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

Comparing AnyEvent/Changes (file contents):
Revision 1.17 by root, Fri Nov 24 14:40:13 2006 UTC vs.
Revision 1.121 by root, Tue May 27 04:59:51 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: testsuite, now that we can be sure that a model is available at all times 3 - add some recommends to the META.yml.
4 - added a a regex read type to AnyEvent::Handle.
5 - always croak in on_error in AnyEvent::Handle.
6 - document how to subclass AnyEvent::Handle.
7 - implement AnyEvent::Util::fork_call.
8 - work around _yet_ _another_ windows perl bug in where empty
9 select masks cause errors in the pure perl backend (this
10 workaround cannot be implemented for the other event loops,
11 but EV also works around this).
4 12
52.1 134.05 Mon May 26 19:44:06 CEST 2008
14 - some platforms ignore the rfc and prepend an extra sa_len member to
15 the sockaddr structure, cater for those. also use sockaddr_family,
16 didn't know it was there, but comes in handy for the workaround.
17 - undo the import WIN32 hack.
18
194.04 Mon May 26 08:03:31 CEST 2008
20 - try to work around yet another windows bug: failed connects are reported
21 as if it were out-of-band data. windows users: you suck. Only
22 EV, Glib, Event and the pure perl backend can handle this condition.
23 - optimize write algorithm in AnyEvent::Handle.
24 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
25 - tcp_connect now properly iterates through all targets.
26 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
27 support entirely. also hardcode some AF_INET6 constants for the
28 sake of perl 5.8.
29 - vastly improved nameserver/suffix detection algorithm for windows
30 (still a hack...)
31 - try to use a fallback dns server if no dns servers could be found.
32 - splatter around some local $SIG{__DIE__}'s to work around bad
33 die handlers.
34
354.03 Sun May 25 05:05:57 CEST 2008
36 - work around perl bugs on windows, where perl returns undocumented
37 error codes for sysread, syswrite etc.
38 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
39 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
40 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
41 - updated documentation.
42 - lots of workarounds for perl 5.8 quirks.
43 - implement a more flexible type system for both reads and writes
44 in AnyEvent::Handle.
45 - support IPv6 in tcp_server.
46 - pass local hostname and port to the prepare callback in tcp_server.
47 - make types extendable in AnyEvent::Handle.
48 - croak, not die, in AnyEvent::Handle;
49
504.0 Sat May 24 19:58:08 CEST 2008
51 - added (experimental) AnyEvent::DNS module.
52 - added (experimental) AnyEvent::Socket module.
53 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
54 - make condvar->end call ->send as documented.
55 - make condvar constructor support the documented cb => argument.
56 - overload condvars so you can use them instead of code references.
57 - support more of the POSIX forms instead of a pure dotted quad for
58 verbatim ip addresses.
59 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
60 - try very hard not to freeze on broken windows perls.
61 - make on_eof optional in AnyEvent::Handle.
62 - support service names in tcp_connect and tcp_server.
63 - work around netbsd/openbsd bug causing unix sockets not to
64 become writable until empty.
65 - only make a dummy read for error detection when getpeername
66 returns ENOTCONN.
67 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal
68 override.
69 - new AnyEvent::Guard::cancel.
70
713.5 Sat May 17 23:17:14 CEST 2008
72 - gracefully deal with systems not defining CLOCK_MONOTONIC
73 (as opposed to not supporting the clock itself, which worked in 3.41).
74 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
75 helpers, to replace the ill-designed AnyEvent::Socket class.
76 - added some examples (eg/connect, eg/handle).
77 - added AnyEvent::Util::guard.
78 - AnyEvent::Util now exports most of its functions.
79 - fixed a bug in *_read_chunk, where the callback was called with $self as
80 first and second argument.
81 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
82 - fix manpage condvars examples to use new syntax.
83 - considerably improve inet_aton with EV::ADNS backend to support
84 cname chains (common dns configuration bug, e.g. with www.google.de)
85 and aaaa records if no a record could be found.
86
873.41 Sun May 11 19:53:13 CEST 2008
88 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
89
903.4 Sun May 11 00:29:25 CEST 2008
91 - complete re-implementation of condvars with a hopefully much mroe useful
92 API (backward compatible functions exist but are undocumented).
93 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
94 (still subject to change).
95 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
96 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
97 allows module authors to avoid forcing event loop detection.
98 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
99 - remove coro backends: Coro now provides generic support for AnyEvent
100 via Coro::AnyEvent.
101
1023.3 Mon Apr 28 09:51:06 CEST 2008
103 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
104 - fix a bug in the pure perl backend that kept watchers alive
105 when multiple watchers were registered for the same fh/poll combo.
106 - add a benchmark section showing AnyEvent overhead and comparing
107 the different event loops with each other.
108 - prefer pure perl over tk when autoprobing, as it's about as fast,
109 but doesn't crash with many watchers.
110 - declare Qt support non-experimental.
111 - clarify the confusing section about the file descriptor being
112 kept alive.
113 - document the race between loading of an event module and
114 child processes exiting.
115 - support POE as "backend" (with some caveats, POE is not generic
116 enough, and darn slow).
117 - support Wx and Prima through POE.
118 - optimise perl backend to use 20% less memory and take advantage
119 of typical timeout behaviour. It can now compete with
120 select/poll-based C event loops in most cases (it is usually
121 faster than Event and Glib :).
122 - roughly cut EV memory use in half and increase its speed by 30%,
123 by removing undocumented functionality. Did something similar
124 to other event loops where possible.
125
1263.2 Thu Apr 24 10:10:40 CEST 2008
127 - do not die when anyevent watchers are destroyed while running callbacks
128 in the pureperl backend (could only happen when two watchers are registered
129 for the same fh/poll combo).
130 - support autoloading for child watchers, was broken in all versions.
131 - implement PERL_ANYEVENT_MODEL env variable.
132 - (experimental) implement interface to the (very crashy/buggy) Event::Lib module.
133 - (experimental) implement interface to the Qt module (cannot be autoprobed).
134 - this release is IO:AnyEvent-proof.
135
1363.12 Tue Apr 22 07:11:46 CEST 2008
137 - reinstate AUTHOR section that got lost somehow.
138 - do not hang in the testsuite with badly broken perls (activestate,
139 strawberry...), but instead diagnose the problem and continue.
140 - use INT instead of CHLD in an attempt to support broken windows perls
141 better (this decreases the test quality, unfortunately).
142 - do not send the signal to the process group
143 (no problem for CHLD, bad for INT :).
144
1453.11 Sat Apr 19 06:57:31 CEST 2008
146 - major documentation rework.
147 - document the fact that child watchers only watch for zombies.
148 - fix the child watcher example.
149
1503.1 Wed Apr 16 17:09:01 CEST 2008
151 - work around recurring bugs in Tk by dup'ing filehandles, the
152 only method with good success chances on Tk (the bugs apparently
153 don't get fixed anytime soon).
154 - lift the restriction of only one watcher per fh direction
155 (as the Tk bug workaround also lifts it and only Tk imposed
156 such strong limits).
157 - changed probe order to prefer coro adaptors.
158 - explain why recursion into the event loop is not supported
159 unless the backend supports it (only Coro::EV does without
160 any restrictions...).
161 - add simple manpages for all backend modules.
162
1633.0 Mon Apr 7 21:30:23 CEST 2008
164 - Coro::Signal changed semantics, roll our own, also cleaning
165 up the Coro implementation in general.
166 - rename Coro backend to CoroEvent.
167 - add some decision helping paragraph to the manpage that should
168 help people to decide whether AnyEvent is the right thing for them.
169
1702.9 Mon Jan 28 13:31:54 CET 2008
171 - update for EV 3.0 API changes.
172
1732.8 Sun Nov 25 15:06:03 CET 2007
174 - waitpid can validly return 0. accept this fact of life
175 instead of reporting it to any watchers.
176
1772.7 Fri Nov 23 11:41:14 CET 2007
178 - force use of AnyEvent::Impl::Perl in testsuite, there is
179 too much breakage outside AnyEvent.
180 - deliver signals synchronously in AnyEvent::Impl::Perl
181 or any other event loop that relies on AnyEvents child
182 watcher emulation. *Could* help with hanging testsuite
183 (except when Event or EV are installed).
184
1852.6 Fri Nov 9 20:36:35 CET 2007
186 - fix bug in testsuite.
187 - move EV adaptor modules to AnyEvent.
188 - add Coro+EV adaptor module.
189
1902.55 Tue Nov 6 17:41:32 CET 2007
191 - add EV to the list of supported event models.
192 - do not auto-reset pid watchers, pass pid and status to them.
193 - allow a pid of zero to watch for all children in child watchers.
194
1952.54 Wed Jul 18 17:36:23 CEST 2007
196 - work around a perl bug that results in
197 BEGIN not safe after errors--compilation aborted without
198 any discernible reason or error message by once
199 again not use'ing strict.
200
2012.53 Sun Jul 8 10:51:53 CEST 2007
202 - make Glib interface work again (spotted by elmex).
203
2042.52 Wed Mar 7 18:36:16 CET 2007
205 - child watchers stopped working permanently when
206 all current child watchers were destroyed.
207
2082.51 Mon Dec 11 21:33:24 CET 2006
209 - work around bugs in perl where eval "require Module" returns
210 true even if the module couldn't be loaded :/.
211
2122.5 Mon Dec 11 02:15:28 CET 2006
213 - avoid $AUTOLOAD because many perls corrupt it.
214 - AnyEvent::detect forces autodetection.
215 - implement signal watchers (experimental)
216 (Tk does not support async signals (<= 804.027 at least),
217 so they might get delayed indefinitely on Tk).
218 - implement child watchers (experimental).
219 - moved default condvar implementation into base module, simplifying
220 most implementation modules.
221
2222.1 Fri Nov 24 15:50:48 CET 2006
6 - better docs. 223 - better docs.
7 - simple testsuite. 224 - simple testsuite.
8 - added AnyEvent->one_event method for special purposes. 225 - added AnyEvent->one_event method for special purposes.
9 226
102.0 Wed Nov 1 02:21:30 CET 2006 2272.0 Wed Nov 1 02:21:30 CET 2006
… …
42 - fixed documentation. 259 - fixed documentation.
43 - clarified some corner cases. 260 - clarified some corner cases.
44 - vastly improved documentation with a long example. 261 - vastly improved documentation with a long example.
45 262
460.2 Thu Dec 1 22:18:49 CET 2005 2630.2 Thu Dec 1 22:18:49 CET 2005
47 - check wether a valid method was called, to avoid endless loops. 264 - check whether a valid method was called, to avoid endless loops.
48 - move condvars into their own class (cleaner, safer). 265 - move condvars into their own class (cleaner, safer).
49 - don't die when some modules are not available. 266 - don't die when some modules are not available.
50 - Tk is too broken to allow for words. Will work once Tk works. 267 - Tk is too broken to allow for words. Will work once Tk works.
51 268
520.1 Thu Dec 1 19:44:57 CET 2005 2690.1 Thu Dec 1 19:44:57 CET 2005

Diff Legend

– Removed lines
+ Added lines
< Changed lines
> Changed lines