1 Sat Jul 10 22:37:47 BST 2004 Nicholas Clark <nick@ccl4.org>
5 1. Store weak references
7 Thu Jun 17 12:26:43 BST 2004 Nicholas Clark <nick@ccl4.org>
11 1. Don't change the type of top level overloaded references to RV -
12 they are perfectly correct as PVMG
13 2. Storable needs to cope with incoming frozen data that happens to be
16 Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org>
20 1. Add regression tests for the auto-require of STORABLE_thaw
21 2. Add auto-require of modules to restore overloading (and tests)
22 3. Change to no context (should give speedup with ithreads)
24 Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org>
28 1. Storing restricted hashes in canonical order would SEGV. Fixed.
29 2. It was impossible to retrieve references to PL_sv_no and and
30 PL_sv_undef from STORABLE_thaw hooks.
31 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique
32 implementation of restricted hashes using PL_sv_undef
33 4. These changes allow a space optimisation for restricted hashes.
35 Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org>
39 1. Thread safety: Storable::CLONE/init_perlinterp() now create
40 a new Perl context for each new ithread.
41 (From Stas Bekman and Jan Dubois.)
42 2. Fix a tag count mismatch with $Storable::Deparse that caused
43 all back-references after a stored sub to be off-by-N (where
44 N was the number of code references in between).
46 3. Prevent CODE references from turning into SCALAR references.
49 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
53 Fix minor problems with the CPAN release
54 1: Make Storable.xs work on 5.8.2 and later (already in the core)
55 2: Ship the linux hints file
56 3: Ship Test::More for the benefit of Perls pre 5.6.2
57 4: Correct Makefile.PL to only install in core for 5.8.0 and later
59 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
63 This release works around a 5.8.0 bug which caused hashes to not
64 be marked as having key flags even though an HEK had HEK_WASUTF8
65 set. (Note that the only reasonable solution is to silently drop
66 the flag from the affected key.)
68 Users of RT 3 who were seeing assertion failures should upgrade.
69 (Perl 5.8.1 will have the bug fixed.)
71 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
75 Minor bugfixes (self-tied objects are now correctly stored, as
76 are the results of additions larger than INT_MAX).
78 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
82 Remove qr// from t/downgrade.t so that it will run on 5.004
83 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
84 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
86 Fix t/integer.t initialisation to actually generate 64 bits of 9c
87 Fix comparison tests to use eval to get around 64 bit IV conversion
88 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
90 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
91 subtests that use $Config{ptrsize}, so that the rest of the test can
93 Change t/malice.t and the error message in check_magic in Storable.xs
94 from "Pointer integer size" to "Pointer size"
95 Remove prerequisite of Test::More from Makefile.PL
96 Ship Test::Builder, Test::Simple and Test::More in t
98 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
102 Adds support for CODE references from Slaven Rezic
103 <slaven.rezic@berlin.de>.
105 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
109 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
110 to fix a recently introduced bug detected by Dave Rolsky.
111 Bug was that for a non threaded build, the class information was
112 being lost at freeze time on the first object with a STORABLE_freeze
113 hook. Consequentially the object was not blessed at all when thawed.
114 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
115 a store-time data lost bug, caused by failure to initialize internal
117 The bug was introduced as development perl change 16442 (on
118 2002/05/07), so has been present since 2.00.
119 Patches to introduce more regression tests to reduce the chance of
120 a reoccurance of this sort of goof are always welcome.
122 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
124 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
126 5.6.x introduced the ability to have IVs as long long. However,
127 Configure still defined BYTEORDER based on the size of a long.
128 Storable uses the BYTEORDER value as part of the header, but
129 doesn't explicity store sizeof(IV) anywhere in the header.
130 Hence on 5.6.x built with IV as long long on a platform that
131 uses Configure (ie most things except VMS and Windows) headers
132 are identical for the different IV sizes, despite the files
133 containing some fields based on sizeof(IV)
135 5.8.0 is consistent; all platforms have BYTEORDER in config.h
136 based on sizeof(IV) rather than sizeof(long). This means that
137 the value of BYTEORDER will change from (say) 4321 to 87654321
138 between 5.6.1 and 5.8.0 built with the same options to Configure
139 on the same machine. This means that the Storable header will
140 differ, and the two versions will wrongly thing that they are
143 For the benefit of long term consistency, Storable now
144 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
145 2.03 onwards default to be incompatible with 2.02 and earlier
146 (ie the large 1.0.x installed base) on the same 5.6.x perl.
148 To allow interworking, a new variable
149 $Storable::interwork_56_64bit is introduced. It defaults to
150 false. Set it to true to read and write old format files. Don't
151 use it unless you have existing stored data written with 5.6.x
152 that you couldn't otherwise read, or you need to interwork with
153 a machine running older Storable on a 5.6.x with long long IVs
154 (i.e., you probably don't need to use it).
156 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
160 Rewrite Storable.xs so that the file header structure for write_magic
161 is built at compile time, and check_magic attempts to the header in
162 blocks rather than byte per byte. These changes make the compiled
163 extension 2.25% smaller, but are not significant enough to give a
166 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
170 - New regression tests integer.t
171 - Add code to safely store large unsigned integers.
172 - Change code not to attempt to store large integers (ie > 32 bits)
173 in network order as 32 bits.
175 *Never* underestimate the value of a pathological test suite carefully
176 crafted with maximum malice before writing a line of real code. It
177 prevents crafty bugs from stowing away in your released code.
178 It's much less embarrassing to find them before you ship.
179 (Well, never underestimate it if you ever want to work for me)
181 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
183 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
185 The perl5 porters have decided to make sure that Storable still
186 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
187 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
188 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
189 due to a bug in 5.6.0 - upgrade to 5.6.1.
191 Jarkko and I have collated the list of changes the perl5 porters have
192 from the perl5 Changes file:
194 - data features of upcoming perl 5.8.0 are supported: Unicode hash
195 keys (Unicode hash values have been supported since Storable 1.0.1)
196 and "restricted hashes" (readonly hashes and hash entries)
197 - a newer version of perl can now be used to serialize data which is
198 not supported in earlier perls: Storable will attempt to do the
199 right thing for as long as possible, croaking only when safe data
200 conversion simply isn't possible. Alternatively earlier perls can
201 opt to have a lossy downgrade data instead of croaking
202 - when built with perls pre 5.7.3 this Storable writes out files
203 with binary format 2.4, the same format as Storable 1.0.8 onwards.
204 This should mean that this Storable will inter-operate seamlessly
205 with any Storable 1.0.8 or newer on perls pre 5.7.3
206 - dclone() now works with empty string scalar objects
207 - retrieving of large hashes is now more efficient
208 - more routines autosplit out of the main module, so Storable should
209 load slightly more quickly
210 - better documentation
211 - the internal context objects are now freed explicitly, rather than
212 relying on thread or process exit
213 - bugs fixed in debugging trace code affecting builds made with 64 bit
215 - code tidy-ups to allow clean compiles with more warning options
216 turned on avoid problems with $@ getting corrupted on 5.005_03 if
217 Carp wasn't already loaded
218 - added &show_file_magic, so you can add to /etc/magic and teach
219 Unix's file command about Storable files
221 We plan to keep Storable on CPAN in sync with the Perl core, so
222 if you encounter bugs or other problems building or using Storable,
223 please let us know at perl5-porters@perl.org
226 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
228 This is the LAST maintenance release of the Storable module.
229 Indeed, Storable is now part of perl 5.8, and will be maintained
230 as part of Perl. The CPAN module will remain available there
231 for people running pre-5.8 perls.
233 Avoid requiring Fcntl upfront, useful to embedded runtimes.
234 Use an eval {} for testing, instead of making Storable.pm
235 simply fail its compilation in the BEGIN block.
237 store_fd() will now correctly autoflush file if needed.
239 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
241 Fixed truncation race with lock_retrieve() in lock_store().
242 The file has to be truncated only once the exclusive lock is held.
244 Removed spurious debugging messages in .xs file.
246 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
248 Systematically use "=over 4" for POD linters.
249 Apparently, POD linters are much stricter than would
250 otherwise be needed, but that's OK.
252 Fixed memory corruption on croaks during thaw(). Thanks
253 to Claudio Garcia for reproducing this bug and providing the
254 code to exercise it. Added test cases for this bug, adapted
257 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
259 Changed tagnum and classnum from I32 to IV in context. Also
262 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
264 Last version was wrongly compiling with assertions on, due
265 to an edit glitch. That did not cause any problem (apart from
266 a slight performance loss) excepted on Win* platforms, where the
267 assertion code does not compile.
269 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
273 Forgot to increase version number at previous patch (there were
274 two of them, which is why we jump from 1.0.8 to 1.0.10).
276 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
278 Version 1.0.8, binary format 2.4.
280 Fixed incorrect error message.
282 Now bless objects ASAP at retrieve time, which is meant to fix
285 * Indirect references to overloaded object were not able to
286 restore overloading if the object was not blessed yet,
287 which was possible since blessing occurred only after the
290 * Storable hooks asking for serialization of blessed ref could
291 get un-blessed refs at retrieval time, for the very same
294 The fix implemented here was suggested by Nick Ing-Simmons.
296 Added support for blessed ref to tied structures. This is the
297 cause for the binary format change.
299 Added EBCDIC version of the compatibility test with 0.6.11,
302 Added tests for the new features, and to make sure the bugs they
303 are meant to fix are indeed fixed.
305 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
307 Removed spurious 'clean' entry in Makefile.PL.
309 Added CAN_FLOCK to determine whether we can flock() or not,
310 by inspecting Perl's configuration parameters, as determined
313 Trace offending package when overloading cannot be restored
316 Made context cleanup safer to avoid dup freeing, mostly in the
317 presence of repeated exceptions during store/retrieve (which can
318 cause memory leaks anyway, so it's just additional safety, not a
321 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
325 Fixed severe "object lost" bug for STORABLE_freeze returns,
326 when refs to lexicals, taken within the hook, were to be
327 serialized by Storable. Enhanced the t/recurse.t test to
328 stress hook a little more with refs to lexicals.
330 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
334 Documented that store() and retrieve() can return undef.
335 That is, the error reporting is not always made via exceptions,
336 as the paragraph on error reporting was implying.
338 Auto requires module of blessed ref when STORABLE_thaw misses.
339 When the Storable engine looks for the STORABLE_thaw hook and
340 does not find it, it now tries to require the package into which
341 the blessed reference is.
343 Just check $^O, in t/lock.t: there's no need to pull the whole
344 Config module for that.
346 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
350 Protected calls to flock() for DOS platform: apparently, the
351 flock/fcnlt emulation is reported to be broken on that
354 Added logcarp emulation if they don't have Log::Agent, since
355 we now use it to carp when lock_store/lock_retrieve is used
358 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
362 Avoid using "tainted" and "dirty" since Perl remaps them via
363 cpp (i.e. #define). This is deeply harmful when threading
364 is enabled. This concerned both the context structure and
365 local variable and argument names. Brrr..., scary!
367 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
371 Fixed spelling in README.
373 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
374 after a proposal from Erik Haugan <erik@solbors.no>.
376 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
378 Fixed stupid typo in the t/utf8.t test.
380 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
382 Version 1.0.1, binary format 2.3.
384 Documented that doubles are stored stringified by nstore().
386 Added Salvador Ortiz Garcia in CREDITS section, He identified
387 a bug in the store hooks and proposed the right fix: the class
388 id was allocated too soon. His bug case was also added to
389 the regression test suite.
391 Now only taint retrieved data when source was tainted. A bug
392 discovered by Marc Lehmann.
394 Added support for UTF-8 strings, a contribution of Marc Lehmann.
395 This is normally only activated in post-5.6 perls.
397 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
399 First official release Storable 1.0, for inclusion in perl 5.7.0.
400 The license scheme is now compatible with Perl's.
402 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
404 ANSI-fied most of the code, preparing for Perl core integration.
405 The next version of Storable will be 0.8, and will be integrated
406 into the Perl core (development branch).
408 Dispatch tables were moved upfront to relieve some compilers,
409 especially on AIX and Windows platforms.
411 Merged 64-bit fixes from perl5-porters.
413 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
415 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
416 the refcnt on the mg_ptr as well.
418 Removed spurious dependency to Devel::Peek, which was used for
419 testing only in t/tied_items.t. Thanks to Conrad Heiney
420 <conrad@fringehead.org> for spotting it first.
422 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
424 Marc Lehmann kindly contributed code to add overloading support
425 and to handle references to tied variables.
427 Rewrote leading blurb about compatibility to make it clearer what
428 "backward compatibility" is about: when I say 0.7 is backward
429 compatible with 0.6, it means the revision 0.7 can read files
432 Mention new Clone(3) extension in SEE ALSO.
434 Was wrongly optimizing for "undef" values in hashes by not
435 fully recursing: as a result, tied "undef" values were incorrectly
438 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
440 First revision of Storable 0.7.
442 The serializing format is new, known as version 2.0. It is fully
443 backward compatible with 0.6. Earlier formats are deprecated and
444 have not even been tested: next version will drop pre-0.6 format.
446 Changes since 0.6@11:
448 - Moved interface to the "beta" status. Some tiny parts are still
449 subject to change, but nothing important enough to warrant an "alpha"
452 - Slightly reduced the size of the Storable image by factorizing
453 object class names and removing final object storage notification due
454 to a redesign of the blessed object storing.
456 - Classes can now redefine how they wish their instances to be serialized
457 and/or deep cloned. Serializing hooks are written in Perl code.
459 - The engine is now fully re-entrant.
461 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
463 Added provision to detect more recent binary formats, since
464 the new upcoming Storable-0.7 will use a different format.
465 In order to prevent attempting the de-serialization of newer
466 formats by older versions, I'm adding this now to the 0.6 series.
468 I'm expecting this revision to be the last of the 0.6 series.
469 Unless it does not work with perl 5.6, which I don't use yet,
470 and therefore against which I cannot test.
472 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
474 Added note about format incompatibilities with old versions
475 (i.e. pre 0.5@9 formats, which cannot be understood as there
476 was no versionning information in the file by then).
478 Protect all $@ variables when eval {} used, to avoid corrupting
479 it when store/retrieve is called within an exception handler.
481 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
482 preventing Perl's patchlevel from being included, which is
483 needed starting from 5.6.
485 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
487 Fixed shared "undef" bug in hashes, which did not remain shared
488 through store/retrieve.
490 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
492 added last_op_in_netorder() predicate
493 documented last_op_in_netorder()
494 added tests for the new last_op_in_netorder() predicate
496 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
498 Forgot to update VERSION
500 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
502 Added mention of japanese translation for the manual page.
504 Fixed typo in macro that made threaded code not compilable,
505 especially on Win32 platforms.
507 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
508 directly instead of relying on internal symbols.
510 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
512 Integrated "thread-safe" patch from Murray Nesbitt.
513 Note that this may not be very efficient for threaded code,
514 see comment in the code.
516 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
517 since I don't have access to such machines.
519 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
521 changed my e-mail to pobox.
523 mentionned it is not thread-safe.
525 updated version number.
527 uses new internal PL_* naming convention.
529 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
531 Updated benchmark figures due to recent optimizations done in
532 store(): tagnums are now stored as-is in the hash table, so
533 no surrounding SV is created. And the "shared keys" mode for
534 hash table was turned off.
536 Fixed backward compatibility (wrt 0.5@9) for retrieval of
537 blessed refs. That old version did something wrong, but the
538 bugfix prevented correct retrieval of the old format.
540 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
542 Changed benchmark figures.
544 Adjust refcnt of tied objects after calling sv_magic() to avoid
545 memory leaks. Contributed by Jeff Gresham.
547 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
549 Added workaround for persistent LVALUE-ness in perl5.004. All
550 scalars tagged as being an lvalue are handled as if they were
551 not an lvalue at all. Added test for that LVALUE bug workaround.
553 Now handles Perl immortal scalars explicitely, by storing &sv_yes
554 as such, explicitely.
556 Retrieval of non-immortal undef cannot be shared. Previous
557 version was over-optimizing by not creating a separate SV for
558 all undefined scalars seen.
560 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
562 Baseline for Storable-0.6@0.
564 This version introduces a binary incompatibility in the generated
565 binary image, which is more compact than older ones by approximatively
566 15%, depending on the exact degree of sharing in your structures.
568 The good news is that your older images can still be retrieved with
569 this version, i.e. backward compatibility is preserved. This version
570 of Storable can only generate new binaries however.
572 Another good news is that the retrieval of data structure is
573 significantly quicker than before, because a Perl array is used
574 instead of a hash table to keep track of retrieved objects, and
575 also because the image being smaller, less I/O function calls are
578 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
580 Version number now got from Storable.pm directly.
582 Fixed overzealous sv_type() optimization, which would make
583 Storable fail when faced with an "upgraded" SV to the PVIV
584 or PVNV kind containing a reference.
586 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
588 Extended the SYNOPSIS section to give quick overview of the
589 routines and their signature.
591 Optimized sv_type() to avoid flags checking when not needed, i.e.
592 when their type makes it impossible for them to be refs or tied.
593 This slightly increases throughput by a few percents when refs
594 and tied variables are marginal occurrences in your data.
596 Stubs for XS now use OutputStream and InputStream file types to
597 make it work when the given file is actually a socket. Perl
598 makes a distinction for sockets in its internal I/O structures
599 by having both a read and a write structure, whereas plain files
602 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
604 Thanks to a contribution from Benjamin A. Holzman, Storable is now
605 able to correctly serialize tied SVs, i.e. tied arrays, hashes
608 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
610 I said SvPOK() had changed to SvPOKp(), but that was a lie...
612 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
614 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
615 in effect mangled the object tags and prevented portability
616 accross 32/64 bit architectures!
618 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
620 Added code example for store_fd() and retrieve_fd() in the
621 man page, to emphasize that file descriptors must be passed as
622 globs, not as plain strings.
624 Cannot use SV addresses as tag when using nstore() on LP64. This
625 was the cause of problems when creating a storable image on an
626 LP64 machine and retrieving it on an ILP32 system, which is
627 exactly what nstore() is meant for...
629 However, we continue to use SV addresses as tags for plain store(),
630 because benchamarking shows that it saves up to 8% of the store
631 time, and store() is meant to be fast at the expense of lack
634 This means there will be approximately an 8% degradation of
635 performance for nstore(), but it's now working as expected.
636 That cost may vary on your machine of course, since it is
637 solely caused by the memory allocation overhead used to create
638 unique SV tags for each distinct stored SV.
640 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
642 Don't use any '_' in version number.
644 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
646 Updated version number.
648 added binmode() calls for systems where it matters.
650 Be sure to pass globs, not plain file strings, to C routines,
651 so that Storable can be used under the Perl debugger.
653 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
655 Fix memory leaks on seen hash table and returned SV refs.
657 Storable did not work properly when tainting enabled.
659 Fixed "Allocation too large" messages in freeze/thaw and added.
660 proper regression test in t/freeze.t.
662 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
664 Updated version number
666 Added freeze/thaw interface and dclone.
668 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
670 Forgot that AutoLoader does not export its own AUTOLOAD.
673 use AutoLoader 'AUTOLOAD';
675 but that would not be backward compatible. So the export is
678 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
680 Empty scalar strings are now "defined" at retrieval time.
682 New test to ensure an empty string is defined when retrieved.
684 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
686 Updated version number
688 Declare VERSION as being used
690 Fixed a typo in the PerlIO_putc remapping.
691 PerlIO_read and perlIO_write inverted size/nb_items.
692 (only relevant for pre-perl5.004 versions)
694 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
696 Updated version number
698 Added VERSION identification
700 Allow build with perl5.003, which is ante perlIO time
702 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
706 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
708 Updated version number in Makefile.PL.
710 Added "thanks to" section to README.
712 Documented new forgive_me variable.
716 Added forgive_me support to allow store() of data structures
717 containing non-storable items like CODE refs.