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