1 Sat Nov 17 02:12:12 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
5 Various broken tests fixed. (No functional changes.)
7 Sat Mar 31 06:11:06 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
11 1. Fixes to Storable::dclone, read_magic, retrieve_lscalar
12 2. Storable 0.1 compatibility
13 3. Miscellaneous compile/leak/test/portability fixes
15 Mon May 23 22:48:49 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
19 Minor changes to address a couple of compile problems.
21 Mon Apr 25 07:29:14 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
25 1. Store weak references
26 2. Add STORABLE_attach hook.
28 Thu Jun 17 12:26:43 BST 2004 Nicholas Clark <nick@ccl4.org>
32 1. Don't change the type of top level overloaded references to RV -
33 they are perfectly correct as PVMG
34 2. Storable needs to cope with incoming frozen data that happens to be
37 Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org>
41 1. Add regression tests for the auto-require of STORABLE_thaw
42 2. Add auto-require of modules to restore overloading (and tests)
43 3. Change to no context (should give speedup with ithreads)
45 Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org>
49 1. Storing restricted hashes in canonical order would SEGV. Fixed.
50 2. It was impossible to retrieve references to PL_sv_no and and
51 PL_sv_undef from STORABLE_thaw hooks.
52 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique
53 implementation of restricted hashes using PL_sv_undef
54 4. These changes allow a space optimisation for restricted hashes.
56 Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org>
60 1. Thread safety: Storable::CLONE/init_perlinterp() now create
61 a new Perl context for each new ithread.
62 (From Stas Bekman and Jan Dubois.)
63 2. Fix a tag count mismatch with $Storable::Deparse that caused
64 all back-references after a stored sub to be off-by-N (where
65 N was the number of code references in between).
67 3. Prevent CODE references from turning into SCALAR references.
70 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
74 Fix minor problems with the CPAN release
75 1: Make Storable.xs work on 5.8.2 and later (already in the core)
76 2: Ship the linux hints file
77 3: Ship Test::More for the benefit of Perls pre 5.6.2
78 4: Correct Makefile.PL to only install in core for 5.8.0 and later
80 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
84 This release works around a 5.8.0 bug which caused hashes to not
85 be marked as having key flags even though an HEK had HEK_WASUTF8
86 set. (Note that the only reasonable solution is to silently drop
87 the flag from the affected key.)
89 Users of RT 3 who were seeing assertion failures should upgrade.
90 (Perl 5.8.1 will have the bug fixed.)
92 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
96 Minor bugfixes (self-tied objects are now correctly stored, as
97 are the results of additions larger than INT_MAX).
99 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
103 Remove qr// from t/downgrade.t so that it will run on 5.004
104 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
105 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
107 Fix t/integer.t initialisation to actually generate 64 bits of 9c
108 Fix comparison tests to use eval to get around 64 bit IV conversion
109 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
111 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
112 subtests that use $Config{ptrsize}, so that the rest of the test can
113 now be run with 5.004
114 Change t/malice.t and the error message in check_magic in Storable.xs
115 from "Pointer integer size" to "Pointer size"
116 Remove prerequisite of Test::More from Makefile.PL
117 Ship Test::Builder, Test::Simple and Test::More in t
119 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
123 Adds support for CODE references from Slaven Rezic
124 <slaven.rezic@berlin.de>.
126 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
130 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
131 to fix a recently introduced bug detected by Dave Rolsky.
132 Bug was that for a non threaded build, the class information was
133 being lost at freeze time on the first object with a STORABLE_freeze
134 hook. Consequentially the object was not blessed at all when thawed.
135 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
136 a store-time data lost bug, caused by failure to initialize internal
138 The bug was introduced as development perl change 16442 (on
139 2002/05/07), so has been present since 2.00.
140 Patches to introduce more regression tests to reduce the chance of
141 a reoccurance of this sort of goof are always welcome.
143 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
145 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
147 5.6.x introduced the ability to have IVs as long long. However,
148 Configure still defined BYTEORDER based on the size of a long.
149 Storable uses the BYTEORDER value as part of the header, but
150 doesn't explicity store sizeof(IV) anywhere in the header.
151 Hence on 5.6.x built with IV as long long on a platform that
152 uses Configure (ie most things except VMS and Windows) headers
153 are identical for the different IV sizes, despite the files
154 containing some fields based on sizeof(IV)
156 5.8.0 is consistent; all platforms have BYTEORDER in config.h
157 based on sizeof(IV) rather than sizeof(long). This means that
158 the value of BYTEORDER will change from (say) 4321 to 87654321
159 between 5.6.1 and 5.8.0 built with the same options to Configure
160 on the same machine. This means that the Storable header will
161 differ, and the two versions will wrongly thing that they are
164 For the benefit of long term consistency, Storable now
165 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
166 2.03 onwards default to be incompatible with 2.02 and earlier
167 (ie the large 1.0.x installed base) on the same 5.6.x perl.
169 To allow interworking, a new variable
170 $Storable::interwork_56_64bit is introduced. It defaults to
171 false. Set it to true to read and write old format files. Don't
172 use it unless you have existing stored data written with 5.6.x
173 that you couldn't otherwise read, or you need to interwork with
174 a machine running older Storable on a 5.6.x with long long IVs
175 (i.e., you probably don't need to use it).
177 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
181 Rewrite Storable.xs so that the file header structure for write_magic
182 is built at compile time, and check_magic attempts to the header in
183 blocks rather than byte per byte. These changes make the compiled
184 extension 2.25% smaller, but are not significant enough to give a
187 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
191 - New regression tests integer.t
192 - Add code to safely store large unsigned integers.
193 - Change code not to attempt to store large integers (ie > 32 bits)
194 in network order as 32 bits.
196 *Never* underestimate the value of a pathological test suite carefully
197 crafted with maximum malice before writing a line of real code. It
198 prevents crafty bugs from stowing away in your released code.
199 It's much less embarrassing to find them before you ship.
200 (Well, never underestimate it if you ever want to work for me)
202 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
204 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
206 The perl5 porters have decided to make sure that Storable still
207 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
208 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
209 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
210 due to a bug in 5.6.0 - upgrade to 5.6.1.
212 Jarkko and I have collated the list of changes the perl5 porters have
213 from the perl5 Changes file:
215 - data features of upcoming perl 5.8.0 are supported: Unicode hash
216 keys (Unicode hash values have been supported since Storable 1.0.1)
217 and "restricted hashes" (readonly hashes and hash entries)
218 - a newer version of perl can now be used to serialize data which is
219 not supported in earlier perls: Storable will attempt to do the
220 right thing for as long as possible, croaking only when safe data
221 conversion simply isn't possible. Alternatively earlier perls can
222 opt to have a lossy downgrade data instead of croaking
223 - when built with perls pre 5.7.3 this Storable writes out files
224 with binary format 2.4, the same format as Storable 1.0.8 onwards.
225 This should mean that this Storable will inter-operate seamlessly
226 with any Storable 1.0.8 or newer on perls pre 5.7.3
227 - dclone() now works with empty string scalar objects
228 - retrieving of large hashes is now more efficient
229 - more routines autosplit out of the main module, so Storable should
230 load slightly more quickly
231 - better documentation
232 - the internal context objects are now freed explicitly, rather than
233 relying on thread or process exit
234 - bugs fixed in debugging trace code affecting builds made with 64 bit
236 - code tidy-ups to allow clean compiles with more warning options
237 turned on avoid problems with $@ getting corrupted on 5.005_03 if
238 Carp wasn't already loaded
239 - added &show_file_magic, so you can add to /etc/magic and teach
240 Unix's file command about Storable files
242 We plan to keep Storable on CPAN in sync with the Perl core, so
243 if you encounter bugs or other problems building or using Storable,
244 please let us know at perl5-porters@perl.org
247 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
249 This is the LAST maintenance release of the Storable module.
250 Indeed, Storable is now part of perl 5.8, and will be maintained
251 as part of Perl. The CPAN module will remain available there
252 for people running pre-5.8 perls.
254 Avoid requiring Fcntl upfront, useful to embedded runtimes.
255 Use an eval {} for testing, instead of making Storable.pm
256 simply fail its compilation in the BEGIN block.
258 store_fd() will now correctly autoflush file if needed.
260 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
262 Fixed truncation race with lock_retrieve() in lock_store().
263 The file has to be truncated only once the exclusive lock is held.
265 Removed spurious debugging messages in .xs file.
267 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
269 Systematically use "=over 4" for POD linters.
270 Apparently, POD linters are much stricter than would
271 otherwise be needed, but that's OK.
273 Fixed memory corruption on croaks during thaw(). Thanks
274 to Claudio Garcia for reproducing this bug and providing the
275 code to exercise it. Added test cases for this bug, adapted
278 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
280 Changed tagnum and classnum from I32 to IV in context. Also
283 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
285 Last version was wrongly compiling with assertions on, due
286 to an edit glitch. That did not cause any problem (apart from
287 a slight performance loss) excepted on Win* platforms, where the
288 assertion code does not compile.
290 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
294 Forgot to increase version number at previous patch (there were
295 two of them, which is why we jump from 1.0.8 to 1.0.10).
297 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
299 Version 1.0.8, binary format 2.4.
301 Fixed incorrect error message.
303 Now bless objects ASAP at retrieve time, which is meant to fix
306 * Indirect references to overloaded object were not able to
307 restore overloading if the object was not blessed yet,
308 which was possible since blessing occurred only after the
311 * Storable hooks asking for serialization of blessed ref could
312 get un-blessed refs at retrieval time, for the very same
315 The fix implemented here was suggested by Nick Ing-Simmons.
317 Added support for blessed ref to tied structures. This is the
318 cause for the binary format change.
320 Added EBCDIC version of the compatibility test with 0.6.11,
323 Added tests for the new features, and to make sure the bugs they
324 are meant to fix are indeed fixed.
326 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
328 Removed spurious 'clean' entry in Makefile.PL.
330 Added CAN_FLOCK to determine whether we can flock() or not,
331 by inspecting Perl's configuration parameters, as determined
334 Trace offending package when overloading cannot be restored
337 Made context cleanup safer to avoid dup freeing, mostly in the
338 presence of repeated exceptions during store/retrieve (which can
339 cause memory leaks anyway, so it's just additional safety, not a
342 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
346 Fixed severe "object lost" bug for STORABLE_freeze returns,
347 when refs to lexicals, taken within the hook, were to be
348 serialized by Storable. Enhanced the t/recurse.t test to
349 stress hook a little more with refs to lexicals.
351 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
355 Documented that store() and retrieve() can return undef.
356 That is, the error reporting is not always made via exceptions,
357 as the paragraph on error reporting was implying.
359 Auto requires module of blessed ref when STORABLE_thaw misses.
360 When the Storable engine looks for the STORABLE_thaw hook and
361 does not find it, it now tries to require the package into which
362 the blessed reference is.
364 Just check $^O, in t/lock.t: there's no need to pull the whole
365 Config module for that.
367 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
371 Protected calls to flock() for DOS platform: apparently, the
372 flock/fcnlt emulation is reported to be broken on that
375 Added logcarp emulation if they don't have Log::Agent, since
376 we now use it to carp when lock_store/lock_retrieve is used
379 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
383 Avoid using "tainted" and "dirty" since Perl remaps them via
384 cpp (i.e. #define). This is deeply harmful when threading
385 is enabled. This concerned both the context structure and
386 local variable and argument names. Brrr..., scary!
388 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
392 Fixed spelling in README.
394 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
395 after a proposal from Erik Haugan <erik@solbors.no>.
397 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
399 Fixed stupid typo in the t/utf8.t test.
401 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
403 Version 1.0.1, binary format 2.3.
405 Documented that doubles are stored stringified by nstore().
407 Added Salvador Ortiz Garcia in CREDITS section, He identified
408 a bug in the store hooks and proposed the right fix: the class
409 id was allocated too soon. His bug case was also added to
410 the regression test suite.
412 Now only taint retrieved data when source was tainted. A bug
413 discovered by Marc Lehmann.
415 Added support for UTF-8 strings, a contribution of Marc Lehmann.
416 This is normally only activated in post-5.6 perls.
418 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
420 First official release Storable 1.0, for inclusion in perl 5.7.0.
421 The license scheme is now compatible with Perl's.
423 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
425 ANSI-fied most of the code, preparing for Perl core integration.
426 The next version of Storable will be 0.8, and will be integrated
427 into the Perl core (development branch).
429 Dispatch tables were moved upfront to relieve some compilers,
430 especially on AIX and Windows platforms.
432 Merged 64-bit fixes from perl5-porters.
434 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
436 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
437 the refcnt on the mg_ptr as well.
439 Removed spurious dependency to Devel::Peek, which was used for
440 testing only in t/tied_items.t. Thanks to Conrad Heiney
441 <conrad@fringehead.org> for spotting it first.
443 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
445 Marc Lehmann kindly contributed code to add overloading support
446 and to handle references to tied variables.
448 Rewrote leading blurb about compatibility to make it clearer what
449 "backward compatibility" is about: when I say 0.7 is backward
450 compatible with 0.6, it means the revision 0.7 can read files
453 Mention new Clone(3) extension in SEE ALSO.
455 Was wrongly optimizing for "undef" values in hashes by not
456 fully recursing: as a result, tied "undef" values were incorrectly
459 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
461 First revision of Storable 0.7.
463 The serializing format is new, known as version 2.0. It is fully
464 backward compatible with 0.6. Earlier formats are deprecated and
465 have not even been tested: next version will drop pre-0.6 format.
467 Changes since 0.6@11:
469 - Moved interface to the "beta" status. Some tiny parts are still
470 subject to change, but nothing important enough to warrant an "alpha"
473 - Slightly reduced the size of the Storable image by factorizing
474 object class names and removing final object storage notification due
475 to a redesign of the blessed object storing.
477 - Classes can now redefine how they wish their instances to be serialized
478 and/or deep cloned. Serializing hooks are written in Perl code.
480 - The engine is now fully re-entrant.
482 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
484 Added provision to detect more recent binary formats, since
485 the new upcoming Storable-0.7 will use a different format.
486 In order to prevent attempting the de-serialization of newer
487 formats by older versions, I'm adding this now to the 0.6 series.
489 I'm expecting this revision to be the last of the 0.6 series.
490 Unless it does not work with perl 5.6, which I don't use yet,
491 and therefore against which I cannot test.
493 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
495 Added note about format incompatibilities with old versions
496 (i.e. pre 0.5@9 formats, which cannot be understood as there
497 was no versionning information in the file by then).
499 Protect all $@ variables when eval {} used, to avoid corrupting
500 it when store/retrieve is called within an exception handler.
502 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
503 preventing Perl's patchlevel from being included, which is
504 needed starting from 5.6.
506 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
508 Fixed shared "undef" bug in hashes, which did not remain shared
509 through store/retrieve.
511 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
513 added last_op_in_netorder() predicate
514 documented last_op_in_netorder()
515 added tests for the new last_op_in_netorder() predicate
517 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
519 Forgot to update VERSION
521 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
523 Added mention of japanese translation for the manual page.
525 Fixed typo in macro that made threaded code not compilable,
526 especially on Win32 platforms.
528 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
529 directly instead of relying on internal symbols.
531 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
533 Integrated "thread-safe" patch from Murray Nesbitt.
534 Note that this may not be very efficient for threaded code,
535 see comment in the code.
537 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
538 since I don't have access to such machines.
540 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
542 changed my e-mail to pobox.
544 mentionned it is not thread-safe.
546 updated version number.
548 uses new internal PL_* naming convention.
550 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
552 Updated benchmark figures due to recent optimizations done in
553 store(): tagnums are now stored as-is in the hash table, so
554 no surrounding SV is created. And the "shared keys" mode for
555 hash table was turned off.
557 Fixed backward compatibility (wrt 0.5@9) for retrieval of
558 blessed refs. That old version did something wrong, but the
559 bugfix prevented correct retrieval of the old format.
561 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
563 Changed benchmark figures.
565 Adjust refcnt of tied objects after calling sv_magic() to avoid
566 memory leaks. Contributed by Jeff Gresham.
568 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
570 Added workaround for persistent LVALUE-ness in perl5.004. All
571 scalars tagged as being an lvalue are handled as if they were
572 not an lvalue at all. Added test for that LVALUE bug workaround.
574 Now handles Perl immortal scalars explicitely, by storing &sv_yes
575 as such, explicitely.
577 Retrieval of non-immortal undef cannot be shared. Previous
578 version was over-optimizing by not creating a separate SV for
579 all undefined scalars seen.
581 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
583 Baseline for Storable-0.6@0.
585 This version introduces a binary incompatibility in the generated
586 binary image, which is more compact than older ones by approximatively
587 15%, depending on the exact degree of sharing in your structures.
589 The good news is that your older images can still be retrieved with
590 this version, i.e. backward compatibility is preserved. This version
591 of Storable can only generate new binaries however.
593 Another good news is that the retrieval of data structure is
594 significantly quicker than before, because a Perl array is used
595 instead of a hash table to keep track of retrieved objects, and
596 also because the image being smaller, less I/O function calls are
599 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
601 Version number now got from Storable.pm directly.
603 Fixed overzealous sv_type() optimization, which would make
604 Storable fail when faced with an "upgraded" SV to the PVIV
605 or PVNV kind containing a reference.
607 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
609 Extended the SYNOPSIS section to give quick overview of the
610 routines and their signature.
612 Optimized sv_type() to avoid flags checking when not needed, i.e.
613 when their type makes it impossible for them to be refs or tied.
614 This slightly increases throughput by a few percents when refs
615 and tied variables are marginal occurrences in your data.
617 Stubs for XS now use OutputStream and InputStream file types to
618 make it work when the given file is actually a socket. Perl
619 makes a distinction for sockets in its internal I/O structures
620 by having both a read and a write structure, whereas plain files
623 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
625 Thanks to a contribution from Benjamin A. Holzman, Storable is now
626 able to correctly serialize tied SVs, i.e. tied arrays, hashes
629 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
631 I said SvPOK() had changed to SvPOKp(), but that was a lie...
633 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
635 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
636 in effect mangled the object tags and prevented portability
637 accross 32/64 bit architectures!
639 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
641 Added code example for store_fd() and retrieve_fd() in the
642 man page, to emphasize that file descriptors must be passed as
643 globs, not as plain strings.
645 Cannot use SV addresses as tag when using nstore() on LP64. This
646 was the cause of problems when creating a storable image on an
647 LP64 machine and retrieving it on an ILP32 system, which is
648 exactly what nstore() is meant for...
650 However, we continue to use SV addresses as tags for plain store(),
651 because benchamarking shows that it saves up to 8% of the store
652 time, and store() is meant to be fast at the expense of lack
655 This means there will be approximately an 8% degradation of
656 performance for nstore(), but it's now working as expected.
657 That cost may vary on your machine of course, since it is
658 solely caused by the memory allocation overhead used to create
659 unique SV tags for each distinct stored SV.
661 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
663 Don't use any '_' in version number.
665 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
667 Updated version number.
669 added binmode() calls for systems where it matters.
671 Be sure to pass globs, not plain file strings, to C routines,
672 so that Storable can be used under the Perl debugger.
674 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
676 Fix memory leaks on seen hash table and returned SV refs.
678 Storable did not work properly when tainting enabled.
680 Fixed "Allocation too large" messages in freeze/thaw and added.
681 proper regression test in t/freeze.t.
683 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
685 Updated version number
687 Added freeze/thaw interface and dclone.
689 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
691 Forgot that AutoLoader does not export its own AUTOLOAD.
694 use AutoLoader 'AUTOLOAD';
696 but that would not be backward compatible. So the export is
699 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
701 Empty scalar strings are now "defined" at retrieval time.
703 New test to ensure an empty string is defined when retrieved.
705 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
707 Updated version number
709 Declare VERSION as being used
711 Fixed a typo in the PerlIO_putc remapping.
712 PerlIO_read and perlIO_write inverted size/nb_items.
713 (only relevant for pre-perl5.004 versions)
715 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
717 Updated version number
719 Added VERSION identification
721 Allow build with perl5.003, which is ante perlIO time
723 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
727 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
729 Updated version number in Makefile.PL.
731 Added "thanks to" section to README.
733 Documented new forgive_me variable.
737 Added forgive_me support to allow store() of data structures
738 containing non-storable items like CODE refs.