| 1 | Revision history for Perl extension AnyEvent. |
1 | Revision history for Perl extension AnyEvent. |
| 2 | |
2 | |
| 3 | 4.2 |
3 | - do not attempt to linger when there is no longer a valid fh. |
|
|
4 | |
|
|
5 | 4.86 Mon Jul 20 23:52:29 CEST 2009 |
|
|
6 | - since the verbose warning is not enough, explicitly document |
|
|
7 | that versions before 1.33 of Net::SSLeay are not secure. |
|
|
8 | - work around signal handling races in Event and (...) Event::Lib. |
|
|
9 | - try to align signal-race timer to full-second boundaries. |
|
|
10 | - work around Tk not liking negative timeouts. |
|
|
11 | - don't complain of different grades of environmental unfriendlyness |
|
|
12 | in IO::ASync. |
|
|
13 | |
|
|
14 | 4.85 Sat Jul 18 06:16:14 CEST 2009 |
|
|
15 | - nail the signal race problem in perl once and for all |
|
|
16 | (see $AnyEvent::MAX_SIGNAL_LATENCY). |
|
|
17 | - take advantage of Async::Interrupt if it is available. |
|
|
18 | - load Time::HiRes and Guard modules on demand only. |
|
|
19 | - add optional/recommended modules section to AnyEvent |
|
|
20 | documentation. |
|
|
21 | - reduce memory usage considerably (and reduce startup penalty) |
|
|
22 | by not using "strict", "warnings" and "overload" modules. |
|
|
23 | - work around buggy windows/openbsd perls and provide EBADMSG |
|
|
24 | and EPROTO ourselves when missing. |
|
|
25 | - improve perl 5.6 compatibility of the core event loop. |
|
|
26 | - made Net::SSLeay version 1.33 a soft requirement. |
|
|
27 | |
|
|
28 | 4.83 Fri Jul 17 16:56:26 CEST 2009 |
|
|
29 | - implement AnyEvent::Socket::getprotobyname. |
|
|
30 | - AnyEvent::CondVar's will now detect recursive blocking |
|
|
31 | waits and will croak, as too many people fall into |
|
|
32 | this trap. |
|
|
33 | - AnyEvent::Handle will now call ->destroy on itself after |
|
|
34 | executing the on_error callback, instead of doing some |
|
|
35 | half-baked internal shutdown, for fatal errors. |
|
|
36 | - clarify on_eof behaviour w.r.t. the read queue and |
|
|
37 | on_read callbacks. |
|
|
38 | - ignore some possible spurious wake-ups in tcp_connect. |
|
|
39 | |
|
|
40 | 4.82 Sat Jul 11 00:34:55 CEST 2009 |
|
|
41 | - POE and Event backends didn't accept some callable objects as |
|
|
42 | callbacks. |
|
|
43 | - use Config module instead of POSIX module to detect signal names |
|
|
44 | in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX |
|
|
45 | module doesn't even have all POSIX signals :/. |
|
|
46 | - use more workarounds around the many refcnt/corruption bugs in |
|
|
47 | Event::Lib. |
|
|
48 | - work around a race condition in perl's select, causing t/03_child.t |
|
|
49 | to rarely fail. |
|
|
50 | |
|
|
51 | 4.81 Thu Jul 9 10:30:30 CEST 2009 |
|
|
52 | - AnyEvent::Handle didn't properly diagnose write errors |
|
|
53 | (it expected -1 from syswrite, how lame... :). |
|
|
54 | - support file descriptors in addition to file handles |
|
|
55 | in AnyEvent->io. |
|
|
56 | - new env variables: PERL_ANYEVENT_RESOLV_CONF, |
|
|
57 | PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE |
|
|
58 | and PERL_ANYEVENT_CA_PATH. |
|
|
59 | - provide a sensible synopsis section for AnyEvent::TLS. |
|
|
60 | - add a "supported backends" section to the manpage. |
|
|
61 | - added simple io watcher test to testsuite, using a |
|
|
62 | portable_socketpair. |
|
|
63 | - tried to improve the stability of the Event::Lib backend, |
|
|
64 | YMMV. |
|
|
65 | |
|
|
66 | 4.8 Mon Jul 6 23:45:16 CEST 2009 |
|
|
67 | - AnyEvent::DNS did not properly follow CNAME records with |
|
|
68 | uppercase targets. |
|
|
69 | - AnyEvent::DNS would errornously return AAAA records |
|
|
70 | with v4 mapped addresses (a faulty record) as ipv4 addresses, |
|
|
71 | causing AnyEvent::Socket to throw an exception. |
|
|
72 | - added new module AnyEvent::TLS for easier SSL/TLS context |
|
|
73 | creation, with many options including hostname verification, |
|
|
74 | secure default configuration, lots of documentation and, |
|
|
75 | predefined diffie-hellman keys for perfect forward security |
|
|
76 | and much more. get it while it's still fresh! |
|
|
77 | - use AnyEvent::TLS in AnyEvent::Handle for context management. |
|
|
78 | - load AnyEvent::Handle only on demand in AnyEvent::DNS, |
|
|
79 | so AnyEvent::Socket users have smaller memory footprint |
|
|
80 | in the common case. |
|
|
81 | - add AnyEvent::Handle->push_shutdown method. |
|
|
82 | - add an additional error message parameter to AnyEvent::Handle's |
|
|
83 | on_error callback (for TLS, $! is still available). |
|
|
84 | - add AnyEvent::Handle on_starttls/on_stoptls callbacks. |
|
|
85 | - make AnyEvent::Handle more robust against early conenction |
|
|
86 | failures (during new), and return C<undef> in that case |
|
|
87 | from the constructor. |
|
|
88 | - AnyEvent::Handle will now try to load only JSON::XS first, |
|
|
89 | then fall back to JSON. |
|
|
90 | - format_ipv4/format_ipv6 are now exported by default, for symmetry, |
|
|
91 | and because it was documented that way. |
|
|
92 | |
|
|
93 | 4.451 Fri Jul 3 00:28:58 CEST 2009 |
|
|
94 | - do not clear rbuf when shutting down an AnyEvent::Handle |
|
|
95 | object - doing so breaks AnyEvent::HTTP. |
|
|
96 | |
|
|
97 | 4.45 Mon Jun 29 22:59:26 CEST 2009 |
|
|
98 | - a write error could cause AnyEvent::Handle to create |
|
|
99 | an I/O watcher with an undefined $fh. |
|
|
100 | - special-case mapped ipv4 addresses in both |
|
|
101 | AnyEvent::Socket::format_address and parse_address, |
|
|
102 | to treat them just like ipv4 addresses. |
|
|
103 | - updated and overhauled the AnyEvent::Intro doc. |
|
|
104 | - implement AnyEvent::Socket::format_ipv4/ipv6. |
|
|
105 | - slightly speed up portable_pipe/socketpair. |
|
|
106 | - expand condvar begin/end documentation. |
|
|
107 | |
|
|
108 | 4.42 Fri Jun 26 08:32:18 CEST 2009 |
|
|
109 | - preliminary and neccesarily incomplete support for IO::Async. |
|
|
110 | - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in |
|
|
111 | case it was set to IGNORE, to ensure we can catch child statuses |
|
|
112 | even when the calling env acts stupidly. |
|
|
113 | - updated benchmarks with IO::Async, which performs very well. |
|
|
114 | |
|
|
115 | 4.412 Wed Jun 24 01:35:57 CEST 2009 |
|
|
116 | - support an "untaint" attribute for AnyEvent::DNS |
|
|
117 | and set it on the default resolver. |
|
|
118 | - implement and document AnyEvent::Impl::Perl::loop. |
|
|
119 | - remove all anyevent-env variables from %ENV |
|
|
120 | when running in tainted mode. |
|
|
121 | - mention and extend the IO::Lambda benchmark. |
|
|
122 | |
|
|
123 | 4.411 Sun Jun 7 18:48:02 CEST 2009 |
|
|
124 | - do not try to use F_SETFD on windows. |
|
|
125 | |
|
|
126 | 4.41 Thu May 14 06:40:11 CEST 2009 |
|
|
127 | - work around issues in older perls (5.8.0?) when |
|
|
128 | a signal handler is deleted from the %SIG hash. |
|
|
129 | - use POSIX::_exit in child test, to avoid running |
|
|
130 | destructors. |
|
|
131 | - speed up CHLD handling by relying on SIGCHLD being |
|
|
132 | synchronously delivered, even when we roll our own |
|
|
133 | implementation. |
|
|
134 | - AnyEvent::DNS: add the "dname" resource record name |
|
|
135 | for cosmetic reasons. |
|
|
136 | |
|
|
137 | 4.4 Sun Apr 26 20:12:33 CEST 2009 |
|
|
138 | - implemented idle watchers, where applicable. |
|
|
139 | - AnyEvent->time died when Event backend was in use. |
|
|
140 | - fix a memleak in the Tk backend. |
|
|
141 | - sped up Tk timer handling. |
|
|
142 | - clip negative "after" values to 0 in AnyEvent::Impl::Event |
|
|
143 | to avoid spamming out warning messages. |
|
|
144 | - fix Qt timers without interval. |
|
|
145 | - avoid Qt zero-timeout special case, in old libqt's. |
|
|
146 | |
|
|
147 | 4.352 Mon Apr 20 16:31:11 CEST 2009 |
|
|
148 | - fix AnyEvent::Strict error messages for child watchers. |
|
|
149 | - fix/update Makefile.PL "recommends" versions. |
|
|
150 | - add AnyEvent->now_update. |
|
|
151 | |
|
|
152 | 4.351 Sat Apr 11 07:56:14 CEST 2009 |
|
|
153 | - actually make the signal pipe work under win32. |
|
|
154 | - localise $! in signal handler to avoid changing $!, although |
|
|
155 | perl itself does not seem to save/restore errno either. |
|
|
156 | - set the cloexec flag on the signal pipe (normally set by perl too). |
|
|
157 | |
|
|
158 | 4.35 Fri Mar 27 11:48:20 CET 2009 |
|
|
159 | - event models relying on AnyEvent's signal watcher emulation did |
|
|
160 | invoke the callback asynchronously, contradicting documentation |
|
|
161 | and causing signals to get lost (this includes AnyEvent's own |
|
|
162 | event loop). AnyEvent now uses the standard pipe trick to make |
|
|
163 | callback execution synchronous to the event loop. |
|
|
164 | - AnyEvent::Handle didn't free TLS context data on DESTROY |
|
|
165 | (patch by Pavel Shaydo). |
|
|
166 | - work around the ever-incompatibly-changing API of MakeMaker. |
|
|
167 | - document that changing global variables without restoring them |
|
|
168 | is a bad idea in Perl (noted by Adam Rosenstein). |
|
|
169 | - AnyEvent::Strict now barfs if ->io is passed a file. |
|
|
170 | |
|
|
171 | 4.34 Thu Feb 12 18:32:45 CET 2009 |
|
|
172 | - separately buffer TLS read data, as otherwise the read queue could |
|
|
173 | deadlock as receiving data is not expected while draining the |
|
|
174 | read queue (which cna only happen with TLS). |
|
|
175 | - raise EBADMSG error on JSON decoding errors. |
|
|
176 | - fix some minor manpage bugs (reported by Maximilian Gaß). |
|
|
177 | - speed up select bitmask parsing quite a bit in the pure perl |
|
|
178 | backend. |
|
|
179 | - use CORE::select instead of just select in AnyEvent::Impl::Perl. |
|
|
180 | |
|
|
181 | 4.331 Tue Jan 6 21:07:25 CET 2009 |
|
|
182 | - socketpair fails on many vista machines because vista has |
|
|
183 | completely broken accept/getpeername and getsockname functions, |
|
|
184 | so we provide our own socketpair emulation that kind of works |
|
|
185 | (AnyEvent::Util::portable_pipe). |
|
|
186 | - new function: AnyEvent::Util::portable_socketpair. |
|
|
187 | - take advantage of the Guard module if it exists. |
|
|
188 | |
|
|
189 | 4.33 Fri Nov 21 02:35:40 CET 2008 |
|
|
190 | - AnyEvent::Strict did errornously flag a fileno of 0 as |
|
|
191 | illegal. |
|
|
192 | - reduce memory usage and slightly speed up the pure perl backend |
|
|
193 | by only storing the file descriptor, not the file handle. |
|
|
194 | - add missing autocork method to AnyEvent::Handle (reported |
|
|
195 | by Adam Rosenstein). |
|
|
196 | - AnyEvent::DNS->resolve errornously documented an $rcode result |
|
|
197 | argument, but there isn't (spotted by Henrik Krohns). |
|
|
198 | - the naked truth about Tk - it's basically unmaintained. |
|
|
199 | |
|
|
200 | 4.32 Mon Nov 3 22:46:32 CET 2008 |
|
|
201 | - fix AnyEvent::Socket::resolve_sockaddr to properly support |
|
|
202 | unix sockets again and choose a proper default for |
|
|
203 | the socket type. |
|
|
204 | - fix call to resolve_sockaddr in tcp_connect. |
|
|
205 | |
|
|
206 | 4.31 Thu Oct 30 04:41:48 CET 2008 |
|
|
207 | - implemented AnyEvent::Handle->destroy method. |
|
|
208 | - hint about unexpected effects in TLS mode. |
|
|
209 | - speed up AnyEvent::Impl::Perl by using more arrays |
|
|
210 | instead of hashes without the slightest loss of readability :) |
|
|
211 | - work around a perl argument refcounting bug. |
|
|
212 | - with some perl patches applied, AnyEvent now seems leak-free. |
|
|
213 | |
|
|
214 | 4.3 Fri Oct 3 09:18:43 CEST 2008 |
|
|
215 | - AnyEvent will now install a no-op signal handler for SIGPIPE, |
|
|
216 | unless one has been installed already. |
|
|
217 | - warn about not loaidng AnyEvent::Impl::POE early enough |
|
|
218 | (patch by Adam Rosenstein). |
|
|
219 | - fixed a great number of bugs and corner cases in AnyEvent::Handle: |
|
|
220 | - fix a bug in where in SSL connect mode, the |
|
|
221 | client would first wait for some data by the server and |
|
|
222 | otherwise hang. (reported and analyzed in an absolutely |
|
|
223 | exemplary manner by Adam Rosenstein). |
|
|
224 | - fix a bug in where SSL EOF would not be treated |
|
|
225 | as stream EOF, putting the connection into a hung state. |
|
|
226 | (reported and analyzed in an absolutely exemplary manner by Adam |
|
|
227 | Rosenstein). |
|
|
228 | - fix a potential segfault when the TLS context |
|
|
229 | would go missing in a read callback (e.g. due to stoptls) - |
|
|
230 | Net::SSLeay of course makes no type checking whatsoever. |
|
|
231 | - AnyEvent::Handle will not stop the read watcher in TLS mode. |
|
|
232 | - AnyEvent::Handle->stoptls will send a TLS close notify |
|
|
233 | instead of simply destroying the stream now. |
|
|
234 | - fix a bug where large blocks of data written in TLS mode |
|
|
235 | would not be sent unless triggered by receives. |
|
|
236 | - on_drain will now take the tls write buffer into account. |
|
|
237 | - SSL operations have been streamlined a bit and should be faster |
|
|
238 | now, more to come, though. |
|
|
239 | - added a FAQ section, will document nontrivial issues in there. |
|
|
240 | |
|
|
241 | 4.234 Mon Sep 29 04:08:13 CEST 2008 |
|
|
242 | - fix child watcher documentation: only child exits will be reported, |
|
|
243 | no trace events. |
|
|
244 | - mention SIGPIPE in AnyEvent::Handle. |
|
|
245 | - perl backend will now use POSIX::times as monotonic clock |
|
|
246 | when available and the monotonic clock option is not. |
|
|
247 | - run even without Time::HiRes in some cases now, but do not |
|
|
248 | provide sub-second accuracy (all *supported* perl versions have |
|
|
249 | Time::HiRes). |
|
|
250 | |
|
|
251 | 4.233 Fri Aug 22 01:48:05 CEST 2008 |
|
|
252 | - fix a bug in the testsuite, causing a hang. |
|
|
253 | - clarified lots of AnyEvent::Handle constructor arguments. |
|
|
254 | |
|
|
255 | 4.232 Thu Aug 21 20:44:25 CEST 2008 |
|
|
256 | - fix a bug in AnyEvent::Handle that could cause two fatal error |
|
|
257 | callbacks on EOF, which caused AnyEvent::HTTP to signal success twice |
|
|
258 | (testcase by Leon Brocard). |
|
|
259 | - clarified on_eof/on_error documentation a bit. |
|
|
260 | |
|
|
261 | 4.231 Tue Jul 29 13:12:15 CEST 2008 |
|
|
262 | - remove some debugging code left in AnyEvent::Util::fork_call |
|
|
263 | (and no, it's impossible to implement with the broken windows |
|
|
264 | perls without resource leaks or worse). |
|
|
265 | |
|
|
266 | 4.23 Tue Jul 29 12:19:59 CEST 2008 |
|
|
267 | - document the first parameter passed to condvar callbacks to be |
|
|
268 | the callback. |
|
|
269 | - add AnyEvent::Socket::{ntoa,aton} aliases. |
|
|
270 | - optimize the AE::Handle->push_read (line) for the default |
|
|
271 | eol marker. |
|
|
272 | - optimize push_read (packstring|storable) for small packets. |
|
|
273 | - invoke on_error callback when no on_eof callback is set. |
|
|
274 | - fix a bug in push_read (storable) of unknown impact. |
|
|
275 | |
|
|
276 | 4.22 Sun Jul 20 16:34:13 CEST 2008 |
|
|
277 | - new function AnyEvent::Socket::parse_hostport. |
|
|
278 | - as the bulkheads at microsoft can't even get getprotobyname reliably |
|
|
279 | working on their shitty fucking broken os we need to hardcode |
|
|
280 | some common protocol numbers in AnyEvent::Socket. How can |
|
|
281 | people even bother with such a pile of shit as windows. |
|
|
282 | |
|
|
283 | 4.21 Thu Jul 17 14:40:05 CEST 2008 |
|
|
284 | - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype |
|
|
285 | which is incompatible to previous invocation syntax. |
|
|
286 | - work around more windows perl bugs in fork_call. I cannot imagine |
|
|
287 | why anybody would bother implementing fork in such an obviously |
|
|
288 | unusable way. |
|
|
289 | - avoid calling $do_retry if the request has already finished in |
|
|
290 | AnyEvent::DNS, thus avoiding an exception (reported by Anatoly |
|
|
291 | K. Sharifulin). |
|
|
292 | |
|
|
293 | 4.2 Sat Jul 12 22:42:11 CEST 2008 |
| 4 | - support an "interval" argument for repeating timers. |
294 | - support an "interval" argument for repeating timers. |
| 5 | - not importing anything from Event causes it to malfunction |
|
|
| 6 | and not use Time::HiRes, work around this bug. |
|
|
| 7 | - fix a bug in the pure perl event loop that caused timers to |
295 | - fix a bug in the pure perl event loop that caused timers to |
| 8 | be delayed under some circumstances. |
296 | be delayed under some circumstances. |
| 9 | - implement optional argument checking via PERL_ANYEVENT_STRICT. |
297 | - implement optional argument checking via PERL_ANYEVENT_STRICT. |
|
|
298 | - not importing anything from Event causes it to malfunction |
|
|
299 | and not use Time::HiRes, work around this bug. |
| 10 | - more bug workaroudns against the endlessly broken Tk module, |
300 | - more bug workarounds against the endlessly broken Tk module, |
| 11 | seems to work now on 32 bit machines, but less so on 64 bit. |
301 | seems to work now on 32 bit machines, but less so on 64 bit. |
| 12 | - minor optimisations applied to most backends. |
302 | - minor optimisations applied to most backends. |
| 13 | |
303 | |
| 14 | 4.161 Fri Jul 4 14:24:48 CEST 2008 |
304 | 4.161 Fri Jul 4 14:24:48 CEST 2008 |
| 15 | - fixed the t/handle/02_write.t, it read after write, with the |
305 | - fixed the t/handle/02_write.t, it read after write, with the |
| … | |
… | |
| 170 | - support service names in tcp_connect and tcp_server. |
460 | - support service names in tcp_connect and tcp_server. |
| 171 | - work around netbsd/openbsd bug causing unix sockets not to |
461 | - work around netbsd/openbsd bug causing unix sockets not to |
| 172 | become writable until empty. |
462 | become writable until empty. |
| 173 | - only make a dummy read for error detection when getpeername |
463 | - only make a dummy read for error detection when getpeername |
| 174 | returns ENOTCONN. |
464 | returns ENOTCONN. |
| 175 | - defend AnyEvent::Util::inet_aton against Coro::LWP brutal |
465 | - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal |
| 176 | override. |
466 | override. |
| 177 | - new AnyEvent::Guard::cancel. |
467 | - new AnyEvent::Guard::cancel. |
| 178 | |
468 | |
| 179 | 3.5 Sat May 17 23:17:14 CEST 2008 |
469 | 3.5 Sat May 17 23:17:14 CEST 2008 |
| 180 | - gracefully deal with systems not defining CLOCK_MONOTONIC |
470 | - gracefully deal with systems not defining CLOCK_MONOTONIC |
| … | |
… | |
| 194 | |
484 | |
| 195 | 3.41 Sun May 11 19:53:13 CEST 2008 |
485 | 3.41 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. |
486 | - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. |
| 197 | |
487 | |
| 198 | 3.4 Sun May 11 00:29:25 CEST 2008 |
488 | 3.4 Sun May 11 00:29:25 CEST 2008 |
| 199 | - complete re-implementation of condvars with a hopefully much mroe useful |
489 | - complete re-implementation of condvars with a hopefully much more useful |
| 200 | API (backward compatible functions exist but are undocumented). |
490 | API (backward compatible functions exist but are undocumented). |
| 201 | - AnyEvent::Handle fully rewritten with a hopefully more powerful API |
491 | - AnyEvent::Handle fully rewritten with a hopefully more powerful API |
| 202 | (still subject to change). |
492 | (still subject to change). |
| 203 | - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. |
493 | - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. |
| 204 | - provide AnyEvent::post_detect and @AnyEvent::post_detect, which |
494 | - provide AnyEvent::post_detect and @AnyEvent::post_detect, which |
