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

Comparing AnyEvent/Changes (file contents):
Revision 1.288 by root, Fri Jul 17 23:12:20 2009 UTC vs.
Revision 1.348 by root, Thu Sep 3 19:49:12 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: _AVOID_, _VERBOSE_ 3 - INCOMPATIBLE CHANGE: do no longer support register_read_type
4 - improve perl 5.6 compatibility of the core event loop. 4 and register_write_type in AnyEvent::Handle, instead support
5 package names (the facility was mostly abused).
6 - implement "packagename-as-read/write type" support in AnyEvent::Handle.
7 - AnyEvent::Handle: new options "keepalive" and "oobinline".
8 - oobinline set by default to avoid security issues.
9
105.12 Tue Sep 1 20:26:50 CEST 2009
11 - be more lenient when parsing resolv.conf files, as some people
12 use hashmarks as comment indicator inside directives (reported
13 by Michael S. Fischer).
14 - use same set of warnings as common::sense 2.0.
15 - fix a potential 32 bit overflow issue due to perl having problems
16 with large hex constants in the Perl backend.
17 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
18 by default.
19 - fix some prototypes.
20 - (optionally) require Async::Interrupt 1.02.
21 - replace opendns fallbacks by verizon/level3 public dns servers.
22
235.112 Fri Aug 21 13:59:04 CEST 2009
24 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
25 using starttls with an illegal/uncreatable tls context.
26 - port to broken win32 perls.
27
285.111 Sun Aug 16 18:44:36 CEST 2009
29 - perl errornously requires sockaddr_un structures returned by the
30 kernel to have a specific length. work around this bug by padding
31 sockaddr_un structures (found to fail on OS X, as perl only uses
32 the correct code for GNU/Linux, but likely to fail on other
33 OSes as well).
34
355.11 Wed Aug 12 17:49:37 CEST 2009
36 - fixed a bug in run_cmd with input from filename.
37 - create a json coder object when none is given in write direction
38 to reduce dependency on JSON.
39
405.1 Tue Aug 11 03:17:32 CEST 2009
41 - tcp_connect now reports the correct error in $!, not an unrelated
42 one (was broken since 4.91).
43 - AE::Handle did not properly clear rtimeout/wtimeout watchers
44 when retrying a tcp_connect.
45 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
46 - fixed wrong documentation in AE::Socket.
47
485.01 Mon Aug 10 03:16:32 CEST 2009
49 - last release broke EV child watchers :/.
50
515.0 Sun Aug 9 17:32:09 CEST 2009
52 - officially introduce and document the AE API.
53 - lowering the timeout at runtime did not have immediate effect
54 in AnyEvent::Handle.
55 - AE::Handle now has separate and independent read and write timeout
56 settings.
57 - probe for Guard module when AnyEvent::Util is loaded, not at
58 runtime (Event doesn't like eval "use" from callbacks, and it avoids
59 exporting the probe function).
60 - Impl::IOAsync's io watchers did not use the filehandle emulation
61 code to work around it's watcher limits.
62 - work around IO::Async's problems with overloaded objects.
63 - try to disable wearnings inside IO::Async, it's just too buggy.
64 - do no longer use Event by default, only EV or Perl, Event is too buggy.
65 - hackishly provide fast AE:: interface when EV or Perl is the backend.
66 - inherit VERSION from AnyEvent::VERSION in most modules.
67
684.91 Thu Aug 6 15:42:45 CEST 2009
69 - AE::Handle::starttls could get out of sync when the read buffer
70 already contains some TLS handshake.
71 - AE::Handle did not properly free the TLS session, causing bigger
72 memory leaks in Net::SSLeay than expected (reported by toaster).
73 - AE::Socket::tcp_connect will now artificially delay invoking
74 the callback to avoid returning after invoking the callback.
75 - convert many internal watcher uses to AE API.
76
774.9 Sat Aug 1 11:07:01 CEST 2009
78 - Glib silently fails when registering a timeout with
79 a negative value, so avoid that.
80 - call condvar callback immediately when it is set
81 after the condition is already signalled.
82 - check rbuf_max condition only after trying to consume
83 data.
84 - normalise signal numbers to names when using the ae's
85 signal handling, but do not document this (yet).
86 - pure perl signal emulation did not properly set nonblocking
87 mode on the signal pipe on win32 (this is unlikely to have caused
88 any issues).
89 - new module: AnyEvent::Debug.
90 - AnyEvent::Strict now checks that a signal was specified
91 by name (not name or number).
92 - reduce memory footprint in typical cases by ~50kb by
93 only compiling signal code when necessary.
94 - add AnyEvent::Handle->rbuf_max.
95 - grab the AE namespace for future new API, implement stubs for
96 said future/alternative API.
97 - new function: AnyEvent::Socket::format_hostport.
98
994.881 Tue Jul 28 12:51:53 CEST 2009
100 - work around a bug in local in pre-5.10 perls, causing
101 AnyEvent::Handle to recurse when it should not
102 (analyzed by elmex).
103
1044.88 Tue Jul 28 04:04:37 CEST 2009
105 - re-bless the handle into a dummy package after calling
106 AnyEvent::Handle::destroy, so the user does not need to check
107 for errors after every push_write etc.
108 - do not attempt to run t/02_signals.t on obviously broken
109 platforms.
110
1114.87 Sun Jul 26 02:06:16 CEST 2009
112 - do not attempt to linger when there is no longer a valid fh.
113 - enforce tls mode to be either accept or connect, do not simply
114 segfault in Net::SSLeay.
115 - AnyEvent::Handle can now call tcp_connect itself (new parameters
116 connect, on_prepare, on_connect and on_connect_error). Updated
117 tutorial accordingly.
118 - add AnyEvent::Impl::Irssi backend.
119
1204.86 Mon Jul 20 23:52:29 CEST 2009
121 - since the verbose warning is not enough, explicitly document
122 that versions before 1.33 of Net::SSLeay are not secure.
123 - work around signal handling races in Event and (...) Event::Lib.
124 - try to align signal-race timer to full-second boundaries.
125 - work around Tk not liking negative timeouts.
126 - don't complain of different grades of environmental unfriendlyness
127 in IO::Async.
128
1294.85 Sat Jul 18 06:16:14 CEST 2009
5 - nail the signal race problem in perl once and for all 130 - nail the signal race problem in perl once and for all
6 (see $AnyEvent::MAX_SIGNAL_LATENCY). 131 (see $AnyEvent::MAX_SIGNAL_LATENCY).
7 - take advantage of Async::Interrupt if it is available. 132 - take advantage of Async::Interrupt if it is available.
8 - load Time::HiRes and Guard modules on demand only. 133 - load Time::HiRes and Guard modules on demand only.
9 - add optional/recommended modules section to AnyEvent 134 - add optional/recommended modules section to AnyEvent
10 documentation. 135 documentation.
11 - reduce memory usage considerably (and reduce startup penalty) 136 - reduce memory usage considerably (and reduce startup penalty)
12 by not using "strict", "warnings" and "overload" modules. 137 by not using "strict", "warnings" and "overload" modules.
13 - work around buggy windows perls and provide EBADMSG and EPROTO 138 - work around buggy windows/openbsd perls and provide EBADMSG
14 ourselves. 139 and EPROTO ourselves when missing.
140 - improve perl 5.6 compatibility of the core event loop.
141 - made Net::SSLeay version 1.33 a soft requirement.
15 142
164.83 Fri Jul 17 16:56:26 CEST 2009 1434.83 Fri Jul 17 16:56:26 CEST 2009
17 - implement AnyEvent::Socket::getprotobyname. 144 - implement AnyEvent::Socket::getprotobyname.
18 - AnyEvent::CondVar's will now detect recursive blocking 145 - AnyEvent::CondVar's will now detect recursive blocking
19 waits and will croak, as too many people fall into 146 waits and will croak, as too many people fall into

Diff Legend

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