| 1 | Revision history for Perl extension AnyEvent. |
1 | Revision history for Perl extension AnyEvent. |
|
|
2 | |
|
|
3 | - actually make the signal pipe work under win32. |
|
|
4 | - localise $! in signal handler to avoid changing $!, although |
|
|
5 | perl itself does not seem to save/restore errno either. |
|
|
6 | - set the cloexec flag on the signal pipe (normally set by perl too). |
|
|
7 | |
|
|
8 | 4.35 Fri Mar 27 11:48:20 CET 2009 |
|
|
9 | - event models relying on AnyEvent's signal watcher emulation did |
|
|
10 | invoke the callback asynchronously, contradicting documentation |
|
|
11 | and causing signals to get lost (this includes AnyEvent's own |
|
|
12 | event loop). AnyEvent now uses the standard pipe trick to make |
|
|
13 | callback execution synchronous to the event loop. |
|
|
14 | - AnyEvent::Handle didn't free TLS context data on DESTROY |
|
|
15 | (patch by Pavel Shaydo). |
|
|
16 | - work around the ever-incompatibly-changing API of MakeMaker. |
|
|
17 | - document that changing global variables without restoring them |
|
|
18 | is a bad idea in Perl (noted by Adam Rosenstein). |
|
|
19 | - AnyEvent::Strict now barfs if ->io is passed a file. |
|
|
20 | |
|
|
21 | 4.34 Thu Feb 12 18:32:45 CET 2009 |
|
|
22 | - separately buffer TLS read data, as otherwise the read queue could |
|
|
23 | deadlock as receiving data is not expected while draining the |
|
|
24 | read queue (which cna only happen with TLS). |
|
|
25 | - raise EBADMSG error on JSON decoding errors. |
|
|
26 | - fix some minor manpage bugs (reported by Maximilian Gaß). |
|
|
27 | - speed up select bitmask parsing quite a bit in the pure perl |
|
|
28 | backend. |
|
|
29 | - use CORE::select instead of just select in AnyEvent::Impl::Perl. |
|
|
30 | |
|
|
31 | 4.331 Tue Jan 6 21:07:25 CET 2009 |
|
|
32 | - socketpair fails on many vista machines because vista has |
|
|
33 | completely broken accept/getpeername and getsockname functions, |
|
|
34 | so we provide our own socketpair emulation that kind of works |
|
|
35 | (AnyEvent::Util::portable_pipe). |
|
|
36 | - new function: AnyEvent::Util::portable_socketpair. |
|
|
37 | - take advantage of the Guard module if it exists. |
|
|
38 | |
|
|
39 | 4.33 Fri Nov 21 02:35:40 CET 2008 |
|
|
40 | - AnyEvent::Strict did errornously flag a fileno of 0 as |
|
|
41 | illegal. |
|
|
42 | - reduce memory usage and slightly speed up the pure perl backend |
|
|
43 | by only storing the file descriptor, not the file handle. |
|
|
44 | - add missing autocork method to AnyEvent::Handle (reported |
|
|
45 | by Adam Rosenstein). |
|
|
46 | - AnyEvent::DNS->resolve errornously documented an $rcode result |
|
|
47 | argument, but there isn't (spotted by Henrik Krohns). |
|
|
48 | - the naked truth about Tk - it's basically unmaintained. |
|
|
49 | |
|
|
50 | 4.32 Mon Nov 3 22:46:32 CET 2008 |
|
|
51 | - fix AnyEvent::Socket::resolve_sockaddr to properly support |
|
|
52 | unix sockets again and choose a proper default for |
|
|
53 | the socket type. |
|
|
54 | - fix call to resolve_sockaddr in tcp_connect. |
|
|
55 | |
|
|
56 | 4.31 Thu Oct 30 04:41:48 CET 2008 |
|
|
57 | - implemented AnyEvent::Handle->destroy method. |
|
|
58 | - hint about unexpected effects in TLS mode. |
|
|
59 | - speed up AnyEvent::Impl::Perl by using more arrays |
|
|
60 | instead of hashes without the slightest loss of readability :) |
|
|
61 | - work around a perl argument refcounting bug. |
|
|
62 | - with some perl patches applied, AnyEvent now seems leak-free. |
|
|
63 | |
|
|
64 | 4.3 Fri Oct 3 09:18:43 CEST 2008 |
|
|
65 | - AnyEvent will now install a no-op signal handler for SIGPIPE, |
|
|
66 | unless one has been installed already. |
|
|
67 | - warn about not loaidng AnyEvent::Impl::POE early enough |
|
|
68 | (patch by Adam Rosenstein). |
|
|
69 | - fixed a great number of bugs and corner cases in AnyEvent::Handle: |
|
|
70 | - fix a bug in where in SSL connect mode, the |
|
|
71 | client would first wait for some data by the server and |
|
|
72 | otherwise hang. (reported and analyzed in an absolutely |
|
|
73 | exemplary manner by Adam Rosenstein). |
|
|
74 | - fix a bug in where SSL EOF would not be treated |
|
|
75 | as stream EOF, putting the connection into a hung state. |
|
|
76 | (reported and analyzed in an absolutely exemplary manner by Adam |
|
|
77 | Rosenstein). |
|
|
78 | - fix a potential segfault when the TLS context |
|
|
79 | would go missing in a read callback (e.g. due to stoptls) - |
|
|
80 | Net::SSLeay of course makes no type checking whatsoever. |
|
|
81 | - AnyEvent::Handle will not stop the read watcher in TLS mode. |
|
|
82 | - AnyEvent::Handle->stoptls will send a TLS close notify |
|
|
83 | instead of simply destroying the stream now. |
|
|
84 | - fix a bug where large blocks of data written in TLS mode |
|
|
85 | would not be sent unless triggered by receives. |
|
|
86 | - on_drain will now take the tls write buffer into account. |
|
|
87 | - SSL operations have been streamlined a bit and should be faster |
|
|
88 | now, more to come, though. |
|
|
89 | - added a FAQ section, will document nontrivial issues in there. |
| 2 | |
90 | |
| 3 | 4.234 Mon Sep 29 04:08:13 CEST 2008 |
91 | 4.234 Mon Sep 29 04:08:13 CEST 2008 |
| 4 | - fix child watcher documentation: only child exits will be reported, |
92 | - fix child watcher documentation: only child exits will be reported, |
| 5 | no trace events. |
93 | no trace events. |
| 6 | - mention SIGPIPE in AnyEvent::Handle. |
94 | - mention SIGPIPE in AnyEvent::Handle. |
