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

Comparing AnyEvent/Changes (file contents):
Revision 1.109 by root, Mon May 26 04:30:30 2008 UTC vs.
Revision 1.377 by root, Sat Jan 30 21:27:59 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: exceptbit 3 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
4 - smaller and faster AnyEvent::Util::fh_nonblocking.
5
65.24 Tue Jan 5 11:39:43 CET 2010
7 - cygwin never reports errors from failed connects. 1.5 just gives
8 you continous readyness and EAGAIN, 1.7 is even more broken and
9 just hangs. work around both issues in a cygwin-specific hack.
10 - improve idle watcher documentation slightly.
11
125.23 Sun Dec 20 23:48:00 CET 2009
13 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
14 - implement punycode_encode/decode, idn_nameprep,
15 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
16 - provide $AE::VERSION.
17 - removed traces of "no strict 'refs'".
18
195.22 Sat Dec 5 03:51:13 CET 2009
20 - downgrade-or-fail in AnyEvent::Handle::push_write, to
21 diagnose encoding failures earlier and more succinctly.
22 (this works around bugs in perl, throwing away encoding info
23 when passing scalar data to extensions).
24 - add more examples to AnyEvent::Socket manpage.
25 - upgrade internal warning set to the same as common::sense 2.03.
26 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
27 (John Beppu).
28 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
29 - add google public dns servers to fallback server set and make sure
30 we load-balance properly between the three sets. also add all
31 fallback dns servers, not just a random one, to each dns config.
32
335.21 Thu Nov 19 02:48:47 CET 2009
34 - fix a problem where socket constants were called with parameters
35 (spotted by David Friedland).
36 - fork_call never use'd POSIX (reported by Daisuke Maki).
37 - improve perl 5.6 compatibility further (but it still won't work
38 unless you rip out everything but the core).
39 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
40 uses 10MB of RAM, but doesn't open a console window. *sigh*.
41
425.202 Wed Oct 14 22:35:44 CEST 2009
43 - AnyEvent::DNS would unexpectedly clobber $_ under windows
44 (analysed by Matthias Waldorf).
45 - AnyEvent::Handle::run_cmd can now pass the PID of the
46 newly-created process, which is much less useful than it might
47 sound (based on patch by Yann Kerherve).
48
495.201 Tue Sep 29 12:09:25 CEST 2009
50 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
51 (reported by Torsten Foertsch).
52 - common::sense 2.0 could cause tcp_server to throw an exception
53 (analysed by elmex).
54
555.2 Mon Sep 14 07:04:49 CEST 2009
56 - INCOMPATIBLE CHANGE: do no longer support register_read_type
57 and register_write_type in AnyEvent::Handle, instead support
58 package names (the facility was mostly abused).
59 - implement "packagename-as-read/write type" support in
60 AnyEvent::Handle.
61 - AnyEvent::Handle: new options "keepalive" and "oobinline".
62 - oobinline set by default to avoid security issues.
63 - the pure-perl event loop backend wrongly detected times() fallback
64 support (spotted by Pavel Boldin).
65
665.12 Tue Sep 1 20:26:50 CEST 2009
67 - be more lenient when parsing resolv.conf files, as some people
68 use hashmarks as comment indicator inside directives (reported
69 by Michael S. Fischer).
70 - use same set of warnings as common::sense 2.0.
71 - fix a potential 32 bit overflow issue due to perl having problems
72 with large hex constants in the Perl backend.
73 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
74 by default.
75 - fix some prototypes.
76 - (optionally) require Async::Interrupt 1.02.
77 - replace opendns fallbacks by verizon/level3 public dns servers.
78
795.112 Fri Aug 21 13:59:04 CEST 2009
80 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
81 using starttls with an illegal/uncreatable tls context.
82 - port to broken win32 perls.
83
845.111 Sun Aug 16 18:44:36 CEST 2009
85 - perl errornously requires sockaddr_un structures returned by the
86 kernel to have a specific length. work around this bug by padding
87 sockaddr_un structures (found to fail on OS X, as perl only uses
88 the correct code for GNU/Linux, but likely to fail on other
89 OSes as well).
90
915.11 Wed Aug 12 17:49:37 CEST 2009
92 - fixed a bug in run_cmd with input from filename.
93 - create a json coder object when none is given in write direction
94 to reduce dependency on JSON.
95
965.1 Tue Aug 11 03:17:32 CEST 2009
97 - tcp_connect now reports the correct error in $!, not an unrelated
98 one (was broken since 4.91).
99 - AE::Handle did not properly clear rtimeout/wtimeout watchers
100 when retrying a tcp_connect.
101 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
102 - fixed wrong documentation in AE::Socket.
103
1045.01 Mon Aug 10 03:16:32 CEST 2009
105 - last release broke EV child watchers :/.
106
1075.0 Sun Aug 9 17:32:09 CEST 2009
108 - officially introduce and document the AE API.
109 - lowering the timeout at runtime did not have immediate effect
110 in AnyEvent::Handle.
111 - AE::Handle now has separate and independent read and write timeout
112 settings.
113 - probe for Guard module when AnyEvent::Util is loaded, not at
114 runtime (Event doesn't like eval "use" from callbacks, and it avoids
115 exporting the probe function).
116 - Impl::IOAsync's io watchers did not use the filehandle emulation
117 code to work around it's watcher limits.
118 - work around IO::Async's problems with overloaded objects.
119 - try to disable wearnings inside IO::Async, it's just too buggy.
120 - do no longer use Event by default, only EV or Perl, Event is too buggy.
121 - hackishly provide fast AE:: interface when EV or Perl is the backend.
122 - inherit VERSION from AnyEvent::VERSION in most modules.
123
1244.91 Thu Aug 6 15:42:45 CEST 2009
125 - AE::Handle::starttls could get out of sync when the read buffer
126 already contains some TLS handshake.
127 - AE::Handle did not properly free the TLS session, causing bigger
128 memory leaks in Net::SSLeay than expected (reported by toaster).
129 - AE::Socket::tcp_connect will now artificially delay invoking
130 the callback to avoid returning after invoking the callback.
131 - convert many internal watcher uses to AE API.
132
1334.9 Sat Aug 1 11:07:01 CEST 2009
134 - Glib silently fails when registering a timeout with
135 a negative value, so avoid that.
136 - call condvar callback immediately when it is set
137 after the condition is already signalled.
138 - check rbuf_max condition only after trying to consume
139 data.
140 - normalise signal numbers to names when using the ae's
141 signal handling, but do not document this (yet).
142 - pure perl signal emulation did not properly set nonblocking
143 mode on the signal pipe on win32 (this is unlikely to have caused
144 any issues).
145 - new module: AnyEvent::Debug.
146 - AnyEvent::Strict now checks that a signal was specified
147 by name (not name or number).
148 - reduce memory footprint in typical cases by ~50kb by
149 only compiling signal code when necessary.
150 - add AnyEvent::Handle->rbuf_max.
151 - grab the AE namespace for future new API, implement stubs for
152 said future/alternative API.
153 - new function: AnyEvent::Socket::format_hostport.
154
1554.881 Tue Jul 28 12:51:53 CEST 2009
156 - work around a bug in local in pre-5.10 perls, causing
157 AnyEvent::Handle to recurse when it should not
158 (analysed by elmex).
159
1604.88 Tue Jul 28 04:04:37 CEST 2009
161 - re-bless the handle into a dummy package after calling
162 AnyEvent::Handle::destroy, so the user does not need to check
163 for errors after every push_write etc.
164 - do not attempt to run t/02_signals.t on obviously broken
165 platforms.
166
1674.87 Sun Jul 26 02:06:16 CEST 2009
168 - do not attempt to linger when there is no longer a valid fh.
169 - enforce tls mode to be either accept or connect, do not simply
170 segfault in Net::SSLeay.
171 - AnyEvent::Handle can now call tcp_connect itself (new parameters
172 connect, on_prepare, on_connect and on_connect_error). Updated
173 tutorial accordingly.
174 - add AnyEvent::Impl::Irssi backend.
175
1764.86 Mon Jul 20 23:52:29 CEST 2009
177 - since the verbose warning is not enough, explicitly document
178 that versions before 1.33 of Net::SSLeay are not secure.
179 - work around signal handling races in Event and (...) Event::Lib.
180 - try to align signal-race timer to full-second boundaries.
181 - work around Tk not liking negative timeouts.
182 - don't complain of different grades of environmental unfriendlyness
183 in IO::Async.
184
1854.85 Sat Jul 18 06:16:14 CEST 2009
186 - nail the signal race problem in perl once and for all
187 (see $AnyEvent::MAX_SIGNAL_LATENCY).
188 - take advantage of Async::Interrupt if it is available.
189 - load Time::HiRes and Guard modules on demand only.
190 - add optional/recommended modules section to AnyEvent
191 documentation.
192 - reduce memory usage considerably (and reduce startup penalty)
193 by not using "strict", "warnings" and "overload" modules.
194 - work around buggy windows/openbsd perls and provide EBADMSG
195 and EPROTO ourselves when missing.
196 - improve perl 5.6 compatibility of the core event loop.
197 - made Net::SSLeay version 1.33 a soft requirement.
198
1994.83 Fri Jul 17 16:56:26 CEST 2009
200 - implement AnyEvent::Socket::getprotobyname.
201 - AnyEvent::CondVar's will now detect recursive blocking
202 waits and will croak, as too many people fall into
203 this trap.
204 - AnyEvent::Handle will now call ->destroy on itself after
205 executing the on_error callback, instead of doing some
206 half-baked internal shutdown, for fatal errors.
207 - clarify on_eof behaviour w.r.t. the read queue and
208 on_read callbacks.
209 - ignore some possible spurious wake-ups in tcp_connect.
210
2114.82 Sat Jul 11 00:34:55 CEST 2009
212 - POE and Event backends didn't accept some callable objects as
213 callbacks.
214 - use Config module instead of POSIX module to detect signal names
215 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
216 module doesn't even have all POSIX signals :/.
217 - use more workarounds around the many refcnt/corruption bugs in
218 Event::Lib.
219 - work around a race condition in perl's select, causing t/03_child.t
220 to rarely fail.
221
2224.81 Thu Jul 9 10:30:30 CEST 2009
223 - AnyEvent::Handle didn't properly diagnose write errors
224 (it expected -1 from syswrite, how lame... :).
225 - support file descriptors in addition to file handles
226 in AnyEvent->io.
227 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
228 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
229 and PERL_ANYEVENT_CA_PATH.
230 - provide a sensible synopsis section for AnyEvent::TLS.
231 - add a "supported backends" section to the manpage.
232 - added simple io watcher test to testsuite, using a
233 portable_socketpair.
234 - tried to improve the stability of the Event::Lib backend,
235 YMMV.
236
2374.8 Mon Jul 6 23:45:16 CEST 2009
238 - AnyEvent::DNS did not properly follow CNAME records with
239 uppercase targets.
240 - AnyEvent::DNS would errornously return AAAA records
241 with v4 mapped addresses (a faulty record) as ipv4 addresses,
242 causing AnyEvent::Socket to throw an exception.
243 - added new module AnyEvent::TLS for easier SSL/TLS context
244 creation, with many options including hostname verification,
245 secure default configuration, lots of documentation and,
246 predefined diffie-hellman keys for perfect forward security
247 and much more. get it while it's still fresh!
248 - use AnyEvent::TLS in AnyEvent::Handle for context management.
249 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
250 so AnyEvent::Socket users have smaller memory footprint
251 in the common case.
252 - add AnyEvent::Handle->push_shutdown method.
253 - add an additional error message parameter to AnyEvent::Handle's
254 on_error callback (for TLS, $! is still available).
255 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
256 - make AnyEvent::Handle more robust against early conenction
257 failures (during new), and return C<undef> in that case
258 from the constructor.
259 - AnyEvent::Handle will now try to load only JSON::XS first,
260 then fall back to JSON.
261 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
262 and because it was documented that way.
263
2644.451 Fri Jul 3 00:28:58 CEST 2009
265 - do not clear rbuf when shutting down an AnyEvent::Handle
266 object - doing so breaks AnyEvent::HTTP.
267
2684.45 Mon Jun 29 22:59:26 CEST 2009
269 - a write error could cause AnyEvent::Handle to create
270 an I/O watcher with an undefined $fh.
271 - special-case mapped ipv4 addresses in both
272 AnyEvent::Socket::format_address and parse_address,
273 to treat them just like ipv4 addresses.
274 - updated and overhauled the AnyEvent::Intro doc.
275 - implement AnyEvent::Socket::format_ipv4/ipv6.
276 - slightly speed up portable_pipe/socketpair.
277 - expand condvar begin/end documentation.
278
2794.42 Fri Jun 26 08:32:18 CEST 2009
280 - preliminary and neccesarily incomplete support for IO::Async.
281 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
282 case it was set to IGNORE, to ensure we can catch child statuses
283 even when the calling env acts stupidly.
284 - updated benchmarks with IO::Async, which performs very well.
285
2864.412 Wed Jun 24 01:35:57 CEST 2009
287 - support an "untaint" attribute for AnyEvent::DNS
288 and set it on the default resolver.
289 - implement and document AnyEvent::Impl::Perl::loop.
290 - remove all anyevent-env variables from %ENV
291 when running in tainted mode.
292 - mention and extend the IO::Lambda benchmark.
293
2944.411 Sun Jun 7 18:48:02 CEST 2009
295 - do not try to use F_SETFD on windows.
296
2974.41 Thu May 14 06:40:11 CEST 2009
298 - work around issues in older perls (5.8.0?) when
299 a signal handler is deleted from the %SIG hash.
300 - use POSIX::_exit in child test, to avoid running
301 destructors.
302 - speed up CHLD handling by relying on SIGCHLD being
303 synchronously delivered, even when we roll our own
304 implementation.
305 - AnyEvent::DNS: add the "dname" resource record name
306 for cosmetic reasons.
307
3084.4 Sun Apr 26 20:12:33 CEST 2009
309 - implemented idle watchers, where applicable.
310 - AnyEvent->time died when Event backend was in use.
311 - fix a memleak in the Tk backend.
312 - sped up Tk timer handling.
313 - clip negative "after" values to 0 in AnyEvent::Impl::Event
314 to avoid spamming out warning messages.
315 - fix Qt timers without interval.
316 - avoid Qt zero-timeout special case, in old libqt's.
317
3184.352 Mon Apr 20 16:31:11 CEST 2009
319 - fix AnyEvent::Strict error messages for child watchers.
320 - fix/update Makefile.PL "recommends" versions.
321 - add AnyEvent->now_update.
322
3234.351 Sat Apr 11 07:56:14 CEST 2009
324 - actually make the signal pipe work under win32.
325 - localise $! in signal handler to avoid changing $!, although
326 perl itself does not seem to save/restore errno either.
327 - set the cloexec flag on the signal pipe (normally set by perl too).
328
3294.35 Fri Mar 27 11:48:20 CET 2009
330 - event models relying on AnyEvent's signal watcher emulation did
331 invoke the callback asynchronously, contradicting documentation
332 and causing signals to get lost (this includes AnyEvent's own
333 event loop). AnyEvent now uses the standard pipe trick to make
334 callback execution synchronous to the event loop.
335 - AnyEvent::Handle didn't free TLS context data on DESTROY
336 (patch by Pavel Shaydo).
337 - work around the ever-incompatibly-changing API of MakeMaker.
338 - document that changing global variables without restoring them
339 is a bad idea in Perl (noted by Adam Rosenstein).
340 - AnyEvent::Strict now barfs if ->io is passed a file.
341
3424.34 Thu Feb 12 18:32:45 CET 2009
343 - separately buffer TLS read data, as otherwise the read queue could
344 deadlock as receiving data is not expected while draining the
345 read queue (which cna only happen with TLS).
346 - raise EBADMSG error on JSON decoding errors.
347 - fix some minor manpage bugs (reported by Maximilian Gaß).
348 - speed up select bitmask parsing quite a bit in the pure perl
349 backend.
350 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
351
3524.331 Tue Jan 6 21:07:25 CET 2009
353 - socketpair fails on many vista machines because vista has
354 completely broken accept/getpeername and getsockname functions,
355 so we provide our own socketpair emulation that kind of works
356 (AnyEvent::Util::portable_pipe).
357 - new function: AnyEvent::Util::portable_socketpair.
358 - take advantage of the Guard module if it exists.
359
3604.33 Fri Nov 21 02:35:40 CET 2008
361 - AnyEvent::Strict did errornously flag a fileno of 0 as
362 illegal.
363 - reduce memory usage and slightly speed up the pure perl backend
364 by only storing the file descriptor, not the file handle.
365 - add missing autocork method to AnyEvent::Handle (reported
366 by Adam Rosenstein).
367 - AnyEvent::DNS->resolve errornously documented an $rcode result
368 argument, but there isn't (spotted by Henrik Krohns).
369 - the naked truth about Tk - it's basically unmaintained.
370
3714.32 Mon Nov 3 22:46:32 CET 2008
372 - fix AnyEvent::Socket::resolve_sockaddr to properly support
373 unix sockets again and choose a proper default for
374 the socket type.
375 - fix call to resolve_sockaddr in tcp_connect.
376
3774.31 Thu Oct 30 04:41:48 CET 2008
378 - implemented AnyEvent::Handle->destroy method.
379 - hint about unexpected effects in TLS mode.
380 - speed up AnyEvent::Impl::Perl by using more arrays
381 instead of hashes without the slightest loss of readability :)
382 - work around a perl argument refcounting bug.
383 - with some perl patches applied, AnyEvent now seems leak-free.
384
3854.3 Fri Oct 3 09:18:43 CEST 2008
386 - AnyEvent will now install a no-op signal handler for SIGPIPE,
387 unless one has been installed already.
388 - warn about not loaidng AnyEvent::Impl::POE early enough
389 (patch by Adam Rosenstein).
390 - fixed a great number of bugs and corner cases in AnyEvent::Handle:
391 - fix a bug in where in SSL connect mode, the
392 client would first wait for some data by the server and
393 otherwise hang. (reported and analysed in an absolutely
394 exemplary manner by Adam Rosenstein).
395 - fix a bug in where SSL EOF would not be treated
396 as stream EOF, putting the connection into a hung state.
397 (reported and analysed in an absolutely exemplary manner by Adam
398 Rosenstein).
399 - fix a potential segfault when the TLS context
400 would go missing in a read callback (e.g. due to stoptls) -
401 Net::SSLeay of course makes no type checking whatsoever.
402 - AnyEvent::Handle will not stop the read watcher in TLS mode.
403 - AnyEvent::Handle->stoptls will send a TLS close notify
404 instead of simply destroying the stream now.
405 - fix a bug where large blocks of data written in TLS mode
406 would not be sent unless triggered by receives.
407 - on_drain will now take the tls write buffer into account.
408 - SSL operations have been streamlined a bit and should be faster
409 now, more to come, though.
410 - added a FAQ section, will document nontrivial issues in there.
411
4124.234 Mon Sep 29 04:08:13 CEST 2008
413 - fix child watcher documentation: only child exits will be reported,
414 no trace events.
415 - mention SIGPIPE in AnyEvent::Handle.
416 - perl backend will now use POSIX::times as monotonic clock
417 when available and the monotonic clock option is not.
418 - run even without Time::HiRes in some cases now, but do not
419 provide sub-second accuracy (all *supported* perl versions have
420 Time::HiRes).
421
4224.233 Fri Aug 22 01:48:05 CEST 2008
423 - fix a bug in the testsuite, causing a hang.
424 - clarified lots of AnyEvent::Handle constructor arguments.
425
4264.232 Thu Aug 21 20:44:25 CEST 2008
427 - fix a bug in AnyEvent::Handle that could cause two fatal error
428 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
429 (testcase by Leon Brocard).
430 - clarified on_eof/on_error documentation a bit.
431
4324.231 Tue Jul 29 13:12:15 CEST 2008
433 - remove some debugging code left in AnyEvent::Util::fork_call
434 (and no, it's impossible to implement with the broken windows
435 perls without resource leaks or worse).
436
4374.23 Tue Jul 29 12:19:59 CEST 2008
438 - document the first parameter passed to condvar callbacks to be
439 the callback.
440 - add AnyEvent::Socket::{ntoa,aton} aliases.
441 - optimize the AE::Handle->push_read (line) for the default
442 eol marker.
443 - optimize push_read (packstring|storable) for small packets.
444 - invoke on_error callback when no on_eof callback is set.
445 - fix a bug in push_read (storable) of unknown impact.
446
4474.22 Sun Jul 20 16:34:13 CEST 2008
448 - new function AnyEvent::Socket::parse_hostport.
449 - as the bulkheads at microsoft can't even get getprotobyname reliably
450 working on their shitty fucking broken os we need to hardcode
451 some common protocol numbers in AnyEvent::Socket. How can
452 people even bother with such a pile of shit as windows.
453
4544.21 Thu Jul 17 14:40:05 CEST 2008
455 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
456 which is incompatible to previous invocation syntax.
457 - work around more windows perl bugs in fork_call. I cannot imagine
458 why anybody would bother implementing fork in such an obviously
459 unusable way.
460 - avoid calling $do_retry if the request has already finished in
461 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
462 K. Sharifulin).
463
4644.2 Sat Jul 12 22:42:11 CEST 2008
465 - support an "interval" argument for repeating timers.
466 - fix a bug in the pure perl event loop that caused timers to
467 be delayed under some circumstances.
468 - implement optional argument checking via PERL_ANYEVENT_STRICT.
469 - not importing anything from Event causes it to malfunction
470 and not use Time::HiRes, work around this bug.
471 - more bug workarounds against the endlessly broken Tk module,
472 seems to work now on 32 bit machines, but less so on 64 bit.
473 - minor optimisations applied to most backends.
474
4754.161 Fri Jul 4 14:24:48 CEST 2008
476 - fixed the t/handle/02_write.t, it read after write, with the
477 assumption that all data could be written before read
478 (reported by Jim Razmus).
479
4804.160 Thu Jul 3 04:02:21 CEST 2008
481 - re-registering signal watchers after unregistering one did
482 not work if a backend falls back on AnyEvent's default
483 implementation (also affected child watchers).
484 - new AnyEvent::Handle options: autocork and no_delay.
485
4864.152 Sun Jun 22 14:15:44 CEST 2008
487 - allow for 32-bit perls that implement shifts differently
488 on different architectures in parse_ipv4 (reported and
489 analysed by Keiichi DAIBA).
490
4914.151 Fri Jun 6 17:34:24 CEST 2008
492 - make sure specifying _only_ on_read and never pushing reads
493 works.
494
4954.15 Fri Jun 6 13:00:46 CEST 2008
496 - the pure perl backend would keep some watchers alive when more than
497 one watcher was registered for the same fd.
498 - new "packstring" and "storable" read and write types
499 for AnyEvent::Handle.
500 - allow on_eof handler to be called after on_error with EPIPE returns.
501 - do not immediately call on_read callback in handle constructor.
502
5034.14 Thu Jun 5 20:29:31 CEST 2008
504 - Fixed a bug in DNS SRV priority sorting.
505 - AnyEvent::Util::guard now reports runtime errors while
506 executing the guard block as warnings.
507 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
508
5094.13 Thu Jun 5 00:47:59 CEST 2008
510 - AnyEvent::DNS only followed cname chains with length 2,
511 contrary to documentation. bumped it up to 10, thanks to
512 microsoft, the current broken-dns-config-king.
513 - AnyEvent::DNS didn't check for socket return status
514 but instead relied on perl not creating filehandles in that
515 case - too bad it gives you a bogus file handle (reported
516 and analysed by Vladimir Timofeev).
517 - fix queue management logic in AnyEvent::Handle:
518 when on_read was registered, the queue was empty
519 and no progress could be made AnyEvent::Handle would
520 enter an endless loop.
521 - correctly start reading again when the handle became
522 busy again after idling.
523 - correctly treat tls shut-downs as EOF condition
524 (for the time being, would be nice to have a callback for that).
525 - correctly call eof callback under all conditions
526 (hopefully).
527 - the timeout callback did not expect that $self can go
528 away any time.
529
5304.12 Tue Jun 3 10:58:04 CEST 2008
531 - include AnyEvent::Intro, a tutorial for anyevent,
532 anyevent::socket and anyevent::handle.
533 - allow more options in on_error.
534
5354.11 Fri May 30 23:42:25 CEST 2008
536 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
537 reverse_lookup and reverse_verify to replace it, support
538 v4mapped and v4compat addresses.
539 - provide more documentation for the resolver class.
540 - really replace longest run of :0: by :: in format_address,
541 also properly convert :: and ::1 again.
542 - support NAPTR record name and decode it.
543 - implement random weight sampling for SRV records, as per
544 rfc 2782.
545 - correctly abort on srv-record targets of ".".
546 - added AnyEvent::DNS::wait_for_slot.
547 - in the unlikely event of a virtual circuit connection
548 being invalidated by a delayed udp reply, AnyEvent::DNS
549 could die. this has been fixed.
550 - plug a probable memleak in the DNS vc code.
551 - use configured timeout also for VC request phase.
552 - implement timeout and max_outstanding methods
553 for resolver class.
554 - update version numbers in all modules.
555
5564.1 Thu May 29 05:45:40 CEST 2008
557 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
558 which mirrors their purpose better (old names still available
559 for a while). Also moved AnyEvent::DNS::addr to
560 AnyEvent::Socket::resolve_sockaddr.
561 - implement AnyEvent->time and AnyEvent->now.
562 - fix IPv6 support in pack_sockaddr.
563 - officially un-experimentalise all newly introduced code.
564 - support unix domain sockets everywhere by specifying
565 a host of "unix/" and the pathname as service.
566 - implement an activity timeout in AnyEvent::Handle.
567 - added a regex read type to AnyEvent::Handle.
568 - added a json read/write type to AnyEvent::Handle.
569 - always croak in on_error in AnyEvent::Handle.
570 - document how to subclass AnyEvent::Handle.
571 - implement AnyEvent::Util::fork_call.
572 - add support for IPv6 nameservers and nameserver statements.
573 - work around _yet_ _another_ windows perl bug in where empty
574 select masks cause errors in the pure perl backend (this
575 workaround cannot be implemented for the other event loops,
576 but EV also works around this).
577 - supply AnyEvent::Socket::address_type and make good use of it.
578 - clarify and fix the Handle documentation w.r.t $self vs. $handle
579 vs. callback arguments.
580 - add some recommends to the META.yml.
581
5824.05 Mon May 26 19:44:06 CEST 2008
583 - some platforms ignore the rfc and prepend an extra sa_len member to
584 the sockaddr structure, cater for those. also use sockaddr_family,
585 didn't know it was there, but comes in handy for the workaround.
586 - undo the import WIN32 hack.
587
5884.04 Mon May 26 08:03:31 CEST 2008
4 - try to work around yet another windows bug: failed connects are reported 589 - try to work around yet another windows bug: failed connects are reported
5 as if it were out-of-band data. windows users: you suck. 590 as if it were out-of-band data. windows users: you suck. Only
591 EV, Glib, Event and the pure perl backend can handle this condition.
592 - optimize write algorithm in AnyEvent::Handle.
6 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly. 593 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
594 - tcp_connect now properly iterates through all targets.
7 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6 595 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
8 support entirely. also hardcode some AF_INEt6 constants for the 596 support entirely. also hardcode some AF_INET6 constants for the
9 sake of perl 5.8. 597 sake of perl 5.8.
10 - vastly improved nameserver/suffix detection algorithm for windows 598 - vastly improved nameserver/suffix detection algorithm for windows
11 (still a hack...) 599 (still a hack...)
12 - try to use a fallback dns server if no dns servers could be found. 600 - try to use a fallback dns server if no dns servers could be found.
13 - splatter around some local $SIG{__DIE__}'s to work around bad 601 - splatter around some local $SIG{__DIE__}'s to work around bad
… …
40 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. 628 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
41 - try very hard not to freeze on broken windows perls. 629 - try very hard not to freeze on broken windows perls.
42 - make on_eof optional in AnyEvent::Handle. 630 - make on_eof optional in AnyEvent::Handle.
43 - support service names in tcp_connect and tcp_server. 631 - support service names in tcp_connect and tcp_server.
44 - work around netbsd/openbsd bug causing unix sockets not to 632 - work around netbsd/openbsd bug causing unix sockets not to
45 become writable untill empty. 633 become writable until empty.
46 - only make a dummy read for error detection when getpeername 634 - only make a dummy read for error detection when getpeername
47 returns ENOTCONN. 635 returns ENOTCONN.
48 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 636 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
49 override. 637 override.
50 - new AnyEvent::Guard::cancel. 638 - new AnyEvent::Guard::cancel.
51 639
523.5 Sat May 17 23:17:14 CEST 2008 6403.5 Sat May 17 23:17:14 CEST 2008
53 - gracefully deal with systems not defining CLOCK_MONOTONIC 641 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
67 655
683.41 Sun May 11 19:53:13 CEST 2008 6563.41 Sun May 11 19:53:13 CEST 2008
69 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 657 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
70 658
713.4 Sun May 11 00:29:25 CEST 2008 6593.4 Sun May 11 00:29:25 CEST 2008
72 - complete re-implementation of condvars with a hopefully much mroe useful 660 - complete re-implementation of condvars with a hopefully much more useful
73 API (backward compatible functions exist but are undocumented). 661 API (backward compatible functions exist but are undocumented).
74 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 662 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
75 (still subject to change). 663 (still subject to change).
76 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 664 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
77 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 665 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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