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