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

Comparing AnyEvent/Changes (file contents):
Revision 1.40 by root, Wed Apr 16 14:04:45 2008 UTC vs.
Revision 1.124 by root, Wed May 28 22:03:31 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: some time-outs in AnyEvent::Handle...
4 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
5 which mirrors their purpose better (old names still available
6 for a while). Also moved AnyEvent::DNS::addr to
7 AnyEvent::Socket::resolve_sockaddr.
8 - fix IPv6 support in pack_sockaddr.
9 - officially un-experimentalise all newly introduced code.
10 - support unix domain sockets everywhere by specifying
11 a host of "unix/" and the pathname as service.
12 - add some recommends to the META.yml.
13 - added a regex read type to AnyEvent::Handle.
14 - added a json read/write type to AnyEvent::Handle.
15 - always croak in on_error in AnyEvent::Handle.
16 - document how to subclass AnyEvent::Handle.
17 - implement AnyEvent::Util::fork_call.
18 - work around _yet_ _another_ windows perl bug in where empty
19 select masks cause errors in the pure perl backend (this
20 workaround cannot be implemented for the other event loops,
21 but EV also works around this).
22 - supply AnyEvent::Socket::address_type and make good use of it.
23 - clarify and fix the Handle documentation w.r.t $self vs. $handle
24 vs. callback arguments.
25
264.05 Mon May 26 19:44:06 CEST 2008
27 - some platforms ignore the rfc and prepend an extra sa_len member to
28 the sockaddr structure, cater for those. also use sockaddr_family,
29 didn't know it was there, but comes in handy for the workaround.
30 - undo the import WIN32 hack.
31
324.04 Mon May 26 08:03:31 CEST 2008
33 - try to work around yet another windows bug: failed connects are reported
34 as if it were out-of-band data. windows users: you suck. Only
35 EV, Glib, Event and the pure perl backend can handle this condition.
36 - optimize write algorithm in AnyEvent::Handle.
37 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
38 - tcp_connect now properly iterates through all targets.
39 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
40 support entirely. also hardcode some AF_INET6 constants for the
41 sake of perl 5.8.
42 - vastly improved nameserver/suffix detection algorithm for windows
43 (still a hack...)
44 - try to use a fallback dns server if no dns servers could be found.
45 - splatter around some local $SIG{__DIE__}'s to work around bad
46 die handlers.
47
484.03 Sun May 25 05:05:57 CEST 2008
49 - work around perl bugs on windows, where perl returns undocumented
50 error codes for sysread, syswrite etc.
51 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
52 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
53 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
54 - updated documentation.
55 - lots of workarounds for perl 5.8 quirks.
56 - implement a more flexible type system for both reads and writes
57 in AnyEvent::Handle.
58 - support IPv6 in tcp_server.
59 - pass local hostname and port to the prepare callback in tcp_server.
60 - make types extendable in AnyEvent::Handle.
61 - croak, not die, in AnyEvent::Handle;
62
634.0 Sat May 24 19:58:08 CEST 2008
64 - added (experimental) AnyEvent::DNS module.
65 - added (experimental) AnyEvent::Socket module.
66 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
67 - make condvar->end call ->send as documented.
68 - make condvar constructor support the documented cb => argument.
69 - overload condvars so you can use them instead of code references.
70 - support more of the POSIX forms instead of a pure dotted quad for
71 verbatim ip addresses.
72 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
73 - try very hard not to freeze on broken windows perls.
74 - make on_eof optional in AnyEvent::Handle.
75 - support service names in tcp_connect and tcp_server.
76 - work around netbsd/openbsd bug causing unix sockets not to
77 become writable until empty.
78 - only make a dummy read for error detection when getpeername
79 returns ENOTCONN.
80 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal
81 override.
82 - new AnyEvent::Guard::cancel.
83
843.5 Sat May 17 23:17:14 CEST 2008
85 - gracefully deal with systems not defining CLOCK_MONOTONIC
86 (as opposed to not supporting the clock itself, which worked in 3.41).
87 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
88 helpers, to replace the ill-designed AnyEvent::Socket class.
89 - added some examples (eg/connect, eg/handle).
90 - added AnyEvent::Util::guard.
91 - AnyEvent::Util now exports most of its functions.
92 - fixed a bug in *_read_chunk, where the callback was called with $self as
93 first and second argument.
94 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
95 - fix manpage condvars examples to use new syntax.
96 - considerably improve inet_aton with EV::ADNS backend to support
97 cname chains (common dns configuration bug, e.g. with www.google.de)
98 and aaaa records if no a record could be found.
99
1003.41 Sun May 11 19:53:13 CEST 2008
101 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
102
1033.4 Sun May 11 00:29:25 CEST 2008
104 - complete re-implementation of condvars with a hopefully much mroe useful
105 API (backward compatible functions exist but are undocumented).
106 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
107 (still subject to change).
108 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
109 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
110 allows module authors to avoid forcing event loop detection.
111 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
112 - remove coro backends: Coro now provides generic support for AnyEvent
113 via Coro::AnyEvent.
114
1153.3 Mon Apr 28 09:51:06 CEST 2008
116 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
117 - fix a bug in the pure perl backend that kept watchers alive
118 when multiple watchers were registered for the same fh/poll combo.
119 - add a benchmark section showing AnyEvent overhead and comparing
120 the different event loops with each other.
121 - prefer pure perl over tk when autoprobing, as it's about as fast,
122 but doesn't crash with many watchers.
123 - declare Qt support non-experimental.
124 - clarify the confusing section about the file descriptor being
125 kept alive.
126 - document the race between loading of an event module and
127 child processes exiting.
128 - support POE as "backend" (with some caveats, POE is not generic
129 enough, and darn slow).
130 - support Wx and Prima through POE.
131 - optimise perl backend to use 20% less memory and take advantage
132 of typical timeout behaviour. It can now compete with
133 select/poll-based C event loops in most cases (it is usually
134 faster than Event and Glib :).
135 - roughly cut EV memory use in half and increase its speed by 30%,
136 by removing undocumented functionality. Did something similar
137 to other event loops where possible.
138
1393.2 Thu Apr 24 10:10:40 CEST 2008
140 - do not die when anyevent watchers are destroyed while running callbacks
141 in the pureperl backend (could only happen when two watchers are registered
142 for the same fh/poll combo).
143 - support autoloading for child watchers, was broken in all versions.
144 - implement PERL_ANYEVENT_MODEL env variable.
145 - (experimental) implement interface to the (very crashy/buggy) Event::Lib module.
146 - (experimental) implement interface to the Qt module (cannot be autoprobed).
147 - this release is IO:AnyEvent-proof.
148
1493.12 Tue Apr 22 07:11:46 CEST 2008
150 - reinstate AUTHOR section that got lost somehow.
151 - do not hang in the testsuite with badly broken perls (activestate,
152 strawberry...), but instead diagnose the problem and continue.
153 - use INT instead of CHLD in an attempt to support broken windows perls
154 better (this decreases the test quality, unfortunately).
155 - do not send the signal to the process group
156 (no problem for CHLD, bad for INT :).
157
1583.11 Sat Apr 19 06:57:31 CEST 2008
159 - major documentation rework.
160 - document the fact that child watchers only watch for zombies.
161 - fix the child watcher example.
162
1633.1 Wed Apr 16 17:09:01 CEST 2008
3 - work around recurring bugs in Tk by dup'ing filehandles, the 164 - work around recurring bugs in Tk by dup'ing filehandles, the
4 only method with good success chances to work on Tk. 165 only method with good success chances on Tk (the bugs apparently
166 don't get fixed anytime soon).
5 - lift the restriction of only one watcher per fh direction. 167 - lift the restriction of only one watcher per fh direction
168 (as the Tk bug workaround also lifts it and only Tk imposed
169 such strong limits).
6 - changed probe ordet to ptefer coro adaptors. 170 - changed probe order to prefer coro adaptors.
171 - explain why recursion into the event loop is not supported
172 unless the backend supports it (only Coro::EV does without
173 any restrictions...).
174 - add simple manpages for all backend modules.
7 175
83.0 Mon Apr 7 21:30:23 CEST 2008 1763.0 Mon Apr 7 21:30:23 CEST 2008
9 - Coro::Signal changed semantics, roll our own, also cleaning 177 - Coro::Signal changed semantics, roll our own, also cleaning
10 up the Coro implementation in general. 178 up the Coro implementation in general.
11 - rename Coro backend to CoroEvent. 179 - rename Coro backend to CoroEvent.
12 - add some decision helping paragraph to the manpage that should 180 - add some decision helping paragraph to the manpage that should
13 help people to decide wether AnyEvent is the right thing for them. 181 help people to decide whether AnyEvent is the right thing for them.
14 182
152.9 Mon Jan 28 13:31:54 CET 2008 1832.9 Mon Jan 28 13:31:54 CET 2008
16 - update for EV 3.0 API changes. 184 - update for EV 3.0 API changes.
17 185
182.8 Sun Nov 25 15:06:03 CET 2007 1862.8 Sun Nov 25 15:06:03 CET 2007
… …
104 - fixed documentation. 272 - fixed documentation.
105 - clarified some corner cases. 273 - clarified some corner cases.
106 - vastly improved documentation with a long example. 274 - vastly improved documentation with a long example.
107 275
1080.2 Thu Dec 1 22:18:49 CET 2005 2760.2 Thu Dec 1 22:18:49 CET 2005
109 - check wether a valid method was called, to avoid endless loops. 277 - check whether a valid method was called, to avoid endless loops.
110 - move condvars into their own class (cleaner, safer). 278 - move condvars into their own class (cleaner, safer).
111 - don't die when some modules are not available. 279 - don't die when some modules are not available.
112 - Tk is too broken to allow for words. Will work once Tk works. 280 - Tk is too broken to allow for words. Will work once Tk works.
113 281
1140.1 Thu Dec 1 19:44:57 CET 2005 2820.1 Thu Dec 1 19:44:57 CET 2005

Diff Legend

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