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

Comparing AnyEvent/Changes (file contents):
Revision 1.59 by root, Sat Apr 26 04:19:02 2008 UTC vs.
Revision 1.184 by root, Wed Aug 20 12:37:21 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.231 Tue Jul 29 13:12:15 CEST 2008
4 - remove some debugging code left in AnyEvent::Util::fork_call
5 (and no, it's impossible to implement with the broken windows
6 perls without resource leaks or worse).
7
84.23 Tue Jul 29 12:19:59 CEST 2008
9 - document the first parameter passed to condvar callbacks to be
10 the callback.
11 - add AnyEvent::Socket::{ntoa,aton} aliases.
12 - optimize the AE::Handle->push_read (line) for the default
13 eol marker.
14 - optimize push_read (packstring|storable) for small packets.
15 - invoke on_error callback when no on_eof callback is set.
16 - fix a bug in push_read (storable) of unknown impact.
17
184.22 Sun Jul 20 16:34:13 CEST 2008
19 - new function AnyEvent::Socket::parse_hostport.
20 - as the bulkheads at microsoft can't even get getprotobyname reliably
21 working on their shitty fucking broken os we need to hardcode
22 some common protocol numbers in AnyEvent::Socket. How can
23 people even bother with such a pile of shit as windows.
24
254.21 Thu Jul 17 14:40:05 CEST 2008
26 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
27 which is incompatible to previous invocation syntax.
28 - work around more windows perl bugs in fork_call. I cannot imagine
29 why anybody would bother implementing fork in such an obviously
30 unusable way.
31 - avoid calling $do_retry if the request has already finished in
32 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
33 K. Sharifulin).
34
354.2 Sat Jul 12 22:42:11 CEST 2008
36 - support an "interval" argument for repeating timers.
37 - fix a bug in the pure perl event loop that caused timers to
38 be delayed under some circumstances.
39 - implement optional argument checking via PERL_ANYEVENT_STRICT.
40 - not importing anything from Event causes it to malfunction
41 and not use Time::HiRes, work around this bug.
42 - more bug workarounds against the endlessly broken Tk module,
43 seems to work now on 32 bit machines, but less so on 64 bit.
44 - minor optimisations applied to most backends.
45
464.161 Fri Jul 4 14:24:48 CEST 2008
47 - fixed the t/handle/02_write.t, it read after write, with the
48 assumption that all data could be written before read
49 (reported by Jim Razmus).
50
514.160 Thu Jul 3 04:02:21 CEST 2008
52 - re-registering signal watchers after unregistering one did
53 not work if a backend falls back on AnyEvent's default
54 implementation (also affected child watchers).
55 - new AnyEvent::Handle options: autocork and no_delay.
56
574.152 Sun Jun 22 14:15:44 CEST 2008
58 - allow for 32-bit perls that implement shifts differently
59 on different architectures in parse_ipv4 (reported and
60 analysed by Keiichi DAIBA).
61
624.151 Fri Jun 6 17:34:24 CEST 2008
63 - make sure specifying _only_ on_read and never pushing reads
64 works.
65
664.15 Fri Jun 6 13:00:46 CEST 2008
67 - the pure perl backend would keep some watchers alive when more than
68 one watcher was registered for the same fd.
69 - new "packstring" and "storable" read and write types
70 for AnyEvent::Handle.
71 - allow on_eof handler to be called after on_error with EPIPE returns.
72 - do not immediately call on_read callback in handle constructor.
73
744.14 Thu Jun 5 20:29:31 CEST 2008
75 - Fixed a bug in DNS SRV priority sorting.
76 - AnyEvent::Util::guard now reports runtime errors while
77 executing the guard block as warnings.
78 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
79
804.13 Thu Jun 5 00:47:59 CEST 2008
81 - AnyEvent::DNS only followed cname chains with length 2,
82 contrary to documentation. bumped it up to 10, thanks to
83 microsoft, the current broken-dns-config-king.
84 - AnyEvent::DNS didn't check for socket return status
85 but instead relied on perl not creating filehandles in that
86 case - too bad it gives you a bogus file handle (reported
87 and analysed by Vladimir Timofeev).
88 - fix queue management logic in AnyEvent::Handle:
89 when on_read was registered, the queue was empty
90 and no progress could be made AnyEvent::Handle would
91 enter an endless loop.
92 - correctly start reading again when the handle became
93 busy again after idling.
94 - correctly treat tls shut-downs as EOF condition
95 (for the time being, would be nice to have a callback for that).
96 - correctly call eof callback under all conditions
97 (hopefully).
98 - the timeout callback did not expect that $self can go
99 away any time.
100
1014.12 Tue Jun 3 10:58:04 CEST 2008
102 - include AnyEvent::Intro, a tutorial for anyevent,
103 anyevent::socket and anyevent::handle.
104 - allow more options in on_error.
105
1064.11 Fri May 30 23:42:25 CEST 2008
107 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
108 reverse_lookup and reverse_verify to replace it, support
109 v4mapped and v4compat addresses.
110 - provide more documentation for the resolver class.
111 - really replace longest run of :0: by :: in format_address,
112 also properly convert :: and ::1 again.
113 - support NAPTR record name and decode it.
114 - implement random weight sampling for SRV records, as per
115 rfc 2782.
116 - correctly abort on srv-record targets of ".".
117 - added AnyEvent::DNS::wait_for_slot.
118 - in the unlikely event of a virtual circuit connection
119 being invalidated by a delayed udp reply, AnyEvent::DNS
120 could die. this has been fixed.
121 - plug a probable memleak in the DNS vc code.
122 - use configured timeout also for VC request phase.
123 - implement timeout and max_outstanding methods
124 for resolver class.
125 - update version numbers in all modules.
126
1274.1 Thu May 29 05:45:40 CEST 2008
128 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
129 which mirrors their purpose better (old names still available
130 for a while). Also moved AnyEvent::DNS::addr to
131 AnyEvent::Socket::resolve_sockaddr.
132 - implement AnyEvent->time and AnyEvent->now.
133 - fix IPv6 support in pack_sockaddr.
134 - officially un-experimentalise all newly introduced code.
135 - support unix domain sockets everywhere by specifying
136 a host of "unix/" and the pathname as service.
137 - implement an activity timeout in AnyEvent::Handle.
138 - added a regex read type to AnyEvent::Handle.
139 - added a json read/write type to AnyEvent::Handle.
140 - always croak in on_error in AnyEvent::Handle.
141 - document how to subclass AnyEvent::Handle.
142 - implement AnyEvent::Util::fork_call.
143 - add support for IPv6 nameservers and nameserver statements.
144 - work around _yet_ _another_ windows perl bug in where empty
145 select masks cause errors in the pure perl backend (this
146 workaround cannot be implemented for the other event loops,
147 but EV also works around this).
148 - supply AnyEvent::Socket::address_type and make good use of it.
149 - clarify and fix the Handle documentation w.r.t $self vs. $handle
150 vs. callback arguments.
151 - add some recommends to the META.yml.
152
1534.05 Mon May 26 19:44:06 CEST 2008
154 - some platforms ignore the rfc and prepend an extra sa_len member to
155 the sockaddr structure, cater for those. also use sockaddr_family,
156 didn't know it was there, but comes in handy for the workaround.
157 - undo the import WIN32 hack.
158
1594.04 Mon May 26 08:03:31 CEST 2008
160 - try to work around yet another windows bug: failed connects are reported
161 as if it were out-of-band data. windows users: you suck. Only
162 EV, Glib, Event and the pure perl backend can handle this condition.
163 - optimize write algorithm in AnyEvent::Handle.
164 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
165 - tcp_connect now properly iterates through all targets.
166 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
167 support entirely. also hardcode some AF_INET6 constants for the
168 sake of perl 5.8.
169 - vastly improved nameserver/suffix detection algorithm for windows
170 (still a hack...)
171 - try to use a fallback dns server if no dns servers could be found.
172 - splatter around some local $SIG{__DIE__}'s to work around bad
173 die handlers.
174
1754.03 Sun May 25 05:05:57 CEST 2008
176 - work around perl bugs on windows, where perl returns undocumented
177 error codes for sysread, syswrite etc.
178 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
179 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
180 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
181 - updated documentation.
182 - lots of workarounds for perl 5.8 quirks.
183 - implement a more flexible type system for both reads and writes
184 in AnyEvent::Handle.
185 - support IPv6 in tcp_server.
186 - pass local hostname and port to the prepare callback in tcp_server.
187 - make types extendable in AnyEvent::Handle.
188 - croak, not die, in AnyEvent::Handle;
189
1904.0 Sat May 24 19:58:08 CEST 2008
191 - added (experimental) AnyEvent::DNS module.
192 - added (experimental) AnyEvent::Socket module.
193 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
194 - make condvar->end call ->send as documented.
195 - make condvar constructor support the documented cb => argument.
196 - overload condvars so you can use them instead of code references.
197 - support more of the POSIX forms instead of a pure dotted quad for
198 verbatim ip addresses.
199 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
200 - try very hard not to freeze on broken windows perls.
201 - make on_eof optional in AnyEvent::Handle.
202 - support service names in tcp_connect and tcp_server.
203 - work around netbsd/openbsd bug causing unix sockets not to
204 become writable until empty.
205 - only make a dummy read for error detection when getpeername
206 returns ENOTCONN.
207 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
208 override.
209 - new AnyEvent::Guard::cancel.
210
2113.5 Sat May 17 23:17:14 CEST 2008
212 - gracefully deal with systems not defining CLOCK_MONOTONIC
213 (as opposed to not supporting the clock itself, which worked in 3.41).
214 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
215 helpers, to replace the ill-designed AnyEvent::Socket class.
216 - added some examples (eg/connect, eg/handle).
217 - added AnyEvent::Util::guard.
218 - AnyEvent::Util now exports most of its functions.
219 - fixed a bug in *_read_chunk, where the callback was called with $self as
220 first and second argument.
221 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
222 - fix manpage condvars examples to use new syntax.
223 - considerably improve inet_aton with EV::ADNS backend to support
224 cname chains (common dns configuration bug, e.g. with www.google.de)
225 and aaaa records if no a record could be found.
226
2273.41 Sun May 11 19:53:13 CEST 2008
228 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
229
2303.4 Sun May 11 00:29:25 CEST 2008
231 - complete re-implementation of condvars with a hopefully much more useful
232 API (backward compatible functions exist but are undocumented).
233 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
234 (still subject to change).
235 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
236 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
237 allows module authors to avoid forcing event loop detection.
238 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
239 - remove coro backends: Coro now provides generic support for AnyEvent
240 via Coro::AnyEvent.
241
2423.3 Mon Apr 28 09:51:06 CEST 2008
243 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
3 - fix a bug in the pure perl backend that kept watchers alive 244 - fix a bug in the pure perl backend that kept watchers alive
4 when multiple watchers were registered for the same fh/poll combo. 245 when multiple watchers were registered for the same fh/poll combo.
5 - add a benchmark section showing AnyEvent overhead and comparing 246 - add a benchmark section showing AnyEvent overhead and comparing
6 the different event loops with each other. 247 the different event loops with each other.
7 - prefer pure perl over tk when autoprobing, as it's about as fast, 248 - prefer pure perl over tk when autoprobing, as it's about as fast,
8 but doesn't crash with many watchers. 249 but doesn't crash with many watchers.
9 - make Qt support non-experimental. 250 - declare Qt support non-experimental.
10 - clarify the confusing section about the file descriptor being 251 - clarify the confusing section about the file descriptor being
11 kept alive. 252 kept alive.
12 - document the race between loading of an event module and 253 - document the race between loading of an event module and
13 child processes exiting. 254 child processes exiting.
14 - support POE as "backend" (with some caveats, POE is not generic 255 - support POE as "backend" (with some caveats, POE is not generic
15 enough, and darn slow). 256 enough, and darn slow).
16 - support Wx and Prima through POE. 257 - support Wx and Prima through POE.
17 - optimise perl backend to use 20% less memory and to be roughly 258 - optimise perl backend to use 20% less memory and take advantage
18 2-3 times faster in the common case of sparsely-active file 259 of typical timeout behaviour. It can now compete with
19 descriptors. It can now compete with select/poll-based C event 260 select/poll-based C event loops in most cases (it is usually
20 loops in most cases (it is usually faster than Event and Glib :). 261 faster than Event and Glib :).
21 - roughly cut EV memory use in half and increase its speed by 30%, 262 - roughly cut EV memory use in half and increase its speed by 30%,
22 by removing undocumented functionality. 263 by removing undocumented functionality. Did something similar
264 to other event loops where possible.
23 265
243.2 Thu Apr 24 10:10:40 CEST 2008 2663.2 Thu Apr 24 10:10:40 CEST 2008
25 - do not die when anyevent watchers are destroyed while running callbacks 267 - do not die when anyevent watchers are destroyed while running callbacks
26 in the pureperl backend (could only happen when two watchers are registered 268 in the pureperl backend (could only happen when two watchers are registered
27 for the same fh/poll combo). 269 for the same fh/poll combo).

Diff Legend

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