1 Sat Mar 31 06:11:06 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
5 1. Fixes to Storable::dclone, read_magic, retrieve_lscalar
6 2. Storable 0.1 compatibility
7 3. Miscellaneous compile/leak/test/portability fixes
9 Mon May 23 22:48:49 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
13 Minor changes to address a couple of compile problems.
15 Mon Apr 25 07:29:14 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
19 1. Store weak references
20 2. Add STORABLE_attach hook.
22 Thu Jun 17 12:26:43 BST 2004 Nicholas Clark <nick@ccl4.org>
26 1. Don't change the type of top level overloaded references to RV -
27 they are perfectly correct as PVMG
28 2. Storable needs to cope with incoming frozen data that happens to be
31 Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org>
35 1. Add regression tests for the auto-require of STORABLE_thaw
36 2. Add auto-require of modules to restore overloading (and tests)
37 3. Change to no context (should give speedup with ithreads)
39 Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org>
43 1. Storing restricted hashes in canonical order would SEGV. Fixed.
44 2. It was impossible to retrieve references to PL_sv_no and and
45 PL_sv_undef from STORABLE_thaw hooks.
46 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique
47 implementation of restricted hashes using PL_sv_undef
48 4. These changes allow a space optimisation for restricted hashes.
50 Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org>
54 1. Thread safety: Storable::CLONE/init_perlinterp() now create
55 a new Perl context for each new ithread.
56 (From Stas Bekman and Jan Dubois.)
57 2. Fix a tag count mismatch with $Storable::Deparse that caused
58 all back-references after a stored sub to be off-by-N (where
59 N was the number of code references in between).
61 3. Prevent CODE references from turning into SCALAR references.
64 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
68 Fix minor problems with the CPAN release
69 1: Make Storable.xs work on 5.8.2 and later (already in the core)
70 2: Ship the linux hints file
71 3: Ship Test::More for the benefit of Perls pre 5.6.2
72 4: Correct Makefile.PL to only install in core for 5.8.0 and later
74 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
78 This release works around a 5.8.0 bug which caused hashes to not
79 be marked as having key flags even though an HEK had HEK_WASUTF8
80 set. (Note that the only reasonable solution is to silently drop
81 the flag from the affected key.)
83 Users of RT 3 who were seeing assertion failures should upgrade.
84 (Perl 5.8.1 will have the bug fixed.)
86 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
90 Minor bugfixes (self-tied objects are now correctly stored, as
91 are the results of additions larger than INT_MAX).
93 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
97 Remove qr// from t/downgrade.t so that it will run on 5.004
98 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
99 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
101 Fix t/integer.t initialisation to actually generate 64 bits of 9c
102 Fix comparison tests to use eval to get around 64 bit IV conversion
103 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
105 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
106 subtests that use $Config{ptrsize}, so that the rest of the test can
107 now be run with 5.004
108 Change t/malice.t and the error message in check_magic in Storable.xs
109 from "Pointer integer size" to "Pointer size"
110 Remove prerequisite of Test::More from Makefile.PL
111 Ship Test::Builder, Test::Simple and Test::More in t
113 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
117 Adds support for CODE references from Slaven Rezic
118 <slaven.rezic@berlin.de>.
120 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
124 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
125 to fix a recently introduced bug detected by Dave Rolsky.
126 Bug was that for a non threaded build, the class information was
127 being lost at freeze time on the first object with a STORABLE_freeze
128 hook. Consequentially the object was not blessed at all when thawed.
129 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
130 a store-time data lost bug, caused by failure to initialize internal
132 The bug was introduced as development perl change 16442 (on
133 2002/05/07), so has been present since 2.00.
134 Patches to introduce more regression tests to reduce the chance of
135 a reoccurance of this sort of goof are always welcome.
137 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
139 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
141 5.6.x introduced the ability to have IVs as long long. However,
142 Configure still defined BYTEORDER based on the size of a long.
143 Storable uses the BYTEORDER value as part of the header, but
144 doesn't explicity store sizeof(IV) anywhere in the header.
145 Hence on 5.6.x built with IV as long long on a platform that
146 uses Configure (ie most things except VMS and Windows) headers
147 are identical for the different IV sizes, despite the files
148 containing some fields based on sizeof(IV)
150 5.8.0 is consistent; all platforms have BYTEORDER in config.h
151 based on sizeof(IV) rather than sizeof(long). This means that
152 the value of BYTEORDER will change from (say) 4321 to 87654321
153 between 5.6.1 and 5.8.0 built with the same options to Configure
154 on the same machine. This means that the Storable header will
155 differ, and the two versions will wrongly thing that they are
158 For the benefit of long term consistency, Storable now
159 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
160 2.03 onwards default to be incompatible with 2.02 and earlier
161 (ie the large 1.0.x installed base) on the same 5.6.x perl.
163 To allow interworking, a new variable
164 $Storable::interwork_56_64bit is introduced. It defaults to
165 false. Set it to true to read and write old format files. Don't
166 use it unless you have existing stored data written with 5.6.x
167 that you couldn't otherwise read, or you need to interwork with
168 a machine running older Storable on a 5.6.x with long long IVs
169 (i.e., you probably don't need to use it).
171 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
175 Rewrite Storable.xs so that the file header structure for write_magic
176 is built at compile time, and check_magic attempts to the header in
177 blocks rather than byte per byte. These changes make the compiled
178 extension 2.25% smaller, but are not significant enough to give a
181 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
185 - New regression tests integer.t
186 - Add code to safely store large unsigned integers.
187 - Change code not to attempt to store large integers (ie > 32 bits)
188 in network order as 32 bits.
190 *Never* underestimate the value of a pathological test suite carefully
191 crafted with maximum malice before writing a line of real code. It
192 prevents crafty bugs from stowing away in your released code.
193 It's much less embarrassing to find them before you ship.
194 (Well, never underestimate it if you ever want to work for me)
196 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
198 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
200 The perl5 porters have decided to make sure that Storable still
201 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
202 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
203 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
204 due to a bug in 5.6.0 - upgrade to 5.6.1.
206 Jarkko and I have collated the list of changes the perl5 porters have
207 from the perl5 Changes file:
209 - data features of upcoming perl 5.8.0 are supported: Unicode hash
210 keys (Unicode hash values have been supported since Storable 1.0.1)
211 and "restricted hashes" (readonly hashes and hash entries)
212 - a newer version of perl can now be used to serialize data which is
213 not supported in earlier perls: Storable will attempt to do the
214 right thing for as long as possible, croaking only when safe data
215 conversion simply isn't possible. Alternatively earlier perls can
216 opt to have a lossy downgrade data instead of croaking
217 - when built with perls pre 5.7.3 this Storable writes out files
218 with binary format 2.4, the same format as Storable 1.0.8 onwards.
219 This should mean that this Storable will inter-operate seamlessly
220 with any Storable 1.0.8 or newer on perls pre 5.7.3
221 - dclone() now works with empty string scalar objects
222 - retrieving of large hashes is now more efficient
223 - more routines autosplit out of the main module, so Storable should
224 load slightly more quickly
225 - better documentation
226 - the internal context objects are now freed explicitly, rather than
227 relying on thread or process exit
228 - bugs fixed in debugging trace code affecting builds made with 64 bit
230 - code tidy-ups to allow clean compiles with more warning options
231 turned on avoid problems with $@ getting corrupted on 5.005_03 if
232 Carp wasn't already loaded
233 - added &show_file_magic, so you can add to /etc/magic and teach
234 Unix's file command about Storable files
236 We plan to keep Storable on CPAN in sync with the Perl core, so
237 if you encounter bugs or other problems building or using Storable,
238 please let us know at perl5-porters@perl.org
241 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
243 This is the LAST maintenance release of the Storable module.
244 Indeed, Storable is now part of perl 5.8, and will be maintained
245 as part of Perl. The CPAN module will remain available there
246 for people running pre-5.8 perls.
248 Avoid requiring Fcntl upfront, useful to embedded runtimes.
249 Use an eval {} for testing, instead of making Storable.pm
250 simply fail its compilation in the BEGIN block.
252 store_fd() will now correctly autoflush file if needed.
254 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
256 Fixed truncation race with lock_retrieve() in lock_store().
257 The file has to be truncated only once the exclusive lock is held.
259 Removed spurious debugging messages in .xs file.
261 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
263 Systematically use "=over 4" for POD linters.
264 Apparently, POD linters are much stricter than would
265 otherwise be needed, but that's OK.
267 Fixed memory corruption on croaks during thaw(). Thanks
268 to Claudio Garcia for reproducing this bug and providing the
269 code to exercise it. Added test cases for this bug, adapted
272 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
274 Changed tagnum and classnum from I32 to IV in context. Also
277 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
279 Last version was wrongly compiling with assertions on, due
280 to an edit glitch. That did not cause any problem (apart from
281 a slight performance loss) excepted on Win* platforms, where the
282 assertion code does not compile.
284 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
288 Forgot to increase version number at previous patch (there were
289 two of them, which is why we jump from 1.0.8 to 1.0.10).
291 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
293 Version 1.0.8, binary format 2.4.
295 Fixed incorrect error message.
297 Now bless objects ASAP at retrieve time, which is meant to fix
300 * Indirect references to overloaded object were not able to
301 restore overloading if the object was not blessed yet,
302 which was possible since blessing occurred only after the
305 * Storable hooks asking for serialization of blessed ref could
306 get un-blessed refs at retrieval time, for the very same
309 The fix implemented here was suggested by Nick Ing-Simmons.
311 Added support for blessed ref to tied structures. This is the
312 cause for the binary format change.
314 Added EBCDIC version of the compatibility test with 0.6.11,
317 Added tests for the new features, and to make sure the bugs they
318 are meant to fix are indeed fixed.
320 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
322 Removed spurious 'clean' entry in Makefile.PL.
324 Added CAN_FLOCK to determine whether we can flock() or not,
325 by inspecting Perl's configuration parameters, as determined
328 Trace offending package when overloading cannot be restored
331 Made context cleanup safer to avoid dup freeing, mostly in the
332 presence of repeated exceptions during store/retrieve (which can
333 cause memory leaks anyway, so it's just additional safety, not a
336 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
340 Fixed severe "object lost" bug for STORABLE_freeze returns,
341 when refs to lexicals, taken within the hook, were to be
342 serialized by Storable. Enhanced the t/recurse.t test to
343 stress hook a little more with refs to lexicals.
345 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
349 Documented that store() and retrieve() can return undef.
350 That is, the error reporting is not always made via exceptions,
351 as the paragraph on error reporting was implying.
353 Auto requires module of blessed ref when STORABLE_thaw misses.
354 When the Storable engine looks for the STORABLE_thaw hook and
355 does not find it, it now tries to require the package into which
356 the blessed reference is.
358 Just check $^O, in t/lock.t: there's no need to pull the whole
359 Config module for that.
361 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
365 Protected calls to flock() for DOS platform: apparently, the
366 flock/fcnlt emulation is reported to be broken on that
369 Added logcarp emulation if they don't have Log::Agent, since
370 we now use it to carp when lock_store/lock_retrieve is used
373 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
377 Avoid using "tainted" and "dirty" since Perl remaps them via
378 cpp (i.e. #define). This is deeply harmful when threading
379 is enabled. This concerned both the context structure and
380 local variable and argument names. Brrr..., scary!
382 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
386 Fixed spelling in README.
388 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
389 after a proposal from Erik Haugan <erik@solbors.no>.
391 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
393 Fixed stupid typo in the t/utf8.t test.
395 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
397 Version 1.0.1, binary format 2.3.
399 Documented that doubles are stored stringified by nstore().
401 Added Salvador Ortiz Garcia in CREDITS section, He identified
402 a bug in the store hooks and proposed the right fix: the class
403 id was allocated too soon. His bug case was also added to
404 the regression test suite.
406 Now only taint retrieved data when source was tainted. A bug
407 discovered by Marc Lehmann.
409 Added support for UTF-8 strings, a contribution of Marc Lehmann.
410 This is normally only activated in post-5.6 perls.
412 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
414 First official release Storable 1.0, for inclusion in perl 5.7.0.
415 The license scheme is now compatible with Perl's.
417 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
419 ANSI-fied most of the code, preparing for Perl core integration.
420 The next version of Storable will be 0.8, and will be integrated
421 into the Perl core (development branch).
423 Dispatch tables were moved upfront to relieve some compilers,
424 especially on AIX and Windows platforms.
426 Merged 64-bit fixes from perl5-porters.
428 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
430 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
431 the refcnt on the mg_ptr as well.
433 Removed spurious dependency to Devel::Peek, which was used for
434 testing only in t/tied_items.t. Thanks to Conrad Heiney
435 <conrad@fringehead.org> for spotting it first.
437 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
439 Marc Lehmann kindly contributed code to add overloading support
440 and to handle references to tied variables.
442 Rewrote leading blurb about compatibility to make it clearer what
443 "backward compatibility" is about: when I say 0.7 is backward
444 compatible with 0.6, it means the revision 0.7 can read files
447 Mention new Clone(3) extension in SEE ALSO.
449 Was wrongly optimizing for "undef" values in hashes by not
450 fully recursing: as a result, tied "undef" values were incorrectly
453 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
455 First revision of Storable 0.7.
457 The serializing format is new, known as version 2.0. It is fully
458 backward compatible with 0.6. Earlier formats are deprecated and
459 have not even been tested: next version will drop pre-0.6 format.
461 Changes since 0.6@11:
463 - Moved interface to the "beta" status. Some tiny parts are still
464 subject to change, but nothing important enough to warrant an "alpha"
467 - Slightly reduced the size of the Storable image by factorizing
468 object class names and removing final object storage notification due
469 to a redesign of the blessed object storing.
471 - Classes can now redefine how they wish their instances to be serialized
472 and/or deep cloned. Serializing hooks are written in Perl code.
474 - The engine is now fully re-entrant.
476 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
478 Added provision to detect more recent binary formats, since
479 the new upcoming Storable-0.7 will use a different format.
480 In order to prevent attempting the de-serialization of newer
481 formats by older versions, I'm adding this now to the 0.6 series.
483 I'm expecting this revision to be the last of the 0.6 series.
484 Unless it does not work with perl 5.6, which I don't use yet,
485 and therefore against which I cannot test.
487 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
489 Added note about format incompatibilities with old versions
490 (i.e. pre 0.5@9 formats, which cannot be understood as there
491 was no versionning information in the file by then).
493 Protect all $@ variables when eval {} used, to avoid corrupting
494 it when store/retrieve is called within an exception handler.
496 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
497 preventing Perl's patchlevel from being included, which is
498 needed starting from 5.6.
500 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
502 Fixed shared "undef" bug in hashes, which did not remain shared
503 through store/retrieve.
505 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
507 added last_op_in_netorder() predicate
508 documented last_op_in_netorder()
509 added tests for the new last_op_in_netorder() predicate
511 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
513 Forgot to update VERSION
515 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
517 Added mention of japanese translation for the manual page.
519 Fixed typo in macro that made threaded code not compilable,
520 especially on Win32 platforms.
522 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
523 directly instead of relying on internal symbols.
525 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
527 Integrated "thread-safe" patch from Murray Nesbitt.
528 Note that this may not be very efficient for threaded code,
529 see comment in the code.
531 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
532 since I don't have access to such machines.
534 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
536 changed my e-mail to pobox.
538 mentionned it is not thread-safe.
540 updated version number.
542 uses new internal PL_* naming convention.
544 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
546 Updated benchmark figures due to recent optimizations done in
547 store(): tagnums are now stored as-is in the hash table, so
548 no surrounding SV is created. And the "shared keys" mode for
549 hash table was turned off.
551 Fixed backward compatibility (wrt 0.5@9) for retrieval of
552 blessed refs. That old version did something wrong, but the
553 bugfix prevented correct retrieval of the old format.
555 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
557 Changed benchmark figures.
559 Adjust refcnt of tied objects after calling sv_magic() to avoid
560 memory leaks. Contributed by Jeff Gresham.
562 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
564 Added workaround for persistent LVALUE-ness in perl5.004. All
565 scalars tagged as being an lvalue are handled as if they were
566 not an lvalue at all. Added test for that LVALUE bug workaround.
568 Now handles Perl immortal scalars explicitely, by storing &sv_yes
569 as such, explicitely.
571 Retrieval of non-immortal undef cannot be shared. Previous
572 version was over-optimizing by not creating a separate SV for
573 all undefined scalars seen.
575 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
577 Baseline for Storable-0.6@0.
579 This version introduces a binary incompatibility in the generated
580 binary image, which is more compact than older ones by approximatively
581 15%, depending on the exact degree of sharing in your structures.
583 The good news is that your older images can still be retrieved with
584 this version, i.e. backward compatibility is preserved. This version
585 of Storable can only generate new binaries however.
587 Another good news is that the retrieval of data structure is
588 significantly quicker than before, because a Perl array is used
589 instead of a hash table to keep track of retrieved objects, and
590 also because the image being smaller, less I/O function calls are
593 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
595 Version number now got from Storable.pm directly.
597 Fixed overzealous sv_type() optimization, which would make
598 Storable fail when faced with an "upgraded" SV to the PVIV
599 or PVNV kind containing a reference.
601 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
603 Extended the SYNOPSIS section to give quick overview of the
604 routines and their signature.
606 Optimized sv_type() to avoid flags checking when not needed, i.e.
607 when their type makes it impossible for them to be refs or tied.
608 This slightly increases throughput by a few percents when refs
609 and tied variables are marginal occurrences in your data.
611 Stubs for XS now use OutputStream and InputStream file types to
612 make it work when the given file is actually a socket. Perl
613 makes a distinction for sockets in its internal I/O structures
614 by having both a read and a write structure, whereas plain files
617 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
619 Thanks to a contribution from Benjamin A. Holzman, Storable is now
620 able to correctly serialize tied SVs, i.e. tied arrays, hashes
623 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
625 I said SvPOK() had changed to SvPOKp(), but that was a lie...
627 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
629 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
630 in effect mangled the object tags and prevented portability
631 accross 32/64 bit architectures!
633 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
635 Added code example for store_fd() and retrieve_fd() in the
636 man page, to emphasize that file descriptors must be passed as
637 globs, not as plain strings.
639 Cannot use SV addresses as tag when using nstore() on LP64. This
640 was the cause of problems when creating a storable image on an
641 LP64 machine and retrieving it on an ILP32 system, which is
642 exactly what nstore() is meant for...
644 However, we continue to use SV addresses as tags for plain store(),
645 because benchamarking shows that it saves up to 8% of the store
646 time, and store() is meant to be fast at the expense of lack
649 This means there will be approximately an 8% degradation of
650 performance for nstore(), but it's now working as expected.
651 That cost may vary on your machine of course, since it is
652 solely caused by the memory allocation overhead used to create
653 unique SV tags for each distinct stored SV.
655 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
657 Don't use any '_' in version number.
659 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
661 Updated version number.
663 added binmode() calls for systems where it matters.
665 Be sure to pass globs, not plain file strings, to C routines,
666 so that Storable can be used under the Perl debugger.
668 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
670 Fix memory leaks on seen hash table and returned SV refs.
672 Storable did not work properly when tainting enabled.
674 Fixed "Allocation too large" messages in freeze/thaw and added.
675 proper regression test in t/freeze.t.
677 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
679 Updated version number
681 Added freeze/thaw interface and dclone.
683 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
685 Forgot that AutoLoader does not export its own AUTOLOAD.
688 use AutoLoader 'AUTOLOAD';
690 but that would not be backward compatible. So the export is
693 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
695 Empty scalar strings are now "defined" at retrieval time.
697 New test to ensure an empty string is defined when retrieved.
699 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
701 Updated version number
703 Declare VERSION as being used
705 Fixed a typo in the PerlIO_putc remapping.
706 PerlIO_read and perlIO_write inverted size/nb_items.
707 (only relevant for pre-perl5.004 versions)
709 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
711 Updated version number
713 Added VERSION identification
715 Allow build with perl5.003, which is ante perlIO time
717 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
721 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
723 Updated version number in Makefile.PL.
725 Added "thanks to" section to README.
727 Documented new forgive_me variable.
731 Added forgive_me support to allow store() of data structures
732 containing non-storable items like CODE refs.