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

Comparing AnyEvent/Changes (file contents):
Revision 1.205 by root, Mon Nov 3 21:49:25 2008 UTC vs.
Revision 1.278 by root, Fri Jul 10 22:35:27 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
34.82 Sat Jul 11 00:34:55 CEST 2009
4 - POE and Event backends didn't accept some callable objects as
5 callbacks.
6 - use Config module instead of POSIX module to detect signal names
7 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
8 module doesn't even have all POSIX signals :/.
9 - use more workarounds around the many refcnt/corruption bugs in
10 Event::Lib.
11 - work around a race condition in perl's select, causing t/03_child.t
12 to rarely fail.
13
144.81 Thu Jul 9 10:30:30 CEST 2009
15 - AnyEvent::Handle didn't properly diagnose write errors
16 (it expected -1 from syswrite, how lame... :).
17 - support file descriptors in addition to file handles
18 in AnyEvent->io.
19 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
20 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
21 and PERL_ANYEVENT_CA_PATH.
22 - provide a sensible synopsis section for AnyEvent::TLS.
23 - add a "supported backends" section to the manpage.
24 - added simple io watcher test to testsuite, using a
25 portable_socketpair.
26 - tried to improve the stability of the Event::Lib backend,
27 YMMV.
28
294.8 Mon Jul 6 23:45:16 CEST 2009
30 - AnyEvent::DNS did not properly follow CNAME records with
31 uppercase targets.
32 - AnyEvent::DNS would errornously return AAAA records
33 with v4 mapped addresses (a faulty record) as ipv4 addresses,
34 causing AnyEvent::Socket to throw an exception.
35 - added new module AnyEvent::TLS for easier SSL/TLS context
36 creation, with many options including hostname verification,
37 secure default configuration, lots of documentation and,
38 predefined diffie-hellman keys for perfect forward security
39 and much more. get it while it's still fresh!
40 - use AnyEvent::TLS in AnyEvent::Handle for context management.
41 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
42 so AnyEvent::Socket users have smaller memory footprint
43 in the common case.
44 - add AnyEvent::Handle->push_shutdown method.
45 - add an additional error message parameter to AnyEvent::Handle's
46 on_error callback (for TLS, $! is still available).
47 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
48 - make AnyEvent::Handle more robust against early conenction
49 failures (during new), and return C<undef> in that case
50 from the constructor.
51 - AnyEvent::Handle will now try to load only JSON::XS first,
52 then fall back to JSON.
53 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
54 and because it was documented that way.
55
564.451 Fri Jul 3 00:28:58 CEST 2009
57 - do not clear rbuf when shutting down an AnyEvent::Handle
58 object - doing so breaks AnyEvent::HTTP.
59
604.45 Mon Jun 29 22:59:26 CEST 2009
61 - a write error could cause AnyEvent::Handle to create
62 an I/O watcher with an undefined $fh.
63 - special-case mapped ipv4 addresses in both
64 AnyEvent::Socket::format_address and parse_address,
65 to treat them just like ipv4 addresses.
66 - updated and overhauled the AnyEvent::Intro doc.
67 - implement AnyEvent::Socket::format_ipv4/ipv6.
68 - slightly speed up portable_pipe/socketpair.
69 - expand condvar begin/end documentation.
70
714.42 Fri Jun 26 08:32:18 CEST 2009
72 - preliminary and neccesarily incomplete support for IO::Async.
73 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
74 case it was set to IGNORE, to ensure we can catch child statuses
75 even when the calling env acts stupidly.
76 - updated benchmarks with IO::Async, which performs very well.
77
784.412 Wed Jun 24 01:35:57 CEST 2009
79 - support an "untaint" attribute for AnyEvent::DNS
80 and set it on the default resolver.
81 - implement and document AnyEvent::Impl::Perl::loop.
82 - remove all anyevent-env variables from %ENV
83 when running in tainted mode.
84 - mention and extend the IO::Lambda benchmark.
85
864.411 Sun Jun 7 18:48:02 CEST 2009
87 - do not try to use F_SETFD on windows.
88
894.41 Thu May 14 06:40:11 CEST 2009
90 - work around issues in older perls (5.8.0?) when
91 a signal handler is deleted from the %SIG hash.
92 - use POSIX::_exit in child test, to avoid running
93 destructors.
94 - speed up CHLD handling by relying on SIGCHLD being
95 synchronously delivered, even when we roll our own
96 implementation.
97 - AnyEvent::DNS: add the "dname" resource record name
98 for cosmetic reasons.
99
1004.4 Sun Apr 26 20:12:33 CEST 2009
101 - implemented idle watchers, where applicable.
102 - AnyEvent->time died when Event backend was in use.
103 - fix a memleak in the Tk backend.
104 - sped up Tk timer handling.
105 - clip negative "after" values to 0 in AnyEvent::Impl::Event
106 to avoid spamming out warning messages.
107 - fix Qt timers without interval.
108 - avoid Qt zero-timeout special case, in old libqt's.
109
1104.352 Mon Apr 20 16:31:11 CEST 2009
111 - fix AnyEvent::Strict error messages for child watchers.
112 - fix/update Makefile.PL "recommends" versions.
113 - add AnyEvent->now_update.
114
1154.351 Sat Apr 11 07:56:14 CEST 2009
116 - actually make the signal pipe work under win32.
117 - localise $! in signal handler to avoid changing $!, although
118 perl itself does not seem to save/restore errno either.
119 - set the cloexec flag on the signal pipe (normally set by perl too).
120
1214.35 Fri Mar 27 11:48:20 CET 2009
122 - event models relying on AnyEvent's signal watcher emulation did
123 invoke the callback asynchronously, contradicting documentation
124 and causing signals to get lost (this includes AnyEvent's own
125 event loop). AnyEvent now uses the standard pipe trick to make
126 callback execution synchronous to the event loop.
127 - AnyEvent::Handle didn't free TLS context data on DESTROY
128 (patch by Pavel Shaydo).
129 - work around the ever-incompatibly-changing API of MakeMaker.
130 - document that changing global variables without restoring them
131 is a bad idea in Perl (noted by Adam Rosenstein).
132 - AnyEvent::Strict now barfs if ->io is passed a file.
133
1344.34 Thu Feb 12 18:32:45 CET 2009
135 - separately buffer TLS read data, as otherwise the read queue could
136 deadlock as receiving data is not expected while draining the
137 read queue (which cna only happen with TLS).
138 - raise EBADMSG error on JSON decoding errors.
139 - fix some minor manpage bugs (reported by Maximilian Gaß).
140 - speed up select bitmask parsing quite a bit in the pure perl
141 backend.
142 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
143
1444.331 Tue Jan 6 21:07:25 CET 2009
145 - socketpair fails on many vista machines because vista has
146 completely broken accept/getpeername and getsockname functions,
147 so we provide our own socketpair emulation that kind of works
148 (AnyEvent::Util::portable_pipe).
149 - new function: AnyEvent::Util::portable_socketpair.
150 - take advantage of the Guard module if it exists.
151
1524.33 Fri Nov 21 02:35:40 CET 2008
153 - AnyEvent::Strict did errornously flag a fileno of 0 as
154 illegal.
155 - reduce memory usage and slightly speed up the pure perl backend
156 by only storing the file descriptor, not the file handle.
157 - add missing autocork method to AnyEvent::Handle (reported
158 by Adam Rosenstein).
159 - AnyEvent::DNS->resolve errornously documented an $rcode result
160 argument, but there isn't (spotted by Henrik Krohns).
161 - the naked truth about Tk - it's basically unmaintained.
2 162
34.32 Mon Nov 3 22:46:32 CET 2008 1634.32 Mon Nov 3 22:46:32 CET 2008
4 - fix AnyEvent::Socket::resolve_sockaddr to properly support 164 - fix AnyEvent::Socket::resolve_sockaddr to properly support
5 unix sockets again and choose a proper default for 165 unix sockets again and choose a proper default for
6 the socket type. 166 the socket type.

Diff Legend

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