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 1.301 by root, Mon Jul 27 22:49:22 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3 - re-bless the handle into a dummy package after calling
4 AnyEvent::Handle::destroy, so the user does not need to check
5 for errors after every push_write etc.
6
74.87 Sun Jul 26 02:06:16 CEST 2009
8 - do not attempt to linger when there is no longer a valid fh.
9 - enforce tls mode to be either accept or connect, do not simply
10 segfault in Net::SSLeay.
11 - AnyEvent::Handle can now call tcp_connect itself (new parameters
12 connect, on_prepare, on_connect and on_connect_error). Updated
13 tutorial accordingly.
14 - add AnyEvent::Impl::Irssi backend.
15
164.86 Mon Jul 20 23:52:29 CEST 2009
17 - since the verbose warning is not enough, explicitly document
18 that versions before 1.33 of Net::SSLeay are not secure.
19 - work around signal handling races in Event and (...) Event::Lib.
20 - try to align signal-race timer to full-second boundaries.
21 - work around Tk not liking negative timeouts.
22 - don't complain of different grades of environmental unfriendlyness
23 in IO::ASync.
24
254.85 Sat Jul 18 06:16:14 CEST 2009
26 - nail the signal race problem in perl once and for all
27 (see $AnyEvent::MAX_SIGNAL_LATENCY).
28 - take advantage of Async::Interrupt if it is available.
29 - load Time::HiRes and Guard modules on demand only.
30 - add optional/recommended modules section to AnyEvent
31 documentation.
32 - reduce memory usage considerably (and reduce startup penalty)
33 by not using "strict", "warnings" and "overload" modules.
34 - work around buggy windows/openbsd perls and provide EBADMSG
35 and EPROTO ourselves when missing.
36 - improve perl 5.6 compatibility of the core event loop.
37 - made Net::SSLeay version 1.33 a soft requirement.
38
394.83 Fri Jul 17 16:56:26 CEST 2009
40 - implement AnyEvent::Socket::getprotobyname.
41 - AnyEvent::CondVar's will now detect recursive blocking
42 waits and will croak, as too many people fall into
43 this trap.
44 - AnyEvent::Handle will now call ->destroy on itself after
45 executing the on_error callback, instead of doing some
46 half-baked internal shutdown, for fatal errors.
47 - clarify on_eof behaviour w.r.t. the read queue and
48 on_read callbacks.
49 - ignore some possible spurious wake-ups in tcp_connect.
50
514.82 Sat Jul 11 00:34:55 CEST 2009
52 - POE and Event backends didn't accept some callable objects as
53 callbacks.
54 - use Config module instead of POSIX module to detect signal names
55 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
56 module doesn't even have all POSIX signals :/.
57 - use more workarounds around the many refcnt/corruption bugs in
58 Event::Lib.
59 - work around a race condition in perl's select, causing t/03_child.t
60 to rarely fail.
61
624.81 Thu Jul 9 10:30:30 CEST 2009
63 - AnyEvent::Handle didn't properly diagnose write errors
64 (it expected -1 from syswrite, how lame... :).
65 - support file descriptors in addition to file handles
66 in AnyEvent->io.
67 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
68 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
69 and PERL_ANYEVENT_CA_PATH.
70 - provide a sensible synopsis section for AnyEvent::TLS.
71 - add a "supported backends" section to the manpage.
72 - added simple io watcher test to testsuite, using a
73 portable_socketpair.
74 - tried to improve the stability of the Event::Lib backend,
75 YMMV.
76
774.8 Mon Jul 6 23:45:16 CEST 2009
78 - AnyEvent::DNS did not properly follow CNAME records with
79 uppercase targets.
80 - AnyEvent::DNS would errornously return AAAA records
81 with v4 mapped addresses (a faulty record) as ipv4 addresses,
82 causing AnyEvent::Socket to throw an exception.
83 - added new module AnyEvent::TLS for easier SSL/TLS context
84 creation, with many options including hostname verification,
85 secure default configuration, lots of documentation and,
86 predefined diffie-hellman keys for perfect forward security
87 and much more. get it while it's still fresh!
88 - use AnyEvent::TLS in AnyEvent::Handle for context management.
89 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
90 so AnyEvent::Socket users have smaller memory footprint
91 in the common case.
92 - add AnyEvent::Handle->push_shutdown method.
93 - add an additional error message parameter to AnyEvent::Handle's
94 on_error callback (for TLS, $! is still available).
95 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
96 - make AnyEvent::Handle more robust against early conenction
97 failures (during new), and return C<undef> in that case
98 from the constructor.
99 - AnyEvent::Handle will now try to load only JSON::XS first,
100 then fall back to JSON.
101 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
102 and because it was documented that way.
103
1044.451 Fri Jul 3 00:28:58 CEST 2009
105 - do not clear rbuf when shutting down an AnyEvent::Handle
106 object - doing so breaks AnyEvent::HTTP.
107
1084.45 Mon Jun 29 22:59:26 CEST 2009
109 - a write error could cause AnyEvent::Handle to create
110 an I/O watcher with an undefined $fh.
111 - special-case mapped ipv4 addresses in both
112 AnyEvent::Socket::format_address and parse_address,
113 to treat them just like ipv4 addresses.
114 - updated and overhauled the AnyEvent::Intro doc.
115 - implement AnyEvent::Socket::format_ipv4/ipv6.
116 - slightly speed up portable_pipe/socketpair.
117 - expand condvar begin/end documentation.
118
1194.42 Fri Jun 26 08:32:18 CEST 2009
120 - preliminary and neccesarily incomplete support for IO::Async.
121 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
122 case it was set to IGNORE, to ensure we can catch child statuses
123 even when the calling env acts stupidly.
124 - updated benchmarks with IO::Async, which performs very well.
125
1264.412 Wed Jun 24 01:35:57 CEST 2009
127 - support an "untaint" attribute for AnyEvent::DNS
128 and set it on the default resolver.
129 - implement and document AnyEvent::Impl::Perl::loop.
130 - remove all anyevent-env variables from %ENV
131 when running in tainted mode.
132 - mention and extend the IO::Lambda benchmark.
133
1344.411 Sun Jun 7 18:48:02 CEST 2009
135 - do not try to use F_SETFD on windows.
136
1374.41 Thu May 14 06:40:11 CEST 2009
138 - work around issues in older perls (5.8.0?) when
139 a signal handler is deleted from the %SIG hash.
140 - use POSIX::_exit in child test, to avoid running
141 destructors.
142 - speed up CHLD handling by relying on SIGCHLD being
143 synchronously delivered, even when we roll our own
144 implementation.
145 - AnyEvent::DNS: add the "dname" resource record name
146 for cosmetic reasons.
147
1484.4 Sun Apr 26 20:12:33 CEST 2009
149 - implemented idle watchers, where applicable.
150 - AnyEvent->time died when Event backend was in use.
151 - fix a memleak in the Tk backend.
152 - sped up Tk timer handling.
153 - clip negative "after" values to 0 in AnyEvent::Impl::Event
154 to avoid spamming out warning messages.
155 - fix Qt timers without interval.
156 - avoid Qt zero-timeout special case, in old libqt's.
157
1584.352 Mon Apr 20 16:31:11 CEST 2009
159 - fix AnyEvent::Strict error messages for child watchers.
160 - fix/update Makefile.PL "recommends" versions.
161 - add AnyEvent->now_update.
162
1634.351 Sat Apr 11 07:56:14 CEST 2009
164 - actually make the signal pipe work under win32.
165 - localise $! in signal handler to avoid changing $!, although
166 perl itself does not seem to save/restore errno either.
167 - set the cloexec flag on the signal pipe (normally set by perl too).
168
1694.35 Fri Mar 27 11:48:20 CET 2009
170 - event models relying on AnyEvent's signal watcher emulation did
171 invoke the callback asynchronously, contradicting documentation
172 and causing signals to get lost (this includes AnyEvent's own
173 event loop). AnyEvent now uses the standard pipe trick to make
174 callback execution synchronous to the event loop.
175 - AnyEvent::Handle didn't free TLS context data on DESTROY
176 (patch by Pavel Shaydo).
177 - work around the ever-incompatibly-changing API of MakeMaker.
178 - document that changing global variables without restoring them
179 is a bad idea in Perl (noted by Adam Rosenstein).
180 - AnyEvent::Strict now barfs if ->io is passed a file.
181
1824.34 Thu Feb 12 18:32:45 CET 2009
183 - separately buffer TLS read data, as otherwise the read queue could
184 deadlock as receiving data is not expected while draining the
185 read queue (which cna only happen with TLS).
186 - raise EBADMSG error on JSON decoding errors.
187 - fix some minor manpage bugs (reported by Maximilian Gaß).
188 - speed up select bitmask parsing quite a bit in the pure perl
189 backend.
190 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
191
1924.331 Tue Jan 6 21:07:25 CET 2009
193 - socketpair fails on many vista machines because vista has
194 completely broken accept/getpeername and getsockname functions,
195 so we provide our own socketpair emulation that kind of works
196 (AnyEvent::Util::portable_pipe).
197 - new function: AnyEvent::Util::portable_socketpair.
198 - take advantage of the Guard module if it exists.
199
2004.33 Fri Nov 21 02:35:40 CET 2008
201 - AnyEvent::Strict did errornously flag a fileno of 0 as
202 illegal.
203 - reduce memory usage and slightly speed up the pure perl backend
204 by only storing the file descriptor, not the file handle.
205 - add missing autocork method to AnyEvent::Handle (reported
206 by Adam Rosenstein).
207 - AnyEvent::DNS->resolve errornously documented an $rcode result
208 argument, but there isn't (spotted by Henrik Krohns).
209 - the naked truth about Tk - it's basically unmaintained.
210
2114.32 Mon Nov 3 22:46:32 CET 2008
212 - fix AnyEvent::Socket::resolve_sockaddr to properly support
213 unix sockets again and choose a proper default for
214 the socket type.
215 - fix call to resolve_sockaddr in tcp_connect.
216
2174.31 Thu Oct 30 04:41:48 CET 2008
218 - implemented AnyEvent::Handle->destroy method.
219 - hint about unexpected effects in TLS mode.
220 - speed up AnyEvent::Impl::Perl by using more arrays
221 instead of hashes without the slightest loss of readability :)
222 - work around a perl argument refcounting bug.
223 - with some perl patches applied, AnyEvent now seems leak-free.
224
2254.3 Fri Oct 3 09:18:43 CEST 2008
226 - AnyEvent will now install a no-op signal handler for SIGPIPE,
227 unless one has been installed already.
228 - warn about not loaidng AnyEvent::Impl::POE early enough
229 (patch by Adam Rosenstein).
230 - fixed a great number of bugs and corner cases in AnyEvent::Handle:
231 - fix a bug in where in SSL connect mode, the
232 client would first wait for some data by the server and
233 otherwise hang. (reported and analyzed in an absolutely
234 exemplary manner by Adam Rosenstein).
235 - fix a bug in where SSL EOF would not be treated
236 as stream EOF, putting the connection into a hung state.
237 (reported and analyzed in an absolutely exemplary manner by Adam
238 Rosenstein).
239 - fix a potential segfault when the TLS context
240 would go missing in a read callback (e.g. due to stoptls) -
241 Net::SSLeay of course makes no type checking whatsoever.
242 - AnyEvent::Handle will not stop the read watcher in TLS mode.
243 - AnyEvent::Handle->stoptls will send a TLS close notify
244 instead of simply destroying the stream now.
245 - fix a bug where large blocks of data written in TLS mode
246 would not be sent unless triggered by receives.
247 - on_drain will now take the tls write buffer into account.
248 - SSL operations have been streamlined a bit and should be faster
249 now, more to come, though.
250 - added a FAQ section, will document nontrivial issues in there.
251
2524.234 Mon Sep 29 04:08:13 CEST 2008
253 - fix child watcher documentation: only child exits will be reported,
254 no trace events.
255 - mention SIGPIPE in AnyEvent::Handle.
256 - perl backend will now use POSIX::times as monotonic clock
257 when available and the monotonic clock option is not.
258 - run even without Time::HiRes in some cases now, but do not
259 provide sub-second accuracy (all *supported* perl versions have
260 Time::HiRes).
261
2624.233 Fri Aug 22 01:48:05 CEST 2008
263 - fix a bug in the testsuite, causing a hang.
264 - clarified lots of AnyEvent::Handle constructor arguments.
265
2664.232 Thu Aug 21 20:44:25 CEST 2008
267 - fix a bug in AnyEvent::Handle that could cause two fatal error
268 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
269 (testcase by Leon Brocard).
270 - clarified on_eof/on_error documentation a bit.
271
2724.231 Tue Jul 29 13:12:15 CEST 2008
273 - remove some debugging code left in AnyEvent::Util::fork_call
274 (and no, it's impossible to implement with the broken windows
275 perls without resource leaks or worse).
276
2774.23 Tue Jul 29 12:19:59 CEST 2008
278 - document the first parameter passed to condvar callbacks to be
279 the callback.
280 - add AnyEvent::Socket::{ntoa,aton} aliases.
281 - optimize the AE::Handle->push_read (line) for the default
282 eol marker.
283 - optimize push_read (packstring|storable) for small packets.
284 - invoke on_error callback when no on_eof callback is set.
285 - fix a bug in push_read (storable) of unknown impact.
286
2874.22 Sun Jul 20 16:34:13 CEST 2008
288 - new function AnyEvent::Socket::parse_hostport.
289 - as the bulkheads at microsoft can't even get getprotobyname reliably
290 working on their shitty fucking broken os we need to hardcode
291 some common protocol numbers in AnyEvent::Socket. How can
292 people even bother with such a pile of shit as windows.
293
2944.21 Thu Jul 17 14:40:05 CEST 2008
295 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
296 which is incompatible to previous invocation syntax.
297 - work around more windows perl bugs in fork_call. I cannot imagine
298 why anybody would bother implementing fork in such an obviously
299 unusable way.
300 - avoid calling $do_retry if the request has already finished in
301 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
302 K. Sharifulin).
303
3044.2 Sat Jul 12 22:42:11 CEST 2008
305 - support an "interval" argument for repeating timers.
306 - fix a bug in the pure perl event loop that caused timers to
307 be delayed under some circumstances.
308 - implement optional argument checking via PERL_ANYEVENT_STRICT.
309 - not importing anything from Event causes it to malfunction
310 and not use Time::HiRes, work around this bug.
311 - more bug workarounds against the endlessly broken Tk module,
312 seems to work now on 32 bit machines, but less so on 64 bit.
313 - minor optimisations applied to most backends.
314
3154.161 Fri Jul 4 14:24:48 CEST 2008
316 - fixed the t/handle/02_write.t, it read after write, with the
317 assumption that all data could be written before read
318 (reported by Jim Razmus).
319
3204.160 Thu Jul 3 04:02:21 CEST 2008
321 - re-registering signal watchers after unregistering one did
322 not work if a backend falls back on AnyEvent's default
323 implementation (also affected child watchers).
324 - new AnyEvent::Handle options: autocork and no_delay.
325
3264.152 Sun Jun 22 14:15:44 CEST 2008
327 - allow for 32-bit perls that implement shifts differently
328 on different architectures in parse_ipv4 (reported and
329 analysed by Keiichi DAIBA).
330
3314.151 Fri Jun 6 17:34:24 CEST 2008
332 - make sure specifying _only_ on_read and never pushing reads
333 works.
334
3354.15 Fri Jun 6 13:00:46 CEST 2008
336 - the pure perl backend would keep some watchers alive when more than
337 one watcher was registered for the same fd.
338 - new "packstring" and "storable" read and write types
339 for AnyEvent::Handle.
340 - allow on_eof handler to be called after on_error with EPIPE returns.
341 - do not immediately call on_read callback in handle constructor.
342
3434.14 Thu Jun 5 20:29:31 CEST 2008
344 - Fixed a bug in DNS SRV priority sorting.
345 - AnyEvent::Util::guard now reports runtime errors while
346 executing the guard block as warnings.
347 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
348
3494.13 Thu Jun 5 00:47:59 CEST 2008
350 - AnyEvent::DNS only followed cname chains with length 2,
351 contrary to documentation. bumped it up to 10, thanks to
352 microsoft, the current broken-dns-config-king.
353 - AnyEvent::DNS didn't check for socket return status
354 but instead relied on perl not creating filehandles in that
355 case - too bad it gives you a bogus file handle (reported
356 and analysed by Vladimir Timofeev).
357 - fix queue management logic in AnyEvent::Handle:
358 when on_read was registered, the queue was empty
359 and no progress could be made AnyEvent::Handle would
360 enter an endless loop.
361 - correctly start reading again when the handle became
362 busy again after idling.
363 - correctly treat tls shut-downs as EOF condition
364 (for the time being, would be nice to have a callback for that).
365 - correctly call eof callback under all conditions
366 (hopefully).
367 - the timeout callback did not expect that $self can go
368 away any time.
369
3704.12 Tue Jun 3 10:58:04 CEST 2008
371 - include AnyEvent::Intro, a tutorial for anyevent,
372 anyevent::socket and anyevent::handle.
373 - allow more options in on_error.
374
3754.11 Fri May 30 23:42:25 CEST 2008
376 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
377 reverse_lookup and reverse_verify to replace it, support
378 v4mapped and v4compat addresses.
379 - provide more documentation for the resolver class.
380 - really replace longest run of :0: by :: in format_address,
381 also properly convert :: and ::1 again.
382 - support NAPTR record name and decode it.
383 - implement random weight sampling for SRV records, as per
384 rfc 2782.
385 - correctly abort on srv-record targets of ".".
386 - added AnyEvent::DNS::wait_for_slot.
387 - in the unlikely event of a virtual circuit connection
388 being invalidated by a delayed udp reply, AnyEvent::DNS
389 could die. this has been fixed.
390 - plug a probable memleak in the DNS vc code.
391 - use configured timeout also for VC request phase.
392 - implement timeout and max_outstanding methods
393 for resolver class.
394 - update version numbers in all modules.
395
3964.1 Thu May 29 05:45:40 CEST 2008
397 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
398 which mirrors their purpose better (old names still available
399 for a while). Also moved AnyEvent::DNS::addr to
400 AnyEvent::Socket::resolve_sockaddr.
401 - implement AnyEvent->time and AnyEvent->now.
402 - fix IPv6 support in pack_sockaddr.
403 - officially un-experimentalise all newly introduced code.
404 - support unix domain sockets everywhere by specifying
405 a host of "unix/" and the pathname as service.
406 - implement an activity timeout in AnyEvent::Handle.
407 - added a regex read type to AnyEvent::Handle.
408 - added a json read/write type to AnyEvent::Handle.
409 - always croak in on_error in AnyEvent::Handle.
410 - document how to subclass AnyEvent::Handle.
411 - implement AnyEvent::Util::fork_call.
412 - add support for IPv6 nameservers and nameserver statements.
413 - work around _yet_ _another_ windows perl bug in where empty
414 select masks cause errors in the pure perl backend (this
415 workaround cannot be implemented for the other event loops,
416 but EV also works around this).
417 - supply AnyEvent::Socket::address_type and make good use of it.
418 - clarify and fix the Handle documentation w.r.t $self vs. $handle
419 vs. callback arguments.
420 - add some recommends to the META.yml.
421
4224.05 Mon May 26 19:44:06 CEST 2008
423 - some platforms ignore the rfc and prepend an extra sa_len member to
424 the sockaddr structure, cater for those. also use sockaddr_family,
425 didn't know it was there, but comes in handy for the workaround.
426 - undo the import WIN32 hack.
427
4284.04 Mon May 26 08:03:31 CEST 2008
429 - try to work around yet another windows bug: failed connects are reported
430 as if it were out-of-band data. windows users: you suck. Only
431 EV, Glib, Event and the pure perl backend can handle this condition.
432 - optimize write algorithm in AnyEvent::Handle.
433 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
434 - tcp_connect now properly iterates through all targets.
435 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
436 support entirely. also hardcode some AF_INET6 constants for the
437 sake of perl 5.8.
438 - vastly improved nameserver/suffix detection algorithm for windows
439 (still a hack...)
440 - try to use a fallback dns server if no dns servers could be found.
441 - splatter around some local $SIG{__DIE__}'s to work around bad
442 die handlers.
443
4444.03 Sun May 25 05:05:57 CEST 2008
445 - work around perl bugs on windows, where perl returns undocumented
446 error codes for sysread, syswrite etc.
447 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
448 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
449 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
450 - updated documentation.
451 - lots of workarounds for perl 5.8 quirks.
452 - implement a more flexible type system for both reads and writes
453 in AnyEvent::Handle.
454 - support IPv6 in tcp_server.
455 - pass local hostname and port to the prepare callback in tcp_server.
456 - make types extendable in AnyEvent::Handle.
457 - croak, not die, in AnyEvent::Handle;
458
4594.0 Sat May 24 19:58:08 CEST 2008
460 - added (experimental) AnyEvent::DNS module.
461 - added (experimental) AnyEvent::Socket module.
462 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
463 - make condvar->end call ->send as documented.
464 - make condvar constructor support the documented cb => argument.
465 - overload condvars so you can use them instead of code references.
466 - support more of the POSIX forms instead of a pure dotted quad for
467 verbatim ip addresses.
468 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
469 - try very hard not to freeze on broken windows perls.
470 - make on_eof optional in AnyEvent::Handle.
471 - support service names in tcp_connect and tcp_server.
472 - work around netbsd/openbsd bug causing unix sockets not to
473 become writable until empty.
474 - only make a dummy read for error detection when getpeername
475 returns ENOTCONN.
476 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
477 override.
478 - new AnyEvent::Guard::cancel.
479
4803.5 Sat May 17 23:17:14 CEST 2008
481 - gracefully deal with systems not defining CLOCK_MONOTONIC
482 (as opposed to not supporting the clock itself, which worked in 3.41).
483 - added AnyEvent::Util::tcp_connect and AnyEvent::Util::tcp_server
484 helpers, to replace the ill-designed AnyEvent::Socket class.
485 - added some examples (eg/connect, eg/handle).
486 - added AnyEvent::Util::guard.
487 - AnyEvent::Util now exports most of its functions.
488 - fixed a bug in *_read_chunk, where the callback was called with $self as
489 first and second argument.
490 - simplified and fixed regex handling in AnyEvent::Handle::*_read_line.
491 - fix manpage condvars examples to use new syntax.
492 - considerably improve inet_aton with EV::ADNS backend to support
493 cname chains (common dns configuration bug, e.g. with www.google.de)
494 and aaaa records if no a record could be found.
495
4963.41 Sun May 11 19:53:13 CEST 2008
497 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
498
4993.4 Sun May 11 00:29:25 CEST 2008
500 - complete re-implementation of condvars with a hopefully much more useful
501 API (backward compatible functions exist but are undocumented).
502 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
503 (still subject to change).
504 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
505 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which
506 allows module authors to avoid forcing event loop detection.
507 (used by Coro::AIO, Coro::BDB, Coro::AnyEvent for example).
508 - remove coro backends: Coro now provides generic support for AnyEvent
509 via Coro::AnyEvent.
510
5113.3 Mon Apr 28 09:51:06 CEST 2008
512 - added AnyEvent::Handle, AnyEvent::Socket and AnyEvent::Util modules.
3 - fix a bug in the pure perl backend that kept watchers alive 513 - fix a bug in the pure perl backend that kept watchers alive
4 when multiple watchers were registered for the same fh/poll combo. 514 when multiple watchers were registered for the same fh/poll combo.
5 - add a benchmark section showing AnyEvent overhead and comparing 515 - add a benchmark section showing AnyEvent overhead and comparing
6 the different event loops with each other. 516 the different event loops with each other.
7 - prefer pure perl over tk when autoprobing, as it's about as fast, 517 - prefer pure perl over tk when autoprobing, as it's about as fast,
8 but doesn't crash with many watchers. 518 but doesn't crash with many watchers.
9 - make Qt support non-experimental. 519 - declare Qt support non-experimental.
10 - clarify the confusing section about the file descriptor being 520 - clarify the confusing section about the file descriptor being
11 kept alive. 521 kept alive.
12 - document the race between loading of an event module and 522 - document the race between loading of an event module and
13 child processes exiting. 523 child processes exiting.
14 - support POE as "backend" (with some caveats, POE is not generic 524 - support POE as "backend" (with some caveats, POE is not generic
15 enough, and darn slow). 525 enough, and darn slow).
16 - support Wx and Prima through POE. 526 - support Wx and Prima through POE.
17 - optimise perl backend to use 20% less memory and to be roughly 527 - optimise perl backend to use 20% less memory and take advantage
18 2-3 times faster in the common case of sparsely-active file 528 of typical timeout behaviour. It can now compete with
19 descriptors. It can now compete with select/poll-based C event 529 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 :). 530 faster than Event and Glib :).
21 - roughly cut EV memory use in half and increase its speed by 30%, 531 - roughly cut EV memory use in half and increase its speed by 30%,
22 by removing undocumented functionality. 532 by removing undocumented functionality. Did something similar
533 to other event loops where possible.
23 534
243.2 Thu Apr 24 10:10:40 CEST 2008 5353.2 Thu Apr 24 10:10:40 CEST 2008
25 - do not die when anyevent watchers are destroyed while running callbacks 536 - do not die when anyevent watchers are destroyed while running callbacks
26 in the pureperl backend (could only happen when two watchers are registered 537 in the pureperl backend (could only happen when two watchers are registered
27 for the same fh/poll combo). 538 for the same fh/poll combo).

Diff Legend

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