1 Sat Jan 3 18:49:18 GMT 2004 Nicholas Clark <nick@ccl4.org>
5 Fix minor problems with the CPAN release
6 1: Make Storable.xs work on 5.8.2 and later (already in the core)
7 2: Ship the linux hints file
8 3: Ship Test::More for the benefit of Perls pre 5.6.2
9 4: Correct Makefile.PL to only install in core for 5.8.0 and later
11 Sat Sep 6 01:08:20 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
15 This release works around a 5.8.0 bug which caused hashes to not
16 be marked as having key flags even though an HEK had HEK_WASUTF8
17 set. (Note that the only reasonable solution is to silently drop
18 the flag from the affected key.)
20 Users of RT 3 who were seeing assertion failures should upgrade.
21 (Perl 5.8.1 will have the bug fixed.)
23 Mon May 5 10:24:16 IST 2003 Abhijit Menon-Sen <ams@wiw.org>
27 Minor bugfixes (self-tied objects are now correctly stored, as
28 are the results of additions larger than INT_MAX).
30 Mon Oct 7 21:56:38 BST 2002 Nicholas Clark <nick@ccl4.org>
34 Remove qr// from t/downgrade.t so that it will run on 5.004
35 Mention $File::Spec::VERSION a second time in t/forgive.t so that it
36 runs without warnings in 5.004 (this may be a 5.00405 bug I'm working
38 Fix t/integer.t initialisation to actually generate 64 bits of 9c
39 Fix comparison tests to use eval to get around 64 bit IV conversion
40 issues on 5.6.x, following my t/integer.t ^ precedence bug found by
42 Alter t/malice.t to work with Test/More.pm in t/, and skip individual
43 subtests that use $Config{ptrsize}, so that the rest of the test can
45 Change t/malice.t and the error message in check_magic in Storable.xs
46 from "Pointer integer size" to "Pointer size"
47 Remove prerequisite of Test::More from Makefile.PL
48 Ship Test::Builder, Test::Simple and Test::More in t
50 Thu Oct 3 08:57:22 IST 2002 Abhijit Menon-Sen <ams@wiw.org>
54 Adds support for CODE references from Slaven Rezic
55 <slaven.rezic@berlin.de>.
57 Fri Jun 7 23:55:41 BST 2002 Nicholas Clark
61 Bug fix from Radu Greab <radu@netsoft.ro> (plus regression test)
62 to fix a recently introduced bug detected by Dave Rolsky.
63 Bug was that for a non threaded build, the class information was
64 being lost at freeze time on the first object with a STORABLE_freeze
65 hook. Consequentially the object was not blessed at all when thawed.
66 (The presence (or lack) of STORABLE_thaw was irrelevant; this was
67 a store-time data lost bug, caused by failure to initialize internal
69 The bug was introduced as development perl change 16442 (on
70 2002/05/07), so has been present since 2.00.
71 Patches to introduce more regression tests to reduce the chance of
72 a reoccurance of this sort of goof are always welcome.
74 Thu May 30 20:31:08 BST 2002 Nicholas Clark <nick@ccl4.org>
76 Version 2.03 Header changes on 5.6.x on Unix where IV is long long
78 5.6.x introduced the ability to have IVs as long long. However,
79 Configure still defined BYTEORDER based on the size of a long.
80 Storable uses the BYTEORDER value as part of the header, but
81 doesn't explicity store sizeof(IV) anywhere in the header.
82 Hence on 5.6.x built with IV as long long on a platform that
83 uses Configure (ie most things except VMS and Windows) headers
84 are identical for the different IV sizes, despite the files
85 containing some fields based on sizeof(IV)
87 5.8.0 is consistent; all platforms have BYTEORDER in config.h
88 based on sizeof(IV) rather than sizeof(long). This means that
89 the value of BYTEORDER will change from (say) 4321 to 87654321
90 between 5.6.1 and 5.8.0 built with the same options to Configure
91 on the same machine. This means that the Storable header will
92 differ, and the two versions will wrongly thing that they are
95 For the benefit of long term consistency, Storable now
96 implements the 5.8.0 BYTEORDER policy on 5.6.x. This means that
97 2.03 onwards default to be incompatible with 2.02 and earlier
98 (ie the large 1.0.x installed base) on the same 5.6.x perl.
100 To allow interworking, a new variable
101 $Storable::interwork_56_64bit is introduced. It defaults to
102 false. Set it to true to read and write old format files. Don't
103 use it unless you have existing stored data written with 5.6.x
104 that you couldn't otherwise read, or you need to interwork with
105 a machine running older Storable on a 5.6.x with long long IVs
106 (i.e., you probably don't need to use it).
108 Sat May 25 22:38:39 BST 2002 Nicholas Clark <nick@ccl4.org>
112 Rewrite Storable.xs so that the file header structure for write_magic
113 is built at compile time, and check_magic attempts to the header in
114 blocks rather than byte per byte. These changes make the compiled
115 extension 2.25% smaller, but are not significant enough to give a
118 Thu May 23 22:50:41 BST 2002 Nicholas Clark <nick@ccl4.org>
122 - New regression tests integer.t
123 - Add code to safely store large unsigned integers.
124 - Change code not to attempt to store large integers (ie > 32 bits)
125 in network order as 32 bits.
127 *Never* underestimate the value of a pathological test suite carefully
128 crafted with maximum malice before writing a line of real code. It
129 prevents crafty bugs from stowing away in your released code.
130 It's much less embarrassing to find them before you ship.
131 (Well, never underestimate it if you ever want to work for me)
133 Fri May 17 22:48:59 BST 2002 Nicholas Clark <nick@ccl4.org>
135 Version 2.0, binary format 2.5 (but writes format 2.4 on pre 5.7.3)
137 The perl5 porters have decided to make sure that Storable still
138 builds on pre-5.8 perls, and make the 5.8 version available on CPAN.
139 The VERSION is now 2.0, and it passes all tests on 5.005_03, 5.6.1
140 and 5.6.1 with threads. On 5.6.0 t/downgrade.t fails tests 34 and 37,
141 due to a bug in 5.6.0 - upgrade to 5.6.1.
143 Jarkko and I have collated the list of changes the perl5 porters have
144 from the perl5 Changes file:
146 - data features of upcoming perl 5.8.0 are supported: Unicode hash
147 keys (Unicode hash values have been supported since Storable 1.0.1)
148 and "restricted hashes" (readonly hashes and hash entries)
149 - a newer version of perl can now be used to serialize data which is
150 not supported in earlier perls: Storable will attempt to do the
151 right thing for as long as possible, croaking only when safe data
152 conversion simply isn't possible. Alternatively earlier perls can
153 opt to have a lossy downgrade data instead of croaking
154 - when built with perls pre 5.7.3 this Storable writes out files
155 with binary format 2.4, the same format as Storable 1.0.8 onwards.
156 This should mean that this Storable will inter-operate seamlessly
157 with any Storable 1.0.8 or newer on perls pre 5.7.3
158 - dclone() now works with empty string scalar objects
159 - retrieving of large hashes is now more efficient
160 - more routines autosplit out of the main module, so Storable should
161 load slightly more quickly
162 - better documentation
163 - the internal context objects are now freed explicitly, rather than
164 relying on thread or process exit
165 - bugs fixed in debugging trace code affecting builds made with 64 bit
167 - code tidy-ups to allow clean compiles with more warning options
168 turned on avoid problems with $@ getting corrupted on 5.005_03 if
169 Carp wasn't already loaded
170 - added &show_file_magic, so you can add to /etc/magic and teach
171 Unix's file command about Storable files
173 We plan to keep Storable on CPAN in sync with the Perl core, so
174 if you encounter bugs or other problems building or using Storable,
175 please let us know at perl5-porters@perl.org
178 Sat Dec 1 14:37:54 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
180 This is the LAST maintenance release of the Storable module.
181 Indeed, Storable is now part of perl 5.8, and will be maintained
182 as part of Perl. The CPAN module will remain available there
183 for people running pre-5.8 perls.
185 Avoid requiring Fcntl upfront, useful to embedded runtimes.
186 Use an eval {} for testing, instead of making Storable.pm
187 simply fail its compilation in the BEGIN block.
189 store_fd() will now correctly autoflush file if needed.
191 Tue Aug 28 23:53:20 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
193 Fixed truncation race with lock_retrieve() in lock_store().
194 The file has to be truncated only once the exclusive lock is held.
196 Removed spurious debugging messages in .xs file.
198 Sun Jul 1 13:27:32 MEST 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
200 Systematically use "=over 4" for POD linters.
201 Apparently, POD linters are much stricter than would
202 otherwise be needed, but that's OK.
204 Fixed memory corruption on croaks during thaw(). Thanks
205 to Claudio Garcia for reproducing this bug and providing the
206 code to exercise it. Added test cases for this bug, adapted
209 Made code compile cleanly with -Wall (from Jarkko Hietaniemi).
211 Changed tagnum and classnum from I32 to IV in context. Also
214 Thu Mar 15 01:22:32 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
216 Last version was wrongly compiling with assertions on, due
217 to an edit glitch. That did not cause any problem (apart from
218 a slight performance loss) excepted on Win* platforms, where the
219 assertion code does not compile.
221 Sat Feb 17 13:37:37 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
225 Forgot to increase version number at previous patch (there were
226 two of them, which is why we jump from 1.0.8 to 1.0.10).
228 Sat Feb 17 13:35:00 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
230 Version 1.0.8, binary format 2.4.
232 Fixed incorrect error message.
234 Now bless objects ASAP at retrieve time, which is meant to fix
237 * Indirect references to overloaded object were not able to
238 restore overloading if the object was not blessed yet,
239 which was possible since blessing occurred only after the
242 * Storable hooks asking for serialization of blessed ref could
243 get un-blessed refs at retrieval time, for the very same
246 The fix implemented here was suggested by Nick Ing-Simmons.
248 Added support for blessed ref to tied structures. This is the
249 cause for the binary format change.
251 Added EBCDIC version of the compatibility test with 0.6.11,
254 Added tests for the new features, and to make sure the bugs they
255 are meant to fix are indeed fixed.
257 Wed Jan 3 10:43:18 MET 2001 Raphael Manfredi <Raphael_Manfredi@pobox.com>
259 Removed spurious 'clean' entry in Makefile.PL.
261 Added CAN_FLOCK to determine whether we can flock() or not,
262 by inspecting Perl's configuration parameters, as determined
265 Trace offending package when overloading cannot be restored
268 Made context cleanup safer to avoid dup freeing, mostly in the
269 presence of repeated exceptions during store/retrieve (which can
270 cause memory leaks anyway, so it's just additional safety, not a
273 Sun Nov 5 18:23:48 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
277 Fixed severe "object lost" bug for STORABLE_freeze returns,
278 when refs to lexicals, taken within the hook, were to be
279 serialized by Storable. Enhanced the t/recurse.t test to
280 stress hook a little more with refs to lexicals.
282 Thu Oct 26 19:14:38 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
286 Documented that store() and retrieve() can return undef.
287 That is, the error reporting is not always made via exceptions,
288 as the paragraph on error reporting was implying.
290 Auto requires module of blessed ref when STORABLE_thaw misses.
291 When the Storable engine looks for the STORABLE_thaw hook and
292 does not find it, it now tries to require the package into which
293 the blessed reference is.
295 Just check $^O, in t/lock.t: there's no need to pull the whole
296 Config module for that.
298 Mon Oct 23 20:03:49 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
302 Protected calls to flock() for DOS platform: apparently, the
303 flock/fcnlt emulation is reported to be broken on that
306 Added logcarp emulation if they don't have Log::Agent, since
307 we now use it to carp when lock_store/lock_retrieve is used
310 Fri Sep 29 21:52:29 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
314 Avoid using "tainted" and "dirty" since Perl remaps them via
315 cpp (i.e. #define). This is deeply harmful when threading
316 is enabled. This concerned both the context structure and
317 local variable and argument names. Brrr..., scary!
319 Thu Sep 28 23:46:39 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
323 Fixed spelling in README.
325 Added lock_store, lock_nstore, and lock_retrieve (advisory locking)
326 after a proposal from Erik Haugan <erik@solbors.no>.
328 Perls before 5.004_04 lack newSVpvn, added remapping in XS.
330 Fixed stupid typo in the t/utf8.t test.
332 Sun Sep 17 18:51:10 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
334 Version 1.0.1, binary format 2.3.
336 Documented that doubles are stored stringified by nstore().
338 Added Salvador Ortiz Garcia in CREDITS section, He identified
339 a bug in the store hooks and proposed the right fix: the class
340 id was allocated too soon. His bug case was also added to
341 the regression test suite.
343 Now only taint retrieved data when source was tainted. A bug
344 discovered by Marc Lehmann.
346 Added support for UTF-8 strings, a contribution of Marc Lehmann.
347 This is normally only activated in post-5.6 perls.
349 Thu Aug 31 23:06:06 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
351 First official release Storable 1.0, for inclusion in perl 5.7.0.
352 The license scheme is now compatible with Perl's.
354 Thu Aug 24 01:02:02 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
356 ANSI-fied most of the code, preparing for Perl core integration.
357 The next version of Storable will be 0.8, and will be integrated
358 into the Perl core (development branch).
360 Dispatch tables were moved upfront to relieve some compilers,
361 especially on AIX and Windows platforms.
363 Merged 64-bit fixes from perl5-porters.
365 Mon Aug 14 09:22:04 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
367 Added a refcnt dec in retrieve_tied_key(): sv_magic() increases
368 the refcnt on the mg_ptr as well.
370 Removed spurious dependency to Devel::Peek, which was used for
371 testing only in t/tied_items.t. Thanks to Conrad Heiney
372 <conrad@fringehead.org> for spotting it first.
374 Sun Aug 13 22:12:59 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
376 Marc Lehmann kindly contributed code to add overloading support
377 and to handle references to tied variables.
379 Rewrote leading blurb about compatibility to make it clearer what
380 "backward compatibility" is about: when I say 0.7 is backward
381 compatible with 0.6, it means the revision 0.7 can read files
384 Mention new Clone(3) extension in SEE ALSO.
386 Was wrongly optimizing for "undef" values in hashes by not
387 fully recursing: as a result, tied "undef" values were incorrectly
390 Sun Jul 30 12:59:17 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
392 First revision of Storable 0.7.
394 The serializing format is new, known as version 2.0. It is fully
395 backward compatible with 0.6. Earlier formats are deprecated and
396 have not even been tested: next version will drop pre-0.6 format.
398 Changes since 0.6@11:
400 - Moved interface to the "beta" status. Some tiny parts are still
401 subject to change, but nothing important enough to warrant an "alpha"
404 - Slightly reduced the size of the Storable image by factorizing
405 object class names and removing final object storage notification due
406 to a redesign of the blessed object storing.
408 - Classes can now redefine how they wish their instances to be serialized
409 and/or deep cloned. Serializing hooks are written in Perl code.
411 - The engine is now fully re-entrant.
413 Sun Apr 2 23:47:50 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
415 Added provision to detect more recent binary formats, since
416 the new upcoming Storable-0.7 will use a different format.
417 In order to prevent attempting the de-serialization of newer
418 formats by older versions, I'm adding this now to the 0.6 series.
420 I'm expecting this revision to be the last of the 0.6 series.
421 Unless it does not work with perl 5.6, which I don't use yet,
422 and therefore against which I cannot test.
424 Wed Mar 29 19:55:21 MEST 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
426 Added note about format incompatibilities with old versions
427 (i.e. pre 0.5@9 formats, which cannot be understood as there
428 was no versionning information in the file by then).
430 Protect all $@ variables when eval {} used, to avoid corrupting
431 it when store/retrieve is called within an exception handler.
433 Mistakenly included "patchlevel.h" instead of <patchlevel.h>,
434 preventing Perl's patchlevel from being included, which is
435 needed starting from 5.6.
437 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
439 Fixed shared "undef" bug in hashes, which did not remain shared
440 through store/retrieve.
442 Thu Feb 10 19:48:16 MET 2000 Raphael Manfredi <Raphael_Manfredi@pobox.com>
444 added last_op_in_netorder() predicate
445 documented last_op_in_netorder()
446 added tests for the new last_op_in_netorder() predicate
448 Wed Oct 20 19:07:36 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
450 Forgot to update VERSION
452 Tue Oct 19 21:25:02 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
454 Added mention of japanese translation for the manual page.
456 Fixed typo in macro that made threaded code not compilable,
457 especially on Win32 platforms.
459 Changed detection of older perls (pre-5.005) by testing PATCHLEVEL
460 directly instead of relying on internal symbols.
462 Tue Sep 14 22:13:28 MEST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
464 Integrated "thread-safe" patch from Murray Nesbitt.
465 Note that this may not be very efficient for threaded code,
466 see comment in the code.
468 Try to avoid compilation warning on 64-bit CPUs. Can't test it,
469 since I don't have access to such machines.
471 Mon Jul 12 14:37:19 METDST 1999 Raphael Manfredi <Raphael_Manfredi@pobox.com>
473 changed my e-mail to pobox.
475 mentionned it is not thread-safe.
477 updated version number.
479 uses new internal PL_* naming convention.
481 Fri Jul 3 13:38:16 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
483 Updated benchmark figures due to recent optimizations done in
484 store(): tagnums are now stored as-is in the hash table, so
485 no surrounding SV is created. And the "shared keys" mode for
486 hash table was turned off.
488 Fixed backward compatibility (wrt 0.5@9) for retrieval of
489 blessed refs. That old version did something wrong, but the
490 bugfix prevented correct retrieval of the old format.
492 Mon Jun 22 11:00:48 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
494 Changed benchmark figures.
496 Adjust refcnt of tied objects after calling sv_magic() to avoid
497 memory leaks. Contributed by Jeff Gresham.
499 Fri Jun 12 11:50:04 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
501 Added workaround for persistent LVALUE-ness in perl5.004. All
502 scalars tagged as being an lvalue are handled as if they were
503 not an lvalue at all. Added test for that LVALUE bug workaround.
505 Now handles Perl immortal scalars explicitely, by storing &sv_yes
506 as such, explicitely.
508 Retrieval of non-immortal undef cannot be shared. Previous
509 version was over-optimizing by not creating a separate SV for
510 all undefined scalars seen.
512 Thu Jun 4 17:21:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
514 Baseline for Storable-0.6@0.
516 This version introduces a binary incompatibility in the generated
517 binary image, which is more compact than older ones by approximatively
518 15%, depending on the exact degree of sharing in your structures.
520 The good news is that your older images can still be retrieved with
521 this version, i.e. backward compatibility is preserved. This version
522 of Storable can only generate new binaries however.
524 Another good news is that the retrieval of data structure is
525 significantly quicker than before, because a Perl array is used
526 instead of a hash table to keep track of retrieved objects, and
527 also because the image being smaller, less I/O function calls are
530 Tue May 12 09:15:15 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
532 Version number now got from Storable.pm directly.
534 Fixed overzealous sv_type() optimization, which would make
535 Storable fail when faced with an "upgraded" SV to the PVIV
536 or PVNV kind containing a reference.
538 Thu Apr 30 15:11:30 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
540 Extended the SYNOPSIS section to give quick overview of the
541 routines and their signature.
543 Optimized sv_type() to avoid flags checking when not needed, i.e.
544 when their type makes it impossible for them to be refs or tied.
545 This slightly increases throughput by a few percents when refs
546 and tied variables are marginal occurrences in your data.
548 Stubs for XS now use OutputStream and InputStream file types to
549 make it work when the given file is actually a socket. Perl
550 makes a distinction for sockets in its internal I/O structures
551 by having both a read and a write structure, whereas plain files
554 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
556 Thanks to a contribution from Benjamin A. Holzman, Storable is now
557 able to correctly serialize tied SVs, i.e. tied arrays, hashes
560 Thu Apr 9 18:07:51 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
562 I said SvPOK() had changed to SvPOKp(), but that was a lie...
564 Wed Apr 8 13:14:29 METDST 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
566 Wrote sizeof(SV *) instead of sizeof(I32) when portable, which
567 in effect mangled the object tags and prevented portability
568 accross 32/64 bit architectures!
570 Wed Mar 25 14:57:02 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
572 Added code example for store_fd() and retrieve_fd() in the
573 man page, to emphasize that file descriptors must be passed as
574 globs, not as plain strings.
576 Cannot use SV addresses as tag when using nstore() on LP64. This
577 was the cause of problems when creating a storable image on an
578 LP64 machine and retrieving it on an ILP32 system, which is
579 exactly what nstore() is meant for...
581 However, we continue to use SV addresses as tags for plain store(),
582 because benchamarking shows that it saves up to 8% of the store
583 time, and store() is meant to be fast at the expense of lack
586 This means there will be approximately an 8% degradation of
587 performance for nstore(), but it's now working as expected.
588 That cost may vary on your machine of course, since it is
589 solely caused by the memory allocation overhead used to create
590 unique SV tags for each distinct stored SV.
592 Tue Jan 20 09:21:53 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
594 Don't use any '_' in version number.
596 Tue Jan 13 17:51:50 MET 1998 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
598 Updated version number.
600 added binmode() calls for systems where it matters.
602 Be sure to pass globs, not plain file strings, to C routines,
603 so that Storable can be used under the Perl debugger.
605 Wed Nov 5 10:53:22 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
607 Fix memory leaks on seen hash table and returned SV refs.
609 Storable did not work properly when tainting enabled.
611 Fixed "Allocation too large" messages in freeze/thaw and added.
612 proper regression test in t/freeze.t.
614 Tue Jun 3 09:41:33 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
616 Updated version number
618 Added freeze/thaw interface and dclone.
620 Fri May 16 10:45:47 METDST 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
622 Forgot that AutoLoader does not export its own AUTOLOAD.
625 use AutoLoader 'AUTOLOAD';
627 but that would not be backward compatible. So the export is
630 Tue Mar 25 11:21:32 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
632 Empty scalar strings are now "defined" at retrieval time.
634 New test to ensure an empty string is defined when retrieved.
636 Thu Feb 27 16:32:44 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
638 Updated version number
640 Declare VERSION as being used
642 Fixed a typo in the PerlIO_putc remapping.
643 PerlIO_read and perlIO_write inverted size/nb_items.
644 (only relevant for pre-perl5.004 versions)
646 Thu Feb 27 15:58:31 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
648 Updated version number
650 Added VERSION identification
652 Allow build with perl5.003, which is ante perlIO time
654 Mon Jan 13 17:53:18 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
658 Wed Jan 22 15:19:56 MET 1997 Raphael Manfredi <Raphael_Manfredi@grenoble.hp.com>
660 Updated version number in Makefile.PL.
662 Added "thanks to" section to README.
664 Documented new forgive_me variable.
668 Added forgive_me support to allow store() of data structures
669 containing non-storable items like CODE refs.