1 |
Revision history for AnyEvent::DBI |
2 |
|
3 |
TODO: why does make test pass even though Slave.pm has a version mismatch? |
4 |
3.04 Mon Apr 23 18:30:40 CEST 2018 |
5 |
- forgot to bump slave version (reported by Felix Ostmann). |
6 |
|
7 |
3.03 Thu Apr 19 06:24:14 CEST 2018 |
8 |
- the main process socket was not actually set to nonblocking mode, |
9 |
causing lockups under load (reported by Felix Ostmann). |
10 |
|
11 |
3.02 Mon Sep 4 13:45:56 CEST 2017 |
12 |
- fixed module version mismatch (reported by Martin Barth). |
13 |
|
14 |
3.01 Sun Aug 27 11:53:26 CEST 2017 |
15 |
- add missing dependency to AnyEvent::Fork. |
16 |
|
17 |
3.0 Tue Aug 15 09:05:30 CEST 2017 |
18 |
- the exec_server constructor argument is now ignored, fork vs. exec |
19 |
is now decided by availability of AnyEvent::Fork::Early or not. |
20 |
- experimental support for fetching (but not setting) statement handle |
21 |
attributes: ->stattr. |
22 |
- use AnyEvent::Fork for process management. |
23 |
- switch to CBOR::XS and Convert::Scalar for further speed and memory |
24 |
improvements - a loop doing "select 0" is now almost twice as fast, |
25 |
more impressive speedups are expected for larger results. |
26 |
This also considerably simplified the code. |
27 |
- t/02_sql_lite.t wrongly assumed sqlite doesn't implement instr(), but |
28 |
nowadays, it does. |
29 |
- added stability canary support. |
30 |
- no longer load Scalar::Util, Fcntl, Socket and POSIX modules. |
31 |
- move child code into it's own module, AnyEvent::DBI::Slave. |
32 |
- no longer load DBI (except in slaves). |
33 |
|
34 |
2.3 Tue Apr 2 02:59:12 CEST 2013 |
35 |
- fix begin_work, commit and rollback, that were broken due |
36 |
to a precedence problem (analyzed by Jerry Lundström). |
37 |
- slightly improve tests. |
38 |
|
39 |
2.2 Thu May 17 04:13:27 CEST 2012 |
40 |
- timeouts didn't trigger (Adam Rosenstein). |
41 |
- do not leave zombie processes (Adam Rosenstein). |
42 |
|
43 |
2.1 Sat Oct 30 22:22:05 CEST 2010 |
44 |
- no longer kill -9 the child, just kill -TERM and close the connection, |
45 |
to avoid database corruption - this also works around a perl bug |
46 |
(global destruction). |
47 |
- convert to the faster AE API - major speedup. |
48 |
- use common::sense. |
49 |
|
50 |
2.0 Mon Jun 29 10:18:58 CEST 2009 |
51 |
- INCOMPATIBLE CHANGE: callbacks now are passed the $dbh |
52 |
as first argument, always. |
53 |
- INCOMPATIBLE CHANGE: really pass rv as third argument to exec |
54 |
callbacks, as documented but not implemented for 1.1. |
55 |
- thanks to viacheslav.t for testing. |
56 |
- major patch by Adam Rosenstein: |
57 |
- add an on_connect callback (for reliable failure notification) |
58 |
- add timeouts to connect and exec. |
59 |
- reaps server processes if you undef your AE::DBI handle and keep |
60 |
running. |
61 |
- option to start the server via exec, otherwise try to emulate |
62 |
close-on-exec behavior. |
63 |
- add the begin_work/commit/rollback methods required to actually use |
64 |
transactions. |
65 |
- add accessors for database handle attibutes such as AutoCommit, |
66 |
RaiseError, etc. |
67 |
- add support for the func() method to interact directly with a driver. |
68 |
- add more/real tests. |
69 |
|
70 |
1.1 Sat Dec 20 22:28:11 CET 2008 |
71 |
- INCOMPATIBLE CHANGE: pass result value in place of %extra |
72 |
to callbacks. |
73 |
- do not call fetchrow_arrayref on non-select statements. |
74 |
- do not die on non-fatal errors, unless there is no error callback |
75 |
(reported by Adam Rosenstein). |
76 |
|
77 |
1.0 Mon Jun 9 16:27:36 CEST 2008 |
78 |
- original version, an AnyEvent::HTTP clone. |