1 Mon May 18 09:38:20 IST 2009 Abhijit Menon-Sen <ams@toroid.org>
5 Fix bug handling blessed references to overloaded objects, plus
6 other miscellaneous fixes.
8 (Version 2.19 was released with 5.8.9.)
10 Thu Nov 22 13:24:18 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
14 Compile fixes for older Perls. (No functional changes.)
16 Sat Nov 17 02:12:12 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
20 Various broken tests fixed. (No functional changes.)
22 Sat Mar 31 06:11:06 IST 2007 Abhijit Menon-Sen <ams@toroid.org>
26 1. Fixes to Storable::dclone, read_magic, retrieve_lscalar
27 2. Storable 0.1 compatibility
28 3. Miscellaneous compile/leak/test/portability fixes
30 Mon May 23 22:48:49 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
34 Minor changes to address a couple of compile problems.
36 Mon Apr 25 07:29:14 IST 2005 Abhijit Menon-Sen <ams@wiw.org>
40 1. Store weak references
41 2. Add STORABLE_attach hook.
43 Thu Jun 17 12:26:43 BST 2004 Nicholas Clark <nick@ccl4.org>
47 1. Don't change the type of top level overloaded references to RV -
48 they are perfectly correct as PVMG
49 2. Storable needs to cope with incoming frozen data that happens to be
52 Wed Mar 17 15:40:29 GMT 2004 Nicholas Clark <nick@ccl4.org>
56 1. Add regression tests for the auto-require of STORABLE_thaw
57 2. Add auto-require of modules to restore overloading (and tests)
58 3. Change to no context (should give speedup with ithreads)
60 Sat Mar 13 20:11:03 GMT 2004 Nicholas Clark <nick@ccl4.org>
64 1. Storing restricted hashes in canonical order would SEGV. Fixed.
65 2. It was impossible to retrieve references to PL_sv_no and and
66 PL_sv_undef from STORABLE_thaw hooks.
67 3. restrict.t was failing on 5.8.0, due to 5.8.0's unique
68 implementation of restricted hashes using PL_sv_undef
69 4. These changes allow a space optimisation for restricted hashes.
71 Sat Jan 24 16:22:32 IST 2004 Abhijit Menon-Sen <ams@wiw.org>
75 1. Thread safety: Storable::CLONE/init_perlinterp() now create
76 a new Perl context for each new ithread.
77 (From Stas Bekman and Jan Dubois.)
78 2. Fix a tag count mismatch with $Storable::Deparse that caused
79 all back-references after a stored sub to be off-by-N (where
80 N was the number of code references in between).
82 3. Prevent CODE references from turning into SCALAR references.
85 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
89 Fix minor problems with the CPAN release
90 1: Make Storable.xs work on 5.8.2 and later (already in the core)
91 2: Ship the linux hints file
92 3: Ship Test::More for the benefit of Perls pre 5.6.2
93 4: Correct Makefile.PL to only install in core for 5.8.0 and later
95 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
99 This release works around a 5.8.0 bug which caused hashes to not
100 be marked as having key flags even though an HEK had HEK_WASUTF8
101 set. (Note that the only reasonable solution is to silently drop
102 the flag from the affected key.)
104 Users of RT 3 who were seeing assertion failures should upgrade.
105 (Perl 5.8.1 will have the bug fixed.)
107 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
111 Minor bugfixes (self-tied objects are now correctly stored, as
112 are the results of additions larger than INT_MAX).
114 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
118 Remove qr// from t/downgrade.t so that it will run on 5.004
119 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
120 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
122 Fix t/integer.t initialisation to actually generate 64 bits of 9c
123 Fix comparison tests to use eval to get around 64 bit IV conversion
124 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
126 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
127 subtests that use $Config{ptrsize}, so that the rest of the test can
128 now be run with 5.004
129 Change t/malice.t and the error message in check_magic in Storable.xs
130 from "Pointer integer size" to "Pointer size"
131 Remove prerequisite of Test::More from Makefile.PL
132 Ship Test::Builder, Test::Simple and Test::More in t
134 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
138 Adds support for CODE references from Slaven Rezic
139 <slaven.rezic@berlin.de>.
141 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
145 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
146 to fix a recently introduced bug detected by Dave Rolsky.
147 Bug was that for a non threaded build, the class information was
148 being lost at freeze time on the first object with a STORABLE_freeze
149 hook. Consequentially the object was not blessed at all when thawed.
150 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
151 a store-time data lost bug, caused by failure to initialize internal
153 The bug was introduced as development perl change 16442 (on
154 2002/05/07), so has been present since 2.00.
155 Patches to introduce more regression tests to reduce the chance of
156 a reoccurance of this sort of goof are always welcome.
158 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
160 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
162 5.6.x introduced the ability to have IVs as long long. However,
163 Configure still defined BYTEORDER based on the size of a long.
164 Storable uses the BYTEORDER value as part of the header, but
165 doesn't explicity store sizeof(IV) anywhere in the header.
166 Hence on 5.6.x built with IV as long long on a platform that
167 uses Configure (ie most things except VMS and Windows) headers
168 are identical for the different IV sizes, despite the files
169 containing some fields based on sizeof(IV)
171 5.8.0 is consistent; all platforms have BYTEORDER in config.h
172 based on sizeof(IV) rather than sizeof(long). This means that
173 the value of BYTEORDER will change from (say) 4321 to 87654321
174 between 5.6.1 and 5.8.0 built with the same options to Configure
175 on the same machine. This means that the Storable header will
176 differ, and the two versions will wrongly thing that they are
179 For the benefit of long term consistency, Storable now
180 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
181 2.03 onwards default to be incompatible with 2.02 and earlier
182 (ie the large 1.0.x installed base) on the same 5.6.x perl.
184 To allow interworking, a new variable
185 $Storable::interwork_56_64bit is introduced. It defaults to
186 false. Set it to true to read and write old format files. Don't
187 use it unless you have existing stored data written with 5.6.x
188 that you couldn't otherwise read, or you need to interwork with
189 a machine running older Storable on a 5.6.x with long long IVs
190 (i.e., you probably don't need to use it).
192 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
196 Rewrite Storable.xs so that the file header structure for write_magic
197 is built at compile time, and check_magic attempts to the header in
198 blocks rather than byte per byte. These changes make the compiled
199 extension 2.25% smaller, but are not significant enough to give a
202 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
206 - New regression tests integer.t
207 - Add code to safely store large unsigned integers.
208 - Change code not to attempt to store large integers (ie > 32 bits)
209 in network order as 32 bits.
211 *Never* underestimate the value of a pathological test suite carefully
212 crafted with maximum malice before writing a line of real code. It
213 prevents crafty bugs from stowing away in your released code.
214 It's much less embarrassing to find them before you ship.
215 (Well, never underestimate it if you ever want to work for me)
217 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
219 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
221 The perl5 porters have decided to make sure that Storable still
222 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
223 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
224 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
225 due to a bug in 5.6.0 - upgrade to 5.6.1.
227 Jarkko and I have collated the list of changes the perl5 porters have
228 from the perl5 Changes file:
230 - data features of upcoming perl 5.8.0 are supported: Unicode hash
231 keys (Unicode hash values have been supported since Storable 1.0.1)
232 and "restricted hashes" (readonly hashes and hash entries)
233 - a newer version of perl can now be used to serialize data which is
234 not supported in earlier perls: Storable will attempt to do the
235 right thing for as long as possible, croaking only when safe data
236 conversion simply isn't possible. Alternatively earlier perls can
237 opt to have a lossy downgrade data instead of croaking
238 - when built with perls pre 5.7.3 this Storable writes out files
239 with binary format 2.4, the same format as Storable 1.0.8 onwards.
240 This should mean that this Storable will inter-operate seamlessly
241 with any Storable 1.0.8 or newer on perls pre 5.7.3
242 - dclone() now works with empty string scalar objects
243 - retrieving of large hashes is now more efficient
244 - more routines autosplit out of the main module, so Storable should
245 load slightly more quickly
246 - better documentation
247 - the internal context objects are now freed explicitly, rather than
248 relying on thread or process exit
249 - bugs fixed in debugging trace code affecting builds made with 64 bit
251 - code tidy-ups to allow clean compiles with more warning options
252 turned on avoid problems with $@ getting corrupted on 5.005_03 if
253 Carp wasn't already loaded
254 - added &show_file_magic, so you can add to /etc/magic and teach
255 Unix's file command about Storable files
257 We plan to keep Storable on CPAN in sync with the Perl core, so
258 if you encounter bugs or other problems building or using Storable,
259 please let us know at perl5-porters@perl.org
262 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
264 This is the LAST maintenance release of the Storable module.
265 Indeed, Storable is now part of perl 5.8, and will be maintained
266 as part of Perl. The CPAN module will remain available there
267 for people running pre-5.8 perls.
269 Avoid requiring Fcntl upfront, useful to embedded runtimes.
270 Use an eval {} for testing, instead of making Storable.pm
271 simply fail its compilation in the BEGIN block.
273 store_fd() will now correctly autoflush file if needed.
275 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
277 Fixed truncation race with lock_retrieve() in lock_store().
278 The file has to be truncated only once the exclusive lock is held.
280 Removed spurious debugging messages in .xs file.
282 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
284 Systematically use "=over 4" for POD linters.
285 Apparently, POD linters are much stricter than would
286 otherwise be needed, but that's OK.
288 Fixed memory corruption on croaks during thaw(). Thanks
289 to Claudio Garcia for reproducing this bug and providing the
290 code to exercise it. Added test cases for this bug, adapted
293 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
295 Changed tagnum and classnum from I32 to IV in context. Also
298 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
300 Last version was wrongly compiling with assertions on, due
301 to an edit glitch. That did not cause any problem (apart from
302 a slight performance loss) excepted on Win* platforms, where the
303 assertion code does not compile.
305 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
309 Forgot to increase version number at previous patch (there were
310 two of them, which is why we jump from 1.0.8 to 1.0.10).
312 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
314 Version 1.0.8, binary format 2.4.
316 Fixed incorrect error message.
318 Now bless objects ASAP at retrieve time, which is meant to fix
321 * Indirect references to overloaded object were not able to
322 restore overloading if the object was not blessed yet,
323 which was possible since blessing occurred only after the
326 * Storable hooks asking for serialization of blessed ref could
327 get un-blessed refs at retrieval time, for the very same
330 The fix implemented here was suggested by Nick Ing-Simmons.
332 Added support for blessed ref to tied structures. This is the
333 cause for the binary format change.
335 Added EBCDIC version of the compatibility test with 0.6.11,
338 Added tests for the new features, and to make sure the bugs they
339 are meant to fix are indeed fixed.
341 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
343 Removed spurious 'clean' entry in Makefile.PL.
345 Added CAN_FLOCK to determine whether we can flock() or not,
346 by inspecting Perl's configuration parameters, as determined
349 Trace offending package when overloading cannot be restored
352 Made context cleanup safer to avoid dup freeing, mostly in the
353 presence of repeated exceptions during store/retrieve (which can
354 cause memory leaks anyway, so it's just additional safety, not a
357 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
361 Fixed severe "object lost" bug for STORABLE_freeze returns,
362 when refs to lexicals, taken within the hook, were to be
363 serialized by Storable. Enhanced the t/recurse.t test to
364 stress hook a little more with refs to lexicals.
366 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
370 Documented that store() and retrieve() can return undef.
371 That is, the error reporting is not always made via exceptions,
372 as the paragraph on error reporting was implying.
374 Auto requires module of blessed ref when STORABLE_thaw misses.
375 When the Storable engine looks for the STORABLE_thaw hook and
376 does not find it, it now tries to require the package into which
377 the blessed reference is.
379 Just check $^O, in t/lock.t: there's no need to pull the whole
380 Config module for that.
382 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
386 Protected calls to flock() for DOS platform: apparently, the
387 flock/fcnlt emulation is reported to be broken on that
390 Added logcarp emulation if they don't have Log::Agent, since
391 we now use it to carp when lock_store/lock_retrieve is used
394 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
398 Avoid using "tainted" and "dirty" since Perl remaps them via
399 cpp (i.e. #define). This is deeply harmful when threading
400 is enabled. This concerned both the context structure and
401 local variable and argument names. Brrr..., scary!
403 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
407 Fixed spelling in README.
409 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
410 after a proposal from Erik Haugan <erik@solbors.no>.
412 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
414 Fixed stupid typo in the t/utf8.t test.
416 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
418 Version 1.0.1, binary format 2.3.
420 Documented that doubles are stored stringified by nstore().
422 Added Salvador Ortiz Garcia in CREDITS section, He identified
423 a bug in the store hooks and proposed the right fix: the class
424 id was allocated too soon. His bug case was also added to
425 the regression test suite.
427 Now only taint retrieved data when source was tainted. A bug
428 discovered by Marc Lehmann.
430 Added support for UTF-8 strings, a contribution of Marc Lehmann.
431 This is normally only activated in post-5.6 perls.
433 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
435 First official release Storable 1.0, for inclusion in perl 5.7.0.
436 The license scheme is now compatible with Perl's.
438 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
440 ANSI-fied most of the code, preparing for Perl core integration.
441 The next version of Storable will be 0.8, and will be integrated
442 into the Perl core (development branch).
444 Dispatch tables were moved upfront to relieve some compilers,
445 especially on AIX and Windows platforms.
447 Merged 64-bit fixes from perl5-porters.
449 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
451 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
452 the refcnt on the mg_ptr as well.
454 Removed spurious dependency to Devel::Peek, which was used for
455 testing only in t/tied_items.t. Thanks to Conrad Heiney
456 <conrad@fringehead.org> for spotting it first.
458 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
460 Marc Lehmann kindly contributed code to add overloading support
461 and to handle references to tied variables.
463 Rewrote leading blurb about compatibility to make it clearer what
464 "backward compatibility" is about: when I say 0.7 is backward
465 compatible with 0.6, it means the revision 0.7 can read files
468 Mention new Clone(3) extension in SEE ALSO.
470 Was wrongly optimizing for "undef" values in hashes by not
471 fully recursing: as a result, tied "undef" values were incorrectly
474 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
476 First revision of Storable 0.7.
478 The serializing format is new, known as version 2.0. It is fully
479 backward compatible with 0.6. Earlier formats are deprecated and
480 have not even been tested: next version will drop pre-0.6 format.
482 Changes since 0.6@11:
484 - Moved interface to the "beta" status. Some tiny parts are still
485 subject to change, but nothing important enough to warrant an "alpha"
488 - Slightly reduced the size of the Storable image by factorizing
489 object class names and removing final object storage notification due
490 to a redesign of the blessed object storing.
492 - Classes can now redefine how they wish their instances to be serialized
493 and/or deep cloned. Serializing hooks are written in Perl code.
495 - The engine is now fully re-entrant.
497 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
499 Added provision to detect more recent binary formats, since
500 the new upcoming Storable-0.7 will use a different format.
501 In order to prevent attempting the de-serialization of newer
502 formats by older versions, I'm adding this now to the 0.6 series.
504 I'm expecting this revision to be the last of the 0.6 series.
505 Unless it does not work with perl 5.6, which I don't use yet,
506 and therefore against which I cannot test.
508 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
510 Added note about format incompatibilities with old versions
511 (i.e. pre 0.5@9 formats, which cannot be understood as there
512 was no versionning information in the file by then).
514 Protect all $@ variables when eval {} used, to avoid corrupting
515 it when store/retrieve is called within an exception handler.
517 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
518 preventing Perl's patchlevel from being included, which is
519 needed starting from 5.6.
521 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
523 Fixed shared "undef" bug in hashes, which did not remain shared
524 through store/retrieve.
526 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
528 added last_op_in_netorder() predicate
529 documented last_op_in_netorder()
530 added tests for the new last_op_in_netorder() predicate
532 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
534 Forgot to update VERSION
536 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
538 Added mention of japanese translation for the manual page.
540 Fixed typo in macro that made threaded code not compilable,
541 especially on Win32 platforms.
543 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
544 directly instead of relying on internal symbols.
546 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
548 Integrated "thread-safe" patch from Murray Nesbitt.
549 Note that this may not be very efficient for threaded code,
550 see comment in the code.
552 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
553 since I don't have access to such machines.
555 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
557 changed my e-mail to pobox.
559 mentionned it is not thread-safe.
561 updated version number.
563 uses new internal PL_* naming convention.
565 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
567 Updated benchmark figures due to recent optimizations done in
568 store(): tagnums are now stored as-is in the hash table, so
569 no surrounding SV is created. And the "shared keys" mode for
570 hash table was turned off.
572 Fixed backward compatibility (wrt 0.5@9) for retrieval of
573 blessed refs. That old version did something wrong, but the
574 bugfix prevented correct retrieval of the old format.
576 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
578 Changed benchmark figures.
580 Adjust refcnt of tied objects after calling sv_magic() to avoid
581 memory leaks. Contributed by Jeff Gresham.
583 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
585 Added workaround for persistent LVALUE-ness in perl5.004. All
586 scalars tagged as being an lvalue are handled as if they were
587 not an lvalue at all. Added test for that LVALUE bug workaround.
589 Now handles Perl immortal scalars explicitely, by storing &sv_yes
590 as such, explicitely.
592 Retrieval of non-immortal undef cannot be shared. Previous
593 version was over-optimizing by not creating a separate SV for
594 all undefined scalars seen.
596 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
598 Baseline for Storable-0.6@0.
600 This version introduces a binary incompatibility in the generated
601 binary image, which is more compact than older ones by approximatively
602 15%, depending on the exact degree of sharing in your structures.
604 The good news is that your older images can still be retrieved with
605 this version, i.e. backward compatibility is preserved. This version
606 of Storable can only generate new binaries however.
608 Another good news is that the retrieval of data structure is
609 significantly quicker than before, because a Perl array is used
610 instead of a hash table to keep track of retrieved objects, and
611 also because the image being smaller, less I/O function calls are
614 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
616 Version number now got from Storable.pm directly.
618 Fixed overzealous sv_type() optimization, which would make
619 Storable fail when faced with an "upgraded" SV to the PVIV
620 or PVNV kind containing a reference.
622 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
624 Extended the SYNOPSIS section to give quick overview of the
625 routines and their signature.
627 Optimized sv_type() to avoid flags checking when not needed, i.e.
628 when their type makes it impossible for them to be refs or tied.
629 This slightly increases throughput by a few percents when refs
630 and tied variables are marginal occurrences in your data.
632 Stubs for XS now use OutputStream and InputStream file types to
633 make it work when the given file is actually a socket. Perl
634 makes a distinction for sockets in its internal I/O structures
635 by having both a read and a write structure, whereas plain files
638 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
640 Thanks to a contribution from Benjamin A. Holzman, Storable is now
641 able to correctly serialize tied SVs, i.e. tied arrays, hashes
644 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
646 I said SvPOK() had changed to SvPOKp(), but that was a lie...
648 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
650 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
651 in effect mangled the object tags and prevented portability
652 accross 32/64 bit architectures!
654 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
656 Added code example for store_fd() and retrieve_fd() in the
657 man page, to emphasize that file descriptors must be passed as
658 globs, not as plain strings.
660 Cannot use SV addresses as tag when using nstore() on LP64. This
661 was the cause of problems when creating a storable image on an
662 LP64 machine and retrieving it on an ILP32 system, which is
663 exactly what nstore() is meant for...
665 However, we continue to use SV addresses as tags for plain store(),
666 because benchamarking shows that it saves up to 8% of the store
667 time, and store() is meant to be fast at the expense of lack
670 This means there will be approximately an 8% degradation of
671 performance for nstore(), but it's now working as expected.
672 That cost may vary on your machine of course, since it is
673 solely caused by the memory allocation overhead used to create
674 unique SV tags for each distinct stored SV.
676 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
678 Don't use any '_' in version number.
680 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
682 Updated version number.
684 added binmode() calls for systems where it matters.
686 Be sure to pass globs, not plain file strings, to C routines,
687 so that Storable can be used under the Perl debugger.
689 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
691 Fix memory leaks on seen hash table and returned SV refs.
693 Storable did not work properly when tainting enabled.
695 Fixed "Allocation too large" messages in freeze/thaw and added.
696 proper regression test in t/freeze.t.
698 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
700 Updated version number
702 Added freeze/thaw interface and dclone.
704 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
706 Forgot that AutoLoader does not export its own AUTOLOAD.
709 use AutoLoader 'AUTOLOAD';
711 but that would not be backward compatible. So the export is
714 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
716 Empty scalar strings are now "defined" at retrieval time.
718 New test to ensure an empty string is defined when retrieved.
720 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
722 Updated version number
724 Declare VERSION as being used
726 Fixed a typo in the PerlIO_putc remapping.
727 PerlIO_read and perlIO_write inverted size/nb_items.
728 (only relevant for pre-perl5.004 versions)
730 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
732 Updated version number
734 Added VERSION identification
736 Allow build with perl5.003, which is ante perlIO time
738 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
742 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
744 Updated version number in Makefile.PL.
746 Added "thanks to" section to README.
748 Documented new forgive_me variable.
752 Added forgive_me support to allow store() of data structures
753 containing non-storable items like CODE refs.