| 1 | Revision history for AnyEvent::DBI |
1 | Revision history for AnyEvent::DBI |
|
|
2 | |
|
|
3 | 3.0 Tue Aug 15 09:05:30 CEST 2017 |
|
|
4 | - the exec_server constructor argument is now ignored, fork vs. exec |
|
|
5 | is now decided by availability of AnyEvent::Fork::Early or not. |
|
|
6 | - experimental support for fetching (but not setting) statement handle |
|
|
7 | attributes: ->stattr. |
|
|
8 | - use AnyEvent::Fork for process management. |
|
|
9 | - switch to CBOR::XS and Convert::Scalar for further speed and memory |
|
|
10 | improvements - a loop doing "select 0" is now almost twice as fast, |
|
|
11 | more impressive speedups are expected for larger results. |
|
|
12 | This also considerably simplified the code. |
|
|
13 | - t/02_sql_lite.t wrongly assumed sqlite doesn't implement instr(), but |
|
|
14 | nowadays, it does. |
|
|
15 | - added stability canary support. |
|
|
16 | - no longer load Scalar::Util, Fcntl, Socket and POSIX modules. |
|
|
17 | - move child code into it's own module, AnyEvent::DBI::Slave. |
|
|
18 | - no longer load DBI (except in slaves). |
| 2 | |
19 | |
| 3 | 2.3 Tue Apr 2 02:59:12 CEST 2013 |
20 | 2.3 Tue Apr 2 02:59:12 CEST 2013 |
| 4 | - fix begin_work, commit and rollback, that were broken due |
21 | - fix begin_work, commit and rollback, that were broken due |
| 5 | to a precedence problem (analyzed by Jerry Lundström). |
22 | to a precedence problem (analyzed by Jerry Lundström). |
| 6 | - slightly improve tests. |
23 | - slightly improve tests. |
