1 |
root |
1.1 |
Revision history for Perl extension AnyEvent. |
2 |
|
|
|
3 |
root |
1.109 |
TODO: exceptbit |
4 |
|
|
- try to work around yet another windows bug: failed connects are reported |
5 |
|
|
as if it were out-of-band data. windows users: you suck. |
6 |
root |
1.106 |
- properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly. |
7 |
root |
1.104 |
- check wether IPv6 sockets can actually be created, otherwise, disable ipv6 |
8 |
root |
1.107 |
support entirely. also hardcode some AF_INEt6 constants for the |
9 |
|
|
sake of perl 5.8. |
10 |
root |
1.106 |
- vastly improved nameserver/suffix detection algorithm for windows |
11 |
|
|
(still a hack...) |
12 |
|
|
- try to use a fallback dns server if no dns servers could be found. |
13 |
root |
1.108 |
- splatter around some local $SIG{__DIE__}'s to work around bad |
14 |
|
|
die handlers. |
15 |
root |
1.104 |
|
16 |
root |
1.102 |
4.03 Sun May 25 05:05:57 CEST 2008 |
17 |
root |
1.100 |
- work around perl bugs on windows, where perl returns undocumented |
18 |
|
|
error codes for sysread, syswrite etc. |
19 |
root |
1.96 |
- fix AnyEvent::Handle::unshift_read to really unshift and not push. |
20 |
root |
1.97 |
- AnyEvent::Handle could sometimes stumble over 0-byte writes. |
21 |
root |
1.101 |
- fix unpack_sockaddr, now works for AF_INET6 sockaddr's. |
22 |
root |
1.100 |
- updated documentation. |
23 |
root |
1.103 |
- lots of workarounds for perl 5.8 quirks. |
24 |
root |
1.98 |
- implement a more flexible type system for both reads and writes |
25 |
|
|
in AnyEvent::Handle. |
26 |
root |
1.101 |
- support IPv6 in tcp_server. |
27 |
|
|
- pass local hostname and port to the prepare callback in tcp_server. |
28 |
root |
1.99 |
- make types extendable in AnyEvent::Handle. |
29 |
root |
1.100 |
- croak, not die, in AnyEvent::Handle; |
30 |
root |
1.96 |
|
31 |
root |
1.95 |
4.0 Sat May 24 19:58:08 CEST 2008 |
32 |
root |
1.88 |
- added (experimental) AnyEvent::DNS module. |
33 |
|
|
- added (experimental) AnyEvent::Socket module. |
34 |
root |
1.93 |
- added (experiemntal) support for TLS/SSL in AnyEvent::Handle. |
35 |
root |
1.90 |
- make condvar->end call ->send as documented. |
36 |
|
|
- make condvar constructor support the documented cb => argument. |
37 |
root |
1.94 |
- overload condvars so you can use them instead of code references. |
38 |
root |
1.93 |
- support more of the POSIX forms instead of a pure dotted quad for |
39 |
root |
1.89 |
verbatim ip addresses. |
40 |
root |
1.88 |
- AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. |
41 |
root |
1.87 |
- try very hard not to freeze on broken windows perls. |
42 |
root |
1.86 |
- make on_eof optional in AnyEvent::Handle. |
43 |
root |
1.83 |
- support service names in tcp_connect and tcp_server. |
44 |
root |
1.84 |
- work around netbsd/openbsd bug causing unix sockets not to |
45 |
|
|
become writable untill empty. |
46 |
root |
1.81 |
- only make a dummy read for error detection when getpeername |
47 |
|
|
returns ENOTCONN. |
48 |
root |
1.82 |
- defend AnyEvent::Util::inet_aton against Coro::LWP brutal |
49 |
|
|
override. |
50 |
root |
1.91 |
- new AnyEvent::Guard::cancel. |
51 |
root |
1.81 |
|
52 |
root |
1.80 |
3.5 Sat May 17 23:17:14 CEST 2008 |
53 |
root |
1.79 |
- gracefully deal with systems not defining CLOCK_MONOTONIC |
54 |
|
|
(as opposed to not supporting the clock itself, which worked in 3.41). |
55 |
root |
1.77 |
- added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server |
56 |
|
|
helpers, to replace the ill-designed AnyEvent::Socket class. |
57 |
|
|
- added some examples (eg/connect, eg/handle). |
58 |
|
|
- added AnyEvent::Util::guard. |
59 |
|
|
- AnyEvent::Util now exports most of its functions. |
60 |
elmex |
1.74 |
- fixed a bug in *_read_chunk, where the callback was called with $self as |
61 |
|
|
first and second argument. |
62 |
root |
1.77 |
- simplified and fixed regex handling in AnyEvent::Handle::*_read_line. |
63 |
root |
1.72 |
- fix manpage condvars examples to use new syntax. |
64 |
root |
1.78 |
- considerably improve inet_aton with EV::ADNS backend to support |
65 |
|
|
cname chains (common dns configuration bug, e.g. with www.google.de) |
66 |
|
|
and aaaa records if no a record could be found. |
67 |
root |
1.72 |
|
68 |
root |
1.71 |
3.41 Sun May 11 19:53:13 CEST 2008 |
69 |
|
|
- work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. |
70 |
|
|
|
71 |
root |
1.70 |
3.4 Sun May 11 00:29:25 CEST 2008 |
72 |
|
|
- complete re-implementation of condvars with a hopefully much mroe useful |
73 |
|
|
API (backward compatible functions exist but are undocumented). |
74 |
root |
1.69 |
- AnyEvent::Handle fully rewritten with a hopefully more powerful API |
75 |
|
|
(still subject to change). |
76 |
root |
1.63 |
- take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. |
77 |
root |
1.68 |
- provide AnyEvent::post_detect and @AnyEvent::post_detect, which |
78 |
root |
1.67 |
allows module authors to avoid forcing event loop detection. |
79 |
|
|
(used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example). |
80 |
root |
1.66 |
- remove coro backends: Coro now provides generic support for AnyEvent |
81 |
|
|
via Coro::AnyEvent. |
82 |
root |
1.63 |
|
83 |
root |
1.62 |
3.3 Mon Apr 28 09:51:06 CEST 2008 |
84 |
|
|
- added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules. |
85 |
root |
1.55 |
- fix a bug in the pure perl backend that kept watchers alive |
86 |
|
|
when multiple watchers were registered for the same fh/poll combo. |
87 |
root |
1.54 |
- add a benchmark section showing AnyEvent overhead and comparing |
88 |
|
|
the different event loops with each other. |
89 |
root |
1.56 |
- prefer pure perl over tk when autoprobing, as it's about as fast, |
90 |
|
|
but doesn't crash with many watchers. |
91 |
root |
1.62 |
- declare Qt support non-experimental. |
92 |
root |
1.56 |
- clarify the confusing section about the file descriptor being |
93 |
|
|
kept alive. |
94 |
|
|
- document the race between loading of an event module and |
95 |
|
|
child processes exiting. |
96 |
root |
1.53 |
- support POE as "backend" (with some caveats, POE is not generic |
97 |
root |
1.54 |
enough, and darn slow). |
98 |
root |
1.53 |
- support Wx and Prima through POE. |
99 |
root |
1.60 |
- optimise perl backend to use 20% less memory and take advantage |
100 |
root |
1.62 |
of typical timeout behaviour. It can now compete with |
101 |
root |
1.60 |
select/poll-based C event loops in most cases (it is usually |
102 |
|
|
faster than Event and Glib :). |
103 |
root |
1.59 |
- roughly cut EV memory use in half and increase its speed by 30%, |
104 |
root |
1.62 |
by removing undocumented functionality. Did something similar |
105 |
|
|
to other event loops where possible. |
106 |
root |
1.53 |
|
107 |
root |
1.50 |
3.2 Thu Apr 24 10:10:40 CEST 2008 |
108 |
root |
1.51 |
- do not die when anyevent watchers are destroyed while running callbacks |
109 |
|
|
in the pureperl backend (could only happen when two watchers are registered |
110 |
|
|
for the same fh/poll combo). |
111 |
|
|
- support autoloading for child watchers, was broken in all versions. |
112 |
root |
1.47 |
- implement PERL_ANYEVENT_MODEL env variable. |
113 |
root |
1.49 |
- (experimental) implement interface to the (very crashy/buggy) Event::Lib module. |
114 |
|
|
- (experimental) implement interface to the Qt module (cannot be autoprobed). |
115 |
root |
1.52 |
- this release is IO:AnyEvent-proof. |
116 |
root |
1.47 |
|
117 |
root |
1.46 |
3.12 Tue Apr 22 07:11:46 CEST 2008 |
118 |
|
|
- reinstate AUTHOR section that got lost somehow. |
119 |
|
|
- do not hang in the testsuite with badly broken perls (activestate, |
120 |
|
|
strawberry...), but instead diagnose the problem and continue. |
121 |
|
|
- use INT instead of CHLD in an attempt to support broken windows perls |
122 |
|
|
better (this decreases the test quality, unfortunately). |
123 |
|
|
- do not send the signal to the process group |
124 |
|
|
(no problem for CHLD, bad for INT :). |
125 |
|
|
|
126 |
root |
1.45 |
3.11 Sat Apr 19 06:57:31 CEST 2008 |
127 |
|
|
- major documentation rework. |
128 |
|
|
- document the fact that child watchers only watch for zombies. |
129 |
|
|
- fix the child watcher example. |
130 |
|
|
|
131 |
root |
1.43 |
3.1 Wed Apr 16 17:09:01 CEST 2008 |
132 |
root |
1.40 |
- work around recurring bugs in Tk by dup'ing filehandles, the |
133 |
root |
1.43 |
only method with good success chances on Tk (the bugs apparently |
134 |
|
|
don't get fixed anytime soon). |
135 |
root |
1.42 |
- lift the restriction of only one watcher per fh direction |
136 |
|
|
(as the Tk bug workaround also lifts it and only Tk imposed |
137 |
|
|
such strong limits). |
138 |
root |
1.41 |
- changed probe order to prefer coro adaptors. |
139 |
|
|
- explain why recursion into the event loop is not supported |
140 |
|
|
unless the backend supports it (only Coro::EV does without |
141 |
|
|
any restrictions...). |
142 |
|
|
- add simple manpages for all backend modules. |
143 |
root |
1.40 |
|
144 |
root |
1.39 |
3.0 Mon Apr 7 21:30:23 CEST 2008 |
145 |
root |
1.38 |
- Coro::Signal changed semantics, roll our own, also cleaning |
146 |
|
|
up the Coro implementation in general. |
147 |
|
|
- rename Coro backend to CoroEvent. |
148 |
|
|
- add some decision helping paragraph to the manpage that should |
149 |
root |
1.44 |
help people to decide whether AnyEvent is the right thing for them. |
150 |
root |
1.38 |
|
151 |
root |
1.37 |
2.9 Mon Jan 28 13:31:54 CET 2008 |
152 |
|
|
- update for EV 3.0 API changes. |
153 |
|
|
|
154 |
root |
1.36 |
2.8 Sun Nov 25 15:06:03 CET 2007 |
155 |
|
|
- waitpid can validly return 0. accept this fact of life |
156 |
|
|
instead of reporting it to any watchers. |
157 |
|
|
|
158 |
root |
1.35 |
2.7 Fri Nov 23 11:41:14 CET 2007 |
159 |
root |
1.34 |
- force use of AnyEvent::Impl::Perl in testsuite, there is |
160 |
|
|
too much breakage outside AnyEvent. |
161 |
root |
1.35 |
- deliver signals synchronously in AnyEvent::Impl::Perl |
162 |
|
|
or any other event loop that relies on AnyEvents child |
163 |
|
|
watcher emulation. *Could* help with hanging testsuite |
164 |
|
|
(except when Event or EV are installed). |
165 |
root |
1.34 |
|
166 |
root |
1.33 |
2.6 Fri Nov 9 20:36:35 CET 2007 |
167 |
root |
1.32 |
- fix bug in testsuite. |
168 |
|
|
- move EV adaptor modules to AnyEvent. |
169 |
|
|
- add Coro+EV adaptor module. |
170 |
root |
1.26 |
|
171 |
root |
1.31 |
2.55 Tue Nov 6 17:41:32 CET 2007 |
172 |
root |
1.27 |
- add EV to the list of supported event models. |
173 |
root |
1.30 |
- do not auto-reset pid watchers, pass pid and status to them. |
174 |
root |
1.29 |
- allow a pid of zero to watch for all children in child watchers. |
175 |
root |
1.27 |
|
176 |
root |
1.25 |
2.54 Wed Jul 18 17:36:23 CEST 2007 |
177 |
|
|
- work around a perl bug that results in |
178 |
|
|
BEGIN not safe after errors--compilation aborted without |
179 |
|
|
any discernible reason or error message by once |
180 |
|
|
again not use'ing strict. |
181 |
|
|
|
182 |
root |
1.24 |
2.53 Sun Jul 8 10:51:53 CEST 2007 |
183 |
|
|
- make Glib interface work again (spotted by elmex). |
184 |
|
|
|
185 |
root |
1.22 |
2.52 Wed Mar 7 18:36:16 CET 2007 |
186 |
|
|
- child watchers stopped working permanently when |
187 |
root |
1.23 |
all current child watchers were destroyed. |
188 |
root |
1.22 |
|
189 |
root |
1.21 |
2.51 Mon Dec 11 21:33:24 CET 2006 |
190 |
|
|
- work around bugs in perl where eval "require Module" returns |
191 |
|
|
true even if the module couldn't be loaded :/. |
192 |
|
|
|
193 |
root |
1.20 |
2.5 Mon Dec 11 02:15:28 CET 2006 |
194 |
root |
1.19 |
- avoid $AUTOLOAD because many perls corrupt it. |
195 |
root |
1.20 |
- AnyEvent::detect forces autodetection. |
196 |
|
|
- implement signal watchers (experimental) |
197 |
|
|
(Tk does not support async signals (<= 804.027 at least), |
198 |
|
|
so they might get delayed indefinitely on Tk). |
199 |
|
|
- implement child watchers (experimental). |
200 |
|
|
- moved default condvar implementation into base module, simplifying |
201 |
|
|
most implementation modules. |
202 |
root |
1.14 |
|
203 |
root |
1.18 |
2.1 Fri Nov 24 15:50:48 CET 2006 |
204 |
root |
1.16 |
- better docs. |
205 |
|
|
- simple testsuite. |
206 |
root |
1.17 |
- added AnyEvent->one_event method for special purposes. |
207 |
root |
1.16 |
|
208 |
root |
1.15 |
2.0 Wed Nov 1 02:21:30 CET 2006 |
209 |
root |
1.13 |
- INCOMPATIBLE CHANGE: poll can now either be "r" or "w" but not both. |
210 |
|
|
- INCOMPATIBLE CHANGE: io watcher callbacks have no arguments anymore. |
211 |
|
|
- EXPERIMENTAL: |
212 |
|
|
new pure-perl model added, AnyEvent now always finds a suitable |
213 |
|
|
event model. |
214 |
root |
1.12 |
- improved documentation. |
215 |
|
|
- reduced cpu overhead. |
216 |
|
|
|
217 |
root |
1.11 |
1.02 Fri Jan 13 14:15:40 CET 2006 |
218 |
|
|
- add COPYING to clarify license. |
219 |
|
|
|
220 |
root |
1.10 |
1.01 Sun Jan 8 05:48:09 CET 2006 |
221 |
|
|
- correctly cancel Event timers, avoid memory leaks. |
222 |
|
|
|
223 |
root |
1.9 |
1.0 Sun Jan 8 05:40:58 CET 2006 |
224 |
root |
1.8 |
- allow third-party interfaces to be added at runtime. |
225 |
|
|
- fix version number. |
226 |
|
|
|
227 |
root |
1.7 |
0.9 Fri Dec 30 02:25:41 CET 2005 |
228 |
root |
1.6 |
- fixed errors in the documentation. |
229 |
root |
1.7 |
- going back to using $VERSION as "load check", |
230 |
|
|
as suggested by Jörn Reder. Let's see why I stopped using |
231 |
|
|
this method. |
232 |
|
|
- add 'hup' to the list of events to receive for glib |
233 |
|
|
and generate 'rw' events for it, as some versions of |
234 |
|
|
glib seem to require this, while this breaks other |
235 |
|
|
(older) versions of glib. |
236 |
|
|
- implemented PERL_ANYEVENT_VERBOSE env variable. |
237 |
root |
1.6 |
|
238 |
root |
1.5 |
0.3 Sun Dec 4 10:43:33 CET 2005 |
239 |
|
|
- no changes to the code. |
240 |
|
|
- fixed documentation. |
241 |
|
|
- clarified some corner cases. |
242 |
|
|
- vastly improved documentation with a long example. |
243 |
|
|
|
244 |
root |
1.3 |
0.2 Thu Dec 1 22:18:49 CET 2005 |
245 |
root |
1.44 |
- check whether a valid method was called, to avoid endless loops. |
246 |
root |
1.4 |
- move condvars into their own class (cleaner, safer). |
247 |
|
|
- don't die when some modules are not available. |
248 |
|
|
- Tk is too broken to allow for words. Will work once Tk works. |
249 |
root |
1.3 |
|
250 |
root |
1.2 |
0.1 Thu Dec 1 19:44:57 CET 2005 |
251 |
|
|
- spiced up and cleaned up for release. |
252 |
|
|
|
253 |
root |
1.1 |
0.01 Tue Sep 21 18:12:09 CEST 2004 |
254 |
|
|
- original version; copied from Convert::Scalar. |
255 |
|
|
|