/[pcre]/code/trunk/maint/README
ViewVC logotype

Contents of /code/trunk/maint/README

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1217 - (show annotations)
Fri Nov 9 17:07:01 2012 UTC (6 years, 9 months ago) by ph10
File size: 16335 byte(s)
Preparations for 8.32-RC1.
1 MAINTENANCE README FOR PCRE
2 ===========================
3
4 The files in the "maint" directory of the PCRE source contain data, scripts,
5 and programs that are used for the maintenance of PCRE, but which do not form
6 part of the PCRE distribution tarballs. This document describes these files and
7 also contains some notes for maintainers. Its contents are:
8
9 Files in the maint directory
10 Updating to a new Unicode release
11 Preparing for a PCRE release
12 Making a PCRE release
13 Long-term ideas (wish list)
14
15
16 Files in the maint directory
17 ============================
18
19 ---------------- This file is now OBSOLETE and no longer used ----------------
20 Builducptable A Perl script that creates the contents of the ucptable.h file
21 from two Unicode data files, which themselves are downloaded
22 from the Unicode web site. Run this script in the "maint"
23 directory.
24 ---------------- This file is now OBSOLETE and no longer used ----------------
25
26 GenerateUtt.py A Python script to generate part of the pcre_tables.c file
27 that contains Unicode script names in a long string with
28 offsets, which is tedious to maintain by hand.
29
30 ManyConfigTests A shell script that runs "configure, make, test" a number of
31 times with different configuration settings.
32
33 MultiStage2.py A Python script that generates the file pcre_ucd.c from three
34 Unicode data tables, which are themselves downloaded from the
35 Unicode web site. Run this script in the "maint" directory.
36 The generated file contains the tables for a 2-stage lookup
37 of Unicode properties.
38
39 pcre_chartables.c.non-standard
40 This is a set of character tables that came from a Windows
41 system. It has characters greater than 128 that are set as
42 spaces, amongst other things. I kept it so that it can be
43 used for testing from time to time.
44
45 README This file.
46
47 Unicode.tables The files in this directory (CaseFolding.txt,
48 DerivedGeneralCategory.txt, GraphemeBreakProperty.txt,
49 Scripts.txt and UnicodeData.txt) were downloaded from the
50 Unicode web site. They contain information about Unicode
51 characters and scripts.
52
53 ucptest.c A short C program for testing the Unicode property macros
54 that do lookups in the pcre_ucd.c data, mainly useful after
55 rebuilding the Unicode property table. Compile and run this in
56 the "maint" directory (see comments at its head).
57
58 ucptestdata A directory containing two files, testinput1 and testoutput1,
59 to use in conjunction with the ucptest program.
60
61 utf8.c A short, freestanding C program for converting a Unicode code
62 point into a sequence of bytes in the UTF-8 encoding, and vice
63 versa. If its argument is a hex number such as 0x1234, it
64 outputs a list of the equivalent UTF-8 bytes. If its argument
65 is sequence of concatenated UTF-8 bytes (e.g. e188b4) it
66 treats them as a UTF-8 character and outputs the equivalent
67 code point in hex.
68
69
70 Updating to a new Unicode release
71 =================================
72
73 When there is a new release of Unicode, the files in Unicode.tables must be
74 refreshed from the web site. If the new version of Unicode adds new character
75 scripts, the source file ucp.h and both the MultiStage2.py and the
76 GenerateUtt.py scripts must be edited to add the new names. Then MultiStage2.py
77 can be run to generate a new version of pcre_ucd.c, and GenerateUtt.py can be
78 run to generate the tricky tables for inclusion in pcre_tables.c.
79
80 If MultiStage2.py gives the error "ValueError: list.index(x): x not in list",
81 the cause is usually a missing (or misspelt) name in the list of scripts. I
82 couldn't find a straightforward list of scripts on the Unicode site, but
83 there's a useful Wikipedia page that list them, and notes the Unicode version
84 in which they were introduced:
85
86 http://en.wikipedia.org/wiki/Unicode_scripts#Table_of_Unicode_scripts
87
88 The ucptest program can be compiled and used to check that the new tables in
89 pcre_ucd.c work properly, using the data files in ucptestdata to check a number
90 of test characters. The source file ucptest.c must be updated whenever new
91 Unicode script names are added.
92
93 Note also that both the pcresyntax.3 and pcrepattern.3 man pages contain lists
94 of Unicode script names.
95
96
97 Preparing for a PCRE release
98 ============================
99
100 This section contains a checklist of things that I consult before building a
101 distribution for a new release.
102
103 . Ensure that the version number and version date are correct in configure.ac.
104
105 . Update the library version numbers in configure.ac according to the rules
106 given below.
107
108 . If new build options have been added, ensure that they are added to the CMake
109 files as well as to the autoconf files. The relevant files are CMakeLists.txt
110 and config-cmake.h.in. After making a release tarball, test it out with CMake
111 if there have been changes here.
112
113 . Run ./autogen.sh to ensure everything is up-to-date.
114
115 . Compile and test with many different config options, and combinations of
116 options. Also, test with valgrind by running "RunTest valgrind" and
117 "RunGrepTest valgrind" (which takes quite a long time). The script
118 maint/ManyConfigTests now encapsulates this testing. It runs tests with
119 different configurations, and it also runs some of them with valgrind, all of
120 which can take quite some time.
121
122 . Run perltest.pl on the test data for tests 1, 4, and 6. The output
123 should match the PCRE test output, apart from the version identification at
124 the start of each test. The other tests are not Perl-compatible (they use
125 various PCRE-specific features or options).
126
127 . It is possible to test with the emulated memmove() function by undefining
128 HAVE_MEMMOVE and HAVE_BCOPY in config.h, though I do not do this often. You
129 may see a number of "pcre_memmove defined but not used" warnings for the
130 modules in which there is no call to memmove(). These can be ignored.
131
132 . Documentation: check AUTHORS, ChangeLog (check version and date), LICENCE,
133 NEWS (check version and date), NON-AUTOTOOLS-BUILD, and README. Many of these
134 won't need changing, but over the long term things do change.
135
136 . I used to test new releases myself on a number of different operating
137 systems, using different compilers as well. For example, on Solaris it is
138 helpful to test using Sun's cc compiler as a change from gcc. Adding
139 -xarch=v9 to the cc options does a 64-bit test, but it also needs -S 64 for
140 pcretest to increase the stack size for test 2. Since I retired I can no
141 longer do this, but instead I rely on putting out release candidates for
142 folks on the pcre-dev list to test.
143
144
145 Updating version info for libtool
146 =================================
147
148 This set of rules for updating library version information came from a web page
149 whose URL I have forgotten. The version information consists of three parts:
150 (current, revision, age).
151
152 1. Start with version information of 0:0:0 for each libtool library.
153
154 2. Update the version information only immediately before a public release of
155 your software. More frequent updates are unnecessary, and only guarantee
156 that the current interface number gets larger faster.
157
158 3. If the library source code has changed at all since the last update, then
159 increment revision; c:r:a becomes c:r+1:a.
160
161 4. If any interfaces have been added, removed, or changed since the last
162 update, increment current, and set revision to 0.
163
164 5. If any interfaces have been added since the last public release, then
165 increment age.
166
167 6. If any interfaces have been removed or changed since the last public
168 release, then set age to 0.
169
170 The following explanation may help in understanding the above rules a bit
171 better. Consider that there are three possible kinds of reaction from users to
172 changes in a shared library:
173
174 1. Programs using the previous version may use the new version as a drop-in
175 replacement, and programs using the new version can also work with the
176 previous one. In other words, no recompiling nor relinking is needed. In
177 this case, increment revision only, don't touch current or age.
178
179 2. Programs using the previous version may use the new version as a drop-in
180 replacement, but programs using the new version may use APIs not present in
181 the previous one. In other words, a program linking against the new version
182 may fail if linked against the old version at run time. In this case, set
183 revision to 0, increment current and age.
184
185 3. Programs may need to be changed, recompiled, relinked in order to use the
186 new version. Increment current, set revision and age to 0.
187
188
189 Making a PCRE release
190 =====================
191
192 Run PrepareRelease and commit the files that it changes (by removing trailing
193 spaces). The first thing this script does is to run CheckMan on the man pages;
194 if it finds any markup errors, it reports them and then aborts.
195
196 Once PrepareRelease has run clean, run "make distcheck" to create the tarballs
197 and the zipball. Double-check with "svn status", then create an SVN tagged
198 copy:
199
200 svn copy svn://vcs.exim.org/pcre/code/trunk \
201 svn://vcs.exim.org/pcre/code/tags/pcre-8.xx
202
203 Don't forget to update Freshmeat when the new release is out, and to tell
204 webmaster@pcre.org and the mailing list. Also, update the list of version
205 numbers in Bugzilla (edit products).
206
207
208 Future ideas (wish list)
209 ========================
210
211 This section records a list of ideas so that they do not get forgotten. They
212 vary enormously in their usefulness and potential for implementation. Some are
213 very sensible; some are rather wacky. Some have been on this list for years;
214 others are relatively new.
215
216 . Optimization
217
218 There are always ideas for new optimizations so as to speed up pattern
219 matching. Most of them try to save work by recognizing a non-match without
220 having to scan all the possibilities. These are some that I've recorded:
221
222 * /((A{0,5}){0,5}){0,5}(something complex)/ on a non-matching string is very
223 slow, though Perl is fast. Can we speed up somehow? Convert to {0,125}?
224 OTOH, this is pathological - the user could easily fix it.
225
226 * Turn ={4} into ==== ? (for speed). I once did an experiment, and it seems
227 to have little effect, and maybe makes things worse.
228
229 * "Ends with literal string" - note that a single character doesn't gain much
230 over the existing "required byte" (reqbyte) feature that just remembers one
231 data unit.
232
233 * These probably need to go in pcre_study():
234
235 o Remember an initial string rather than just 1 char?
236
237 o A required data unit from alternatives - not just the last unit, but an
238 earlier one if common to all alternatives.
239
240 o Friedl contains other ideas.
241
242 * pcre_study() does not set initial byte flags for Unicode property types
243 such as \p; I don't know how much benefit there would be for, for example,
244 setting the bits for 0-9 and all bytes >= xC0 when a pattern starts with
245 \p{N}.
246
247 * There is scope for more "auto-possessifying" in connection with \p and \P.
248
249 . If Perl gets to a consistent state over the settings of capturing sub-
250 patterns inside repeats, see if we can match it. One example of the
251 difference is the matching of /(main(O)?)+/ against mainOmain, where PCRE
252 leaves $2 set. In Perl, it's unset. Changing this in PCRE will be very hard
253 because I think it needs much more state to be remembered.
254
255 . Perl 6 will be a revolution. Is it a revolution too far for PCRE?
256
257 . Allow errorptr and erroroffset to be NULL. I don't like this idea.
258
259 . Line endings:
260
261 * Option to use NUL as a line terminator in subject strings. This could now
262 be done relatively easily since the extension to support LF, CR, and CRLF.
263 If it is done, a suitable option for pcregrep is also required.
264
265 . Option to provide the pattern with a length instead of with a NUL terminator.
266 This affects quite a few places in the code and is not trivial.
267
268 . Catch SIGSEGV for stack overflows?
269
270 . A feature to suspend a match via a callout was once requested.
271
272 . Option to convert results into character offsets and character lengths.
273
274 . Option for pcregrep to scan only the start of a file. I am not keen - this is
275 the job of "head".
276
277 . A (non-Unix) user wanted pcregrep options to (a) list a file name just once,
278 preceded by a blank line, instead of adding it to every matched line, and (b)
279 support --outputfile=name.
280
281 . Consider making UTF-8 and UCP the default for PCRE n.0 for some n > 8.
282 (And now presumably UTF-16 and UCP for the 16-bit library, and UTF-32 and UCP
283 for the 32-bit library.)
284
285 . Add a user pointer to pcre_malloc/free functions -- some option would be
286 needed to retain backward compatibility.
287
288 . Define a union for the results from pcre_fullinfo().
289
290 . Provide a "random access to the subject" facility so that the way in which it
291 is stored is independent of PCRE. For efficiency, it probably isn't possible
292 to switch this dynamically. It would have to be specified when PCRE was
293 compiled. PCRE would then call a function every time it wanted a character.
294
295 . Wild thought: the ability to compile from PCRE's internal byte code to a real
296 FSM and a very fast (third) matcher to process the result. There would be
297 even more restrictions than for pcre_dfa_exec(), however. This is not easy.
298 This is probably obsolete now that we have the JIT support.
299
300 . Should pcretest have some private locale data, to avoid relying on the
301 available locales for the test data, since different OS have different ideas?
302 This won't be as thorough a test, but perhaps that doesn't really matter.
303
304 . pcregrep: add -rs for a sorted recurse? Having to store file names and sort
305 them will of course slow it down.
306
307 . Someone suggested --disable-callout to save code space when callouts are
308 never wanted. This seems rather marginal.
309
310 . Check names that consist entirely of digits: PCRE allows, but do Perl and
311 Python, etc?
312
313 . A user suggested a parameter to limit the length of string matched, for
314 example if the parameter is N, the current match should fail if the matched
315 substring exceeds N. This could apply to both match functions. The value
316 could be a new field in the extra block.
317
318 . Callouts with arguments: (?Cn:ARG) for instance.
319
320 . A user is going to supply a patch to generalize the API for user-specific
321 memory allocation so that it is more flexible in threaded environments. This
322 was promised a long time ago, and never appeared. However, this is a live
323 issue not only for threaded environments, but for libraries that use PCRE and
324 want not to be beholden to their caller's memory allocation.
325
326 . Write a wrapper to maintain a structure with specified runtime parameters,
327 such as recurse limit, and pass these to PCRE each time it is called. Also
328 maybe malloc and free. A user sent a prototype. This relates the the previous
329 item.
330
331 . Write a function that generates random matching strings for a compiled regex.
332
333 . Pcregrep: an option to specify the output line separator, either as a string
334 or select from a fixed list. This is not dead easy, because at the moment it
335 outputs whatever is in the input file.
336
337 . Improve the code for duplicate checking in pcre_dfa_exec(). An incomplete,
338 non-thread-safe patch showed that this can help performance for patterns
339 where there are many alternatives. However, a simple thread-safe
340 implementation that I tried made things worse in many simple cases, so this
341 is not an obviously good thing.
342
343 . PCRE cannot at present distinguish between subpatterns with different names,
344 but the same number (created by the use of ?|). In order to do so, a way of
345 remembering *which* subpattern numbered n matched is needed. Bugzilla #760.
346 Now that (*MARK) has been implemented, it can perhaps be used as a way round
347 this problem.
348
349 . Instead of having #ifdef HAVE_CONFIG_H in each module, put #include
350 "something" and the the #ifdef appears only in one place, in "something".
351
352 Philip Hazel
353 Email local part: ph10
354 Email domain: cam.ac.uk
355 Last updated: 09 November 2012

  ViewVC Help
Powered by ViewVC 1.1.5