 309
299 ph10 304 - Fixed PH's badly added libz and libbz2 support.
300     - Fixed a problem with static linking.
301 ph10 312 - Added pcredemo. [But later removed - see 7 below.]
302 ph10 304 - Fixed dftables problem and added an option.
303     - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
305     - Added readline support for pcretest.
306 ph10 309 - Added an listing of the option settings after cmake has run.
308 ph10 314 4. A user submitted a patch to Makefile that makes it easy to create
309     "pcre.dll" under mingw when using Configure/Make. I added stuff to
310     Makefile.am that cause it to include this special target, without
311     affecting anything else. Note that the same mingw target plus all
312     the other distribution libraries and programs are now supported
313     when configuring with CMake (see 6 below) instead of with
314     Configure/Make.
315 ph10 309
316 ph10 308 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
317     This is an attempt to solve the reported problem "pcrecpp::no_arg is not
318 ph10 309 exported in the Windows port". It has not yet been confirmed that the patch
319     solves the problem, but it does no harm.
320 ph10 313
321 ph10 311 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
322 ph10 319 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
323 ph10 314 with CMake, and also correct the comment about stack recursion.
324 ph10 313
325 ph10 312 7. Remove the automatic building of pcredemo from the ./configure system and
326 ph10 313 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
327     of a program that users should build themselves after PCRE is installed, so
328     building it automatically is not really right. What is more, it gave
329 ph10 312 trouble in some build environments.
330 ph10 300
331 ph10 319 8. Further tidies to CMakeLists.txt from Sheri and Christian.
332 ph10 308
333 ph10 319
334 ph10 298 Version 7.5 10-Jan-08
335 ph10 263 ---------------------
337     1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
338     values in parens when parsing an RE using the C++ wrapper."
339 ph10 286
340 ph10 264 2. Negative specials like \S did not work in character classes in UTF-8 mode.
341     Characters greater than 255 were excluded from the class instead of being
342     included.
343 ph10 286
344     3. The same bug as (2) above applied to negated POSIX classes such as
345 ph10 265 [:^space:].
346 ph10 286
347 ph10 267 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
348     defined or documented. It seems to have been a typo for PCRE_STATIC, so
349 ph10 286 I have changed it.
351 ph10 268 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
352 ph10 286 first named subpattern) and a construct such as (?&a) would reference the
353     first named subpattern whose name started with "a" (in other words, the
354 ph10 272 length check was missing). Both these problems are fixed. "Subpattern name
355     expected" is now given for (?&) (a zero-length name), and this patch also
356     makes it give the same error for \k'' (previously it complained that that
357     was a reference to a non-existent subpattern).
358 ph10 286
359 ph10 269 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
360 ph10 286 this is right because (?- can be followed by option settings as well as by
361 ph10 269 digits. I have, however, made the messages clearer.
362 ph10 286
363 ph10 270 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
364     than the number used in the conditional) now cause a compile-time error.
365     This is actually not compatible with Perl, which accepts such patterns, but
366     treats the conditional as always being FALSE (as PCRE used to), but it
367     seems to me that giving a diagnostic is better.
368 ph10 286
369 ph10 274 8. Change "alphameric" to the more common word "alphanumeric" in comments
370 ph10 275 and messages.
371 ph10 286
372     9. Fix two occurrences of "backslash" in comments that should have been
373     "backspace".
375     10. Remove two redundant lines of code that can never be obeyed (their function
376     was moved elsewhere).
378     11. The program that makes PCRE's Unicode character property table had a bug
379     which caused it to generate incorrect table entries for sequences of
380 ph10 277 characters that have the same character type, but are in different scripts.
381 ph10 286 It amalgamated them into a single range, with the script of the first of
382 ph10 277 them. In other words, some characters were in the wrong script. There were
383     thirteen such cases, affecting characters in the following ranges:
384 ph10 286
385 ph10 277 U+002b0 - U+002c1
386     U+0060c - U+0060d
387 ph10 286 U+0061e - U+00612
388 ph10 277 U+0064b - U+0065e
389     U+0074d - U+0076d
390     U+01800 - U+01805
391     U+01d00 - U+01d77
392     U+01d9b - U+01dbf
393     U+0200b - U+0200f
394     U+030fc - U+030fe
395     U+03260 - U+0327f
396     U+0fb46 - U+0fbb1
397     U+10450 - U+1049d
398 ph10 286
399 ph10 279 12. The -o option (show only the matching part of a line) for pcregrep was not
400     compatible with GNU grep in that, if there was more than one match in a
401     line, it showed only the first of them. It now behaves in the same way as
402     GNU grep.
403 ph10 286
404     13. If the -o and -v options were combined for pcregrep, it printed a blank
405     line for every non-matching line. GNU grep prints nothing, and pcregrep now
406     does the same. The return code can be used to tell if there were any
407     non-matching lines.
409 ph10 289 14. Added --file-offsets and --line-offsets to pcregrep.
411     15. The pattern (?=something)(?R) was not being diagnosed as a potentially
412 ph10 282 infinitely looping recursion. The bug was that positive lookaheads were not
413 ph10 286 being skipped when checking for a possible empty match (negative lookaheads
414     and both kinds of lookbehind were skipped).
416 ph10 289 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
417 ph10 284 inclusion of <windows.h> to before rather than after the definition of
418     INVALID_FILE_ATTRIBUTES (patch from David Byron).
419 ph10 263
420 ph10 289 17. Specifying a possessive quantifier with a specific limit for a Unicode
421 ph10 286 character property caused pcre_compile() to compile bad code, which led at
422     runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
423     are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
424     caused the error; without that there was no problem.
425 ph10 263
426 ph10 289 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
427 ph10 286
428 ph10 289 19. Added --enable-pcretest-libreadline.
429 ph10 286
430 ph10 289 20. In pcrecpp.cc, the variable 'count' was incremented twice in
431 ph10 288 RE::GlobalReplace(). As a result, the number of replacements returned was
432 ph10 298 double what it should be. I removed one of the increments, but Craig sent a
433     later patch that removed the other one (the right fix) and added unit tests
434     that check the return values (which was not done before).
435 ph10 292
436 ph10 291 21. Several CMake things:
437 ph10 287
438 ph10 291 (1) Arranged that, when cmake is used on Unix, the libraries end up with
439 ph10 292 the names libpcre and libpcreposix, not just pcre and pcreposix.
440 ph10 288
441 ph10 292 (2) The above change means that pcretest and pcregrep are now correctly
442     linked with the newly-built libraries, not previously installed ones.
443 ph10 291
445 ph10 298
446     22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
447     crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
448     UTF-8 newline character). The key issue is that the pattern starts .*;
449     this means that the match must be either at the beginning, or after a
450     newline. The bug was in the code for advancing after a failed match and
451     checking that the new position followed a newline. It was not taking
452 ph10 294 account of UTF-8 characters correctly.
453 ph10 298
454     23. PCRE was behaving differently from Perl in the way it recognized POSIX
455     character classes. PCRE was not treating the sequence [:...:] as a
456     character class unless the ... were all letters. Perl, however, seems to
457     allow any characters between [: and :], though of course it rejects as
458     unknown any "names" that contain non-letters, because all the known class
459     names consist only of letters. Thus, Perl gives an error for [[:1234:]],
460     for example, whereas PCRE did not - it did not recognize a POSIX character
461     class. This seemed a bit dangerous, so the code has been changed to be
462     closer to Perl. The behaviour is not identical to Perl, because PCRE will
463     diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
464     treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
465     Perl does, and where it didn't before.
467 ph10 296 24. Rewrite so as to remove the single use of %n from pcregrep because in some
468     Windows environments %n is disabled by default.
469 ph10 292
471 ph10 260 Version 7.4 21-Sep-07
472 ph10 230 ---------------------
474     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
475 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
476     LF". That isn't really right - the whole point of the change was to try to
477     help when there was an actual mention of one of the two characters. So now
478     the change happens only if \r or \n (or a literal CR or LF) character is
479 ph10 230 encountered.
480 ph10 231
481     2. The 32-bit options word was also used for 6 internal flags, but the numbers
482     of both had grown to the point where there were only 3 bits left.
483     Fortunately, there was spare space in the data structure, and so I have
484     moved the internal flags into a new 16-bit field to free up more option
485 ph10 230 bits.
486 ph10 231
487     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
488     but did not set the internal JCHANGED flag - either of these is enough to
489     control the way the "get" function works - but the PCRE_INFO_JCHANGED
490     facility is supposed to tell if (?J) was ever used, so now (?J) at the
491 ph10 230 start sets both bits.
493 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
494     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
495 ph10 230
496 ph10 243 5. doc/pcresyntax.html was missing from the distribution.
498     6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
499 ph10 233 compatibility, even though it is no longer used.
500 ph10 243
501     7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
502 ph10 254 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
503 ph10 259 windows.h file is present (where different names are used). [This was
504 ph10 257 reversed later after testing - see 16 below.]
505 ph10 243
506     8. Changed all #include <config.h> to #include "config.h". There were also
507 ph10 236 some further <pcre.h> cases that I changed to "pcre.h".
508 ph10 243
509 ph10 239 9. When pcregrep was used with the --colour option, it missed the line ending
510 ph10 243 sequence off the lines that it output.
512     10. It was pointed out to me that arrays of string pointers cause lots of
513     relocations when a shared library is dynamically loaded. A technique of
514     using a single long string with a table of offsets can drastically reduce
515     these. I have refactored PCRE in four places to do this. The result is
516 ph10 240 dramatic:
517 ph10 243
518 ph10 240 Originally: 290
519     After changing UCP table: 187
520 ph10 243 After changing error message table: 43
521 ph10 240 After changing table of "verbs" 36
522     After changing table of Posix names 22
523 ph10 243
524 ph10 240 Thanks to the folks working on Gregex for glib for this insight.
525 ph10 247
526 ph10 244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
527 ph10 247 unicode-properties was also set.
529 ph10 248 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
530 ph10 231
531 ph10 253 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
532     checked only for CRLF.
533 ph10 233
534 ph10 259 14. Added casts to pcretest.c to avoid compiler warnings.
536 ph10 256 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
537 ph10 253
538 ph10 257 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
539 ph10 259 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
540     entirely. This removes changes made in 7 above.
541 ph10 256
542 ph10 261 17. The CMake files have been updated, and there is now more information about
543     building with CMake in the NON-UNIX-USE document.
544 ph10 257
545 ph10 261
546 ph10 228 Version 7.3 28-Aug-07
547 ph10 157 ---------------------
549 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
550     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
551     brackets there is not right, since it causes compilers to look for an
552     installed pcre.h, not the version that is in the source that is being
553     compiled (which of course may be different). I have changed it back to:
554 ph10 197
555 ph10 189 #include "pcre.h"
556 ph10 197
557     I have a vague recollection that the change was concerned with compiling in
558     different directories, but in the new build system, that is taken care of
559     by the VPATH setting the Makefile.
561 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
562     when the subject happened to end in the byte 0x85 (e.g. if the last
563     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
564     characters but of course it shouldn't be taken as a newline when it is part
565     of another character. The bug was that, for an unlimited repeat of . in
566     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
567     characters when looking for a newline.
568 ph10 197
569     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
571     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
572     in debug output.
574 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
575     long printing in the pcrecpp unittest when running under MinGW.
576 ph10 189
577 ph10 197 6. ESC_K was left out of the EBCDIC table.
578 ph10 189
579 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
580     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
581     limit also applies to "virtual nesting" when a pattern is recursive, and in
582     this case 1000 isn't so big. I have been able to remove this limit at the
583     expense of backing off one optimization in certain circumstances. Normally,
584     when pcre_exec() would call its internal match() function recursively and
585     immediately return the result unconditionally, it uses a "tail recursion"
586     feature to save stack. However, when a subpattern that can match an empty
587     string has an unlimited repetition quantifier, it no longer makes this
588     optimization. That gives it a stack frame in which to save the data for
589     checking that an empty string has been matched. Previously this was taken
590     from the 1000-entry workspace that had been reserved. So now there is no
591     explicit limit, but more stack is used.
592 ph10 189
593 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
594     syntax that is required for Windows, and which was going wrong for the
595     pcreposix and pcrecpp parts of the library. These were overlooked when this
596     problem was solved for the main library.
597 ph10 197
598 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
599     the size of patterns that contain repeated groups with explicit upper
600     limits. As the maximum quantifier is 65535, the maximum group length was
601     set at 30,000 so that the product of these two numbers did not overflow a
602     32-bit integer. However, it turns out that people want to use groups that
603     are longer than 30,000 bytes (though not repeat them that many times).
604     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
605     made it possible to implement the integer overflow checks in a much more
606     dynamic way, which I have now done. The artificial limitation on group
607     length has been removed - we now have only the limit on the total length of
608     the compiled pattern, which depends on the LINK_SIZE setting.
609 ph10 208
610     10. Fixed a bug in the documentation for get/copy named substring when
611     duplicate names are permitted. If none of the named substrings are set, the
612     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
613     empty string.
615     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
616     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
617     because the ] is interpreted as the first data character and the
618     terminating ] is not found. PCRE has been made compatible with Perl in this
619     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
620     cause memory overwriting.
622 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
623     string has been matched (to stop an infinite loop). It was not recognizing
624 ph10 208 a conditional subpattern that could match an empty string if that
625 ph10 206 subpattern was within another subpattern. For example, it looped when
626 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
627 ph10 206 condition was not nested. This bug has been fixed.
628 ph10 208
629 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
630     past the start of the subject in the presence of bytes with the top bit
631     set, for example "\x8aBCD".
632 ph10 211
633 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
634     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
635 ph10 200
636 ph10 211 14. Optimized (?!) to (*FAIL).
637 ph10 202
638 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
639     This restricts code points to be within the range 0 to 0x10FFFF, excluding
640     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
641     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
642     does: it's just the validity check that is more restrictive.
643 ph10 220
644     16. Inserted checks for integer overflows during escape sequence (backslash)
645     processing, and also fixed erroneous offset values for syntax errors during
646     backslash processing.
648 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
649 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
651 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
652     caused an overrun.
653 ph10 220
654     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
655     something other than just ASCII characters) inside a group that had an
656     unlimited repeat caused a loop at compile time (while checking to see
657     whether the group could match an empty string).
659 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
660     [\P{Any}] did so. (Error in the code for printing property names.)
661 ph10 210
662 ph10 220 21. An orphan \E inside a character class could cause a crash.
664     22. A repeated capturing bracket such as (A)? could cause a wild memory
665 ph10 218 reference during compilation.
666 ph10 220
667     23. There are several functions in pcre_compile() that scan along a compiled
668     expression for various reasons (e.g. to see if it's fixed length for look
669 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
670 ph10 220 present in the pattern. These operators have additional parameters compared
671     with \d, etc, and these were not being taken into account when moving along
672 ph10 218 the compiled data. Specifically:
673 ph10 220
674     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
675     length.
677     (b) An item such as \pL+ within a repeated group could cause crashes or
678 ph10 218 loops.
679 ph10 220
680 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
681 ph10 220 "reference to non-existent subpattern" error.
683 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
685 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
686 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
687 ph10 211
688 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
689 ph10 218
690 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
691 ph10 223 character were causing crashes (broken optimization).
692 ph10 227
693 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
694 ph10 227 \p or \P) caused a compile-time loop.
696 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
697     break. For example, the unstudied pattern [\r\n]A does not match the string
698     "\r\nA" because change 7.0/46 below moves the current point on by two
699     characters after failing to match at the start. However, the pattern \nA
700 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
701     the same is true. There doesn't seem any very clean way out of this, but
702     what I have chosen to do makes the common cases work: PCRE now takes note
703     of whether there can be an explicit match for \r or \n anywhere in the
704     pattern, and if so, 7.0/46 no longer applies. As part of this change,
705     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
706     pattern has explicit CR or LF references.
707 ph10 222
708 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
709 ph10 223
710 ph10 227
711 ph10 189 Version 7.2 19-Jun-07
712     ---------------------
714 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
715     which is apparently normally available under Windows.
716 ph10 159
717     2. Re-jig the pcregrep tests with different newline settings in an attempt
718     to make them independent of the local environment's newline setting.
719 ph10 157
720 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
721 ph10 165
722 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
723     was not 2, because the output contained actual offsets. The recent new
724 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
725     usable with all link sizes.
727 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
728     stack recursion. This gives a massive performance boost under BSD, but just
729 ph10 165 a small improvement under Linux. However, it saves one field in the frame
730 ph10 164 in all cases.
731 ph10 172
732 ph10 166 6. Added more features from the forthcoming Perl 5.10:
733 ph10 172
734 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
735     recursion call. It refers to the nth most recently opened parentheses.
736 ph10 172
737 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
738 ph10 172 to be opened parentheses.
740     (c) Conditions that refer to capturing parentheses can be specified
741 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
742 ph10 172
743 ph10 168 (d) \K resets the start of the current match so that everything before
744 ph10 172 is not part of it.
746 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
747 ph10 172
748 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
749 ph10 172 reference syntax.
750 ph10 182
751 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
752 ph10 182 alternative starts with the same number.
753 ph10 172
754 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
756 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
759     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
760     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
761 ph10 170 for detecting groups that can match an empty string.
762 ph10 159
763 ph10 172 9. A pattern with a very large number of alternatives (more than several
764     hundred) was running out of internal workspace during the pre-compile
765     phase, where pcre_compile() figures out how much memory will be needed. A
766     bit of new cunning has reduced the workspace needed for groups with
767     alternatives. The 1000-alternative test pattern now uses 12 bytes of
768     workspace instead of running out of the 4096 that are available.
769 ph10 182
770 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
771 ph10 172
772 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
773     The report of the bug said:
774 ph10 182
775 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
776     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
777 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
778 ph10 185
779 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
780 ph10 185 it matched the wrong number of bytes.
781 ph10 172
782 ph10 179
783 ph10 155 Version 7.1 24-Apr-07
784 ph10 98 ---------------------
786 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
787 ph10 122 that is more "standard", making use of automake and other Autotools. There
788 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
789     on this.
790 ph10 111
791     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
792     for recursive directory scanning broke on some systems because the files
793     are not scanned in any specific order and on different systems the order
794     was different. A call to "sort" has been inserted into RunGrepTest for the
795     approprate test as a short-term fix. In the longer term there may be an
796 ph10 100 alternative.
797 ph10 111
798 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
799 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
800     people do their own conversions for various reasons). The problems
801     concerned the use of low-level troff macros .br and .in. I have therefore
802     removed all such uses from the man pages (some were redundant, some could
803 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
804     HTML has been updated to handle .nf/.fi and to complain if it encounters
805     .br or .in.
806 ph10 111
807 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
808 ph10 123 arranged for config.h to be included in the distribution, with the name
809 ph10 111 config.h.generic, for the benefit of those who have to compile without
810     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
812     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
813     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
814 ph10 127 makevp.bat for the current PCRE, using the additional files
815 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
816 ph10 111
817     6. A Windows user reported a minor discrepancy with test 2, which turned out
818     to be caused by a trailing space on an input line that had got lost in his
819 ph10 102 copy. The trailing space was an accident, so I've just removed it.
820 ph10 111
821 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
822 ph10 111 that is needed.
824 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
825     as "const" (a) because they are and (b) because it helps the PHP
826     maintainers who have recently made a script to detect big data structures
827 ph10 111 in the php code that should be moved to the .rodata section. I remembered
828     to update Builducptable as well, so it won't revert if ucptable.h is ever
829 ph10 105 re-created.
830 ph10 111
831     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
832 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
833 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
834     support is not required. This saves 1.5-2K of code, which is important in
835 ph10 107 some applications.
836 ph10 111
837 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
838 ph10 111 so as not to refer to the tables, even though these functions will never be
839     called when UTF-8 support is disabled. Otherwise there are problems with a
840     shared library.
842 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
844     (a) It was defining its arguments as char * instead of void *.
846 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
847     a long time ago when I wrote it, but is no longer the case.
849 ph10 118 The emulated memove() is provided for those environments that have neither
850 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
851 ph10 118 is clearly not the case, as these two bugs were recently reported.
852 ph10 123
853 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
854 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
855     pages, and it removes trailing spaces from listed files. It also creates
856     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
857     case, it wraps all the #defines with #ifndefs. This script should be run
858 ph10 111 before "make dist".
859 ph10 123
860 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
861     with Unicode property support.
862 ph10 123
863     (a) For a maximizing quantifier, if the two different cases of the
864     character were of different lengths in their UTF-8 codings (there are
865     some cases like this - I found 11), and the matching function had to
866 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
867     were both the same length.
868 ph10 123
869     (b) When PCRE was configured to use the heap rather than the stack for
870     recursion during matching, it was not correctly preserving the data for
871     the other case of a UTF-8 character when checking ahead for a match
872     while processing a minimizing repeat. If the check also involved
873 ph10 115 matching a wide character, but failed, corruption could cause an
874     erroneous result when trying to check for a repeat of the original
875     character.
876 ph10 123
877 ph10 116 13. Some tidying changes to the testing mechanism:
878 ph10 98
879 ph10 116 (a) The RunTest script now detects the internal link size and whether there
880     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
881 ph10 123 values substituted by "configure". (The RunGrepTest script already did
882     this for UTF-8.) The configure.ac script no longer substitutes the
883     relevant variables.
885 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
886     with length and offset values. This means that the output is different
887     for different internal link sizes. Test 2 is skipped for link sizes
888     other than 2 because of this, bypassing the problem. Unfortunately,
889     there was also a test in test 3 (the locale tests) that used /B and
890 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
891     I have added a new /Z option to pcretest that replaces the length and
892     offset values with spaces. This is now used to make test 3 independent
893 ph10 122 of link size. (Test 2 will be tidied up later.)
894 ph10 123
895     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
896 ph10 122 segmentation fault instead of returning the appropriate error message.
897 ph10 134
898 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
899 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
900     This doesn't seem right; it now treats the CRLF combination as the line
901     ending, and so does not match in that case. It's only a pattern such as ^$
902     that would hit this one: something like ^ABC$ would have failed after \r
903     and then tried again after \r\n.
905 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
906 ph10 134 in an attempt to make files that differ only in their line terminators
907     compare equal. This works on Linux.
908 ph10 142
909 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
910     as it exited. This is now fixed, thanks to valgrind.
911 ph10 142
912 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
913 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
914     was because its rules for how to advance for /g after matching an empty
915 ph10 143 string at the end of a line did not allow for this case. They now check for
916     it specially.
917 ph10 150
918     20. pcretest is supposed to handle patterns and data of any length, by
919     extending its buffers when necessary. It was getting this wrong when the
920 ph10 147 buffer for a data line had to be extended.
921 ph10 150
922 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
923 ph10 150 CRLF as a newline sequence.
924 ph10 152
925 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
926 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
927     I have nevertheless tidied it up.
928 ph10 111
929 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
930 ph10 123
931 ph10 153 24. Added a man page for pcre-config.
932 ph10 152
933 ph10 153
934 nigel 93 Version 7.0 19-Dec-06
935     ---------------------
937     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
938     moving to gcc 4.1.1.
940     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
941     sys/time.h, which is documented as needed for this function. It doesn't
942     seem to matter on Linux, but it showed up on some releases of OS X.
944     3. It seems that there are systems where bytes whose values are greater than
945     127 match isprint() in the "C" locale. The "C" locale should be the
946     default when a C program starts up. In most systems, only ASCII printing
947     characters match isprint(). This difference caused the output from pcretest
948     to vary, making some of the tests fail. I have changed pcretest so that:
950     (a) When it is outputting text in the compiled version of a pattern, bytes
951     other than 32-126 are always shown as hex escapes.
953     (b) When it is outputting text that is a matched part of a subject string,
954     it does the same, unless a different locale has been set for the match
955     (using the /L modifier). In this case, it uses isprint() to decide.
957     4. Fixed a major bug that caused incorrect computation of the amount of memory
958     required for a compiled pattern when options that changed within the
959     pattern affected the logic of the preliminary scan that determines the
960     length. The relevant options are -x, and -i in UTF-8 mode. The result was
961     that the computed length was too small. The symptoms of this bug were
962     either the PCRE error "internal error: code overflow" from pcre_compile(),
963     or a glibc crash with a message such as "pcretest: free(): invalid next
964     size (fast)". Examples of patterns that provoked this bug (shown in
965     pcretest format) are:
967     /(?-x: )/x
968     /(?x)(?-x: \s*#\s*)/
969     /((?i)[\x{c0}])/8
970     /(?i:[\x{c0}])/8
972     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
973     is now done differently.
975     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
976     wrapper classes; (b) implement a new function in the C++ scanner that is
977     more efficient than the old way of doing things because it avoids levels of
978     recursion in the regex matching; (c) add a paragraph to the documentation
979     for the FullMatch() function.
981     6. The escape sequence \n was being treated as whatever was defined as
982     "newline". Not only was this contrary to the documentation, which states
983     that \n is character 10 (hex 0A), but it also went horribly wrong when
984     "newline" was defined as CRLF. This has been fixed.
986     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
987     was being set to -1 for the "end of line" case (supposedly a value that no
988     character can have). Though this value is never used (the check for end of
989     line is "zero bytes in current character"), it caused compiler complaints.
990     I've changed it to 0xffffffff.
992     8. In pcre_version.c, the version string was being built by a sequence of
993     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
994     string (as it is for production releases) called a macro with an empty
995     argument. The C standard says the result of this is undefined. The gcc
996     compiler treats it as an empty string (which was what was wanted) but it is
997     reported that Visual C gives an error. The source has been hacked around to
998     avoid this problem.
1000     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
1001     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
1002     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
1003     of them did).
1005     10. Originally, pcretest opened its input and output without "b"; then I was
1006     told that "b" was needed in some environments, so it was added for release
1007     5.0 to both the input and output. (It makes no difference on Unix-like
1008     systems.) Later I was told that it is wrong for the input on Windows. I've
1009     now abstracted the modes into two macros, to make it easier to fiddle with
1010     them, and removed "b" from the input mode under Windows.
1012     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
1014     12. Added -help and --help to pcretest as an official way of being reminded
1015     of the options.
1017     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
1018     and pcrecpp.cc because they annoy compilers at high warning levels.
1020     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
1022     15. Fixed an occurrence of == in configure.ac that should have been = (shell
1023     scripts are not C programs :-) and which was not noticed because it works
1024     on Linux.
1026     16. pcretest is supposed to handle any length of pattern and data line (as one
1027     line or as a continued sequence of lines) by extending its input buffer if
1028     necessary. This feature was broken for very long pattern lines, leading to
1029     a string of junk being passed to pcre_compile() if the pattern was longer
1030     than about 50K.
1032     17. I have done a major re-factoring of the way pcre_compile() computes the
1033     amount of memory needed for a compiled pattern. Previously, there was code
1034     that made a preliminary scan of the pattern in order to do this. That was
1035     OK when PCRE was new, but as the facilities have expanded, it has become
1036     harder and harder to keep it in step with the real compile phase, and there
1037     have been a number of bugs (see for example, 4 above). I have now found a
1038     cunning way of running the real compile function in a "fake" mode that
1039     enables it to compute how much memory it would need, while actually only
1040     ever using a few hundred bytes of working memory and without too many
1041     tests of the mode. This should make future maintenance and development
1042     easier. A side effect of this work is that the limit of 200 on the nesting
1043     depth of parentheses has been removed (though this was never a serious
1044     limitation, I suspect). However, there is a downside: pcre_compile() now
1045     runs more slowly than before (30% or more, depending on the pattern). I
1046     hope this isn't a big issue. There is no effect on runtime performance.
1048     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
1049     newline (only possible for the last line of a file) and it was a
1050     pattern that set a locale (followed by /Lsomething), pcretest crashed.
1052     19. Added additional timing features to pcretest. (1) The -tm option now times
1053     matching only, not compiling. (2) Both -t and -tm can be followed, as a
1054     separate command line item, by a number that specifies the number of
1055     repeats to use when timing. The default is 50000; this gives better
1056     precision, but takes uncomfortably long for very large patterns.
1058     20. Extended pcre_study() to be more clever in cases where a branch of a
1059     subpattern has no definite first character. For example, (a*|b*)[cd] would
1060     previously give no result from pcre_study(). Now it recognizes that the
1061     first character must be a, b, c, or d.
1063     21. There was an incorrect error "recursive call could loop indefinitely" if
1064     a subpattern (or the entire pattern) that was being tested for matching an
1065     empty string contained only one non-empty item after a nested subpattern.
1066     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
1067     incorrectly, because the \d was being skipped in the check.
1069     22. The pcretest program now has a new pattern option /B and a command line
1070     option -b, which is equivalent to adding /B to every pattern. This causes
1071     it to show the compiled bytecode, without the additional information that
1072     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
1073     is the same as /B/I).
1075     23. A new optimization is now able automatically to treat some sequences such
1076     as a*b as a*+b. More specifically, if something simple (such as a character
1077     or a simple class like \d) has an unlimited quantifier, and is followed by
1078     something that cannot possibly match the quantified thing, the quantifier
1079     is automatically "possessified".
1081     24. A recursive reference to a subpattern whose number was greater than 39
1082     went wrong under certain circumstances in UTF-8 mode. This bug could also
1083     have affected the operation of pcre_study().
1085     25. Realized that a little bit of performance could be had by replacing
1086     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
1088     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
1090     27. Possessive quantifiers such as a++ were previously implemented by turning
1091     them into atomic groups such as ($>a+). Now they have their own opcodes,
1092     which improves performance. This includes the automatically created ones
1093     from 23 above.
1095     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
1096     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
1097     the first matched character to be a colon. This applied both to named and
1098     numbered groups.
1100     29. The ucpinternal.h header file was missing its idempotency #ifdef.
1102     30. I was sent a "project" file called libpcre.a.dev which I understand makes
1103     building PCRE on Windows easier, so I have included it in the distribution.
1105     31. There is now a check in pcretest against a ridiculously large number being
1106     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
1107     loop, the loop is abandoned.
1109     32. Forward references to subpatterns in conditions such as (?(2)...) where
1110     subpattern 2 is defined later cause pcre_compile() to search forwards in
1111     the pattern for the relevant set of parentheses. This search went wrong
1112     when there were unescaped parentheses in a character class, parentheses
1113     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
1115     33. "Subroutine" calls and backreferences were previously restricted to
1116     referencing subpatterns earlier in the regex. This restriction has now
1117     been removed.
1119     34. Added a number of extra features that are going to be in Perl 5.10. On the
1120     whole, these are just syntactic alternatives for features that PCRE had
1121     previously implemented using the Python syntax or my own invention. The
1122     other formats are all retained for compatibility.
1124     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1125     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1126     also .NET compatible.
1128     (b) A recursion or subroutine call to a named group can now be defined as
1129     (?&name) as well as (?P>name).
1131     (c) A backreference to a named group can now be defined as \k<name> or
1132     \k'name' as well as (?P=name). The new forms, as well as being in Perl
1133     5.10, are also .NET compatible.
1135     (d) A conditional reference to a named group can now use the syntax
1136     (?(<name>) or (?('name') as well as (?(name).
1138     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1139     groups (named and numbered) that are never evaluated inline, but can be
1140     called as "subroutines" from elsewhere. In effect, the DEFINE condition
1141     is always false. There may be only one alternative in such a group.
1143     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1144     as the simple (?(R). The condition is true only if the most recent
1145     recursion is that of the given number or name. It does not search out
1146     through the entire recursion stack.
1148     (g) The escape \gN or \g{N} has been added, where N is a positive or
1149     negative number, specifying an absolute or relative reference.
1151     35. Tidied to get rid of some further signed/unsigned compiler warnings and
1152     some "unreachable code" warnings.
1154     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1155     things, this adds five new scripts.
1157     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1158     There were also incompatibilities regarding the handling of \Q..\E inside
1159     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1160     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1162     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1163     matches an empty string, and forcibly breaks the loop. There were bugs in
1164     this code in non-simple cases. For a pattern such as ^(a()*)* matched
1165     against aaaa the result was just "a" rather than "aaaa", for example. Two
1166     separate and independent bugs (that affected different cases) have been
1167     fixed.
1169     39. Refactored the code to abolish the use of different opcodes for small
1170     capturing bracket numbers. This is a tidy that I avoided doing when I
1171     removed the limit on the number of capturing brackets for 3.5 back in 2001.
1172     The new approach is not only tidier, it makes it possible to reduce the
1173     memory needed to fix the previous bug (38).
1175     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1176     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1177     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1178     mode.
1180     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1181     report.
1183     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1184     copy construction and assignment in the C++ wrapper.
1186     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1187     couple of bugs that could have given wrong results in the "--newline=crlf"
1188     case.
1190     44. Added a number of casts and did some reorganization of signed/unsigned int
1191     variables following suggestions from Dair Grant. Also renamed the variable
1192     "this" as "item" because it is a C++ keyword.
1194     45. Arranged for dftables to add
1196     #include "pcre_internal.h"
1198     to pcre_chartables.c because without it, gcc 4.x may remove the array
1199     definition from the final binary if PCRE is built into a static library and
1200     dead code stripping is activated.
1202     46. For an unanchored pattern, if a match attempt fails at the start of a
1203     newline sequence, and the newline setting is CRLF or ANY, and the next two
1204     characters are CRLF, advance by two characters instead of one.
1207 nigel 91 Version 6.7 04-Jul-06
1208     ---------------------
1210     1. In order to handle tests when input lines are enormously long, pcretest has
1211     been re-factored so that it automatically extends its buffers when
1212     necessary. The code is crude, but this _is_ just a test program. The
1213     default size has been increased from 32K to 50K.
1215     2. The code in pcre_study() was using the value of the re argument before
1216     testing it for NULL. (Of course, in any sensible call of the function, it
1217     won't be NULL.)
1219     3. The memmove() emulation function in pcre_internal.h, which is used on
1220     systems that lack both memmove() and bcopy() - that is, hardly ever -
1221     was missing a "static" storage class specifier.
1223     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1224     containing an extended class (one that cannot be represented by a bitmap
1225     because it contains high-valued characters or Unicode property items, e.g.
1226     [\pZ]). Almost always one would set UTF-8 mode when processing such a
1227     pattern, but PCRE should not loop if you do not (it no longer does).
1228     [Detail: two cases were found: (a) a repeated subpattern containing an
1229     extended class; (b) a recursive reference to a subpattern that followed a
1230     previous extended class. It wasn't skipping over the extended class
1231     correctly when UTF-8 mode was not set.]
1233     5. A negated single-character class was not being recognized as fixed-length
1234     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1235     compile error "lookbehind assertion is not fixed length".
1237     6. The RunPerlTest auxiliary script was showing an unexpected difference
1238     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1239     write a Perl script that can interpret lines of an input file either as
1240     byte characters or as UTF-8, which is what "perltest" was being required to
1241     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1242     can't do is switch easily at run time between having the "use utf8;" pragma
1243     or not. In the end, I fudged it by using the RunPerlTest script to insert
1244     "use utf8;" explicitly for the UTF-8 tests.
1246     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1247     the end of the subject string, contrary to the documentation and to what
1248     Perl does. This was true of both matching functions. Now it matches only at
1249     the start of the subject and immediately after *internal* newlines.
1251     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1252     a pointer to an int instead of a pointer to an unsigned long int. This
1253     caused problems on 64-bit systems.
1255     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1256     instance of the 'standard' template library not being so standard".
1258     10. There was no check on the number of named subpatterns nor the maximum
1259     length of a subpattern name. The product of these values is used to compute
1260     the size of the memory block for a compiled pattern. By supplying a very
1261     long subpattern name and a large number of named subpatterns, the size
1262     computation could be caused to overflow. This is now prevented by limiting
1263     the length of names to 32 characters, and the number of named subpatterns
1264     to 10,000.
1266     11. Subpatterns that are repeated with specific counts have to be replicated in
1267     the compiled pattern. The size of memory for this was computed from the
1268     length of the subpattern and the repeat count. The latter is limited to
1269     65535, but there was no limit on the former, meaning that integer overflow
1270     could in principle occur. The compiled length of a repeated subpattern is
1271     now limited to 30,000 bytes in order to prevent this.
1273     12. Added the optional facility to have named substrings with the same name.
1275     13. Added the ability to use a named substring as a condition, using the
1276     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1277     are numbers (not recommended). Forward references are permitted.
1279     14. Added forward references in named backreferences (if you see what I mean).
1281     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1282     pattern could run off the end of the subject. For example, the pattern
1283     "(?s)(.{1,5})"8 did this with the subject "ab".
1285     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1286     PCRE_CASELESS was set when matching characters that were quantified with ?
1287     or *.
1289     17. A character class other than a single negated character that had a minimum
1290     but no maximum quantifier - for example [ab]{6,} - was not handled
1291     correctly by pce_dfa_exec(). It would match only one character.
1293     18. A valid (though odd) pattern that looked like a POSIX character
1294     class but used an invalid character after [ (for example [[,abc,]]) caused
1295     pcre_compile() to give the error "Failed: internal error: code overflow" or
1296     in some cases to crash with a glibc free() error. This could even happen if
1297     the pattern terminated after [[ but there just happened to be a sequence of
1298     letters, a binary zero, and a closing ] in the memory that followed.
1300     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1301     over the years. Originally (before any Unicode support), just the bottom 8
1302     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1303     output from "man perlunicode" includes this:
1305     The regular expression compiler produces polymorphic opcodes. That
1306     is, the pattern adapts to the data and automatically switches to
1307     the Unicode character scheme when presented with Unicode data--or
1308     instead uses a traditional byte scheme when presented with byte
1309     data.
1311     Sadly, a wide octal escape does not cause a switch, and in a string with
1312     no other multibyte characters, these octal escapes are treated as before.
1313     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1314     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1315     Unicode string.
1317     I have not perpetrated such confusion in PCRE. Up till now, it took just
1318     the bottom 8 bits, as in old Perl. I have now made octal escapes with
1319     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1320     translate to the appropriate multibyte character.
1322     29. Applied some refactoring to reduce the number of warnings from Microsoft
1323     and Borland compilers. This has included removing the fudge introduced
1324     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1325     a warning about an unused variable.
1327     21. PCRE has not included VT (character 0x0b) in the set of whitespace
1328     characters since release 4.0, because Perl (from release 5.004) does not.
1329     [Or at least, is documented not to: some releases seem to be in conflict
1330     with the documentation.] However, when a pattern was studied with
1331     pcre_study() and all its branches started with \s, PCRE still included VT
1332     as a possible starting character. Of course, this did no harm; it just
1333     caused an unnecessary match attempt.
1335     22. Removed a now-redundant internal flag bit that recorded the fact that case
1336     dependency changed within the pattern. This was once needed for "required
1337     byte" processing, but is no longer used. This recovers a now-scarce options
1338     bit. Also moved the least significant internal flag bit to the most-
1339     significant bit of the word, which was not previously used (hangover from
1340     the days when it was an int rather than a uint) to free up another bit for
1341     the future.
1343     23. Added support for CRLF line endings as well as CR and LF. As well as the
1344     default being selectable at build time, it can now be changed at runtime
1345     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1346     specify that it is scanning data with non-default line endings.
1348     24. Changed the definition of CXXLINK to make it agree with the definition of
1349     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1351     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1352     recursions. This makes a big different to stack usage for some patterns.
1354     26. If a subpattern containing a named recursion or subroutine reference such
1355     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1356     the space required for the compiled pattern went wrong and gave too small a
1357     value. Depending on the environment, this could lead to "Failed: internal
1358     error: code overflow at offset 49" or "glibc detected double free or
1359     corruption" errors.
1361     27. Applied patches from Google (a) to support the new newline modes and (b) to
1362     advance over multibyte UTF-8 characters in GlobalReplace.
1364     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1365     difference for some implementation of PCRE in some Windows version.
1367     29. Added some extra testing facilities to pcretest:
1369     \q<number> in a data line sets the "match limit" value
1370     \Q<number> in a data line sets the "match recursion limt" value
1371     -S <number> sets the stack size, where <number> is in megabytes
1373     The -S option isn't available for Windows.
1376 nigel 89 Version 6.6 06-Feb-06
1377     ---------------------
1379     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1380     in pcreposix.h. I have copied the definition from pcre.h.
1382     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1383     because pcre.h is no longer a built file.
1385     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1386     not normally included in the compiled code.
1389 nigel 87 Version 6.5 01-Feb-06
1390     ---------------------
1392     1. When using the partial match feature with pcre_dfa_exec(), it was not
1393     anchoring the second and subsequent partial matches at the new starting
1394     point. This could lead to incorrect results. For example, with the pattern
1395     /1234/, partially matching against "123" and then "a4" gave a match.
1397     2. Changes to pcregrep:
1399     (a) All non-match returns from pcre_exec() were being treated as failures
1400     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1401     error message is output. Some extra information is given for the
1403     probably the only errors that are likely to be caused by users (by
1404     specifying a regex that has nested indefinite repeats, for instance).
1405     If there are more than 20 of these errors, pcregrep is abandoned.
1407     (b) A binary zero was treated as data while matching, but terminated the
1408     output line if it was written out. This has been fixed: binary zeroes
1409     are now no different to any other data bytes.
1411     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1412     used to set a locale for matching. The --locale=xxxx long option has
1413     been added (no short equivalent) to specify a locale explicitly on the
1414     pcregrep command, overriding the environment variables.
1416     (d) When -B was used with -n, some line numbers in the output were one less
1417     than they should have been.
1419     (e) Added the -o (--only-matching) option.
1421     (f) If -A or -C was used with -c (count only), some lines of context were
1422     accidentally printed for the final match.
1424     (g) Added the -H (--with-filename) option.
1426     (h) The combination of options -rh failed to suppress file names for files
1427     that were found from directory arguments.
1429     (i) Added the -D (--devices) and -d (--directories) options.
1431     (j) Added the -F (--fixed-strings) option.
1433     (k) Allow "-" to be used as a file name for -f as well as for a data file.
1435     (l) Added the --colo(u)r option.
1437     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1438     is not present by default.
1440     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1441     items such as (?R) or (?1), when the recursion could match a number of
1442     alternatives. If it matched one of the alternatives, but subsequently,
1443     outside the recursion, there was a failure, the code tried to back up into
1444     the recursion. However, because of the way PCRE is implemented, this is not
1445     possible, and the result was an incorrect result from the match.
1447     In order to prevent this happening, the specification of recursion has
1448     been changed so that all such subpatterns are automatically treated as
1449     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1451     4. I had overlooked the fact that, in some locales, there are characters for
1452     which isalpha() is true but neither isupper() nor islower() are true. In
1453     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1454     and ordfeminine) are like this. This affected the treatment of \w and \W
1455     when they appeared in character classes, but not when they appeared outside
1456     a character class. The bit map for "word" characters is now created
1457     separately from the results of isalnum() instead of just taking it from the
1458     upper, lower, and digit maps. (Plus the underscore character, of course.)
1460     5. The above bug also affected the handling of POSIX character classes such as
1461     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1462     permanent tables. Instead, the bit maps for such a class were previously
1463     created as the appropriate unions of the upper, lower, and digit bitmaps.
1464     Now they are created by subtraction from the [[:word:]] class, which has
1465     its own bitmap.
1467     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1468     It is created by subtracting the vertical space characters (\x09, \x0a,
1469     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1470     subtraction was done in the overall bitmap for a character class, meaning
1471     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1472     be recognized. This bug has been fixed.
1474     7. Patches from the folks at Google:
1476     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1477     real life, but is still worth protecting against".
1479     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1480     regular expressions".
1482     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1483     have it.
1485     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1486     "configure" and the latter not, in order to fix a problem somebody had
1487     with compiling the Arg class on HP-UX.
1489     (e) Improve the error-handling of the C++ wrapper a little bit.
1491     (f) New tests for checking recursion limiting.
1493     8. The pcre_memmove() function, which is used only if the environment does not
1494     have a standard memmove() function (and is therefore rarely compiled),
1495     contained two bugs: (a) use of int instead of size_t, and (b) it was not
1496     returning a result (though PCRE never actually uses the result).
1498     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1499     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1500     returned instead of calling malloc() with an overflowing number that would
1501     most likely cause subsequent chaos.
1503     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1505     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1506     with this option is matched, the nmatch and pmatch options of regexec() are
1507     ignored.
1509     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1510     provided in case anyone wants to the the POSIX interface with UTF-8
1511     strings.
1513     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1514     C++ linking (needed for some HP-UX environments).
1516     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1517     (unused parameter) and in the pcre_printint() function (omitted "default"
1518     switch label when the default is to do nothing).
1520     15. Added some code to make it possible, when PCRE is compiled as a C++
1521     library, to replace subject pointers for pcre_exec() with a smart pointer
1522     class, thus making it possible to process discontinuous strings.
1524     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1525     much the same function. They were added by different people who were trying
1526     to make PCRE easy to compile on non-Unix systems. It has been suggested
1527     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1528     for compiling on Windows systems. I have therefore replaced it with
1529     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1530     defaults to "extern" for C or "extern C" for C++, which works fine on
1531     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1532     SCOPE with something explicit in config.h. In addition:
1534     (a) pcreposix.h still had just "extern" instead of either of these macros;
1535     I have replaced it with PCRE_DATA_SCOPE.
1537     (b) Functions such as _pcre_xclass(), which are internal to the library,
1538     but external in the C sense, all had PCRE_EXPORT in their definitions.
1539     This is apparently wrong for the Windows case, so I have removed it.
1540     (It makes no difference on Unix-like systems.)
1542     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1543     of recursive calls to match(). This is different to MATCH_LIMIT because
1544     that limits the total number of calls to match(), not all of which increase
1545     the depth of recursion. Limiting the recursion depth limits the amount of
1546     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1547     when PCRE is compiled, and changed at run time. A patch from Google adds
1548     this functionality to the C++ interface.
1550     18. Changes to the handling of Unicode character properties:
1552     (a) Updated the table to Unicode 4.1.0.
1554     (b) Recognize characters that are not in the table as "Cn" (undefined).
1556     (c) I revised the way the table is implemented to a much improved format
1557     which includes recognition of ranges. It now supports the ranges that
1558     are defined in UnicodeData.txt, and it also amalgamates other
1559     characters into ranges. This has reduced the number of entries in the
1560     table from around 16,000 to around 3,000, thus reducing its size
1561     considerably. I realized I did not need to use a tree structure after
1562     all - a binary chop search is just as efficient. Having reduced the
1563     number of entries, I extended their size from 6 bytes to 8 bytes to
1564     allow for more data.
1566     (d) Added support for Unicode script names via properties such as \p{Han}.
1568     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1569     matching that character.
1571     20. When matching a repeated Unicode property with a minimum greater than zero,
1572     (for example \pL{2,}), PCRE could look past the end of the subject if it
1573     reached it while seeking the minimum number of characters. This could
1574     happen only if some of the characters were more than one byte long, because
1575     there is a check for at least the minimum number of bytes.
1577     21. Refactored the implementation of \p and \P so as to be more general, to
1578     allow for more different types of property in future. This has changed the
1579     compiled form incompatibly. Anybody with saved compiled patterns that use
1580     \p or \P will have to recompile them.
1582     22. Added "Any" and "L&" to the supported property types.
1584     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1585     but give a compile time error if the value is greater than 0xff.
1587     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1588     accidentally not being installed or uninstalled.
1590     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1591     made were to insert the current release number. This seemed silly, because
1592     it made things harder for people building PCRE on systems that don't run
1593     "configure". I have turned pcre.h into a distributed file, no longer built
1594     by "configure", with the version identification directly included. There is
1595     no longer a pcre.h.in file.
1597     However, this change necessitated a change to the pcre-config script as
1598     well. It is built from pcre-config.in, and one of the substitutions was the
1599     release number. I have updated configure.ac so that ./configure now finds
1600     the release number by grepping pcre.h.
1602     26. Added the ability to run the tests under valgrind.
1605 nigel 85 Version 6.4 05-Sep-05
1606     ---------------------
1608     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1609     "--" to be printed when multiple files were scanned, even when none of the
1610     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1611     consider it to be a bug, and have restored the previous behaviour.
1613     2. A couple of code tidies to get rid of compiler warnings.
1615     3. The pcretest program used to cheat by referring to symbols in the library
1616     whose names begin with _pcre_. These are internal symbols that are not
1617     really supposed to be visible externally, and in some environments it is
1618     possible to suppress them. The cheating is now confined to including
1619     certain files from the library's source, which is a bit cleaner.
1621     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1622     file's purpose clearer.
1624     5. Reorganized pcre_ucp_findchar().
1627 nigel 83 Version 6.3 15-Aug-05
1628     ---------------------
1630     1. The file libpcre.pc.in did not have general read permission in the tarball.
1632     2. There were some problems when building without C++ support:
1634     (a) If C++ support was not built, "make install" and "make test" still
1635     tried to test it.
1637     (b) There were problems when the value of CXX was explicitly set. Some
1638     changes have been made to try to fix these, and ...
1640     (c) --disable-cpp can now be used to explicitly disable C++ support.
1642     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1643     backslash in a target when C++ was disabled. This confuses some
1644     versions of "make", apparently. Using an intermediate variable solves
1645     this. (Same for CPP_LOBJ.)
1647     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1648     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1649     necessary on certain architectures.
1651     4. Added a setting of -export-symbols-regex to the link command to remove
1652     those symbols that are exported in the C sense, but actually are local
1653     within the library, and not documented. Their names all begin with
1654     "_pcre_". This is not a perfect job, because (a) we have to except some
1655     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1656     available (and never for static libraries). I have made a note to try to
1657     find a way round (a) in the future.
1660 nigel 81 Version 6.2 01-Aug-05
1661     ---------------------
1663     1. There was no test for integer overflow of quantifier values. A construction
1664     such as {1111111111111111} would give undefined results. What is worse, if
1665     a minimum quantifier for a parenthesized subpattern overflowed and became
1666     negative, the calculation of the memory size went wrong. This could have
1667     led to memory overwriting.
1669     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1671     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1672     operating environments where this matters.
1674     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1675     PCRE options from within the C++ wrapper.
1677     5. Named capturing subpatterns were not being correctly counted when a pattern
1678     was compiled. This caused two problems: (a) If there were more than 100
1679     such subpatterns, the calculation of the memory needed for the whole
1680     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1681     back references of the form \12, where the number was greater than 9, were
1682     not recognized as back references, even though there were sufficient
1683     previous subpatterns.
1685     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1686     versions of gcc, e.g. 2.95.4.
1689 nigel 79 Version 6.1 21-Jun-05
1690     ---------------------
1692     1. There was one reference to the variable "posix" in pcretest.c that was not
1693     surrounded by "#if !defined NOPOSIX".
1695     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1696     the cross-check on the old pcre_info() function, for the benefit of the
1697     cut-down version of PCRE that is currently imported into Exim.
1699     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1700     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1701     patterns that start (?i) but I don't think that matters. The use of (?i) is
1702     just an example; this all applies to the other options as well.
1704     4. Since libtool seems to echo the compile commands it is issuing, the output
1705     from "make" can be reduced a bit by putting "@" in front of each libtool
1706     compile command.
1708     5. Patch from the folks at Google for configure.in to be a bit more thorough
1709     in checking for a suitable C++ installation before trying to compile the
1710     C++ stuff. This should fix a reported problem when a compiler was present,
1711     but no suitable headers.
1713     6. The man pages all had just "PCRE" as their title. I have changed them to
1714     be the relevant file name. I have also arranged that these names are
1715     retained in the file doc/pcre.txt, which is a concatenation in text format
1716     of all the man pages except the little individual ones for each function.
1718     7. The NON-UNIX-USE file had not been updated for the different set of source
1719     files that come with release 6. I also added a few comments about the C++
1720     wrapper.
1723 nigel 77 Version 6.0 07-Jun-05
1724     ---------------------
1726     1. Some minor internal re-organization to help with my DFA experiments.
1728     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1729     didn't matter for the library itself when fully configured, but did matter
1730     when compiling without UCP support, or within Exim, where the ucp files are
1731     not imported.
1733     3. Refactoring of the library code to split up the various functions into
1734     different source modules. The addition of the new DFA matching code (see
1735     below) to a single monolithic source would have made it really too
1736     unwieldy, quite apart from causing all the code to be include in a
1737     statically linked application, when only some functions are used. This is
1738     relevant even without the DFA addition now that patterns can be compiled in
1739     one application and matched in another.
1741     The downside of splitting up is that there have to be some external
1742     functions and data tables that are used internally in different modules of
1743     the library but which are not part of the API. These have all had their
1744     names changed to start with "_pcre_" so that they are unlikely to clash
1745     with other external names.
1747     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1748     a different (DFA) algorithm. Although it is slower than the original
1749     function, it does have some advantages for certain types of matching
1750     problem.
1752     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1753     including restarting after a partial match.
1755     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1756     defined when compiling for Windows was sent to me. I have put it into the
1757     code, though I have no means of testing or verifying it.
1759     7. Added the pcre_refcount() auxiliary function.
1761     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1762     match before or at the first newline in the subject string. In pcretest,
1763     the /f option on a pattern can be used to set this.
1765     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1766     would behave wrongly. This has been present in PCRE since release 4.0.
1768     10. A number of changes to the pcregrep command:
1770     (a) Refactored how -x works; insert ^(...)$ instead of setting
1771     PCRE_ANCHORED and checking the length, in preparation for adding
1772     something similar for -w.
1774     (b) Added the -w (match as a word) option.
1776     (c) Refactored the way lines are read and buffered so as to have more
1777     than one at a time available.
1779     (d) Implemented a pcregrep test script.
1781     (e) Added the -M (multiline match) option. This allows patterns to match
1782     over several lines of the subject. The buffering ensures that at least
1783     8K, or the rest of the document (whichever is the shorter) is available
1784     for matching (and similarly the previous 8K for lookbehind assertions).
1786     (f) Changed the --help output so that it now says
1788     -w, --word-regex(p)
1790     instead of two lines, one with "regex" and the other with "regexp"
1791     because that confused at least one person since the short forms are the
1792     same. (This required a bit of code, as the output is generated
1793     automatically from a table. It wasn't just a text change.)
1795     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1796     option but starts with a hyphen. Could be a pattern or a path name
1797     starting with a hyphen, for instance.
1799     (h) "-" can be given as a file name to represent stdin.
1801     (i) When file names are being printed, "(standard input)" is used for
1802     the standard input, for compatibility with GNU grep. Previously
1803     "<stdin>" was used.
1805     (j) The option --label=xxx can be used to supply a name to be used for
1806     stdin when file names are being printed. There is no short form.
1808     (k) Re-factored the options decoding logic because we are going to add
1809     two more options that take data. Such options can now be given in four
1810     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1812     (l) Added the -A, -B, and -C options for requesting that lines of context
1813     around matches be printed.
1815     (m) Added the -L option to print the names of files that do not contain
1816     any matching lines, that is, the complement of -l.
1818     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1819     continue to scan other files.
1821     (o) The -s option was incorrectly implemented. For compatibility with other
1822     greps, it now suppresses the error message for a non-existent or non-
1823     accessible file (but not the return code). There is a new option called
1824     -q that suppresses the output of matching lines, which was what -s was
1825     previously doing.
1827     (p) Added --include and --exclude options to specify files for inclusion
1828     and exclusion when recursing.
1830     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1831     Hopefully, it now does.
1833     12. Missing cast in pcre_study().
1835     13. Added an "uninstall" target to the makefile.
1837     14. Replaced "extern" in the function prototypes in Makefile.in with
1838     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1839     world, but is set differently for Windows.
1841     15. Added a second compiling function called pcre_compile2(). The only
1842     difference is that it has an extra argument, which is a pointer to an
1843     integer error code. When there is a compile-time failure, this is set
1844     non-zero, in addition to the error test pointer being set to point to an
1845     error message. The new argument may be NULL if no error number is required
1846     (but then you may as well call pcre_compile(), which is now just a
1847     wrapper). This facility is provided because some applications need a
1848     numeric error indication, but it has also enabled me to tidy up the way
1849     compile-time errors are handled in the POSIX wrapper.
1851     16. Added VPATH=.libs to the makefile; this should help when building with one
1852     prefix path and installing with another. (Or so I'm told by someone who
1853     knows more about this stuff than I do.)
1855     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1856     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1857     match everything, including newlines. This is not POSIX-compatible, but
1858     somebody wanted the feature. From pcretest it can be activated by using
1859     both the P and the s flags.
1861     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1863     19. libpcre.pc was being incorrectly installed as executable.
1865     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1866     it now also looks for '\r' so that it will work unmodified on Windows.
1868     21. Added Google's contributed C++ wrapper to the distribution.
1870     22. Added some untidy missing memory free() calls in pcretest, to keep
1871     Electric Fence happy when testing.
1875 nigel 75 Version 5.0 13-Sep-04
1876     ---------------------
1878     1. Internal change: literal characters are no longer packed up into items
1879     containing multiple characters in a single byte-string. Each character
1880     is now matched using a separate opcode. However, there may be more than one
1881     byte in the character in UTF-8 mode.
1883     2. The pcre_callout_block structure has two new fields: pattern_position and
1884     next_item_length. These contain the offset in the pattern to the next match
1885     item, and its length, respectively.
1887     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1888     insertion of callouts before each pattern item. Added the /C option to
1889     pcretest to make use of this.
1891     4. On the advice of a Windows user, the lines
1893     #if defined(_WIN32) || defined(WIN32)
1894     _setmode( _fileno( stdout ), 0x8000 );
1895     #endif /* defined(_WIN32) || defined(WIN32) */
1897     have been added to the source of pcretest. This apparently does useful
1898     magic in relation to line terminators.
1900     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1901     for the benefit of those environments where the "b" makes a difference.
1903     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1904     to know about it. I have put a hack into configure.in that adds in code
1905     to set GCC=yes if CC=icc. This seems to end up at a point in the
1906     generated configure script that is early enough to affect the setting of
1907     compiler options, which is what is needed, but I have no means of testing
1908     whether it really works. (The user who reported this had patched the
1909     generated configure script, which of course I cannot do.)
1911     LATER: After change 22 below (new libtool files), the configure script
1912     seems to know about icc (and also ecc). Therefore, I have commented out
1913     this hack in configure.in.
1915     7. Added support for pkg-config (2 patches were sent in).
1917     8. Negated POSIX character classes that used a combination of internal tables
1918     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1919     [[:^ascii]]. Typically, they would match almost any characters. The other
1920     POSIX classes were not broken in this way.
1922     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1923     to find the match, as PCRE was deluded into thinking that the match had to
1924     start at the start point or following a newline. The same bug applied to
1925     patterns with negative forward assertions or any backward assertions
1926     preceding ".*" at the start, unless the pattern required a fixed first
1927     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1929     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1930     starting at the last subject character, bytes beyond the end of the subject
1931     string were read.
1933     11. Renamed the variable "class" as "classbits" to make life easier for C++
1934     users. (Previously there was a macro definition, but it apparently wasn't
1935     enough.)
1937     12. Added the new field "tables" to the extra data so that tables can be passed
1938     in at exec time, or the internal tables can be re-selected. This allows
1939     a compiled regex to be saved and re-used at a later time by a different
1940     program that might have everything at different addresses.
1942     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1943     -R library as well as a -L library.
1945     14. The debugging options of pcretest (-d on the command line or D on a
1946     pattern) showed incorrect output for anything following an extended class
1947     that contained multibyte characters and which was followed by a quantifier.
1949     15. Added optional support for general category Unicode character properties
1950     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1951     support. It adds about 90K to the size of the library. The meanings of the
1952     inbuilt class escapes such as \d and \s have NOT been changed.
1954     16. Updated pcredemo.c to include calls to free() to release the memory for the
1955     compiled pattern.
1957     17. The generated file chartables.c was being created in the source directory
1958     instead of in the building directory. This caused the build to fail if the
1959     source directory was different from the building directory, and was
1960     read-only.
1962     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1963     file. No doubt somebody will tell me if they don't make sense... Also added
1964     Dan Mooney's comments about building on OpenVMS.
1966     19. Added support for partial matching via the PCRE_PARTIAL option for
1967     pcre_exec() and the \P data escape in pcretest.
1969     20. Extended pcretest with 3 new pattern features:
1971     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1972     write the compiled pattern to the file whose name is "rest-of-line".
1973     This is a straight binary dump of the data, with the saved pointer to
1974     the character tables forced to be NULL. The study data, if any, is
1975     written too. After writing, pcretest reads a new pattern.
1977     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1978     compiled pattern from the given file. There must not be any
1979     occurrences of "<" in the file name (pretty unlikely); if there are,
1980     pcretest will instead treat the initial "<" as a pattern delimiter.
1981     After reading in the pattern, pcretest goes on to read data lines as
1982     usual.
1984     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1985     and 16-bit fields in a compiled pattern, to simulate a pattern that
1986     was compiled on a host of opposite endianness.
1988     21. The pcre-exec() function can now cope with patterns that were compiled on
1989     hosts of opposite endianness, with this restriction:
1991     As for any compiled expression that is saved and used later, the tables
1992     pointer field cannot be preserved; the extra_data field in the arguments
1993     to pcre_exec() should be used to pass in a tables address if a value
1994     other than the default internal tables were used at compile time.
1996     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1997     now diagnosed as an error. Previously, most of the time, a negative number
1998     would have been treated as zero, but if in addition "ovector" was passed as
1999     NULL, a crash could occur.
2001     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
2002     new versions from the libtool 1.5 distribution (the last one is a copy of
2003     a file called libtool.m4). This seems to have fixed the need to patch
2004     "configure" to support Darwin 1.3 (which I used to do). However, I still
2005     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
2006     workstation).
2008     24. Changed the PCRE licence to be the more standard "BSD" licence.
2011 nigel 73 Version 4.5 01-Dec-03
2012     ---------------------
2014     1. There has been some re-arrangement of the code for the match() function so
2015     that it can be compiled in a version that does not call itself recursively.
2016     Instead, it keeps those local variables that need separate instances for
2017     each "recursion" in a frame on the heap, and gets/frees frames whenever it
2018     needs to "recurse". Keeping track of where control must go is done by means
2019     of setjmp/longjmp. The whole thing is implemented by a set of macros that
2020     hide most of the details from the main code, and operates only if
2021     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
2022     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
2023     operating.
2025     To make it easier for callers to provide specially tailored get/free
2026     functions for this usage, two new functions, pcre_stack_malloc, and
2027     pcre_stack_free, are used. They are always called in strict stacking order,
2028     and the size of block requested is always the same.
2030     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
2031     PCRE has been compiled to use the stack or the heap for recursion. The
2032     -C option of pcretest uses this to show which version is compiled.
2034     A new data escape \S, is added to pcretest; it causes the amounts of store
2035     obtained and freed by both kinds of malloc/free at match time to be added
2036     to the output.
2038     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
2039     what's available on my current Linux desktop machine.
2041     3. When matching a UTF-8 string, the test for a valid string at the start has
2042     been extended. If start_offset is not zero, PCRE now checks that it points
2043     to a byte that is the start of a UTF-8 character. If not, it returns
2044     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
2045     this is necessary because there may be backward assertions in the pattern.
2046     When matching the same subject several times, it may save resources to use
2047     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
2049     4. The code for checking the validity of UTF-8 strings has been tightened so
2050     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
2051     containing "overlong sequences".
2053     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
2054     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
2055     should have been "&", but it just so happened that all the cases this let
2056     through by mistake were picked up later in the function.
2058     6. I had used a variable called "isblank" - this is a C99 function, causing
2059     some compilers to warn. To avoid this, I renamed it (as "blankclass").
2061     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
2062     prompting; (b) run "./pcretest /dev/null" at the start of the test script
2063     so the version is shown; (c) stop "make test" echoing "./RunTest".
2065     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
2067     9. The prototype for memmove() for systems that don't have it was using
2068     size_t, but the inclusion of the header that defines size_t was later. I've
2069     moved the #includes for the C headers earlier to avoid this.
2071     10. Added some adjustments to the code to make it easier to compiler on certain
2072     special systems:
2074     (a) Some "const" qualifiers were missing.
2075     (b) Added the macro EXPORT before all exported functions; by default this
2076     is defined to be empty.
2077     (c) Changed the dftables auxiliary program (that builds chartables.c) so
2078     that it reads its output file name as an argument instead of writing
2079     to the standard output and assuming this can be redirected.
2081     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
2082     class containing characters with values greater than 255, PCRE compilation
2083     went into a loop.
2085     12. A recursive reference to a subpattern that was within another subpattern
2086     that had a minimum quantifier of zero caused PCRE to crash. For example,
2087     (x(y(?2))z)? provoked this bug with a subject that got as far as the
2088     recursion. If the recursively-called subpattern itself had a zero repeat,
2089     that was OK.
2091     13. In pcretest, the buffer for reading a data line was set at 30K, but the
2092     buffer into which it was copied (for escape processing) was still set at
2093     1024, so long lines caused crashes.
2095     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
2096     "internal error: code overflow...". This applied to any character class
2097     that was followed by a possessive quantifier.
2099     15. Modified the Makefile to add libpcre.la as a prerequisite for
2100     libpcreposix.la because I was told this is needed for a parallel build to
2101     work.
2103     16. If a pattern that contained .* following optional items at the start was
2104     studied, the wrong optimizing data was generated, leading to matching
2105     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
2106     matching string must start with a or b or c. The correct conclusion for
2107     this pattern is that a match can start with any character.
2110 nigel 71 Version 4.4 13-Aug-03
2111     ---------------------
2113     1. In UTF-8 mode, a character class containing characters with values between
2114     127 and 255 was not handled correctly if the compiled pattern was studied.
2115     In fixing this, I have also improved the studying algorithm for such
2116     classes (slightly).
2118     2. Three internal functions had redundant arguments passed to them. Removal
2119     might give a very teeny performance improvement.
2121     3. Documentation bug: the value of the capture_top field in a callout is *one
2122     more than* the number of the hightest numbered captured substring.
2124     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2125     in incorrectly linking with a previously installed version. They now link
2126     explicitly with libpcre.la.
2128     5. configure.in no longer needs to recognize Cygwin specially.
2130     6. A problem in pcre.in for Windows platforms is fixed.
2132     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2133     pcretest, it used to include the size of the study block as part of its
2134     output. Unfortunately, the structure contains a field that has a different
2135     size on different hardware architectures. This meant that the tests that
2136     showed this size failed. As the block is currently always of a fixed size,
2137     this information isn't actually particularly useful in pcretest output, so
2138     I have just removed it.
2140     8. Three pre-processor statements accidentally did not start in column 1.
2141     Sadly, there are *still* compilers around that complain, even though
2142     standard C has not required this for well over a decade. Sigh.
2144     9. In pcretest, the code for checking callouts passed small integers in the
2145     callout_data field, which is a void * field. However, some picky compilers
2146     complained about the casts involved for this on 64-bit systems. Now
2147     pcretest passes the address of the small integer instead, which should get
2148     rid of the warnings.
2150     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2151     both compile and run time, and gives an error if an invalid UTF-8 sequence
2152     is found. There is a option for disabling this check in cases where the
2153     string is known to be correct and/or the maximum performance is wanted.
2155     11. In response to a bug report, I changed one line in Makefile.in from
2157     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2158     to
2159     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2161     to look similar to other lines, but I have no way of telling whether this
2162     is the right thing to do, as I do not use Windows. No doubt I'll get told
2163     if it's wrong...
2166 nigel 69 Version 4.3 21-May-03
2167     ---------------------
2169     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2170     Makefile.
2172     2. Some refactoring to improve the quality of the code:
2174     (i) The utf8_table... variables are now declared "const".
2176     (ii) The code for \cx, which used the "case flipping" table to upper case
2177     lower case letters, now just substracts 32. This is ASCII-specific,
2178     but the whole concept of \cx is ASCII-specific, so it seems
2179     reasonable.
2181     (iii) PCRE was using its character types table to recognize decimal and
2182     hexadecimal digits in the pattern. This is silly, because it handles
2183     only 0-9, a-f, and A-F, but the character types table is locale-
2184     specific, which means strange things might happen. A private
2185     table is now used for this - though it costs 256 bytes, a table is
2186     much faster than multiple explicit tests. Of course, the standard
2187     character types table is still used for matching digits in subject
2188     strings against \d.
2190     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2191     ending in _t are). So I've renamed it as ESC_tee.
2193     3. The first argument for regexec() in the POSIX wrapper should have been
2194     defined as "const".
2196     4. Changed pcretest to use malloc() for its buffers so that they can be
2197     Electric Fenced for debugging.
2199     5. There were several places in the code where, in UTF-8 mode, PCRE would try
2200     to read one or more bytes before the start of the subject string. Often this
2201     had no effect on PCRE's behaviour, but in some circumstances it could
2202     provoke a segmentation fault.
2204     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2205     to try to read one or more bytes before the start of the subject string.
2207     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2208     UTF-8 support could misbehave in various ways if the subject string
2209     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2210     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2211     back over UTF-8 characters.)
2214 nigel 67 Version 4.2 14-Apr-03
2215     ---------------------
2217     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2219     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2220     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2221     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2222     [WIN_PREFIX]: new variable, "cyg" for Cygwin
2223     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2224     and BUILD_EXEEXT
2225     Note: automatic setting of the BUILD variables is not yet working
2226     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2227     compile-time but not at link-time
2228     [LINK]: use for linking executables only
2229     make different versions for Windows and non-Windows
2230     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2231     libraries
2232     [LINK_FOR_BUILD]: new variable
2233     [OBJEXT]: use throughout
2234     [EXEEXT]: use throughout
2235     <winshared>: new target
2236     <wininstall>: new target
2237     <dftables.o>: use native compiler
2238     <dftables>: use native linker
2239     <install>: handle Windows platform correctly
2240     <clean>: ditto
2241     <check>: ditto
2242     copy DLL to top builddir before testing
2244     As part of these changes, -no-undefined was removed again. This was reported
2245     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2246     in any case.
2248     3. Some tidies to get rid of compiler warnings:
2250     . In the match_data structure, match_limit was an unsigned long int, whereas
2251     match_call_count was an int. I've made them both unsigned long ints.
2253     . In pcretest the fact that a const uschar * doesn't automatically cast to
2254     a void * provoked a warning.
2256     . Turning on some more compiler warnings threw up some "shadow" variables
2257     and a few more missing casts.
2259     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2260     option, a class that contained a single character with a value between 128
2261     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2263     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2264     option, a class that contained several characters, but with at least one
2265     whose value was between 128 and 255 caused PCRE to crash.
2268 nigel 65 Version 4.1 12-Mar-03
2269     ---------------------
2270 nigel 63
2271 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
2272     needed, and a string in dftables.c that was longer than standard compilers are
2273     required to support.
2275     2. Compiling with Sun's compiler found a few more places where the code could
2276     be tidied up in order to avoid warnings.
2278     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2279     first of these names is deprecated in the latest Autoconf in favour of the name
2280     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2281     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2282     analogy I have changed it to CFLAGS_FOR_BUILD.
2284     4. Added -no-undefined to the linking command in the Makefile, because this is
2285     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2286     linking step for the pcreposix library.
2288     5. PCRE was failing to diagnose the case of two named groups with the same
2289     name.
2291     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2292     literal character that is needed in the subject for a match, and scans along to
2293     ensure that it is present before embarking on the full matching process. This
2294     saves time in cases of nested unlimited repeats that are never going to match.
2295     Problem: the scan can take a lot of time if the subject is very long (e.g.
2296     megabytes), thus penalizing straightforward matches. It is now done only if the
2297     amount of subject to be scanned is less than 1000 bytes.
2299     7. A lesser problem with the same optimization is that it was recording the
2300     first character of an anchored pattern as "needed", thus provoking a search
2301     right along the subject, even when the first match of the pattern was going to
2302     fail. The "needed" character is now not set for anchored patterns, unless it
2303     follows something in the pattern that is of non-fixed length. Thus, it still
2304     fulfils its original purpose of finding quick non-matches in cases of nested
2305     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2308     Version 4.0 17-Feb-03
2309     ---------------------
2311 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
2312     extended to the end of string, PCRE compiled incorrect data. This could lead to
2313     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2315     2. Moved to autoconf 2.53 and libtool 1.4.2.
2317     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2318     the special perltest8 script is no longer needed - all the tests can be run
2319     from a single perltest script.
2321     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2322     by \s. It has now been removed in PCRE. This means it isn't recognized as
2323     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2324     class [:space:] *does* include VT, thereby creating a mess.
2326     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2327     space and tab.
2329     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2330     its new features into the main test script, reducing the number of scripts.
2332     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2333     were backward compatible, and made the (?i) apply to the whole pattern, as if
2334     /i were given. Now it behaves more logically, and applies the option setting
2335     only to what follows. PCRE has been changed to follow suit. However, if it
2336     finds options settings right at the start of the pattern, it extracts them into
2337     the global options, as before. Thus, they show up in the info data.
2339     8. Added support for the \Q...\E escape sequence. Characters in between are
2340     treated as literals. This is slightly different from Perl in that $ and @ are
2341     also handled as literals inside the quotes. In Perl, they will cause variable
2342     interpolation. Note the following examples:
2344     Pattern PCRE matches Perl matches
2346     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2347     \Qabc\$xyz\E abc\$xyz abc\$xyz
2348     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2350     For compatibility with Perl, \Q...\E sequences are recognized inside character
2351     classes as well as outside them.
2353     9. Re-organized 3 code statements in pcretest to avoid "overflow in
2354     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2355     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2356     signed/unsigned warnings.
2358     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2359     option for pcretest, so I've replaced it by a simple function that does just
2360     that job.
2362     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2363     "pcregrep -".
2365     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2366     Java package. This provides some syntactic sugar for simple cases of what my
2367     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2368     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2369     item, you can use this simplified notation. Note that only makes sense with
2370     greedy quantifiers. Consequently, the use of the possessive quantifier forces
2371     greediness, whatever the setting of the PCRE_UNGREEDY option.
2373     13. A change of greediness default within a pattern was not taking effect at
2374     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2375     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2376     was abstracted outside.
2378     14. PCRE now supports the \G assertion. It is true when the current matching
2379     position is at the start point of the match. This differs from \A when the
2380     starting offset is non-zero. Used with the /g option of pcretest (or similar
2381     code), it works in the same way as it does for Perl's /g option. If all
2382     alternatives of a regex begin with \G, the expression is anchored to the start
2383     match position, and the "anchored" flag is set in the compiled expression.
2385     15. Some bugs concerning the handling of certain option changes within patterns
2386     have been fixed. These applied to options other than (?ims). For example,
2387     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2388     been the other way round. Some of this was related to change 7 above.
2390     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2391     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2392     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2393     POSIX classes only within a class (e.g. /[[:alpha:]]/).
2395     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2396     mode. Unlike ".", it always matches newline, whatever the setting of
2397     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2398     assertions. Perl allows it, but it doesn't (in general) work because it can't
2399     calculate the length of the lookbehind. At least, that's the case for Perl
2400     5.8.0 - I've been told they are going to document that it doesn't work in
2401     future.
2403     18. Added an error diagnosis for escapes that PCRE does not support: these are
2404     \L, \l, \N, \P, \p, \U, \u, and \X.
2406     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2407     reading past the end of the pattern in cases such as /[abcd/.
2409     20. PCRE was getting more memory than necessary for patterns with classes that
2410     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2412     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2413     compiling PCRE for use with Virtual Pascal.
2415     22. Small fix to the Makefile to make it work properly if the build is done
2416     outside the source tree.
2418     23. Added a new extension: a condition to go with recursion. If a conditional
2419     subpattern starts with (?(R) the "true" branch is used if recursion has
2420     happened, whereas the "false" branch is used only at the top level.
2422     24. When there was a very long string of literal characters (over 255 bytes
2423     without UTF support, over 250 bytes with UTF support), the computation of how
2424     much memory was required could be incorrect, leading to segfaults or other
2425     strange effects.
2427     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2428     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2429     there was a subsequent back reference to those brackets. This meant that, for
2430     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2431     possible to check for precisely this case. All we can do is abandon the
2432     optimization if .* occurs inside capturing brackets when there are any back
2433     references whatsoever. (See below for a better fix that came later.)
2435     26. The handling of the optimization for finding the first character of a
2436     non-anchored pattern, and for finding a character that is required later in the
2437     match were failing in some cases. This didn't break the matching; it just
2438     failed to optimize when it could. The way this is done has been re-implemented.
2440     27. Fixed typo in error message for invalid (?R item (it said "(?p").
2442     28. Added a new feature that provides some of the functionality that Perl
2443     provides with (?{...}). The facility is termed a "callout". The way it is done
2444     in PCRE is for the caller to provide an optional function, by setting
2445     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2446     global variable. By default it is unset, which disables all calling out. To get
2447     the function called, the regex must include (?C) at appropriate points. This
2448     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2449     This provides a means of identifying different callout points. When PCRE
2450     reaches such a point in the regex, if pcre_callout has been set, the external
2451     function is called. It is provided with data in a structure called
2452     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2453     matching continues; if it returns a non-zero value, the match at the current
2454     point fails. However, backtracking will occur if possible. [This was changed
2455     later and other features added - see item 49 below.]
2457     29. pcretest is upgraded to test the callout functionality. It provides a
2458     callout function that displays information. By default, it shows the start of
2459     the match and the current position in the text. There are some new data escapes
2460     to vary what happens:
2462     \C+ in addition, show current contents of captured substrings
2463     \C- do not supply a callout function
2464     \C!n return 1 when callout number n is reached
2465     \C!n!m return 1 when callout number n is reached for the mth time
2467     30. If pcregrep was called with the -l option and just a single file name, it
2468     output "<stdin>" if a match was found, instead of the file name.
2470     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2471     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2472     pcre_exec(). This saves a malloc/free per call. The default value of
2473     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2474     when configuring.
2476     32. The default maximum size of a compiled pattern is 64K. There have been a
2477     few cases of people hitting this limit. The code now uses macros to handle the
2478     storing of links as offsets within the compiled pattern. It defaults to 2-byte
2479     links, but this can be changed to 3 or 4 bytes by --with-link-size when
2480     configuring. Tests 2 and 5 work only with 2-byte links because they output
2481     debugging information about compiled patterns.
2483     33. Internal code re-arrangements:
2485     (a) Moved the debugging function for printing out a compiled regex into
2486     its own source file (printint.c) and used #include to pull it into
2487     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2488     separate copies.
2490     (b) Defined the list of op-code names for debugging as a macro in
2491     internal.h so that it is next to the definition of the opcodes.
2493     (c) Defined a table of op-code lengths for simpler skipping along compiled
2494     code. This is again a macro in internal.h so that it is next to the
2495     definition of the opcodes.
2497     34. Added support for recursive calls to individual subpatterns, along the
2498     lines of Robin Houston's patch (but implemented somewhat differently).
2500     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2501     allow it to read and process whole directories in Win32. This code was
2502     contributed by Lionel Fourquaux; it has not been tested by me.
2504     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2505     used to name a group. Names consist of alphanumerics and underscores, and must
2506     be unique. Back references use the syntax (?P=name) and recursive calls use
2507     (?P>name) which is a PCRE extension to the Python extension. Groups still have
2508     numbers. The function pcre_fullinfo() can be used after compilation to extract
2509     a name/number map. There are three relevant calls:
2511     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2512     PCRE_INFO_NAMECOUNT yields the number of entries
2513     PCRE_INFO_NAMETABLE yields a pointer to the map.
2515     The map is a vector of fixed-size entries. The size of each entry depends on
2516     the length of the longest name used. The first two bytes of each entry are the
2517     group number, most significant byte first. There follows the corresponding
2518     name, zero terminated. The names are in alphabetical order.
2520     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2521     case instead of 255. Making it the same both with and without UTF-8 support
2522     means that the same test output works with both.
2524     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2525     calling malloc() with a zero argument.
2527     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2528     optimization. I've improved things by keeping a bitmap of backreferences with
2529     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2530     fact referenced, the optimization can be applied. It is unlikely that a
2531     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2532     the match to follow \n) will appear inside brackets with a number greater than
2533     31, but if it does, any back reference > 31 suppresses the optimization.
2535     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2536     of disabling numbered capturing parentheses. Any opening parenthesis that is
2537     not followed by ? behaves as if it were followed by ?: but named parentheses
2538     can still be used for capturing (and they will acquire numbers in the usual
2539     way).
2541     41. Redesigned the return codes from the match() function into yes/no/error so
2542     that errors can be passed back from deep inside the nested calls. A malloc
2543     failure while inside a recursive subpattern call now causes the
2544     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2546     42. It is now possible to set a limit on the number of times the match()
2547     function is called in a call to pcre_exec(). This facility makes it possible to
2548     limit the amount of recursion and backtracking, though not in a directly
2549     obvious way, because the match() function is used in a number of different
2550     circumstances. The count starts from zero for each position in the subject
2551     string (for non-anchored patterns). The default limit is, for compatibility, a
2552     large number, namely 10 000 000. You can change this in two ways:
2554     (a) When configuring PCRE before making, you can use --with-match-limit=n
2555     to set a default value for the compiled library.
2557     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2558     a different value is set. See 45 below.
2560     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2562     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2563     of things that can be changed at compile time. The first argument specifies
2564     what is wanted and the second points to where the information is to be placed.
2565     The current list of available information is:
2569     The output is an integer that is set to one if UTF-8 support is available;
2570     otherwise it is set to zero.
2574     The output is an integer that it set to the value of the code that is used for
2575     newline. It is either LF (10) or CR (13).
2579     The output is an integer that contains the number of bytes used for internal
2580     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2584     The output is an integer that contains the threshold above which the POSIX
2585     interface uses malloc() for output vectors. See item 31 above.
2589     The output is an unsigned integer that contains the default limit of the number
2590     of match() calls in a pcre_exec() execution. See 42 above.
2592     44. pcretest has been upgraded by the addition of the -C option. This causes it
2593     to extract all the available output from the new pcre_config() function, and to
2594     output it. The program then exits immediately.
2596     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2597     order to support additional features. One way would have been to define
2598     pcre_exec2() (for example) with extra arguments, but this would not have been
2599     extensible, and would also have required all calls to the original function to
2600     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2601     is used for passing in "extra" data from pcre_study().
2603     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2604     contains the following fields:
2606     flags a bitmap indicating which of the following fields are set
2607     study_data opaque data from pcre_study()
2608     match_limit a way of specifying a limit on match() calls for a specific
2609     call to pcre_exec()
2610     callout_data data for callouts (see 49 below)
2612     The flag bits are also defined in pcre.h, and are
2618     The pcre_study() function now returns one of these new pcre_extra blocks, with
2619     the actual study data pointed to by the study_data field, and the
2620     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2621     before. That is, this change is entirely upwards-compatible and requires no
2622     change to existing code.
2624     If you want to pass in additional data to pcre_exec(), you can either place it
2625     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2626     block.
2628     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2629     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2630     times with different match limits, until it finds the minimum value needed for
2631     pcre_exec() to complete. The value is then output. This can be instructive; for
2632     most simple matches the number is quite small, but for pathological cases it
2633     gets very large very quickly.
2635     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2636     returns the size of the data block pointed to by the study_data field in a
2637     pcre_extra block, that is, the value that was passed as the argument to
2638     pcre_malloc() when PCRE was getting memory in which to place the information
2639     created by pcre_study(). The fourth argument should point to a size_t variable.
2640     pcretest has been extended so that this information is shown after a successful
2641     pcre_study() call when information about the compiled regex is being displayed.
2643     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2644     because what follows is always an absolute path. (Later: it turns out that this
2645     is more than cosmetic for MinGW, because it doesn't like empty path
2646     components.)
2648     49. Some changes have been made to the callout feature (see 28 above):
2650     (i) A callout function now has three choices for what it returns:
2652     0 => success, carry on matching
2653     > 0 => failure at this point, but backtrack if possible
2654     < 0 => serious error, return this value from pcre_exec()
2656     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2657     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2658     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2659     use by callout functions. It will never be used by PCRE itself.
2661     (ii) The pcre_extra structure (see 45 above) has a void * field called
2662     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2663     pcre_callout_block structure has a field of the same name. The contents of
2664     the field passed in the pcre_extra structure are passed to the callout
2665     function in the corresponding field in the callout block. This makes it
2666     easier to use the same callout-containing regex from multiple threads. For
2667     testing, the pcretest program has a new data escape
2669     \C*n pass the number n (may be negative) as callout_data
2671     If the callout function in pcretest receives a non-zero value as
2672     callout_data, it returns that value.
2674     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2675     there were some redundant $(CFLAGS) in commands that are now specified as
2676     $(LINK), which already includes $(CFLAGS).
2678     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2679     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2680     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2681     one-byte characters throughout. Note that case-insensitive matching applies
2682     only to characters whose values are less than 256. PCRE doesn't support the
2683     notion of cases for higher-valued characters.
2685     (i) A character class whose characters are all within 0-255 is handled as
2686     a bit map, and the map is inverted for negative classes. Previously, a
2687     character > 255 always failed to match such a class; however it should
2688     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2690     (ii) A negated character class with a single character < 255 is coded as
2691     "not this character" (OP_NOT). This wasn't working properly when the test
2692     character was multibyte, either singly or repeated.
2694     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2695     mode, for example: \x{100}{2,3}.
2697     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2698     singly or repeated) now correctly test multibyte characters. However,
2699     PCRE doesn't recognize any characters with values greater than 255 as
2700     digits, spaces, or word characters. Such characters always match \D, \S,
2701     and \W, and never match \d, \s, or \w.
2703     (v) Classes may now contain characters and character ranges with values
2704     greater than 255. For example: [ab\x{100}-\x{400}].
2706     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2707     PCRE in UTF-8 mode.
2709     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2710     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2711     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2712     value.)
2714     53. The single man page has become too large. I have therefore split it up into
2715     a number of separate man pages. These also give rise to individual HTML pages;
2716     these are now put in a separate directory, and there is an index.html page that
2717     lists them all. Some hyperlinking between the pages has been installed.
2719     54. Added convenience functions for handling named capturing parentheses.
2721     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2722     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2723     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2724     are faulted.
2726     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2727     calling configure. These values are used when compiling the dftables.c program
2728     which is run to generate the source of the default character tables. They
2729     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2730     you will need to set these values.
2732     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2735     Version 3.9 02-Jan-02
2736 nigel 61 ---------------------
2738     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2740     2. If --disable-static was given, the building process failed when trying to
2741     build pcretest and pcregrep. (For some reason it was using libtool to compile
2742     them, which is not right, as they aren't part of the library.)
2745 nigel 59 Version 3.8 18-Dec-01
2746     ---------------------
2748     1. The experimental UTF-8 code was completely screwed up. It was packing the
2749     bytes in the wrong order. How dumb can you get?
2752 nigel 57 Version 3.7 29-Oct-01
2753     ---------------------
2755     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2756     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2757     this didn't happen under Solaris 8, where I normally test things.
2759 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2760     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2761     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2762     this fix an hour or so after the initial 3.7 release.)
2763 nigel 57
2764 nigel 59
2765 nigel 55 Version 3.6 23-Oct-01
2766     ---------------------
2768     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2769     offsets passed as NULL with zero offset count.
2771     2. The config.guess and config.sub files had not been updated when I moved to
2772     the latest autoconf.
2775 nigel 53 Version 3.5 15-Aug-01
2776     ---------------------
2777 nigel 5
2778 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2779     had been forgotten.
2781     2. By using declared but undefined structures, we can avoid using "void"
2782     definitions in pcre.h while keeping the internal definitions of the structures
2783     private.
2785     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2786     user point of view, this means that both static and shared libraries are built
2787     by default, but this can be individually controlled. More of the work of
2788     handling this static/shared cases is now inside libtool instead of PCRE's make
2789     file.
2791     4. The pcretest utility is now installed along with pcregrep because it is
2792     useful for users (to test regexs) and by doing this, it automatically gets
2793     relinked by libtool. The documentation has been turned into a man page, so
2794     there are now .1, .txt, and .html versions in /doc.
2796     5. Upgrades to pcregrep:
2797     (i) Added long-form option names like gnu grep.
2798     (ii) Added --help to list all options with an explanatory phrase.
2799     (iii) Added -r, --recursive to recurse into sub-directories.
2800     (iv) Added -f, --file to read patterns from a file.
2802     6. pcre_exec() was referring to its "code" argument before testing that
2803     argument for NULL (and giving an error if it was NULL).
2805     7. Upgraded Makefile.in to allow for compiling in a different directory from
2806     the source directory.
2808     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2809     options bits, the pointer it was passed was to an int instead of to an unsigned
2810     long int. This mattered only on 64-bit systems.
2812     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2813     generated) instead of pcre.in, which it its source. Also made the same change
2814     in several of the .c files.
2816     10. A new release of gcc defines printf() as a macro, which broke pcretest
2817     because it had an ifdef in the middle of a string argument for printf(). Fixed
2818     by using separate calls to printf().
2820     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2821     script, to force use of CR or LF instead of \n in the source. On non-Unix
2822     systems, the value can be set in config.h.
2824     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2825     absolute limit. Changed the text of the error message to make this clear, and
2826     likewise updated the man page.
2828     13. The limit of 99 on the number of capturing subpatterns has been removed.
2829     The new limit is 65535, which I hope will not be a "real" limit.
2832 nigel 51 Version 3.4 22-Aug-00
2833     ---------------------
2835     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2837     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2840 nigel 49 Version 3.3 01-Aug-00
2841     ---------------------
2843     1. If an octal character was given, but the value was greater than \377, it
2844     was not getting masked to the least significant bits, as documented. This could
2845     lead to crashes in some systems.
2847     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2848     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2850     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2851     These just pass their arguments on to (pcre_free)(), but they are provided
2852     because some uses of PCRE bind it to non-C systems that can call its functions,
2853     but cannot call free() or pcre_free() directly.
2855     4. Add "make test" as a synonym for "make check". Corrected some comments in
2856     the Makefile.
2858     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2859     Makefile.
2861     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2862     command called pgrep for grepping around the active processes.
2864     7. Added the beginnings of support for UTF-8 character strings.
2866     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2867     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2868     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2869     out for the ar command.)
2872 nigel 47 Version 3.2 12-May-00
2873     ---------------------
2875     This is purely a bug fixing release.
2877     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2878     of ZA. This was just one example of several cases that could provoke this bug,
2879     which was introduced by change 9 of version 2.00. The code for breaking
2880     infinite loops after an iteration that matches an empty string was't working
2881     correctly.
2883     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2884     when matched against abbab (for example). After matching an empty string, it
2885     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2886     caused it to match further down the string than it should.
2888     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2889     was there because it doesn't seem to be needed, and it causes trouble on some
2890     systems, as it is not a Standard C header. It has been removed.
2892     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2893     were reported on the Macintosh. The changes were from
2895     while ((c = *(++ptr)) != 0 && c != '\n');
2896     to
2897     while ((c = *(++ptr)) != 0 && c != '\n') ;
2899     Totally extraordinary, but if that's what it takes...
2901     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2902     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2903     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2904     assumes the way PCRE uses memmove() (always moving upwards).
2906     6. PCRE is being used in one environment where strchr() is not available. There
2907     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2908     faster code anyway.
2911 nigel 45 Version 3.1 09-Feb-00
2912     ---------------------
2914     The only change in this release is the fixing of some bugs in Makefile.in for
2915     the "install" target:
2917     (1) It was failing to install pcreposix.h.
2919     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2922 nigel 43 Version 3.0 01-Feb-00
2923     ---------------------
2924 nigel 41
2925     1. Add support for the /+ modifier to perltest (to output $` like it does in
2926     pcretest).
2928     2. Add support for the /g modifier to perltest.
2930     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2931     matches null strings.
2933     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2934     pattern. Perl treats empty patterns specially - it reuses the most recent
2935     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2936     effect.
2938     5. The POSIX interface was broken in that it was just handing over the POSIX
2939     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2940     required a bigger vector, with some working space on the end. This means that
2941     the POSIX wrapper now has to get and free some memory, and copy the results.
2943 nigel 43 6. Added some simple autoconf support, placing the test data and the
2944     documentation in separate directories, re-organizing some of the
2945     information files, and making it build pcre-config (a GNU standard). Also added
2946     libtool support for building PCRE as a shared library, which is now the
2947     default.
2948 nigel 41
2949 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2950     09 are not valid octal constants. Single digits will be used for minor values
2951     less than 10.
2953     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2954     existing programs that set these in the POSIX interface can use PCRE without
2955     modification.
2957     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2958     return all that pcre_info() returns, plus additional data. The pcre_info()
2959     function is retained for compatibility, but is considered to be obsolete.
2961     10. Added experimental recursion feature (?R) to handle one common case that
2962     Perl 5.6 will be able to do with (?p{...}).
2964     11. Added support for POSIX character classes like [:alpha:], which Perl is
2965     adopting.
2968 nigel 39 Version 2.08 31-Aug-99
2969     ----------------------
2971     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2972     trying to match at the startoffset position, but instead was moving forward to
2973     the next newline as if a previous match had failed.
2975     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2976     and could get into a loop if a null string was matched other than at the start
2977     of the subject.
2979     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2980     be distinguished at compile time, and for completeness also added PCRE_DATE.
2982     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2983     in GnuWin32 environments.
2986 nigel 37 Version 2.07 29-Jul-99
2987     ----------------------
2989     1. The documentation is now supplied in plain text form and HTML as well as in
2990     the form of man page sources.
2992     2. C++ compilers don't like assigning (void *) values to other pointer types.
2993     In particular this affects malloc(). Although there is no problem in Standard
2994     C, I've put in casts to keep C++ compilers happy.
2996     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2997     should be (const char *).
2999     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
3000     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
3001     However, I haven't made this a standard facility. The documentation doesn't
3002     mention it, and the Makefile doesn't support it.
3004     5. The Makefile now contains an "install" target, with editable destinations at
3005     the top of the file. The pcretest program is not installed.
3007     6. pgrep -V now gives the PCRE version number and date.
3009     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
3010     causing the entire string to be ignored, instead of just the last character.
3012     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
3013     non-matching string, it can take a very, very long time, even for strings of
3014     quite modest length, because of the nested recursion. PCRE now does better in
3015     some of these cases. It does this by remembering the last required literal
3016     character in the pattern, and pre-searching the subject to ensure it is present
3017     before running the real match. In other words, it applies a heuristic to detect
3018     some types of certain failure quickly, and in the above example, if presented
3019     with a string that has no trailing " it gives "no match" very quickly.
3021     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
3022     other alternatives are tried instead.
3025 nigel 35 Version 2.06 09-Jun-99
3026     ----------------------
3028     1. Change pcretest's output for amount of store used to show just the code
3029     space, because the remainder (the data block) varies in size between 32-bit and
3030     64-bit systems.
3032     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
3033     start matching at. This allows lookbehinds to work when searching for multiple
3034     occurrences in a string.
3036     3. Added additional options to pcretest for testing multiple occurrences:
3038     /+ outputs the rest of the string that follows a match
3039     /g loops for multiple occurrences, using the new startoffset argument
3040     /G loops for multiple occurrences by passing an incremented pointer
3042     4. PCRE wasn't doing the "first character" optimization for patterns starting
3043     with \b or \B, though it was doing it for other lookbehind assertions. That is,
3044     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
3045     the letter 'x'. On long subject strings, this gives a significant speed-up.
3048 nigel 33 Version 2.05 21-Apr-99
3049     ----------------------
3051     1. Changed the type of magic_number from int to long int so that it works
3052     properly on 16-bit systems.
3054     2. Fixed a bug which caused patterns starting with .* not to work correctly
3055     when the subject string contained newline characters. PCRE was assuming
3056     anchoring for such patterns in all cases, which is not correct because .* will
3057     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
3058     DOTALL is set at top level; otherwise it knows that patterns starting with .*
3059     must be retried after every newline in the subject.
3062 nigel 31 Version 2.04 18-Feb-99
3063     ----------------------
3065     1. For parenthesized subpatterns with repeats whose minimum was zero, the
3066     computation of the store needed to hold the pattern was incorrect (too large).
3067     If such patterns were nested a few deep, this could multiply and become a real
3068     problem.
3070     2. Added /M option to pcretest to show the memory requirement of a specific
3071     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
3073     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
3074     compiled in such a way that the backtracking after subsequent failure was
3075     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
3076     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
3079 nigel 29 Version 2.03 02-Feb-99
3080     ----------------------
3082     1. Fixed typo and small mistake in man page.
3084 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
3085     LICENCE file containing the conditions.
3086 nigel 29
3087     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
3088     Perl, that is the internal \ allows the delimiter to be included in the
3089     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
3090     the final delimiter, add \ to the end of the pattern (to test the error).
3092     4. Added the convenience functions for extracting substrings after a successful
3093     match. Updated pcretest to make it able to test these functions.
3096 nigel 27 Version 2.02 14-Jan-99
3097     ----------------------
3099     1. Initialized the working variables associated with each extraction so that
3100     their saving and restoring doesn't refer to uninitialized store.
3102     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
3103     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
3104     fix the problem.
3106     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
3107     calls, and wasn't printing the correct value for compiling calls. Increased the
3108     default value of LOOPREPEAT, and the number of significant figures in the
3109     times.
3111     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
3113     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
3114     a building problem on Windows NT with a FAT file system.
3117 nigel 25 Version 2.01 21-Oct-98
3118     ----------------------
3120     1. Changed the API for pcre_compile() to allow for the provision of a pointer
3121     to character tables built by pcre_maketables() in the current locale. If NULL
3122     is passed, the default tables are used.
3125 nigel 23 Version 2.00 24-Sep-98
3126 nigel 21 ----------------------
3128 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3129     it any more.
3131     2. Allow quantification of (?>) groups, and make it work correctly.
3133     3. The first character computation wasn't working for (?>) groups.
3135     4. Correct the implementation of \Z (it is permitted to match on the \n at the
3136     end of the subject) and add 5.005's \z, which really does match only at the
3137     very end of the subject.
3139     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3141     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3142     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3143     localized options. All options to pcre_study() were also removed.
3145     7. Add other new features from 5.005:
3147     $(?<= positive lookbehind
3148     $(?<! negative lookbehind
3149     (?imsx-imsx) added the unsetting capability
3150     such a setting is global if at outer level; local otherwise
3151     (?imsx-imsx:) non-capturing groups with option setting
3152     (?(cond)re|re) conditional pattern matching
3154     A backreference to itself in a repeated group matches the previous
3155     captured string.
3157     8. General tidying up of studying (both automatic and via "study")
3158     consequential on the addition of new assertions.
3160     9. As in 5.005, unlimited repeated groups that could match an empty substring
3161     are no longer faulted at compile time. Instead, the loop is forcibly broken at
3162     runtime if any iteration does actually match an empty substring.
3164     10. Include the RunTest script in the distribution.
3166     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3167     discrepancies, some of which were old and were also with respect to 5.004. They
3168     have now been fixed.
3171     Version 1.09 28-Apr-98
3172     ----------------------
3174 nigel 21 1. A negated single character class followed by a quantifier with a minimum
3175     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3176     program crashes, or just wrong answers. This did not apply to negated classes
3177     containing more than one character, or to minima other than one.
3180 nigel 19 Version 1.08 27-Mar-98
3181     ----------------------
3183     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3185     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3186     latter must appear before anything that relies on it in the pattern.
3189 nigel 17 Version 1.07 16-Feb-98
3190     ----------------------
3192     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3193     repeat of a potentially empty string).
3196 nigel 15 Version 1.06 23-Jan-98
3197     ----------------------
3199     1. Added Markus Oberhumer's little patches for C++.
3201     2. Literal strings longer than 255 characters were broken.
3204 nigel 13 Version 1.05 23-Dec-97
3205     ----------------------
3207     1. Negated character classes containing more than one character were failing if
3208     PCRE_CASELESS was set at run time.
3211 nigel 11 Version 1.04 19-Dec-97
3212     ----------------------
3214     1. Corrected the man page, where some "const" qualifiers had been omitted.
3216     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3217     input syntax.
3219     3. Fixed memory leak which occurred when a regex with back references was
3220     matched with an offsets vector that wasn't big enough. The temporary memory
3221     that is used in this case wasn't being freed if the match failed.
3223     4. Tidied pcretest to ensure it frees memory that it gets.
3225     5. Temporary memory was being obtained in the case where the passed offsets
3226     vector was exactly big enough.
3228     6. Corrected definition of offsetof() from change 5 below.
3230     7. I had screwed up change 6 below and broken the rules for the use of
3231     setjmp(). Now fixed.
3234 nigel 9 Version 1.03 18-Dec-97
3235     ----------------------
3237     1. A erroneous regex with a missing opening parenthesis was correctly
3238     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3239     on some systems.
3241     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3242     it was reported that one broken compiler failed on the former because "code" is
3243     also an independent variable.
3245     3. The erroneous regex a[]b caused an array overrun reference.
3247     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3248     fail on data ending with that character. (It was going on too far, and checking
3249     the next character, typically a binary zero.) This was specific to the
3250     optimized code for single-character negative classes.
3252     5. Added a contributed patch from the TIN world which does the following:
3254     + Add an undef for memmove, in case the the system defines a macro for it.
3256     + Add a definition of offsetof(), in case there isn't one. (I don't know
3257     the reason behind this - offsetof() is part of the ANSI standard - but
3258     it does no harm).
3260     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3261     most of the places where whitespace preceded '#'. I have given up and
3262     allowed the remaining 2 cases to be at the margin.
3264     + Rename some variables in pcre to eliminate shadowing. This seems very
3265     pedantic, but does no harm, of course.
3267     6. Moved the call to setjmp() into its own function, to get rid of warnings
3268     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3270     7. Constructs such as \d{8,} were compiling into the equivalent of
3271 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3272 nigel 9 outcome, but in this particular case used more store than had been allocated,
3273     which caused the bug to be discovered because it threw up an internal error.
3275     8. The debugging code in both pcre and pcretest for outputting the compiled
3276     form of a regex was going wrong in the case of back references followed by
3277     curly-bracketed repeats.
3280 nigel 7 Version 1.02 12-Dec-97
3281     ----------------------
3283     1. Typos in pcre.3 and comments in the source fixed.
3285     2. Applied a contributed patch to get rid of places where it used to remove
3286     'const' from variables, and fixed some signed/unsigned and uninitialized
3287     variable warnings.
3289     3. Added the "runtest" target to Makefile.
3291     4. Set default compiler flag to -O2 rather than just -O.
3294 nigel 5 Version 1.01 19-Nov-97
3295     ----------------------
3297     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3298     like /([ab]*)*/, that is, for classes with more than one character in them.
3300     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3301     as /((?>a*))*/ (a PCRE_EXTRA facility).
3304     Version 1.00 18-Nov-97
3305     ----------------------
3307     1. Added compile-time macros to support systems such as SunOS4 which don't have
3308     memmove() or strerror() but have other things that can be used instead.
3310     2. Arranged that "make clean" removes the executables.
3313 nigel 3 Version 0.99 27-Oct-97
3314     ----------------------
3316     1. Fixed bug in code for optimizing classes with only one character. It was
3317     initializing a 32-byte map regardless, which could cause it to run off the end
3318     of the memory it had got.
3320     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3323     Version 0.98 22-Oct-97
3324     ----------------------
3326     1. Fixed bug in code for handling temporary memory usage when there are more
3327     back references than supplied space in the ovector. This could cause segfaults.
3330     Version 0.97 21-Oct-97
3331     ----------------------
3333     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3335     2. Optimized negated single characters not to use a bit map.
3337     3. Brought error texts together as macro definitions; clarified some of them;
3338     fixed one that was wrong - it said "range out of order" when it meant "invalid
3339     escape sequence".
3341     4. Changed some char * arguments to const char *.
3343     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3345     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3346     pcretest.
3349     Version 0.96 16-Oct-97
3350     ----------------------
3352     1. Added a simple "pgrep" utility to the distribution.
3354     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3355     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3356     where "ddd" means "one or more decimal digits".
3358     3. Fixed serious bug. If a pattern had a back reference, but the call to
3359     pcre_exec() didn't supply a large enough ovector to record the related
3360     identifying subpattern, the match always failed. PCRE now remembers the number
3361     of the largest back reference, and gets some temporary memory in which to save
3362     the offsets during matching if necessary, in order to ensure that
3363     backreferences always work.
3365     4. Increased the compatibility with Perl in a number of ways:
3367     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3368     to request this handling. The option can be set at compile or exec time.
3370     (b) $ matches before a terminating newline by default; an option
3371     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3372     mode). The option can be set at compile or exec time.
3374     (c) The handling of \ followed by a digit other than 0 is now supposed to be
3375     the same as Perl's. If the decimal number it represents is less than 10
3376     or there aren't that many previous left capturing parentheses, an octal
3377     escape is read. Inside a character class, it's always an octal escape,
3378     even if it is a single digit.
3380     (d) An escaped but undefined alphabetic character is taken as a literal,
3381     unless PCRE_EXTRA is set. Currently this just reserves the remaining
3382     escapes.
3384     (e) {0} is now permitted. (The previous item is removed from the compiled
3385     pattern).
3387     5. Changed all the names of code files so that the basic parts are no longer
3388     than 10 characters, and abolished the teeny "globals.c" file.
3390     6. Changed the handling of character classes; they are now done with a 32-byte
3391     bit map always.
3393     7. Added the -d and /D options to pcretest to make it possible to look at the
3394     internals of compilation without having to recompile pcre.
3397     Version 0.95 23-Sep-97
3398     ----------------------
3400     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3401     \x20 at the start of a run of normal characters. These were being treated as
3402     real characters, instead of the source characters being re-checked.
3405     Version 0.94 18-Sep-97
3406     ----------------------
3408     1. The functions are now thread-safe, with the caveat that the global variables
3409     containing pointers to malloc() and free() or alternative functions are the
3410     same for all threads.
3412     2. Get pcre_study() to generate a bitmap of initial characters for non-
3413     anchored patterns when this is possible, and use it if passed to pcre_exec().
3416     Version 0.93 15-Sep-97
3417     ----------------------
3419     1. /(b)|(:+)/ was computing an incorrect first character.
3421     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3422     but not actually doing anything yet.
3424     3. Treat "-" characters in classes that cannot be part of ranges as literals,
3425     as Perl does (e.g. [-az] or [az-]).
3427     4. Set the anchored flag if a branch starts with .* or .*? because that tests
3428     all possible positions.
3430     5. Split up into different modules to avoid including unneeded functions in a
3431     compiled binary. However, compile and exec are still in one module. The "study"
3432     function is split off.
3434     6. The character tables are now in a separate module whose source is generated
3435     by an auxiliary program - but can then be edited by hand if required. There are
3436     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3437     toupper() in the code.
3439     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3440     make them global. Abolish the function for setting them, as the caller can now
3441     set them directly.
3444     Version 0.92 11-Sep-97
3445     ----------------------
3447     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3448     (e.g. /a{1,3}/) was broken (I mis-optimized it).
3450     2. Caseless matching was not working in character classes if the characters in
3451     the pattern were in upper case.
3453     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3455     4. Make PCRE_ANCHORED public and accept as a compile option.
3457     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3458     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3459     pass them.
3461     6. Give an error if bad option bits passed at compile or run time.
3463     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3464     pcretest to cause it to pass that flag.
3466     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3467     options, and the first character, if set.
3469     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3472     Version 0.91 10-Sep-97
3473     ----------------------
3475     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3476     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3478     2. PCRE was looping on encountering an indefinitely repeated back reference to
3479     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3480     Perl does - treats the match as successful.
3482     ****


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

ViewVC Help
Powered by ViewVC 1.1.12