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

Contents of /code/trunk/ChangeLog

Parent Directory Parent Directory | Revision Log Revision Log


Revision 7 - (show annotations) (download)
Sat Feb 24 21:38:09 2007 UTC (7 years, 1 month ago) by nigel
File size: 7245 byte(s)
Load pcre-1.02 into code/trunk.

1 ChangeLog for PCRE
2 ------------------
3
4
5 Version 1.02 12-Dec-97
6 ----------------------
7
8 1. Typos in pcre.3 and comments in the source fixed.
9
10 2. Applied a contributed patch to get rid of places where it used to remove
11 'const' from variables, and fixed some signed/unsigned and uninitialized
12 variable warnings.
13
14 3. Added the "runtest" target to Makefile.
15
16 4. Set default compiler flag to -O2 rather than just -O.
17
18
19 Version 1.01 19-Nov-97
20 ----------------------
21
22 1. PCRE was failing to diagnose unlimited repeat of empty string for patterns
23 like /([ab]*)*/, that is, for classes with more than one character in them.
24
25 2. Likewise, it wasn't diagnosing patterns with "once-only" subpatterns, such
26 as /((?>a*))*/ (a PCRE_EXTRA facility).
27
28
29 Version 1.00 18-Nov-97
30 ----------------------
31
32 1. Added compile-time macros to support systems such as SunOS4 which don't have
33 memmove() or strerror() but have other things that can be used instead.
34
35 2. Arranged that "make clean" removes the executables.
36
37
38 Version 0.99 27-Oct-97
39 ----------------------
40
41 1. Fixed bug in code for optimizing classes with only one character. It was
42 initializing a 32-byte map regardless, which could cause it to run off the end
43 of the memory it had got.
44
45 2. Added, conditional on PCRE_EXTRA, the proposed (?>REGEX) construction.
46
47
48 Version 0.98 22-Oct-97
49 ----------------------
50
51 1. Fixed bug in code for handling temporary memory usage when there are more
52 back references than supplied space in the ovector. This could cause segfaults.
53
54
55 Version 0.97 21-Oct-97
56 ----------------------
57
58 1. Added the \X "cut" facility, conditional on PCRE_EXTRA.
59
60 2. Optimized negated single characters not to use a bit map.
61
62 3. Brought error texts together as macro definitions; clarified some of them;
63 fixed one that was wrong - it said "range out of order" when it meant "invalid
64 escape sequence".
65
66 4. Changed some char * arguments to const char *.
67
68 5. Added PCRE_NOTBOL and PCRE_NOTEOL (from POSIX).
69
70 6. Added the POSIX-style API wrapper in pcreposix.a and testing facilities in
71 pcretest.
72
73
74 Version 0.96 16-Oct-97
75 ----------------------
76
77 1. Added a simple "pgrep" utility to the distribution.
78
79 2. Fixed an incompatibility with Perl: "{" is now treated as a normal character
80 unless it appears in one of the precise forms "{ddd}", "{ddd,}", or "{ddd,ddd}"
81 where "ddd" means "one or more decimal digits".
82
83 3. Fixed serious bug. If a pattern had a back reference, but the call to
84 pcre_exec() didn't supply a large enough ovector to record the related
85 identifying subpattern, the match always failed. PCRE now remembers the number
86 of the largest back reference, and gets some temporary memory in which to save
87 the offsets during matching if necessary, in order to ensure that
88 backreferences always work.
89
90 4. Increased the compatibility with Perl in a number of ways:
91
92 (a) . no longer matches \n by default; an option PCRE_DOTALL is provided
93 to request this handling. The option can be set at compile or exec time.
94
95 (b) $ matches before a terminating newline by default; an option
96 PCRE_DOLLAR_ENDONLY is provided to override this (but not in multiline
97 mode). The option can be set at compile or exec time.
98
99 (c) The handling of \ followed by a digit other than 0 is now supposed to be
100 the same as Perl's. If the decimal number it represents is less than 10
101 or there aren't that many previous left capturing parentheses, an octal
102 escape is read. Inside a character class, it's always an octal escape,
103 even if it is a single digit.
104
105 (d) An escaped but undefined alphabetic character is taken as a literal,
106 unless PCRE_EXTRA is set. Currently this just reserves the remaining
107 escapes.
108
109 (e) {0} is now permitted. (The previous item is removed from the compiled
110 pattern).
111
112 5. Changed all the names of code files so that the basic parts are no longer
113 than 10 characters, and abolished the teeny "globals.c" file.
114
115 6. Changed the handling of character classes; they are now done with a 32-byte
116 bit map always.
117
118 7. Added the -d and /D options to pcretest to make it possible to look at the
119 internals of compilation without having to recompile pcre.
120
121
122 Version 0.95 23-Sep-97
123 ----------------------
124
125 1. Fixed bug in pre-pass concerning escaped "normal" characters such as \x5c or
126 \x20 at the start of a run of normal characters. These were being treated as
127 real characters, instead of the source characters being re-checked.
128
129
130 Version 0.94 18-Sep-97
131 ----------------------
132
133 1. The functions are now thread-safe, with the caveat that the global variables
134 containing pointers to malloc() and free() or alternative functions are the
135 same for all threads.
136
137 2. Get pcre_study() to generate a bitmap of initial characters for non-
138 anchored patterns when this is possible, and use it if passed to pcre_exec().
139
140
141 Version 0.93 15-Sep-97
142 ----------------------
143
144 1. /(b)|(:+)/ was computing an incorrect first character.
145
146 2. Add pcre_study() to the API and the passing of pcre_extra to pcre_exec(),
147 but not actually doing anything yet.
148
149 3. Treat "-" characters in classes that cannot be part of ranges as literals,
150 as Perl does (e.g. [-az] or [az-]).
151
152 4. Set the anchored flag if a branch starts with .* or .*? because that tests
153 all possible positions.
154
155 5. Split up into different modules to avoid including unneeded functions in a
156 compiled binary. However, compile and exec are still in one module. The "study"
157 function is split off.
158
159 6. The character tables are now in a separate module whose source is generated
160 by an auxiliary program - but can then be edited by hand if required. There are
161 now no calls to isalnum(), isspace(), isdigit(), isxdigit(), tolower() or
162 toupper() in the code.
163
164 7. Turn the malloc/free funtions variables into pcre_malloc and pcre_free and
165 make them global. Abolish the function for setting them, as the caller can now
166 set them directly.
167
168
169 Version 0.92 11-Sep-97
170 ----------------------
171
172 1. A repeat with a fixed maximum and a minimum of 1 for an ordinary character
173 (e.g. /a{1,3}/) was broken (I mis-optimized it).
174
175 2. Caseless matching was not working in character classes if the characters in
176 the pattern were in upper case.
177
178 3. Make ranges like [W-c] work in the same way as Perl for caseless matching.
179
180 4. Make PCRE_ANCHORED public and accept as a compile option.
181
182 5. Add an options word to pcre_exec() and accept PCRE_ANCHORED and
183 PCRE_CASELESS at run time. Add escapes \A and \I to pcretest to cause it to
184 pass them.
185
186 6. Give an error if bad option bits passed at compile or run time.
187
188 7. Add PCRE_MULTILINE at compile and exec time, and (?m) as well. Add \M to
189 pcretest to cause it to pass that flag.
190
191 8. Add pcre_info(), to get the number of identifying subpatterns, the stored
192 options, and the first character, if set.
193
194 9. Recognize C+ or C{n,m} where n >= 1 as providing a fixed starting character.
195
196
197 Version 0.91 10-Sep-97
198 ----------------------
199
200 1. PCRE was failing to diagnose unlimited repeats of subpatterns that could
201 match the empty string as in /(a*)*/. It was looping and ultimately crashing.
202
203 2. PCRE was looping on encountering an indefinitely repeated back reference to
204 a subpattern that had matched an empty string, e.g. /(a|)\1*/. It now does what
205 Perl does - treats the match as successful.
206
207 ****

webmaster@exim.org
ViewVC Help
Powered by ViewVC 1.1.12