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

Comparing AnyEvent/Changes (file contents):
Revision 1.295 by root, Wed Jul 22 05:37:32 2009 UTC vs.
Revision 1.443 by root, Thu Feb 3 00:29:33 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent strict and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict checking for $_ changes?
7
8 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
9 object when it's already gone (analyzed by Ryan Bullock).
10 - update idna algorithm(s) to UTS#46 version 6.0.0.
11 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
12 uts46 says it's an error, and testsuite makes more sense.
13 - remove leading dots in idn_to_ascii - testsuite says to remove,
14 uts46 says it's an error, and testsuite makes more sense.
15 - fixed a bug in idn ignored character processing.
16 - passes the uts46 6.0.0 testsuite except for three cases which
17 apparently are bugs in the testsuite itself.
18 - passes the uts46 6.0.1 testsuite except for eleven cases which
19 apparently are bugs in the testsuite itself.
20
215.31 Sun Jan 23 11:49:19 CET 2011
22 - work around a POE bug where POE would first poll for new events and
23 invoke handlers BEFORE handling the message queue.
24 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
25 - stop_read now does just that in TLS mode. see its
26 documentation for caveats.
27 - update to common::sense 3.4 warning set (some warnings in perl impose
28 an arbitrary 32kb limit on regex matches, which affects
29 AnyEvent::Handle).
30 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
31 - new experimental AE::Handle::resettls method.
32
335.3 Fri Dec 31 05:48:13 CET 2010
34 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
35 issues with IO::Async have been solved in that version, including
36 the availabilty of a default loop!
37 - implement new wbuf_max parameter for AnyEvent::Handle.
38 - added Cocoa::EventLoop backend by Daisuke Murase.
39 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
40 - clarify handle on_prepare documentation.
41 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
42 to avoid leaking them to any child processes.
43 - try to force files into the arch-dependend lib directory, for the
44 half-"turly obsessive compulsive person"(s) that depend on it.
45
465.29 Sun Dec 5 10:49:21 CET 2010
47 - convert EV backend to EV 4.00 API (so better upgrade EV too).
48 - AE::Handle: implement dynamic read_size adjustment, add
49 max_read_size parameter, reduce default read_size to 2048.
50 - add some FAQ entries for servers and/or on_eof vs. on_error.
51 - work around OS bugs (cygwin again) when lingering in AE::Handle.
52 - reduce memory usage slightly when connect was used in AE::Handle.
53 - wrok around more segfaults in Tk.
54 - document tls_ctx => undef as valid.
55 - detect WNOHANG value at compiletime.
56 - include some probably linux-only support for building AnyEvent
57 as part of the perl core.
58 - improve compatibility of t/08_idna.t to perl 5.8.
59 - make AnyEvent::Strict truly optional (in case somebody builds a
60 minimal perl...).
61
625.28 Wed Oct 13 04:14:23 CEST 2010
63 - due to a glitch, AnyEvent's internal getprotobyname wasn't
64 actually used in tcp_server.
65 - implement AnyEvent::CondVar->new method, which was documented but
66 missing.
67 - support autoloading of read types in unshift_read.
68 - AnyEvent::Strict: do the test for invalid fh arguments differently,
69 resulting in more informative error messages.
70 - new AnyEvent::FAQ manpage.
71 - clarify that you can actually call push_read etc. in on_read.
72 - put keys and certs in /tmp, not into the current directory,
73 and document this in the SECURITY CONSIDERATIONS sections
74 of AnyEvent::TLS.
75 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
76 requested by Ron Isaacson.
77 - document $AnyEvent::DNS::RESOLVER variable.
78 - incorporated major typo patches by Abhijit Menon-Sen.
79 - AnyEvent::Handle now croaks when negative timeouts are passed.
80 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
81 undocumented, but who knows when they might come in handy.
82
835.271 Tue Jun 8 12:05:46 CEST 2010
84 - backport to perl 5.8.x.
85
865.27 Sun Jun 6 12:12:05 CEST 2010
87 - postpone differently in AnyEvent::Socket now, as
88 when not, canceling the connection attempt might fail
89 (found by Felix Antonius Wilhelm Ostmann).
90 - explicitly check for non-stream sockets in AE::Handle, too many
91 clueless people fell into the trap of this somehow working.
92 - simplified and reworked the "OTHER MODULES" section.
93 - better/more condvar examples.
94
955.261 Wed Apr 28 16:13:36 CEST 2010
96 - AF_INET6 was not properly used from Socket6 during configuration
97 time (found by Andy Grundman).
98
995.26 Mon Apr 12 04:49:35 CEST 2010
100 - don't generate spurious readyness notifications when select
101 returns EINTR in the pure perl backend (analysed and testcase
102 by Bas Denissen).
103 - give same examples in SYNOPSIS sections of AnyEvent and AE.
104 - provider faster implementations for AE::time/now/now_update for
105 EV and Perl backends.
106 - provide faster AE::cv and AE::time implementations for all
107 backends.
108 - no longer support lower/mixed-case signal names (this was
109 never documented nor universally supported).
110 - some more memory ยต-optimisations, and somewhat less messy
111 code to cope with both AnyEvent and AE APIs.
112
1135.251 Sat Mar 13 00:58:21 CET 2010
114 - make test failed if it was run before make install on perl <=
115 5.8.8, and older versions of perl make our live very difficult,
116 so write constants.pl during Makefile.PL time.
117
1185.25 Sat Mar 13 00:23:14 CET 2010
119 - fix a race condition in AnyEvent::Handle that would cause
120 a "bio_read: unsupported method" error (in _another_
121 ssl connection) after user code threw an exception.
122 - added AnyEvent::Handle->destroyed method.
123 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
124 - the AnyEvent::Util::fh_nonblocking function worked only
125 by lucky accident on win32.
126 - smaller and faster AnyEvent::Util::fh_nonblocking.
127 - when the (required!) Time::HiRes module is missing, AnyEvent
128 did not fall back to built-in time properly.
129 - do not load Fcntl at runtime, saving memory and loading time.
130 - precompile a number of constants and use them instead of runtime
131 detection and eval.
132 - free detection code after detection and similar memory optimisations.
133 - Perl backend timer interval best effort drift has been improved
134 (same algorithm as EV).
135 - update unicode idna mapping table.
136
1375.24 Tue Jan 5 11:39:43 CET 2010
138 - cygwin never reports errors from failed connects. 1.5 just gives
139 you continous readyness and EAGAIN, 1.7 is even more broken and
140 just hangs. work around both issues in a cygwin-specific hack.
141 - improve idle watcher documentation slightly.
142
1435.23 Sun Dec 20 23:48:00 CET 2009
144 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
145 - implement punycode_encode/decode, idn_nameprep,
146 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
147 - provide $AE::VERSION.
148 - removed traces of "no strict 'refs'".
149
1505.22 Sat Dec 5 03:51:13 CET 2009
151 - downgrade-or-fail in AnyEvent::Handle::push_write, to
152 diagnose encoding failures earlier and more succinctly.
153 (this works around bugs in perl, throwing away encoding info
154 when passing scalar data to extensions).
155 - add more examples to AnyEvent::Socket manpage.
156 - upgrade internal warning set to the same as common::sense 2.03.
157 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
158 (John Beppu).
159 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
160 - add google public dns servers to fallback server set and make sure
161 we load-balance properly between the three sets. also add all
162 fallback dns servers, not just a random one, to each dns config.
163
1645.21 Thu Nov 19 02:48:47 CET 2009
165 - fix a problem where socket constants were called with parameters
166 (spotted by David Friedland).
167 - fork_call never use'd POSIX (reported by Daisuke Maki).
168 - improve perl 5.6 compatibility further (but it still won't work
169 unless you rip out everything but the core).
170 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
171 uses 10MB of RAM, but doesn't open a console window. *sigh*.
172
1735.202 Wed Oct 14 22:35:44 CEST 2009
174 - AnyEvent::DNS would unexpectedly clobber $_ under windows
175 (analysed by Matthias Waldorf).
176 - AnyEvent::Handle::run_cmd can now pass the PID of the
177 newly-created process, which is much less useful than it might
178 sound (based on patch by Yann Kerherve).
179
1805.201 Tue Sep 29 12:09:25 CEST 2009
181 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
182 (reported by Torsten Foertsch).
183 - common::sense 2.0 could cause tcp_server to throw an exception
184 (analysed by elmex).
185
1865.2 Mon Sep 14 07:04:49 CEST 2009
187 - INCOMPATIBLE CHANGE: do no longer support register_read_type
188 and register_write_type in AnyEvent::Handle, instead support
189 package names (the facility was mostly abused).
190 - implement "packagename-as-read/write type" support in
191 AnyEvent::Handle.
192 - AnyEvent::Handle: new options "keepalive" and "oobinline".
193 - oobinline set by default to avoid security issues.
194 - the pure-perl event loop backend wrongly detected times() fallback
195 support (spotted by Pavel Boldin).
196
1975.12 Tue Sep 1 20:26:50 CEST 2009
198 - be more lenient when parsing resolv.conf files, as some people
199 use hashmarks as comment indicator inside directives (reported
200 by Michael S. Fischer).
201 - use same set of warnings as common::sense 2.0.
202 - fix a potential 32 bit overflow issue due to perl having problems
203 with large hex constants in the Perl backend.
204 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
205 by default.
206 - fix some prototypes.
207 - (optionally) require Async::Interrupt 1.02.
208 - replace opendns fallbacks by verizon/level3 public dns servers.
209
2105.112 Fri Aug 21 13:59:04 CEST 2009
211 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
212 using starttls with an illegal/uncreatable tls context.
213 - port to broken win32 perls.
214
2155.111 Sun Aug 16 18:44:36 CEST 2009
216 - perl errornously requires sockaddr_un structures returned by the
217 kernel to have a specific length. work around this bug by padding
218 sockaddr_un structures (found to fail on OS X, as perl only uses
219 the correct code for GNU/Linux, but likely to fail on other
220 OSes as well).
221
2225.11 Wed Aug 12 17:49:37 CEST 2009
223 - fixed a bug in run_cmd with input from filename.
224 - create a json coder object when none is given in write direction
225 to reduce dependency on JSON.
226
2275.1 Tue Aug 11 03:17:32 CEST 2009
228 - tcp_connect now reports the correct error in $!, not an unrelated
229 one (was broken since 4.91).
230 - AE::Handle did not properly clear rtimeout/wtimeout watchers
231 when retrying a tcp_connect.
232 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
233 - fixed wrong documentation in AE::Socket.
234
2355.01 Mon Aug 10 03:16:32 CEST 2009
236 - last release broke EV child watchers :/.
237
2385.0 Sun Aug 9 17:32:09 CEST 2009
239 - officially introduce and document the AE API.
240 - lowering the timeout at runtime did not have immediate effect
241 in AnyEvent::Handle.
242 - AE::Handle now has separate and independent read and write timeout
243 settings.
244 - probe for Guard module when AnyEvent::Util is loaded, not at
245 runtime (Event doesn't like eval "use" from callbacks, and it avoids
246 exporting the probe function).
247 - Impl::IOAsync's io watchers did not use the filehandle emulation
248 code to work around it's watcher limits.
249 - work around IO::Async's problems with overloaded objects.
250 - try to disable wearnings inside IO::Async, it's just too buggy.
251 - do no longer use Event by default, only EV or Perl, Event is too buggy.
252 - hackishly provide fast AE:: interface when EV or Perl is the backend.
253 - inherit VERSION from AnyEvent::VERSION in most modules.
254
2554.91 Thu Aug 6 15:42:45 CEST 2009
256 - AE::Handle::starttls could get out of sync when the read buffer
257 already contains some TLS handshake.
258 - AE::Handle did not properly free the TLS session, causing bigger
259 memory leaks in Net::SSLeay than expected (reported by toaster).
260 - AE::Socket::tcp_connect will now artificially delay invoking
261 the callback to avoid returning after invoking the callback.
262 - convert many internal watcher uses to AE API.
263
2644.9 Sat Aug 1 11:07:01 CEST 2009
265 - Glib silently fails when registering a timeout with
266 a negative value, so avoid that.
267 - call condvar callback immediately when it is set
268 after the condition is already signalled.
269 - check rbuf_max condition only after trying to consume
270 data.
271 - normalise signal numbers to names when using the ae's
272 signal handling, but do not document this (yet).
273 - pure perl signal emulation did not properly set nonblocking
274 mode on the signal pipe on win32 (this is unlikely to have caused
275 any issues).
276 - new module: AnyEvent::Debug.
277 - AnyEvent::Strict now checks that a signal was specified
278 by name (not name or number).
279 - reduce memory footprint in typical cases by ~50kb by
280 only compiling signal code when necessary.
281 - add AnyEvent::Handle->rbuf_max.
282 - grab the AE namespace for future new API, implement stubs for
283 said future/alternative API.
284 - new function: AnyEvent::Socket::format_hostport.
285
2864.881 Tue Jul 28 12:51:53 CEST 2009
287 - work around a bug in local in pre-5.10 perls, causing
288 AnyEvent::Handle to recurse when it should not
289 (analysed by elmex).
290
2914.88 Tue Jul 28 04:04:37 CEST 2009
292 - re-bless the handle into a dummy package after calling
293 AnyEvent::Handle::destroy, so the user does not need to check
294 for errors after every push_write etc.
295 - do not attempt to run t/02_signals.t on obviously broken
296 platforms.
297
2984.87 Sun Jul 26 02:06:16 CEST 2009
3 - do not attempt to linger when there is no longer a valid fh. 299 - do not attempt to linger when there is no longer a valid fh.
300 - enforce tls mode to be either accept or connect, do not simply
301 segfault in Net::SSLeay.
302 - AnyEvent::Handle can now call tcp_connect itself (new parameters
303 connect, on_prepare, on_connect and on_connect_error). Updated
304 tutorial accordingly.
305 - add AnyEvent::Impl::Irssi backend.
4 306
54.86 Mon Jul 20 23:52:29 CEST 2009 3074.86 Mon Jul 20 23:52:29 CEST 2009
6 - since the verbose warning is not enough, explicitly document 308 - since the verbose warning is not enough, explicitly document
7 that versions before 1.33 of Net::SSLeay are not secure. 309 that versions before 1.33 of Net::SSLeay are not secure.
8 - work around signal handling races in Event and (...) Event::Lib. 310 - work around signal handling races in Event and (...) Event::Lib.
9 - try to align signal-race timer to full-second boundaries. 311 - try to align signal-race timer to full-second boundaries.
10 - work around Tk not liking negative timeouts. 312 - work around Tk not liking negative timeouts.
11 - don't complain of different grades of environmental unfriendlyness 313 - don't complain of different grades of environmental unfriendlyness
12 in IO::ASync. 314 in IO::Async.
13 315
144.85 Sat Jul 18 06:16:14 CEST 2009 3164.85 Sat Jul 18 06:16:14 CEST 2009
15 - nail the signal race problem in perl once and for all 317 - nail the signal race problem in perl once and for all
16 (see $AnyEvent::MAX_SIGNAL_LATENCY). 318 (see $AnyEvent::MAX_SIGNAL_LATENCY).
17 - take advantage of Async::Interrupt if it is available. 319 - take advantage of Async::Interrupt if it is available.
… …
2144.3 Fri Oct 3 09:18:43 CEST 2008 5164.3 Fri Oct 3 09:18:43 CEST 2008
215 - AnyEvent will now install a no-op signal handler for SIGPIPE, 517 - AnyEvent will now install a no-op signal handler for SIGPIPE,
216 unless one has been installed already. 518 unless one has been installed already.
217 - warn about not loaidng AnyEvent::Impl::POE early enough 519 - warn about not loaidng AnyEvent::Impl::POE early enough
218 (patch by Adam Rosenstein). 520 (patch by Adam Rosenstein).
219 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 521 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
220 - fix a bug in where in SSL connect mode, the 522 - fix a bug in where in SSL connect mode, the
221 client would first wait for some data by the server and 523 client would first wait for some data by the server and
222 otherwise hang. (reported and analyzed in an absolutely 524 otherwise hang. (reported and analysed in an absolutely
223 exemplary manner by Adam Rosenstein). 525 exemplary manner by Adam Rosenstein).
224 - fix a bug in where SSL EOF would not be treated 526 - fix a bug in where SSL EOF would not be treated
225 as stream EOF, putting the connection into a hung state. 527 as stream EOF, putting the connection into a hung state.
226 (reported and analyzed in an absolutely exemplary manner by Adam 528 (reported and analysed in an absolutely exemplary manner by Adam
227 Rosenstein). 529 Rosenstein).
228 - fix a potential segfault when the TLS context 530 - fix a potential segfault when the TLS context
229 would go missing in a read callback (e.g. due to stoptls) - 531 would go missing in a read callback (e.g. due to stoptls) -
230 Net::SSLeay of course makes no type checking whatsoever. 532 Net::SSLeay of course makes no type checking whatsoever.
231 - AnyEvent::Handle will not stop the read watcher in TLS mode. 533 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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