1 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
5 Version 1.0.8, binary format 2.4.
7 Fixed incorrect error message.
9 Now bless objects ASAP at retrieve time, which is meant to fix
12 * Indirect references to overloaded object were not able to
13 restore overloading if the object was not blessed yet,
14 which was possible since blessing occurred only after the
17 * Storable hooks asking for serialization of blessed ref could
18 get un-blessed refs at retrieval time, for the very same
21 The fix implemented here was suggested by Nick Ing-Simmons.
23 Added support for blessed ref to tied structures. This is the
24 cause for the binary format change.
26 Added EBCDIC version of the compatibility test with 0.6.11,
29 Added tests for the new features, and to make sure the bugs they
30 are meant to fix are indeed fixed.
32 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
36 Removed spurious 'clean' entry in Makefile.PL.
38 Added CAN_FLOCK to determine whether we can flock() or not,
39 by inspecting Perl's configuration parameters, as determined
42 Trace offending package when overloading cannot be restored
45 Made context cleanup safer to avoid dup freeing, mostly in the
46 presence of repeated exceptions during store/retrieve (which can
47 cause memory leaks anyway, so it's just additional safety, not a
50 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
56 Fixed severe "object lost" bug for STORABLE_freeze returns,
57 when refs to lexicals, taken within the hook, were to be
58 serialized by Storable. Enhanced the t/recurse.t test to
59 stress hook a little more with refs to lexicals.
61 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
67 Documented that store() and retrieve() can return undef.
68 That is, the error reporting is not always made via exceptions,
69 as the paragraph on error reporting was implying.
71 Auto requires module of blessed ref when STORABLE_thaw misses.
72 When the Storable engine looks for the STORABLE_thaw hook and
73 does not find it, it now tries to require the package into which
74 the blessed reference is.
76 Just check $^O, in t/lock.t: there's no need to pull the whole
77 Config module for that.
79 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
85 Avoid using "tainted" and "dirty" since Perl remaps them via
86 cpp (i.e. #define). This is deeply harmful when threading
87 is enabled. This concerned both the context structure and
88 local variable and argument names. Brrr..., scary!
90 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
96 Fixed spelling in README.
98 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
99 after a proposal from Erik Haugan <erik@solbors.no>.
101 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
103 Fixed stupid typo in the t/utf8.t test.
105 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
109 Version 1.0.1, binary format 2.3.
111 Documented that doubles are stored stringified by nstore().
113 Added Salvador Ortiz Garcia in CREDITS section, He identified
114 a bug in the store hooks and proposed the right fix: the class
115 id was allocated too soon. His bug case was also added to
116 the regression test suite.
118 Now only taint retrieved data when source was tainted. A bug
119 discovered by Marc Lehmann.
121 Added support for UTF-8 strings, a contribution of Marc Lehmann.
122 This is normally only activated in post-5.6 perls.
124 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
126 First official release Storable 1.0, for inclusion in perl 5.7.0.
127 The license scheme is now compatible with Perl's.
129 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
134 ANSI-fied most of the code, preparing for Perl core integration.
135 The next version of Storable will be 0.8, and will be integrated
136 into the Perl core (development branch).
138 Dispatch tables were moved upfront to relieve some compilers,
139 especially on AIX and Windows platforms.
141 Merged 64-bit fixes from perl5-porters.
143 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
147 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
148 the refcnt on the mg_ptr as well.
150 Removed spurious dependency to Devel::Peek, which was used for
151 testing only in t/tied_items.t. Thanks to Conrad Heiney
152 <conrad@fringehead.org> for spotting it first.
154 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
158 Marc Lehmann kindly contributed code to add overloading support
159 and to handle references to tied variables.
161 Rewrote leading blurb about compatibility to make it clearer what
162 "backward compatibility" is about: when I say 0.7 is backward
163 compatible with 0.6, it means the revision 0.7 can read files
166 Mention new Clone(3) extension in SEE ALSO.
168 Was wrongly optimizing for "undef" values in hashes by not
169 fully recursing: as a result, tied "undef" values were incorrectly
172 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
174 First revision of Storable 0.7.
176 The serializing format is new, known as version 2.0. It is fully
177 backward compatible with 0.6. Earlier formats are deprecated and
178 have not even been tested: next version will drop pre-0.6 format.
180 Changes since 0.6@11:
182 - Moved interface to the "beta" status. Some tiny parts are still
183 subject to change, but nothing important enough to warrant an "alpha"
186 - Slightly reduced the size of the Storable image by factorizing
187 object class names and removing final object storage notification due
188 to a redesign of the blessed object storing.
190 - Classes can now redefine how they wish their instances to be serialized
191 and/or deep cloned. Serializing hooks are written in Perl code.
193 - The engine is now fully re-entrant.
195 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
199 Added provision to detect more recent binary formats, since
200 the new upcoming Storable-0.7 will use a different format.
201 In order to prevent attempting the de-serialization of newer
202 formats by older versions, I'm adding this now to the 0.6 series.
204 I'm expecting this revision to be the last of the 0.6 series.
205 Unless it does not work with perl 5.6, which I don't use yet,
206 and therefore against which I cannot test.
208 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
212 Added note about format incompatibilities with old versions
213 (i.e. pre 0.5@9 formats, which cannot be understood as there
214 was no versionning information in the file by then).
216 Protect all $@ variables when eval {} used, to avoid corrupting
217 it when store/retrieve is called within an exception handler.
219 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
220 preventing Perl's patchlevel from being included, which is
221 needed starting from 5.6.
223 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
227 Fixed shared "undef" bug in hashes, which did not remain shared
228 through store/retrieve.
230 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
234 added last_op_in_netorder() predicate
235 documented last_op_in_netorder()
236 added tests for the new last_op_in_netorder() predicate
238 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
242 Forgot to update VERSION
244 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
248 Added mention of japanese translation for the manual page.
250 Fixed typo in macro that made threaded code not compilable,
251 especially on Win32 platforms.
253 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
254 directly instead of relying on internal symbols.
256 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
260 Integrated "thread-safe" patch from Murray Nesbitt.
261 Note that this may not be very efficient for threaded code,
262 see comment in the code.
264 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
265 since I don't have access to such machines.
267 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
271 changed my e-mail to pobox.
273 mentionned it is not thread-safe.
275 updated version number.
277 uses new internal PL_* naming convention.
279 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
283 Updated benchmark figures due to recent optimizations done in
284 store(): tagnums are now stored as-is in the hash table, so
285 no surrounding SV is created. And the "shared keys" mode for
286 hash table was turned off.
288 Fixed backward compatibility (wrt 0.5@9) for retrieval of
289 blessed refs. That old version did something wrong, but the
290 bugfix prevented correct retrieval of the old format.
292 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
296 Changed benchmark figures.
298 Adjust refcnt of tied objects after calling sv_magic() to avoid
299 memory leaks. Contributed by Jeff Gresham.
301 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
305 Added workaround for persistent LVALUE-ness in perl5.004. All
306 scalars tagged as being an lvalue are handled as if they were
307 not an lvalue at all. Added test for that LVALUE bug workaround.
309 Now handles Perl immortal scalars explicitely, by storing &sv_yes
310 as such, explicitely.
312 Retrieval of non-immortal undef cannot be shared. Previous
313 version was over-optimizing by not creating a separate SV for
314 all undefined scalars seen.
316 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
320 Baseline for Storable-0.6@0.
322 This version introduces a binary incompatibility in the generated
323 binary image, which is more compact than older ones by approximatively
324 15%, depending on the exact degree of sharing in your structures.
326 The good news is that your older images can still be retrieved with
327 this version, i.e. backward compatibility is preserved. This version
328 of Storable can only generate new binaries however.
330 Another good news is that the retrieval of data structure is
331 significantly quicker than before, because a Perl array is used
332 instead of a hash table to keep track of retrieved objects, and
333 also because the image being smaller, less I/O function calls are
336 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
340 Version number now got from Storable.pm directly.
342 Fixed overzealous sv_type() optimization, which would make
343 Storable fail when faced with an "upgraded" SV to the PVIV
344 or PVNV kind containing a reference.
346 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
350 Extended the SYNOPSIS section to give quick overview of the
351 routines and their signature.
353 Optimized sv_type() to avoid flags checking when not needed, i.e.
354 when their type makes it impossible for them to be refs or tied.
355 This slightly increases throughput by a few percents when refs
356 and tied variables are marginal occurrences in your data.
358 Stubs for XS now use OutputStream and InputStream file types to
359 make it work when the given file is actually a socket. Perl
360 makes a distinction for sockets in its internal I/O structures
361 by having both a read and a write structure, whereas plain files
364 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
368 Thanks to a contribution from Benjamin A. Holzman, Storable is now
369 able to correctly serialize tied SVs, i.e. tied arrays, hashes
372 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
376 I said SvPOK() had changed to SvPOKp(), but that was a lie...
378 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
382 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
383 in effect mangled the object tags and prevented portability
384 accross 32/64 bit architectures!
386 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
390 Added code example for store_fd() and retrieve_fd() in the
391 man page, to emphasize that file descriptors must be passed as
392 globs, not as plain strings.
394 Cannot use SV addresses as tag when using nstore() on LP64. This
395 was the cause of problems when creating a storable image on an
396 LP64 machine and retrieving it on an ILP32 system, which is
397 exactly what nstore() is meant for...
399 However, we continue to use SV addresses as tags for plain store(),
400 because benchamarking shows that it saves up to 8% of the store
401 time, and store() is meant to be fast at the expense of lack
404 This means there will be approximately an 8% degradation of
405 performance for nstore(), but it's now working as expected.
406 That cost may vary on your machine of course, since it is
407 solely caused by the memory allocation overhead used to create
408 unique SV tags for each distinct stored SV.
410 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
414 Don't use any '_' in version number.
416 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
420 Updated version number.
422 added binmode() calls for systems where it matters.
424 Be sure to pass globs, not plain file strings, to C routines,
425 so that Storable can be used under the Perl debugger.
427 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
431 Fix memory leaks on seen hash table and returned SV refs.
433 Storable did not work properly when tainting enabled.
435 Fixed "Allocation too large" messages in freeze/thaw and added.
436 proper regression test in t/freeze.t.
438 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
442 Updated version number
444 Added freeze/thaw interface and dclone.
446 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
450 Forgot that AutoLoader does not export its own AUTOLOAD.
453 use AutoLoader 'AUTOLOAD';
455 but that would not be backward compatible. So the export is
458 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
462 Empty scalar strings are now "defined" at retrieval time.
464 New test to ensure an empty string is defined when retrieved.
466 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
470 Updated version number
472 Declare VERSION as being used
474 Fixed a typo in the PerlIO_putc remapping.
475 PerlIO_read and perlIO_write inverted size/nb_items.
476 (only relevant for pre-perl5.004 versions)
478 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
482 Updated version number
484 Added VERSION identification
486 Allow build with perl5.003, which is ante perlIO time
488 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
494 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
498 Updated version number in Makefile.PL.
500 Added "thanks to" section to README.
502 Documented new forgive_me variable.
506 Added forgive_me support to allow store() of data structures
507 containing non-storable items like CODE refs.