1 Mon May 23 22:48:49 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
5 Minor changes to address a couple of compile problems.
7 Mon Apr 25 07:29:14 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
11 1. Store weak references
12 2. Add STORABLE_attach hook.
14 Thu Jun 17 12:26:43 BST 2004 Nicholas Clark <nick@ccl4.org>
18 1. Don't change the type of top level overloaded references to RV -
19 they are perfectly correct as PVMG
20 2. Storable needs to cope with incoming frozen data that happens to be
23 Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org>
27 1. Add regression tests for the auto-require of STORABLE_thaw
28 2. Add auto-require of modules to restore overloading (and tests)
29 3. Change to no context (should give speedup with ithreads)
31 Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org>
35 1. Storing restricted hashes in canonical order would SEGV. Fixed.
36 2. It was impossible to retrieve references to PL_sv_no and and
37 PL_sv_undef from STORABLE_thaw hooks.
38 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique
39 implementation of restricted hashes using PL_sv_undef
40 4. These changes allow a space optimisation for restricted hashes.
42 Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org>
46 1. Thread safety: Storable::CLONE/init_perlinterp() now create
47 a new Perl context for each new ithread.
48 (From Stas Bekman and Jan Dubois.)
49 2. Fix a tag count mismatch with $Storable::Deparse that caused
50 all back-references after a stored sub to be off-by-N (where
51 N was the number of code references in between).
53 3. Prevent CODE references from turning into SCALAR references.
56 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
60 Fix minor problems with the CPAN release
61 1: Make Storable.xs work on 5.8.2 and later (already in the core)
62 2: Ship the linux hints file
63 3: Ship Test::More for the benefit of Perls pre 5.6.2
64 4: Correct Makefile.PL to only install in core for 5.8.0 and later
66 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
70 This release works around a 5.8.0 bug which caused hashes to not
71 be marked as having key flags even though an HEK had HEK_WASUTF8
72 set. (Note that the only reasonable solution is to silently drop
73 the flag from the affected key.)
75 Users of RT 3 who were seeing assertion failures should upgrade.
76 (Perl 5.8.1 will have the bug fixed.)
78 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
82 Minor bugfixes (self-tied objects are now correctly stored, as
83 are the results of additions larger than INT_MAX).
85 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
89 Remove qr// from t/downgrade.t so that it will run on 5.004
90 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
91 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
93 Fix t/integer.t initialisation to actually generate 64 bits of 9c
94 Fix comparison tests to use eval to get around 64 bit IV conversion
95 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
97 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
98 subtests that use $Config{ptrsize}, so that the rest of the test can
100 Change t/malice.t and the error message in check_magic in Storable.xs
101 from "Pointer integer size" to "Pointer size"
102 Remove prerequisite of Test::More from Makefile.PL
103 Ship Test::Builder, Test::Simple and Test::More in t
105 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
109 Adds support for CODE references from Slaven Rezic
110 <slaven.rezic@berlin.de>.
112 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
116 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
117 to fix a recently introduced bug detected by Dave Rolsky.
118 Bug was that for a non threaded build, the class information was
119 being lost at freeze time on the first object with a STORABLE_freeze
120 hook. Consequentially the object was not blessed at all when thawed.
121 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
122 a store-time data lost bug, caused by failure to initialize internal
124 The bug was introduced as development perl change 16442 (on
125 2002/05/07), so has been present since 2.00.
126 Patches to introduce more regression tests to reduce the chance of
127 a reoccurance of this sort of goof are always welcome.
129 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
131 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
133 5.6.x introduced the ability to have IVs as long long. However,
134 Configure still defined BYTEORDER based on the size of a long.
135 Storable uses the BYTEORDER value as part of the header, but
136 doesn't explicity store sizeof(IV) anywhere in the header.
137 Hence on 5.6.x built with IV as long long on a platform that
138 uses Configure (ie most things except VMS and Windows) headers
139 are identical for the different IV sizes, despite the files
140 containing some fields based on sizeof(IV)
142 5.8.0 is consistent; all platforms have BYTEORDER in config.h
143 based on sizeof(IV) rather than sizeof(long). This means that
144 the value of BYTEORDER will change from (say) 4321 to 87654321
145 between 5.6.1 and 5.8.0 built with the same options to Configure
146 on the same machine. This means that the Storable header will
147 differ, and the two versions will wrongly thing that they are
150 For the benefit of long term consistency, Storable now
151 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
152 2.03 onwards default to be incompatible with 2.02 and earlier
153 (ie the large 1.0.x installed base) on the same 5.6.x perl.
155 To allow interworking, a new variable
156 $Storable::interwork_56_64bit is introduced. It defaults to
157 false. Set it to true to read and write old format files. Don't
158 use it unless you have existing stored data written with 5.6.x
159 that you couldn't otherwise read, or you need to interwork with
160 a machine running older Storable on a 5.6.x with long long IVs
161 (i.e., you probably don't need to use it).
163 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
167 Rewrite Storable.xs so that the file header structure for write_magic
168 is built at compile time, and check_magic attempts to the header in
169 blocks rather than byte per byte. These changes make the compiled
170 extension 2.25% smaller, but are not significant enough to give a
173 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
177 - New regression tests integer.t
178 - Add code to safely store large unsigned integers.
179 - Change code not to attempt to store large integers (ie > 32 bits)
180 in network order as 32 bits.
182 *Never* underestimate the value of a pathological test suite carefully
183 crafted with maximum malice before writing a line of real code. It
184 prevents crafty bugs from stowing away in your released code.
185 It's much less embarrassing to find them before you ship.
186 (Well, never underestimate it if you ever want to work for me)
188 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
190 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
192 The perl5 porters have decided to make sure that Storable still
193 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
194 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
195 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
196 due to a bug in 5.6.0 - upgrade to 5.6.1.
198 Jarkko and I have collated the list of changes the perl5 porters have
199 from the perl5 Changes file:
201 - data features of upcoming perl 5.8.0 are supported: Unicode hash
202 keys (Unicode hash values have been supported since Storable 1.0.1)
203 and "restricted hashes" (readonly hashes and hash entries)
204 - a newer version of perl can now be used to serialize data which is
205 not supported in earlier perls: Storable will attempt to do the
206 right thing for as long as possible, croaking only when safe data
207 conversion simply isn't possible. Alternatively earlier perls can
208 opt to have a lossy downgrade data instead of croaking
209 - when built with perls pre 5.7.3 this Storable writes out files
210 with binary format 2.4, the same format as Storable 1.0.8 onwards.
211 This should mean that this Storable will inter-operate seamlessly
212 with any Storable 1.0.8 or newer on perls pre 5.7.3
213 - dclone() now works with empty string scalar objects
214 - retrieving of large hashes is now more efficient
215 - more routines autosplit out of the main module, so Storable should
216 load slightly more quickly
217 - better documentation
218 - the internal context objects are now freed explicitly, rather than
219 relying on thread or process exit
220 - bugs fixed in debugging trace code affecting builds made with 64 bit
222 - code tidy-ups to allow clean compiles with more warning options
223 turned on avoid problems with $@ getting corrupted on 5.005_03 if
224 Carp wasn't already loaded
225 - added &show_file_magic, so you can add to /etc/magic and teach
226 Unix's file command about Storable files
228 We plan to keep Storable on CPAN in sync with the Perl core, so
229 if you encounter bugs or other problems building or using Storable,
230 please let us know at perl5-porters@perl.org
233 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
235 This is the LAST maintenance release of the Storable module.
236 Indeed, Storable is now part of perl 5.8, and will be maintained
237 as part of Perl. The CPAN module will remain available there
238 for people running pre-5.8 perls.
240 Avoid requiring Fcntl upfront, useful to embedded runtimes.
241 Use an eval {} for testing, instead of making Storable.pm
242 simply fail its compilation in the BEGIN block.
244 store_fd() will now correctly autoflush file if needed.
246 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
248 Fixed truncation race with lock_retrieve() in lock_store().
249 The file has to be truncated only once the exclusive lock is held.
251 Removed spurious debugging messages in .xs file.
253 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
255 Systematically use "=over 4" for POD linters.
256 Apparently, POD linters are much stricter than would
257 otherwise be needed, but that's OK.
259 Fixed memory corruption on croaks during thaw(). Thanks
260 to Claudio Garcia for reproducing this bug and providing the
261 code to exercise it. Added test cases for this bug, adapted
264 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
266 Changed tagnum and classnum from I32 to IV in context. Also
269 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
271 Last version was wrongly compiling with assertions on, due
272 to an edit glitch. That did not cause any problem (apart from
273 a slight performance loss) excepted on Win* platforms, where the
274 assertion code does not compile.
276 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
280 Forgot to increase version number at previous patch (there were
281 two of them, which is why we jump from 1.0.8 to 1.0.10).
283 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
285 Version 1.0.8, binary format 2.4.
287 Fixed incorrect error message.
289 Now bless objects ASAP at retrieve time, which is meant to fix
292 * Indirect references to overloaded object were not able to
293 restore overloading if the object was not blessed yet,
294 which was possible since blessing occurred only after the
297 * Storable hooks asking for serialization of blessed ref could
298 get un-blessed refs at retrieval time, for the very same
301 The fix implemented here was suggested by Nick Ing-Simmons.
303 Added support for blessed ref to tied structures. This is the
304 cause for the binary format change.
306 Added EBCDIC version of the compatibility test with 0.6.11,
309 Added tests for the new features, and to make sure the bugs they
310 are meant to fix are indeed fixed.
312 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
314 Removed spurious 'clean' entry in Makefile.PL.
316 Added CAN_FLOCK to determine whether we can flock() or not,
317 by inspecting Perl's configuration parameters, as determined
320 Trace offending package when overloading cannot be restored
323 Made context cleanup safer to avoid dup freeing, mostly in the
324 presence of repeated exceptions during store/retrieve (which can
325 cause memory leaks anyway, so it's just additional safety, not a
328 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
332 Fixed severe "object lost" bug for STORABLE_freeze returns,
333 when refs to lexicals, taken within the hook, were to be
334 serialized by Storable. Enhanced the t/recurse.t test to
335 stress hook a little more with refs to lexicals.
337 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
341 Documented that store() and retrieve() can return undef.
342 That is, the error reporting is not always made via exceptions,
343 as the paragraph on error reporting was implying.
345 Auto requires module of blessed ref when STORABLE_thaw misses.
346 When the Storable engine looks for the STORABLE_thaw hook and
347 does not find it, it now tries to require the package into which
348 the blessed reference is.
350 Just check $^O, in t/lock.t: there's no need to pull the whole
351 Config module for that.
353 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
357 Protected calls to flock() for DOS platform: apparently, the
358 flock/fcnlt emulation is reported to be broken on that
361 Added logcarp emulation if they don't have Log::Agent, since
362 we now use it to carp when lock_store/lock_retrieve is used
365 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
369 Avoid using "tainted" and "dirty" since Perl remaps them via
370 cpp (i.e. #define). This is deeply harmful when threading
371 is enabled. This concerned both the context structure and
372 local variable and argument names. Brrr..., scary!
374 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
378 Fixed spelling in README.
380 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
381 after a proposal from Erik Haugan <erik@solbors.no>.
383 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
385 Fixed stupid typo in the t/utf8.t test.
387 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
389 Version 1.0.1, binary format 2.3.
391 Documented that doubles are stored stringified by nstore().
393 Added Salvador Ortiz Garcia in CREDITS section, He identified
394 a bug in the store hooks and proposed the right fix: the class
395 id was allocated too soon. His bug case was also added to
396 the regression test suite.
398 Now only taint retrieved data when source was tainted. A bug
399 discovered by Marc Lehmann.
401 Added support for UTF-8 strings, a contribution of Marc Lehmann.
402 This is normally only activated in post-5.6 perls.
404 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
406 First official release Storable 1.0, for inclusion in perl 5.7.0.
407 The license scheme is now compatible with Perl's.
409 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
411 ANSI-fied most of the code, preparing for Perl core integration.
412 The next version of Storable will be 0.8, and will be integrated
413 into the Perl core (development branch).
415 Dispatch tables were moved upfront to relieve some compilers,
416 especially on AIX and Windows platforms.
418 Merged 64-bit fixes from perl5-porters.
420 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
422 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
423 the refcnt on the mg_ptr as well.
425 Removed spurious dependency to Devel::Peek, which was used for
426 testing only in t/tied_items.t. Thanks to Conrad Heiney
427 <conrad@fringehead.org> for spotting it first.
429 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
431 Marc Lehmann kindly contributed code to add overloading support
432 and to handle references to tied variables.
434 Rewrote leading blurb about compatibility to make it clearer what
435 "backward compatibility" is about: when I say 0.7 is backward
436 compatible with 0.6, it means the revision 0.7 can read files
439 Mention new Clone(3) extension in SEE ALSO.
441 Was wrongly optimizing for "undef" values in hashes by not
442 fully recursing: as a result, tied "undef" values were incorrectly
445 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
447 First revision of Storable 0.7.
449 The serializing format is new, known as version 2.0. It is fully
450 backward compatible with 0.6. Earlier formats are deprecated and
451 have not even been tested: next version will drop pre-0.6 format.
453 Changes since 0.6@11:
455 - Moved interface to the "beta" status. Some tiny parts are still
456 subject to change, but nothing important enough to warrant an "alpha"
459 - Slightly reduced the size of the Storable image by factorizing
460 object class names and removing final object storage notification due
461 to a redesign of the blessed object storing.
463 - Classes can now redefine how they wish their instances to be serialized
464 and/or deep cloned. Serializing hooks are written in Perl code.
466 - The engine is now fully re-entrant.
468 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
470 Added provision to detect more recent binary formats, since
471 the new upcoming Storable-0.7 will use a different format.
472 In order to prevent attempting the de-serialization of newer
473 formats by older versions, I'm adding this now to the 0.6 series.
475 I'm expecting this revision to be the last of the 0.6 series.
476 Unless it does not work with perl 5.6, which I don't use yet,
477 and therefore against which I cannot test.
479 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
481 Added note about format incompatibilities with old versions
482 (i.e. pre 0.5@9 formats, which cannot be understood as there
483 was no versionning information in the file by then).
485 Protect all $@ variables when eval {} used, to avoid corrupting
486 it when store/retrieve is called within an exception handler.
488 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
489 preventing Perl's patchlevel from being included, which is
490 needed starting from 5.6.
492 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
494 Fixed shared "undef" bug in hashes, which did not remain shared
495 through store/retrieve.
497 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
499 added last_op_in_netorder() predicate
500 documented last_op_in_netorder()
501 added tests for the new last_op_in_netorder() predicate
503 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
505 Forgot to update VERSION
507 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
509 Added mention of japanese translation for the manual page.
511 Fixed typo in macro that made threaded code not compilable,
512 especially on Win32 platforms.
514 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
515 directly instead of relying on internal symbols.
517 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
519 Integrated "thread-safe" patch from Murray Nesbitt.
520 Note that this may not be very efficient for threaded code,
521 see comment in the code.
523 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
524 since I don't have access to such machines.
526 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
528 changed my e-mail to pobox.
530 mentionned it is not thread-safe.
532 updated version number.
534 uses new internal PL_* naming convention.
536 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
538 Updated benchmark figures due to recent optimizations done in
539 store(): tagnums are now stored as-is in the hash table, so
540 no surrounding SV is created. And the "shared keys" mode for
541 hash table was turned off.
543 Fixed backward compatibility (wrt 0.5@9) for retrieval of
544 blessed refs. That old version did something wrong, but the
545 bugfix prevented correct retrieval of the old format.
547 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
549 Changed benchmark figures.
551 Adjust refcnt of tied objects after calling sv_magic() to avoid
552 memory leaks. Contributed by Jeff Gresham.
554 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
556 Added workaround for persistent LVALUE-ness in perl5.004. All
557 scalars tagged as being an lvalue are handled as if they were
558 not an lvalue at all. Added test for that LVALUE bug workaround.
560 Now handles Perl immortal scalars explicitely, by storing &sv_yes
561 as such, explicitely.
563 Retrieval of non-immortal undef cannot be shared. Previous
564 version was over-optimizing by not creating a separate SV for
565 all undefined scalars seen.
567 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
569 Baseline for Storable-0.6@0.
571 This version introduces a binary incompatibility in the generated
572 binary image, which is more compact than older ones by approximatively
573 15%, depending on the exact degree of sharing in your structures.
575 The good news is that your older images can still be retrieved with
576 this version, i.e. backward compatibility is preserved. This version
577 of Storable can only generate new binaries however.
579 Another good news is that the retrieval of data structure is
580 significantly quicker than before, because a Perl array is used
581 instead of a hash table to keep track of retrieved objects, and
582 also because the image being smaller, less I/O function calls are
585 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
587 Version number now got from Storable.pm directly.
589 Fixed overzealous sv_type() optimization, which would make
590 Storable fail when faced with an "upgraded" SV to the PVIV
591 or PVNV kind containing a reference.
593 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
595 Extended the SYNOPSIS section to give quick overview of the
596 routines and their signature.
598 Optimized sv_type() to avoid flags checking when not needed, i.e.
599 when their type makes it impossible for them to be refs or tied.
600 This slightly increases throughput by a few percents when refs
601 and tied variables are marginal occurrences in your data.
603 Stubs for XS now use OutputStream and InputStream file types to
604 make it work when the given file is actually a socket. Perl
605 makes a distinction for sockets in its internal I/O structures
606 by having both a read and a write structure, whereas plain files
609 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
611 Thanks to a contribution from Benjamin A. Holzman, Storable is now
612 able to correctly serialize tied SVs, i.e. tied arrays, hashes
615 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
617 I said SvPOK() had changed to SvPOKp(), but that was a lie...
619 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
621 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
622 in effect mangled the object tags and prevented portability
623 accross 32/64 bit architectures!
625 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
627 Added code example for store_fd() and retrieve_fd() in the
628 man page, to emphasize that file descriptors must be passed as
629 globs, not as plain strings.
631 Cannot use SV addresses as tag when using nstore() on LP64. This
632 was the cause of problems when creating a storable image on an
633 LP64 machine and retrieving it on an ILP32 system, which is
634 exactly what nstore() is meant for...
636 However, we continue to use SV addresses as tags for plain store(),
637 because benchamarking shows that it saves up to 8% of the store
638 time, and store() is meant to be fast at the expense of lack
641 This means there will be approximately an 8% degradation of
642 performance for nstore(), but it's now working as expected.
643 That cost may vary on your machine of course, since it is
644 solely caused by the memory allocation overhead used to create
645 unique SV tags for each distinct stored SV.
647 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
649 Don't use any '_' in version number.
651 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
653 Updated version number.
655 added binmode() calls for systems where it matters.
657 Be sure to pass globs, not plain file strings, to C routines,
658 so that Storable can be used under the Perl debugger.
660 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
662 Fix memory leaks on seen hash table and returned SV refs.
664 Storable did not work properly when tainting enabled.
666 Fixed "Allocation too large" messages in freeze/thaw and added.
667 proper regression test in t/freeze.t.
669 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
671 Updated version number
673 Added freeze/thaw interface and dclone.
675 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
677 Forgot that AutoLoader does not export its own AUTOLOAD.
680 use AutoLoader 'AUTOLOAD';
682 but that would not be backward compatible. So the export is
685 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
687 Empty scalar strings are now "defined" at retrieval time.
689 New test to ensure an empty string is defined when retrieved.
691 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
693 Updated version number
695 Declare VERSION as being used
697 Fixed a typo in the PerlIO_putc remapping.
698 PerlIO_read and perlIO_write inverted size/nb_items.
699 (only relevant for pre-perl5.004 versions)
701 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
703 Updated version number
705 Added VERSION identification
707 Allow build with perl5.003, which is ante perlIO time
709 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
713 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
715 Updated version number in Makefile.PL.
717 Added "thanks to" section to README.
719 Documented new forgive_me variable.
723 Added forgive_me support to allow store() of data structures
724 containing non-storable items like CODE refs.