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

Comparing AnyEvent/Changes (file contents):
Revision 1.91 by root, Sat May 24 01:15:19 2008 UTC vs.
Revision 1.402 by root, Tue Jun 8 10:06:15 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: preferences, retry-callback, memleaks 35.271 Tue Jun 8 12:05:46 CEST 2010
43.51 4 - backport to perl 5.8.x.
5
65.27 Sun Jun 6 12:12:05 CEST 2010
7 - postpone differently in AnyEvent::Socket now, as
8 when not, canceling the connection attempt might fail
9 (found by Felix Antonius Wilhelm Ostmann).
10 - explicitly check for non-stream sockets in AE::Handle, too many
11 clueless people fell into the trap of this somehow working.
12 - simplified and reworked the "OTHER MODULES" section.
13 - better/more condvar examples.
14
155.261 Wed Apr 28 16:13:36 CEST 2010
16 - AF_INET6 was not properly used from Socket6 during configuration
17 time (found by Andy Grundman).
18
195.26 Mon Apr 12 04:49:35 CEST 2010
20 - don't generate spurious readyness notifications when select
21 returns EINTR in the pure perl backend (analysed and testcase
22 by Bas Denissen).
23 - give same examples in SYNOPSIS sections of AnyEvent and AE.
24 - provider faster implementations for AE::time/now/now_update for
25 EV and Perl backends.
26 - provide faster AE::cv and AE::time implementations for all
27 backends.
28 - no longer support lower/mixed-case signal names (this was
29 never documented nor universally supported).
30 - some more memory µ-optimisations, and somewhat less messy
31 code to cope with both AnyEvent and AE APIs.
32
335.251 Sat Mar 13 00:58:21 CET 2010
34 - make test failed if it was run before make install on perl <=
35 5.8.8, and older versions of perl make our live very difficult,
36 so write constants.pl during Makefile.PL time.
37
385.25 Sat Mar 13 00:23:14 CET 2010
39 - fix a race condition in AnyEvent::Handle that would cause
40 a "bio_read: unsupported method" error (in _another_
41 ssl connection) after user code threw an exception.
42 - added AnyEvent::Handle->destroyed method.
43 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
44 - the AnyEvent::Util::fh_nonblocking function worked only
45 by lucky accident on win32.
46 - smaller and faster AnyEvent::Util::fh_nonblocking.
47 - when the (required!) Time::HiRes module is missing, AnyEvent
48 did not fall back to built-in time properly.
49 - do not load Fcntl at runtime, saving memory and loading time.
50 - precompile a number of constants and use them instead of runtime
51 detection and eval.
52 - free detection code after detection and similar memory optimisations.
53 - Perl backend timer interval best effort drift has been improved
54 (same algorithm as EV).
55 - update unicode idna mapping table.
56
575.24 Tue Jan 5 11:39:43 CET 2010
58 - cygwin never reports errors from failed connects. 1.5 just gives
59 you continous readyness and EAGAIN, 1.7 is even more broken and
60 just hangs. work around both issues in a cygwin-specific hack.
61 - improve idle watcher documentation slightly.
62
635.23 Sun Dec 20 23:48:00 CET 2009
64 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
65 - implement punycode_encode/decode, idn_nameprep,
66 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
67 - provide $AE::VERSION.
68 - removed traces of "no strict 'refs'".
69
705.22 Sat Dec 5 03:51:13 CET 2009
71 - downgrade-or-fail in AnyEvent::Handle::push_write, to
72 diagnose encoding failures earlier and more succinctly.
73 (this works around bugs in perl, throwing away encoding info
74 when passing scalar data to extensions).
75 - add more examples to AnyEvent::Socket manpage.
76 - upgrade internal warning set to the same as common::sense 2.03.
77 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
78 (John Beppu).
79 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
80 - add google public dns servers to fallback server set and make sure
81 we load-balance properly between the three sets. also add all
82 fallback dns servers, not just a random one, to each dns config.
83
845.21 Thu Nov 19 02:48:47 CET 2009
85 - fix a problem where socket constants were called with parameters
86 (spotted by David Friedland).
87 - fork_call never use'd POSIX (reported by Daisuke Maki).
88 - improve perl 5.6 compatibility further (but it still won't work
89 unless you rip out everything but the core).
90 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
91 uses 10MB of RAM, but doesn't open a console window. *sigh*.
92
935.202 Wed Oct 14 22:35:44 CEST 2009
94 - AnyEvent::DNS would unexpectedly clobber $_ under windows
95 (analysed by Matthias Waldorf).
96 - AnyEvent::Handle::run_cmd can now pass the PID of the
97 newly-created process, which is much less useful than it might
98 sound (based on patch by Yann Kerherve).
99
1005.201 Tue Sep 29 12:09:25 CEST 2009
101 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
102 (reported by Torsten Foertsch).
103 - common::sense 2.0 could cause tcp_server to throw an exception
104 (analysed by elmex).
105
1065.2 Mon Sep 14 07:04:49 CEST 2009
107 - INCOMPATIBLE CHANGE: do no longer support register_read_type
108 and register_write_type in AnyEvent::Handle, instead support
109 package names (the facility was mostly abused).
110 - implement "packagename-as-read/write type" support in
111 AnyEvent::Handle.
112 - AnyEvent::Handle: new options "keepalive" and "oobinline".
113 - oobinline set by default to avoid security issues.
114 - the pure-perl event loop backend wrongly detected times() fallback
115 support (spotted by Pavel Boldin).
116
1175.12 Tue Sep 1 20:26:50 CEST 2009
118 - be more lenient when parsing resolv.conf files, as some people
119 use hashmarks as comment indicator inside directives (reported
120 by Michael S. Fischer).
121 - use same set of warnings as common::sense 2.0.
122 - fix a potential 32 bit overflow issue due to perl having problems
123 with large hex constants in the Perl backend.
124 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
125 by default.
126 - fix some prototypes.
127 - (optionally) require Async::Interrupt 1.02.
128 - replace opendns fallbacks by verizon/level3 public dns servers.
129
1305.112 Fri Aug 21 13:59:04 CEST 2009
131 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
132 using starttls with an illegal/uncreatable tls context.
133 - port to broken win32 perls.
134
1355.111 Sun Aug 16 18:44:36 CEST 2009
136 - perl errornously requires sockaddr_un structures returned by the
137 kernel to have a specific length. work around this bug by padding
138 sockaddr_un structures (found to fail on OS X, as perl only uses
139 the correct code for GNU/Linux, but likely to fail on other
140 OSes as well).
141
1425.11 Wed Aug 12 17:49:37 CEST 2009
143 - fixed a bug in run_cmd with input from filename.
144 - create a json coder object when none is given in write direction
145 to reduce dependency on JSON.
146
1475.1 Tue Aug 11 03:17:32 CEST 2009
148 - tcp_connect now reports the correct error in $!, not an unrelated
149 one (was broken since 4.91).
150 - AE::Handle did not properly clear rtimeout/wtimeout watchers
151 when retrying a tcp_connect.
152 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
153 - fixed wrong documentation in AE::Socket.
154
1555.01 Mon Aug 10 03:16:32 CEST 2009
156 - last release broke EV child watchers :/.
157
1585.0 Sun Aug 9 17:32:09 CEST 2009
159 - officially introduce and document the AE API.
160 - lowering the timeout at runtime did not have immediate effect
161 in AnyEvent::Handle.
162 - AE::Handle now has separate and independent read and write timeout
163 settings.
164 - probe for Guard module when AnyEvent::Util is loaded, not at
165 runtime (Event doesn't like eval "use" from callbacks, and it avoids
166 exporting the probe function).
167 - Impl::IOAsync's io watchers did not use the filehandle emulation
168 code to work around it's watcher limits.
169 - work around IO::Async's problems with overloaded objects.
170 - try to disable wearnings inside IO::Async, it's just too buggy.
171 - do no longer use Event by default, only EV or Perl, Event is too buggy.
172 - hackishly provide fast AE:: interface when EV or Perl is the backend.
173 - inherit VERSION from AnyEvent::VERSION in most modules.
174
1754.91 Thu Aug 6 15:42:45 CEST 2009
176 - AE::Handle::starttls could get out of sync when the read buffer
177 already contains some TLS handshake.
178 - AE::Handle did not properly free the TLS session, causing bigger
179 memory leaks in Net::SSLeay than expected (reported by toaster).
180 - AE::Socket::tcp_connect will now artificially delay invoking
181 the callback to avoid returning after invoking the callback.
182 - convert many internal watcher uses to AE API.
183
1844.9 Sat Aug 1 11:07:01 CEST 2009
185 - Glib silently fails when registering a timeout with
186 a negative value, so avoid that.
187 - call condvar callback immediately when it is set
188 after the condition is already signalled.
189 - check rbuf_max condition only after trying to consume
190 data.
191 - normalise signal numbers to names when using the ae's
192 signal handling, but do not document this (yet).
193 - pure perl signal emulation did not properly set nonblocking
194 mode on the signal pipe on win32 (this is unlikely to have caused
195 any issues).
196 - new module: AnyEvent::Debug.
197 - AnyEvent::Strict now checks that a signal was specified
198 by name (not name or number).
199 - reduce memory footprint in typical cases by ~50kb by
200 only compiling signal code when necessary.
201 - add AnyEvent::Handle->rbuf_max.
202 - grab the AE namespace for future new API, implement stubs for
203 said future/alternative API.
204 - new function: AnyEvent::Socket::format_hostport.
205
2064.881 Tue Jul 28 12:51:53 CEST 2009
207 - work around a bug in local in pre-5.10 perls, causing
208 AnyEvent::Handle to recurse when it should not
209 (analysed by elmex).
210
2114.88 Tue Jul 28 04:04:37 CEST 2009
212 - re-bless the handle into a dummy package after calling
213 AnyEvent::Handle::destroy, so the user does not need to check
214 for errors after every push_write etc.
215 - do not attempt to run t/02_signals.t on obviously broken
216 platforms.
217
2184.87 Sun Jul 26 02:06:16 CEST 2009
219 - do not attempt to linger when there is no longer a valid fh.
220 - enforce tls mode to be either accept or connect, do not simply
221 segfault in Net::SSLeay.
222 - AnyEvent::Handle can now call tcp_connect itself (new parameters
223 connect, on_prepare, on_connect and on_connect_error). Updated
224 tutorial accordingly.
225 - add AnyEvent::Impl::Irssi backend.
226
2274.86 Mon Jul 20 23:52:29 CEST 2009
228 - since the verbose warning is not enough, explicitly document
229 that versions before 1.33 of Net::SSLeay are not secure.
230 - work around signal handling races in Event and (...) Event::Lib.
231 - try to align signal-race timer to full-second boundaries.
232 - work around Tk not liking negative timeouts.
233 - don't complain of different grades of environmental unfriendlyness
234 in IO::Async.
235
2364.85 Sat Jul 18 06:16:14 CEST 2009
237 - nail the signal race problem in perl once and for all
238 (see $AnyEvent::MAX_SIGNAL_LATENCY).
239 - take advantage of Async::Interrupt if it is available.
240 - load Time::HiRes and Guard modules on demand only.
241 - add optional/recommended modules section to AnyEvent
242 documentation.
243 - reduce memory usage considerably (and reduce startup penalty)
244 by not using "strict", "warnings" and "overload" modules.
245 - work around buggy windows/openbsd perls and provide EBADMSG
246 and EPROTO ourselves when missing.
247 - improve perl 5.6 compatibility of the core event loop.
248 - made Net::SSLeay version 1.33 a soft requirement.
249
2504.83 Fri Jul 17 16:56:26 CEST 2009
251 - implement AnyEvent::Socket::getprotobyname.
252 - AnyEvent::CondVar's will now detect recursive blocking
253 waits and will croak, as too many people fall into
254 this trap.
255 - AnyEvent::Handle will now call ->destroy on itself after
256 executing the on_error callback, instead of doing some
257 half-baked internal shutdown, for fatal errors.
258 - clarify on_eof behaviour w.r.t. the read queue and
259 on_read callbacks.
260 - ignore some possible spurious wake-ups in tcp_connect.
261
2624.82 Sat Jul 11 00:34:55 CEST 2009
263 - POE and Event backends didn't accept some callable objects as
264 callbacks.
265 - use Config module instead of POSIX module to detect signal names
266 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
267 module doesn't even have all POSIX signals :/.
268 - use more workarounds around the many refcnt/corruption bugs in
269 Event::Lib.
270 - work around a race condition in perl's select, causing t/03_child.t
271 to rarely fail.
272
2734.81 Thu Jul 9 10:30:30 CEST 2009
274 - AnyEvent::Handle didn't properly diagnose write errors
275 (it expected -1 from syswrite, how lame... :).
276 - support file descriptors in addition to file handles
277 in AnyEvent->io.
278 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
279 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
280 and PERL_ANYEVENT_CA_PATH.
281 - provide a sensible synopsis section for AnyEvent::TLS.
282 - add a "supported backends" section to the manpage.
283 - added simple io watcher test to testsuite, using a
284 portable_socketpair.
285 - tried to improve the stability of the Event::Lib backend,
286 YMMV.
287
2884.8 Mon Jul 6 23:45:16 CEST 2009
289 - AnyEvent::DNS did not properly follow CNAME records with
290 uppercase targets.
291 - AnyEvent::DNS would errornously return AAAA records
292 with v4 mapped addresses (a faulty record) as ipv4 addresses,
293 causing AnyEvent::Socket to throw an exception.
294 - added new module AnyEvent::TLS for easier SSL/TLS context
295 creation, with many options including hostname verification,
296 secure default configuration, lots of documentation and,
297 predefined diffie-hellman keys for perfect forward security
298 and much more. get it while it's still fresh!
299 - use AnyEvent::TLS in AnyEvent::Handle for context management.
300 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
301 so AnyEvent::Socket users have smaller memory footprint
302 in the common case.
303 - add AnyEvent::Handle->push_shutdown method.
304 - add an additional error message parameter to AnyEvent::Handle's
305 on_error callback (for TLS, $! is still available).
306 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
307 - make AnyEvent::Handle more robust against early conenction
308 failures (during new), and return C<undef> in that case
309 from the constructor.
310 - AnyEvent::Handle will now try to load only JSON::XS first,
311 then fall back to JSON.
312 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
313 and because it was documented that way.
314
3154.451 Fri Jul 3 00:28:58 CEST 2009
316 - do not clear rbuf when shutting down an AnyEvent::Handle
317 object - doing so breaks AnyEvent::HTTP.
318
3194.45 Mon Jun 29 22:59:26 CEST 2009
320 - a write error could cause AnyEvent::Handle to create
321 an I/O watcher with an undefined $fh.
322 - special-case mapped ipv4 addresses in both
323 AnyEvent::Socket::format_address and parse_address,
324 to treat them just like ipv4 addresses.
325 - updated and overhauled the AnyEvent::Intro doc.
326 - implement AnyEvent::Socket::format_ipv4/ipv6.
327 - slightly speed up portable_pipe/socketpair.
328 - expand condvar begin/end documentation.
329
3304.42 Fri Jun 26 08:32:18 CEST 2009
331 - preliminary and neccesarily incomplete support for IO::Async.
332 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
333 case it was set to IGNORE, to ensure we can catch child statuses
334 even when the calling env acts stupidly.
335 - updated benchmarks with IO::Async, which performs very well.
336
3374.412 Wed Jun 24 01:35:57 CEST 2009
338 - support an "untaint" attribute for AnyEvent::DNS
339 and set it on the default resolver.
340 - implement and document AnyEvent::Impl::Perl::loop.
341 - remove all anyevent-env variables from %ENV
342 when running in tainted mode.
343 - mention and extend the IO::Lambda benchmark.
344
3454.411 Sun Jun 7 18:48:02 CEST 2009
346 - do not try to use F_SETFD on windows.
347
3484.41 Thu May 14 06:40:11 CEST 2009
349 - work around issues in older perls (5.8.0?) when
350 a signal handler is deleted from the %SIG hash.
351 - use POSIX::_exit in child test, to avoid running
352 destructors.
353 - speed up CHLD handling by relying on SIGCHLD being
354 synchronously delivered, even when we roll our own
355 implementation.
356 - AnyEvent::DNS: add the "dname" resource record name
357 for cosmetic reasons.
358
3594.4 Sun Apr 26 20:12:33 CEST 2009
360 - implemented idle watchers, where applicable.
361 - AnyEvent->time died when Event backend was in use.
362 - fix a memleak in the Tk backend.
363 - sped up Tk timer handling.
364 - clip negative "after" values to 0 in AnyEvent::Impl::Event
365 to avoid spamming out warning messages.
366 - fix Qt timers without interval.
367 - avoid Qt zero-timeout special case, in old libqt's.
368
3694.352 Mon Apr 20 16:31:11 CEST 2009
370 - fix AnyEvent::Strict error messages for child watchers.
371 - fix/update Makefile.PL "recommends" versions.
372 - add AnyEvent->now_update.
373
3744.351 Sat Apr 11 07:56:14 CEST 2009
375 - actually make the signal pipe work under win32.
376 - localise $! in signal handler to avoid changing $!, although
377 perl itself does not seem to save/restore errno either.
378 - set the cloexec flag on the signal pipe (normally set by perl too).
379
3804.35 Fri Mar 27 11:48:20 CET 2009
381 - event models relying on AnyEvent's signal watcher emulation did
382 invoke the callback asynchronously, contradicting documentation
383 and causing signals to get lost (this includes AnyEvent's own
384 event loop). AnyEvent now uses the standard pipe trick to make
385 callback execution synchronous to the event loop.
386 - AnyEvent::Handle didn't free TLS context data on DESTROY
387 (patch by Pavel Shaydo).
388 - work around the ever-incompatibly-changing API of MakeMaker.
389 - document that changing global variables without restoring them
390 is a bad idea in Perl (noted by Adam Rosenstein).
391 - AnyEvent::Strict now barfs if ->io is passed a file.
392
3934.34 Thu Feb 12 18:32:45 CET 2009
394 - separately buffer TLS read data, as otherwise the read queue could
395 deadlock as receiving data is not expected while draining the
396 read queue (which cna only happen with TLS).
397 - raise EBADMSG error on JSON decoding errors.
398 - fix some minor manpage bugs (reported by Maximilian Gaß).
399 - speed up select bitmask parsing quite a bit in the pure perl
400 backend.
401 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
402
4034.331 Tue Jan 6 21:07:25 CET 2009
404 - socketpair fails on many vista machines because vista has
405 completely broken accept/getpeername and getsockname functions,
406 so we provide our own socketpair emulation that kind of works
407 (AnyEvent::Util::portable_pipe).
408 - new function: AnyEvent::Util::portable_socketpair.
409 - take advantage of the Guard module if it exists.
410
4114.33 Fri Nov 21 02:35:40 CET 2008
412 - AnyEvent::Strict did errornously flag a fileno of 0 as
413 illegal.
414 - reduce memory usage and slightly speed up the pure perl backend
415 by only storing the file descriptor, not the file handle.
416 - add missing autocork method to AnyEvent::Handle (reported
417 by Adam Rosenstein).
418 - AnyEvent::DNS->resolve errornously documented an $rcode result
419 argument, but there isn't (spotted by Henrik Krohns).
420 - the naked truth about Tk - it's basically unmaintained.
421
4224.32 Mon Nov 3 22:46:32 CET 2008
423 - fix AnyEvent::Socket::resolve_sockaddr to properly support
424 unix sockets again and choose a proper default for
425 the socket type.
426 - fix call to resolve_sockaddr in tcp_connect.
427
4284.31 Thu Oct 30 04:41:48 CET 2008
429 - implemented AnyEvent::Handle->destroy method.
430 - hint about unexpected effects in TLS mode.
431 - speed up AnyEvent::Impl::Perl by using more arrays
432 instead of hashes without the slightest loss of readability :)
433 - work around a perl argument refcounting bug.
434 - with some perl patches applied, AnyEvent now seems leak-free.
435
4364.3 Fri Oct 3 09:18:43 CEST 2008
437 - AnyEvent will now install a no-op signal handler for SIGPIPE,
438 unless one has been installed already.
439 - warn about not loaidng AnyEvent::Impl::POE early enough
440 (patch by Adam Rosenstein).
441 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
442 - fix a bug in where in SSL connect mode, the
443 client would first wait for some data by the server and
444 otherwise hang. (reported and analysed in an absolutely
445 exemplary manner by Adam Rosenstein).
446 - fix a bug in where SSL EOF would not be treated
447 as stream EOF, putting the connection into a hung state.
448 (reported and analysed in an absolutely exemplary manner by Adam
449 Rosenstein).
450 - fix a potential segfault when the TLS context
451 would go missing in a read callback (e.g. due to stoptls) -
452 Net::SSLeay of course makes no type checking whatsoever.
453 - AnyEvent::Handle will not stop the read watcher in TLS mode.
454 - AnyEvent::Handle->stoptls will send a TLS close notify
455 instead of simply destroying the stream now.
456 - fix a bug where large blocks of data written in TLS mode
457 would not be sent unless triggered by receives.
458 - on_drain will now take the tls write buffer into account.
459 - SSL operations have been streamlined a bit and should be faster
460 now, more to come, though.
461 - added a FAQ section, will document nontrivial issues in there.
462
4634.234 Mon Sep 29 04:08:13 CEST 2008
464 - fix child watcher documentation: only child exits will be reported,
465 no trace events.
466 - mention SIGPIPE in AnyEvent::Handle.
467 - perl backend will now use POSIX::times as monotonic clock
468 when available and the monotonic clock option is not.
469 - run even without Time::HiRes in some cases now, but do not
470 provide sub-second accuracy (all *supported* perl versions have
471 Time::HiRes).
472
4734.233 Fri Aug 22 01:48:05 CEST 2008
474 - fix a bug in the testsuite, causing a hang.
475 - clarified lots of AnyEvent::Handle constructor arguments.
476
4774.232 Thu Aug 21 20:44:25 CEST 2008
478 - fix a bug in AnyEvent::Handle that could cause two fatal error
479 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
480 (testcase by Leon Brocard).
481 - clarified on_eof/on_error documentation a bit.
482
4834.231 Tue Jul 29 13:12:15 CEST 2008
484 - remove some debugging code left in AnyEvent::Util::fork_call
485 (and no, it's impossible to implement with the broken windows
486 perls without resource leaks or worse).
487
4884.23 Tue Jul 29 12:19:59 CEST 2008
489 - document the first parameter passed to condvar callbacks to be
490 the callback.
491 - add AnyEvent::Socket::{ntoa,aton} aliases.
492 - optimize the AE::Handle->push_read (line) for the default
493 eol marker.
494 - optimize push_read (packstring|storable) for small packets.
495 - invoke on_error callback when no on_eof callback is set.
496 - fix a bug in push_read (storable) of unknown impact.
497
4984.22 Sun Jul 20 16:34:13 CEST 2008
499 - new function AnyEvent::Socket::parse_hostport.
500 - as the bulkheads at microsoft can't even get getprotobyname reliably
501 working on their shitty fucking broken os we need to hardcode
502 some common protocol numbers in AnyEvent::Socket. How can
503 people even bother with such a pile of shit as windows.
504
5054.21 Thu Jul 17 14:40:05 CEST 2008
506 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
507 which is incompatible to previous invocation syntax.
508 - work around more windows perl bugs in fork_call. I cannot imagine
509 why anybody would bother implementing fork in such an obviously
510 unusable way.
511 - avoid calling $do_retry if the request has already finished in
512 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
513 K. Sharifulin).
514
5154.2 Sat Jul 12 22:42:11 CEST 2008
516 - support an "interval" argument for repeating timers.
517 - fix a bug in the pure perl event loop that caused timers to
518 be delayed under some circumstances.
519 - implement optional argument checking via PERL_ANYEVENT_STRICT.
520 - not importing anything from Event causes it to malfunction
521 and not use Time::HiRes, work around this bug.
522 - more bug workarounds against the endlessly broken Tk module,
523 seems to work now on 32 bit machines, but less so on 64 bit.
524 - minor optimisations applied to most backends.
525
5264.161 Fri Jul 4 14:24:48 CEST 2008
527 - fixed the t/handle/02_write.t, it read after write, with the
528 assumption that all data could be written before read
529 (reported by Jim Razmus).
530
5314.160 Thu Jul 3 04:02:21 CEST 2008
532 - re-registering signal watchers after unregistering one did
533 not work if a backend falls back on AnyEvent's default
534 implementation (also affected child watchers).
535 - new AnyEvent::Handle options: autocork and no_delay.
536
5374.152 Sun Jun 22 14:15:44 CEST 2008
538 - allow for 32-bit perls that implement shifts differently
539 on different architectures in parse_ipv4 (reported and
540 analysed by Keiichi DAIBA).
541
5424.151 Fri Jun 6 17:34:24 CEST 2008
543 - make sure specifying _only_ on_read and never pushing reads
544 works.
545
5464.15 Fri Jun 6 13:00:46 CEST 2008
547 - the pure perl backend would keep some watchers alive when more than
548 one watcher was registered for the same fd.
549 - new "packstring" and "storable" read and write types
550 for AnyEvent::Handle.
551 - allow on_eof handler to be called after on_error with EPIPE returns.
552 - do not immediately call on_read callback in handle constructor.
553
5544.14 Thu Jun 5 20:29:31 CEST 2008
555 - Fixed a bug in DNS SRV priority sorting.
556 - AnyEvent::Util::guard now reports runtime errors while
557 executing the guard block as warnings.
558 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
559
5604.13 Thu Jun 5 00:47:59 CEST 2008
561 - AnyEvent::DNS only followed cname chains with length 2,
562 contrary to documentation. bumped it up to 10, thanks to
563 microsoft, the current broken-dns-config-king.
564 - AnyEvent::DNS didn't check for socket return status
565 but instead relied on perl not creating filehandles in that
566 case - too bad it gives you a bogus file handle (reported
567 and analysed by Vladimir Timofeev).
568 - fix queue management logic in AnyEvent::Handle:
569 when on_read was registered, the queue was empty
570 and no progress could be made AnyEvent::Handle would
571 enter an endless loop.
572 - correctly start reading again when the handle became
573 busy again after idling.
574 - correctly treat tls shut-downs as EOF condition
575 (for the time being, would be nice to have a callback for that).
576 - correctly call eof callback under all conditions
577 (hopefully).
578 - the timeout callback did not expect that $self can go
579 away any time.
580
5814.12 Tue Jun 3 10:58:04 CEST 2008
582 - include AnyEvent::Intro, a tutorial for anyevent,
583 anyevent::socket and anyevent::handle.
584 - allow more options in on_error.
585
5864.11 Fri May 30 23:42:25 CEST 2008
587 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
588 reverse_lookup and reverse_verify to replace it, support
589 v4mapped and v4compat addresses.
590 - provide more documentation for the resolver class.
591 - really replace longest run of :0: by :: in format_address,
592 also properly convert :: and ::1 again.
593 - support NAPTR record name and decode it.
594 - implement random weight sampling for SRV records, as per
595 rfc 2782.
596 - correctly abort on srv-record targets of ".".
597 - added AnyEvent::DNS::wait_for_slot.
598 - in the unlikely event of a virtual circuit connection
599 being invalidated by a delayed udp reply, AnyEvent::DNS
600 could die. this has been fixed.
601 - plug a probable memleak in the DNS vc code.
602 - use configured timeout also for VC request phase.
603 - implement timeout and max_outstanding methods
604 for resolver class.
605 - update version numbers in all modules.
606
6074.1 Thu May 29 05:45:40 CEST 2008
608 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
609 which mirrors their purpose better (old names still available
610 for a while). Also moved AnyEvent::DNS::addr to
611 AnyEvent::Socket::resolve_sockaddr.
612 - implement AnyEvent->time and AnyEvent->now.
613 - fix IPv6 support in pack_sockaddr.
614 - officially un-experimentalise all newly introduced code.
615 - support unix domain sockets everywhere by specifying
616 a host of "unix/" and the pathname as service.
617 - implement an activity timeout in AnyEvent::Handle.
618 - added a regex read type to AnyEvent::Handle.
619 - added a json read/write type to AnyEvent::Handle.
620 - always croak in on_error in AnyEvent::Handle.
621 - document how to subclass AnyEvent::Handle.
622 - implement AnyEvent::Util::fork_call.
623 - add support for IPv6 nameservers and nameserver statements.
624 - work around _yet_ _another_ windows perl bug in where empty
625 select masks cause errors in the pure perl backend (this
626 workaround cannot be implemented for the other event loops,
627 but EV also works around this).
628 - supply AnyEvent::Socket::address_type and make good use of it.
629 - clarify and fix the Handle documentation w.r.t $self vs. $handle
630 vs. callback arguments.
631 - add some recommends to the META.yml.
632
6334.05 Mon May 26 19:44:06 CEST 2008
634 - some platforms ignore the rfc and prepend an extra sa_len member to
635 the sockaddr structure, cater for those. also use sockaddr_family,
636 didn't know it was there, but comes in handy for the workaround.
637 - undo the import WIN32 hack.
638
6394.04 Mon May 26 08:03:31 CEST 2008
640 - try to work around yet another windows bug: failed connects are reported
641 as if it were out-of-band data. windows users: you suck. Only
642 EV, Glib, Event and the pure perl backend can handle this condition.
643 - optimize write algorithm in AnyEvent::Handle.
644 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
645 - tcp_connect now properly iterates through all targets.
646 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
647 support entirely. also hardcode some AF_INET6 constants for the
648 sake of perl 5.8.
649 - vastly improved nameserver/suffix detection algorithm for windows
650 (still a hack...)
651 - try to use a fallback dns server if no dns servers could be found.
652 - splatter around some local $SIG{__DIE__}'s to work around bad
653 die handlers.
654
6554.03 Sun May 25 05:05:57 CEST 2008
656 - work around perl bugs on windows, where perl returns undocumented
657 error codes for sysread, syswrite etc.
658 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
659 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
660 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
661 - updated documentation.
662 - lots of workarounds for perl 5.8 quirks.
663 - implement a more flexible type system for both reads and writes
664 in AnyEvent::Handle.
665 - support IPv6 in tcp_server.
666 - pass local hostname and port to the prepare callback in tcp_server.
667 - make types extendable in AnyEvent::Handle.
668 - croak, not die, in AnyEvent::Handle;
669
6704.0 Sat May 24 19:58:08 CEST 2008
5 - added (experimental) AnyEvent::DNS module. 671 - added (experimental) AnyEvent::DNS module.
6 - added (experimental) AnyEvent::Socket module. 672 - added (experimental) AnyEvent::Socket module.
673 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
7 - make condvar->end call ->send as documented. 674 - make condvar->end call ->send as documented.
8 - make condvar constructor support the documented cb => argument. 675 - make condvar constructor support the documented cb => argument.
676 - overload condvars so you can use them instead of code references.
9 - support more of the POSIX forms instead of a pure dooted quad for 677 - support more of the POSIX forms instead of a pure dotted quad for
10 verbatim ip addresses. 678 verbatim ip addresses.
11 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. 679 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
12 - try very hard not to freeze on broken windows perls. 680 - try very hard not to freeze on broken windows perls.
13 - make on_eof optional in AnyEvent::Handle. 681 - make on_eof optional in AnyEvent::Handle.
14 - support service names in tcp_connect and tcp_server. 682 - support service names in tcp_connect and tcp_server.
15 - work around netbsd/openbsd bug causing unix sockets not to 683 - work around netbsd/openbsd bug causing unix sockets not to
16 become writable untill empty. 684 become writable until empty.
17 - only make a dummy read for error detection when getpeername 685 - only make a dummy read for error detection when getpeername
18 returns ENOTCONN. 686 returns ENOTCONN.
19 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 687 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
20 override. 688 override.
21 - new AnyEvent::Guard::cancel. 689 - new AnyEvent::Guard::cancel.
22 690
233.5 Sat May 17 23:17:14 CEST 2008 6913.5 Sat May 17 23:17:14 CEST 2008
24 - gracefully deal with systems not defining CLOCK_MONOTONIC 692 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
38 706
393.41 Sun May 11 19:53:13 CEST 2008 7073.41 Sun May 11 19:53:13 CEST 2008
40 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 708 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
41 709
423.4 Sun May 11 00:29:25 CEST 2008 7103.4 Sun May 11 00:29:25 CEST 2008
43 - complete re-implementation of condvars with a hopefully much mroe useful 711 - complete re-implementation of condvars with a hopefully much more useful
44 API (backward compatible functions exist but are undocumented). 712 API (backward compatible functions exist but are undocumented).
45 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 713 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
46 (still subject to change). 714 (still subject to change).
47 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 715 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
48 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 716 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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