[ViewVC] Diff of: cvs/AnyEvent/Changes

Comparing AnyEvent/Changes (file contents):
Revision 1.33 by root, Fri Nov 9 19:37:05 2007 UTC vs.
Revision 1.106 by root, Mon May 26 02:18:41 2008 UTC

…		…
1	Revision history for Perl extension AnyEvent.	1	Revision history for Perl extension AnyEvent.
		2
		3	- properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
		4	- check wether IPv6 sockets can actually be created, otherwise, disable ipv6
		5	support entirely.
		6	- vastly improved nameserver/suffix detection algorithm for windows
		7	(still a hack...)
		8	- try to use a fallback dns server if no dns servers could be found.
		9
		10	4.03 Sun May 25 05:05:57 CEST 2008
		11	- work around perl bugs on windows, where perl returns undocumented
		12	error codes for sysread, syswrite etc.
		13	- fix AnyEvent::Handle::unshift_read to really unshift and not push.
		14	- AnyEvent::Handle could sometimes stumble over 0-byte writes.
		15	- fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
		16	- updated documentation.
		17	- lots of workarounds for perl 5.8 quirks.
		18	- implement a more flexible type system for both reads and writes
		19	in AnyEvent::Handle.
		20	- support IPv6 in tcp_server.
		21	- pass local hostname and port to the prepare callback in tcp_server.
		22	- make types extendable in AnyEvent::Handle.
		23	- croak, not die, in AnyEvent::Handle;
		24
		25	4.0 Sat May 24 19:58:08 CEST 2008
		26	- added (experimental) AnyEvent::DNS module.
		27	- added (experimental) AnyEvent::Socket module.
		28	- added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
		29	- make condvar->end call ->send as documented.
		30	- make condvar constructor support the documented cb => argument.
		31	- overload condvars so you can use them instead of code references.
		32	- support more of the POSIX forms instead of a pure dotted quad for
		33	verbatim ip addresses.
		34	- AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
		35	- try very hard not to freeze on broken windows perls.
		36	- make on_eof optional in AnyEvent::Handle.
		37	- support service names in tcp_connect and tcp_server.
		38	- work around netbsd/openbsd bug causing unix sockets not to
		39	become writable untill empty.
		40	- only make a dummy read for error detection when getpeername
		41	returns ENOTCONN.
		42	- defend AnyEvent::Util::inet_aton against Coro::LWP brutal
		43	override.
		44	- new AnyEvent::Guard::cancel.
		45
		46	3.5 Sat May 17 23:17:14 CEST 2008
		47	- gracefully deal with systems not defining CLOCK_MONOTONIC
		48	(as opposed to not supporting the clock itself, which worked in 3.41).
		49	- added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
		50	helpers, to replace the ill-designed AnyEvent::Socket class.
		51	- added some examples (eg/connect, eg/handle).
		52	- added AnyEvent::Util::guard.
		53	- AnyEvent::Util now exports most of its functions.
		54	- fixed a bug in *_read_chunk, where the callback was called with $self as
		55	first and second argument.
		56	- simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
		57	- fix manpage condvars examples to use new syntax.
		58	- considerably improve inet_aton with EV::ADNS backend to support
		59	cname chains (common dns configuration bug, e.g. with www.google.de)
		60	and aaaa records if no a record could be found.
		61
		62	3.41 Sun May 11 19:53:13 CEST 2008
		63	- work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
		64
		65	3.4 Sun May 11 00:29:25 CEST 2008
		66	- complete re-implementation of condvars with a hopefully much mroe useful
		67	API (backward compatible functions exist but are undocumented).
		68	- AnyEvent::Handle fully rewritten with a hopefully more powerful API
		69	(still subject to change).
		70	- take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
		71	- provide AnyEvent::post_detect and @AnyEvent::post_detect, which
		72	allows module authors to avoid forcing event loop detection.
		73	(used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
		74	- remove coro backends: Coro now provides generic support for AnyEvent
		75	via Coro::AnyEvent.
		76
		77	3.3 Mon Apr 28 09:51:06 CEST 2008
		78	- added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
		79	- fix a bug in the pure perl backend that kept watchers alive
		80	when multiple watchers were registered for the same fh/poll combo.
		81	- add a benchmark section showing AnyEvent overhead and comparing
		82	the different event loops with each other.
		83	- prefer pure perl over tk when autoprobing, as it's about as fast,
		84	but doesn't crash with many watchers.
		85	- declare Qt support non-experimental.
		86	- clarify the confusing section about the file descriptor being
		87	kept alive.
		88	- document the race between loading of an event module and
		89	child processes exiting.
		90	- support POE as "backend" (with some caveats, POE is not generic
		91	enough, and darn slow).
		92	- support Wx and Prima through POE.
		93	- optimise perl backend to use 20% less memory and take advantage
		94	of typical timeout behaviour. It can now compete with
		95	select/poll-based C event loops in most cases (it is usually
		96	faster than Event and Glib :).
		97	- roughly cut EV memory use in half and increase its speed by 30%,
		98	by removing undocumented functionality. Did something similar
		99	to other event loops where possible.
		100
		101	3.2 Thu Apr 24 10:10:40 CEST 2008
		102	- do not die when anyevent watchers are destroyed while running callbacks
		103	in the pureperl backend (could only happen when two watchers are registered
		104	for the same fh/poll combo).
		105	- support autoloading for child watchers, was broken in all versions.
		106	- implement PERL_ANYEVENT_MODEL env variable.
		107	- (experimental) implement interface to the (very crashy/buggy) Event::Lib module.
		108	- (experimental) implement interface to the Qt module (cannot be autoprobed).
		109	- this release is IO:AnyEvent-proof.
		110
		111	3.12 Tue Apr 22 07:11:46 CEST 2008
		112	- reinstate AUTHOR section that got lost somehow.
		113	- do not hang in the testsuite with badly broken perls (activestate,
		114	strawberry...), but instead diagnose the problem and continue.
		115	- use INT instead of CHLD in an attempt to support broken windows perls
		116	better (this decreases the test quality, unfortunately).
		117	- do not send the signal to the process group
		118	(no problem for CHLD, bad for INT :).
		119
		120	3.11 Sat Apr 19 06:57:31 CEST 2008
		121	- major documentation rework.
		122	- document the fact that child watchers only watch for zombies.
		123	- fix the child watcher example.
		124
		125	3.1 Wed Apr 16 17:09:01 CEST 2008
		126	- work around recurring bugs in Tk by dup'ing filehandles, the
		127	only method with good success chances on Tk (the bugs apparently
		128	don't get fixed anytime soon).
		129	- lift the restriction of only one watcher per fh direction
		130	(as the Tk bug workaround also lifts it and only Tk imposed
		131	such strong limits).
		132	- changed probe order to prefer coro adaptors.
		133	- explain why recursion into the event loop is not supported
		134	unless the backend supports it (only Coro::EV does without
		135	any restrictions...).
		136	- add simple manpages for all backend modules.
		137
		138	3.0 Mon Apr 7 21:30:23 CEST 2008
		139	- Coro::Signal changed semantics, roll our own, also cleaning
		140	up the Coro implementation in general.
		141	- rename Coro backend to CoroEvent.
		142	- add some decision helping paragraph to the manpage that should
		143	help people to decide whether AnyEvent is the right thing for them.
		144
		145	2.9 Mon Jan 28 13:31:54 CET 2008
		146	- update for EV 3.0 API changes.
		147
		148	2.8 Sun Nov 25 15:06:03 CET 2007
		149	- waitpid can validly return 0. accept this fact of life
		150	instead of reporting it to any watchers.
		151
		152	2.7 Fri Nov 23 11:41:14 CET 2007
		153	- force use of AnyEvent::Impl::Perl in testsuite, there is
		154	too much breakage outside AnyEvent.
		155	- deliver signals synchronously in AnyEvent::Impl::Perl
		156	or any other event loop that relies on AnyEvents child
		157	watcher emulation. Could help with hanging testsuite
		158	(except when Event or EV are installed).
2		159
3	2.6 Fri Nov 9 20:36:35 CET 2007	160	2.6 Fri Nov 9 20:36:35 CET 2007
4	- fix bug in testsuite.	161	- fix bug in testsuite.
5	- move EV adaptor modules to AnyEvent.	162	- move EV adaptor modules to AnyEvent.
6	- add Coro+EV adaptor module.	163	- add Coro+EV adaptor module.
77	- fixed documentation.	234	- fixed documentation.
78	- clarified some corner cases.	235	- clarified some corner cases.
79	- vastly improved documentation with a long example.	236	- vastly improved documentation with a long example.
80		237
81	0.2 Thu Dec 1 22:18:49 CET 2005	238	0.2 Thu Dec 1 22:18:49 CET 2005
82	- check wether a valid method was called, to avoid endless loops.	239	- check whether a valid method was called, to avoid endless loops.
83	- move condvars into their own class (cleaner, safer).	240	- move condvars into their own class (cleaner, safer).
84	- don't die when some modules are not available.	241	- don't die when some modules are not available.
85	- Tk is too broken to allow for words. Will work once Tk works.	242	- Tk is too broken to allow for words. Will work once Tk works.
86		243
87	0.1 Thu Dec 1 19:44:57 CET 2005	244	0.1 Thu Dec 1 19:44:57 CET 2005

Diff Legend

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

Comparing AnyEvent/Changes (file contents): Revision 1.33 by root, Fri Nov 9 19:37:05 2007 UTC vs. Revision 1.106 by root, Mon May 26 02:18:41 2008 UTC

Diff Legend

Comparing AnyEvent/Changes (file contents):
Revision 1.33 by root, Fri Nov 9 19:37:05 2007 UTC vs.
Revision 1.106 by root, Mon May 26 02:18:41 2008 UTC