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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 21 - (hide annotations) (download)
Sat Feb 24 21:38:37 2007 UTC (7 years, 8 months ago) by nigel
File size: 11125 byte(s)
Load pcre-1.09 into code/trunk.

1 nigel 3 ChangeLog for PCRE
2     ------------------
3    
4 nigel 5
5 nigel 21 Version 1.09 28 Apr-98
6     ----------------------
7    
8     1. A negated single character class followed by a quantifier with a minimum
9     value of one (e.g. [^x]{1,6} ) was not compiled correctly. This could lead to
10     program crashes, or just wrong answers. This did not apply to negated classes
11     containing more than one character, or to minima other than one.
12    
13    
14 nigel 19 Version 1.08 27-Mar-98
15     ----------------------
16    
17     1. Add PCRE_UNGREEDY to invert the greediness of quantifiers.
18    
19     2. Add (?U) and (?X) to set PCRE_UNGREEDY and PCRE_EXTRA respectively. The
20     latter must appear before anything that relies on it in the pattern.
21    
22    
23 nigel 17 Version 1.07 16-Feb-98
24     ----------------------
25    
26     1. A pattern such as /((a)*)*/ was not being diagnosed as in error (unlimited
27     repeat of a potentially empty string).
28    
29    
30 nigel 15 Version 1.06 23-Jan-98
31     ----------------------
32    
33     1. Added Markus Oberhumer's little patches for C++.
34    
35     2. Literal strings longer than 255 characters were broken.
36    
37    
38 nigel 13 Version 1.05 23-Dec-97
39     ----------------------
40    
41     1. Negated character classes containing more than one character were failing if
42     PCRE_CASELESS was set at run time.
43    
44    
45 nigel 11 Version 1.04 19-Dec-97
46     ----------------------
47    
48     1. Corrected the man page, where some "const" qualifiers had been omitted.
49    
50     2. Made debugging output print "{0,xxx}" instead of just "{,xxx}" to agree with
51     input syntax.
52    
53     3. Fixed memory leak which occurred when a regex with back references was
54     matched with an offsets vector that wasn't big enough. The temporary memory
55     that is used in this case wasn't being freed if the match failed.
56    
57     4. Tidied pcretest to ensure it frees memory that it gets.
58    
59     5. Temporary memory was being obtained in the case where the passed offsets
60     vector was exactly big enough.
61    
62     6. Corrected definition of offsetof() from change 5 below.
63    
64     7. I had screwed up change 6 below and broken the rules for the use of
65     setjmp(). Now fixed.
66    
67    
68 nigel 9 Version 1.03 18-Dec-97
69     ----------------------
70    
71     1. A erroneous regex with a missing opening parenthesis was correctly
72     diagnosed, but PCRE attempted to access brastack[-1], which could cause crashes
73     on some systems.
74    
75     2. Replaced offsetof(real_pcre, code) by offsetof(real_pcre, code[0]) because
76     it was reported that one broken compiler failed on the former because "code" is
77     also an independent variable.
78    
79     3. The erroneous regex a[]b caused an array overrun reference.
80    
81     4. A regex ending with a one-character negative class (e.g. /[^k]$/) did not
82     fail on data ending with that character. (It was going on too far, and checking
83     the next character, typically a binary zero.) This was specific to the
84     optimized code for single-character negative classes.
85    
86     5. Added a contributed patch from the TIN world which does the following:
87    
88     + Add an undef for memmove, in case the the system defines a macro for it.
89    
90     + Add a definition of offsetof(), in case there isn't one. (I don't know
91     the reason behind this - offsetof() is part of the ANSI standard - but
92     it does no harm).
93    
94     + Reduce the ifdef's in pcre.c using macro DPRINTF, thereby eliminating
95     most of the places where whitespace preceded '#'. I have given up and
96     allowed the remaining 2 cases to be at the margin.
97    
98     + Rename some variables in pcre to eliminate shadowing. This seems very
99     pedantic, but does no harm, of course.
100    
101     6. Moved the call to setjmp() into its own function, to get rid of warnings
102     from gcc -Wall, and avoided calling it at all unless PCRE_EXTRA is used.
103    
104     7. Constructs such as \d{8,} were compiling into the equivalent of
105 nigel 11 \d{8}\d{0,65527} instead of \d{8}\d* which didn't make much difference to the
106 nigel 9 outcome, but in this particular case used more store than had been allocated,
107     which caused the bug to be discovered because it threw up an internal error.
108    
109     8. The debugging code in both pcre and pcretest for outputting the compiled
110     form of a regex was going wrong in the case of back references followed by
111     curly-bracketed repeats.
112    
113    
114 nigel 7 Version 1.02 12-Dec-97
115     ----------------------
116    
117     1. Typos in pcre.3 and comments in the source fixed.
118    
119     2. Applied a contributed patch to get rid of places where it used to remove
120     'const' from variables, and fixed some signed/unsigned and uninitialized
121     variable warnings.
122    
123     3. Added the "runtest" target to Makefile.
124    
125     4. Set default compiler flag to -O2 rather than just -O.
126    
127    
128 nigel 5 Version 1.01 19-Nov-97
129     ----------------------
130    
131     1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
132     like /([ab]*)*/, that is, for classes with more than one character in them.
133    
134     2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
135     as /((?>a*))*/ (a PCRE_EXTRA facility).
136    
137    
138     Version 1.00 18-Nov-97
139     ----------------------
140    
141     1. Added compile-time macros to support systems such as SunOS4 which don't have
142     memmove() or strerror() but have other things that can be used instead.
143    
144     2. Arranged that "make clean" removes the executables.
145    
146    
147 nigel 3 Version 0.99 27-Oct-97
148     ----------------------
149    
150     1. Fixed bug in code for optimizing classes with only one character. It was
151     initializing a 32-byte map regardless, which could cause it to run off the end
152     of the memory it had got.
153    
154     2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
155    
156    
157     Version 0.98 22-Oct-97
158     ----------------------
159    
160     1. Fixed bug in code for handling temporary memory usage when there are more
161     back references than supplied space in the ovector. This could cause segfaults.
162    
163    
164     Version 0.97 21-Oct-97
165     ----------------------
166    
167     1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
168    
169     2. Optimized negated single characters not to use a bit map.
170    
171     3. Brought error texts together as macro definitions; clarified some of them;
172     fixed one that was wrong - it said "range out of order" when it meant "invalid
173     escape sequence".
174    
175     4. Changed some char * arguments to const char *.
176    
177     5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
178    
179     6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
180     pcretest.
181    
182    
183     Version 0.96 16-Oct-97
184     ----------------------
185    
186     1. Added a simple "pgrep" utility to the distribution.
187    
188     2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
189     unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
190     where "ddd" means "one or more decimal digits".
191    
192     3. Fixed serious bug. If a pattern had a back reference, but the call to
193     pcre_exec() didn't supply a large enough ovector to record the related
194     identifying subpattern, the match always failed. PCRE now remembers the number
195     of the largest back reference, and gets some temporary memory in which to save
196     the offsets during matching if necessary, in order to ensure that
197     backreferences always work.
198    
199     4. Increased the compatibility with Perl in a number of ways:
200    
201     (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
202     to request this handling. The option can be set at compile or exec time.
203    
204     (b) $ matches before a terminating newline by default; an option
205     PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
206     mode). The option can be set at compile or exec time.
207    
208     (c) The handling of \ followed by a digit other than 0 is now supposed to be
209     the same as Perl's. If the decimal number it represents is less than 10
210     or there aren't that many previous left capturing parentheses, an octal
211     escape is read. Inside a character class, it's always an octal escape,
212     even if it is a single digit.
213    
214     (d) An escaped but undefined alphabetic character is taken as a literal,
215     unless PCRE_EXTRA is set. Currently this just reserves the remaining
216     escapes.
217    
218     (e) {0} is now permitted. (The previous item is removed from the compiled
219     pattern).
220    
221     5. Changed all the names of code files so that the basic parts are no longer
222     than 10 characters, and abolished the teeny "globals.c" file.
223    
224     6. Changed the handling of character classes; they are now done with a 32-byte
225     bit map always.
226    
227     7. Added the -d and /D options to pcretest to make it possible to look at the
228     internals of compilation without having to recompile pcre.
229    
230    
231     Version 0.95 23-Sep-97
232     ----------------------
233    
234     1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
235     \x20 at the start of a run of normal characters. These were being treated as
236     real characters, instead of the source characters being re-checked.
237    
238    
239     Version 0.94 18-Sep-97
240     ----------------------
241    
242     1. The functions are now thread-safe, with the caveat that the global variables
243     containing pointers to malloc() and free() or alternative functions are the
244     same for all threads.
245    
246     2. Get pcre_study() to generate a bitmap of initial characters for non-
247     anchored patterns when this is possible, and use it if passed to pcre_exec().
248    
249    
250     Version 0.93 15-Sep-97
251     ----------------------
252    
253     1. /(b)|(:+)/ was computing an incorrect first character.
254    
255     2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
256     but not actually doing anything yet.
257    
258     3. Treat "-" characters in classes that cannot be part of ranges as literals,
259     as Perl does (e.g. [-az] or [az-]).
260    
261     4. Set the anchored flag if a branch starts with .* or .*? because that tests
262     all possible positions.
263    
264     5. Split up into different modules to avoid including unneeded functions in a
265     compiled binary. However, compile and exec are still in one module. The "study"
266     function is split off.
267    
268     6. The character tables are now in a separate module whose source is generated
269     by an auxiliary program - but can then be edited by hand if required. There are
270     now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
271     toupper() in the code.
272    
273     7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
274     make them global. Abolish the function for setting them, as the caller can now
275     set them directly.
276    
277    
278     Version 0.92 11-Sep-97
279     ----------------------
280    
281     1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
282     (e.g. /a{1,3}/) was broken (I mis-optimized it).
283    
284     2. Caseless matching was not working in character classes if the characters in
285     the pattern were in upper case.
286    
287     3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
288    
289     4. Make PCRE_ANCHORED public and accept as a compile option.
290    
291     5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
292     PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
293     pass them.
294    
295     6. Give an error if bad option bits passed at compile or run time.
296    
297     7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
298     pcretest to cause it to pass that flag.
299    
300     8. Add pcre_info(), to get the number of identifying subpatterns, the stored
301     options, and the first character, if set.
302    
303     9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
304    
305    
306     Version 0.91 10-Sep-97
307     ----------------------
308    
309     1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
310     match the empty string as in /(a*)*/. It was looping and ultimately crashing.
311    
312     2. PCRE was looping on encountering an indefinitely repeated back reference to
313     a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
314     Perl does - treats the match as successful.
315    
316     ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12