ViewVC logotype

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log

Revision 360 - (hide annotations) (download)
Wed Jul 9 20:00:28 2008 UTC (6 years, 9 months ago) by ph10
File size: 158348 byte(s)
Fix CVE-2008-2371 (outer level option with alternatives caused crash).

1 nigel 3 ChangeLog for PCRE
2     ------------------
4 ph10 349 Version 8.0 02 Jul-08
5     ---------------------
7     1. Replaced UCP searching code with optimized version as implemented for Ad
8     Muncher (http://www.admuncher.com/) by Peter Kankowski. This uses a two-
9 ph10 351 stage table and inline lookup instead of a function, giving speed ups of 2
10     to 5 times on some simple patterns that I tested. Permission was given to
11     distribute the MultiStage2.py script that generates the tables (it's not in
12     the tarball, but is in the Subversion repository).
13 ph10 349
14 ph10 351 2. Updated the Unicode datatables to Unicode 5.1.0. This adds yet more
15     scripts.
16 ph10 358
17 ph10 353 3. Change 12 for 7.7 introduced a bug in pcre_study() when a pattern contained
18     a group with a zero qualifier. The result of the study could be incorrect,
19 ph10 358 or the function might crash, depending on the pattern.
21     4. Caseless matching was not working for non-ASCII characters in back
22 ph10 354 references. For example, /(\x{de})\1/8i was not matching \x{de}\x{fe}.
23 ph10 358 It now works when Unicode Property Support is available.
25 ph10 355 5. In pcretest, an escape such as \x{de} in the data was always generating
26     a UTF-8 string, even in non-UTF-8 mode. Now it generates a single byte in
27     non-UTF-8 mode. If the value is greater than 255, it gives a warning about
28 ph10 358 truncation.
29 ph10 349
30 ph10 358 6. Minor bugfix in pcrecpp.cc (change "" == ... to NULL == ...).
32 ph10 357 7. Added two (int) casts to pcregrep when printing the difference of two
33     pointers, in case they are 64-bit values.
34 ph10 358
35     8. Added comments about Mac OS X stack usage to the pcrestack man page and to
36     test 2 if it fails.
37 ph10 359
38     9. Added PCRE_CALL_CONVENTION just before the names of all exported functions,
39     and a #define of that name to empty if it is not externally set. This is to
40     allow users of MSVC to set it if necessary.
42     10. The PCRE_EXP_DEFN macro which precedes exported functions was missing from
43     the convenience functions in the pcre_get.c source file.
44 ph10 360
45     11. An option change at the start of a pattern that had top-level alternatives
46     could cause overwriting and/or a crash. This command provoked a crash in
47     some environments:
49     printf "/(?i)[\xc3\xa9\xc3\xbd]|[\xc3\xa9\xc3\xbdA]/8\n" | pcretest
51     This potential security problem was recorded as CVE-2008-2371.
52 ph10 351
53 ph10 357
54 ph10 347 Version 7.7 07-May-08
55 ph10 321 ---------------------
57     1. Applied Craig's patch to sort out a long long problem: "If we can't convert
58 ph10 345 a string to a long long, pretend we don't even have a long long." This is
59 ph10 321 done by checking for the strtoq, strtoll, and _strtoi64 functions.
60 ph10 345
61 ph10 322 2. Applied Craig's patch to pcrecpp.cc to restore ABI compatibility with
62     pre-7.6 versions, which defined a global no_arg variable instead of putting
63 ph10 332 it in the RE class. (See also #8 below.)
64 ph10 345
65     3. Remove a line of dead code, identified by coverity and reported by Nuno
66     Lopes.
68 ph10 324 4. Fixed two related pcregrep bugs involving -r with --include or --exclude:
69 ph10 321
70 ph10 324 (1) The include/exclude patterns were being applied to the whole pathnames
71 ph10 345 of files, instead of just to the final components.
73 ph10 324 (2) If there was more than one level of directory, the subdirectories were
74     skipped unless they satisfied the include/exclude conditions. This is
75     inconsistent with GNU grep (and could even be seen as contrary to the
76     pcregrep specification - which I improved to make it absolutely clear).
77     The action now is always to scan all levels of directory, and just
78     apply the include/exclude patterns to regular files.
79 ph10 345
80 ph10 325 5. Added the --include_dir and --exclude_dir patterns to pcregrep, and used
81 ph10 345 --exclude_dir in the tests to avoid scanning .svn directories.
83 ph10 326 6. Applied Craig's patch to the QuoteMeta function so that it escapes the
84 ph10 345 NUL character as backslash + 0 rather than backslash + NUL, because PCRE
85 ph10 326 doesn't support NULs in patterns.
86 ph10 345
87     7. Added some missing "const"s to declarations of static tables in
88     pcre_compile.c and pcre_dfa_exec.c.
90 ph10 328 8. Applied Craig's patch to pcrecpp.cc to fix a problem in OS X that was
91 ph10 329 caused by fix #2 above. (Subsequently also a second patch to fix the
92 ph10 345 first patch. And a third patch - this was a messy problem.)
93 ph10 321
94 ph10 345 9. Applied Craig's patch to remove the use of push_back().
96 ph10 332 10. Applied Alan Lehotsky's patch to add REG_STARTEND support to the POSIX
97     matching function regexec().
98 ph10 345
99     11. Added support for the Oniguruma syntax \g<name>, \g<n>, \g'name', \g'n',
100     which, however, unlike Perl's \g{...}, are subroutine calls, not back
101 ph10 333 references. PCRE supports relative numbers with this syntax (I don't think
102     Oniguruma does).
103 ph10 345
104     12. Previously, a group with a zero repeat such as (...){0} was completely
105 ph10 335 omitted from the compiled regex. However, this means that if the group
106     was called as a subroutine from elsewhere in the pattern, things went wrong
107 ph10 345 (an internal error was given). Such groups are now left in the compiled
108     pattern, with a new opcode that causes them to be skipped at execution
109 ph10 335 time.
110 ph10 345
111 ph10 341 13. Added the PCRE_JAVASCRIPT_COMPAT option. This makes the following changes
112     to the way PCRE behaves:
113 ph10 345
114 ph10 336 (a) A lone ] character is dis-allowed (Perl treats it as data).
115 ph10 345
116     (b) A back reference to an unmatched subpattern matches an empty string
117 ph10 336 (Perl fails the current match path).
118 ph10 345
119 ph10 341 (c) A data ] in a character class must be notated as \] because if the
120 ph10 345 first data character in a class is ], it defines an empty class. (In
121 ph10 341 Perl it is not possible to have an empty class.) The empty class []
122     never matches; it forces failure and is equivalent to (*FAIL) or (?!).
123     The negative empty class [^] matches any one character, independently
124     of the DOTALL setting.
125 ph10 345
126     14. A pattern such as /(?2)[]a()b](abc)/ which had a forward reference to a
127 ph10 340 non-existent subpattern following a character class starting with ']' and
128     containing () gave an internal compiling error instead of "reference to
129     non-existent subpattern". Fortunately, when the pattern did exist, the
130 ph10 345 compiled code was correct. (When scanning forwards to check for the
131 ph10 340 existencd of the subpattern, it was treating the data ']' as terminating
132 ph10 345 the class, so got the count wrong. When actually compiling, the reference
133 ph10 340 was subsequently set up correctly.)
134 ph10 345
135 ph10 341 15. The "always fail" assertion (?!) is optimzed to (*FAIL) by pcre_compile;
136 ph10 345 it was being rejected as not supported by pcre_dfa_exec(), even though
137     other assertions are supported. I have made pcre_dfa_exec() support
138     (*FAIL).
140 ph10 342 16. The implementation of 13c above involved the invention of a new opcode,
141 ph10 345 OP_ALLANY, which is like OP_ANY but doesn't check the /s flag. Since /s
142     cannot be changed at match time, I realized I could make a small
143     improvement to matching performance by compiling OP_ALLANY instead of
144     OP_ANY for "." when DOTALL was set, and then removing the runtime tests
145     on the OP_ANY path.
147     17. Compiling pcretest on Windows with readline support failed without the
148 ph10 343 following two fixes: (1) Make the unistd.h include conditional on
149 ph10 345 HAVE_UNISTD_H; (2) #define isatty and fileno as _isatty and _fileno.
151 ph10 344 18. Changed CMakeLists.txt and cmake/FindReadline.cmake to arrange for the
152 ph10 345 ncurses library to be included for pcretest when ReadLine support is
153     requested, but also to allow for it to be overridden. This patch came from
154     Daniel Bergström.
155 ph10 347
156 ph10 346 19. There was a typo in the file ucpinternal.h where f0_rangeflag was defined
157 ph10 347 as 0x00f00000 instead of 0x00800000. Luckily, this would not have caused
158     any errors with the current Unicode tables. Thanks to Peter Kankowski for
159     spotting this.
160 ph10 324
161 ph10 332
162 ph10 319 Version 7.6 28-Jan-08
163 ph10 300 ---------------------
165 ph10 302 1. A character class containing a very large number of characters with
166     codepoints greater than 255 (in UTF-8 mode, of course) caused a buffer
167     overflow.
168 ph10 309
169     2. Patch to cut out the "long long" test in pcrecpp_unittest when
170     HAVE_LONG_LONG is not defined.
172 ph10 303 3. Applied Christian Ehrlicher's patch to update the CMake build files to
173 ph10 304 bring them up to date and include new features. This patch includes:
174 ph10 309
175 ph10 304 - Fixed PH's badly added libz and libbz2 support.
176     - Fixed a problem with static linking.
177 ph10 312 - Added pcredemo. [But later removed - see 7 below.]
178 ph10 304 - Fixed dftables problem and added an option.
179     - Added a number of HAVE_XXX tests, including HAVE_WINDOWS_H and
181     - Added readline support for pcretest.
182 ph10 309 - Added an listing of the option settings after cmake has run.
184 ph10 314 4. A user submitted a patch to Makefile that makes it easy to create
185     "pcre.dll" under mingw when using Configure/Make. I added stuff to
186     Makefile.am that cause it to include this special target, without
187     affecting anything else. Note that the same mingw target plus all
188     the other distribution libraries and programs are now supported
189     when configuring with CMake (see 6 below) instead of with
190     Configure/Make.
191 ph10 309
192 ph10 308 5. Applied Craig's patch that moves no_arg into the RE class in the C++ code.
193     This is an attempt to solve the reported problem "pcrecpp::no_arg is not
194 ph10 309 exported in the Windows port". It has not yet been confirmed that the patch
195     solves the problem, but it does no harm.
196 ph10 313
197 ph10 311 6. Applied Sheri's patch to CMakeLists.txt to add NON_STANDARD_LIB_PREFIX and
198 ph10 319 NON_STANDARD_LIB_SUFFIX for dll names built with mingw when configured
199 ph10 314 with CMake, and also correct the comment about stack recursion.
200 ph10 313
201 ph10 312 7. Remove the automatic building of pcredemo from the ./configure system and
202 ph10 313 from CMakeLists.txt. The whole idea of pcredemo.c is that it is an example
203     of a program that users should build themselves after PCRE is installed, so
204     building it automatically is not really right. What is more, it gave
205 ph10 312 trouble in some build environments.
206 ph10 300
207 ph10 319 8. Further tidies to CMakeLists.txt from Sheri and Christian.
208 ph10 308
209 ph10 319
210 ph10 298 Version 7.5 10-Jan-08
211 ph10 263 ---------------------
213     1. Applied a patch from Craig: "This patch makes it possible to 'ignore'
214     values in parens when parsing an RE using the C++ wrapper."
215 ph10 286
216 ph10 264 2. Negative specials like \S did not work in character classes in UTF-8 mode.
217     Characters greater than 255 were excluded from the class instead of being
218     included.
219 ph10 286
220     3. The same bug as (2) above applied to negated POSIX classes such as
221 ph10 265 [:^space:].
222 ph10 286
223 ph10 267 4. PCRECPP_STATIC was referenced in pcrecpp_internal.h, but nowhere was it
224     defined or documented. It seems to have been a typo for PCRE_STATIC, so
225 ph10 286 I have changed it.
227 ph10 268 5. The construct (?&) was not diagnosed as a syntax error (it referenced the
228 ph10 286 first named subpattern) and a construct such as (?&a) would reference the
229     first named subpattern whose name started with "a" (in other words, the
230 ph10 272 length check was missing). Both these problems are fixed. "Subpattern name
231     expected" is now given for (?&) (a zero-length name), and this patch also
232     makes it give the same error for \k'' (previously it complained that that
233     was a reference to a non-existent subpattern).
234 ph10 286
235 ph10 269 6. The erroneous patterns (?+-a) and (?-+a) give different error messages;
236 ph10 286 this is right because (?- can be followed by option settings as well as by
237 ph10 269 digits. I have, however, made the messages clearer.
238 ph10 286
239 ph10 270 7. Patterns such as (?(1)a|b) (a pattern that contains fewer subpatterns
240     than the number used in the conditional) now cause a compile-time error.
241     This is actually not compatible with Perl, which accepts such patterns, but
242     treats the conditional as always being FALSE (as PCRE used to), but it
243     seems to me that giving a diagnostic is better.
244 ph10 286
245 ph10 274 8. Change "alphameric" to the more common word "alphanumeric" in comments
246 ph10 275 and messages.
247 ph10 286
248     9. Fix two occurrences of "backslash" in comments that should have been
249     "backspace".
251     10. Remove two redundant lines of code that can never be obeyed (their function
252     was moved elsewhere).
254     11. The program that makes PCRE's Unicode character property table had a bug
255     which caused it to generate incorrect table entries for sequences of
256 ph10 277 characters that have the same character type, but are in different scripts.
257 ph10 286 It amalgamated them into a single range, with the script of the first of
258 ph10 277 them. In other words, some characters were in the wrong script. There were
259     thirteen such cases, affecting characters in the following ranges:
260 ph10 286
261 ph10 277 U+002b0 - U+002c1
262     U+0060c - U+0060d
263 ph10 286 U+0061e - U+00612
264 ph10 277 U+0064b - U+0065e
265     U+0074d - U+0076d
266     U+01800 - U+01805
267     U+01d00 - U+01d77
268     U+01d9b - U+01dbf
269     U+0200b - U+0200f
270     U+030fc - U+030fe
271     U+03260 - U+0327f
272     U+0fb46 - U+0fbb1
273     U+10450 - U+1049d
274 ph10 286
275 ph10 279 12. The -o option (show only the matching part of a line) for pcregrep was not
276     compatible with GNU grep in that, if there was more than one match in a
277     line, it showed only the first of them. It now behaves in the same way as
278     GNU grep.
279 ph10 286
280     13. If the -o and -v options were combined for pcregrep, it printed a blank
281     line for every non-matching line. GNU grep prints nothing, and pcregrep now
282     does the same. The return code can be used to tell if there were any
283     non-matching lines.
285 ph10 289 14. Added --file-offsets and --line-offsets to pcregrep.
287     15. The pattern (?=something)(?R) was not being diagnosed as a potentially
288 ph10 282 infinitely looping recursion. The bug was that positive lookaheads were not
289 ph10 286 being skipped when checking for a possible empty match (negative lookaheads
290     and both kinds of lookbehind were skipped).
292 ph10 289 16. Fixed two typos in the Windows-only code in pcregrep.c, and moved the
293 ph10 284 inclusion of <windows.h> to before rather than after the definition of
294     INVALID_FILE_ATTRIBUTES (patch from David Byron).
295 ph10 263
296 ph10 289 17. Specifying a possessive quantifier with a specific limit for a Unicode
297 ph10 286 character property caused pcre_compile() to compile bad code, which led at
298     runtime to PCRE_ERROR_INTERNAL (-14). Examples of patterns that caused this
299     are: /\p{Zl}{2,3}+/8 and /\p{Cc}{2}+/8. It was the possessive "+" that
300     caused the error; without that there was no problem.
301 ph10 263
302 ph10 289 18. Added --enable-pcregrep-libz and --enable-pcregrep-libbz2.
303 ph10 286
304 ph10 289 19. Added --enable-pcretest-libreadline.
305 ph10 286
306 ph10 289 20. In pcrecpp.cc, the variable 'count' was incremented twice in
307 ph10 288 RE::GlobalReplace(). As a result, the number of replacements returned was
308 ph10 298 double what it should be. I removed one of the increments, but Craig sent a
309     later patch that removed the other one (the right fix) and added unit tests
310     that check the return values (which was not done before).
311 ph10 292
312 ph10 291 21. Several CMake things:
313 ph10 287
314 ph10 291 (1) Arranged that, when cmake is used on Unix, the libraries end up with
315 ph10 292 the names libpcre and libpcreposix, not just pcre and pcreposix.
316 ph10 288
317 ph10 292 (2) The above change means that pcretest and pcregrep are now correctly
318     linked with the newly-built libraries, not previously installed ones.
319 ph10 291
321 ph10 298
322     22. In UTF-8 mode, with newline set to "any", a pattern such as .*a.*=.b.*
323     crashed when matching a string such as a\x{2029}b (note that \x{2029} is a
324     UTF-8 newline character). The key issue is that the pattern starts .*;
325     this means that the match must be either at the beginning, or after a
326     newline. The bug was in the code for advancing after a failed match and
327     checking that the new position followed a newline. It was not taking
328 ph10 294 account of UTF-8 characters correctly.
329 ph10 298
330     23. PCRE was behaving differently from Perl in the way it recognized POSIX
331     character classes. PCRE was not treating the sequence [:...:] as a
332     character class unless the ... were all letters. Perl, however, seems to
333     allow any characters between [: and :], though of course it rejects as
334     unknown any "names" that contain non-letters, because all the known class
335     names consist only of letters. Thus, Perl gives an error for [[:1234:]],
336     for example, whereas PCRE did not - it did not recognize a POSIX character
337     class. This seemed a bit dangerous, so the code has been changed to be
338     closer to Perl. The behaviour is not identical to Perl, because PCRE will
339     diagnose an unknown class for, for example, [[:l\ower:]] where Perl will
340     treat it as [[:lower:]]. However, PCRE does now give "unknown" errors where
341     Perl does, and where it didn't before.
343 ph10 296 24. Rewrite so as to remove the single use of %n from pcregrep because in some
344     Windows environments %n is disabled by default.
345 ph10 292
347 ph10 260 Version 7.4 21-Sep-07
348 ph10 230 ---------------------
350     1. Change 7.3/28 was implemented for classes by looking at the bitmap. This
351 ph10 231 means that a class such as [\s] counted as "explicit reference to CR or
352     LF". That isn't really right - the whole point of the change was to try to
353     help when there was an actual mention of one of the two characters. So now
354     the change happens only if \r or \n (or a literal CR or LF) character is
355 ph10 230 encountered.
356 ph10 231
357     2. The 32-bit options word was also used for 6 internal flags, but the numbers
358     of both had grown to the point where there were only 3 bits left.
359     Fortunately, there was spare space in the data structure, and so I have
360     moved the internal flags into a new 16-bit field to free up more option
361 ph10 230 bits.
362 ph10 231
363     3. The appearance of (?J) at the start of a pattern set the DUPNAMES option,
364     but did not set the internal JCHANGED flag - either of these is enough to
365     control the way the "get" function works - but the PCRE_INFO_JCHANGED
366     facility is supposed to tell if (?J) was ever used, so now (?J) at the
367 ph10 230 start sets both bits.
369 ph10 231 4. Added options (at build time, compile time, exec time) to change \R from
370     matching any Unicode line ending sequence to just matching CR, LF, or CRLF.
371 ph10 230
372 ph10 243 5. doc/pcresyntax.html was missing from the distribution.
374     6. Put back the definition of PCRE_ERROR_NULLWSLIMIT, for backward
375 ph10 233 compatibility, even though it is no longer used.
376 ph10 243
377     7. Added macro for snprintf to pcrecpp_unittest.cc and also for strtoll and
378 ph10 254 strtoull to pcrecpp.cc to select the available functions in WIN32 when the
379 ph10 259 windows.h file is present (where different names are used). [This was
380 ph10 257 reversed later after testing - see 16 below.]
381 ph10 243
382     8. Changed all #include <config.h> to #include "config.h". There were also
383 ph10 236 some further <pcre.h> cases that I changed to "pcre.h".
384 ph10 243
385 ph10 239 9. When pcregrep was used with the --colour option, it missed the line ending
386 ph10 243 sequence off the lines that it output.
388     10. It was pointed out to me that arrays of string pointers cause lots of
389     relocations when a shared library is dynamically loaded. A technique of
390     using a single long string with a table of offsets can drastically reduce
391     these. I have refactored PCRE in four places to do this. The result is
392 ph10 240 dramatic:
393 ph10 243
394 ph10 240 Originally: 290
395     After changing UCP table: 187
396 ph10 243 After changing error message table: 43
397 ph10 240 After changing table of "verbs" 36
398     After changing table of Posix names 22
399 ph10 243
400 ph10 240 Thanks to the folks working on Gregex for glib for this insight.
401 ph10 247
402 ph10 244 11. --disable-stack-for-recursion caused compiling to fail unless -enable-
403 ph10 247 unicode-properties was also set.
405 ph10 248 12. Updated the tests so that they work when \R is defaulted to ANYCRLF.
406 ph10 231
407 ph10 253 13. Added checks for ANY and ANYCRLF to pcrecpp.cc where it previously
408     checked only for CRLF.
409 ph10 233
410 ph10 259 14. Added casts to pcretest.c to avoid compiler warnings.
412 ph10 256 15. Added Craig's patch to various pcrecpp modules to avoid compiler warnings.
413 ph10 253
414 ph10 257 16. Added Craig's patch to remove the WINDOWS_H tests, that were not working,
415 ph10 259 and instead check for _strtoi64 explicitly, and avoid the use of snprintf()
416     entirely. This removes changes made in 7 above.
417 ph10 256
418 ph10 261 17. The CMake files have been updated, and there is now more information about
419     building with CMake in the NON-UNIX-USE document.
420 ph10 257
421 ph10 261
422 ph10 228 Version 7.3 28-Aug-07
423 ph10 157 ---------------------
425 ph10 189 1. In the rejigging of the build system that eventually resulted in 7.1, the
426     line "#include <pcre.h>" was included in pcre_internal.h. The use of angle
427     brackets there is not right, since it causes compilers to look for an
428     installed pcre.h, not the version that is in the source that is being
429     compiled (which of course may be different). I have changed it back to:
430 ph10 197
431 ph10 189 #include "pcre.h"
432 ph10 197
433     I have a vague recollection that the change was concerned with compiling in
434     different directories, but in the new build system, that is taken care of
435     by the VPATH setting the Makefile.
437 ph10 190 2. The pattern .*$ when run in not-DOTALL UTF-8 mode with newline=any failed
438     when the subject happened to end in the byte 0x85 (e.g. if the last
439     character was \x{1ec5}). *Character* 0x85 is one of the "any" newline
440     characters but of course it shouldn't be taken as a newline when it is part
441     of another character. The bug was that, for an unlimited repeat of . in
442     not-DOTALL UTF-8 mode, PCRE was advancing by bytes rather than by
443     characters when looking for a newline.
444 ph10 197
445     3. A small performance improvement in the DOTALL UTF-8 mode .* case.
447     4. Debugging: adjusted the names of opcodes for different kinds of parentheses
448     in debug output.
450 ph10 201 5. Arrange to use "%I64d" instead of "%lld" and "%I64u" instead of "%llu" for
451     long printing in the pcrecpp unittest when running under MinGW.
452 ph10 189
453 ph10 197 6. ESC_K was left out of the EBCDIC table.
454 ph10 189
455 ph10 197 7. Change 7.0/38 introduced a new limit on the number of nested non-capturing
456     parentheses; I made it 1000, which seemed large enough. Unfortunately, the
457     limit also applies to "virtual nesting" when a pattern is recursive, and in
458     this case 1000 isn't so big. I have been able to remove this limit at the
459     expense of backing off one optimization in certain circumstances. Normally,
460     when pcre_exec() would call its internal match() function recursively and
461     immediately return the result unconditionally, it uses a "tail recursion"
462     feature to save stack. However, when a subpattern that can match an empty
463     string has an unlimited repetition quantifier, it no longer makes this
464     optimization. That gives it a stack frame in which to save the data for
465     checking that an empty string has been matched. Previously this was taken
466     from the 1000-entry workspace that had been reserved. So now there is no
467     explicit limit, but more stack is used.
468 ph10 189
469 ph10 200 8. Applied Daniel's patches to solve problems with the import/export magic
470     syntax that is required for Windows, and which was going wrong for the
471     pcreposix and pcrecpp parts of the library. These were overlooked when this
472     problem was solved for the main library.
473 ph10 197
474 ph10 202 9. There were some crude static tests to avoid integer overflow when computing
475     the size of patterns that contain repeated groups with explicit upper
476     limits. As the maximum quantifier is 65535, the maximum group length was
477     set at 30,000 so that the product of these two numbers did not overflow a
478     32-bit integer. However, it turns out that people want to use groups that
479     are longer than 30,000 bytes (though not repeat them that many times).
480     Change 7.0/17 (the refactoring of the way the pattern size is computed) has
481     made it possible to implement the integer overflow checks in a much more
482     dynamic way, which I have now done. The artificial limitation on group
483     length has been removed - we now have only the limit on the total length of
484     the compiled pattern, which depends on the LINK_SIZE setting.
485 ph10 208
486     10. Fixed a bug in the documentation for get/copy named substring when
487     duplicate names are permitted. If none of the named substrings are set, the
488     functions return PCRE_ERROR_NOSUBSTRING (7); the doc said they returned an
489     empty string.
491     11. Because Perl interprets \Q...\E at a high level, and ignores orphan \E
492     instances, patterns such as [\Q\E] or [\E] or even [^\E] cause an error,
493     because the ] is interpreted as the first data character and the
494     terminating ] is not found. PCRE has been made compatible with Perl in this
495     regard. Previously, it interpreted [\Q\E] as an empty class, and [\E] could
496     cause memory overwriting.
498 ph10 206 10. Like Perl, PCRE automatically breaks an unlimited repeat after an empty
499     string has been matched (to stop an infinite loop). It was not recognizing
500 ph10 208 a conditional subpattern that could match an empty string if that
501 ph10 206 subpattern was within another subpattern. For example, it looped when
502 ph10 208 trying to match (((?(1)X|))*) but it was OK with ((?(1)X|)*) where the
503 ph10 206 condition was not nested. This bug has been fixed.
504 ph10 208
505 ph10 207 12. A pattern like \X?\d or \P{L}?\d in non-UTF-8 mode could cause a backtrack
506     past the start of the subject in the presence of bytes with the top bit
507     set, for example "\x8aBCD".
508 ph10 211
509 ph10 210 13. Added Perl 5.10 experimental backtracking controls (*FAIL), (*F), (*PRUNE),
510     (*SKIP), (*THEN), (*COMMIT), and (*ACCEPT).
511 ph10 200
512 ph10 211 14. Optimized (?!) to (*FAIL).
513 ph10 202
514 ph10 212 15. Updated the test for a valid UTF-8 string to conform to the later RFC 3629.
515     This restricts code points to be within the range 0 to 0x10FFFF, excluding
516     the "low surrogate" sequence 0xD800 to 0xDFFF. Previously, PCRE allowed the
517     full range 0 to 0x7FFFFFFF, as defined by RFC 2279. Internally, it still
518     does: it's just the validity check that is more restrictive.
519 ph10 220
520     16. Inserted checks for integer overflows during escape sequence (backslash)
521     processing, and also fixed erroneous offset values for syntax errors during
522     backslash processing.
524 ph10 214 17. Fixed another case of looking too far back in non-UTF-8 mode (cf 12 above)
525 ph10 220 for patterns like [\PPP\x8a]{1,}\x80 with the subject "A\x80".
527 ph10 215 18. An unterminated class in a pattern like (?1)\c[ with a "forward reference"
528     caused an overrun.
529 ph10 220
530     19. A pattern like (?:[\PPa*]*){8,} which had an "extended class" (one with
531     something other than just ASCII characters) inside a group that had an
532     unlimited repeat caused a loop at compile time (while checking to see
533     whether the group could match an empty string).
535 ph10 217 20. Debugging a pattern containing \p or \P could cause a crash. For example,
536     [\P{Any}] did so. (Error in the code for printing property names.)
537 ph10 210
538 ph10 220 21. An orphan \E inside a character class could cause a crash.
540     22. A repeated capturing bracket such as (A)? could cause a wild memory
541 ph10 218 reference during compilation.
542 ph10 220
543     23. There are several functions in pcre_compile() that scan along a compiled
544     expression for various reasons (e.g. to see if it's fixed length for look
545 ph10 218 behind). There were bugs in these functions when a repeated \p or \P was
546 ph10 220 present in the pattern. These operators have additional parameters compared
547     with \d, etc, and these were not being taken into account when moving along
548 ph10 218 the compiled data. Specifically:
549 ph10 220
550     (a) A item such as \p{Yi}{3} in a lookbehind was not treated as fixed
551     length.
553     (b) An item such as \pL+ within a repeated group could cause crashes or
554 ph10 218 loops.
555 ph10 220
556 ph10 218 (c) A pattern such as \p{Yi}+(\P{Yi}+)(?1) could give an incorrect
557 ph10 220 "reference to non-existent subpattern" error.
559 ph10 221 (d) A pattern like (\P{Yi}{2}\277)? could loop at compile time.
561 ph10 219 24. A repeated \S or \W in UTF-8 mode could give wrong answers when multibyte
562 ph10 220 characters were involved (for example /\S{2}/8g with "A\x{a3}BC").
563 ph10 211
564 ph10 222 25. Using pcregrep in multiline, inverted mode (-Mv) caused it to loop.
565 ph10 218
566 ph10 227 26. Patterns such as [\P{Yi}A] which include \p or \P and just one other
567 ph10 223 character were causing crashes (broken optimization).
568 ph10 227
569 ph10 224 27. Patterns such as (\P{Yi}*\277)* (group with possible zero repeat containing
570 ph10 227 \p or \P) caused a compile-time loop.
572 ph10 226 28. More problems have arisen in unanchored patterns when CRLF is a valid line
573     break. For example, the unstudied pattern [\r\n]A does not match the string
574     "\r\nA" because change 7.0/46 below moves the current point on by two
575     characters after failing to match at the start. However, the pattern \nA
576 ph10 227 *does* match, because it doesn't start till \n, and if [\r\n]A is studied,
577     the same is true. There doesn't seem any very clean way out of this, but
578     what I have chosen to do makes the common cases work: PCRE now takes note
579     of whether there can be an explicit match for \r or \n anywhere in the
580     pattern, and if so, 7.0/46 no longer applies. As part of this change,
581     there's a new PCRE_INFO_HASCRORLF option for finding out whether a compiled
582     pattern has explicit CR or LF references.
583 ph10 222
584 ph10 227 29. Added (*CR) etc for changing newline setting at start of pattern.
585 ph10 223
586 ph10 227
587 ph10 189 Version 7.2 19-Jun-07
588     ---------------------
590 ph10 157 1. If the fr_FR locale cannot be found for test 3, try the "french" locale,
591     which is apparently normally available under Windows.
592 ph10 159
593     2. Re-jig the pcregrep tests with different newline settings in an attempt
594     to make them independent of the local environment's newline setting.
595 ph10 157
596 ph10 160 3. Add code to configure.ac to remove -g from the CFLAGS default settings.
597 ph10 165
598 ph10 161 4. Some of the "internals" tests were previously cut out when the link size
599     was not 2, because the output contained actual offsets. The recent new
600 ph10 165 "Z" feature of pcretest means that these can be cut out, making the tests
601     usable with all link sizes.
603 ph10 164 5. Implemented Stan Switzer's goto replacement for longjmp() when not using
604     stack recursion. This gives a massive performance boost under BSD, but just
605 ph10 165 a small improvement under Linux. However, it saves one field in the frame
606 ph10 164 in all cases.
607 ph10 172
608 ph10 166 6. Added more features from the forthcoming Perl 5.10:
609 ph10 172
610 ph10 166 (a) (?-n) (where n is a string of digits) is a relative subroutine or
611     recursion call. It refers to the nth most recently opened parentheses.
612 ph10 172
613 ph10 166 (b) (?+n) is also a relative subroutine call; it refers to the nth next
614 ph10 172 to be opened parentheses.
616     (c) Conditions that refer to capturing parentheses can be specified
617 ph10 167 relatively, for example, (?(-2)... or (?(+3)...
618 ph10 172
619 ph10 168 (d) \K resets the start of the current match so that everything before
620 ph10 172 is not part of it.
622 ph10 171 (e) \k{name} is synonymous with \k<name> and \k'name' (.NET compatible).
623 ph10 172
624 ph10 171 (f) \g{name} is another synonym - part of Perl 5.10's unification of
625 ph10 172 reference syntax.
626 ph10 182
627 ph10 175 (g) (?| introduces a group in which the numbering of parentheses in each
628 ph10 182 alternative starts with the same number.
629 ph10 172
630 ph10 182 (h) \h, \H, \v, and \V match horizontal and vertical whitespace.
632 ph10 172 7. Added two new calls to pcre_fullinfo(): PCRE_INFO_OKPARTIAL and
635     8. A pattern such as (.*(.)?)* caused pcre_exec() to fail by either not
636     terminating or by crashing. Diagnosed by Viktor Griph; it was in the code
637 ph10 170 for detecting groups that can match an empty string.
638 ph10 159
639 ph10 172 9. A pattern with a very large number of alternatives (more than several
640     hundred) was running out of internal workspace during the pre-compile
641     phase, where pcre_compile() figures out how much memory will be needed. A
642     bit of new cunning has reduced the workspace needed for groups with
643     alternatives. The 1000-alternative test pattern now uses 12 bytes of
644     workspace instead of running out of the 4096 that are available.
645 ph10 182
646 ph10 176 10. Inserted some missing (unsigned int) casts to get rid of compiler warnings.
647 ph10 172
648 ph10 179 11. Applied patch from Google to remove an optimization that didn't quite work.
649     The report of the bug said:
650 ph10 182
651 ph10 179 pcrecpp::RE("a*").FullMatch("aaa") matches, while
652     pcrecpp::RE("a*?").FullMatch("aaa") does not, and
653 ph10 182 pcrecpp::RE("a*?\\z").FullMatch("aaa") does again.
654 ph10 185
655 ph10 184 12. If \p or \P was used in non-UTF-8 mode on a character greater than 127
656 ph10 185 it matched the wrong number of bytes.
657 ph10 172
658 ph10 179
659 ph10 155 Version 7.1 24-Apr-07
660 ph10 98 ---------------------
662 ph10 111 1. Applied Bob Rossi and Daniel G's patches to convert the build system to one
663 ph10 122 that is more "standard", making use of automake and other Autotools. There
664 ph10 99 is some re-arrangement of the files and adjustment of comments consequent
665     on this.
666 ph10 111
667     2. Part of the patch fixed a problem with the pcregrep tests. The test of -r
668     for recursive directory scanning broke on some systems because the files
669     are not scanned in any specific order and on different systems the order
670     was different. A call to "sort" has been inserted into RunGrepTest for the
671     approprate test as a short-term fix. In the longer term there may be an
672 ph10 100 alternative.
673 ph10 111
674 ph10 100 3. I had an email from Eric Raymond about problems translating some of PCRE's
675 ph10 111 man pages to HTML (despite the fact that I distribute HTML pages, some
676     people do their own conversions for various reasons). The problems
677     concerned the use of low-level troff macros .br and .in. I have therefore
678     removed all such uses from the man pages (some were redundant, some could
679 ph10 113 be replaced by .nf/.fi pairs). The 132html script that I use to generate
680     HTML has been updated to handle .nf/.fi and to complain if it encounters
681     .br or .in.
682 ph10 111
683 ph10 100 4. Updated comments in configure.ac that get placed in config.h.in and also
684 ph10 123 arranged for config.h to be included in the distribution, with the name
685 ph10 111 config.h.generic, for the benefit of those who have to compile without
686     Autotools (compare pcre.h, which is now distributed as pcre.h.generic).
688     5. Updated the support (such as it is) for Virtual Pascal, thanks to Stefan
689     Weber: (1) pcre_internal.h was missing some function renames; (2) updated
690 ph10 127 makevp.bat for the current PCRE, using the additional files
691 ph10 135 makevp_c.txt, makevp_l.txt, and pcregexp.pas.
692 ph10 111
693     6. A Windows user reported a minor discrepancy with test 2, which turned out
694     to be caused by a trailing space on an input line that had got lost in his
695 ph10 102 copy. The trailing space was an accident, so I've just removed it.
696 ph10 111
697 ph10 104 7. Add -Wl,-R... flags in pcre-config.in for *BSD* systems, as I'm told
698 ph10 111 that is needed.
700 ph10 105 8. Mark ucp_table (in ucptable.h) and ucp_gentype (in pcre_ucp_searchfuncs.c)
701     as "const" (a) because they are and (b) because it helps the PHP
702     maintainers who have recently made a script to detect big data structures
703 ph10 111 in the php code that should be moved to the .rodata section. I remembered
704     to update Builducptable as well, so it won't revert if ucptable.h is ever
705 ph10 105 re-created.
706 ph10 111
707     9. Added some extra #ifdef SUPPORT_UTF8 conditionals into pcretest.c,
708 ph10 107 pcre_printint.src, pcre_compile.c, pcre_study.c, and pcre_tables.c, in
709 ph10 111 order to be able to cut out the UTF-8 tables in the latter when UTF-8
710     support is not required. This saves 1.5-2K of code, which is important in
711 ph10 107 some applications.
712 ph10 111
713 ph10 110 Later: more #ifdefs are needed in pcre_ord2utf8.c and pcre_valid_utf8.c
714 ph10 111 so as not to refer to the tables, even though these functions will never be
715     called when UTF-8 support is disabled. Otherwise there are problems with a
716     shared library.
718 ph10 118 10. Fixed two bugs in the emulated memmove() function in pcre_internal.h:
720     (a) It was defining its arguments as char * instead of void *.
722 ph10 123 (b) It was assuming that all moves were upwards in memory; this was true
723     a long time ago when I wrote it, but is no longer the case.
725 ph10 118 The emulated memove() is provided for those environments that have neither
726 ph10 123 memmove() nor bcopy(). I didn't think anyone used it these days, but that
727 ph10 118 is clearly not the case, as these two bugs were recently reported.
728 ph10 123
729 ph10 111 11. The script PrepareRelease is now distributed: it calls 132html, CleanTxt,
730 ph10 123 and Detrail to create the HTML documentation, the .txt form of the man
731     pages, and it removes trailing spaces from listed files. It also creates
732     pcre.h.generic and config.h.generic from pcre.h and config.h. In the latter
733     case, it wraps all the #defines with #ifndefs. This script should be run
734 ph10 111 before "make dist".
735 ph10 123
736 ph10 115 12. Fixed two fairly obscure bugs concerned with quantified caseless matching
737     with Unicode property support.
738 ph10 123
739     (a) For a maximizing quantifier, if the two different cases of the
740     character were of different lengths in their UTF-8 codings (there are
741     some cases like this - I found 11), and the matching function had to
742 ph10 115 back up over a mixture of the two cases, it incorrectly assumed they
743     were both the same length.
744 ph10 123
745     (b) When PCRE was configured to use the heap rather than the stack for
746     recursion during matching, it was not correctly preserving the data for
747     the other case of a UTF-8 character when checking ahead for a match
748     while processing a minimizing repeat. If the check also involved
749 ph10 115 matching a wide character, but failed, corruption could cause an
750     erroneous result when trying to check for a repeat of the original
751     character.
752 ph10 123
753 ph10 116 13. Some tidying changes to the testing mechanism:
754 ph10 98
755 ph10 116 (a) The RunTest script now detects the internal link size and whether there
756     is UTF-8 and UCP support by running ./pcretest -C instead of relying on
757 ph10 123 values substituted by "configure". (The RunGrepTest script already did
758     this for UTF-8.) The configure.ac script no longer substitutes the
759     relevant variables.
761 ph10 116 (b) The debugging options /B and /D in pcretest show the compiled bytecode
762     with length and offset values. This means that the output is different
763     for different internal link sizes. Test 2 is skipped for link sizes
764     other than 2 because of this, bypassing the problem. Unfortunately,
765     there was also a test in test 3 (the locale tests) that used /B and
766 ph10 123 failed for link sizes other than 2. Rather than cut the whole test out,
767     I have added a new /Z option to pcretest that replaces the length and
768     offset values with spaces. This is now used to make test 3 independent
769 ph10 122 of link size. (Test 2 will be tidied up later.)
770 ph10 123
771     14. If erroroffset was passed as NULL to pcre_compile, it provoked a
772 ph10 122 segmentation fault instead of returning the appropriate error message.
773 ph10 134
774 ph10 131 15. In multiline mode when the newline sequence was set to "any", the pattern
775 ph10 134 ^$ would give a match between the \r and \n of a subject such as "A\r\nB".
776     This doesn't seem right; it now treats the CRLF combination as the line
777     ending, and so does not match in that case. It's only a pattern such as ^$
778     that would hit this one: something like ^ABC$ would have failed after \r
779     and then tried again after \r\n.
781 ph10 131 16. Changed the comparison command for RunGrepTest from "diff -u" to "diff -ub"
782 ph10 134 in an attempt to make files that differ only in their line terminators
783     compare equal. This works on Linux.
784 ph10 142
785 ph10 141 17. Under certain error circumstances pcregrep might try to free random memory
786     as it exited. This is now fixed, thanks to valgrind.
787 ph10 142
788 ph10 141 19. In pcretest, if the pattern /(?m)^$/g<any> was matched against the string
789 ph10 142 "abc\r\n\r\n", it found an unwanted second match after the second \r. This
790     was because its rules for how to advance for /g after matching an empty
791 ph10 143 string at the end of a line did not allow for this case. They now check for
792     it specially.
793 ph10 150
794     20. pcretest is supposed to handle patterns and data of any length, by
795     extending its buffers when necessary. It was getting this wrong when the
796 ph10 147 buffer for a data line had to be extended.
797 ph10 150
798 ph10 149 21. Added PCRE_NEWLINE_ANYCRLF which is like ANY, but matches only CR, LF, or
799 ph10 150 CRLF as a newline sequence.
800 ph10 152
801 ph10 151 22. Code for handling Unicode properties in pcre_dfa_exec() wasn't being cut
802 ph10 152 out by #ifdef SUPPORT_UCP. This did no harm, as it could never be used, but
803     I have nevertheless tidied it up.
804 ph10 111
805 ph10 152 23. Added some casts to kill warnings from HP-UX ia64 compiler.
806 ph10 123
807 ph10 153 24. Added a man page for pcre-config.
808 ph10 152
809 ph10 153
810 nigel 93 Version 7.0 19-Dec-06
811     ---------------------
813     1. Fixed a signed/unsigned compiler warning in pcre_compile.c, shown up by
814     moving to gcc 4.1.1.
816     2. The -S option for pcretest uses setrlimit(); I had omitted to #include
817     sys/time.h, which is documented as needed for this function. It doesn't
818     seem to matter on Linux, but it showed up on some releases of OS X.
820     3. It seems that there are systems where bytes whose values are greater than
821     127 match isprint() in the "C" locale. The "C" locale should be the
822     default when a C program starts up. In most systems, only ASCII printing
823     characters match isprint(). This difference caused the output from pcretest
824     to vary, making some of the tests fail. I have changed pcretest so that:
826     (a) When it is outputting text in the compiled version of a pattern, bytes
827     other than 32-126 are always shown as hex escapes.
829     (b) When it is outputting text that is a matched part of a subject string,
830     it does the same, unless a different locale has been set for the match
831     (using the /L modifier). In this case, it uses isprint() to decide.
833     4. Fixed a major bug that caused incorrect computation of the amount of memory
834     required for a compiled pattern when options that changed within the
835     pattern affected the logic of the preliminary scan that determines the
836     length. The relevant options are -x, and -i in UTF-8 mode. The result was
837     that the computed length was too small. The symptoms of this bug were
838     either the PCRE error "internal error: code overflow" from pcre_compile(),
839     or a glibc crash with a message such as "pcretest: free(): invalid next
840     size (fast)". Examples of patterns that provoked this bug (shown in
841     pcretest format) are:
843     /(?-x: )/x
844     /(?x)(?-x: \s*#\s*)/
845     /((?i)[\x{c0}])/8
846     /(?i:[\x{c0}])/8
848     HOWEVER: Change 17 below makes this fix obsolete as the memory computation
849     is now done differently.
851     5. Applied patches from Google to: (a) add a QuoteMeta function to the C++
852     wrapper classes; (b) implement a new function in the C++ scanner that is
853     more efficient than the old way of doing things because it avoids levels of
854     recursion in the regex matching; (c) add a paragraph to the documentation
855     for the FullMatch() function.
857     6. The escape sequence \n was being treated as whatever was defined as
858     "newline". Not only was this contrary to the documentation, which states
859     that \n is character 10 (hex 0A), but it also went horribly wrong when
860     "newline" was defined as CRLF. This has been fixed.
862     7. In pcre_dfa_exec.c the value of an unsigned integer (the variable called c)
863     was being set to -1 for the "end of line" case (supposedly a value that no
864     character can have). Though this value is never used (the check for end of
865     line is "zero bytes in current character"), it caused compiler complaints.
866     I've changed it to 0xffffffff.
868     8. In pcre_version.c, the version string was being built by a sequence of
869     C macros that, in the event of PCRE_PRERELEASE being defined as an empty
870     string (as it is for production releases) called a macro with an empty
871     argument. The C standard says the result of this is undefined. The gcc
872     compiler treats it as an empty string (which was what was wanted) but it is
873     reported that Visual C gives an error. The source has been hacked around to
874     avoid this problem.
876     9. On the advice of a Windows user, included <io.h> and <fcntl.h> in Windows
877     builds of pcretest, and changed the call to _setmode() to use _O_BINARY
878     instead of 0x8000. Made all the #ifdefs test both _WIN32 and WIN32 (not all
879     of them did).
881     10. Originally, pcretest opened its input and output without "b"; then I was
882     told that "b" was needed in some environments, so it was added for release
883     5.0 to both the input and output. (It makes no difference on Unix-like
884     systems.) Later I was told that it is wrong for the input on Windows. I've
885     now abstracted the modes into two macros, to make it easier to fiddle with
886     them, and removed "b" from the input mode under Windows.
888     11. Added pkgconfig support for the C++ wrapper library, libpcrecpp.
890     12. Added -help and --help to pcretest as an official way of being reminded
891     of the options.
893     13. Removed some redundant semicolons after macro calls in pcrecpparg.h.in
894     and pcrecpp.cc because they annoy compilers at high warning levels.
896     14. A bit of tidying/refactoring in pcre_exec.c in the main bumpalong loop.
898     15. Fixed an occurrence of == in configure.ac that should have been = (shell
899     scripts are not C programs :-) and which was not noticed because it works
900     on Linux.
902     16. pcretest is supposed to handle any length of pattern and data line (as one
903     line or as a continued sequence of lines) by extending its input buffer if
904     necessary. This feature was broken for very long pattern lines, leading to
905     a string of junk being passed to pcre_compile() if the pattern was longer
906     than about 50K.
908     17. I have done a major re-factoring of the way pcre_compile() computes the
909     amount of memory needed for a compiled pattern. Previously, there was code
910     that made a preliminary scan of the pattern in order to do this. That was
911     OK when PCRE was new, but as the facilities have expanded, it has become
912     harder and harder to keep it in step with the real compile phase, and there
913     have been a number of bugs (see for example, 4 above). I have now found a
914     cunning way of running the real compile function in a "fake" mode that
915     enables it to compute how much memory it would need, while actually only
916     ever using a few hundred bytes of working memory and without too many
917     tests of the mode. This should make future maintenance and development
918     easier. A side effect of this work is that the limit of 200 on the nesting
919     depth of parentheses has been removed (though this was never a serious
920     limitation, I suspect). However, there is a downside: pcre_compile() now
921     runs more slowly than before (30% or more, depending on the pattern). I
922     hope this isn't a big issue. There is no effect on runtime performance.
924     18. Fixed a minor bug in pcretest: if a pattern line was not terminated by a
925     newline (only possible for the last line of a file) and it was a
926     pattern that set a locale (followed by /Lsomething), pcretest crashed.
928     19. Added additional timing features to pcretest. (1) The -tm option now times
929     matching only, not compiling. (2) Both -t and -tm can be followed, as a
930     separate command line item, by a number that specifies the number of
931     repeats to use when timing. The default is 50000; this gives better
932     precision, but takes uncomfortably long for very large patterns.
934     20. Extended pcre_study() to be more clever in cases where a branch of a
935     subpattern has no definite first character. For example, (a*|b*)[cd] would
936     previously give no result from pcre_study(). Now it recognizes that the
937     first character must be a, b, c, or d.
939     21. There was an incorrect error "recursive call could loop indefinitely" if
940     a subpattern (or the entire pattern) that was being tested for matching an
941     empty string contained only one non-empty item after a nested subpattern.
942     For example, the pattern (?>\x{100}*)\d(?R) provoked this error
943     incorrectly, because the \d was being skipped in the check.
945     22. The pcretest program now has a new pattern option /B and a command line
946     option -b, which is equivalent to adding /B to every pattern. This causes
947     it to show the compiled bytecode, without the additional information that
948     -d shows. The effect of -d is now the same as -b with -i (and similarly, /D
949     is the same as /B/I).
951     23. A new optimization is now able automatically to treat some sequences such
952     as a*b as a*+b. More specifically, if something simple (such as a character
953     or a simple class like \d) has an unlimited quantifier, and is followed by
954     something that cannot possibly match the quantified thing, the quantifier
955     is automatically "possessified".
957     24. A recursive reference to a subpattern whose number was greater than 39
958     went wrong under certain circumstances in UTF-8 mode. This bug could also
959     have affected the operation of pcre_study().
961     25. Realized that a little bit of performance could be had by replacing
962     (c & 0xc0) == 0xc0 with c >= 0xc0 when processing UTF-8 characters.
964     26. Timing data from pcretest is now shown to 4 decimal places instead of 3.
966     27. Possessive quantifiers such as a++ were previously implemented by turning
967     them into atomic groups such as ($>a+). Now they have their own opcodes,
968     which improves performance. This includes the automatically created ones
969     from 23 above.
971     28. A pattern such as (?=(\w+))\1: which simulates an atomic group using a
972     lookahead was broken if it was not anchored. PCRE was mistakenly expecting
973     the first matched character to be a colon. This applied both to named and
974     numbered groups.
976     29. The ucpinternal.h header file was missing its idempotency #ifdef.
978     30. I was sent a "project" file called libpcre.a.dev which I understand makes
979     building PCRE on Windows easier, so I have included it in the distribution.
981     31. There is now a check in pcretest against a ridiculously large number being
982     returned by pcre_exec() or pcre_dfa_exec(). If this happens in a /g or /G
983     loop, the loop is abandoned.
985     32. Forward references to subpatterns in conditions such as (?(2)...) where
986     subpattern 2 is defined later cause pcre_compile() to search forwards in
987     the pattern for the relevant set of parentheses. This search went wrong
988     when there were unescaped parentheses in a character class, parentheses
989     escaped with \Q...\E, or parentheses in a #-comment in /x mode.
991     33. "Subroutine" calls and backreferences were previously restricted to
992     referencing subpatterns earlier in the regex. This restriction has now
993     been removed.
995     34. Added a number of extra features that are going to be in Perl 5.10. On the
996     whole, these are just syntactic alternatives for features that PCRE had
997     previously implemented using the Python syntax or my own invention. The
998     other formats are all retained for compatibility.
1000     (a) Named groups can now be defined as (?<name>...) or (?'name'...) as well
1001     as (?P<name>...). The new forms, as well as being in Perl 5.10, are
1002     also .NET compatible.
1004     (b) A recursion or subroutine call to a named group can now be defined as
1005     (?&name) as well as (?P>name).
1007     (c) A backreference to a named group can now be defined as \k<name> or
1008     \k'name' as well as (?P=name). The new forms, as well as being in Perl
1009     5.10, are also .NET compatible.
1011     (d) A conditional reference to a named group can now use the syntax
1012     (?(<name>) or (?('name') as well as (?(name).
1014     (e) A "conditional group" of the form (?(DEFINE)...) can be used to define
1015     groups (named and numbered) that are never evaluated inline, but can be
1016     called as "subroutines" from elsewhere. In effect, the DEFINE condition
1017     is always false. There may be only one alternative in such a group.
1019     (f) A test for recursion can be given as (?(R1).. or (?(R&name)... as well
1020     as the simple (?(R). The condition is true only if the most recent
1021     recursion is that of the given number or name. It does not search out
1022     through the entire recursion stack.
1024     (g) The escape \gN or \g{N} has been added, where N is a positive or
1025     negative number, specifying an absolute or relative reference.
1027     35. Tidied to get rid of some further signed/unsigned compiler warnings and
1028     some "unreachable code" warnings.
1030     36. Updated the Unicode property tables to Unicode version 5.0.0. Amongst other
1031     things, this adds five new scripts.
1033     37. Perl ignores orphaned \E escapes completely. PCRE now does the same.
1034     There were also incompatibilities regarding the handling of \Q..\E inside
1035     character classes, for example with patterns like [\Qa\E-\Qz\E] where the
1036     hyphen was adjacent to \Q or \E. I hope I've cleared all this up now.
1038     38. Like Perl, PCRE detects when an indefinitely repeated parenthesized group
1039     matches an empty string, and forcibly breaks the loop. There were bugs in
1040     this code in non-simple cases. For a pattern such as ^(a()*)* matched
1041     against aaaa the result was just "a" rather than "aaaa", for example. Two
1042     separate and independent bugs (that affected different cases) have been
1043     fixed.
1045     39. Refactored the code to abolish the use of different opcodes for small
1046     capturing bracket numbers. This is a tidy that I avoided doing when I
1047     removed the limit on the number of capturing brackets for 3.5 back in 2001.
1048     The new approach is not only tidier, it makes it possible to reduce the
1049     memory needed to fix the previous bug (38).
1051     40. Implemented PCRE_NEWLINE_ANY to recognize any of the Unicode newline
1052     sequences (http://unicode.org/unicode/reports/tr18/) as "newline" when
1053     processing dot, circumflex, or dollar metacharacters, or #-comments in /x
1054     mode.
1056     41. Add \R to match any Unicode newline sequence, as suggested in the Unicode
1057     report.
1059     42. Applied patch, originally from Ari Pollak, modified by Google, to allow
1060     copy construction and assignment in the C++ wrapper.
1062     43. Updated pcregrep to support "--newline=any". In the process, I fixed a
1063     couple of bugs that could have given wrong results in the "--newline=crlf"
1064     case.
1066     44. Added a number of casts and did some reorganization of signed/unsigned int
1067     variables following suggestions from Dair Grant. Also renamed the variable
1068     "this" as "item" because it is a C++ keyword.
1070     45. Arranged for dftables to add
1072     #include "pcre_internal.h"
1074     to pcre_chartables.c because without it, gcc 4.x may remove the array
1075     definition from the final binary if PCRE is built into a static library and
1076     dead code stripping is activated.
1078     46. For an unanchored pattern, if a match attempt fails at the start of a
1079     newline sequence, and the newline setting is CRLF or ANY, and the next two
1080     characters are CRLF, advance by two characters instead of one.
1083 nigel 91 Version 6.7 04-Jul-06
1084     ---------------------
1086     1. In order to handle tests when input lines are enormously long, pcretest has
1087     been re-factored so that it automatically extends its buffers when
1088     necessary. The code is crude, but this _is_ just a test program. The
1089     default size has been increased from 32K to 50K.
1091     2. The code in pcre_study() was using the value of the re argument before
1092     testing it for NULL. (Of course, in any sensible call of the function, it
1093     won't be NULL.)
1095     3. The memmove() emulation function in pcre_internal.h, which is used on
1096     systems that lack both memmove() and bcopy() - that is, hardly ever -
1097     was missing a "static" storage class specifier.
1099     4. When UTF-8 mode was not set, PCRE looped when compiling certain patterns
1100     containing an extended class (one that cannot be represented by a bitmap
1101     because it contains high-valued characters or Unicode property items, e.g.
1102     [\pZ]). Almost always one would set UTF-8 mode when processing such a
1103     pattern, but PCRE should not loop if you do not (it no longer does).
1104     [Detail: two cases were found: (a) a repeated subpattern containing an
1105     extended class; (b) a recursive reference to a subpattern that followed a
1106     previous extended class. It wasn't skipping over the extended class
1107     correctly when UTF-8 mode was not set.]
1109     5. A negated single-character class was not being recognized as fixed-length
1110     in lookbehind assertions such as (?<=[^f]), leading to an incorrect
1111     compile error "lookbehind assertion is not fixed length".
1113     6. The RunPerlTest auxiliary script was showing an unexpected difference
1114     between PCRE and Perl for UTF-8 tests. It turns out that it is hard to
1115     write a Perl script that can interpret lines of an input file either as
1116     byte characters or as UTF-8, which is what "perltest" was being required to
1117     do for the non-UTF-8 and UTF-8 tests, respectively. Essentially what you
1118     can't do is switch easily at run time between having the "use utf8;" pragma
1119     or not. In the end, I fudged it by using the RunPerlTest script to insert
1120     "use utf8;" explicitly for the UTF-8 tests.
1122     7. In multiline (/m) mode, PCRE was matching ^ after a terminating newline at
1123     the end of the subject string, contrary to the documentation and to what
1124     Perl does. This was true of both matching functions. Now it matches only at
1125     the start of the subject and immediately after *internal* newlines.
1127     8. A call of pcre_fullinfo() from pcretest to get the option bits was passing
1128     a pointer to an int instead of a pointer to an unsigned long int. This
1129     caused problems on 64-bit systems.
1131     9. Applied a patch from the folks at Google to pcrecpp.cc, to fix "another
1132     instance of the 'standard' template library not being so standard".
1134     10. There was no check on the number of named subpatterns nor the maximum
1135     length of a subpattern name. The product of these values is used to compute
1136     the size of the memory block for a compiled pattern. By supplying a very
1137     long subpattern name and a large number of named subpatterns, the size
1138     computation could be caused to overflow. This is now prevented by limiting
1139     the length of names to 32 characters, and the number of named subpatterns
1140     to 10,000.
1142     11. Subpatterns that are repeated with specific counts have to be replicated in
1143     the compiled pattern. The size of memory for this was computed from the
1144     length of the subpattern and the repeat count. The latter is limited to
1145     65535, but there was no limit on the former, meaning that integer overflow
1146     could in principle occur. The compiled length of a repeated subpattern is
1147     now limited to 30,000 bytes in order to prevent this.
1149     12. Added the optional facility to have named substrings with the same name.
1151     13. Added the ability to use a named substring as a condition, using the
1152     Python syntax: (?(name)yes|no). This overloads (?(R)... and names that
1153     are numbers (not recommended). Forward references are permitted.
1155     14. Added forward references in named backreferences (if you see what I mean).
1157     15. In UTF-8 mode, with the PCRE_DOTALL option set, a quantified dot in the
1158     pattern could run off the end of the subject. For example, the pattern
1159     "(?s)(.{1,5})"8 did this with the subject "ab".
1161     16. If PCRE_DOTALL or PCRE_MULTILINE were set, pcre_dfa_exec() behaved as if
1162     PCRE_CASELESS was set when matching characters that were quantified with ?
1163     or *.
1165     17. A character class other than a single negated character that had a minimum
1166     but no maximum quantifier - for example [ab]{6,} - was not handled
1167     correctly by pce_dfa_exec(). It would match only one character.
1169     18. A valid (though odd) pattern that looked like a POSIX character
1170     class but used an invalid character after [ (for example [[,abc,]]) caused
1171     pcre_compile() to give the error "Failed: internal error: code overflow" or
1172     in some cases to crash with a glibc free() error. This could even happen if
1173     the pattern terminated after [[ but there just happened to be a sequence of
1174     letters, a binary zero, and a closing ] in the memory that followed.
1176     19. Perl's treatment of octal escapes in the range \400 to \777 has changed
1177     over the years. Originally (before any Unicode support), just the bottom 8
1178     bits were taken. Thus, for example, \500 really meant \100. Nowadays the
1179     output from "man perlunicode" includes this:
1181     The regular expression compiler produces polymorphic opcodes. That
1182     is, the pattern adapts to the data and automatically switches to
1183     the Unicode character scheme when presented with Unicode data--or
1184     instead uses a traditional byte scheme when presented with byte
1185     data.
1187     Sadly, a wide octal escape does not cause a switch, and in a string with
1188     no other multibyte characters, these octal escapes are treated as before.
1189     Thus, in Perl, the pattern /\500/ actually matches \100 but the pattern
1190     /\500|\x{1ff}/ matches \500 or \777 because the whole thing is treated as a
1191     Unicode string.
1193     I have not perpetrated such confusion in PCRE. Up till now, it took just
1194     the bottom 8 bits, as in old Perl. I have now made octal escapes with
1195     values greater than \377 illegal in non-UTF-8 mode. In UTF-8 mode they
1196     translate to the appropriate multibyte character.
1198     29. Applied some refactoring to reduce the number of warnings from Microsoft
1199     and Borland compilers. This has included removing the fudge introduced
1200     seven years ago for the OS/2 compiler (see 2.02/2 below) because it caused
1201     a warning about an unused variable.
1203     21. PCRE has not included VT (character 0x0b) in the set of whitespace
1204     characters since release 4.0, because Perl (from release 5.004) does not.
1205     [Or at least, is documented not to: some releases seem to be in conflict
1206     with the documentation.] However, when a pattern was studied with
1207     pcre_study() and all its branches started with \s, PCRE still included VT
1208     as a possible starting character. Of course, this did no harm; it just
1209     caused an unnecessary match attempt.
1211     22. Removed a now-redundant internal flag bit that recorded the fact that case
1212     dependency changed within the pattern. This was once needed for "required
1213     byte" processing, but is no longer used. This recovers a now-scarce options
1214     bit. Also moved the least significant internal flag bit to the most-
1215     significant bit of the word, which was not previously used (hangover from
1216     the days when it was an int rather than a uint) to free up another bit for
1217     the future.
1219     23. Added support for CRLF line endings as well as CR and LF. As well as the
1220     default being selectable at build time, it can now be changed at runtime
1221     via the PCRE_NEWLINE_xxx flags. There are now options for pcregrep to
1222     specify that it is scanning data with non-default line endings.
1224     24. Changed the definition of CXXLINK to make it agree with the definition of
1225     LINK in the Makefile, by replacing LDFLAGS to CXXFLAGS.
1227     25. Applied Ian Taylor's patches to avoid using another stack frame for tail
1228     recursions. This makes a big different to stack usage for some patterns.
1230     26. If a subpattern containing a named recursion or subroutine reference such
1231     as (?P>B) was quantified, for example (xxx(?P>B)){3}, the calculation of
1232     the space required for the compiled pattern went wrong and gave too small a
1233     value. Depending on the environment, this could lead to "Failed: internal
1234     error: code overflow at offset 49" or "glibc detected double free or
1235     corruption" errors.
1237     27. Applied patches from Google (a) to support the new newline modes and (b) to
1238     advance over multibyte UTF-8 characters in GlobalReplace.
1240     28. Change free() to pcre_free() in pcredemo.c. Apparently this makes a
1241     difference for some implementation of PCRE in some Windows version.
1243     29. Added some extra testing facilities to pcretest:
1245     \q<number> in a data line sets the "match limit" value
1246     \Q<number> in a data line sets the "match recursion limt" value
1247     -S <number> sets the stack size, where <number> is in megabytes
1249     The -S option isn't available for Windows.
1252 nigel 89 Version 6.6 06-Feb-06
1253     ---------------------
1255     1. Change 16(a) for 6.5 broke things, because PCRE_DATA_SCOPE was not defined
1256     in pcreposix.h. I have copied the definition from pcre.h.
1258     2. Change 25 for 6.5 broke compilation in a build directory out-of-tree
1259     because pcre.h is no longer a built file.
1261     3. Added Jeff Friedl's additional debugging patches to pcregrep. These are
1262     not normally included in the compiled code.
1265 nigel 87 Version 6.5 01-Feb-06
1266     ---------------------
1268     1. When using the partial match feature with pcre_dfa_exec(), it was not
1269     anchoring the second and subsequent partial matches at the new starting
1270     point. This could lead to incorrect results. For example, with the pattern
1271     /1234/, partially matching against "123" and then "a4" gave a match.
1273     2. Changes to pcregrep:
1275     (a) All non-match returns from pcre_exec() were being treated as failures
1276     to match the line. Now, unless the error is PCRE_ERROR_NOMATCH, an
1277     error message is output. Some extra information is given for the
1279     probably the only errors that are likely to be caused by users (by
1280     specifying a regex that has nested indefinite repeats, for instance).
1281     If there are more than 20 of these errors, pcregrep is abandoned.
1283     (b) A binary zero was treated as data while matching, but terminated the
1284     output line if it was written out. This has been fixed: binary zeroes
1285     are now no different to any other data bytes.
1287     (c) Whichever of the LC_ALL or LC_CTYPE environment variables is set is
1288     used to set a locale for matching. The --locale=xxxx long option has
1289     been added (no short equivalent) to specify a locale explicitly on the
1290     pcregrep command, overriding the environment variables.
1292     (d) When -B was used with -n, some line numbers in the output were one less
1293     than they should have been.
1295     (e) Added the -o (--only-matching) option.
1297     (f) If -A or -C was used with -c (count only), some lines of context were
1298     accidentally printed for the final match.
1300     (g) Added the -H (--with-filename) option.
1302     (h) The combination of options -rh failed to suppress file names for files
1303     that were found from directory arguments.
1305     (i) Added the -D (--devices) and -d (--directories) options.
1307     (j) Added the -F (--fixed-strings) option.
1309     (k) Allow "-" to be used as a file name for -f as well as for a data file.
1311     (l) Added the --colo(u)r option.
1313     (m) Added Jeffrey Friedl's -S testing option, but within #ifdefs so that it
1314     is not present by default.
1316     3. A nasty bug was discovered in the handling of recursive patterns, that is,
1317     items such as (?R) or (?1), when the recursion could match a number of
1318     alternatives. If it matched one of the alternatives, but subsequently,
1319     outside the recursion, there was a failure, the code tried to back up into
1320     the recursion. However, because of the way PCRE is implemented, this is not
1321     possible, and the result was an incorrect result from the match.
1323     In order to prevent this happening, the specification of recursion has
1324     been changed so that all such subpatterns are automatically treated as
1325     atomic groups. Thus, for example, (?R) is treated as if it were (?>(?R)).
1327     4. I had overlooked the fact that, in some locales, there are characters for
1328     which isalpha() is true but neither isupper() nor islower() are true. In
1329     the fr_FR locale, for instance, the \xAA and \xBA characters (ordmasculine
1330     and ordfeminine) are like this. This affected the treatment of \w and \W
1331     when they appeared in character classes, but not when they appeared outside
1332     a character class. The bit map for "word" characters is now created
1333     separately from the results of isalnum() instead of just taking it from the
1334     upper, lower, and digit maps. (Plus the underscore character, of course.)
1336     5. The above bug also affected the handling of POSIX character classes such as
1337     [[:alpha:]] and [[:alnum:]]. These do not have their own bit maps in PCRE's
1338     permanent tables. Instead, the bit maps for such a class were previously
1339     created as the appropriate unions of the upper, lower, and digit bitmaps.
1340     Now they are created by subtraction from the [[:word:]] class, which has
1341     its own bitmap.
1343     6. The [[:blank:]] character class matches horizontal, but not vertical space.
1344     It is created by subtracting the vertical space characters (\x09, \x0a,
1345     \x0b, \x0c) from the [[:space:]] bitmap. Previously, however, the
1346     subtraction was done in the overall bitmap for a character class, meaning
1347     that a class such as [\x0c[:blank:]] was incorrect because \x0c would not
1348     be recognized. This bug has been fixed.
1350     7. Patches from the folks at Google:
1352     (a) pcrecpp.cc: "to handle a corner case that may or may not happen in
1353     real life, but is still worth protecting against".
1355     (b) pcrecpp.cc: "corrects a bug when negative radixes are used with
1356     regular expressions".
1358     (c) pcre_scanner.cc: avoid use of std::count() because not all systems
1359     have it.
1361     (d) Split off pcrecpparg.h from pcrecpp.h and had the former built by
1362     "configure" and the latter not, in order to fix a problem somebody had
1363     with compiling the Arg class on HP-UX.
1365     (e) Improve the error-handling of the C++ wrapper a little bit.
1367     (f) New tests for checking recursion limiting.
1369     8. The pcre_memmove() function, which is used only if the environment does not
1370     have a standard memmove() function (and is therefore rarely compiled),
1371     contained two bugs: (a) use of int instead of size_t, and (b) it was not
1372     returning a result (though PCRE never actually uses the result).
1374     9. In the POSIX regexec() interface, if nmatch is specified as a ridiculously
1375     large number - greater than INT_MAX/(3*sizeof(int)) - REG_ESPACE is
1376     returned instead of calling malloc() with an overflowing number that would
1377     most likely cause subsequent chaos.
1379     10. The debugging option of pcretest was not showing the NO_AUTO_CAPTURE flag.
1381     11. The POSIX flag REG_NOSUB is now supported. When a pattern that was compiled
1382     with this option is matched, the nmatch and pmatch options of regexec() are
1383     ignored.
1385     12. Added REG_UTF8 to the POSIX interface. This is not defined by POSIX, but is
1386     provided in case anyone wants to the the POSIX interface with UTF-8
1387     strings.
1389     13. Added CXXLDFLAGS to the Makefile parameters to provide settings only on the
1390     C++ linking (needed for some HP-UX environments).
1392     14. Avoid compiler warnings in get_ucpname() when compiled without UCP support
1393     (unused parameter) and in the pcre_printint() function (omitted "default"
1394     switch label when the default is to do nothing).
1396     15. Added some code to make it possible, when PCRE is compiled as a C++
1397     library, to replace subject pointers for pcre_exec() with a smart pointer
1398     class, thus making it possible to process discontinuous strings.
1400     16. The two macros PCRE_EXPORT and PCRE_DATA_SCOPE are confusing, and perform
1401     much the same function. They were added by different people who were trying
1402     to make PCRE easy to compile on non-Unix systems. It has been suggested
1403     that PCRE_EXPORT be abolished now that there is more automatic apparatus
1404     for compiling on Windows systems. I have therefore replaced it with
1405     PCRE_DATA_SCOPE. This is set automatically for Windows; if not set it
1406     defaults to "extern" for C or "extern C" for C++, which works fine on
1407     Unix-like systems. It is now possible to override the value of PCRE_DATA_
1408     SCOPE with something explicit in config.h. In addition:
1410     (a) pcreposix.h still had just "extern" instead of either of these macros;
1411     I have replaced it with PCRE_DATA_SCOPE.
1413     (b) Functions such as _pcre_xclass(), which are internal to the library,
1414     but external in the C sense, all had PCRE_EXPORT in their definitions.
1415     This is apparently wrong for the Windows case, so I have removed it.
1416     (It makes no difference on Unix-like systems.)
1418     17. Added a new limit, MATCH_LIMIT_RECURSION, which limits the depth of nesting
1419     of recursive calls to match(). This is different to MATCH_LIMIT because
1420     that limits the total number of calls to match(), not all of which increase
1421     the depth of recursion. Limiting the recursion depth limits the amount of
1422     stack (or heap if NO_RECURSE is set) that is used. The default can be set
1423     when PCRE is compiled, and changed at run time. A patch from Google adds
1424     this functionality to the C++ interface.
1426     18. Changes to the handling of Unicode character properties:
1428     (a) Updated the table to Unicode 4.1.0.
1430     (b) Recognize characters that are not in the table as "Cn" (undefined).
1432     (c) I revised the way the table is implemented to a much improved format
1433     which includes recognition of ranges. It now supports the ranges that
1434     are defined in UnicodeData.txt, and it also amalgamates other
1435     characters into ranges. This has reduced the number of entries in the
1436     table from around 16,000 to around 3,000, thus reducing its size
1437     considerably. I realized I did not need to use a tree structure after
1438     all - a binary chop search is just as efficient. Having reduced the
1439     number of entries, I extended their size from 6 bytes to 8 bytes to
1440     allow for more data.
1442     (d) Added support for Unicode script names via properties such as \p{Han}.
1444     19. In UTF-8 mode, a backslash followed by a non-Ascii character was not
1445     matching that character.
1447     20. When matching a repeated Unicode property with a minimum greater than zero,
1448     (for example \pL{2,}), PCRE could look past the end of the subject if it
1449     reached it while seeking the minimum number of characters. This could
1450     happen only if some of the characters were more than one byte long, because
1451     there is a check for at least the minimum number of bytes.
1453     21. Refactored the implementation of \p and \P so as to be more general, to
1454     allow for more different types of property in future. This has changed the
1455     compiled form incompatibly. Anybody with saved compiled patterns that use
1456     \p or \P will have to recompile them.
1458     22. Added "Any" and "L&" to the supported property types.
1460     23. Recognize \x{...} as a code point specifier, even when not in UTF-8 mode,
1461     but give a compile time error if the value is greater than 0xff.
1463     24. The man pages for pcrepartial, pcreprecompile, and pcre_compile2 were
1464     accidentally not being installed or uninstalled.
1466     25. The pcre.h file was built from pcre.h.in, but the only changes that were
1467     made were to insert the current release number. This seemed silly, because
1468     it made things harder for people building PCRE on systems that don't run
1469     "configure". I have turned pcre.h into a distributed file, no longer built
1470     by "configure", with the version identification directly included. There is
1471     no longer a pcre.h.in file.
1473     However, this change necessitated a change to the pcre-config script as
1474     well. It is built from pcre-config.in, and one of the substitutions was the
1475     release number. I have updated configure.ac so that ./configure now finds
1476     the release number by grepping pcre.h.
1478     26. Added the ability to run the tests under valgrind.
1481 nigel 85 Version 6.4 05-Sep-05
1482     ---------------------
1484     1. Change 6.0/10/(l) to pcregrep introduced a bug that caused separator lines
1485     "--" to be printed when multiple files were scanned, even when none of the
1486     -A, -B, or -C options were used. This is not compatible with Gnu grep, so I
1487     consider it to be a bug, and have restored the previous behaviour.
1489     2. A couple of code tidies to get rid of compiler warnings.
1491     3. The pcretest program used to cheat by referring to symbols in the library
1492     whose names begin with _pcre_. These are internal symbols that are not
1493     really supposed to be visible externally, and in some environments it is
1494     possible to suppress them. The cheating is now confined to including
1495     certain files from the library's source, which is a bit cleaner.
1497     4. Renamed pcre.in as pcre.h.in to go with pcrecpp.h.in; it also makes the
1498     file's purpose clearer.
1500     5. Reorganized pcre_ucp_findchar().
1503 nigel 83 Version 6.3 15-Aug-05
1504     ---------------------
1506     1. The file libpcre.pc.in did not have general read permission in the tarball.
1508     2. There were some problems when building without C++ support:
1510     (a) If C++ support was not built, "make install" and "make test" still
1511     tried to test it.
1513     (b) There were problems when the value of CXX was explicitly set. Some
1514     changes have been made to try to fix these, and ...
1516     (c) --disable-cpp can now be used to explicitly disable C++ support.
1518     (d) The use of @CPP_OBJ@ directly caused a blank line preceded by a
1519     backslash in a target when C++ was disabled. This confuses some
1520     versions of "make", apparently. Using an intermediate variable solves
1521     this. (Same for CPP_LOBJ.)
1523     3. $(LINK_FOR_BUILD) now includes $(CFLAGS_FOR_BUILD) and $(LINK)
1524     (non-Windows) now includes $(CFLAGS) because these flags are sometimes
1525     necessary on certain architectures.
1527     4. Added a setting of -export-symbols-regex to the link command to remove
1528     those symbols that are exported in the C sense, but actually are local
1529     within the library, and not documented. Their names all begin with
1530     "_pcre_". This is not a perfect job, because (a) we have to except some
1531     symbols that pcretest ("illegally") uses, and (b) the facility isn't always
1532     available (and never for static libraries). I have made a note to try to
1533     find a way round (a) in the future.
1536 nigel 81 Version 6.2 01-Aug-05
1537     ---------------------
1539     1. There was no test for integer overflow of quantifier values. A construction
1540     such as {1111111111111111} would give undefined results. What is worse, if
1541     a minimum quantifier for a parenthesized subpattern overflowed and became
1542     negative, the calculation of the memory size went wrong. This could have
1543     led to memory overwriting.
1545     2. Building PCRE using VPATH was broken. Hopefully it is now fixed.
1547     3. Added "b" to the 2nd argument of fopen() in dftables.c, for non-Unix-like
1548     operating environments where this matters.
1550     4. Applied Giuseppe Maxia's patch to add additional features for controlling
1551     PCRE options from within the C++ wrapper.
1553     5. Named capturing subpatterns were not being correctly counted when a pattern
1554     was compiled. This caused two problems: (a) If there were more than 100
1555     such subpatterns, the calculation of the memory needed for the whole
1556     compiled pattern went wrong, leading to an overflow error. (b) Numerical
1557     back references of the form \12, where the number was greater than 9, were
1558     not recognized as back references, even though there were sufficient
1559     previous subpatterns.
1561     6. Two minor patches to pcrecpp.cc in order to allow it to compile on older
1562     versions of gcc, e.g. 2.95.4.
1565 nigel 79 Version 6.1 21-Jun-05
1566     ---------------------
1568     1. There was one reference to the variable "posix" in pcretest.c that was not
1569     surrounded by "#if !defined NOPOSIX".
1571     2. Make it possible to compile pcretest without DFA support, UTF8 support, or
1572     the cross-check on the old pcre_info() function, for the benefit of the
1573     cut-down version of PCRE that is currently imported into Exim.
1575     3. A (silly) pattern starting with (?i)(?-i) caused an internal space
1576     allocation error. I've done the easy fix, which wastes 2 bytes for sensible
1577     patterns that start (?i) but I don't think that matters. The use of (?i) is
1578     just an example; this all applies to the other options as well.
1580     4. Since libtool seems to echo the compile commands it is issuing, the output
1581     from "make" can be reduced a bit by putting "@" in front of each libtool
1582     compile command.
1584     5. Patch from the folks at Google for configure.in to be a bit more thorough
1585     in checking for a suitable C++ installation before trying to compile the
1586     C++ stuff. This should fix a reported problem when a compiler was present,
1587     but no suitable headers.
1589     6. The man pages all had just "PCRE" as their title. I have changed them to
1590     be the relevant file name. I have also arranged that these names are
1591     retained in the file doc/pcre.txt, which is a concatenation in text format
1592     of all the man pages except the little individual ones for each function.
1594     7. The NON-UNIX-USE file had not been updated for the different set of source
1595     files that come with release 6. I also added a few comments about the C++
1596     wrapper.
1599 nigel 77 Version 6.0 07-Jun-05
1600     ---------------------
1602     1. Some minor internal re-organization to help with my DFA experiments.
1604     2. Some missing #ifdef SUPPORT_UCP conditionals in pcretest and printint that
1605     didn't matter for the library itself when fully configured, but did matter
1606     when compiling without UCP support, or within Exim, where the ucp files are
1607     not imported.
1609     3. Refactoring of the library code to split up the various functions into
1610     different source modules. The addition of the new DFA matching code (see
1611     below) to a single monolithic source would have made it really too
1612     unwieldy, quite apart from causing all the code to be include in a
1613     statically linked application, when only some functions are used. This is
1614     relevant even without the DFA addition now that patterns can be compiled in
1615     one application and matched in another.
1617     The downside of splitting up is that there have to be some external
1618     functions and data tables that are used internally in different modules of
1619     the library but which are not part of the API. These have all had their
1620     names changed to start with "_pcre_" so that they are unlikely to clash
1621     with other external names.
1623     4. Added an alternate matching function, pcre_dfa_exec(), which matches using
1624     a different (DFA) algorithm. Although it is slower than the original
1625     function, it does have some advantages for certain types of matching
1626     problem.
1628     5. Upgrades to pcretest in order to test the features of pcre_dfa_exec(),
1629     including restarting after a partial match.
1631     6. A patch for pcregrep that defines INVALID_FILE_ATTRIBUTES if it is not
1632     defined when compiling for Windows was sent to me. I have put it into the
1633     code, though I have no means of testing or verifying it.
1635     7. Added the pcre_refcount() auxiliary function.
1637     8. Added the PCRE_FIRSTLINE option. This constrains an unanchored pattern to
1638     match before or at the first newline in the subject string. In pcretest,
1639     the /f option on a pattern can be used to set this.
1641     9. A repeated \w when used in UTF-8 mode with characters greater than 256
1642     would behave wrongly. This has been present in PCRE since release 4.0.
1644     10. A number of changes to the pcregrep command:
1646     (a) Refactored how -x works; insert ^(...)$ instead of setting
1647     PCRE_ANCHORED and checking the length, in preparation for adding
1648     something similar for -w.
1650     (b) Added the -w (match as a word) option.
1652     (c) Refactored the way lines are read and buffered so as to have more
1653     than one at a time available.
1655     (d) Implemented a pcregrep test script.
1657     (e) Added the -M (multiline match) option. This allows patterns to match
1658     over several lines of the subject. The buffering ensures that at least
1659     8K, or the rest of the document (whichever is the shorter) is available
1660     for matching (and similarly the previous 8K for lookbehind assertions).
1662     (f) Changed the --help output so that it now says
1664     -w, --word-regex(p)
1666     instead of two lines, one with "regex" and the other with "regexp"
1667     because that confused at least one person since the short forms are the
1668     same. (This required a bit of code, as the output is generated
1669     automatically from a table. It wasn't just a text change.)
1671     (g) -- can be used to terminate pcregrep options if the next thing isn't an
1672     option but starts with a hyphen. Could be a pattern or a path name
1673     starting with a hyphen, for instance.
1675     (h) "-" can be given as a file name to represent stdin.
1677     (i) When file names are being printed, "(standard input)" is used for
1678     the standard input, for compatibility with GNU grep. Previously
1679     "<stdin>" was used.
1681     (j) The option --label=xxx can be used to supply a name to be used for
1682     stdin when file names are being printed. There is no short form.
1684     (k) Re-factored the options decoding logic because we are going to add
1685     two more options that take data. Such options can now be given in four
1686     different ways, e.g. "-fname", "-f name", "--file=name", "--file name".
1688     (l) Added the -A, -B, and -C options for requesting that lines of context
1689     around matches be printed.
1691     (m) Added the -L option to print the names of files that do not contain
1692     any matching lines, that is, the complement of -l.
1694     (n) The return code is 2 if any file cannot be opened, but pcregrep does
1695     continue to scan other files.
1697     (o) The -s option was incorrectly implemented. For compatibility with other
1698     greps, it now suppresses the error message for a non-existent or non-
1699     accessible file (but not the return code). There is a new option called
1700     -q that suppresses the output of matching lines, which was what -s was
1701     previously doing.
1703     (p) Added --include and --exclude options to specify files for inclusion
1704     and exclusion when recursing.
1706     11. The Makefile was not using the Autoconf-supported LDFLAGS macro properly.
1707     Hopefully, it now does.
1709     12. Missing cast in pcre_study().
1711     13. Added an "uninstall" target to the makefile.
1713     14. Replaced "extern" in the function prototypes in Makefile.in with
1714     "PCRE_DATA_SCOPE", which defaults to 'extern' or 'extern "C"' in the Unix
1715     world, but is set differently for Windows.
1717     15. Added a second compiling function called pcre_compile2(). The only
1718     difference is that it has an extra argument, which is a pointer to an
1719     integer error code. When there is a compile-time failure, this is set
1720     non-zero, in addition to the error test pointer being set to point to an
1721     error message. The new argument may be NULL if no error number is required
1722     (but then you may as well call pcre_compile(), which is now just a
1723     wrapper). This facility is provided because some applications need a
1724     numeric error indication, but it has also enabled me to tidy up the way
1725     compile-time errors are handled in the POSIX wrapper.
1727     16. Added VPATH=.libs to the makefile; this should help when building with one
1728     prefix path and installing with another. (Or so I'm told by someone who
1729     knows more about this stuff than I do.)
1731     17. Added a new option, REG_DOTALL, to the POSIX function regcomp(). This
1732     passes PCRE_DOTALL to the pcre_compile() function, making the "." character
1733     match everything, including newlines. This is not POSIX-compatible, but
1734     somebody wanted the feature. From pcretest it can be activated by using
1735     both the P and the s flags.
1737     18. AC_PROG_LIBTOOL appeared twice in Makefile.in. Removed one.
1739     19. libpcre.pc was being incorrectly installed as executable.
1741     20. A couple of places in pcretest check for end-of-line by looking for '\n';
1742     it now also looks for '\r' so that it will work unmodified on Windows.
1744     21. Added Google's contributed C++ wrapper to the distribution.
1746     22. Added some untidy missing memory free() calls in pcretest, to keep
1747     Electric Fence happy when testing.
1751 nigel 75 Version 5.0 13-Sep-04
1752     ---------------------
1754     1. Internal change: literal characters are no longer packed up into items
1755     containing multiple characters in a single byte-string. Each character
1756     is now matched using a separate opcode. However, there may be more than one
1757     byte in the character in UTF-8 mode.
1759     2. The pcre_callout_block structure has two new fields: pattern_position and
1760     next_item_length. These contain the offset in the pattern to the next match
1761     item, and its length, respectively.
1763     3. The PCRE_AUTO_CALLOUT option for pcre_compile() requests the automatic
1764     insertion of callouts before each pattern item. Added the /C option to
1765     pcretest to make use of this.
1767     4. On the advice of a Windows user, the lines
1769     #if defined(_WIN32) || defined(WIN32)
1770     _setmode( _fileno( stdout ), 0x8000 );
1771     #endif /* defined(_WIN32) || defined(WIN32) */
1773     have been added to the source of pcretest. This apparently does useful
1774     magic in relation to line terminators.
1776     5. Changed "r" and "w" in the calls to fopen() in pcretest to "rb" and "wb"
1777     for the benefit of those environments where the "b" makes a difference.
1779     6. The icc compiler has the same options as gcc, but "configure" doesn't seem
1780     to know about it. I have put a hack into configure.in that adds in code
1781     to set GCC=yes if CC=icc. This seems to end up at a point in the
1782     generated configure script that is early enough to affect the setting of
1783     compiler options, which is what is needed, but I have no means of testing
1784     whether it really works. (The user who reported this had patched the
1785     generated configure script, which of course I cannot do.)
1787     LATER: After change 22 below (new libtool files), the configure script
1788     seems to know about icc (and also ecc). Therefore, I have commented out
1789     this hack in configure.in.
1791     7. Added support for pkg-config (2 patches were sent in).
1793     8. Negated POSIX character classes that used a combination of internal tables
1794     were completely broken. These were [[:^alpha:]], [[:^alnum:]], and
1795     [[:^ascii]]. Typically, they would match almost any characters. The other
1796     POSIX classes were not broken in this way.
1798     9. Matching the pattern "\b.*?" against "ab cd", starting at offset 1, failed
1799     to find the match, as PCRE was deluded into thinking that the match had to
1800     start at the start point or following a newline. The same bug applied to
1801     patterns with negative forward assertions or any backward assertions
1802     preceding ".*" at the start, unless the pattern required a fixed first
1803     character. This was a failing pattern: "(?!.bcd).*". The bug is now fixed.
1805     10. In UTF-8 mode, when moving forwards in the subject after a failed match
1806     starting at the last subject character, bytes beyond the end of the subject
1807     string were read.
1809     11. Renamed the variable "class" as "classbits" to make life easier for C++
1810     users. (Previously there was a macro definition, but it apparently wasn't
1811     enough.)
1813     12. Added the new field "tables" to the extra data so that tables can be passed
1814     in at exec time, or the internal tables can be re-selected. This allows
1815     a compiled regex to be saved and re-used at a later time by a different
1816     program that might have everything at different addresses.
1818     13. Modified the pcre-config script so that, when run on Solaris, it shows a
1819     -R library as well as a -L library.
1821     14. The debugging options of pcretest (-d on the command line or D on a
1822     pattern) showed incorrect output for anything following an extended class
1823     that contained multibyte characters and which was followed by a quantifier.
1825     15. Added optional support for general category Unicode character properties
1826     via the \p, \P, and \X escapes. Unicode property support implies UTF-8
1827     support. It adds about 90K to the size of the library. The meanings of the
1828     inbuilt class escapes such as \d and \s have NOT been changed.
1830     16. Updated pcredemo.c to include calls to free() to release the memory for the
1831     compiled pattern.
1833     17. The generated file chartables.c was being created in the source directory
1834     instead of in the building directory. This caused the build to fail if the
1835     source directory was different from the building directory, and was
1836     read-only.
1838     18. Added some sample Win commands from Mark Tetrode into the NON-UNIX-USE
1839     file. No doubt somebody will tell me if they don't make sense... Also added
1840     Dan Mooney's comments about building on OpenVMS.
1842     19. Added support for partial matching via the PCRE_PARTIAL option for
1843     pcre_exec() and the \P data escape in pcretest.
1845     20. Extended pcretest with 3 new pattern features:
1847     (i) A pattern option of the form ">rest-of-line" causes pcretest to
1848     write the compiled pattern to the file whose name is "rest-of-line".
1849     This is a straight binary dump of the data, with the saved pointer to
1850     the character tables forced to be NULL. The study data, if any, is
1851     written too. After writing, pcretest reads a new pattern.
1853     (ii) If, instead of a pattern, "<rest-of-line" is given, pcretest reads a
1854     compiled pattern from the given file. There must not be any
1855     occurrences of "<" in the file name (pretty unlikely); if there are,
1856     pcretest will instead treat the initial "<" as a pattern delimiter.
1857     After reading in the pattern, pcretest goes on to read data lines as
1858     usual.
1860     (iii) The F pattern option causes pcretest to flip the bytes in the 32-bit
1861     and 16-bit fields in a compiled pattern, to simulate a pattern that
1862     was compiled on a host of opposite endianness.
1864     21. The pcre-exec() function can now cope with patterns that were compiled on
1865     hosts of opposite endianness, with this restriction:
1867     As for any compiled expression that is saved and used later, the tables
1868     pointer field cannot be preserved; the extra_data field in the arguments
1869     to pcre_exec() should be used to pass in a tables address if a value
1870     other than the default internal tables were used at compile time.
1872     22. Calling pcre_exec() with a negative value of the "ovecsize" parameter is
1873     now diagnosed as an error. Previously, most of the time, a negative number
1874     would have been treated as zero, but if in addition "ovector" was passed as
1875     NULL, a crash could occur.
1877     23. Updated the files ltmain.sh, config.sub, config.guess, and aclocal.m4 with
1878     new versions from the libtool 1.5 distribution (the last one is a copy of
1879     a file called libtool.m4). This seems to have fixed the need to patch
1880     "configure" to support Darwin 1.3 (which I used to do). However, I still
1881     had to patch ltmain.sh to ensure that ${SED} is set (it isn't on my
1882     workstation).
1884     24. Changed the PCRE licence to be the more standard "BSD" licence.
1887 nigel 73 Version 4.5 01-Dec-03
1888     ---------------------
1890     1. There has been some re-arrangement of the code for the match() function so
1891     that it can be compiled in a version that does not call itself recursively.
1892     Instead, it keeps those local variables that need separate instances for
1893     each "recursion" in a frame on the heap, and gets/frees frames whenever it
1894     needs to "recurse". Keeping track of where control must go is done by means
1895     of setjmp/longjmp. The whole thing is implemented by a set of macros that
1896     hide most of the details from the main code, and operates only if
1897     NO_RECURSE is defined while compiling pcre.c. If PCRE is built using the
1898     "configure" mechanism, "--disable-stack-for-recursion" turns on this way of
1899     operating.
1901     To make it easier for callers to provide specially tailored get/free
1902     functions for this usage, two new functions, pcre_stack_malloc, and
1903     pcre_stack_free, are used. They are always called in strict stacking order,
1904     and the size of block requested is always the same.
1906     The PCRE_CONFIG_STACKRECURSE info parameter can be used to find out whether
1907     PCRE has been compiled to use the stack or the heap for recursion. The
1908     -C option of pcretest uses this to show which version is compiled.
1910     A new data escape \S, is added to pcretest; it causes the amounts of store
1911     obtained and freed by both kinds of malloc/free at match time to be added
1912     to the output.
1914     2. Changed the locale test to use "fr_FR" instead of "fr" because that's
1915     what's available on my current Linux desktop machine.
1917     3. When matching a UTF-8 string, the test for a valid string at the start has
1918     been extended. If start_offset is not zero, PCRE now checks that it points
1919     to a byte that is the start of a UTF-8 character. If not, it returns
1920     PCRE_ERROR_BADUTF8_OFFSET (-11). Note: the whole string is still checked;
1921     this is necessary because there may be backward assertions in the pattern.
1922     When matching the same subject several times, it may save resources to use
1923     PCRE_NO_UTF8_CHECK on all but the first call if the string is long.
1925     4. The code for checking the validity of UTF-8 strings has been tightened so
1926     that it rejects (a) strings containing 0xfe or 0xff bytes and (b) strings
1927     containing "overlong sequences".
1929     5. Fixed a bug (appearing twice) that I could not find any way of exploiting!
1930     I had written "if ((digitab[*p++] && chtab_digit) == 0)" where the "&&"
1931     should have been "&", but it just so happened that all the cases this let
1932     through by mistake were picked up later in the function.
1934     6. I had used a variable called "isblank" - this is a C99 function, causing
1935     some compilers to warn. To avoid this, I renamed it (as "blankclass").
1937     7. Cosmetic: (a) only output another newline at the end of pcretest if it is
1938     prompting; (b) run "./pcretest /dev/null" at the start of the test script
1939     so the version is shown; (c) stop "make test" echoing "./RunTest".
1941     8. Added patches from David Burgess to enable PCRE to run on EBCDIC systems.
1943     9. The prototype for memmove() for systems that don't have it was using
1944     size_t, but the inclusion of the header that defines size_t was later. I've
1945     moved the #includes for the C headers earlier to avoid this.
1947     10. Added some adjustments to the code to make it easier to compiler on certain
1948     special systems:
1950     (a) Some "const" qualifiers were missing.
1951     (b) Added the macro EXPORT before all exported functions; by default this
1952     is defined to be empty.
1953     (c) Changed the dftables auxiliary program (that builds chartables.c) so
1954     that it reads its output file name as an argument instead of writing
1955     to the standard output and assuming this can be redirected.
1957     11. In UTF-8 mode, if a recursive reference (e.g. (?1)) followed a character
1958     class containing characters with values greater than 255, PCRE compilation
1959     went into a loop.
1961     12. A recursive reference to a subpattern that was within another subpattern
1962     that had a minimum quantifier of zero caused PCRE to crash. For example,
1963     (x(y(?2))z)? provoked this bug with a subject that got as far as the
1964     recursion. If the recursively-called subpattern itself had a zero repeat,
1965     that was OK.
1967     13. In pcretest, the buffer for reading a data line was set at 30K, but the
1968     buffer into which it was copied (for escape processing) was still set at
1969     1024, so long lines caused crashes.
1971     14. A pattern such as /[ab]{1,3}+/ failed to compile, giving the error
1972     "internal error: code overflow...". This applied to any character class
1973     that was followed by a possessive quantifier.
1975     15. Modified the Makefile to add libpcre.la as a prerequisite for
1976     libpcreposix.la because I was told this is needed for a parallel build to
1977     work.
1979     16. If a pattern that contained .* following optional items at the start was
1980     studied, the wrong optimizing data was generated, leading to matching
1981     errors. For example, studying /[ab]*.*c/ concluded, erroneously, that any
1982     matching string must start with a or b or c. The correct conclusion for
1983     this pattern is that a match can start with any character.
1986 nigel 71 Version 4.4 13-Aug-03
1987     ---------------------
1989     1. In UTF-8 mode, a character class containing characters with values between
1990     127 and 255 was not handled correctly if the compiled pattern was studied.
1991     In fixing this, I have also improved the studying algorithm for such
1992     classes (slightly).
1994     2. Three internal functions had redundant arguments passed to them. Removal
1995     might give a very teeny performance improvement.
1997     3. Documentation bug: the value of the capture_top field in a callout is *one
1998     more than* the number of the hightest numbered captured substring.
2000     4. The Makefile linked pcretest and pcregrep with -lpcre, which could result
2001     in incorrectly linking with a previously installed version. They now link
2002     explicitly with libpcre.la.
2004     5. configure.in no longer needs to recognize Cygwin specially.
2006     6. A problem in pcre.in for Windows platforms is fixed.
2008     7. If a pattern was successfully studied, and the -d (or /D) flag was given to
2009     pcretest, it used to include the size of the study block as part of its
2010     output. Unfortunately, the structure contains a field that has a different
2011     size on different hardware architectures. This meant that the tests that
2012     showed this size failed. As the block is currently always of a fixed size,
2013     this information isn't actually particularly useful in pcretest output, so
2014     I have just removed it.
2016     8. Three pre-processor statements accidentally did not start in column 1.
2017     Sadly, there are *still* compilers around that complain, even though
2018     standard C has not required this for well over a decade. Sigh.
2020     9. In pcretest, the code for checking callouts passed small integers in the
2021     callout_data field, which is a void * field. However, some picky compilers
2022     complained about the casts involved for this on 64-bit systems. Now
2023     pcretest passes the address of the small integer instead, which should get
2024     rid of the warnings.
2026     10. By default, when in UTF-8 mode, PCRE now checks for valid UTF-8 strings at
2027     both compile and run time, and gives an error if an invalid UTF-8 sequence
2028     is found. There is a option for disabling this check in cases where the
2029     string is known to be correct and/or the maximum performance is wanted.
2031     11. In response to a bug report, I changed one line in Makefile.in from
2033     -Wl,--out-implib,.libs/lib@WIN_PREFIX@pcreposix.dll.a \
2034     to
2035     -Wl,--out-implib,.libs/@WIN_PREFIX@libpcreposix.dll.a \
2037     to look similar to other lines, but I have no way of telling whether this
2038     is the right thing to do, as I do not use Windows. No doubt I'll get told
2039     if it's wrong...
2042 nigel 69 Version 4.3 21-May-03
2043     ---------------------
2045     1. Two instances of @WIN_PREFIX@ omitted from the Windows targets in the
2046     Makefile.
2048     2. Some refactoring to improve the quality of the code:
2050     (i) The utf8_table... variables are now declared "const".
2052     (ii) The code for \cx, which used the "case flipping" table to upper case
2053     lower case letters, now just substracts 32. This is ASCII-specific,
2054     but the whole concept of \cx is ASCII-specific, so it seems
2055     reasonable.
2057     (iii) PCRE was using its character types table to recognize decimal and
2058     hexadecimal digits in the pattern. This is silly, because it handles
2059     only 0-9, a-f, and A-F, but the character types table is locale-
2060     specific, which means strange things might happen. A private
2061     table is now used for this - though it costs 256 bytes, a table is
2062     much faster than multiple explicit tests. Of course, the standard
2063     character types table is still used for matching digits in subject
2064     strings against \d.
2066     (iv) Strictly, the identifier ESC_t is reserved by POSIX (all identifiers
2067     ending in _t are). So I've renamed it as ESC_tee.
2069     3. The first argument for regexec() in the POSIX wrapper should have been
2070     defined as "const".
2072     4. Changed pcretest to use malloc() for its buffers so that they can be
2073     Electric Fenced for debugging.
2075     5. There were several places in the code where, in UTF-8 mode, PCRE would try
2076     to read one or more bytes before the start of the subject string. Often this
2077     had no effect on PCRE's behaviour, but in some circumstances it could
2078     provoke a segmentation fault.
2080     6. A lookbehind at the start of a pattern in UTF-8 mode could also cause PCRE
2081     to try to read one or more bytes before the start of the subject string.
2083     7. A lookbehind in a pattern matched in non-UTF-8 mode on a PCRE compiled with
2084     UTF-8 support could misbehave in various ways if the subject string
2085     contained bytes with the 0x80 bit set and the 0x40 bit unset in a lookbehind
2086     area. (PCRE was not checking for the UTF-8 mode flag, and trying to move
2087     back over UTF-8 characters.)
2090 nigel 67 Version 4.2 14-Apr-03
2091     ---------------------
2093     1. Typo "#if SUPPORT_UTF8" instead of "#ifdef SUPPORT_UTF8" fixed.
2095     2. Changes to the building process, supplied by Ronald Landheer-Cieslak
2096     [ON_WINDOWS]: new variable, "#" on non-Windows platforms
2097     [NOT_ON_WINDOWS]: new variable, "#" on Windows platforms
2098     [WIN_PREFIX]: new variable, "cyg" for Cygwin
2099     * Makefile.in: use autoconf substitution for OBJEXT, EXEEXT, BUILD_OBJEXT
2100     and BUILD_EXEEXT
2101     Note: automatic setting of the BUILD variables is not yet working
2102     set CPPFLAGS and BUILD_CPPFLAGS (but don't use yet) - should be used at
2103     compile-time but not at link-time
2104     [LINK]: use for linking executables only
2105     make different versions for Windows and non-Windows
2106     [LINKLIB]: new variable, copy of UNIX-style LINK, used for linking
2107     libraries
2108     [LINK_FOR_BUILD]: new variable
2109     [OBJEXT]: use throughout
2110     [EXEEXT]: use throughout
2111     <winshared>: new target
2112     <wininstall>: new target
2113     <dftables.o>: use native compiler
2114     <dftables>: use native linker
2115     <install>: handle Windows platform correctly
2116     <clean>: ditto
2117     <check>: ditto
2118     copy DLL to top builddir before testing
2120     As part of these changes, -no-undefined was removed again. This was reported
2121     to give trouble on HP-UX 11.0, so getting rid of it seems like a good idea
2122     in any case.
2124     3. Some tidies to get rid of compiler warnings:
2126     . In the match_data structure, match_limit was an unsigned long int, whereas
2127     match_call_count was an int. I've made them both unsigned long ints.
2129     . In pcretest the fact that a const uschar * doesn't automatically cast to
2130     a void * provoked a warning.
2132     . Turning on some more compiler warnings threw up some "shadow" variables
2133     and a few more missing casts.
2135     4. If PCRE was complied with UTF-8 support, but called without the PCRE_UTF8
2136     option, a class that contained a single character with a value between 128
2137     and 255 (e.g. /[\xFF]/) caused PCRE to crash.
2139     5. If PCRE was compiled with UTF-8 support, but called without the PCRE_UTF8
2140     option, a class that contained several characters, but with at least one
2141     whose value was between 128 and 255 caused PCRE to crash.
2144 nigel 65 Version 4.1 12-Mar-03
2145     ---------------------
2146 nigel 63
2147 nigel 65 1. Compiling with gcc -pedantic found a couple of places where casts were
2148     needed, and a string in dftables.c that was longer than standard compilers are
2149     required to support.
2151     2. Compiling with Sun's compiler found a few more places where the code could
2152     be tidied up in order to avoid warnings.
2154     3. The variables for cross-compiling were called HOST_CC and HOST_CFLAGS; the
2155     first of these names is deprecated in the latest Autoconf in favour of the name
2156     CC_FOR_BUILD, because "host" is typically used to mean the system on which the
2157     compiled code will be run. I can't find a reference for HOST_CFLAGS, but by
2158     analogy I have changed it to CFLAGS_FOR_BUILD.
2160     4. Added -no-undefined to the linking command in the Makefile, because this is
2161     apparently helpful for Windows. To make it work, also added "-L. -lpcre" to the
2162     linking step for the pcreposix library.
2164     5. PCRE was failing to diagnose the case of two named groups with the same
2165     name.
2167     6. A problem with one of PCRE's optimizations was discovered. PCRE remembers a
2168     literal character that is needed in the subject for a match, and scans along to
2169     ensure that it is present before embarking on the full matching process. This
2170     saves time in cases of nested unlimited repeats that are never going to match.
2171     Problem: the scan can take a lot of time if the subject is very long (e.g.
2172     megabytes), thus penalizing straightforward matches. It is now done only if the
2173     amount of subject to be scanned is less than 1000 bytes.
2175     7. A lesser problem with the same optimization is that it was recording the
2176     first character of an anchored pattern as "needed", thus provoking a search
2177     right along the subject, even when the first match of the pattern was going to
2178     fail. The "needed" character is now not set for anchored patterns, unless it
2179     follows something in the pattern that is of non-fixed length. Thus, it still
2180     fulfils its original purpose of finding quick non-matches in cases of nested
2181     unlimited repeats, but isn't used for simple anchored patterns such as /^abc/.
2184     Version 4.0 17-Feb-03
2185     ---------------------
2187 nigel 63 1. If a comment in an extended regex that started immediately after a meta-item
2188     extended to the end of string, PCRE compiled incorrect data. This could lead to
2189     all kinds of weird effects. Example: /#/ was bad; /()#/ was bad; /a#/ was not.
2191     2. Moved to autoconf 2.53 and libtool 1.4.2.
2193     3. Perl 5.8 no longer needs "use utf8" for doing UTF-8 things. Consequently,
2194     the special perltest8 script is no longer needed - all the tests can be run
2195     from a single perltest script.
2197     4. From 5.004, Perl has not included the VT character (0x0b) in the set defined
2198     by \s. It has now been removed in PCRE. This means it isn't recognized as
2199     whitespace in /x regexes too, which is the same as Perl. Note that the POSIX
2200     class [:space:] *does* include VT, thereby creating a mess.
2202     5. Added the class [:blank:] (a GNU extension from Perl 5.8) to match only
2203     space and tab.
2205     6. Perl 5.005 was a long time ago. It's time to amalgamate the tests that use
2206     its new features into the main test script, reducing the number of scripts.
2208     7. Perl 5.8 has changed the meaning of patterns like /a(?i)b/. Earlier versions
2209     were backward compatible, and made the (?i) apply to the whole pattern, as if
2210     /i were given. Now it behaves more logically, and applies the option setting
2211     only to what follows. PCRE has been changed to follow suit. However, if it
2212     finds options settings right at the start of the pattern, it extracts them into
2213     the global options, as before. Thus, they show up in the info data.
2215     8. Added support for the \Q...\E escape sequence. Characters in between are
2216     treated as literals. This is slightly different from Perl in that $ and @ are
2217     also handled as literals inside the quotes. In Perl, they will cause variable
2218     interpolation. Note the following examples:
2220     Pattern PCRE matches Perl matches
2222     \Qabc$xyz\E abc$xyz abc followed by the contents of $xyz
2223     \Qabc\$xyz\E abc\$xyz abc\$xyz
2224     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2226     For compatibility with Perl, \Q...\E sequences are recognized inside character
2227     classes as well as outside them.
2229     9. Re-organized 3 code statements in pcretest to avoid "overflow in
2230     floating-point constant arithmetic" warnings from a Microsoft compiler. Added a
2231     (size_t) cast to one statement in pcretest and one in pcreposix to avoid
2232     signed/unsigned warnings.
2234     10. SunOS4 doesn't have strtoul(). This was used only for unpicking the -o
2235     option for pcretest, so I've replaced it by a simple function that does just
2236     that job.
2238     11. pcregrep was ending with code 0 instead of 2 for the commands "pcregrep" or
2239     "pcregrep -".
2241     12. Added "possessive quantifiers" ?+, *+, ++, and {,}+ which come from Sun's
2242     Java package. This provides some syntactic sugar for simple cases of what my
2243     documentation calls "once-only subpatterns". A pattern such as x*+ is the same
2244     as (?>x*). In other words, if what is inside (?>...) is just a single repeated
2245     item, you can use this simplified notation. Note that only makes sense with
2246     greedy quantifiers. Consequently, the use of the possessive quantifier forces
2247     greediness, whatever the setting of the PCRE_UNGREEDY option.
2249     13. A change of greediness default within a pattern was not taking effect at
2250     the current level for patterns like /(b+(?U)a+)/. It did apply to parenthesized
2251     subpatterns that followed. Patterns like /b+(?U)a+/ worked because the option
2252     was abstracted outside.
2254     14. PCRE now supports the \G assertion. It is true when the current matching
2255     position is at the start point of the match. This differs from \A when the
2256     starting offset is non-zero. Used with the /g option of pcretest (or similar
2257     code), it works in the same way as it does for Perl's /g option. If all
2258     alternatives of a regex begin with \G, the expression is anchored to the start
2259     match position, and the "anchored" flag is set in the compiled expression.
2261     15. Some bugs concerning the handling of certain option changes within patterns
2262     have been fixed. These applied to options other than (?ims). For example,
2263     "a(?x: b c )d" did not match "XabcdY" but did match "Xa b c dY". It should have
2264     been the other way round. Some of this was related to change 7 above.
2266     16. PCRE now gives errors for /[.x.]/ and /[=x=]/ as unsupported POSIX
2267     features, as Perl does. Previously, PCRE gave the warnings only for /[[.x.]]/
2268     and /[[=x=]]/. PCRE now also gives an error for /[:name:]/ because it supports
2269     POSIX classes only within a class (e.g. /[[:alpha:]]/).
2271     17. Added support for Perl's \C escape. This matches one byte, even in UTF8
2272     mode. Unlike ".", it always matches newline, whatever the setting of
2273     PCRE_DOTALL. However, PCRE does not permit \C to appear in lookbehind
2274     assertions. Perl allows it, but it doesn't (in general) work because it can't
2275     calculate the length of the lookbehind. At least, that's the case for Perl
2276     5.8.0 - I've been told they are going to document that it doesn't work in
2277     future.
2279     18. Added an error diagnosis for escapes that PCRE does not support: these are
2280     \L, \l, \N, \P, \p, \U, \u, and \X.
2282     19. Although correctly diagnosing a missing ']' in a character class, PCRE was
2283     reading past the end of the pattern in cases such as /[abcd/.
2285     20. PCRE was getting more memory than necessary for patterns with classes that
2286     contained both POSIX named classes and other characters, e.g. /[[:space:]abc/.
2288     21. Added some code, conditional on #ifdef VPCOMPAT, to make life easier for
2289     compiling PCRE for use with Virtual Pascal.
2291     22. Small fix to the Makefile to make it work properly if the build is done
2292     outside the source tree.
2294     23. Added a new extension: a condition to go with recursion. If a conditional
2295     subpattern starts with (?(R) the "true" branch is used if recursion has
2296     happened, whereas the "false" branch is used only at the top level.
2298     24. When there was a very long string of literal characters (over 255 bytes
2299     without UTF support, over 250 bytes with UTF support), the computation of how
2300     much memory was required could be incorrect, leading to segfaults or other
2301     strange effects.
2303     25. PCRE was incorrectly assuming anchoring (either to start of subject or to
2304     start of line for a non-DOTALL pattern) when a pattern started with (.*) and
2305     there was a subsequent back reference to those brackets. This meant that, for
2306     example, /(.*)\d+\1/ failed to match "abc123bc". Unfortunately, it isn't
2307     possible to check for precisely this case. All we can do is abandon the
2308     optimization if .* occurs inside capturing brackets when there are any back
2309     references whatsoever. (See below for a better fix that came later.)
2311     26. The handling of the optimization for finding the first character of a
2312     non-anchored pattern, and for finding a character that is required later in the
2313     match were failing in some cases. This didn't break the matching; it just
2314     failed to optimize when it could. The way this is done has been re-implemented.
2316     27. Fixed typo in error message for invalid (?R item (it said "(?p").
2318     28. Added a new feature that provides some of the functionality that Perl
2319     provides with (?{...}). The facility is termed a "callout". The way it is done
2320     in PCRE is for the caller to provide an optional function, by setting
2321     pcre_callout to its entry point. Like pcre_malloc and pcre_free, this is a
2322     global variable. By default it is unset, which disables all calling out. To get
2323     the function called, the regex must include (?C) at appropriate points. This
2324     is, in fact, equivalent to (?C0), and any number <= 255 may be given with (?C).
2325     This provides a means of identifying different callout points. When PCRE
2326     reaches such a point in the regex, if pcre_callout has been set, the external
2327     function is called. It is provided with data in a structure called
2328     pcre_callout_block, which is defined in pcre.h. If the function returns 0,
2329     matching continues; if it returns a non-zero value, the match at the current
2330     point fails. However, backtracking will occur if possible. [This was changed
2331     later and other features added - see item 49 below.]
2333     29. pcretest is upgraded to test the callout functionality. It provides a
2334     callout function that displays information. By default, it shows the start of
2335     the match and the current position in the text. There are some new data escapes
2336     to vary what happens:
2338     \C+ in addition, show current contents of captured substrings
2339     \C- do not supply a callout function
2340     \C!n return 1 when callout number n is reached
2341     \C!n!m return 1 when callout number n is reached for the mth time
2343     30. If pcregrep was called with the -l option and just a single file name, it
2344     output "<stdin>" if a match was found, instead of the file name.
2346     31. Improve the efficiency of the POSIX API to PCRE. If the number of capturing
2347     slots is less than POSIX_MALLOC_THRESHOLD, use a block on the stack to pass to
2348     pcre_exec(). This saves a malloc/free per call. The default value of
2349     POSIX_MALLOC_THRESHOLD is 10; it can be changed by --with-posix-malloc-threshold
2350     when configuring.
2352     32. The default maximum size of a compiled pattern is 64K. There have been a
2353     few cases of people hitting this limit. The code now uses macros to handle the
2354     storing of links as offsets within the compiled pattern. It defaults to 2-byte
2355     links, but this can be changed to 3 or 4 bytes by --with-link-size when
2356     configuring. Tests 2 and 5 work only with 2-byte links because they output
2357     debugging information about compiled patterns.
2359     33. Internal code re-arrangements:
2361     (a) Moved the debugging function for printing out a compiled regex into
2362     its own source file (printint.c) and used #include to pull it into
2363     pcretest.c and, when DEBUG is defined, into pcre.c, instead of having two
2364     separate copies.
2366     (b) Defined the list of op-code names for debugging as a macro in
2367     internal.h so that it is next to the definition of the opcodes.
2369     (c) Defined a table of op-code lengths for simpler skipping along compiled
2370     code. This is again a macro in internal.h so that it is next to the
2371     definition of the opcodes.
2373     34. Added support for recursive calls to individual subpatterns, along the
2374     lines of Robin Houston's patch (but implemented somewhat differently).
2376     35. Further mods to the Makefile to help Win32. Also, added code to pcregrep to
2377     allow it to read and process whole directories in Win32. This code was
2378     contributed by Lionel Fourquaux; it has not been tested by me.
2380     36. Added support for named subpatterns. The Python syntax (?P<name>...) is
2381     used to name a group. Names consist of alphanumerics and underscores, and must
2382     be unique. Back references use the syntax (?P=name) and recursive calls use
2383     (?P>name) which is a PCRE extension to the Python extension. Groups still have
2384     numbers. The function pcre_fullinfo() can be used after compilation to extract
2385     a name/number map. There are three relevant calls:
2387     PCRE_INFO_NAMEENTRYSIZE yields the size of each entry in the map
2388     PCRE_INFO_NAMECOUNT yields the number of entries
2389     PCRE_INFO_NAMETABLE yields a pointer to the map.
2391     The map is a vector of fixed-size entries. The size of each entry depends on
2392     the length of the longest name used. The first two bytes of each entry are the
2393     group number, most significant byte first. There follows the corresponding
2394     name, zero terminated. The names are in alphabetical order.
2396     37. Make the maximum literal string in the compiled code 250 for the non-UTF-8
2397     case instead of 255. Making it the same both with and without UTF-8 support
2398     means that the same test output works with both.
2400     38. There was a case of malloc(0) in the POSIX testing code in pcretest. Avoid
2401     calling malloc() with a zero argument.
2403     39. Change 25 above had to resort to a heavy-handed test for the .* anchoring
2404     optimization. I've improved things by keeping a bitmap of backreferences with
2405     numbers 1-31 so that if .* occurs inside capturing brackets that are not in
2406     fact referenced, the optimization can be applied. It is unlikely that a
2407     relevant occurrence of .* (i.e. one which might indicate anchoring or forcing
2408     the match to follow \n) will appear inside brackets with a number greater than
2409     31, but if it does, any back reference > 31 suppresses the optimization.
2411     40. Added a new compile-time option PCRE_NO_AUTO_CAPTURE. This has the effect
2412     of disabling numbered capturing parentheses. Any opening parenthesis that is
2413     not followed by ? behaves as if it were followed by ?: but named parentheses
2414     can still be used for capturing (and they will acquire numbers in the usual
2415     way).
2417     41. Redesigned the return codes from the match() function into yes/no/error so
2418     that errors can be passed back from deep inside the nested calls. A malloc
2419     failure while inside a recursive subpattern call now causes the
2420     PCRE_ERROR_NOMEMORY return instead of quietly going wrong.
2422     42. It is now possible to set a limit on the number of times the match()
2423     function is called in a call to pcre_exec(). This facility makes it possible to
2424     limit the amount of recursion and backtracking, though not in a directly
2425     obvious way, because the match() function is used in a number of different
2426     circumstances. The count starts from zero for each position in the subject
2427     string (for non-anchored patterns). The default limit is, for compatibility, a
2428     large number, namely 10 000 000. You can change this in two ways:
2430     (a) When configuring PCRE before making, you can use --with-match-limit=n
2431     to set a default value for the compiled library.
2433     (b) For each call to pcre_exec(), you can pass a pcre_extra block in which
2434     a different value is set. See 45 below.
2436     If the limit is exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2438     43. Added a new function pcre_config(int, void *) to enable run-time extraction
2439     of things that can be changed at compile time. The first argument specifies
2440     what is wanted and the second points to where the information is to be placed.
2441     The current list of available information is:
2445     The output is an integer that is set to one if UTF-8 support is available;
2446     otherwise it is set to zero.
2450     The output is an integer that it set to the value of the code that is used for
2451     newline. It is either LF (10) or CR (13).
2455     The output is an integer that contains the number of bytes used for internal
2456     linkage in compiled expressions. The value is 2, 3, or 4. See item 32 above.
2460     The output is an integer that contains the threshold above which the POSIX
2461     interface uses malloc() for output vectors. See item 31 above.
2465     The output is an unsigned integer that contains the default limit of the number
2466     of match() calls in a pcre_exec() execution. See 42 above.
2468     44. pcretest has been upgraded by the addition of the -C option. This causes it
2469     to extract all the available output from the new pcre_config() function, and to
2470     output it. The program then exits immediately.
2472     45. A need has arisen to pass over additional data with calls to pcre_exec() in
2473     order to support additional features. One way would have been to define
2474     pcre_exec2() (for example) with extra arguments, but this would not have been
2475     extensible, and would also have required all calls to the original function to
2476     be mapped to the new one. Instead, I have chosen to extend the mechanism that
2477     is used for passing in "extra" data from pcre_study().
2479     The pcre_extra structure is now exposed and defined in pcre.h. It currently
2480     contains the following fields:
2482     flags a bitmap indicating which of the following fields are set
2483     study_data opaque data from pcre_study()
2484     match_limit a way of specifying a limit on match() calls for a specific
2485     call to pcre_exec()
2486     callout_data data for callouts (see 49 below)
2488     The flag bits are also defined in pcre.h, and are
2494     The pcre_study() function now returns one of these new pcre_extra blocks, with
2495     the actual study data pointed to by the study_data field, and the
2496     PCRE_EXTRA_STUDY_DATA flag set. This can be passed directly to pcre_exec() as
2497     before. That is, this change is entirely upwards-compatible and requires no
2498     change to existing code.
2500     If you want to pass in additional data to pcre_exec(), you can either place it
2501     in a pcre_extra block provided by pcre_study(), or create your own pcre_extra
2502     block.
2504     46. pcretest has been extended to test the PCRE_EXTRA_MATCH_LIMIT feature. If a
2505     data string contains the escape sequence \M, pcretest calls pcre_exec() several
2506     times with different match limits, until it finds the minimum value needed for
2507     pcre_exec() to complete. The value is then output. This can be instructive; for
2508     most simple matches the number is quite small, but for pathological cases it
2509     gets very large very quickly.
2511     47. There's a new option for pcre_fullinfo() called PCRE_INFO_STUDYSIZE. It
2512     returns the size of the data block pointed to by the study_data field in a
2513     pcre_extra block, that is, the value that was passed as the argument to
2514     pcre_malloc() when PCRE was getting memory in which to place the information
2515     created by pcre_study(). The fourth argument should point to a size_t variable.
2516     pcretest has been extended so that this information is shown after a successful
2517     pcre_study() call when information about the compiled regex is being displayed.
2519     48. Cosmetic change to Makefile: there's no need to have / after $(DESTDIR)
2520     because what follows is always an absolute path. (Later: it turns out that this
2521     is more than cosmetic for MinGW, because it doesn't like empty path
2522     components.)
2524     49. Some changes have been made to the callout feature (see 28 above):
2526     (i) A callout function now has three choices for what it returns:
2528     0 => success, carry on matching
2529     > 0 => failure at this point, but backtrack if possible
2530     < 0 => serious error, return this value from pcre_exec()
2532     Negative values should normally be chosen from the set of PCRE_ERROR_xxx
2533     values. In particular, returning PCRE_ERROR_NOMATCH forces a standard
2534     "match failed" error. The error number PCRE_ERROR_CALLOUT is reserved for
2535     use by callout functions. It will never be used by PCRE itself.
2537     (ii) The pcre_extra structure (see 45 above) has a void * field called
2538     callout_data, with corresponding flag bit PCRE_EXTRA_CALLOUT_DATA. The
2539     pcre_callout_block structure has a field of the same name. The contents of
2540     the field passed in the pcre_extra structure are passed to the callout
2541     function in the corresponding field in the callout block. This makes it
2542     easier to use the same callout-containing regex from multiple threads. For
2543     testing, the pcretest program has a new data escape
2545     \C*n pass the number n (may be negative) as callout_data
2547     If the callout function in pcretest receives a non-zero value as
2548     callout_data, it returns that value.
2550     50. Makefile wasn't handling CFLAGS properly when compiling dftables. Also,
2551     there were some redundant $(CFLAGS) in commands that are now specified as
2552     $(LINK), which already includes $(CFLAGS).
2554     51. Extensions to UTF-8 support are listed below. These all apply when (a) PCRE
2555     has been compiled with UTF-8 support *and* pcre_compile() has been compiled
2556     with the PCRE_UTF8 flag. Patterns that are compiled without that flag assume
2557     one-byte characters throughout. Note that case-insensitive matching applies
2558     only to characters whose values are less than 256. PCRE doesn't support the
2559     notion of cases for higher-valued characters.
2561     (i) A character class whose characters are all within 0-255 is handled as
2562     a bit map, and the map is inverted for negative classes. Previously, a
2563     character > 255 always failed to match such a class; however it should
2564     match if the class was a negative one (e.g. [^ab]). This has been fixed.
2566     (ii) A negated character class with a single character < 255 is coded as
2567     "not this character" (OP_NOT). This wasn't working properly when the test
2568     character was multibyte, either singly or repeated.
2570     (iii) Repeats of multibyte characters are now handled correctly in UTF-8
2571     mode, for example: \x{100}{2,3}.
2573     (iv) The character escapes \b, \B, \d, \D, \s, \S, \w, and \W (either
2574     singly or repeated) now correctly test multibyte characters. However,
2575     PCRE doesn't recognize any characters with values greater than 255 as
2576     digits, spaces, or word characters. Such characters always match \D, \S,
2577     and \W, and never match \d, \s, or \w.
2579     (v) Classes may now contain characters and character ranges with values
2580     greater than 255. For example: [ab\x{100}-\x{400}].
2582     (vi) pcregrep now has a --utf-8 option (synonym -u) which makes it call
2583     PCRE in UTF-8 mode.
2585     52. The info request value PCRE_INFO_FIRSTCHAR has been renamed
2586     PCRE_INFO_FIRSTBYTE because it is a byte value. However, the old name is
2587     retained for backwards compatibility. (Note that LASTLITERAL is also a byte
2588     value.)
2590     53. The single man page has become too large. I have therefore split it up into
2591     a number of separate man pages. These also give rise to individual HTML pages;
2592     these are now put in a separate directory, and there is an index.html page that
2593     lists them all. Some hyperlinking between the pages has been installed.
2595     54. Added convenience functions for handling named capturing parentheses.
2597     55. Unknown escapes inside character classes (e.g. [\M]) and escapes that
2598     aren't interpreted therein (e.g. [\C]) are literals in Perl. This is now also
2599     true in PCRE, except when the PCRE_EXTENDED option is set, in which case they
2600     are faulted.
2602     56. Introduced HOST_CC and HOST_CFLAGS which can be set in the environment when
2603     calling configure. These values are used when compiling the dftables.c program
2604     which is run to generate the source of the default character tables. They
2605     default to the values of CC and CFLAGS. If you are cross-compiling PCRE,
2606     you will need to set these values.
2608     57. Updated the building process for Windows DLL, as provided by Fred Cox.
2611     Version 3.9 02-Jan-02
2612 nigel 61 ---------------------
2614     1. A bit of extraneous text had somehow crept into the pcregrep documentation.
2616     2. If --disable-static was given, the building process failed when trying to
2617     build pcretest and pcregrep. (For some reason it was using libtool to compile
2618     them, which is not right, as they aren't part of the library.)
2621 nigel 59 Version 3.8 18-Dec-01
2622     ---------------------
2624     1. The experimental UTF-8 code was completely screwed up. It was packing the
2625     bytes in the wrong order. How dumb can you get?
2628 nigel 57 Version 3.7 29-Oct-01
2629     ---------------------
2631     1. In updating pcretest to check change 1 of version 3.6, I screwed up.
2632     This caused pcretest, when used on the test data, to segfault. Unfortunately,
2633     this didn't happen under Solaris 8, where I normally test things.
2635 nigel 59 2. The Makefile had to be changed to make it work on BSD systems, where 'make'
2636     doesn't seem to recognize that ./xxx and xxx are the same file. (This entry
2637     isn't in ChangeLog distributed with 3.7 because I forgot when I hastily made
2638     this fix an hour or so after the initial 3.7 release.)
2639 nigel 57
2640 nigel 59
2641 nigel 55 Version 3.6 23-Oct-01
2642     ---------------------
2644     1. Crashed with /(sens|respons)e and \1ibility/ and "sense and sensibility" if
2645     offsets passed as NULL with zero offset count.
2647     2. The config.guess and config.sub files had not been updated when I moved to
2648     the latest autoconf.
2651 nigel 53 Version 3.5 15-Aug-01
2652     ---------------------
2653 nigel 5
2654 nigel 53 1. Added some missing #if !defined NOPOSIX conditionals in pcretest.c that
2655     had been forgotten.
2657     2. By using declared but undefined structures, we can avoid using "void"
2658     definitions in pcre.h while keeping the internal definitions of the structures
2659     private.
2661     3. The distribution is now built using autoconf 2.50 and libtool 1.4. From a
2662     user point of view, this means that both static and shared libraries are built
2663     by default, but this can be individually controlled. More of the work of
2664     handling this static/shared cases is now inside libtool instead of PCRE's make
2665     file.
2667     4. The pcretest utility is now installed along with pcregrep because it is
2668     useful for users (to test regexs) and by doing this, it automatically gets
2669     relinked by libtool. The documentation has been turned into a man page, so
2670     there are now .1, .txt, and .html versions in /doc.
2672     5. Upgrades to pcregrep:
2673     (i) Added long-form option names like gnu grep.
2674     (ii) Added --help to list all options with an explanatory phrase.
2675     (iii) Added -r, --recursive to recurse into sub-directories.
2676     (iv) Added -f, --file to read patterns from a file.
2678     6. pcre_exec() was referring to its "code" argument before testing that
2679     argument for NULL (and giving an error if it was NULL).
2681     7. Upgraded Makefile.in to allow for compiling in a different directory from
2682     the source directory.
2684     8. Tiny buglet in pcretest: when pcre_fullinfo() was called to retrieve the
2685     options bits, the pointer it was passed was to an int instead of to an unsigned
2686     long int. This mattered only on 64-bit systems.
2688     9. Fixed typo (3.4/1) in pcre.h again. Sigh. I had changed pcre.h (which is
2689     generated) instead of pcre.in, which it its source. Also made the same change
2690     in several of the .c files.
2692     10. A new release of gcc defines printf() as a macro, which broke pcretest
2693     because it had an ifdef in the middle of a string argument for printf(). Fixed
2694     by using separate calls to printf().
2696     11. Added --enable-newline-is-cr and --enable-newline-is-lf to the configure
2697     script, to force use of CR or LF instead of \n in the source. On non-Unix
2698     systems, the value can be set in config.h.
2700     12. The limit of 200 on non-capturing parentheses is a _nesting_ limit, not an
2701     absolute limit. Changed the text of the error message to make this clear, and
2702     likewise updated the man page.
2704     13. The limit of 99 on the number of capturing subpatterns has been removed.
2705     The new limit is 65535, which I hope will not be a "real" limit.
2708 nigel 51 Version 3.4 22-Aug-00
2709     ---------------------
2711     1. Fixed typo in pcre.h: unsigned const char * changed to const unsigned char *.
2713     2. Diagnose condition (?(0) as an error instead of crashing on matching.
2716 nigel 49 Version 3.3 01-Aug-00
2717     ---------------------
2719     1. If an octal character was given, but the value was greater than \377, it
2720     was not getting masked to the least significant bits, as documented. This could
2721     lead to crashes in some systems.
2723     2. Perl 5.6 (if not earlier versions) accepts classes like [a-\d] and treats
2724     the hyphen as a literal. PCRE used to give an error; it now behaves like Perl.
2726     3. Added the functions pcre_free_substring() and pcre_free_substring_list().
2727     These just pass their arguments on to (pcre_free)(), but they are provided
2728     because some uses of PCRE bind it to non-C systems that can call its functions,
2729     but cannot call free() or pcre_free() directly.
2731     4. Add "make test" as a synonym for "make check". Corrected some comments in
2732     the Makefile.
2734     5. Add $(DESTDIR)/ in front of all the paths in the "install" target in the
2735     Makefile.
2737     6. Changed the name of pgrep to pcregrep, because Solaris has introduced a
2738     command called pgrep for grepping around the active processes.
2740     7. Added the beginnings of support for UTF-8 character strings.
2742     8. Arranged for the Makefile to pass over the settings of CC, CFLAGS, and
2743     RANLIB to ./ltconfig so that they are used by libtool. I think these are all
2744     the relevant ones. (AR is not passed because ./ltconfig does its own figuring
2745     out for the ar command.)
2748 nigel 47 Version 3.2 12-May-00
2749     ---------------------
2751     This is purely a bug fixing release.
2753     1. If the pattern /((Z)+|A)*/ was matched agained ZABCDEFG it matched Z instead
2754     of ZA. This was just one example of several cases that could provoke this bug,
2755     which was introduced by change 9 of version 2.00. The code for breaking
2756     infinite loops after an iteration that matches an empty string was't working
2757     correctly.
2759     2. The pcretest program was not imitating Perl correctly for the pattern /a*/g
2760     when matched against abbab (for example). After matching an empty string, it
2761     wasn't forcing anchoring when setting PCRE_NOTEMPTY for the next attempt; this
2762     caused it to match further down the string than it should.
2764     3. The code contained an inclusion of sys/types.h. It isn't clear why this
2765     was there because it doesn't seem to be needed, and it causes trouble on some
2766     systems, as it is not a Standard C header. It has been removed.
2768     4. Made 4 silly changes to the source to avoid stupid compiler warnings that
2769     were reported on the Macintosh. The changes were from
2771     while ((c = *(++ptr)) != 0 && c != '\n');
2772     to
2773     while ((c = *(++ptr)) != 0 && c != '\n') ;
2775     Totally extraordinary, but if that's what it takes...
2777     5. PCRE is being used in one environment where neither memmove() nor bcopy() is
2778     available. Added HAVE_BCOPY and an autoconf test for it; if neither
2779     HAVE_MEMMOVE nor HAVE_BCOPY is set, use a built-in emulation function which
2780     assumes the way PCRE uses memmove() (always moving upwards).
2782     6. PCRE is being used in one environment where strchr() is not available. There
2783     was only one use in pcre.c, and writing it out to avoid strchr() probably gives
2784     faster code anyway.
2787 nigel 45 Version 3.1 09-Feb-00
2788     ---------------------
2790     The only change in this release is the fixing of some bugs in Makefile.in for
2791     the "install" target:
2793     (1) It was failing to install pcreposix.h.
2795     (2) It was overwriting the pcre.3 man page with the pcreposix.3 man page.
2798 nigel 43 Version 3.0 01-Feb-00
2799     ---------------------
2800 nigel 41
2801     1. Add support for the /+ modifier to perltest (to output $` like it does in
2802     pcretest).
2804     2. Add support for the /g modifier to perltest.
2806     3. Fix pcretest so that it behaves even more like Perl for /g when the pattern
2807     matches null strings.
2809     4. Fix perltest so that it doesn't do unwanted things when fed an empty
2810     pattern. Perl treats empty patterns specially - it reuses the most recent
2811     pattern, which is not what we want. Replace // by /(?#)/ in order to avoid this
2812     effect.
2814     5. The POSIX interface was broken in that it was just handing over the POSIX
2815     captured string vector to pcre_exec(), but (since release 2.00) PCRE has
2816     required a bigger vector, with some working space on the end. This means that
2817     the POSIX wrapper now has to get and free some memory, and copy the results.
2819 nigel 43 6. Added some simple autoconf support, placing the test data and the
2820     documentation in separate directories, re-organizing some of the
2821     information files, and making it build pcre-config (a GNU standard). Also added
2822     libtool support for building PCRE as a shared library, which is now the
2823     default.
2824 nigel 41
2825 nigel 43 7. Got rid of the leading zero in the definition of PCRE_MINOR because 08 and
2826     09 are not valid octal constants. Single digits will be used for minor values
2827     less than 10.
2829     8. Defined REG_EXTENDED and REG_NOSUB as zero in the POSIX header, so that
2830     existing programs that set these in the POSIX interface can use PCRE without
2831     modification.
2833     9. Added a new function, pcre_fullinfo() with an extensible interface. It can
2834     return all that pcre_info() returns, plus additional data. The pcre_info()
2835     function is retained for compatibility, but is considered to be obsolete.
2837     10. Added experimental recursion feature (?R) to handle one common case that
2838     Perl 5.6 will be able to do with (?p{...}).
2840     11. Added support for POSIX character classes like [:alpha:], which Perl is
2841     adopting.
2844 nigel 39 Version 2.08 31-Aug-99
2845     ----------------------
2847     1. When startoffset was not zero and the pattern began with ".*", PCRE was not
2848     trying to match at the startoffset position, but instead was moving forward to
2849     the next newline as if a previous match had failed.
2851     2. pcretest was not making use of PCRE_NOTEMPTY when repeating for /g and /G,
2852     and could get into a loop if a null string was matched other than at the start
2853     of the subject.
2855     3. Added definitions of PCRE_MAJOR and PCRE_MINOR to pcre.h so the version can
2856     be distinguished at compile time, and for completeness also added PCRE_DATE.
2858     5. Added Paul Sokolovsky's minor changes to make it easy to compile a Win32 DLL
2859     in GnuWin32 environments.
2862 nigel 37 Version 2.07 29-Jul-99
2863     ----------------------
2865     1. The documentation is now supplied in plain text form and HTML as well as in
2866     the form of man page sources.
2868     2. C++ compilers don't like assigning (void *) values to other pointer types.
2869     In particular this affects malloc(). Although there is no problem in Standard
2870     C, I've put in casts to keep C++ compilers happy.
2872     3. Typo on pcretest.c; a cast of (unsigned char *) in the POSIX regexec() call
2873     should be (const char *).
2875     4. If NOPOSIX is defined, pcretest.c compiles without POSIX support. This may
2876     be useful for non-Unix systems who don't want to bother with the POSIX stuff.
2877     However, I haven't made this a standard facility. The documentation doesn't
2878     mention it, and the Makefile doesn't support it.
2880     5. The Makefile now contains an "install" target, with editable destinations at
2881     the top of the file. The pcretest program is not installed.
2883     6. pgrep -V now gives the PCRE version number and date.
2885     7. Fixed bug: a zero repetition after a literal string (e.g. /abcde{0}/) was
2886     causing the entire string to be ignored, instead of just the last character.
2888     8. If a pattern like /"([^\\"]+|\\.)*"/ is applied in the normal way to a
2889     non-matching string, it can take a very, very long time, even for strings of
2890     quite modest length, because of the nested recursion. PCRE now does better in
2891     some of these cases. It does this by remembering the last required literal
2892     character in the pattern, and pre-searching the subject to ensure it is present
2893     before running the real match. In other words, it applies a heuristic to detect
2894     some types of certain failure quickly, and in the above example, if presented
2895     with a string that has no trailing " it gives "no match" very quickly.
2897     9. A new runtime option PCRE_NOTEMPTY causes null string matches to be ignored;
2898     other alternatives are tried instead.
2901 nigel 35 Version 2.06 09-Jun-99
2902     ----------------------
2904     1. Change pcretest's output for amount of store used to show just the code
2905     space, because the remainder (the data block) varies in size between 32-bit and
2906     64-bit systems.
2908     2. Added an extra argument to pcre_exec() to supply an offset in the subject to
2909     start matching at. This allows lookbehinds to work when searching for multiple
2910     occurrences in a string.
2912     3. Added additional options to pcretest for testing multiple occurrences:
2914     /+ outputs the rest of the string that follows a match
2915     /g loops for multiple occurrences, using the new startoffset argument
2916     /G loops for multiple occurrences by passing an incremented pointer
2918     4. PCRE wasn't doing the "first character" optimization for patterns starting
2919     with \b or \B, though it was doing it for other lookbehind assertions. That is,
2920     it wasn't noticing that a match for a pattern such as /\bxyz/ has to start with
2921     the letter 'x'. On long subject strings, this gives a significant speed-up.
2924 nigel 33 Version 2.05 21-Apr-99
2925     ----------------------
2927     1. Changed the type of magic_number from int to long int so that it works
2928     properly on 16-bit systems.
2930     2. Fixed a bug which caused patterns starting with .* not to work correctly
2931     when the subject string contained newline characters. PCRE was assuming
2932     anchoring for such patterns in all cases, which is not correct because .* will
2933     not pass a newline unless PCRE_DOTALL is set. It now assumes anchoring only if
2934     DOTALL is set at top level; otherwise it knows that patterns starting with .*
2935     must be retried after every newline in the subject.
2938 nigel 31 Version 2.04 18-Feb-99
2939     ----------------------
2941     1. For parenthesized subpatterns with repeats whose minimum was zero, the
2942     computation of the store needed to hold the pattern was incorrect (too large).
2943     If such patterns were nested a few deep, this could multiply and become a real
2944     problem.
2946     2. Added /M option to pcretest to show the memory requirement of a specific
2947     pattern. Made -m a synonym of -s (which does this globally) for compatibility.
2949     3. Subpatterns of the form (regex){n,m} (i.e. limited maximum) were being
2950     compiled in such a way that the backtracking after subsequent failure was
2951     pessimal. Something like (a){0,3} was compiled as (a)?(a)?(a)? instead of
2952     ((a)((a)(a)?)?)? with disastrous performance if the maximum was of any size.
2955 nigel 29 Version 2.03 02-Feb-99
2956     ----------------------
2958     1. Fixed typo and small mistake in man page.
2960 nigel 31 2. Added 4th condition (GPL supersedes if conflict) and created separate
2961     LICENCE file containing the conditions.
2962 nigel 29
2963     3. Updated pcretest so that patterns such as /abc\/def/ work like they do in
2964     Perl, that is the internal \ allows the delimiter to be included in the
2965     pattern. Locked out the use of \ as a delimiter. If \ immediately follows
2966     the final delimiter, add \ to the end of the pattern (to test the error).
2968     4. Added the convenience functions for extracting substrings after a successful
2969     match. Updated pcretest to make it able to test these functions.
2972 nigel 27 Version 2.02 14-Jan-99
2973     ----------------------
2975     1. Initialized the working variables associated with each extraction so that
2976     their saving and restoring doesn't refer to uninitialized store.
2978     2. Put dummy code into study.c in order to trick the optimizer of the IBM C
2979     compiler for OS/2 into generating correct code. Apparently IBM isn't going to
2980     fix the problem.
2982     3. Pcretest: the timing code wasn't using LOOPREPEAT for timing execution
2983     calls, and wasn't printing the correct value for compiling calls. Increased the
2984     default value of LOOPREPEAT, and the number of significant figures in the
2985     times.
2987     4. Changed "/bin/rm" in the Makefile to "-rm" so it works on Windows NT.
2989     5. Renamed "deftables" as "dftables" to get it down to 8 characters, to avoid
2990     a building problem on Windows NT with a FAT file system.
2993 nigel 25 Version 2.01 21-Oct-98
2994     ----------------------
2996     1. Changed the API for pcre_compile() to allow for the provision of a pointer
2997     to character tables built by pcre_maketables() in the current locale. If NULL
2998     is passed, the default tables are used.
3001 nigel 23 Version 2.00 24-Sep-98
3002 nigel 21 ----------------------
3004 nigel 23 1. Since the (>?) facility is in Perl 5.005, don't require PCRE_EXTRA to enable
3005     it any more.
3007     2. Allow quantification of (?>) groups, and make it work correctly.
3009     3. The first character computation wasn't working for (?>) groups.
3011     4. Correct the implementation of \Z (it is permitted to match on the \n at the
3012     end of the subject) and add 5.005's \z, which really does match only at the
3013     very end of the subject.
3015     5. Remove the \X "cut" facility; Perl doesn't have it, and (?> is neater.
3017     6. Remove the ability to specify CASELESS, MULTILINE, DOTALL, and
3018     DOLLAR_END_ONLY at runtime, to make it possible to implement the Perl 5.005
3019     localized options. All options to pcre_study() were also removed.
3021     7. Add other new features from 5.005:
3023     $(?<= positive lookbehind
3024     $(?<! negative lookbehind
3025     (?imsx-imsx) added the unsetting capability
3026     such a setting is global if at outer level; local otherwise
3027     (?imsx-imsx:) non-capturing groups with option setting
3028     (?(cond)re|re) conditional pattern matching
3030     A backreference to itself in a repeated group matches the previous
3031     captured string.
3033     8. General tidying up of studying (both automatic and via "study")
3034     consequential on the addition of new assertions.
3036     9. As in 5.005, unlimited repeated groups that could match an empty substring
3037     are no longer faulted at compile time. Instead, the loop is forcibly broken at
3038     runtime if any iteration does actually match an empty substring.
3040     10. Include the RunTest script in the distribution.
3042     11. Added tests from the Perl 5.005_02 distribution. This showed up a few
3043     discrepancies, some of which were old and were also with respect to 5.004. They
3044     have now been fixed.
3047     Version 1.09 28-Apr-98
3048     ----------------------
3050 nigel 21 1. A negated single character class followed by a quantifier with a minimum
3051     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
3052     program crashes, or just wrong answers. This did not apply to negated classes
3053     containing more than one character, or to minima other than one.
3056 nigel 19 Version 1.08 27-Mar-98
3057     ----------------------
3059     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
3061     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
3062     latter must appear before anything that relies on it in the pattern.
3065 nigel 17 Version 1.07 16-Feb-98
3066     ----------------------
3068     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
3069     repeat of a potentially empty string).
3072 nigel 15 Version 1.06 23-Jan-98
3073     ----------------------
3075     1. Added Markus Oberhumer's little patches for C++.
3077     2. Literal strings longer than 255 characters were broken.
3080 nigel 13 Version 1.05 23-Dec-97
3081     ----------------------
3083     1. Negated character classes containing more than one character were failing if
3084     PCRE_CASELESS was set at run time.
3087 nigel 11 Version 1.04 19-Dec-97
3088     ----------------------
3090     1. Corrected the man page, where some "const" qualifiers had been omitted.
3092     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
3093     input syntax.
3095     3. Fixed memory leak which occurred when a regex with back references was
3096     matched with an offsets vector that wasn't big enough. The temporary memory
3097     that is used in this case wasn't being freed if the match failed.
3099     4. Tidied pcretest to ensure it frees memory that it gets.
3101     5. Temporary memory was being obtained in the case where the passed offsets
3102     vector was exactly big enough.
3104     6. Corrected definition of offsetof() from change 5 below.
3106     7. I had screwed up change 6 below and broken the rules for the use of
3107     setjmp(). Now fixed.
3110 nigel 9 Version 1.03 18-Dec-97
3111     ----------------------
3113     1. A erroneous regex with a missing opening parenthesis was correctly
3114     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
3115     on some systems.
3117     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
3118     it was reported that one broken compiler failed on the former because "code" is
3119     also an independent variable.
3121     3. The erroneous regex a[]b caused an array overrun reference.
3123     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
3124     fail on data ending with that character. (It was going on too far, and checking
3125     the next character, typically a binary zero.) This was specific to the
3126     optimized code for single-character negative classes.
3128     5. Added a contributed patch from the TIN world which does the following:
3130     + Add an undef for memmove, in case the the system defines a macro for it.
3132     + Add a definition of offsetof(), in case there isn't one. (I don't know
3133     the reason behind this - offsetof() is part of the ANSI standard - but
3134     it does no harm).
3136     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
3137     most of the places where whitespace preceded '#'. I have given up and
3138     allowed the remaining 2 cases to be at the margin.
3140     + Rename some variables in pcre to eliminate shadowing. This seems very
3141     pedantic, but does no harm, of course.
3143     6. Moved the call to setjmp() into its own function, to get rid of warnings
3144     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
3146     7. Constructs such as \d{8,} were compiling into the equivalent of
3147 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
3148 nigel 9 outcome, but in this particular case used more store than had been allocated,
3149     which caused the bug to be discovered because it threw up an internal error.
3151     8. The debugging code in both pcre and pcretest for outputting the compiled
3152     form of a regex was going wrong in the case of back references followed by
3153     curly-bracketed repeats.
3156 nigel 7 Version 1.02 12-Dec-97
3157     ----------------------
3159     1. Typos in pcre.3 and comments in the source fixed.
3161     2. Applied a contributed patch to get rid of places where it used to remove
3162     'const' from variables, and fixed some signed/unsigned and uninitialized
3163     variable warnings.
3165     3. Added the "runtest" target to Makefile.
3167     4. Set default compiler flag to -O2 rather than just -O.
3170 nigel 5 Version 1.01 19-Nov-97
3171     ----------------------
3173     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
3174     like /([ab]*)*/, that is, for classes with more than one character in them.
3176     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
3177     as /((?>a*))*/ (a PCRE_EXTRA facility).
3180     Version 1.00 18-Nov-97
3181     ----------------------
3183     1. Added compile-time macros to support systems such as SunOS4 which don't have
3184     memmove() or strerror() but have other things that can be used instead.
3186     2. Arranged that "make clean" removes the executables.
3189 nigel 3 Version 0.99 27-Oct-97
3190     ----------------------
3192     1. Fixed bug in code for optimizing classes with only one character. It was
3193     initializing a 32-byte map regardless, which could cause it to run off the end
3194     of the memory it had got.
3196     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
3199     Version 0.98 22-Oct-97
3200     ----------------------
3202     1. Fixed bug in code for handling temporary memory usage when there are more
3203     back references than supplied space in the ovector. This could cause segfaults.
3206     Version 0.97 21-Oct-97
3207     ----------------------
3209     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
3211     2. Optimized negated single characters not to use a bit map.
3213     3. Brought error texts together as macro definitions; clarified some of them;
3214     fixed one that was wrong - it said "range out of order" when it meant "invalid
3215     escape sequence".
3217     4. Changed some char * arguments to const char *.
3219     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
3221     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
3222     pcretest.
3225     Version 0.96 16-Oct-97
3226     ----------------------
3228     1. Added a simple "pgrep" utility to the distribution.
3230     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
3231     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
3232     where "ddd" means "one or more decimal digits".
3234     3. Fixed serious bug. If a pattern had a back reference, but the call to
3235     pcre_exec() didn't supply a large enough ovector to record the related
3236     identifying subpattern, the match always failed. PCRE now remembers the number
3237     of the largest back reference, and gets some temporary memory in which to save
3238     the offsets during matching if necessary, in order to ensure that
3239     backreferences always work.
3241     4. Increased the compatibility with Perl in a number of ways:
3243     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
3244     to request this handling. The option can be set at compile or exec time.
3246     (b) $ matches before a terminating newline by default; an option
3247     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
3248     mode). The option can be set at compile or exec time.
3250     (c) The handling of \ followed by a digit other than 0 is now supposed to be
3251     the same as Perl's. If the decimal number it represents is less than 10
3252     or there aren't that many previous left capturing parentheses, an octal
3253     escape is read. Inside a character class, it's always an octal escape,
3254     even if it is a single digit.
3256     (d) An escaped but undefined alphabetic character is taken as a literal,
3257     unless PCRE_EXTRA is set. Currently this just reserves the remaining
3258     escapes.
3260     (e) {0} is now permitted. (The previous item is removed from the compiled
3261     pattern).
3263     5. Changed all the names of code files so that the basic parts are no longer
3264     than 10 characters, and abolished the teeny "globals.c" file.
3266     6. Changed the handling of character classes; they are now done with a 32-byte
3267     bit map always.
3269     7. Added the -d and /D options to pcretest to make it possible to look at the
3270     internals of compilation without having to recompile pcre.
3273     Version 0.95 23-Sep-97
3274     ----------------------
3276     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
3277     \x20 at the start of a run of normal characters. These were being treated as
3278     real characters, instead of the source characters being re-checked.
3281     Version 0.94 18-Sep-97
3282     ----------------------
3284     1. The functions are now thread-safe, with the caveat that the global variables
3285     containing pointers to malloc() and free() or alternative functions are the
3286     same for all threads.
3288     2. Get pcre_study() to generate a bitmap of initial characters for non-
3289     anchored patterns when this is possible, and use it if passed to pcre_exec().
3292     Version 0.93 15-Sep-97
3293     ----------------------
3295     1. /(b)|(:+)/ was computing an incorrect first character.
3297     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
3298     but not actually doing anything yet.
3300     3. Treat "-" characters in classes that cannot be part of ranges as literals,
3301     as Perl does (e.g. [-az] or [az-]).
3303     4. Set the anchored flag if a branch starts with .* or .*? because that tests
3304     all possible positions.
3306     5. Split up into different modules to avoid including unneeded functions in a
3307     compiled binary. However, compile and exec are still in one module. The "study"
3308     function is split off.
3310     6. The character tables are now in a separate module whose source is generated
3311     by an auxiliary program - but can then be edited by hand if required. There are
3312     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
3313     toupper() in the code.
3315     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
3316     make them global. Abolish the function for setting them, as the caller can now
3317     set them directly.
3320     Version 0.92 11-Sep-97
3321     ----------------------
3323     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
3324     (e.g. /a{1,3}/) was broken (I mis-optimized it).
3326     2. Caseless matching was not working in character classes if the characters in
3327     the pattern were in upper case.
3329     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
3331     4. Make PCRE_ANCHORED public and accept as a compile option.
3333     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
3334     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
3335     pass them.
3337     6. Give an error if bad option bits passed at compile or run time.
3339     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
3340     pcretest to cause it to pass that flag.
3342     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
3343     options, and the first character, if set.
3345     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
3348     Version 0.91 10-Sep-97
3349     ----------------------
3351     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
3352     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
3354     2. PCRE was looping on encountering an indefinitely repeated back reference to
3355     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
3356     Perl does - treats the match as successful.
3358     ****


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

ViewVC Help
Powered by ViewVC 1.1.12