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

Comparing AnyEvent/Changes (file contents):
Revision 1.223 by root, Fri Mar 27 08:33:41 2009 UTC vs.
Revision 1.326 by root, Sun Aug 9 10:42:10 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: strict checks for file, and valid fd 34.92
44.341 4 - lowering the timeout at runtime did not have immediate effect
5 in AnyEvent::Handle.
6 - AE::Handle now has separate and independent read and write timeout
7 settings.
8 - probe for Guard module when AnyEvent::Util is loaded, not at
9 runtime (Event doesn't like eval "use" from callbacks, and it avoids
10 exporting the probe function).
11 - hackishly provide fast AE:: interface when EV is the backend.
12 - inherit VERSION from AnyEvent::VERSION in most modules.
13
144.91 Thu Aug 6 15:42:45 CEST 2009
15 - AE::Handle::starttls could get out of sync when the read buffer
16 already contains some TLS handshake.
17 - AE::Handle did not properly free the TLS session, causing bigger
18 memory leaks in Net::SSLeay than expected (reported by toaster).
19 - AE::Socket::tcp_connect will now artificially delay invoking
20 the callback to avoid returning after invoking the callback.
21 - convert many internal watcher uses to AE API.
22
234.9 Sat Aug 1 11:07:01 CEST 2009
24 - Glib silently fails when registering a timeout with
25 a negative value, so avoid that.
26 - call condvar callback immediately when it is set
27 after the condition is already signalled.
28 - check rbuf_max condition only after trying to consume
29 data.
30 - normalise signal numbers to names when using the ae's
31 signal handling, but do not document this (yet).
32 - pure perl signal emulation did not properly set nonblocking
33 mode on the signal pipe on win32 (this is unlikely to have caused
34 any issues).
35 - new module: AnyEvent::Debug.
36 - AnyEvent::Strict now checks that a signal was specified
37 by name (not name or number).
38 - reduce memory footprint in typical cases by ~50kb by
39 only compiling signal code when necessary.
40 - add AnyEvent::Handle->rbuf_max.
41 - grab the AE namespace for future new API, implement stubs for
42 said future/alternative API.
43 - new function: AnyEvent::Socket::format_hostport.
44
454.881 Tue Jul 28 12:51:53 CEST 2009
46 - work around a bug in local in pre-5.10 perls, causing
47 AnyEvent::Handle to recurse when it should not
48 (analyzed by elmex).
49
504.88 Tue Jul 28 04:04:37 CEST 2009
51 - re-bless the handle into a dummy package after calling
52 AnyEvent::Handle::destroy, so the user does not need to check
53 for errors after every push_write etc.
54 - do not attempt to run t/02_signals.t on obviously broken
55 platforms.
56
574.87 Sun Jul 26 02:06:16 CEST 2009
58 - do not attempt to linger when there is no longer a valid fh.
59 - enforce tls mode to be either accept or connect, do not simply
60 segfault in Net::SSLeay.
61 - AnyEvent::Handle can now call tcp_connect itself (new parameters
62 connect, on_prepare, on_connect and on_connect_error). Updated
63 tutorial accordingly.
64 - add AnyEvent::Impl::Irssi backend.
65
664.86 Mon Jul 20 23:52:29 CEST 2009
67 - since the verbose warning is not enough, explicitly document
68 that versions before 1.33 of Net::SSLeay are not secure.
69 - work around signal handling races in Event and (...) Event::Lib.
70 - try to align signal-race timer to full-second boundaries.
71 - work around Tk not liking negative timeouts.
72 - don't complain of different grades of environmental unfriendlyness
73 in IO::Async.
74
754.85 Sat Jul 18 06:16:14 CEST 2009
76 - nail the signal race problem in perl once and for all
77 (see $AnyEvent::MAX_SIGNAL_LATENCY).
78 - take advantage of Async::Interrupt if it is available.
79 - load Time::HiRes and Guard modules on demand only.
80 - add optional/recommended modules section to AnyEvent
81 documentation.
82 - reduce memory usage considerably (and reduce startup penalty)
83 by not using "strict", "warnings" and "overload" modules.
84 - work around buggy windows/openbsd perls and provide EBADMSG
85 and EPROTO ourselves when missing.
86 - improve perl 5.6 compatibility of the core event loop.
87 - made Net::SSLeay version 1.33 a soft requirement.
88
894.83 Fri Jul 17 16:56:26 CEST 2009
90 - implement AnyEvent::Socket::getprotobyname.
91 - AnyEvent::CondVar's will now detect recursive blocking
92 waits and will croak, as too many people fall into
93 this trap.
94 - AnyEvent::Handle will now call ->destroy on itself after
95 executing the on_error callback, instead of doing some
96 half-baked internal shutdown, for fatal errors.
97 - clarify on_eof behaviour w.r.t. the read queue and
98 on_read callbacks.
99 - ignore some possible spurious wake-ups in tcp_connect.
100
1014.82 Sat Jul 11 00:34:55 CEST 2009
102 - POE and Event backends didn't accept some callable objects as
103 callbacks.
104 - use Config module instead of POSIX module to detect signal names
105 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
106 module doesn't even have all POSIX signals :/.
107 - use more workarounds around the many refcnt/corruption bugs in
108 Event::Lib.
109 - work around a race condition in perl's select, causing t/03_child.t
110 to rarely fail.
111
1124.81 Thu Jul 9 10:30:30 CEST 2009
113 - AnyEvent::Handle didn't properly diagnose write errors
114 (it expected -1 from syswrite, how lame... :).
115 - support file descriptors in addition to file handles
116 in AnyEvent->io.
117 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
118 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
119 and PERL_ANYEVENT_CA_PATH.
120 - provide a sensible synopsis section for AnyEvent::TLS.
121 - add a "supported backends" section to the manpage.
122 - added simple io watcher test to testsuite, using a
123 portable_socketpair.
124 - tried to improve the stability of the Event::Lib backend,
125 YMMV.
126
1274.8 Mon Jul 6 23:45:16 CEST 2009
128 - AnyEvent::DNS did not properly follow CNAME records with
129 uppercase targets.
130 - AnyEvent::DNS would errornously return AAAA records
131 with v4 mapped addresses (a faulty record) as ipv4 addresses,
132 causing AnyEvent::Socket to throw an exception.
133 - added new module AnyEvent::TLS for easier SSL/TLS context
134 creation, with many options including hostname verification,
135 secure default configuration, lots of documentation and,
136 predefined diffie-hellman keys for perfect forward security
137 and much more. get it while it's still fresh!
138 - use AnyEvent::TLS in AnyEvent::Handle for context management.
139 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
140 so AnyEvent::Socket users have smaller memory footprint
141 in the common case.
142 - add AnyEvent::Handle->push_shutdown method.
143 - add an additional error message parameter to AnyEvent::Handle's
144 on_error callback (for TLS, $! is still available).
145 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
146 - make AnyEvent::Handle more robust against early conenction
147 failures (during new), and return C<undef> in that case
148 from the constructor.
149 - AnyEvent::Handle will now try to load only JSON::XS first,
150 then fall back to JSON.
151 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
152 and because it was documented that way.
153
1544.451 Fri Jul 3 00:28:58 CEST 2009
155 - do not clear rbuf when shutting down an AnyEvent::Handle
156 object - doing so breaks AnyEvent::HTTP.
157
1584.45 Mon Jun 29 22:59:26 CEST 2009
159 - a write error could cause AnyEvent::Handle to create
160 an I/O watcher with an undefined $fh.
161 - special-case mapped ipv4 addresses in both
162 AnyEvent::Socket::format_address and parse_address,
163 to treat them just like ipv4 addresses.
164 - updated and overhauled the AnyEvent::Intro doc.
165 - implement AnyEvent::Socket::format_ipv4/ipv6.
166 - slightly speed up portable_pipe/socketpair.
167 - expand condvar begin/end documentation.
168
1694.42 Fri Jun 26 08:32:18 CEST 2009
170 - preliminary and neccesarily incomplete support for IO::Async.
171 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
172 case it was set to IGNORE, to ensure we can catch child statuses
173 even when the calling env acts stupidly.
174 - updated benchmarks with IO::Async, which performs very well.
175
1764.412 Wed Jun 24 01:35:57 CEST 2009
177 - support an "untaint" attribute for AnyEvent::DNS
178 and set it on the default resolver.
179 - implement and document AnyEvent::Impl::Perl::loop.
180 - remove all anyevent-env variables from %ENV
181 when running in tainted mode.
182 - mention and extend the IO::Lambda benchmark.
183
1844.411 Sun Jun 7 18:48:02 CEST 2009
185 - do not try to use F_SETFD on windows.
186
1874.41 Thu May 14 06:40:11 CEST 2009
188 - work around issues in older perls (5.8.0?) when
189 a signal handler is deleted from the %SIG hash.
190 - use POSIX::_exit in child test, to avoid running
191 destructors.
192 - speed up CHLD handling by relying on SIGCHLD being
193 synchronously delivered, even when we roll our own
194 implementation.
195 - AnyEvent::DNS: add the "dname" resource record name
196 for cosmetic reasons.
197
1984.4 Sun Apr 26 20:12:33 CEST 2009
199 - implemented idle watchers, where applicable.
200 - AnyEvent->time died when Event backend was in use.
201 - fix a memleak in the Tk backend.
202 - sped up Tk timer handling.
203 - clip negative "after" values to 0 in AnyEvent::Impl::Event
204 to avoid spamming out warning messages.
205 - fix Qt timers without interval.
206 - avoid Qt zero-timeout special case, in old libqt's.
207
2084.352 Mon Apr 20 16:31:11 CEST 2009
209 - fix AnyEvent::Strict error messages for child watchers.
210 - fix/update Makefile.PL "recommends" versions.
211 - add AnyEvent->now_update.
212
2134.351 Sat Apr 11 07:56:14 CEST 2009
214 - actually make the signal pipe work under win32.
215 - localise $! in signal handler to avoid changing $!, although
216 perl itself does not seem to save/restore errno either.
217 - set the cloexec flag on the signal pipe (normally set by perl too).
218
2194.35 Fri Mar 27 11:48:20 CET 2009
5 - event models relying on AnyEvent's signal watcher emulation did 220 - event models relying on AnyEvent's signal watcher emulation did
6 invoke the callback asynchronously, contradicting documentation 221 invoke the callback asynchronously, contradicting documentation
7 and causing signals to get lost (this includes AnyEvent's own 222 and causing signals to get lost (this includes AnyEvent's own
8 event loop). AnyEvent now uses the standard pipe trick to make 223 event loop). AnyEvent now uses the standard pipe trick to make
9 callback execution synchronous to the event loop. 224 callback execution synchronous to the event loop.
10 - AnyEvent::Handle didn't free tls context data on DESTROY 225 - AnyEvent::Handle didn't free TLS context data on DESTROY
11 (patch by Pavel Shaydo). 226 (patch by Pavel Shaydo).
12 - work around the ever-incompatibly-changing API of MakeMaker. 227 - work around the ever-incompatibly-changing API of MakeMaker.
13 - document that changing global variables without restoring them 228 - document that changing global variables without restoring them
14 is a bad idea in Perl (noted by Adam Rosenstein). 229 is a bad idea in Perl (noted by Adam Rosenstein).
230 - AnyEvent::Strict now barfs if ->io is passed a file.
15 231
164.34 Thu Feb 12 18:32:45 CET 2009 2324.34 Thu Feb 12 18:32:45 CET 2009
17 - separately buffer TLS read data, as otherwise the read queue could 233 - separately buffer TLS read data, as otherwise the read queue could
18 deadlock as receiving data is not expected while draining the 234 deadlock as receiving data is not expected while draining the
19 read queue (which cna only happen with TLS). 235 read queue (which cna only happen with TLS).

Diff Legend

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