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 1.198 by root, Thu Oct 23 02:41:00 2008 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3 - implement AnyEvent::Handle->destroy method.
4 - hint about unexpected effects in tls mode.
5
64.3 Fri Oct 3 09:18:43 CEST 2008
7 - AnyEvent will now install a no-op signal handler for SIGPIPE,
8 unless one has been installed already.
9 - warn about not loaidng AnyEvent::Impl::POE early enough
10 (patch by Adam Rosenstein).
11 - fixed a great number of bugs and corner cases in AnyEvent::Handle:
12 - fix a bug in where in SSL connect mode, the
13 client would first wait for some data by the server and
14 otherwise hang. (reported and analyzed in an absolutely
15 exemplary manner by Adam Rosenstein).
16 - fix a bug in where SSL EOF would not be treated
17 as stream EOF, putting the connection into a hung state.
18 (reported and analyzed in an absolutely exemplary manner by Adam
19 Rosenstein).
20 - fix a potential segfault when the TLS context
21 would go missing in a read callback (e.g. due to stoptls) -
22 Net::SSLeay of course makes no type checking whatsoever.
23 - AnyEvent::Handle will not stop the read watcher in TLS mode.
24 - AnyEvent::Handle->stoptls will send a TLS close notify
25 instead of simply destroying the stream now.
26 - fix a bug where large blocks of data written in TLS mode
27 would not be sent unless triggered by receives.
28 - on_drain will now take the tls write buffer into account.
29 - SSL operations have been streamlined a bit and should be faster
30 now, more to come, though.
31 - added a FAQ section, will document nontrivial issues in there.
32
334.234 Mon Sep 29 04:08:13 CEST 2008
34 - fix child watcher documentation: only child exits will be reported,
35 no trace events.
36 - mention SIGPIPE in AnyEvent::Handle.
37 - perl backend will now use POSIX::times as monotonic clock
38 when available and the monotonic clock option is not.
39 - run even without Time::HiRes in some cases now, but do not
40 provide sub-second accuracy (all *supported* perl versions have
41 Time::HiRes).
42
434.233 Fri Aug 22 01:48:05 CEST 2008
44 - fix a bug in the testsuite, causing a hang.
45 - clarified lots of AnyEvent::Handle constructor arguments.
46
474.232 Thu Aug 21 20:44:25 CEST 2008
48 - fix a bug in AnyEvent::Handle that could cause two fatal error
49 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
50 (testcase by Leon Brocard).
51 - clarified on_eof/on_error documentation a bit.
52
534.231 Tue Jul 29 13:12:15 CEST 2008
54 - remove some debugging code left in AnyEvent::Util::fork_call
55 (and no, it's impossible to implement with the broken windows
56 perls without resource leaks or worse).
57
584.23 Tue Jul 29 12:19:59 CEST 2008
59 - document the first parameter passed to condvar callbacks to be
60 the callback.
61 - add AnyEvent::Socket::{ntoa,aton} aliases.
62 - optimize the AE::Handle->push_read (line) for the default
63 eol marker.
64 - optimize push_read (packstring|storable) for small packets.
65 - invoke on_error callback when no on_eof callback is set.
66 - fix a bug in push_read (storable) of unknown impact.
67
684.22 Sun Jul 20 16:34:13 CEST 2008
69 - new function AnyEvent::Socket::parse_hostport.
70 - as the bulkheads at microsoft can't even get getprotobyname reliably
71 working on their shitty fucking broken os we need to hardcode
72 some common protocol numbers in AnyEvent::Socket. How can
73 people even bother with such a pile of shit as windows.
74
754.21 Thu Jul 17 14:40:05 CEST 2008
76 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
77 which is incompatible to previous invocation syntax.
78 - work around more windows perl bugs in fork_call. I cannot imagine
79 why anybody would bother implementing fork in such an obviously
80 unusable way.
81 - avoid calling $do_retry if the request has already finished in
82 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
83 K. Sharifulin).
84
854.2 Sat Jul 12 22:42:11 CEST 2008
86 - support an "interval" argument for repeating timers.
87 - fix a bug in the pure perl event loop that caused timers to
88 be delayed under some circumstances.
89 - implement optional argument checking via PERL_ANYEVENT_STRICT.
90 - not importing anything from Event causes it to malfunction
91 and not use Time::HiRes, work around this bug.
92 - more bug workarounds against the endlessly broken Tk module,
93 seems to work now on 32 bit machines, but less so on 64 bit.
94 - minor optimisations applied to most backends.
95
964.161 Fri Jul 4 14:24:48 CEST 2008
97 - fixed the t/handle/02_write.t, it read after write, with the
98 assumption that all data could be written before read
99 (reported by Jim Razmus).
100
1014.160 Thu Jul 3 04:02:21 CEST 2008
102 - re-registering signal watchers after unregistering one did
103 not work if a backend falls back on AnyEvent's default
104 implementation (also affected child watchers).
105 - new AnyEvent::Handle options: autocork and no_delay.
106
1074.152 Sun Jun 22 14:15:44 CEST 2008
108 - allow for 32-bit perls that implement shifts differently
109 on different architectures in parse_ipv4 (reported and
110 analysed by Keiichi DAIBA).
111
1124.151 Fri Jun 6 17:34:24 CEST 2008
113 - make sure specifying _only_ on_read and never pushing reads
114 works.
115
1164.15 Fri Jun 6 13:00:46 CEST 2008
117 - the pure perl backend would keep some watchers alive when more than
118 one watcher was registered for the same fd.
119 - new "packstring" and "storable" read and write types
120 for AnyEvent::Handle.
121 - allow on_eof handler to be called after on_error with EPIPE returns.
122 - do not immediately call on_read callback in handle constructor.
123
1244.14 Thu Jun 5 20:29:31 CEST 2008
125 - Fixed a bug in DNS SRV priority sorting.
126 - AnyEvent::Util::guard now reports runtime errors while
127 executing the guard block as warnings.
128 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
129
1304.13 Thu Jun 5 00:47:59 CEST 2008
131 - AnyEvent::DNS only followed cname chains with length 2,
132 contrary to documentation. bumped it up to 10, thanks to
133 microsoft, the current broken-dns-config-king.
134 - AnyEvent::DNS didn't check for socket return status
135 but instead relied on perl not creating filehandles in that
136 case - too bad it gives you a bogus file handle (reported
137 and analysed by Vladimir Timofeev).
138 - fix queue management logic in AnyEvent::Handle:
139 when on_read was registered, the queue was empty
140 and no progress could be made AnyEvent::Handle would
141 enter an endless loop.
142 - correctly start reading again when the handle became
143 busy again after idling.
144 - correctly treat tls shut-downs as EOF condition
145 (for the time being, would be nice to have a callback for that).
146 - correctly call eof callback under all conditions
147 (hopefully).
148 - the timeout callback did not expect that $self can go
149 away any time.
150
1514.12 Tue Jun 3 10:58:04 CEST 2008
152 - include AnyEvent::Intro, a tutorial for anyevent,
153 anyevent::socket and anyevent::handle.
154 - allow more options in on_error.
155
1564.11 Fri May 30 23:42:25 CEST 2008
157 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
158 reverse_lookup and reverse_verify to replace it, support
159 v4mapped and v4compat addresses.
160 - provide more documentation for the resolver class.
161 - really replace longest run of :0: by :: in format_address,
162 also properly convert :: and ::1 again.
163 - support NAPTR record name and decode it.
164 - implement random weight sampling for SRV records, as per
165 rfc 2782.
166 - correctly abort on srv-record targets of ".".
167 - added AnyEvent::DNS::wait_for_slot.
168 - in the unlikely event of a virtual circuit connection
169 being invalidated by a delayed udp reply, AnyEvent::DNS
170 could die. this has been fixed.
171 - plug a probable memleak in the DNS vc code.
172 - use configured timeout also for VC request phase.
173 - implement timeout and max_outstanding methods
174 for resolver class.
175 - update version numbers in all modules.
176
1774.1 Thu May 29 05:45:40 CEST 2008
178 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
179 which mirrors their purpose better (old names still available
180 for a while). Also moved AnyEvent::DNS::addr to
181 AnyEvent::Socket::resolve_sockaddr.
182 - implement AnyEvent->time and AnyEvent->now.
183 - fix IPv6 support in pack_sockaddr.
184 - officially un-experimentalise all newly introduced code.
185 - support unix domain sockets everywhere by specifying
186 a host of "unix/" and the pathname as service.
187 - implement an activity timeout in AnyEvent::Handle.
188 - added a regex read type to AnyEvent::Handle.
189 - added a json read/write type to AnyEvent::Handle.
190 - always croak in on_error in AnyEvent::Handle.
191 - document how to subclass AnyEvent::Handle.
192 - implement AnyEvent::Util::fork_call.
193 - add support for IPv6 nameservers and nameserver statements.
194 - work around _yet_ _another_ windows perl bug in where empty
195 select masks cause errors in the pure perl backend (this
196 workaround cannot be implemented for the other event loops,
197 but EV also works around this).
198 - supply AnyEvent::Socket::address_type and make good use of it.
199 - clarify and fix the Handle documentation w.r.t $self vs. $handle
200 vs. callback arguments.
201 - add some recommends to the META.yml.
202
2034.05 Mon May 26 19:44:06 CEST 2008
204 - some platforms ignore the rfc and prepend an extra sa_len member to
205 the sockaddr structure, cater for those. also use sockaddr_family,
206 didn't know it was there, but comes in handy for the workaround.
207 - undo the import WIN32 hack.
208
2094.04 Mon May 26 08:03:31 CEST 2008
210 - try to work around yet another windows bug: failed connects are reported
211 as if it were out-of-band data. windows users: you suck. Only
212 EV, Glib, Event and the pure perl backend can handle this condition.
213 - optimize write algorithm in AnyEvent::Handle.
214 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
215 - tcp_connect now properly iterates through all targets.
216 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
217 support entirely. also hardcode some AF_INET6 constants for the
218 sake of perl 5.8.
219 - vastly improved nameserver/suffix detection algorithm for windows
220 (still a hack...)
221 - try to use a fallback dns server if no dns servers could be found.
222 - splatter around some local $SIG{__DIE__}'s to work around bad
223 die handlers.
224
2254.03 Sun May 25 05:05:57 CEST 2008
226 - work around perl bugs on windows, where perl returns undocumented
227 error codes for sysread, syswrite etc.
228 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
229 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
230 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
231 - updated documentation.
232 - lots of workarounds for perl 5.8 quirks.
233 - implement a more flexible type system for both reads and writes
234 in AnyEvent::Handle.
235 - support IPv6 in tcp_server.
236 - pass local hostname and port to the prepare callback in tcp_server.
237 - make types extendable in AnyEvent::Handle.
238 - croak, not die, in AnyEvent::Handle;
239
2404.0 Sat May 24 19:58:08 CEST 2008
241 - added (experimental) AnyEvent::DNS module.
242 - added (experimental) AnyEvent::Socket module.
243 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
244 - make condvar->end call ->send as documented.
245 - make condvar constructor support the documented cb => argument.
246 - overload condvars so you can use them instead of code references.
247 - support more of the POSIX forms instead of a pure dotted quad for
248 verbatim ip addresses.
249 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
250 - try very hard not to freeze on broken windows perls.
251 - make on_eof optional in AnyEvent::Handle.
252 - support service names in tcp_connect and tcp_server.
253 - work around netbsd/openbsd bug causing unix sockets not to
254 become writable until empty.
255 - only make a dummy read for error detection when getpeername
256 returns ENOTCONN.
257 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
258 override.
259 - new AnyEvent::Guard::cancel.
260
2613.5 Sat May 17 23:17:14 CEST 2008
262 - gracefully deal with systems not defining CLOCK_MONOTONIC
263 (as opposed to not supporting the clock itself, which worked in 3.41).
264 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
265 helpers, to replace the ill-designed AnyEvent::Socket class.
266 - added some examples (eg/connect, eg/handle).
267 - added AnyEvent::Util::guard.
268 - AnyEvent::Util now exports most of its functions.
269 - fixed a bug in *_read_chunk, where the callback was called with $self as
270 first and second argument.
271 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
272 - fix manpage condvars examples to use new syntax.
273 - considerably improve inet_aton with EV::ADNS backend to support
274 cname chains (common dns configuration bug, e.g. with www.google.de)
275 and aaaa records if no a record could be found.
276
2773.41 Sun May 11 19:53:13 CEST 2008
278 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
279
2803.4 Sun May 11 00:29:25 CEST 2008
281 - complete re-implementation of condvars with a hopefully much more useful
282 API (backward compatible functions exist but are undocumented).
283 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
284 (still subject to change).
285 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
286 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
287 allows module authors to avoid forcing event loop detection.
288 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
289 - remove coro backends: Coro now provides generic support for AnyEvent
290 via Coro::AnyEvent.
291
2923.3 Mon Apr 28 09:51:06 CEST 2008
293 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
3 - fix a bug in the pure perl backend that kept watchers alive 294 - fix a bug in the pure perl backend that kept watchers alive
4 when multiple watchers were registered for the same fh/poll combo. 295 when multiple watchers were registered for the same fh/poll combo.
5 - add a benchmark section showing AnyEvent overhead and comparing 296 - add a benchmark section showing AnyEvent overhead and comparing
6 the different event loops with each other. 297 the different event loops with each other.
7 - prefer pure perl over tk when autoprobing, as it's about as fast, 298 - prefer pure perl over tk when autoprobing, as it's about as fast,
8 but doesn't crash with many watchers. 299 but doesn't crash with many watchers.
9 - make Qt support non-experimental. 300 - declare Qt support non-experimental.
10 - clarify the confusing section about the file descriptor being 301 - clarify the confusing section about the file descriptor being
11 kept alive. 302 kept alive.
12 - document the race between loading of an event module and 303 - document the race between loading of an event module and
13 child processes exiting. 304 child processes exiting.
14 - support POE as "backend" (with some caveats, POE is not generic 305 - support POE as "backend" (with some caveats, POE is not generic
15 enough, and darn slow). 306 enough, and darn slow).
16 - support Wx and Prima through POE. 307 - support Wx and Prima through POE.
17 - optimise perl backend to use 20% less memory and to be roughly 308 - optimise perl backend to use 20% less memory and take advantage
18 2-3 times faster in the common case of sparsely-active file 309 of typical timeout behaviour. It can now compete with
19 descriptors. It can now compete with select/poll-based C event 310 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 :). 311 faster than Event and Glib :).
21 - roughly cut EV memory use in half and increase its speed by 30%, 312 - roughly cut EV memory use in half and increase its speed by 30%,
22 by removing undocumented functionality. 313 by removing undocumented functionality. Did something similar
314 to other event loops where possible.
23 315
243.2 Thu Apr 24 10:10:40 CEST 2008 3163.2 Thu Apr 24 10:10:40 CEST 2008
25 - do not die when anyevent watchers are destroyed while running callbacks 317 - do not die when anyevent watchers are destroyed while running callbacks
26 in the pureperl backend (could only happen when two watchers are registered 318 in the pureperl backend (could only happen when two watchers are registered
27 for the same fh/poll combo). 319 for the same fh/poll combo).

Diff Legend

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