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

Comparing AnyEvent/Changes (file contents):
Revision 1.57 by root, Fri Apr 25 13:39:08 2008 UTC vs.
Revision 1.161 by root, Thu Jul 3 02:03:33 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.160 Thu Jul 3 04:02:21 CEST 2008
4 - re-registering signal watchers after unregistering one did
5 not work if a backend falls back on AnyEvent's default
6 implementation (also affected child watchers).
7 - new AnyEvent::Handle options: autocork and no_delay.
8
94.152 Sun Jun 22 14:15:44 CEST 2008
10 - allow for 32-bit perls that implement shifts differently
11 on different architectures in parse_ipv4 (reported and
12 analysed by Keiichi DAIBA).
13
144.151 Fri Jun 6 17:34:24 CEST 2008
15 - make sure specifying _only_ on_read and never pushing reads
16 works.
17
184.15 Fri Jun 6 13:00:46 CEST 2008
19 - the pure perl backend would keep some watchers alive when more than
20 one watcher was registered for the same fd.
21 - new "packstring" and "storable" read and write types
22 for AnyEvent::Handle.
23 - allow on_eof handler to be called after on_error with EPIPE returns.
24 - do not immediately call on_read callback in handle constructor.
25
264.14 Thu Jun 5 20:29:31 CEST 2008
27 - Fixed a bug in DNS SRV priority sorting.
28 - AnyEvent::Util::guard now reports runtime errors while
29 executing the guard block as warnings.
30 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
31
324.13 Thu Jun 5 00:47:59 CEST 2008
33 - AnyEvent::DNS only followed cname chains with length 2,
34 contrary to documentation. bumped it up to 10, thanks to
35 microsoft, the current broken-dns-config-king.
36 - AnyEvent::DNS didn't check for socket return status
37 but instead relied on perl not creating filehandles in that
38 case - too bad it gives you a bogus file handle (reported
39 and analysed by Vladimir Timofeev).
40 - fix queue management logic in AnyEvent::Handle:
41 when on_read was registered, the queue was empty
42 and no progress could be made AnyEvent::Handle would
43 enter an endless loop.
44 - correctly start reading again when the handle became
45 busy again after idling.
46 - correctly treat tls shut-downs as EOF condition
47 (for the time being, would be nice to have a callback for that).
48 - correctly call eof callback under all conditions
49 (hopefully).
50 - the timeout callback did not expect that $self can go
51 away any time.
52
534.12 Tue Jun 3 10:58:04 CEST 2008
54 - include AnyEvent::Intro, a tutorial for anyevent,
55 anyevent::socket and anyevent::handle.
56 - allow more options in on_error.
57
584.11 Fri May 30 23:42:25 CEST 2008
59 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
60 reverse_lookup and reverse_verify to replace it, support
61 v4mapped and v4compat addresses.
62 - provide more documentation for the resolver class.
63 - really replace longest run of :0: by :: in format_address,
64 also properly convert :: and ::1 again.
65 - support NAPTR record name and decode it.
66 - implement random weight sampling for SRV records, as per
67 rfc 2782.
68 - correctly abort on srv-record targets of ".".
69 - added AnyEvent::DNS::wait_for_slot.
70 - in the unlikely event of a virtual circuit connection
71 being invalidated by a delayed udp reply, AnyEvent::DNS
72 could die. this has been fixed.
73 - plug a probable memleak in the DNS vc code.
74 - use configured timeout also for VC request phase.
75 - implement timeout and max_outstanding methods
76 for resolver class.
77 - update version numbers in all modules.
78
794.1 Thu May 29 05:45:40 CEST 2008
80 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
81 which mirrors their purpose better (old names still available
82 for a while). Also moved AnyEvent::DNS::addr to
83 AnyEvent::Socket::resolve_sockaddr.
84 - implement AnyEvent->time and AnyEvent->now.
85 - fix IPv6 support in pack_sockaddr.
86 - officially un-experimentalise all newly introduced code.
87 - support unix domain sockets everywhere by specifying
88 a host of "unix/" and the pathname as service.
89 - implement an activity timeout in AnyEvent::Handle.
90 - added a regex read type to AnyEvent::Handle.
91 - added a json read/write type to AnyEvent::Handle.
92 - always croak in on_error in AnyEvent::Handle.
93 - document how to subclass AnyEvent::Handle.
94 - implement AnyEvent::Util::fork_call.
95 - add support for IPv6 nameservers and nameserver statements.
96 - work around _yet_ _another_ windows perl bug in where empty
97 select masks cause errors in the pure perl backend (this
98 workaround cannot be implemented for the other event loops,
99 but EV also works around this).
100 - supply AnyEvent::Socket::address_type and make good use of it.
101 - clarify and fix the Handle documentation w.r.t $self vs. $handle
102 vs. callback arguments.
103 - add some recommends to the META.yml.
104
1054.05 Mon May 26 19:44:06 CEST 2008
106 - some platforms ignore the rfc and prepend an extra sa_len member to
107 the sockaddr structure, cater for those. also use sockaddr_family,
108 didn't know it was there, but comes in handy for the workaround.
109 - undo the import WIN32 hack.
110
1114.04 Mon May 26 08:03:31 CEST 2008
112 - try to work around yet another windows bug: failed connects are reported
113 as if it were out-of-band data. windows users: you suck. Only
114 EV, Glib, Event and the pure perl backend can handle this condition.
115 - optimize write algorithm in AnyEvent::Handle.
116 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
117 - tcp_connect now properly iterates through all targets.
118 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
119 support entirely. also hardcode some AF_INET6 constants for the
120 sake of perl 5.8.
121 - vastly improved nameserver/suffix detection algorithm for windows
122 (still a hack...)
123 - try to use a fallback dns server if no dns servers could be found.
124 - splatter around some local $SIG{__DIE__}'s to work around bad
125 die handlers.
126
1274.03 Sun May 25 05:05:57 CEST 2008
128 - work around perl bugs on windows, where perl returns undocumented
129 error codes for sysread, syswrite etc.
130 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
131 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
132 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
133 - updated documentation.
134 - lots of workarounds for perl 5.8 quirks.
135 - implement a more flexible type system for both reads and writes
136 in AnyEvent::Handle.
137 - support IPv6 in tcp_server.
138 - pass local hostname and port to the prepare callback in tcp_server.
139 - make types extendable in AnyEvent::Handle.
140 - croak, not die, in AnyEvent::Handle;
141
1424.0 Sat May 24 19:58:08 CEST 2008
143 - added (experimental) AnyEvent::DNS module.
144 - added (experimental) AnyEvent::Socket module.
145 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
146 - make condvar->end call ->send as documented.
147 - make condvar constructor support the documented cb => argument.
148 - overload condvars so you can use them instead of code references.
149 - support more of the POSIX forms instead of a pure dotted quad for
150 verbatim ip addresses.
151 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
152 - try very hard not to freeze on broken windows perls.
153 - make on_eof optional in AnyEvent::Handle.
154 - support service names in tcp_connect and tcp_server.
155 - work around netbsd/openbsd bug causing unix sockets not to
156 become writable until empty.
157 - only make a dummy read for error detection when getpeername
158 returns ENOTCONN.
159 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal
160 override.
161 - new AnyEvent::Guard::cancel.
162
1633.5 Sat May 17 23:17:14 CEST 2008
164 - gracefully deal with systems not defining CLOCK_MONOTONIC
165 (as opposed to not supporting the clock itself, which worked in 3.41).
166 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
167 helpers, to replace the ill-designed AnyEvent::Socket class.
168 - added some examples (eg/connect, eg/handle).
169 - added AnyEvent::Util::guard.
170 - AnyEvent::Util now exports most of its functions.
171 - fixed a bug in *_read_chunk, where the callback was called with $self as
172 first and second argument.
173 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
174 - fix manpage condvars examples to use new syntax.
175 - considerably improve inet_aton with EV::ADNS backend to support
176 cname chains (common dns configuration bug, e.g. with www.google.de)
177 and aaaa records if no a record could be found.
178
1793.41 Sun May 11 19:53:13 CEST 2008
180 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
181
1823.4 Sun May 11 00:29:25 CEST 2008
183 - complete re-implementation of condvars with a hopefully much mroe useful
184 API (backward compatible functions exist but are undocumented).
185 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
186 (still subject to change).
187 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
188 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
189 allows module authors to avoid forcing event loop detection.
190 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
191 - remove coro backends: Coro now provides generic support for AnyEvent
192 via Coro::AnyEvent.
193
1943.3 Mon Apr 28 09:51:06 CEST 2008
195 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
3 - fix a bug in the pure perl backend that kept watchers alive 196 - fix a bug in the pure perl backend that kept watchers alive
4 when multiple watchers were registered for the same fh/poll combo. 197 when multiple watchers were registered for the same fh/poll combo.
5 - add a benchmark section showing AnyEvent overhead and comparing 198 - add a benchmark section showing AnyEvent overhead and comparing
6 the different event loops with each other. 199 the different event loops with each other.
7 - prefer pure perl over tk when autoprobing, as it's about as fast, 200 - prefer pure perl over tk when autoprobing, as it's about as fast,
8 but doesn't crash with many watchers. 201 but doesn't crash with many watchers.
9 - make Qt support non-experimental. 202 - declare Qt support non-experimental.
10 - clarify the confusing section about the file descriptor being 203 - clarify the confusing section about the file descriptor being
11 kept alive. 204 kept alive.
12 - document the race between loading of an event module and 205 - document the race between loading of an event module and
13 child processes exiting. 206 child processes exiting.
14 - support POE as "backend" (with some caveats, POE is not generic 207 - support POE as "backend" (with some caveats, POE is not generic
15 enough, and darn slow). 208 enough, and darn slow).
16 - support Wx and Prima through POE. 209 - support Wx and Prima through POE.
17 - optimise perl backend a bit (less memory). 210 - optimise perl backend to use 20% less memory and take advantage
211 of typical timeout behaviour. It can now compete with
212 select/poll-based C event loops in most cases (it is usually
213 faster than Event and Glib :).
18 - roughly cut EV memory use in half and icrease its speed by 30%, 214 - roughly cut EV memory use in half and increase its speed by 30%,
19 by removing undocumented functionality. 215 by removing undocumented functionality. Did something similar
216 to other event loops where possible.
20 217
213.2 Thu Apr 24 10:10:40 CEST 2008 2183.2 Thu Apr 24 10:10:40 CEST 2008
22 - do not die when anyevent watchers are destroyed while running callbacks 219 - do not die when anyevent watchers are destroyed while running callbacks
23 in the pureperl backend (could only happen when two watchers are registered 220 in the pureperl backend (could only happen when two watchers are registered
24 for the same fh/poll combo). 221 for the same fh/poll combo).

Diff Legend

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