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

Comparing AnyEvent/Changes (file contents):
Revision 1.192 by root, Wed Oct 1 14:49:23 2008 UTC vs.
Revision 1.309 by root, Wed Jul 29 13:15:55 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.3 3TODO: release a:i 1.0
4 - Glib silently fails when registering a timeout with
5 a negative value, so avoid that.
6 - normalise signal numbers to names when using the ae's
7 signal handling, but do not document this (yet).
8 - pure perl signal emulation did not properly set nonblocking
9 mode on the signal pipe on win32 (this is unlikely to have caused
10 any issues).
11 - AnyEvent::Strict now checks that a signal was specified
12 by name (not name or number).
13 - reduce memory footprint in typical cases by ~50kb by
14 only compiling signal code when necessary.
15
164.881 Tue Jul 28 12:51:53 CEST 2009
17 - work around a bug in local in pre-5.10 perls, causing
18 AnyEvent::Handle to recurse when it should not
19 (analyzed by elmex).
20
214.88 Tue Jul 28 04:04:37 CEST 2009
22 - re-bless the handle into a dummy package after calling
23 AnyEvent::Handle::destroy, so the user does not need to check
24 for errors after every push_write etc.
25 - do not attempt to run t/02_signals.t on obviously broken
26 platforms.
27
284.87 Sun Jul 26 02:06:16 CEST 2009
29 - do not attempt to linger when there is no longer a valid fh.
30 - enforce tls mode to be either accept or connect, do not simply
31 segfault in Net::SSLeay.
32 - AnyEvent::Handle can now call tcp_connect itself (new parameters
33 connect, on_prepare, on_connect and on_connect_error). Updated
34 tutorial accordingly.
35 - add AnyEvent::Impl::Irssi backend.
36
374.86 Mon Jul 20 23:52:29 CEST 2009
38 - since the verbose warning is not enough, explicitly document
39 that versions before 1.33 of Net::SSLeay are not secure.
40 - work around signal handling races in Event and (...) Event::Lib.
41 - try to align signal-race timer to full-second boundaries.
42 - work around Tk not liking negative timeouts.
43 - don't complain of different grades of environmental unfriendlyness
44 in IO::Async.
45
464.85 Sat Jul 18 06:16:14 CEST 2009
47 - nail the signal race problem in perl once and for all
48 (see $AnyEvent::MAX_SIGNAL_LATENCY).
49 - take advantage of Async::Interrupt if it is available.
50 - load Time::HiRes and Guard modules on demand only.
51 - add optional/recommended modules section to AnyEvent
52 documentation.
53 - reduce memory usage considerably (and reduce startup penalty)
54 by not using "strict", "warnings" and "overload" modules.
55 - work around buggy windows/openbsd perls and provide EBADMSG
56 and EPROTO ourselves when missing.
57 - improve perl 5.6 compatibility of the core event loop.
58 - made Net::SSLeay version 1.33 a soft requirement.
59
604.83 Fri Jul 17 16:56:26 CEST 2009
61 - implement AnyEvent::Socket::getprotobyname.
62 - AnyEvent::CondVar's will now detect recursive blocking
63 waits and will croak, as too many people fall into
64 this trap.
65 - AnyEvent::Handle will now call ->destroy on itself after
66 executing the on_error callback, instead of doing some
67 half-baked internal shutdown, for fatal errors.
68 - clarify on_eof behaviour w.r.t. the read queue and
69 on_read callbacks.
70 - ignore some possible spurious wake-ups in tcp_connect.
71
724.82 Sat Jul 11 00:34:55 CEST 2009
73 - POE and Event backends didn't accept some callable objects as
74 callbacks.
75 - use Config module instead of POSIX module to detect signal names
76 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
77 module doesn't even have all POSIX signals :/.
78 - use more workarounds around the many refcnt/corruption bugs in
79 Event::Lib.
80 - work around a race condition in perl's select, causing t/03_child.t
81 to rarely fail.
82
834.81 Thu Jul 9 10:30:30 CEST 2009
84 - AnyEvent::Handle didn't properly diagnose write errors
85 (it expected -1 from syswrite, how lame... :).
86 - support file descriptors in addition to file handles
87 in AnyEvent->io.
88 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
89 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
90 and PERL_ANYEVENT_CA_PATH.
91 - provide a sensible synopsis section for AnyEvent::TLS.
92 - add a "supported backends" section to the manpage.
93 - added simple io watcher test to testsuite, using a
94 portable_socketpair.
95 - tried to improve the stability of the Event::Lib backend,
96 YMMV.
97
984.8 Mon Jul 6 23:45:16 CEST 2009
99 - AnyEvent::DNS did not properly follow CNAME records with
100 uppercase targets.
101 - AnyEvent::DNS would errornously return AAAA records
102 with v4 mapped addresses (a faulty record) as ipv4 addresses,
103 causing AnyEvent::Socket to throw an exception.
104 - added new module AnyEvent::TLS for easier SSL/TLS context
105 creation, with many options including hostname verification,
106 secure default configuration, lots of documentation and,
107 predefined diffie-hellman keys for perfect forward security
108 and much more. get it while it's still fresh!
109 - use AnyEvent::TLS in AnyEvent::Handle for context management.
110 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
111 so AnyEvent::Socket users have smaller memory footprint
112 in the common case.
113 - add AnyEvent::Handle->push_shutdown method.
114 - add an additional error message parameter to AnyEvent::Handle's
115 on_error callback (for TLS, $! is still available).
116 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
117 - make AnyEvent::Handle more robust against early conenction
118 failures (during new), and return C<undef> in that case
119 from the constructor.
120 - AnyEvent::Handle will now try to load only JSON::XS first,
121 then fall back to JSON.
122 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
123 and because it was documented that way.
124
1254.451 Fri Jul 3 00:28:58 CEST 2009
126 - do not clear rbuf when shutting down an AnyEvent::Handle
127 object - doing so breaks AnyEvent::HTTP.
128
1294.45 Mon Jun 29 22:59:26 CEST 2009
130 - a write error could cause AnyEvent::Handle to create
131 an I/O watcher with an undefined $fh.
132 - special-case mapped ipv4 addresses in both
133 AnyEvent::Socket::format_address and parse_address,
134 to treat them just like ipv4 addresses.
135 - updated and overhauled the AnyEvent::Intro doc.
136 - implement AnyEvent::Socket::format_ipv4/ipv6.
137 - slightly speed up portable_pipe/socketpair.
138 - expand condvar begin/end documentation.
139
1404.42 Fri Jun 26 08:32:18 CEST 2009
141 - preliminary and neccesarily incomplete support for IO::Async.
142 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
143 case it was set to IGNORE, to ensure we can catch child statuses
144 even when the calling env acts stupidly.
145 - updated benchmarks with IO::Async, which performs very well.
146
1474.412 Wed Jun 24 01:35:57 CEST 2009
148 - support an "untaint" attribute for AnyEvent::DNS
149 and set it on the default resolver.
150 - implement and document AnyEvent::Impl::Perl::loop.
151 - remove all anyevent-env variables from %ENV
152 when running in tainted mode.
153 - mention and extend the IO::Lambda benchmark.
154
1554.411 Sun Jun 7 18:48:02 CEST 2009
156 - do not try to use F_SETFD on windows.
157
1584.41 Thu May 14 06:40:11 CEST 2009
159 - work around issues in older perls (5.8.0?) when
160 a signal handler is deleted from the %SIG hash.
161 - use POSIX::_exit in child test, to avoid running
162 destructors.
163 - speed up CHLD handling by relying on SIGCHLD being
164 synchronously delivered, even when we roll our own
165 implementation.
166 - AnyEvent::DNS: add the "dname" resource record name
167 for cosmetic reasons.
168
1694.4 Sun Apr 26 20:12:33 CEST 2009
170 - implemented idle watchers, where applicable.
171 - AnyEvent->time died when Event backend was in use.
172 - fix a memleak in the Tk backend.
173 - sped up Tk timer handling.
174 - clip negative "after" values to 0 in AnyEvent::Impl::Event
175 to avoid spamming out warning messages.
176 - fix Qt timers without interval.
177 - avoid Qt zero-timeout special case, in old libqt's.
178
1794.352 Mon Apr 20 16:31:11 CEST 2009
180 - fix AnyEvent::Strict error messages for child watchers.
181 - fix/update Makefile.PL "recommends" versions.
182 - add AnyEvent->now_update.
183
1844.351 Sat Apr 11 07:56:14 CEST 2009
185 - actually make the signal pipe work under win32.
186 - localise $! in signal handler to avoid changing $!, although
187 perl itself does not seem to save/restore errno either.
188 - set the cloexec flag on the signal pipe (normally set by perl too).
189
1904.35 Fri Mar 27 11:48:20 CET 2009
191 - event models relying on AnyEvent's signal watcher emulation did
192 invoke the callback asynchronously, contradicting documentation
193 and causing signals to get lost (this includes AnyEvent's own
194 event loop). AnyEvent now uses the standard pipe trick to make
195 callback execution synchronous to the event loop.
196 - AnyEvent::Handle didn't free TLS context data on DESTROY
197 (patch by Pavel Shaydo).
198 - work around the ever-incompatibly-changing API of MakeMaker.
199 - document that changing global variables without restoring them
200 is a bad idea in Perl (noted by Adam Rosenstein).
201 - AnyEvent::Strict now barfs if ->io is passed a file.
202
2034.34 Thu Feb 12 18:32:45 CET 2009
204 - separately buffer TLS read data, as otherwise the read queue could
205 deadlock as receiving data is not expected while draining the
206 read queue (which cna only happen with TLS).
207 - raise EBADMSG error on JSON decoding errors.
208 - fix some minor manpage bugs (reported by Maximilian Gaß).
209 - speed up select bitmask parsing quite a bit in the pure perl
210 backend.
211 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
212
2134.331 Tue Jan 6 21:07:25 CET 2009
214 - socketpair fails on many vista machines because vista has
215 completely broken accept/getpeername and getsockname functions,
216 so we provide our own socketpair emulation that kind of works
217 (AnyEvent::Util::portable_pipe).
218 - new function: AnyEvent::Util::portable_socketpair.
219 - take advantage of the Guard module if it exists.
220
2214.33 Fri Nov 21 02:35:40 CET 2008
222 - AnyEvent::Strict did errornously flag a fileno of 0 as
223 illegal.
224 - reduce memory usage and slightly speed up the pure perl backend
225 by only storing the file descriptor, not the file handle.
226 - add missing autocork method to AnyEvent::Handle (reported
227 by Adam Rosenstein).
228 - AnyEvent::DNS->resolve errornously documented an $rcode result
229 argument, but there isn't (spotted by Henrik Krohns).
230 - the naked truth about Tk - it's basically unmaintained.
231
2324.32 Mon Nov 3 22:46:32 CET 2008
233 - fix AnyEvent::Socket::resolve_sockaddr to properly support
234 unix sockets again and choose a proper default for
235 the socket type.
236 - fix call to resolve_sockaddr in tcp_connect.
237
2384.31 Thu Oct 30 04:41:48 CET 2008
239 - implemented AnyEvent::Handle->destroy method.
240 - hint about unexpected effects in TLS mode.
241 - speed up AnyEvent::Impl::Perl by using more arrays
242 instead of hashes without the slightest loss of readability :)
243 - work around a perl argument refcounting bug.
244 - with some perl patches applied, AnyEvent now seems leak-free.
245
2464.3 Fri Oct 3 09:18:43 CEST 2008
247 - AnyEvent will now install a no-op signal handler for SIGPIPE,
248 unless one has been installed already.
249 - warn about not loaidng AnyEvent::Impl::POE early enough
250 (patch by Adam Rosenstein).
251 - fixed a great number of bugs and corner cases in AnyEvent::Handle:
4 - fix a bug in AnyEvent::Handle where in SSL connect mode, the 252 - fix a bug in where in SSL connect mode, the
5 client would first wait for some data by the server and 253 client would first wait for some data by the server and
6 otherwise hang. (Reported and analyzed in an absolutely 254 otherwise hang. (reported and analyzed in an absolutely
7 exemplary manner by Adam Rosenstein). 255 exemplary manner by Adam Rosenstein).
8 - fix a bug in AnyEvent::Handle where SSL EOF would not be treated 256 - fix a bug in where SSL EOF would not be treated
9 as stream EOF, putting the connection into a hung state. 257 as stream EOF, putting the connection into a hung state.
10 (Reported and analyzed in an absolutely exemplary manner by Adam 258 (reported and analyzed in an absolutely exemplary manner by Adam
11 Rosenstein). 259 Rosenstein).
12 - fix a potential segfault in AnyEvent::Handle when the TLS context 260 - fix a potential segfault when the TLS context
13 would go missing in a read callback (e.g. due to stoptls) - 261 would go missing in a read callback (e.g. due to stoptls) -
14 Net::SSLeay of course makes no type checking whatsoever. 262 Net::SSLeay of course makes no type checking whatsoever.
15 - AnyEvent::Handle will not stop the read watcher in TLS mode. 263 - AnyEvent::Handle will not stop the read watcher in TLS mode.
16 - AnyEvent::Handle->stoptls will send a TLS close notify 264 - AnyEvent::Handle->stoptls will send a TLS close notify
17 instead of simply destroying the stream now. 265 instead of simply destroying the stream now.
266 - fix a bug where large blocks of data written in TLS mode
267 would not be sent unless triggered by receives.
268 - on_drain will now take the tls write buffer into account.
269 - SSL operations have been streamlined a bit and should be faster
270 now, more to come, though.
271 - added a FAQ section, will document nontrivial issues in there.
18 272
194.234 Mon Sep 29 04:08:13 CEST 2008 2734.234 Mon Sep 29 04:08:13 CEST 2008
20 - fix child watcher documentation: only child exits will be reported, 274 - fix child watcher documentation: only child exits will be reported,
21 no trace events. 275 no trace events.
22 - mention SIGPIPE in AnyEvent::Handle. 276 - mention SIGPIPE in AnyEvent::Handle.

Diff Legend

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