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

Comparing AnyEvent/Changes (file contents):
Revision 1.27 by root, Sat Oct 27 15:10:09 2007 UTC vs.
Revision 1.169 by root, Sat Jul 12 20:45:27 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3MAYBE: http://svn.xantus.org/shortbus/trunk/cometd-perl/lab/aio.pl 34.2 Sat Jul 12 22:42:11 CEST 2008
4 (poe makes their own "anyevent", unfortunately). 4 - support an "interval" argument for repeating timers.
5 - fix a bug in the pure perl event loop that caused timers to
6 be delayed under some circumstances.
7 - implement optional argument checking via PERL_ANYEVENT_STRICT.
8 - not importing anything from Event causes it to malfunction
9 and not use Time::HiRes, work around this bug.
10 - more bug workarounds against the endlessly broken Tk module,
11 seems to work now on 32 bit machines, but less so on 64 bit.
12 - minor optimisations applied to most backends.
5 13
62.55 144.161 Fri Jul 4 14:24:48 CEST 2008
15 - fixed the t/handle/02_write.t, it read after write, with the
16 assumption that all data could be written before read
17 (reported by Jim Razmus).
18
194.160 Thu Jul 3 04:02:21 CEST 2008
20 - re-registering signal watchers after unregistering one did
21 not work if a backend falls back on AnyEvent's default
22 implementation (also affected child watchers).
23 - new AnyEvent::Handle options: autocork and no_delay.
24
254.152 Sun Jun 22 14:15:44 CEST 2008
26 - allow for 32-bit perls that implement shifts differently
27 on different architectures in parse_ipv4 (reported and
28 analysed by Keiichi DAIBA).
29
304.151 Fri Jun 6 17:34:24 CEST 2008
31 - make sure specifying _only_ on_read and never pushing reads
32 works.
33
344.15 Fri Jun 6 13:00:46 CEST 2008
35 - the pure perl backend would keep some watchers alive when more than
36 one watcher was registered for the same fd.
37 - new "packstring" and "storable" read and write types
38 for AnyEvent::Handle.
39 - allow on_eof handler to be called after on_error with EPIPE returns.
40 - do not immediately call on_read callback in handle constructor.
41
424.14 Thu Jun 5 20:29:31 CEST 2008
43 - Fixed a bug in DNS SRV priority sorting.
44 - AnyEvent::Util::guard now reports runtime errors while
45 executing the guard block as warnings.
46 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
47
484.13 Thu Jun 5 00:47:59 CEST 2008
49 - AnyEvent::DNS only followed cname chains with length 2,
50 contrary to documentation. bumped it up to 10, thanks to
51 microsoft, the current broken-dns-config-king.
52 - AnyEvent::DNS didn't check for socket return status
53 but instead relied on perl not creating filehandles in that
54 case - too bad it gives you a bogus file handle (reported
55 and analysed by Vladimir Timofeev).
56 - fix queue management logic in AnyEvent::Handle:
57 when on_read was registered, the queue was empty
58 and no progress could be made AnyEvent::Handle would
59 enter an endless loop.
60 - correctly start reading again when the handle became
61 busy again after idling.
62 - correctly treat tls shut-downs as EOF condition
63 (for the time being, would be nice to have a callback for that).
64 - correctly call eof callback under all conditions
65 (hopefully).
66 - the timeout callback did not expect that $self can go
67 away any time.
68
694.12 Tue Jun 3 10:58:04 CEST 2008
70 - include AnyEvent::Intro, a tutorial for anyevent,
71 anyevent::socket and anyevent::handle.
72 - allow more options in on_error.
73
744.11 Fri May 30 23:42:25 CEST 2008
75 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
76 reverse_lookup and reverse_verify to replace it, support
77 v4mapped and v4compat addresses.
78 - provide more documentation for the resolver class.
79 - really replace longest run of :0: by :: in format_address,
80 also properly convert :: and ::1 again.
81 - support NAPTR record name and decode it.
82 - implement random weight sampling for SRV records, as per
83 rfc 2782.
84 - correctly abort on srv-record targets of ".".
85 - added AnyEvent::DNS::wait_for_slot.
86 - in the unlikely event of a virtual circuit connection
87 being invalidated by a delayed udp reply, AnyEvent::DNS
88 could die. this has been fixed.
89 - plug a probable memleak in the DNS vc code.
90 - use configured timeout also for VC request phase.
91 - implement timeout and max_outstanding methods
92 for resolver class.
93 - update version numbers in all modules.
94
954.1 Thu May 29 05:45:40 CEST 2008
96 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
97 which mirrors their purpose better (old names still available
98 for a while). Also moved AnyEvent::DNS::addr to
99 AnyEvent::Socket::resolve_sockaddr.
100 - implement AnyEvent->time and AnyEvent->now.
101 - fix IPv6 support in pack_sockaddr.
102 - officially un-experimentalise all newly introduced code.
103 - support unix domain sockets everywhere by specifying
104 a host of "unix/" and the pathname as service.
105 - implement an activity timeout in AnyEvent::Handle.
106 - added a regex read type to AnyEvent::Handle.
107 - added a json read/write type to AnyEvent::Handle.
108 - always croak in on_error in AnyEvent::Handle.
109 - document how to subclass AnyEvent::Handle.
110 - implement AnyEvent::Util::fork_call.
111 - add support for IPv6 nameservers and nameserver statements.
112 - work around _yet_ _another_ windows perl bug in where empty
113 select masks cause errors in the pure perl backend (this
114 workaround cannot be implemented for the other event loops,
115 but EV also works around this).
116 - supply AnyEvent::Socket::address_type and make good use of it.
117 - clarify and fix the Handle documentation w.r.t $self vs. $handle
118 vs. callback arguments.
119 - add some recommends to the META.yml.
120
1214.05 Mon May 26 19:44:06 CEST 2008
122 - some platforms ignore the rfc and prepend an extra sa_len member to
123 the sockaddr structure, cater for those. also use sockaddr_family,
124 didn't know it was there, but comes in handy for the workaround.
125 - undo the import WIN32 hack.
126
1274.04 Mon May 26 08:03:31 CEST 2008
128 - try to work around yet another windows bug: failed connects are reported
129 as if it were out-of-band data. windows users: you suck. Only
130 EV, Glib, Event and the pure perl backend can handle this condition.
131 - optimize write algorithm in AnyEvent::Handle.
132 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
133 - tcp_connect now properly iterates through all targets.
134 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
135 support entirely. also hardcode some AF_INET6 constants for the
136 sake of perl 5.8.
137 - vastly improved nameserver/suffix detection algorithm for windows
138 (still a hack...)
139 - try to use a fallback dns server if no dns servers could be found.
140 - splatter around some local $SIG{__DIE__}'s to work around bad
141 die handlers.
142
1434.03 Sun May 25 05:05:57 CEST 2008
144 - work around perl bugs on windows, where perl returns undocumented
145 error codes for sysread, syswrite etc.
146 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
147 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
148 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
149 - updated documentation.
150 - lots of workarounds for perl 5.8 quirks.
151 - implement a more flexible type system for both reads and writes
152 in AnyEvent::Handle.
153 - support IPv6 in tcp_server.
154 - pass local hostname and port to the prepare callback in tcp_server.
155 - make types extendable in AnyEvent::Handle.
156 - croak, not die, in AnyEvent::Handle;
157
1584.0 Sat May 24 19:58:08 CEST 2008
159 - added (experimental) AnyEvent::DNS module.
160 - added (experimental) AnyEvent::Socket module.
161 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
162 - make condvar->end call ->send as documented.
163 - make condvar constructor support the documented cb => argument.
164 - overload condvars so you can use them instead of code references.
165 - support more of the POSIX forms instead of a pure dotted quad for
166 verbatim ip addresses.
167 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
168 - try very hard not to freeze on broken windows perls.
169 - make on_eof optional in AnyEvent::Handle.
170 - support service names in tcp_connect and tcp_server.
171 - work around netbsd/openbsd bug causing unix sockets not to
172 become writable until empty.
173 - only make a dummy read for error detection when getpeername
174 returns ENOTCONN.
175 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal
176 override.
177 - new AnyEvent::Guard::cancel.
178
1793.5 Sat May 17 23:17:14 CEST 2008
180 - gracefully deal with systems not defining CLOCK_MONOTONIC
181 (as opposed to not supporting the clock itself, which worked in 3.41).
182 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
183 helpers, to replace the ill-designed AnyEvent::Socket class.
184 - added some examples (eg/connect, eg/handle).
185 - added AnyEvent::Util::guard.
186 - AnyEvent::Util now exports most of its functions.
187 - fixed a bug in *_read_chunk, where the callback was called with $self as
188 first and second argument.
189 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
190 - fix manpage condvars examples to use new syntax.
191 - considerably improve inet_aton with EV::ADNS backend to support
192 cname chains (common dns configuration bug, e.g. with www.google.de)
193 and aaaa records if no a record could be found.
194
1953.41 Sun May 11 19:53:13 CEST 2008
196 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
197
1983.4 Sun May 11 00:29:25 CEST 2008
199 - complete re-implementation of condvars with a hopefully much mroe useful
200 API (backward compatible functions exist but are undocumented).
201 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
202 (still subject to change).
203 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
204 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
205 allows module authors to avoid forcing event loop detection.
206 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
207 - remove coro backends: Coro now provides generic support for AnyEvent
208 via Coro::AnyEvent.
209
2103.3 Mon Apr 28 09:51:06 CEST 2008
211 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
212 - fix a bug in the pure perl backend that kept watchers alive
213 when multiple watchers were registered for the same fh/poll combo.
214 - add a benchmark section showing AnyEvent overhead and comparing
215 the different event loops with each other.
216 - prefer pure perl over tk when autoprobing, as it's about as fast,
217 but doesn't crash with many watchers.
218 - declare Qt support non-experimental.
219 - clarify the confusing section about the file descriptor being
220 kept alive.
221 - document the race between loading of an event module and
222 child processes exiting.
223 - support POE as "backend" (with some caveats, POE is not generic
224 enough, and darn slow).
225 - support Wx and Prima through POE.
226 - optimise perl backend to use 20% less memory and take advantage
227 of typical timeout behaviour. It can now compete with
228 select/poll-based C event loops in most cases (it is usually
229 faster than Event and Glib :).
230 - roughly cut EV memory use in half and increase its speed by 30%,
231 by removing undocumented functionality. Did something similar
232 to other event loops where possible.
233
2343.2 Thu Apr 24 10:10:40 CEST 2008
235 - do not die when anyevent watchers are destroyed while running callbacks
236 in the pureperl backend (could only happen when two watchers are registered
237 for the same fh/poll combo).
238 - support autoloading for child watchers, was broken in all versions.
239 - implement PERL_ANYEVENT_MODEL env variable.
240 - (experimental) implement interface to the (very crashy/buggy) Event::Lib module.
241 - (experimental) implement interface to the Qt module (cannot be autoprobed).
242 - this release is IO:AnyEvent-proof.
243
2443.12 Tue Apr 22 07:11:46 CEST 2008
245 - reinstate AUTHOR section that got lost somehow.
246 - do not hang in the testsuite with badly broken perls (activestate,
247 strawberry...), but instead diagnose the problem and continue.
248 - use INT instead of CHLD in an attempt to support broken windows perls
249 better (this decreases the test quality, unfortunately).
250 - do not send the signal to the process group
251 (no problem for CHLD, bad for INT :).
252
2533.11 Sat Apr 19 06:57:31 CEST 2008
254 - major documentation rework.
255 - document the fact that child watchers only watch for zombies.
256 - fix the child watcher example.
257
2583.1 Wed Apr 16 17:09:01 CEST 2008
259 - work around recurring bugs in Tk by dup'ing filehandles, the
260 only method with good success chances on Tk (the bugs apparently
261 don't get fixed anytime soon).
262 - lift the restriction of only one watcher per fh direction
263 (as the Tk bug workaround also lifts it and only Tk imposed
264 such strong limits).
265 - changed probe order to prefer coro adaptors.
266 - explain why recursion into the event loop is not supported
267 unless the backend supports it (only Coro::EV does without
268 any restrictions...).
269 - add simple manpages for all backend modules.
270
2713.0 Mon Apr 7 21:30:23 CEST 2008
272 - Coro::Signal changed semantics, roll our own, also cleaning
273 up the Coro implementation in general.
274 - rename Coro backend to CoroEvent.
275 - add some decision helping paragraph to the manpage that should
276 help people to decide whether AnyEvent is the right thing for them.
277
2782.9 Mon Jan 28 13:31:54 CET 2008
279 - update for EV 3.0 API changes.
280
2812.8 Sun Nov 25 15:06:03 CET 2007
282 - waitpid can validly return 0. accept this fact of life
283 instead of reporting it to any watchers.
284
2852.7 Fri Nov 23 11:41:14 CET 2007
286 - force use of AnyEvent::Impl::Perl in testsuite, there is
287 too much breakage outside AnyEvent.
288 - deliver signals synchronously in AnyEvent::Impl::Perl
289 or any other event loop that relies on AnyEvents child
290 watcher emulation. *Could* help with hanging testsuite
291 (except when Event or EV are installed).
292
2932.6 Fri Nov 9 20:36:35 CET 2007
294 - fix bug in testsuite.
295 - move EV adaptor modules to AnyEvent.
296 - add Coro+EV adaptor module.
297
2982.55 Tue Nov 6 17:41:32 CET 2007
7 - add EV to the list of supported event models. 299 - add EV to the list of supported event models.
300 - do not auto-reset pid watchers, pass pid and status to them.
301 - allow a pid of zero to watch for all children in child watchers.
8 302
92.54 Wed Jul 18 17:36:23 CEST 2007 3032.54 Wed Jul 18 17:36:23 CEST 2007
10 - work around a perl bug that results in 304 - work around a perl bug that results in
11 BEGIN not safe after errors--compilation aborted without 305 BEGIN not safe after errors--compilation aborted without
12 any discernible reason or error message by once 306 any discernible reason or error message by once
… …
73 - fixed documentation. 367 - fixed documentation.
74 - clarified some corner cases. 368 - clarified some corner cases.
75 - vastly improved documentation with a long example. 369 - vastly improved documentation with a long example.
76 370
770.2 Thu Dec 1 22:18:49 CET 2005 3710.2 Thu Dec 1 22:18:49 CET 2005
78 - check wether a valid method was called, to avoid endless loops. 372 - check whether a valid method was called, to avoid endless loops.
79 - move condvars into their own class (cleaner, safer). 373 - move condvars into their own class (cleaner, safer).
80 - don't die when some modules are not available. 374 - don't die when some modules are not available.
81 - Tk is too broken to allow for words. Will work once Tk works. 375 - Tk is too broken to allow for words. Will work once Tk works.
82 376
830.1 Thu Dec 1 19:44:57 CET 2005 3770.1 Thu Dec 1 19:44:57 CET 2005

Diff Legend

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