ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/AnyEvent/Changes
(Generate patch)

Comparing AnyEvent/Changes (file contents):
Revision 1.24 by root, Sun Jul 8 08:52:10 2007 UTC vs.
Revision 1.76 by elmex, Thu May 15 13:50:23 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
3 - added AnyEvent::Util::tcp_connect helper function
4 - added AnyEvent::Util::listen (and an example eg/listen)
5 - fixed a bug in *_read_chunk, where the callback was called with $self as
6 first and second argument.
7 - added AnyEvent::Util::connect and some examples (eg/connect, eg/handle).
8 - fix manpage condvars examples to use new syntax.
9
103.41 Sun May 11 19:53:13 CEST 2008
11 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
12
133.4 Sun May 11 00:29:25 CEST 2008
14 - complete re-implementation of condvars with a hopefully much mroe useful
15 API (backward compatible functions exist but are undocumented).
16 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
17 (still subject to change).
18 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
19 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
20 allows module authors to avoid forcing event loop detection.
21 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
22 - remove coro backends: Coro now provides generic support for AnyEvent
23 via Coro::AnyEvent.
24
253.3 Mon Apr 28 09:51:06 CEST 2008
26 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
27 - fix a bug in the pure perl backend that kept watchers alive
28 when multiple watchers were registered for the same fh/poll combo.
29 - add a benchmark section showing AnyEvent overhead and comparing
30 the different event loops with each other.
31 - prefer pure perl over tk when autoprobing, as it's about as fast,
32 but doesn't crash with many watchers.
33 - declare Qt support non-experimental.
34 - clarify the confusing section about the file descriptor being
35 kept alive.
36 - document the race between loading of an event module and
37 child processes exiting.
38 - support POE as "backend" (with some caveats, POE is not generic
39 enough, and darn slow).
40 - support Wx and Prima through POE.
41 - optimise perl backend to use 20% less memory and take advantage
42 of typical timeout behaviour. It can now compete with
43 select/poll-based C event loops in most cases (it is usually
44 faster than Event and Glib :).
45 - roughly cut EV memory use in half and increase its speed by 30%,
46 by removing undocumented functionality. Did something similar
47 to other event loops where possible.
48
493.2 Thu Apr 24 10:10:40 CEST 2008
50 - do not die when anyevent watchers are destroyed while running callbacks
51 in the pureperl backend (could only happen when two watchers are registered
52 for the same fh/poll combo).
53 - support autoloading for child watchers, was broken in all versions.
54 - implement PERL_ANYEVENT_MODEL env variable.
55 - (experimental) implement interface to the (very crashy/buggy) Event::Lib module.
56 - (experimental) implement interface to the Qt module (cannot be autoprobed).
57 - this release is IO:AnyEvent-proof.
58
593.12 Tue Apr 22 07:11:46 CEST 2008
60 - reinstate AUTHOR section that got lost somehow.
61 - do not hang in the testsuite with badly broken perls (activestate,
62 strawberry...), but instead diagnose the problem and continue.
63 - use INT instead of CHLD in an attempt to support broken windows perls
64 better (this decreases the test quality, unfortunately).
65 - do not send the signal to the process group
66 (no problem for CHLD, bad for INT :).
67
683.11 Sat Apr 19 06:57:31 CEST 2008
69 - major documentation rework.
70 - document the fact that child watchers only watch for zombies.
71 - fix the child watcher example.
72
733.1 Wed Apr 16 17:09:01 CEST 2008
74 - work around recurring bugs in Tk by dup'ing filehandles, the
75 only method with good success chances on Tk (the bugs apparently
76 don't get fixed anytime soon).
77 - lift the restriction of only one watcher per fh direction
78 (as the Tk bug workaround also lifts it and only Tk imposed
79 such strong limits).
80 - changed probe order to prefer coro adaptors.
81 - explain why recursion into the event loop is not supported
82 unless the backend supports it (only Coro::EV does without
83 any restrictions...).
84 - add simple manpages for all backend modules.
85
863.0 Mon Apr 7 21:30:23 CEST 2008
87 - Coro::Signal changed semantics, roll our own, also cleaning
88 up the Coro implementation in general.
89 - rename Coro backend to CoroEvent.
90 - add some decision helping paragraph to the manpage that should
91 help people to decide whether AnyEvent is the right thing for them.
92
932.9 Mon Jan 28 13:31:54 CET 2008
94 - update for EV 3.0 API changes.
95
962.8 Sun Nov 25 15:06:03 CET 2007
97 - waitpid can validly return 0. accept this fact of life
98 instead of reporting it to any watchers.
99
1002.7 Fri Nov 23 11:41:14 CET 2007
101 - force use of AnyEvent::Impl::Perl in testsuite, there is
102 too much breakage outside AnyEvent.
103 - deliver signals synchronously in AnyEvent::Impl::Perl
104 or any other event loop that relies on AnyEvents child
105 watcher emulation. *Could* help with hanging testsuite
106 (except when Event or EV are installed).
107
1082.6 Fri Nov 9 20:36:35 CET 2007
109 - fix bug in testsuite.
110 - move EV adaptor modules to AnyEvent.
111 - add Coro+EV adaptor module.
112
1132.55 Tue Nov 6 17:41:32 CET 2007
114 - add EV to the list of supported event models.
115 - do not auto-reset pid watchers, pass pid and status to them.
116 - allow a pid of zero to watch for all children in child watchers.
117
1182.54 Wed Jul 18 17:36:23 CEST 2007
119 - work around a perl bug that results in
120 BEGIN not safe after errors--compilation aborted without
121 any discernible reason or error message by once
122 again not use'ing strict.
2 123
32.53 Sun Jul 8 10:51:53 CEST 2007 1242.53 Sun Jul 8 10:51:53 CEST 2007
4 - make Glib interface work again (spotted by elmex). 125 - make Glib interface work again (spotted by elmex).
5 126
62.52 Wed Mar 7 18:36:16 CET 2007 1272.52 Wed Mar 7 18:36:16 CET 2007
61 - fixed documentation. 182 - fixed documentation.
62 - clarified some corner cases. 183 - clarified some corner cases.
63 - vastly improved documentation with a long example. 184 - vastly improved documentation with a long example.
64 185
650.2 Thu Dec 1 22:18:49 CET 2005 1860.2 Thu Dec 1 22:18:49 CET 2005
66 - check wether a valid method was called, to avoid endless loops. 187 - check whether a valid method was called, to avoid endless loops.
67 - move condvars into their own class (cleaner, safer). 188 - move condvars into their own class (cleaner, safer).
68 - don't die when some modules are not available. 189 - don't die when some modules are not available.
69 - Tk is too broken to allow for words. Will work once Tk works. 190 - Tk is too broken to allow for words. Will work once Tk works.
70 191
710.1 Thu Dec 1 19:44:57 CET 2005 1920.1 Thu Dec 1 19:44:57 CET 2005

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines