| 1 | Revision history for perl module BDB. |
1 | Revision history for perl module BDB. |
| 2 | |
2 | |
|
|
3 | TODO: DB_RECNUM/DB_SET_RECNO? |
|
|
4 | TODO: find changes for 5.2 - but documentation has been removed from oracles website |
|
|
5 | TODO: libdb5.3 |
|
|
6 | |
|
|
7 | 1.9 Mon Dec 26 02:46:09 CET 2011 |
|
|
8 | - partial: update for libdb 5.2 (to make it compile only). |
|
|
9 | |
|
|
10 | 1.89 Fri Jul 29 11:29:21 CEST 2011 |
|
|
11 | - make it compile and test successfully with libdb 5.1, but |
|
|
12 | do not add any new symbols or functionality. |
|
|
13 | |
|
|
14 | 1.88 Wed Mar 31 02:43:42 CEST 2010 |
|
|
15 | - a double fork partially killed the event pipe (great testcase |
|
|
16 | by dormando). affects IO::AIO, BDB and Async::Interrupt. |
|
|
17 | - suck steve's dick and rename our symbols to avoid colliding |
|
|
18 | with nonstandard identifier spam on the broken os x pseudo-os. |
|
|
19 | affects IO::AIO and BDB. |
|
|
20 | |
|
|
21 | 1.87 Tue Dec 22 03:06:09 CET 2009 |
|
|
22 | - db 4.8 has removed XA resource manager functionality, |
|
|
23 | cope with it. |
|
|
24 | - add LOGVERSION, LOGVERSION_LATCHING, LOGOLDVER |
|
|
25 | symbols. |
|
|
26 | |
|
|
27 | 1.86 Fri Dec 4 16:03:21 CET 2009 |
|
|
28 | - port to broken loser platform. |
|
|
29 | |
|
|
30 | 1.85 Thu Nov 12 03:22:21 CET 2009 |
|
|
31 | - speed up object creation/type checking considerably |
|
|
32 | by caching the stashes. |
|
|
33 | |
|
|
34 | 1.84 Wed Aug 5 13:52:21 CEST 2009 |
|
|
35 | - require common sense. |
|
|
36 | - use schmorp.h for pipe, taking advantage of eventfd. |
|
|
37 | - fix a bug in event pipe creation code on win32. |
|
|
38 | |
|
|
39 | 1.83 Sun Jan 11 01:52:56 CET 2009 |
|
|
40 | - fix db_upgrade, add db_verify. |
|
|
41 | |
|
|
42 | 1.82 Sun Jan 4 11:22:44 CET 2009 |
|
|
43 | - use our own socketpair emulation (stolen from libev), as perl's |
|
|
44 | doesn't work on vista, which returns fantasy port numbers |
|
|
45 | in getpeer/sockname. |
|
|
46 | |
|
|
47 | 1.81 Tue Oct 21 04:20:25 CEST 2008 |
|
|
48 | - check that mutable arguments can be downgraded - this |
|
|
49 | /very indirectly) ensures that output variables won't |
|
|
50 | have spurious utf-8 flags. |
|
|
51 | - fix opmode checdk in db_c_get, db_c_pget again. |
|
|
52 | - allow undef/default arguments for db_compact, for a |
|
|
53 | crash-free tomorrow! |
|
|
54 | - fix a memleak in db_key_range. |
|
|
55 | - started a common pitfdalls section. |
|
|
56 | - added db_env_log_archive. |
|
|
57 | - added db->set_pagesize. |
|
|
58 | - added BDB::IGNORE_LEASE. |
|
|
59 | - reworked documentation structure. |
|
|
60 | |
|
|
61 | 1.801 Mon Sep 29 04:58:03 CEST 2008 |
|
|
62 | - removed ptrace.h and user.h includes thta have been left in |
|
|
63 | erronously. |
|
|
64 | - "port" to buggy windows compilers. again. |
|
|
65 | |
|
|
66 | 1.8 Sat Sep 27 14:01:37 CEST 2008 |
|
|
67 | - non-callback requests will now - by default - be executed in the |
|
|
68 | foreground, which is much faster, but doesn't allow parallel |
|
|
69 | processing of other, event-based, requests. |
|
|
70 | - functions that modify their arguments now croak when a constant |
|
|
71 | is passed to them. |
|
|
72 | - db_c_get/db_c_pget requests did not correctly test the operation flags. |
|
|
73 | - db_pget now correctly updates the pkey argument. |
|
|
74 | - keep references to the passed env/db/etc. objects so they |
|
|
75 | can't get destroyed while requests are still outstanding. |
|
|
76 | - fixed a few minor stack issues. |
|
|
77 | |
|
|
78 | 1.71 Tue Jul 29 05:30:24 CEST 2008 |
| 3 | - add const char * to typemap for possible 5.6 support. |
79 | - add const char * to typemap for possible 5.6 support. |
| 4 | - untested support for db4.3. |
80 | - untested support for db4.3. |
|
|
81 | - berkelydb needs seriously larger stacksizes (on x86 at least) |
|
|
82 | then what the default xthread.h provides. |
| 5 | |
83 | |
| 6 | 1.7 Thu Jul 17 14:09:25 CEST 2008 |
84 | 1.7 Thu Jul 17 14:09:25 CEST 2008 |
| 7 | - use a totally different mechanism to pass callbacks to |
85 | - use a totally different mechanism to pass callbacks to |
| 8 | requests (see manpage). |
86 | requests (see manpage). |
| 9 | - optional arguments before the callback can now be left out. |
87 | - optional arguments before the callback can now be left out. |
