Applied patch by Wulfram Humann for improving reindex_entry()
[dbsrgits/DBM-Deep.git] / Changes
CommitLineData
ffed8b01 1Revision history for DBM::Deep.
2
695c88b1 31.0013 Jun 18 00:00:00 2008 EST
4 - (This version is compatible with 1.0012)
5 - Fix for RT#30144 (Optimization failure on Win32)
6 - Fixed a bug in reindex_entry (Thanks, Wulfram Humann!)
7
08164b50 81.0012 Jun 09 15:00:00 2008 EST
9 - (This version is compatible with 1.0011)
10 - Fix for RT#30085 (Remove dependency on XS module)
11 - Thank you very much tachyon-II@Perlmonks!!
12 - Updated the POD with fixes that were made, but still on the TODO list.
13 - Bypass for RT#36419 (t/44_upgrade_db.t fails on *BSD)
14 - We're just going to skip that for now. It's a number-of-processes-open
15 issue. Best is to do a port to another module to do the heavy lifting.
16
171.0011 May 27 15:00:00 2008 EST
18 - (This version is compatible with 1.0010)
19 - A test has the wrong plan.
20
edd45134 211.0010 May 27 12:00:00 2008 EST
b8370759 22 - (This version is compatible with 1.0009)
23 - Fix for RT#35140 (invalid POD links)
edd45134 24 - Fix for RT#34819 (Cannot assign the same value back to the same location)
25 - Fix for RT#29957 (Cannot assign the same value back to the same location)
26 - Fix for RT#33863 (Cannot shift an arrayref from an array)
27 - When something is deleted from a DB, the value is export()ed, allowing it
28 to be saved.
29 - This exporting is only done if the refcount == 0 after the deletion.
b8370759 30
6e6789b0 311.0009 Mar 19 12:00:00 2008 EDT
32 - (This version is compatible with 1.0008)
33 - Internal refactorings to prepare for some optimizations.
34 - _fh() has been removed. It was marked as private, so don't complain.
35 - Skip a test that was spuriously failing on Win32 (Thanks, Alias!)
36
45f047f8 371.0008 Mar 09 20:00:00 2008 EDT
38 - (This version is compatible with 1.0007)
39 - Fixed a number of Win32 issues (Reported by Steven Samelson - thank you!)
40 - Much thanks to Nigel Sandever and David Golden for their help
41 debugging the issues, particularly with DBM::Deep's usage of
42 File::Temp (which removes a number of warnings).
43 - Autovivification now works on Win32. It turns out that when a
44 process takes a shared flock on a file, it's not allowed to write to
45 it under Win32, unlike *nix. This is probably a good catch.
46 - Note: The fix is a hack. All locks are now exclusive until a
47 better fix is found.
48
5a70a6c0 491.0007 Jan 10 00:00:00 2008 EDT
50 - (This version is compatible with 1.0006)
51 - Applied a patch+failing test submitted by sprout@cpan.org. Thanks!
52 - Turns out that the case of 17 keys with the same first character in the
53 MD5 hash wasn't being tested for. This was a crashbug.
54 - A fix has been made to upgrade_db.pl (RT# 30067)
55 - The version determinations were in the wrong order or evaluation. This
56 meant that upgrade_db.pl wouldn't work as expected (or at all).
57 - Added a minimum Pod::Usage requirement (RT# 29976)
58 - It's an optional item in Build.PL
59 - utils/upgrade_db.pl now checks for that version, as does the test.
60
e00d0eb3 611.0006 Oct 01 23:15:00 2007 EDT
62 - (This version is compatible with 1.0005)
63 - Removed Clone and replaced it with a hand-written datastructure walker.
64 - This greatly reduces the footprint of a large import
65 - This bypasses a failure of Clone under Perl 5.9.5
66 - Moved t/37_delete_edge_cases.t to t_attic because it wasn't really used
67 - import() has a stricter API now. This is a potentially incompatible API
68 change. Only HASH and ARRAY refs are now allowed and they must match the type
69 of the object being imported into.
70
c57b19c6 711.0005 Oct 01 11:15:00 2007 EDT
72 - (This version is compatible with 1.0004)
73 - Added proper singleton support. This means that the following now works:
74 $db->{foo} = [ 1 .. 3];
75 my $x = $db->{foo};
76 my $y = $db->{foo};
77 is( $x, $y ); # Now passes
78 - This means that Data::Dumper now properly reports when $db->{foo} = $db->{bar}
79
888453b9 801.0004 Sep 28 12:15:00 2007 EDT
81 - (This version is compatible with 1.0003)
82 - Fixed the Changes file (wrong version was displayed for 1.0003)
83 - Added filter sugar methods to be more API-compatible with other DBMs
84 - This was added to support a patch provided to IO::All so it can
85 use DBM::Deep as a DBM provider.
86 - Implemented _dump_file in order to display the file structure. As a
87 result, the following bugs were fixed:
88 - Arrays and hashes now clean up after themselves better.
89 - Bucketlists now clean up after themselves better.
90 - Reindexing properly clears the old bucketlist before freeing it.
91
921.0003 Sep 24 14:00:00 2007 EDT
1cff45d7 93 - THIS VERSION IS INCOMPATIBLE WITH FILES FROM ALL OTHER PRIOR VERSIONS.
94 - Further fixes for unshift/shift/splice and references (RT# 29583)
95 - To fix that, I had to put support for real references in.
96 - the 16 and 22 tests are now re-enabled.
97 - Yes, this means that real references work. See t/45_references.t
98
807f63a7 991.0002 Sep 20 22:00:00 2007 EDT
100 - (This version is compatible with 1.0001)
101 - Expanded _throw_error() so that it provides better information.
102 (Thanks brian d foy!)
103 - Fixed how shift, unshift, and splice work when there are references
104 being moved. It now no longer dies.
105 - Added diag in t/17_import.t to note that the failing tests on blead
106 are due to Clone being broken, not DBM::Deep. The tests will still
107 fail because I don't want users to install something that's broken
108 and deal with those bug reports.
109
151e0077 1101.0001 Mar 12 16:15:00 2007 EDT
111 - (This version is compatible with 1.0000)
112 - Added a missing dependency on IO::Scalar (RT #25387)
113 - Fixed how t/44_upgrade_db.t and utils/upgrade_db.pl worked
114 - utils/upgrade_db.pl now uses #!perl, not #!/usr/bin/perl
115 - t/44_upgrade_db.t now explicitly calls $^X
116 (Thanks merlyn!)
117
1181.0000 Feb 26 22:30:00 2007 EST
e9b0b5f0 119 - THIS VERSION IS INCOMPATIBLE WITH FILES FROM ALL OTHER PRIOR VERSIONS.
120 - To aid in this form of upgrades, DBM::Deep now checks the file format
121 version to make sure that it knows how to read it.
122 - db_upgrade.pl was added to utils/. This will -NOT- install onto
123 your system. This is deliberate.
124 - db_upgrade.pl will not handle developer release file formats. This
125 is due to the fact that all developer releases in preparation for a
126 given release share the same file version, even though the file
127 format may change. This is deliberate.
128 - Importing no longer takes place within a transaction
129 - The following parameters were added:
130 - data_sector_size - this determines the default size of a data sector.
131 - Correctly handle opening readonly files
132
151e0077 1330.99_04 Jan 24 22:30:00 2007 EST
2120a181 134 - Added the missing lib/DBM/Deep.pod file to the MANIFEST
135 - Fixed a poorly-designed test that was failing depending on what Clone::Any
e9b0b5f0 136 was using.
2120a181 137 - All "use 5.6.0;" lines are now "use 5.006_000;" to avoid warnings about
e9b0b5f0 138 unsupported vstrings in 5.9.x
2120a181 139
151e0077 1400.99_03 Jan 23 22:30:00 2007 EST
2120a181 141 - THIS VERSION IS INCOMPATIBLE WITH FILES FROM ALL OTHER PRIOR VERSIONS.
142 - The fileformat changed completely. I will be writing a converter, but
143 it's not there right now. Do NOT expect that this module will
144 correctly detect older versions and handle them sanely. Sanity will be
145 there for 1.00, but we're not there yet, are we?
a8fdabda 146 - Converted to use FileHandle::Fmode to handle filehandle status checks
13ff93d5 147 - Fixed bug with deleting already-deleted items on Win32 (reported by Nigel Sandever)
2120a181 148 - The guts of how transactions work has been rewritten to better handle
149 some edgecases. This required a complete rewrite of the engine.
150 - Freespace management is now in place. It's not perfect, but it's there.
151 - The rewrite of the engine required a rewrite of how first_key/next_key
152 was implemented. This should result in significant speed improvements.
153 - Self-reference has been removed. This means you cannot do:
154 $db->{foo} = { x => 'y' };
155 $db->{bar} = $db->{foo};
156 I hope to be able to return this functionality by 1.00, but I cannot
157 promise anything. To do this properly, it requires refcounting in order
158 to correctly handle deletions and transactions. Once you move away from
159 a simple tree, everything becomes really hard.
a8fdabda 160
c8602313 1610.99_02 Apr 28 05:00:00 2006 Pacific
162 - Added missing file to the MANIFEST
163
1640.99_01 Apr 27 18:00:00 2006 Pacific
d2aefba0 165 - Added explicit dependency on Perl 5.6.0
e06824f8 166 - Digest::MD5 requires 5.6.0
167 - Sub::Uplevel (dep of Test::Exception) requires 5.6.0
95967a5e 168 - Removed error()/clear_error()
d2aefba0 169 - All error-handling is done with die()
e06824f8 170 - Broke out DBM::Deep's code into DBM::Deep::Engine
d2aefba0 171 - Tied variables can no longer be assigned to a DBM::Deep object.
172 - This includes cross-file assignments.
173 - Autovivification now works
174 - This is a consequence of the fact that all assignments are tied.
175 - set_pack() and set_digest() have been removed.
176 - Instead, you will now pass the appropriate values into new()
177 - A pack_size parameter has been added to make 64-bit files easier
c8602313 178 - Transactions now work
95967a5e 179
a8fdabda 1800.983 Apr 10 20:00:00 2006 Pacific
181 - Added patch inspired by Jeff Janes (Thanks!)
182 - Autovivification now works correctly
183 - The following now works correctly
184 my %hash = ( a => 1 );
185 $db->{hash} = \%hash;
186 $hash{b} = 2;
187 cmp_ok( $db->{hash}{b}, '==', 2 );
188 - (RT#18530) - DBM::Deep now plays nicely with -l
189
1900.982 Mar 08 11:00:00 2006 Pacific
191 - Fixed smoketests that were failing on Win32
192 - Added restriction for Perl 5.6.0 or higher.
193 - Digest::MD5 and Sub::Uplevel (dep of Test::Exception) require 5.6+
194
1950.981 Mar 06 11:00:00 2006 Pacific
196 - (RT#17947) - Fixed test that was failing on older Perls
197
9be51a89 1980.98 Feb 28 11:00:00 2006 Pacific
a8bd842e 199 - Added in patch by David Cantrell to allow use of DATA filehandle
bcbe5355 200 - Fixed bug where attempting to export() a structure that used autobless would die
e3cb84dc 201 - Fixed arraytest slowness by localizing $SIG{__DIE__} to prevent Test::Builder's
202 $SIG{__DIE__} from being called. (q.v. http://perldoc.perl.org/functions/eval.html)
4d35d856 203 - More methods have been made private:
204 - root() is now _root()
205 - base_offset() is now _base_offset()
206 - fh() is now _fh()
207 - type() is now _type()
208 - precalc_sizes() is now _precalc_sizes()
e3cb84dc 209
6d049f2c 2100.97 Feb 24 10:00:00 2006 Pacific
e096f245 211 - Reorganization of distribution to a more standard layout
ffed8b01 212 - Migration to Module::Build with EU::MM backwards compatibility
e096f245 213 - Migration of all tests to use Test::More and Test::Exception
214 - Added Devel::Cover report to DBM::Deep POD
215 - Test coverage improved to 89.6% (and climbing)
8b957036 216 - The following methods have been renamed to reflect their private nature:
217 - init() is now _init()
14a3acb6 218 - open() is now _open()
0207d648 219 - close() is now _close()
261d1296 220 - load_tag() is now _load_tag()
221 - index_lookup() is now _index_lookup()
222 - add_bucket() is now _add_bucket()
223 - get_bucket_value() is now _get_bucket_value()
224 - delete_bucket() is now _delete_bucket()
225 - bucket_exists() is now _bucket_exists()
226 - find_bucket_list() is now _find_bucket_list()
227 - traverse_index() is now _traverse_index()
228 - get_next_key() is now _get_next_key()
229 - copy_node() is now _copy_node()
230 - throw_error() is now _throw_error()
217fef02 231 - The various tied classes have been broken out. This means that testing
232 "ref( $obj ) eq 'DBM::Deep'" will now fail. The correct test is
233 "eval { $obj->isa( 'DBM::Deep' ) }".
234 - The various methods like push and delete now have the same return values as
235 the standard builtins.
9d085683 236 - TIEARRAY and TIEHASH now check their parameters more thoroughly
e096f245 237 - Negative indices for arrays works as expected, including throwing the appropriate
238 errors.
26138c00 239 - RT #16877 is fixed (DBM::Deep broken with Perl 5.9.3+).
240 - RT #14893 is fixed (tie() and new() use different parameter lists).
241 - A bug with optimize and threading is fixed.
242 - autobless has received some attention, resulting in a number of bugs fixed.
90f93b43 243 - Removed mode option as it just caused confusion.
3ea61317 244 - Removed volatile option as it is pretty useless (use locking instead)
245 - Locking now implicitly enables autoflush
ffed8b01 246
2470.96 Oct 14 09:55:00 2005 Pacific
248 - Fixed build (OS X hidden files killed it)
249 - You can now pass in an optional filehandle to the constructor
250
2510.95 Oct 12 13:58:00 2005 Pacific
252 - Added optional autobless flag to preserve and restore blessed hashes
253 - Fixed bug where 0 could not be fetched using get_next_key
254 - Fixed bug where tie() constructor didn't accept a hash ref for args
255 - optimize() now preserves user/group/permissions
256 - Errors are now FATAL (meaning it calls die()), unless you set debug flag
257
2580.94 Apr 13 19:00:26 2004 Pacific
259 - Fixed bug reported by John Cardenas (corruption at key level when
260 replace of less data was done on bucket)
261
2620.93 Feb 15 19:53:17 2004 Pacific
263 - Fixed optmize() on Win32 where orig file couldn't be overwritten unless
264 filehandle was closed first. This change introduces a potential race
265 condition when using locking and optmize() on Win32, but it can be
266 fixed in the future using a soft copy instead of Perl's rename().
267
2680.92 Feb 12 19:10:22 2004 Pacific
269 - Fixed bug where passing a reference to a different DBM::Deep object
270 would still result in an internal reference.
271 - Added export() method for recursively extracting hashes/arrays into
272 standard in-memory Perl structures.
273 - Added import() method for recursively importing existing Perl hash/
274 array structures
275 - Fixed bug where optimize() wouldn't work if base level of DB was
276 an array instead of a hash.
277
2780.91 Feb 12 02:30:22 2004 Pacific
279 - Fixed bug with splice() when length of removed section was 0
280 - Updated POD re: circular refs and optimize()
281 - Had to jump version numbers to 0.91 because in previous releases
282 I used only a single digit after the decimal which was confusing
283 the CPAN indexer.
284
2850.10 Feb 11 08:58:35 2004 Pacific
286 - Fixed bug where default file mode was CLEARING files (Thanks Rich!)
287 - Added experimental support for circular references
288 - Fixed bugs in shift(), unshift() and splice() where nested objects
289 in array would be recursively re-stored as basic hashes/arrays
290 - Fixed typos in POD docs
291
2920.9 Feb 10 03:25:48 2004 Pacific
293 - Added Filters for storing/fetching keys/values
294 - Added hook for supplying own hashing algorithm
295 - FIxed some typos in POD docs, added new sections
296
2970.8 Feb 8 02:38:22 2004 Pacific
298 - Renamed to DBM::Deep for CPAN
299 - Added optimize() method for rekindling unused space
300 - Now returning hybrid tie()/OO object from new()
301 - Basic error handling introduced
302 - Added debug mode for printing errors to STDERR
303 - Added TYPE_HASH and TYPE_ARRAY constants for "type" param
304 - Added clone() method for safe copying of objects
305 - Wrote POD documentation
306 - Added set_pack() function for manipulating LONG_SIZE / LONG_PACK
307 - Added aliases for most tied functions for public use
308 - Now setting binmode() on FileHandle for Win32
309 - Added 45 unit tests
310
3110.7 Jan 4 11:31:50 2003 UTC
312 - Renamed to DeepDB
313 - Changed file signature to DPDB (not compatible with older versions)
314 - Converted array length to packed long instead of sprintf()ed string
315
3160.6 Dec 31 15:12:03 2002 UTC
317 - Some misc optimizations for speed
318
3190.5 Oct 18 08:55:29 2002 UTC
320 - support for force_return_next parameter in traverse_index() method for
321 ultra-fast combined key search/removal
322
3230.4 Oct 15 20:07:47 2002 UTC
324 - now making sure filehandle is open for all DB calls
325
3260.3 Oct 3 19:04:13 2002 UTC
327 - fixed bug that could cause corrupted data when using locking
328
3290.2 Aug 6 16:37:32 2002 UTC
330 - Removed base index caching, as it can cause problems when two processes
331 are populating the db at the same time (even with locking)
332
3330.1 Jun 3 08:06:26 2002 UTC
334 - initial release