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

Comparing AnyEvent/Changes (file contents):
Revision 1.301 by root, Mon Jul 27 22:49:22 2009 UTC vs.
Revision 1.409 by root, Mon Oct 11 03:41:39 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3FAQ: my program exits beforee doing anything
4FAQ: tcp_connect never calls my callback (context)
5FAQ: high cpu usage with AE::cv->recv
6
7TODO: condvar->new method is documented and should be there.
8TODO: check for -w? somehow?
9TODO: _start might call on_error with ETIMEDOUT??
10cubic reno lp bic highspeed htcp hybla illinois scalable vegas veno westwood yeah
11TODO: document constants
12TODO: test tcp_nodelay/tcp_congestion
13
14AnyEvent::Handle uncaught error: Connection timed out at
15 /usr/local/lib/perl5/site_perl/5.8.8/AnyEvent/HTTP.pm line 473
16 - due to a glitch, AnyEvent's internal getprotobyname wasn't
17 actually used in tcp_server.
18 - support autoloading of read types in unshift_read.
19 - AnyEvent::Strict: do the test for invalid fh arguments differently,
20 resulting in more informative error messages.
21 - clarify that you can actually call push_read etc. in on_read.
22 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
23 requested by Ron Isaacson.
24 - document $AnyEvent::DNS::RESOLVER variable.
25 - incorporated major typo patches by Abhijit Menon-Sen.
26 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
27 undocumented, but who knows when they might come in handy.
28
295.271 Tue Jun 8 12:05:46 CEST 2010
30 - backport to perl 5.8.x.
31
325.27 Sun Jun 6 12:12:05 CEST 2010
33 - postpone differently in AnyEvent::Socket now, as
34 when not, canceling the connection attempt might fail
35 (found by Felix Antonius Wilhelm Ostmann).
36 - explicitly check for non-stream sockets in AE::Handle, too many
37 clueless people fell into the trap of this somehow working.
38 - simplified and reworked the "OTHER MODULES" section.
39 - better/more condvar examples.
40
415.261 Wed Apr 28 16:13:36 CEST 2010
42 - AF_INET6 was not properly used from Socket6 during configuration
43 time (found by Andy Grundman).
44
455.26 Mon Apr 12 04:49:35 CEST 2010
46 - don't generate spurious readyness notifications when select
47 returns EINTR in the pure perl backend (analysed and testcase
48 by Bas Denissen).
49 - give same examples in SYNOPSIS sections of AnyEvent and AE.
50 - provider faster implementations for AE::time/now/now_update for
51 EV and Perl backends.
52 - provide faster AE::cv and AE::time implementations for all
53 backends.
54 - no longer support lower/mixed-case signal names (this was
55 never documented nor universally supported).
56 - some more memory ยต-optimisations, and somewhat less messy
57 code to cope with both AnyEvent and AE APIs.
58
595.251 Sat Mar 13 00:58:21 CET 2010
60 - make test failed if it was run before make install on perl <=
61 5.8.8, and older versions of perl make our live very difficult,
62 so write constants.pl during Makefile.PL time.
63
645.25 Sat Mar 13 00:23:14 CET 2010
65 - fix a race condition in AnyEvent::Handle that would cause
66 a "bio_read: unsupported method" error (in _another_
67 ssl connection) after user code threw an exception.
68 - added AnyEvent::Handle->destroyed method.
69 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
70 - the AnyEvent::Util::fh_nonblocking function worked only
71 by lucky accident on win32.
72 - smaller and faster AnyEvent::Util::fh_nonblocking.
73 - when the (required!) Time::HiRes module is missing, AnyEvent
74 did not fall back to built-in time properly.
75 - do not load Fcntl at runtime, saving memory and loading time.
76 - precompile a number of constants and use them instead of runtime
77 detection and eval.
78 - free detection code after detection and similar memory optimisations.
79 - Perl backend timer interval best effort drift has been improved
80 (same algorithm as EV).
81 - update unicode idna mapping table.
82
835.24 Tue Jan 5 11:39:43 CET 2010
84 - cygwin never reports errors from failed connects. 1.5 just gives
85 you continous readyness and EAGAIN, 1.7 is even more broken and
86 just hangs. work around both issues in a cygwin-specific hack.
87 - improve idle watcher documentation slightly.
88
895.23 Sun Dec 20 23:48:00 CET 2009
90 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
91 - implement punycode_encode/decode, idn_nameprep,
92 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
93 - provide $AE::VERSION.
94 - removed traces of "no strict 'refs'".
95
965.22 Sat Dec 5 03:51:13 CET 2009
97 - downgrade-or-fail in AnyEvent::Handle::push_write, to
98 diagnose encoding failures earlier and more succinctly.
99 (this works around bugs in perl, throwing away encoding info
100 when passing scalar data to extensions).
101 - add more examples to AnyEvent::Socket manpage.
102 - upgrade internal warning set to the same as common::sense 2.03.
103 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
104 (John Beppu).
105 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
106 - add google public dns servers to fallback server set and make sure
107 we load-balance properly between the three sets. also add all
108 fallback dns servers, not just a random one, to each dns config.
109
1105.21 Thu Nov 19 02:48:47 CET 2009
111 - fix a problem where socket constants were called with parameters
112 (spotted by David Friedland).
113 - fork_call never use'd POSIX (reported by Daisuke Maki).
114 - improve perl 5.6 compatibility further (but it still won't work
115 unless you rip out everything but the core).
116 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
117 uses 10MB of RAM, but doesn't open a console window. *sigh*.
118
1195.202 Wed Oct 14 22:35:44 CEST 2009
120 - AnyEvent::DNS would unexpectedly clobber $_ under windows
121 (analysed by Matthias Waldorf).
122 - AnyEvent::Handle::run_cmd can now pass the PID of the
123 newly-created process, which is much less useful than it might
124 sound (based on patch by Yann Kerherve).
125
1265.201 Tue Sep 29 12:09:25 CEST 2009
127 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
128 (reported by Torsten Foertsch).
129 - common::sense 2.0 could cause tcp_server to throw an exception
130 (analysed by elmex).
131
1325.2 Mon Sep 14 07:04:49 CEST 2009
133 - INCOMPATIBLE CHANGE: do no longer support register_read_type
134 and register_write_type in AnyEvent::Handle, instead support
135 package names (the facility was mostly abused).
136 - implement "packagename-as-read/write type" support in
137 AnyEvent::Handle.
138 - AnyEvent::Handle: new options "keepalive" and "oobinline".
139 - oobinline set by default to avoid security issues.
140 - the pure-perl event loop backend wrongly detected times() fallback
141 support (spotted by Pavel Boldin).
142
1435.12 Tue Sep 1 20:26:50 CEST 2009
144 - be more lenient when parsing resolv.conf files, as some people
145 use hashmarks as comment indicator inside directives (reported
146 by Michael S. Fischer).
147 - use same set of warnings as common::sense 2.0.
148 - fix a potential 32 bit overflow issue due to perl having problems
149 with large hex constants in the Perl backend.
150 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
151 by default.
152 - fix some prototypes.
153 - (optionally) require Async::Interrupt 1.02.
154 - replace opendns fallbacks by verizon/level3 public dns servers.
155
1565.112 Fri Aug 21 13:59:04 CEST 2009
157 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
158 using starttls with an illegal/uncreatable tls context.
159 - port to broken win32 perls.
160
1615.111 Sun Aug 16 18:44:36 CEST 2009
162 - perl errornously requires sockaddr_un structures returned by the
163 kernel to have a specific length. work around this bug by padding
164 sockaddr_un structures (found to fail on OS X, as perl only uses
165 the correct code for GNU/Linux, but likely to fail on other
166 OSes as well).
167
1685.11 Wed Aug 12 17:49:37 CEST 2009
169 - fixed a bug in run_cmd with input from filename.
170 - create a json coder object when none is given in write direction
171 to reduce dependency on JSON.
172
1735.1 Tue Aug 11 03:17:32 CEST 2009
174 - tcp_connect now reports the correct error in $!, not an unrelated
175 one (was broken since 4.91).
176 - AE::Handle did not properly clear rtimeout/wtimeout watchers
177 when retrying a tcp_connect.
178 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
179 - fixed wrong documentation in AE::Socket.
180
1815.01 Mon Aug 10 03:16:32 CEST 2009
182 - last release broke EV child watchers :/.
183
1845.0 Sun Aug 9 17:32:09 CEST 2009
185 - officially introduce and document the AE API.
186 - lowering the timeout at runtime did not have immediate effect
187 in AnyEvent::Handle.
188 - AE::Handle now has separate and independent read and write timeout
189 settings.
190 - probe for Guard module when AnyEvent::Util is loaded, not at
191 runtime (Event doesn't like eval "use" from callbacks, and it avoids
192 exporting the probe function).
193 - Impl::IOAsync's io watchers did not use the filehandle emulation
194 code to work around it's watcher limits.
195 - work around IO::Async's problems with overloaded objects.
196 - try to disable wearnings inside IO::Async, it's just too buggy.
197 - do no longer use Event by default, only EV or Perl, Event is too buggy.
198 - hackishly provide fast AE:: interface when EV or Perl is the backend.
199 - inherit VERSION from AnyEvent::VERSION in most modules.
200
2014.91 Thu Aug 6 15:42:45 CEST 2009
202 - AE::Handle::starttls could get out of sync when the read buffer
203 already contains some TLS handshake.
204 - AE::Handle did not properly free the TLS session, causing bigger
205 memory leaks in Net::SSLeay than expected (reported by toaster).
206 - AE::Socket::tcp_connect will now artificially delay invoking
207 the callback to avoid returning after invoking the callback.
208 - convert many internal watcher uses to AE API.
209
2104.9 Sat Aug 1 11:07:01 CEST 2009
211 - Glib silently fails when registering a timeout with
212 a negative value, so avoid that.
213 - call condvar callback immediately when it is set
214 after the condition is already signalled.
215 - check rbuf_max condition only after trying to consume
216 data.
217 - normalise signal numbers to names when using the ae's
218 signal handling, but do not document this (yet).
219 - pure perl signal emulation did not properly set nonblocking
220 mode on the signal pipe on win32 (this is unlikely to have caused
221 any issues).
222 - new module: AnyEvent::Debug.
223 - AnyEvent::Strict now checks that a signal was specified
224 by name (not name or number).
225 - reduce memory footprint in typical cases by ~50kb by
226 only compiling signal code when necessary.
227 - add AnyEvent::Handle->rbuf_max.
228 - grab the AE namespace for future new API, implement stubs for
229 said future/alternative API.
230 - new function: AnyEvent::Socket::format_hostport.
231
2324.881 Tue Jul 28 12:51:53 CEST 2009
233 - work around a bug in local in pre-5.10 perls, causing
234 AnyEvent::Handle to recurse when it should not
235 (analysed by elmex).
236
2374.88 Tue Jul 28 04:04:37 CEST 2009
3 - re-bless the handle into a dummy package after calling 238 - re-bless the handle into a dummy package after calling
4 AnyEvent::Handle::destroy, so the user does not need to check 239 AnyEvent::Handle::destroy, so the user does not need to check
5 for errors after every push_write etc. 240 for errors after every push_write etc.
241 - do not attempt to run t/02_signals.t on obviously broken
242 platforms.
6 243
74.87 Sun Jul 26 02:06:16 CEST 2009 2444.87 Sun Jul 26 02:06:16 CEST 2009
8 - do not attempt to linger when there is no longer a valid fh. 245 - do not attempt to linger when there is no longer a valid fh.
9 - enforce tls mode to be either accept or connect, do not simply 246 - enforce tls mode to be either accept or connect, do not simply
10 segfault in Net::SSLeay. 247 segfault in Net::SSLeay.
… …
18 that versions before 1.33 of Net::SSLeay are not secure. 255 that versions before 1.33 of Net::SSLeay are not secure.
19 - work around signal handling races in Event and (...) Event::Lib. 256 - work around signal handling races in Event and (...) Event::Lib.
20 - try to align signal-race timer to full-second boundaries. 257 - try to align signal-race timer to full-second boundaries.
21 - work around Tk not liking negative timeouts. 258 - work around Tk not liking negative timeouts.
22 - don't complain of different grades of environmental unfriendlyness 259 - don't complain of different grades of environmental unfriendlyness
23 in IO::ASync. 260 in IO::Async.
24 261
254.85 Sat Jul 18 06:16:14 CEST 2009 2624.85 Sat Jul 18 06:16:14 CEST 2009
26 - nail the signal race problem in perl once and for all 263 - nail the signal race problem in perl once and for all
27 (see $AnyEvent::MAX_SIGNAL_LATENCY). 264 (see $AnyEvent::MAX_SIGNAL_LATENCY).
28 - take advantage of Async::Interrupt if it is available. 265 - take advantage of Async::Interrupt if it is available.
… …
2254.3 Fri Oct 3 09:18:43 CEST 2008 4624.3 Fri Oct 3 09:18:43 CEST 2008
226 - AnyEvent will now install a no-op signal handler for SIGPIPE, 463 - AnyEvent will now install a no-op signal handler for SIGPIPE,
227 unless one has been installed already. 464 unless one has been installed already.
228 - warn about not loaidng AnyEvent::Impl::POE early enough 465 - warn about not loaidng AnyEvent::Impl::POE early enough
229 (patch by Adam Rosenstein). 466 (patch by Adam Rosenstein).
230 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 467 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
231 - fix a bug in where in SSL connect mode, the 468 - fix a bug in where in SSL connect mode, the
232 client would first wait for some data by the server and 469 client would first wait for some data by the server and
233 otherwise hang. (reported and analyzed in an absolutely 470 otherwise hang. (reported and analysed in an absolutely
234 exemplary manner by Adam Rosenstein). 471 exemplary manner by Adam Rosenstein).
235 - fix a bug in where SSL EOF would not be treated 472 - fix a bug in where SSL EOF would not be treated
236 as stream EOF, putting the connection into a hung state. 473 as stream EOF, putting the connection into a hung state.
237 (reported and analyzed in an absolutely exemplary manner by Adam 474 (reported and analysed in an absolutely exemplary manner by Adam
238 Rosenstein). 475 Rosenstein).
239 - fix a potential segfault when the TLS context 476 - fix a potential segfault when the TLS context
240 would go missing in a read callback (e.g. due to stoptls) - 477 would go missing in a read callback (e.g. due to stoptls) -
241 Net::SSLeay of course makes no type checking whatsoever. 478 Net::SSLeay of course makes no type checking whatsoever.
242 - AnyEvent::Handle will not stop the read watcher in TLS mode. 479 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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