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

Comparing AnyEvent/Changes (file contents):
Revision 1.259 by root, Fri Jul 3 08:51:48 2009 UTC vs.
Revision 1.340 by root, Tue Aug 25 12:05:29 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3ZERO_RETURN - stoptls? 3TODO: pass values to anyevent condvar callback, make this count as recv.
45.112 Fri Aug 21 13:59:04 CEST 2009
5 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
6 using starttls with an illegal/uncreatable tls context.
7 - port to broken win32 perls.
8
95.111 Sun Aug 16 18:44:36 CEST 2009
10 - perl errornously requires sockaddr_un structures returned by the
11 kernel to have a specific length. work around this bug by padding
12 sockaddr_un structures (found to fail on OS X, as perl only uses
13 the correct code for GNU/Linux, but likely to fail on other
14 OSes as well).
15
165.11 Wed Aug 12 17:49:37 CEST 2009
17 - fixed a bug in run_cmd with input from filename.
18 - create a json coder object when none is given in write direction
19 to reduce dependency on JSON.
20
215.1 Tue Aug 11 03:17:32 CEST 2009
22 - tcp_connect now reports the correct error in $!, not an unrelated
23 one (was broken since 4.91).
24 - AE::Handle did not properly clear rtimeout/wtimeout watchers
25 when retrying a tcp_connect.
26 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
27 - fixed wrong documentation in AE::Socket.
28
295.01 Mon Aug 10 03:16:32 CEST 2009
30 - last release broke EV child watchers :/.
31
325.0 Sun Aug 9 17:32:09 CEST 2009
33 - officially introduce and document the AE API.
34 - lowering the timeout at runtime did not have immediate effect
35 in AnyEvent::Handle.
36 - AE::Handle now has separate and independent read and write timeout
37 settings.
38 - probe for Guard module when AnyEvent::Util is loaded, not at
39 runtime (Event doesn't like eval "use" from callbacks, and it avoids
40 exporting the probe function).
41 - Impl::IOAsync's io watchers did not use the filehandle emulation
42 code to work around it's watcher limits.
43 - work around IO::Async's problems with overloaded objects.
44 - try to disable wearnings inside IO::Async, it's just too buggy.
45 - do no longer use Event by default, only EV or Perl, Event is too buggy.
46 - hackishly provide fast AE:: interface when EV or Perl is the backend.
47 - inherit VERSION from AnyEvent::VERSION in most modules.
48
494.91 Thu Aug 6 15:42:45 CEST 2009
50 - AE::Handle::starttls could get out of sync when the read buffer
51 already contains some TLS handshake.
52 - AE::Handle did not properly free the TLS session, causing bigger
53 memory leaks in Net::SSLeay than expected (reported by toaster).
54 - AE::Socket::tcp_connect will now artificially delay invoking
55 the callback to avoid returning after invoking the callback.
56 - convert many internal watcher uses to AE API.
57
584.9 Sat Aug 1 11:07:01 CEST 2009
59 - Glib silently fails when registering a timeout with
60 a negative value, so avoid that.
61 - call condvar callback immediately when it is set
62 after the condition is already signalled.
63 - check rbuf_max condition only after trying to consume
64 data.
65 - normalise signal numbers to names when using the ae's
66 signal handling, but do not document this (yet).
67 - pure perl signal emulation did not properly set nonblocking
68 mode on the signal pipe on win32 (this is unlikely to have caused
69 any issues).
70 - new module: AnyEvent::Debug.
71 - AnyEvent::Strict now checks that a signal was specified
72 by name (not name or number).
73 - reduce memory footprint in typical cases by ~50kb by
74 only compiling signal code when necessary.
75 - add AnyEvent::Handle->rbuf_max.
76 - grab the AE namespace for future new API, implement stubs for
77 said future/alternative API.
78 - new function: AnyEvent::Socket::format_hostport.
79
804.881 Tue Jul 28 12:51:53 CEST 2009
81 - work around a bug in local in pre-5.10 perls, causing
82 AnyEvent::Handle to recurse when it should not
83 (analyzed by elmex).
84
854.88 Tue Jul 28 04:04:37 CEST 2009
86 - re-bless the handle into a dummy package after calling
87 AnyEvent::Handle::destroy, so the user does not need to check
88 for errors after every push_write etc.
89 - do not attempt to run t/02_signals.t on obviously broken
90 platforms.
91
924.87 Sun Jul 26 02:06:16 CEST 2009
93 - do not attempt to linger when there is no longer a valid fh.
94 - enforce tls mode to be either accept or connect, do not simply
95 segfault in Net::SSLeay.
96 - AnyEvent::Handle can now call tcp_connect itself (new parameters
97 connect, on_prepare, on_connect and on_connect_error). Updated
98 tutorial accordingly.
99 - add AnyEvent::Impl::Irssi backend.
100
1014.86 Mon Jul 20 23:52:29 CEST 2009
102 - since the verbose warning is not enough, explicitly document
103 that versions before 1.33 of Net::SSLeay are not secure.
104 - work around signal handling races in Event and (...) Event::Lib.
105 - try to align signal-race timer to full-second boundaries.
106 - work around Tk not liking negative timeouts.
107 - don't complain of different grades of environmental unfriendlyness
108 in IO::Async.
109
1104.85 Sat Jul 18 06:16:14 CEST 2009
111 - nail the signal race problem in perl once and for all
112 (see $AnyEvent::MAX_SIGNAL_LATENCY).
113 - take advantage of Async::Interrupt if it is available.
114 - load Time::HiRes and Guard modules on demand only.
115 - add optional/recommended modules section to AnyEvent
116 documentation.
117 - reduce memory usage considerably (and reduce startup penalty)
118 by not using "strict", "warnings" and "overload" modules.
119 - work around buggy windows/openbsd perls and provide EBADMSG
120 and EPROTO ourselves when missing.
121 - improve perl 5.6 compatibility of the core event loop.
122 - made Net::SSLeay version 1.33 a soft requirement.
123
1244.83 Fri Jul 17 16:56:26 CEST 2009
125 - implement AnyEvent::Socket::getprotobyname.
126 - AnyEvent::CondVar's will now detect recursive blocking
127 waits and will croak, as too many people fall into
128 this trap.
129 - AnyEvent::Handle will now call ->destroy on itself after
130 executing the on_error callback, instead of doing some
131 half-baked internal shutdown, for fatal errors.
132 - clarify on_eof behaviour w.r.t. the read queue and
133 on_read callbacks.
134 - ignore some possible spurious wake-ups in tcp_connect.
135
1364.82 Sat Jul 11 00:34:55 CEST 2009
137 - POE and Event backends didn't accept some callable objects as
138 callbacks.
139 - use Config module instead of POSIX module to detect signal names
140 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
141 module doesn't even have all POSIX signals :/.
142 - use more workarounds around the many refcnt/corruption bugs in
143 Event::Lib.
144 - work around a race condition in perl's select, causing t/03_child.t
145 to rarely fail.
146
1474.81 Thu Jul 9 10:30:30 CEST 2009
148 - AnyEvent::Handle didn't properly diagnose write errors
149 (it expected -1 from syswrite, how lame... :).
150 - support file descriptors in addition to file handles
151 in AnyEvent->io.
152 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
153 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
154 and PERL_ANYEVENT_CA_PATH.
155 - provide a sensible synopsis section for AnyEvent::TLS.
156 - add a "supported backends" section to the manpage.
157 - added simple io watcher test to testsuite, using a
158 portable_socketpair.
159 - tried to improve the stability of the Event::Lib backend,
160 YMMV.
161
1624.8 Mon Jul 6 23:45:16 CEST 2009
163 - AnyEvent::DNS did not properly follow CNAME records with
164 uppercase targets.
165 - AnyEvent::DNS would errornously return AAAA records
166 with v4 mapped addresses (a faulty record) as ipv4 addresses,
167 causing AnyEvent::Socket to throw an exception.
168 - added new module AnyEvent::TLS for easier SSL/TLS context
169 creation, with many options including hostname verification,
170 secure default configuration, lots of documentation and,
171 predefined diffie-hellman keys for perfect forward security
172 and much more. get it while it's still fresh!
173 - use AnyEvent::TLS in AnyEvent::Handle for context management.
4 - load AnyEvent::Handle only on demand in AnyEvent::DNS, 174 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
5 so AnyEvent::Socket users have smaller memory footprint 175 so AnyEvent::Socket users have smaller memory footprint
6 in the common case. 176 in the common case.
7 - add AnyEvent::Handle->push_shutdown method. 177 - add AnyEvent::Handle->push_shutdown method.
8 - add an additional error message parameter to AnyEvent::Handle's 178 - add an additional error message parameter to AnyEvent::Handle's
9 on_error callback (for TLS, $! is still available). 179 on_error callback (for TLS, $! is still available).
180 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
10 - make AnyEvent::Handle more robust against early conenction 181 - make AnyEvent::Handle more robust against early conenction
11 failures (during new), and return C<undef> in that case 182 failures (during new), and return C<undef> in that case
12 from the constructor. 183 from the constructor.
13 - AnyEvent::Handle will now try to load only JSON::XS first, 184 - AnyEvent::Handle will now try to load only JSON::XS first,
14 then fall back to JSON. 185 then fall back to JSON.
186 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
187 and because it was documented that way.
15 188
164.451 Fri Jul 3 00:28:58 CEST 2009 1894.451 Fri Jul 3 00:28:58 CEST 2009
17 - do not clear rbuf when shutting down an AnyEvent::Handle 190 - do not clear rbuf when shutting down an AnyEvent::Handle
18 object - doing so breaks AnyEvent::HTTP. 191 object - doing so breaks AnyEvent::HTTP.
19 192

Diff Legend

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