| 1 | Revision history for Perl extension AnyEvent. |
1 | Revision history for Perl extension AnyEvent. |
|
|
2 | |
|
|
3 | - undo the import WIN32 hack. |
|
|
4 | - use sockaddr_family instead of relying on structure layout (because, |
|
|
5 | of course, darwin is broken again). |
|
|
6 | |
|
|
7 | 4.04 Mon May 26 08:03:31 CEST 2008 |
|
|
8 | - try to work around yet another windows bug: failed connects are reported |
|
|
9 | as if it were out-of-band data. windows users: you suck. Only |
|
|
10 | EV, Glib, Event and the pure perl backend can handle this condition. |
|
|
11 | - optimize write algorithm in AnyEvent::Handle. |
|
|
12 | - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly. |
|
|
13 | - tcp_connect now properly iterates through all targets. |
|
|
14 | - check wether IPv6 sockets can actually be created, otherwise, disable ipv6 |
|
|
15 | support entirely. also hardcode some AF_INET6 constants for the |
|
|
16 | sake of perl 5.8. |
|
|
17 | - vastly improved nameserver/suffix detection algorithm for windows |
|
|
18 | (still a hack...) |
|
|
19 | - try to use a fallback dns server if no dns servers could be found. |
|
|
20 | - splatter around some local $SIG{__DIE__}'s to work around bad |
|
|
21 | die handlers. |
|
|
22 | |
|
|
23 | 4.03 Sun May 25 05:05:57 CEST 2008 |
|
|
24 | - work around perl bugs on windows, where perl returns undocumented |
|
|
25 | error codes for sysread, syswrite etc. |
|
|
26 | - fix AnyEvent::Handle::unshift_read to really unshift and not push. |
|
|
27 | - AnyEvent::Handle could sometimes stumble over 0-byte writes. |
|
|
28 | - fix unpack_sockaddr, now works for AF_INET6 sockaddr's. |
|
|
29 | - updated documentation. |
|
|
30 | - lots of workarounds for perl 5.8 quirks. |
|
|
31 | - implement a more flexible type system for both reads and writes |
|
|
32 | in AnyEvent::Handle. |
|
|
33 | - support IPv6 in tcp_server. |
|
|
34 | - pass local hostname and port to the prepare callback in tcp_server. |
|
|
35 | - make types extendable in AnyEvent::Handle. |
|
|
36 | - croak, not die, in AnyEvent::Handle; |
|
|
37 | |
|
|
38 | 4.0 Sat May 24 19:58:08 CEST 2008 |
|
|
39 | - added (experimental) AnyEvent::DNS module. |
|
|
40 | - added (experimental) AnyEvent::Socket module. |
|
|
41 | - added (experiemntal) support for TLS/SSL in AnyEvent::Handle. |
|
|
42 | - make condvar->end call ->send as documented. |
|
|
43 | - make condvar constructor support the documented cb => argument. |
|
|
44 | - overload condvars so you can use them instead of code references. |
|
|
45 | - support more of the POSIX forms instead of a pure dotted quad for |
|
|
46 | verbatim ip addresses. |
|
|
47 | - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. |
|
|
48 | - try very hard not to freeze on broken windows perls. |
|
|
49 | - make on_eof optional in AnyEvent::Handle. |
|
|
50 | - support service names in tcp_connect and tcp_server. |
|
|
51 | - work around netbsd/openbsd bug causing unix sockets not to |
|
|
52 | become writable untill empty. |
|
|
53 | - only make a dummy read for error detection when getpeername |
|
|
54 | returns ENOTCONN. |
|
|
55 | - defend AnyEvent::Util::inet_aton against Coro::LWP brutal |
|
|
56 | override. |
|
|
57 | - new AnyEvent::Guard::cancel. |
|
|
58 | |
|
|
59 | 3.5 Sat May 17 23:17:14 CEST 2008 |
|
|
60 | - gracefully deal with systems not defining CLOCK_MONOTONIC |
|
|
61 | (as opposed to not supporting the clock itself, which worked in 3.41). |
|
|
62 | - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server |
|
|
63 | helpers, to replace the ill-designed AnyEvent::Socket class. |
|
|
64 | - added some examples (eg/connect, eg/handle). |
|
|
65 | - added AnyEvent::Util::guard. |
|
|
66 | - AnyEvent::Util now exports most of its functions. |
|
|
67 | - fixed a bug in *_read_chunk, where the callback was called with $self as |
|
|
68 | first and second argument. |
|
|
69 | - simplified and fixed regex handling in AnyEvent::Handle::*_read_line. |
|
|
70 | - fix manpage condvars examples to use new syntax. |
|
|
71 | - considerably improve inet_aton with EV::ADNS backend to support |
|
|
72 | cname chains (common dns configuration bug, e.g. with www.google.de) |
|
|
73 | and aaaa records if no a record could be found. |
|
|
74 | |
|
|
75 | 3.41 Sun May 11 19:53:13 CEST 2008 |
|
|
76 | - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. |
|
|
77 | |
|
|
78 | 3.4 Sun May 11 00:29:25 CEST 2008 |
|
|
79 | - complete re-implementation of condvars with a hopefully much mroe useful |
|
|
80 | API (backward compatible functions exist but are undocumented). |
|
|
81 | - AnyEvent::Handle fully rewritten with a hopefully more powerful API |
|
|
82 | (still subject to change). |
|
|
83 | - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. |
|
|
84 | - provide AnyEvent::post_detect and @AnyEvent::post_detect, which |
|
|
85 | allows module authors to avoid forcing event loop detection. |
|
|
86 | (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example). |
|
|
87 | - remove coro backends: Coro now provides generic support for AnyEvent |
|
|
88 | via Coro::AnyEvent. |
|
|
89 | |
|
|
90 | 3.3 Mon Apr 28 09:51:06 CEST 2008 |
|
|
91 | - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules. |
|
|
92 | - fix a bug in the pure perl backend that kept watchers alive |
|
|
93 | when multiple watchers were registered for the same fh/poll combo. |
|
|
94 | - add a benchmark section showing AnyEvent overhead and comparing |
|
|
95 | the different event loops with each other. |
|
|
96 | - prefer pure perl over tk when autoprobing, as it's about as fast, |
|
|
97 | but doesn't crash with many watchers. |
|
|
98 | - declare Qt support non-experimental. |
|
|
99 | - clarify the confusing section about the file descriptor being |
|
|
100 | kept alive. |
|
|
101 | - document the race between loading of an event module and |
|
|
102 | child processes exiting. |
|
|
103 | - support POE as "backend" (with some caveats, POE is not generic |
|
|
104 | enough, and darn slow). |
|
|
105 | - support Wx and Prima through POE. |
|
|
106 | - optimise perl backend to use 20% less memory and take advantage |
|
|
107 | of typical timeout behaviour. It can now compete with |
|
|
108 | select/poll-based C event loops in most cases (it is usually |
|
|
109 | faster than Event and Glib :). |
|
|
110 | - roughly cut EV memory use in half and increase its speed by 30%, |
|
|
111 | by removing undocumented functionality. Did something similar |
|
|
112 | to other event loops where possible. |
| 2 | |
113 | |
| 3 | 3.2 Thu Apr 24 10:10:40 CEST 2008 |
114 | 3.2 Thu Apr 24 10:10:40 CEST 2008 |
| 4 | - do not die when anyevent watchers are destroyed while running callbacks |
115 | - do not die when anyevent watchers are destroyed while running callbacks |
| 5 | in the pureperl backend (could only happen when two watchers are registered |
116 | in the pureperl backend (could only happen when two watchers are registered |
| 6 | for the same fh/poll combo). |
117 | for the same fh/poll combo). |
