Commit | Line | Data |
754c00ca |
1 | Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org> |
2 | |
3 | 1. Add regression tests for the auto-require of STORABLE_thaw |
165cc789 |
4 | 2. Add auto-require of modules to restore overloading (and tests) |
754c00ca |
5 | |
dfd91409 |
6 | Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org> |
7 | |
8 | Version 2.11 |
9 | |
10 | 1. Storing restricted hashes in canonical order would SEGV. Fixed. |
11 | 2. It was impossible to retrieve references to PL_sv_no and and |
12 | PL_sv_undef from STORABLE_thaw hooks. |
13 | 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique |
14 | implementation of restricted hashes using PL_sv_undef |
15 | 4. These changes allow a space optimisation for restricted hashes. |
16 | |
a8b7ef86 |
17 | Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org> |
18 | |
19 | Version 2.10 |
20 | |
21 | 1. Thread safety: Storable::CLONE/init_perlinterp() now create |
22 | a new Perl context for each new ithread. |
23 | (From Stas Bekman and Jan Dubois.) |
24 | 2. Fix a tag count mismatch with $Storable::Deparse that caused |
25 | all back-references after a stored sub to be off-by-N (where |
26 | N was the number of code references in between). |
27 | (From Sam Vilain.) |
28 | 3. Prevent CODE references from turning into SCALAR references. |
29 | (From Slaven Rezic.) |
30 | |
dcf2e277 |
31 | Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org> |
32 | |
33 | Version 2.09 |
34 | |
35 | Fix minor problems with the CPAN release |
7224c650 |
36 | 1: Make Storable.xs work on 5.8.2 and later (already in the core) |
dcf2e277 |
37 | 2: Ship the linux hints file |
38 | 3: Ship Test::More for the benefit of Perls pre 5.6.2 |
7224c650 |
39 | 4: Correct Makefile.PL to only install in core for 5.8.0 and later |
dcf2e277 |
40 | |
fcaa57e7 |
41 | Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org> |
42 | |
43 | Version 2.08 |
44 | |
45 | This release works around a 5.8.0 bug which caused hashes to not |
46 | be marked as having key flags even though an HEK had HEK_WASUTF8 |
47 | set. (Note that the only reasonable solution is to silently drop |
48 | the flag from the affected key.) |
49 | |
50 | Users of RT 3 who were seeing assertion failures should upgrade. |
51 | (Perl 5.8.1 will have the bug fixed.) |
52 | |
a5271820 |
53 | Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org> |
54 | |
55 | Version 2.07 |
56 | |
57 | Minor bugfixes (self-tied objects are now correctly stored, as |
58 | are the results of additions larger than INT_MAX). |
59 | |
a2307be4 |
60 | Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org> |
61 | |
62 | Version 2.06 |
63 | |
64 | Remove qr// from t/downgrade.t so that it will run on 5.004 |
65 | Mention $File::Spec::VERSION a second time in t/forgive.t so that it |
66 | runs without warnings in 5.004 (this may be a 5.00405 bug I'm working |
67 | round) |
68 | Fix t/integer.t initialisation to actually generate 64 bits of 9c |
69 | Fix comparison tests to use eval to get around 64 bit IV conversion |
70 | issues on 5.6.x, following my t/integer.t ^ precedence bug found by |
71 | Rafael Garcia-Suarez |
72 | Alter t/malice.t to work with Test/More.pm in t/, and skip individual |
73 | subtests that use $Config{ptrsize}, so that the rest of the test can |
74 | now be run with 5.004 |
75 | Change t/malice.t and the error message in check_magic in Storable.xs |
76 | from "Pointer integer size" to "Pointer size" |
77 | Remove prerequisite of Test::More from Makefile.PL |
78 | Ship Test::Builder, Test::Simple and Test::More in t |
79 | |
9cfdba97 |
80 | Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org> |
81 | |
82 | Version 2.05 |
83 | |
84 | Adds support for CODE references from Slaven Rezic |
85 | <slaven.rezic@berlin.de>. |
86 | |
17374ab3 |
87 | Fri Jun 7 23:55:41 BST 2002 Nicholas Clark |
88 | |
89 | Version 2.04 |
90 | |
91 | Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test) |
92 | to fix a recently introduced bug detected by Dave Rolsky. |
93 | Bug was that for a non threaded build, the class information was |
94 | being lost at freeze time on the first object with a STORABLE_freeze |
95 | hook. Consequentially the object was not blessed at all when thawed. |
96 | (The presence (or lack) of STORABLE_thaw was irrelevant; this was |
97 | a store-time data lost bug, caused by failure to initialize internal |
98 | context) |
99 | The bug was introduced as development perl change 16442 (on |
100 | 2002/05/07), so has been present since 2.00. |
101 | Patches to introduce more regression tests to reduce the chance of |
102 | a reoccurance of this sort of goof are always welcome. |
103 | |
ee0f7aac |
104 | Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org> |
105 | |
8932d55c |
106 | Version 2.03 Header changes on 5.6.x on Unix where IV is long long |
107 | |
108 | 5.6.x introduced the ability to have IVs as long long. However, |
109 | Configure still defined BYTEORDER based on the size of a long. |
110 | Storable uses the BYTEORDER value as part of the header, but |
111 | doesn't explicity store sizeof(IV) anywhere in the header. |
112 | Hence on 5.6.x built with IV as long long on a platform that |
113 | uses Configure (ie most things except VMS and Windows) headers |
114 | are identical for the different IV sizes, despite the files |
115 | containing some fields based on sizeof(IV) |
116 | |
117 | 5.8.0 is consistent; all platforms have BYTEORDER in config.h |
118 | based on sizeof(IV) rather than sizeof(long). This means that |
119 | the value of BYTEORDER will change from (say) 4321 to 87654321 |
120 | between 5.6.1 and 5.8.0 built with the same options to Configure |
121 | on the same machine. This means that the Storable header will |
122 | differ, and the two versions will wrongly thing that they are |
123 | incompatible. |
124 | |
125 | For the benefit of long term consistency, Storable now |
126 | implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that |
127 | 2.03 onwards default to be incompatible with 2.02 and earlier |
128 | (ie the large 1.0.x installed base) on the same 5.6.x perl. |
129 | |
130 | To allow interworking, a new variable |
131 | $Storable::interwork_56_64bit is introduced. It defaults to |
132 | false. Set it to true to read and write old format files. Don't |
133 | use it unless you have existing stored data written with 5.6.x |
134 | that you couldn't otherwise read, or you need to interwork with |
135 | a machine running older Storable on a 5.6.x with long long IVs |
136 | (i.e., you probably don't need to use it). |
137 | |
2aeb6432 |
138 | Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org> |
139 | |
25f64a11 |
140 | Version 2.02 |
141 | |
142 | Rewrite Storable.xs so that the file header structure for write_magic |
143 | is built at compile time, and check_magic attempts to the header in |
144 | blocks rather than byte per byte. These changes make the compiled |
145 | extension 2.25% smaller, but are not significant enough to give a |
146 | noticeable speed up. |
2aeb6432 |
147 | |
db670f21 |
148 | Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org> |
149 | |
25f64a11 |
150 | Version 2.01 |
151 | |
152 | - New regression tests integer.t |
153 | - Add code to safely store large unsigned integers. |
154 | - Change code not to attempt to store large integers (ie > 32 bits) |
155 | in network order as 32 bits. |
156 | |
157 | *Never* underestimate the value of a pathological test suite carefully |
158 | crafted with maximum malice before writing a line of real code. It |
159 | prevents crafty bugs from stowing away in your released code. |
160 | It's much less embarrassing to find them before you ship. |
161 | (Well, never underestimate it if you ever want to work for me) |
162 | |
8989364b |
163 | Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org> |
164 | |
25f64a11 |
165 | Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3) |
166 | |
167 | The perl5 porters have decided to make sure that Storable still |
168 | builds on pre-5.8 perls, and make the 5.8 version available on CPAN. |
169 | The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1 |
170 | and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37, |
171 | due to a bug in 5.6.0 - upgrade to 5.6.1. |
172 | |
173 | Jarkko and I have collated the list of changes the perl5 porters have |
174 | from the perl5 Changes file: |
175 | |
176 | - data features of upcoming perl 5.8.0 are supported: Unicode hash |
177 | keys (Unicode hash values have been supported since Storable 1.0.1) |
178 | and "restricted hashes" (readonly hashes and hash entries) |
179 | - a newer version of perl can now be used to serialize data which is |
180 | not supported in earlier perls: Storable will attempt to do the |
181 | right thing for as long as possible, croaking only when safe data |
182 | conversion simply isn't possible. Alternatively earlier perls can |
183 | opt to have a lossy downgrade data instead of croaking |
184 | - when built with perls pre 5.7.3 this Storable writes out files |
185 | with binary format 2.4, the same format as Storable 1.0.8 onwards. |
186 | This should mean that this Storable will inter-operate seamlessly |
187 | with any Storable 1.0.8 or newer on perls pre 5.7.3 |
188 | - dclone() now works with empty string scalar objects |
189 | - retrieving of large hashes is now more efficient |
190 | - more routines autosplit out of the main module, so Storable should |
191 | load slightly more quickly |
192 | - better documentation |
193 | - the internal context objects are now freed explicitly, rather than |
194 | relying on thread or process exit |
195 | - bugs fixed in debugging trace code affecting builds made with 64 bit |
196 | IVs |
197 | - code tidy-ups to allow clean compiles with more warning options |
198 | turned on avoid problems with $@ getting corrupted on 5.005_03 if |
199 | Carp wasn't already loaded |
200 | - added &show_file_magic, so you can add to /etc/magic and teach |
201 | Unix's file command about Storable files |
202 | |
203 | We plan to keep Storable on CPAN in sync with the Perl core, so |
204 | if you encounter bugs or other problems building or using Storable, |
205 | please let us know at perl5-porters@perl.org |
206 | Patches welcome! |
207 | |
596596d5 |
208 | Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
209 | |
25f64a11 |
210 | This is the LAST maintenance release of the Storable module. |
211 | Indeed, Storable is now part of perl 5.8, and will be maintained |
212 | as part of Perl. The CPAN module will remain available there |
213 | for people running pre-5.8 perls. |
596596d5 |
214 | |
25f64a11 |
215 | Avoid requiring Fcntl upfront, useful to embedded runtimes. |
216 | Use an eval {} for testing, instead of making Storable.pm |
217 | simply fail its compilation in the BEGIN block. |
596596d5 |
218 | |
25f64a11 |
219 | store_fd() will now correctly autoflush file if needed. |
596596d5 |
220 | |
6e0ac6f5 |
221 | Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
222 | |
25f64a11 |
223 | Fixed truncation race with lock_retrieve() in lock_store(). |
224 | The file has to be truncated only once the exclusive lock is held. |
6e0ac6f5 |
225 | |
25f64a11 |
226 | Removed spurious debugging messages in .xs file. |
6e0ac6f5 |
227 | |
e993d95c |
228 | Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
229 | |
25f64a11 |
230 | Systematically use "=over 4" for POD linters. |
231 | Apparently, POD linters are much stricter than would |
232 | otherwise be needed, but that's OK. |
e993d95c |
233 | |
25f64a11 |
234 | Fixed memory corruption on croaks during thaw(). Thanks |
235 | to Claudio Garcia for reproducing this bug and providing the |
236 | code to exercise it. Added test cases for this bug, adapted |
237 | from Claudio's code. |
e993d95c |
238 | |
25f64a11 |
239 | Made code compile cleanly with -Wall (from Jarkko Hietaniemi). |
e993d95c |
240 | |
25f64a11 |
241 | Changed tagnum and classnum from I32 to IV in context. Also |
242 | from Jarkko. |
e993d95c |
243 | |
8be2b38b |
244 | Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
245 | |
25f64a11 |
246 | Last version was wrongly compiling with assertions on, due |
247 | to an edit glitch. That did not cause any problem (apart from |
248 | a slight performance loss) excepted on Win* platforms, where the |
249 | assertion code does not compile. |
8be2b38b |
250 | |
251 | Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
252 | |
25f64a11 |
253 | Version 1.0.10. |
8be2b38b |
254 | |
25f64a11 |
255 | Forgot to increase version number at previous patch (there were |
256 | two of them, which is why we jump from 1.0.8 to 1.0.10). |
8be2b38b |
257 | |
b12202d0 |
258 | Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
259 | |
25f64a11 |
260 | Version 1.0.8, binary format 2.4. |
b12202d0 |
261 | |
25f64a11 |
262 | Fixed incorrect error message. |
b12202d0 |
263 | |
25f64a11 |
264 | Now bless objects ASAP at retrieve time, which is meant to fix |
265 | two bugs: |
b12202d0 |
266 | |
25f64a11 |
267 | * Indirect references to overloaded object were not able to |
268 | restore overloading if the object was not blessed yet, |
269 | which was possible since blessing occurred only after the |
270 | recursive retrieval. |
b12202d0 |
271 | |
25f64a11 |
272 | * Storable hooks asking for serialization of blessed ref could |
273 | get un-blessed refs at retrieval time, for the very same |
274 | reason. |
b12202d0 |
275 | |
25f64a11 |
276 | The fix implemented here was suggested by Nick Ing-Simmons. |
b12202d0 |
277 | |
25f64a11 |
278 | Added support for blessed ref to tied structures. This is the |
279 | cause for the binary format change. |
b12202d0 |
280 | |
25f64a11 |
281 | Added EBCDIC version of the compatibility test with 0.6.11, |
282 | from Peter Prymmer |
b12202d0 |
283 | |
25f64a11 |
284 | Added tests for the new features, and to make sure the bugs they |
285 | are meant to fix are indeed fixed. |
b12202d0 |
286 | |
862382c7 |
287 | Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
288 | |
25f64a11 |
289 | Removed spurious 'clean' entry in Makefile.PL. |
862382c7 |
290 | |
25f64a11 |
291 | Added CAN_FLOCK to determine whether we can flock() or not, |
292 | by inspecting Perl's configuration parameters, as determined |
293 | by Configure. |
862382c7 |
294 | |
25f64a11 |
295 | Trace offending package when overloading cannot be restored |
296 | on a scalar. |
862382c7 |
297 | |
25f64a11 |
298 | Made context cleanup safer to avoid dup freeing, mostly in the |
299 | presence of repeated exceptions during store/retrieve (which can |
300 | cause memory leaks anyway, so it's just additional safety, not a |
301 | definite fix). |
862382c7 |
302 | |
90826881 |
303 | Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
304 | |
25f64a11 |
305 | Version 1.0.6. |
90826881 |
306 | |
25f64a11 |
307 | Fixed severe "object lost" bug for STORABLE_freeze returns, |
308 | when refs to lexicals, taken within the hook, were to be |
309 | serialized by Storable. Enhanced the t/recurse.t test to |
310 | stress hook a little more with refs to lexicals. |
90826881 |
311 | |
212e9bde |
312 | Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
313 | |
25f64a11 |
314 | Version 1.0.5. |
212e9bde |
315 | |
25f64a11 |
316 | Documented that store() and retrieve() can return undef. |
317 | That is, the error reporting is not always made via exceptions, |
318 | as the paragraph on error reporting was implying. |
212e9bde |
319 | |
25f64a11 |
320 | Auto requires module of blessed ref when STORABLE_thaw misses. |
321 | When the Storable engine looks for the STORABLE_thaw hook and |
322 | does not find it, it now tries to require the package into which |
323 | the blessed reference is. |
212e9bde |
324 | |
25f64a11 |
325 | Just check $^O, in t/lock.t: there's no need to pull the whole |
326 | Config module for that. |
212e9bde |
327 | |
8be2b38b |
328 | Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
329 | |
25f64a11 |
330 | Version 1.0.4. |
8be2b38b |
331 | |
25f64a11 |
332 | Protected calls to flock() for DOS platform: apparently, the |
333 | flock/fcnlt emulation is reported to be broken on that |
334 | platform. |
8be2b38b |
335 | |
25f64a11 |
336 | Added logcarp emulation if they don't have Log::Agent, since |
337 | we now use it to carp when lock_store/lock_retrieve is used |
338 | on DOS. |
8be2b38b |
339 | |
dd19458b |
340 | Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
341 | |
25f64a11 |
342 | Version 1.0.3. |
dd19458b |
343 | |
25f64a11 |
344 | Avoid using "tainted" and "dirty" since Perl remaps them via |
345 | cpp (i.e. #define). This is deeply harmful when threading |
346 | is enabled. This concerned both the context structure and |
347 | local variable and argument names. Brrr..., scary! |
dd19458b |
348 | |
349 | Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
350 | |
25f64a11 |
351 | Version 1.0.2. |
dd19458b |
352 | |
25f64a11 |
353 | Fixed spelling in README. |
dd19458b |
354 | |
25f64a11 |
355 | Added lock_store, lock_nstore, and lock_retrieve (advisory locking) |
356 | after a proposal from Erik Haugan <erik@solbors.no>. |
dd19458b |
357 | |
25f64a11 |
358 | Perls before 5.004_04 lack newSVpvn, added remapping in XS. |
dd19458b |
359 | |
25f64a11 |
360 | Fixed stupid typo in the t/utf8.t test. |
dd19458b |
361 | |
362 | Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
363 | |
25f64a11 |
364 | Version 1.0.1, binary format 2.3. |
dd19458b |
365 | |
25f64a11 |
366 | Documented that doubles are stored stringified by nstore(). |
dd19458b |
367 | |
25f64a11 |
368 | Added Salvador Ortiz Garcia in CREDITS section, He identified |
369 | a bug in the store hooks and proposed the right fix: the class |
370 | id was allocated too soon. His bug case was also added to |
371 | the regression test suite. |
dd19458b |
372 | |
25f64a11 |
373 | Now only taint retrieved data when source was tainted. A bug |
374 | discovered by Marc Lehmann. |
dd19458b |
375 | |
25f64a11 |
376 | Added support for UTF-8 strings, a contribution of Marc Lehmann. |
377 | This is normally only activated in post-5.6 perls. |
dd19458b |
378 | |
9e21b3d0 |
379 | Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
380 | |
25f64a11 |
381 | First official release Storable 1.0, for inclusion in perl 5.7.0. |
382 | The license scheme is now compatible with Perl's. |
9e21b3d0 |
383 | |
f0ffaed8 |
384 | Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
385 | |
25f64a11 |
386 | ANSI-fied most of the code, preparing for Perl core integration. |
387 | The next version of Storable will be 0.8, and will be integrated |
388 | into the Perl core (development branch). |
f0ffaed8 |
389 | |
25f64a11 |
390 | Dispatch tables were moved upfront to relieve some compilers, |
391 | especially on AIX and Windows platforms. |
f0ffaed8 |
392 | |
25f64a11 |
393 | Merged 64-bit fixes from perl5-porters. |
f0ffaed8 |
394 | |
7a6a85bf |
395 | Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
396 | |
25f64a11 |
397 | Added a refcnt dec in retrieve_tied_key(): sv_magic() increases |
398 | the refcnt on the mg_ptr as well. |
7a6a85bf |
399 | |
25f64a11 |
400 | Removed spurious dependency to Devel::Peek, which was used for |
401 | testing only in t/tied_items.t. Thanks to Conrad Heiney |
402 | <conrad@fringehead.org> for spotting it first. |
7a6a85bf |
403 | |
404 | Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
405 | |
25f64a11 |
406 | Marc Lehmann kindly contributed code to add overloading support |
407 | and to handle references to tied variables. |
7a6a85bf |
408 | |
25f64a11 |
409 | Rewrote leading blurb about compatibility to make it clearer what |
410 | "backward compatibility" is about: when I say 0.7 is backward |
411 | compatible with 0.6, it means the revision 0.7 can read files |
412 | produced by 0.6. |
7a6a85bf |
413 | |
25f64a11 |
414 | Mention new Clone(3) extension in SEE ALSO. |
7a6a85bf |
415 | |
25f64a11 |
416 | Was wrongly optimizing for "undef" values in hashes by not |
417 | fully recursing: as a result, tied "undef" values were incorrectly |
418 | serialized. |
7a6a85bf |
419 | |
420 | Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
421 | |
25f64a11 |
422 | First revision of Storable 0.7. |
7a6a85bf |
423 | |
25f64a11 |
424 | The serializing format is new, known as version 2.0. It is fully |
425 | backward compatible with 0.6. Earlier formats are deprecated and |
426 | have not even been tested: next version will drop pre-0.6 format. |
7a6a85bf |
427 | |
25f64a11 |
428 | Changes since 0.6@11: |
7a6a85bf |
429 | |
25f64a11 |
430 | - Moved interface to the "beta" status. Some tiny parts are still |
431 | subject to change, but nothing important enough to warrant an "alpha" |
432 | status any longer. |
7a6a85bf |
433 | |
25f64a11 |
434 | - Slightly reduced the size of the Storable image by factorizing |
435 | object class names and removing final object storage notification due |
436 | to a redesign of the blessed object storing. |
7a6a85bf |
437 | |
25f64a11 |
438 | - Classes can now redefine how they wish their instances to be serialized |
439 | and/or deep cloned. Serializing hooks are written in Perl code. |
7a6a85bf |
440 | |
25f64a11 |
441 | - The engine is now fully re-entrant. |
7a6a85bf |
442 | |
443 | Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
444 | |
25f64a11 |
445 | Added provision to detect more recent binary formats, since |
446 | the new upcoming Storable-0.7 will use a different format. |
447 | In order to prevent attempting the de-serialization of newer |
448 | formats by older versions, I'm adding this now to the 0.6 series. |
7a6a85bf |
449 | |
25f64a11 |
450 | I'm expecting this revision to be the last of the 0.6 series. |
451 | Unless it does not work with perl 5.6, which I don't use yet, |
452 | and therefore against which I cannot test. |
7a6a85bf |
453 | |
454 | Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
455 | |
25f64a11 |
456 | Added note about format incompatibilities with old versions |
457 | (i.e. pre 0.5@9 formats, which cannot be understood as there |
458 | was no versionning information in the file by then). |
7a6a85bf |
459 | |
25f64a11 |
460 | Protect all $@ variables when eval {} used, to avoid corrupting |
461 | it when store/retrieve is called within an exception handler. |
7a6a85bf |
462 | |
25f64a11 |
463 | Mistakenly included "patchlevel.h" instead of <patchlevel.h>, |
464 | preventing Perl's patchlevel from being included, which is |
465 | needed starting from 5.6. |
7a6a85bf |
466 | |
467 | Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
468 | |
25f64a11 |
469 | Fixed shared "undef" bug in hashes, which did not remain shared |
470 | through store/retrieve. |
7a6a85bf |
471 | |
472 | Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
473 | |
25f64a11 |
474 | added last_op_in_netorder() predicate |
475 | documented last_op_in_netorder() |
476 | added tests for the new last_op_in_netorder() predicate |
7a6a85bf |
477 | |
478 | Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
479 | |
25f64a11 |
480 | Forgot to update VERSION |
7a6a85bf |
481 | |
482 | Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
483 | |
25f64a11 |
484 | Added mention of japanese translation for the manual page. |
7a6a85bf |
485 | |
25f64a11 |
486 | Fixed typo in macro that made threaded code not compilable, |
487 | especially on Win32 platforms. |
7a6a85bf |
488 | |
25f64a11 |
489 | Changed detection of older perls (pre-5.005) by testing PATCHLEVEL |
490 | directly instead of relying on internal symbols. |
7a6a85bf |
491 | |
492 | Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
493 | |
25f64a11 |
494 | Integrated "thread-safe" patch from Murray Nesbitt. |
495 | Note that this may not be very efficient for threaded code, |
496 | see comment in the code. |
7a6a85bf |
497 | |
25f64a11 |
498 | Try to avoid compilation warning on 64-bit CPUs. Can't test it, |
499 | since I don't have access to such machines. |
7a6a85bf |
500 | |
501 | Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com> |
502 | |
25f64a11 |
503 | changed my e-mail to pobox. |
7a6a85bf |
504 | |
25f64a11 |
505 | mentionned it is not thread-safe. |
7a6a85bf |
506 | |
25f64a11 |
507 | updated version number. |
7a6a85bf |
508 | |
25f64a11 |
509 | uses new internal PL_* naming convention. |
7a6a85bf |
510 | |
511 | Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
512 | |
25f64a11 |
513 | Updated benchmark figures due to recent optimizations done in |
514 | store(): tagnums are now stored as-is in the hash table, so |
515 | no surrounding SV is created. And the "shared keys" mode for |
516 | hash table was turned off. |
7a6a85bf |
517 | |
25f64a11 |
518 | Fixed backward compatibility (wrt 0.5@9) for retrieval of |
519 | blessed refs. That old version did something wrong, but the |
520 | bugfix prevented correct retrieval of the old format. |
7a6a85bf |
521 | |
522 | Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
523 | |
25f64a11 |
524 | Changed benchmark figures. |
7a6a85bf |
525 | |
25f64a11 |
526 | Adjust refcnt of tied objects after calling sv_magic() to avoid |
527 | memory leaks. Contributed by Jeff Gresham. |
7a6a85bf |
528 | |
529 | Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
530 | |
25f64a11 |
531 | Added workaround for persistent LVALUE-ness in perl5.004. All |
532 | scalars tagged as being an lvalue are handled as if they were |
533 | not an lvalue at all. Added test for that LVALUE bug workaround. |
7a6a85bf |
534 | |
25f64a11 |
535 | Now handles Perl immortal scalars explicitely, by storing &sv_yes |
536 | as such, explicitely. |
7a6a85bf |
537 | |
25f64a11 |
538 | Retrieval of non-immortal undef cannot be shared. Previous |
539 | version was over-optimizing by not creating a separate SV for |
540 | all undefined scalars seen. |
7a6a85bf |
541 | |
542 | Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
543 | |
25f64a11 |
544 | Baseline for Storable-0.6@0. |
7a6a85bf |
545 | |
25f64a11 |
546 | This version introduces a binary incompatibility in the generated |
547 | binary image, which is more compact than older ones by approximatively |
548 | 15%, depending on the exact degree of sharing in your structures. |
7a6a85bf |
549 | |
25f64a11 |
550 | The good news is that your older images can still be retrieved with |
551 | this version, i.e. backward compatibility is preserved. This version |
552 | of Storable can only generate new binaries however. |
7a6a85bf |
553 | |
25f64a11 |
554 | Another good news is that the retrieval of data structure is |
555 | significantly quicker than before, because a Perl array is used |
556 | instead of a hash table to keep track of retrieved objects, and |
557 | also because the image being smaller, less I/O function calls are |
558 | made. |
7a6a85bf |
559 | |
560 | Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
561 | |
25f64a11 |
562 | Version number now got from Storable.pm directly. |
7a6a85bf |
563 | |
25f64a11 |
564 | Fixed overzealous sv_type() optimization, which would make |
565 | Storable fail when faced with an "upgraded" SV to the PVIV |
566 | or PVNV kind containing a reference. |
7a6a85bf |
567 | |
568 | Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
569 | |
25f64a11 |
570 | Extended the SYNOPSIS section to give quick overview of the |
571 | routines and their signature. |
7a6a85bf |
572 | |
25f64a11 |
573 | Optimized sv_type() to avoid flags checking when not needed, i.e. |
574 | when their type makes it impossible for them to be refs or tied. |
575 | This slightly increases throughput by a few percents when refs |
576 | and tied variables are marginal occurrences in your data. |
7a6a85bf |
577 | |
25f64a11 |
578 | Stubs for XS now use OutputStream and InputStream file types to |
579 | make it work when the given file is actually a socket. Perl |
580 | makes a distinction for sockets in its internal I/O structures |
581 | by having both a read and a write structure, whereas plain files |
582 | share the same one. |
7a6a85bf |
583 | |
584 | Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
585 | |
25f64a11 |
586 | Thanks to a contribution from Benjamin A. Holzman, Storable is now |
587 | able to correctly serialize tied SVs, i.e. tied arrays, hashes |
588 | and scalars. |
7a6a85bf |
589 | |
590 | Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
591 | |
25f64a11 |
592 | I said SvPOK() had changed to SvPOKp(), but that was a lie... |
7a6a85bf |
593 | |
594 | Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
595 | |
25f64a11 |
596 | Wrote sizeof(SV *) instead of sizeof(I32) when portable, which |
597 | in effect mangled the object tags and prevented portability |
598 | accross 32/64 bit architectures! |
7a6a85bf |
599 | |
600 | Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
601 | |
25f64a11 |
602 | Added code example for store_fd() and retrieve_fd() in the |
603 | man page, to emphasize that file descriptors must be passed as |
604 | globs, not as plain strings. |
7a6a85bf |
605 | |
25f64a11 |
606 | Cannot use SV addresses as tag when using nstore() on LP64. This |
607 | was the cause of problems when creating a storable image on an |
608 | LP64 machine and retrieving it on an ILP32 system, which is |
609 | exactly what nstore() is meant for... |
7a6a85bf |
610 | |
25f64a11 |
611 | However, we continue to use SV addresses as tags for plain store(), |
612 | because benchamarking shows that it saves up to 8% of the store |
613 | time, and store() is meant to be fast at the expense of lack |
614 | of portability. |
7a6a85bf |
615 | |
25f64a11 |
616 | This means there will be approximately an 8% degradation of |
617 | performance for nstore(), but it's now working as expected. |
618 | That cost may vary on your machine of course, since it is |
619 | solely caused by the memory allocation overhead used to create |
620 | unique SV tags for each distinct stored SV. |
7a6a85bf |
621 | |
622 | Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
623 | |
25f64a11 |
624 | Don't use any '_' in version number. |
7a6a85bf |
625 | |
626 | Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
627 | |
25f64a11 |
628 | Updated version number. |
7a6a85bf |
629 | |
25f64a11 |
630 | added binmode() calls for systems where it matters. |
7a6a85bf |
631 | |
25f64a11 |
632 | Be sure to pass globs, not plain file strings, to C routines, |
633 | so that Storable can be used under the Perl debugger. |
7a6a85bf |
634 | |
635 | Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
636 | |
25f64a11 |
637 | Fix memory leaks on seen hash table and returned SV refs. |
7a6a85bf |
638 | |
25f64a11 |
639 | Storable did not work properly when tainting enabled. |
7a6a85bf |
640 | |
25f64a11 |
641 | Fixed "Allocation too large" messages in freeze/thaw and added. |
642 | proper regression test in t/freeze.t. |
7a6a85bf |
643 | |
644 | Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
645 | |
25f64a11 |
646 | Updated version number |
7a6a85bf |
647 | |
25f64a11 |
648 | Added freeze/thaw interface and dclone. |
7a6a85bf |
649 | |
650 | Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
651 | |
25f64a11 |
652 | Forgot that AutoLoader does not export its own AUTOLOAD. |
653 | I could use |
7a6a85bf |
654 | |
25f64a11 |
655 | use AutoLoader 'AUTOLOAD'; |
656 | |
657 | but that would not be backward compatible. So the export is |
658 | done by hand... |
7a6a85bf |
659 | |
660 | Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
661 | |
25f64a11 |
662 | Empty scalar strings are now "defined" at retrieval time. |
7a6a85bf |
663 | |
25f64a11 |
664 | New test to ensure an empty string is defined when retrieved. |
7a6a85bf |
665 | |
666 | Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
667 | |
25f64a11 |
668 | Updated version number |
7a6a85bf |
669 | |
25f64a11 |
670 | Declare VERSION as being used |
7a6a85bf |
671 | |
25f64a11 |
672 | Fixed a typo in the PerlIO_putc remapping. |
673 | PerlIO_read and perlIO_write inverted size/nb_items. |
674 | (only relevant for pre-perl5.004 versions) |
7a6a85bf |
675 | |
676 | Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
677 | |
25f64a11 |
678 | Updated version number |
7a6a85bf |
679 | |
25f64a11 |
680 | Added VERSION identification |
7a6a85bf |
681 | |
25f64a11 |
682 | Allow build with perl5.003, which is ante perlIO time |
7a6a85bf |
683 | |
684 | Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
685 | |
25f64a11 |
686 | Random code fixes. |
7a6a85bf |
687 | |
688 | Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com> |
689 | |
25f64a11 |
690 | Updated version number in Makefile.PL. |
7a6a85bf |
691 | |
25f64a11 |
692 | Added "thanks to" section to README. |
7a6a85bf |
693 | |
25f64a11 |
694 | Documented new forgive_me variable. |
7a6a85bf |
695 | |
25f64a11 |
696 | Made 64-bit clean. |
7a6a85bf |
697 | |
25f64a11 |
698 | Added forgive_me support to allow store() of data structures |
699 | containing non-storable items like CODE refs. |