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

Comparing AnyEvent/Changes (file contents):
Revision 1.94 by root, Sat May 24 17:48:38 2008 UTC vs.
Revision 1.339 by root, Fri Aug 21 11:59:24 2009 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.0 35.112 Fri Aug 21 13:59:04 CEST 2009
4 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
5 using starttls with an illegal/uncreatable tls context.
6 - port to broken win32 perls.
7
85.111 Sun Aug 16 18:44:36 CEST 2009
9 - perl errornously requires sockaddr_un structures returned by the
10 kernel to have a specific length. work around this bug by padding
11 sockaddr_un structures (found to fail on OS X, as perl only uses
12 the correct code for GNU/Linux, but likely to fail on other
13 OSes as well).
14
155.11 Wed Aug 12 17:49:37 CEST 2009
16 - fixed a bug in run_cmd with input from filename.
17 - create a json coder object when none is given in write direction
18 to reduce dependency on JSON.
19
205.1 Tue Aug 11 03:17:32 CEST 2009
21 - tcp_connect now reports the correct error in $!, not an unrelated
22 one (was broken since 4.91).
23 - AE::Handle did not properly clear rtimeout/wtimeout watchers
24 when retrying a tcp_connect.
25 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
26 - fixed wrong documentation in AE::Socket.
27
285.01 Mon Aug 10 03:16:32 CEST 2009
29 - last release broke EV child watchers :/.
30
315.0 Sun Aug 9 17:32:09 CEST 2009
32 - officially introduce and document the AE API.
33 - lowering the timeout at runtime did not have immediate effect
34 in AnyEvent::Handle.
35 - AE::Handle now has separate and independent read and write timeout
36 settings.
37 - probe for Guard module when AnyEvent::Util is loaded, not at
38 runtime (Event doesn't like eval "use" from callbacks, and it avoids
39 exporting the probe function).
40 - Impl::IOAsync's io watchers did not use the filehandle emulation
41 code to work around it's watcher limits.
42 - work around IO::Async's problems with overloaded objects.
43 - try to disable wearnings inside IO::Async, it's just too buggy.
44 - do no longer use Event by default, only EV or Perl, Event is too buggy.
45 - hackishly provide fast AE:: interface when EV or Perl is the backend.
46 - inherit VERSION from AnyEvent::VERSION in most modules.
47
484.91 Thu Aug 6 15:42:45 CEST 2009
49 - AE::Handle::starttls could get out of sync when the read buffer
50 already contains some TLS handshake.
51 - AE::Handle did not properly free the TLS session, causing bigger
52 memory leaks in Net::SSLeay than expected (reported by toaster).
53 - AE::Socket::tcp_connect will now artificially delay invoking
54 the callback to avoid returning after invoking the callback.
55 - convert many internal watcher uses to AE API.
56
574.9 Sat Aug 1 11:07:01 CEST 2009
58 - Glib silently fails when registering a timeout with
59 a negative value, so avoid that.
60 - call condvar callback immediately when it is set
61 after the condition is already signalled.
62 - check rbuf_max condition only after trying to consume
63 data.
64 - normalise signal numbers to names when using the ae's
65 signal handling, but do not document this (yet).
66 - pure perl signal emulation did not properly set nonblocking
67 mode on the signal pipe on win32 (this is unlikely to have caused
68 any issues).
69 - new module: AnyEvent::Debug.
70 - AnyEvent::Strict now checks that a signal was specified
71 by name (not name or number).
72 - reduce memory footprint in typical cases by ~50kb by
73 only compiling signal code when necessary.
74 - add AnyEvent::Handle->rbuf_max.
75 - grab the AE namespace for future new API, implement stubs for
76 said future/alternative API.
77 - new function: AnyEvent::Socket::format_hostport.
78
794.881 Tue Jul 28 12:51:53 CEST 2009
80 - work around a bug in local in pre-5.10 perls, causing
81 AnyEvent::Handle to recurse when it should not
82 (analyzed by elmex).
83
844.88 Tue Jul 28 04:04:37 CEST 2009
85 - re-bless the handle into a dummy package after calling
86 AnyEvent::Handle::destroy, so the user does not need to check
87 for errors after every push_write etc.
88 - do not attempt to run t/02_signals.t on obviously broken
89 platforms.
90
914.87 Sun Jul 26 02:06:16 CEST 2009
92 - do not attempt to linger when there is no longer a valid fh.
93 - enforce tls mode to be either accept or connect, do not simply
94 segfault in Net::SSLeay.
95 - AnyEvent::Handle can now call tcp_connect itself (new parameters
96 connect, on_prepare, on_connect and on_connect_error). Updated
97 tutorial accordingly.
98 - add AnyEvent::Impl::Irssi backend.
99
1004.86 Mon Jul 20 23:52:29 CEST 2009
101 - since the verbose warning is not enough, explicitly document
102 that versions before 1.33 of Net::SSLeay are not secure.
103 - work around signal handling races in Event and (...) Event::Lib.
104 - try to align signal-race timer to full-second boundaries.
105 - work around Tk not liking negative timeouts.
106 - don't complain of different grades of environmental unfriendlyness
107 in IO::Async.
108
1094.85 Sat Jul 18 06:16:14 CEST 2009
110 - nail the signal race problem in perl once and for all
111 (see $AnyEvent::MAX_SIGNAL_LATENCY).
112 - take advantage of Async::Interrupt if it is available.
113 - load Time::HiRes and Guard modules on demand only.
114 - add optional/recommended modules section to AnyEvent
115 documentation.
116 - reduce memory usage considerably (and reduce startup penalty)
117 by not using "strict", "warnings" and "overload" modules.
118 - work around buggy windows/openbsd perls and provide EBADMSG
119 and EPROTO ourselves when missing.
120 - improve perl 5.6 compatibility of the core event loop.
121 - made Net::SSLeay version 1.33 a soft requirement.
122
1234.83 Fri Jul 17 16:56:26 CEST 2009
124 - implement AnyEvent::Socket::getprotobyname.
125 - AnyEvent::CondVar's will now detect recursive blocking
126 waits and will croak, as too many people fall into
127 this trap.
128 - AnyEvent::Handle will now call ->destroy on itself after
129 executing the on_error callback, instead of doing some
130 half-baked internal shutdown, for fatal errors.
131 - clarify on_eof behaviour w.r.t. the read queue and
132 on_read callbacks.
133 - ignore some possible spurious wake-ups in tcp_connect.
134
1354.82 Sat Jul 11 00:34:55 CEST 2009
136 - POE and Event backends didn't accept some callable objects as
137 callbacks.
138 - use Config module instead of POSIX module to detect signal names
139 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
140 module doesn't even have all POSIX signals :/.
141 - use more workarounds around the many refcnt/corruption bugs in
142 Event::Lib.
143 - work around a race condition in perl's select, causing t/03_child.t
144 to rarely fail.
145
1464.81 Thu Jul 9 10:30:30 CEST 2009
147 - AnyEvent::Handle didn't properly diagnose write errors
148 (it expected -1 from syswrite, how lame... :).
149 - support file descriptors in addition to file handles
150 in AnyEvent->io.
151 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
152 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
153 and PERL_ANYEVENT_CA_PATH.
154 - provide a sensible synopsis section for AnyEvent::TLS.
155 - add a "supported backends" section to the manpage.
156 - added simple io watcher test to testsuite, using a
157 portable_socketpair.
158 - tried to improve the stability of the Event::Lib backend,
159 YMMV.
160
1614.8 Mon Jul 6 23:45:16 CEST 2009
162 - AnyEvent::DNS did not properly follow CNAME records with
163 uppercase targets.
164 - AnyEvent::DNS would errornously return AAAA records
165 with v4 mapped addresses (a faulty record) as ipv4 addresses,
166 causing AnyEvent::Socket to throw an exception.
167 - added new module AnyEvent::TLS for easier SSL/TLS context
168 creation, with many options including hostname verification,
169 secure default configuration, lots of documentation and,
170 predefined diffie-hellman keys for perfect forward security
171 and much more. get it while it's still fresh!
172 - use AnyEvent::TLS in AnyEvent::Handle for context management.
173 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
174 so AnyEvent::Socket users have smaller memory footprint
175 in the common case.
176 - add AnyEvent::Handle->push_shutdown method.
177 - add an additional error message parameter to AnyEvent::Handle's
178 on_error callback (for TLS, $! is still available).
179 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
180 - make AnyEvent::Handle more robust against early conenction
181 failures (during new), and return C<undef> in that case
182 from the constructor.
183 - AnyEvent::Handle will now try to load only JSON::XS first,
184 then fall back to JSON.
185 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
186 and because it was documented that way.
187
1884.451 Fri Jul 3 00:28:58 CEST 2009
189 - do not clear rbuf when shutting down an AnyEvent::Handle
190 object - doing so breaks AnyEvent::HTTP.
191
1924.45 Mon Jun 29 22:59:26 CEST 2009
193 - a write error could cause AnyEvent::Handle to create
194 an I/O watcher with an undefined $fh.
195 - special-case mapped ipv4 addresses in both
196 AnyEvent::Socket::format_address and parse_address,
197 to treat them just like ipv4 addresses.
198 - updated and overhauled the AnyEvent::Intro doc.
199 - implement AnyEvent::Socket::format_ipv4/ipv6.
200 - slightly speed up portable_pipe/socketpair.
201 - expand condvar begin/end documentation.
202
2034.42 Fri Jun 26 08:32:18 CEST 2009
204 - preliminary and neccesarily incomplete support for IO::Async.
205 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
206 case it was set to IGNORE, to ensure we can catch child statuses
207 even when the calling env acts stupidly.
208 - updated benchmarks with IO::Async, which performs very well.
209
2104.412 Wed Jun 24 01:35:57 CEST 2009
211 - support an "untaint" attribute for AnyEvent::DNS
212 and set it on the default resolver.
213 - implement and document AnyEvent::Impl::Perl::loop.
214 - remove all anyevent-env variables from %ENV
215 when running in tainted mode.
216 - mention and extend the IO::Lambda benchmark.
217
2184.411 Sun Jun 7 18:48:02 CEST 2009
219 - do not try to use F_SETFD on windows.
220
2214.41 Thu May 14 06:40:11 CEST 2009
222 - work around issues in older perls (5.8.0?) when
223 a signal handler is deleted from the %SIG hash.
224 - use POSIX::_exit in child test, to avoid running
225 destructors.
226 - speed up CHLD handling by relying on SIGCHLD being
227 synchronously delivered, even when we roll our own
228 implementation.
229 - AnyEvent::DNS: add the "dname" resource record name
230 for cosmetic reasons.
231
2324.4 Sun Apr 26 20:12:33 CEST 2009
233 - implemented idle watchers, where applicable.
234 - AnyEvent->time died when Event backend was in use.
235 - fix a memleak in the Tk backend.
236 - sped up Tk timer handling.
237 - clip negative "after" values to 0 in AnyEvent::Impl::Event
238 to avoid spamming out warning messages.
239 - fix Qt timers without interval.
240 - avoid Qt zero-timeout special case, in old libqt's.
241
2424.352 Mon Apr 20 16:31:11 CEST 2009
243 - fix AnyEvent::Strict error messages for child watchers.
244 - fix/update Makefile.PL "recommends" versions.
245 - add AnyEvent->now_update.
246
2474.351 Sat Apr 11 07:56:14 CEST 2009
248 - actually make the signal pipe work under win32.
249 - localise $! in signal handler to avoid changing $!, although
250 perl itself does not seem to save/restore errno either.
251 - set the cloexec flag on the signal pipe (normally set by perl too).
252
2534.35 Fri Mar 27 11:48:20 CET 2009
254 - event models relying on AnyEvent's signal watcher emulation did
255 invoke the callback asynchronously, contradicting documentation
256 and causing signals to get lost (this includes AnyEvent's own
257 event loop). AnyEvent now uses the standard pipe trick to make
258 callback execution synchronous to the event loop.
259 - AnyEvent::Handle didn't free TLS context data on DESTROY
260 (patch by Pavel Shaydo).
261 - work around the ever-incompatibly-changing API of MakeMaker.
262 - document that changing global variables without restoring them
263 is a bad idea in Perl (noted by Adam Rosenstein).
264 - AnyEvent::Strict now barfs if ->io is passed a file.
265
2664.34 Thu Feb 12 18:32:45 CET 2009
267 - separately buffer TLS read data, as otherwise the read queue could
268 deadlock as receiving data is not expected while draining the
269 read queue (which cna only happen with TLS).
270 - raise EBADMSG error on JSON decoding errors.
271 - fix some minor manpage bugs (reported by Maximilian Gaß).
272 - speed up select bitmask parsing quite a bit in the pure perl
273 backend.
274 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
275
2764.331 Tue Jan 6 21:07:25 CET 2009
277 - socketpair fails on many vista machines because vista has
278 completely broken accept/getpeername and getsockname functions,
279 so we provide our own socketpair emulation that kind of works
280 (AnyEvent::Util::portable_pipe).
281 - new function: AnyEvent::Util::portable_socketpair.
282 - take advantage of the Guard module if it exists.
283
2844.33 Fri Nov 21 02:35:40 CET 2008
285 - AnyEvent::Strict did errornously flag a fileno of 0 as
286 illegal.
287 - reduce memory usage and slightly speed up the pure perl backend
288 by only storing the file descriptor, not the file handle.
289 - add missing autocork method to AnyEvent::Handle (reported
290 by Adam Rosenstein).
291 - AnyEvent::DNS->resolve errornously documented an $rcode result
292 argument, but there isn't (spotted by Henrik Krohns).
293 - the naked truth about Tk - it's basically unmaintained.
294
2954.32 Mon Nov 3 22:46:32 CET 2008
296 - fix AnyEvent::Socket::resolve_sockaddr to properly support
297 unix sockets again and choose a proper default for
298 the socket type.
299 - fix call to resolve_sockaddr in tcp_connect.
300
3014.31 Thu Oct 30 04:41:48 CET 2008
302 - implemented AnyEvent::Handle->destroy method.
303 - hint about unexpected effects in TLS mode.
304 - speed up AnyEvent::Impl::Perl by using more arrays
305 instead of hashes without the slightest loss of readability :)
306 - work around a perl argument refcounting bug.
307 - with some perl patches applied, AnyEvent now seems leak-free.
308
3094.3 Fri Oct 3 09:18:43 CEST 2008
310 - AnyEvent will now install a no-op signal handler for SIGPIPE,
311 unless one has been installed already.
312 - warn about not loaidng AnyEvent::Impl::POE early enough
313 (patch by Adam Rosenstein).
314 - fixed a great number of bugs and corner cases in AnyEvent::Handle:
315 - fix a bug in where in SSL connect mode, the
316 client would first wait for some data by the server and
317 otherwise hang. (reported and analyzed in an absolutely
318 exemplary manner by Adam Rosenstein).
319 - fix a bug in where SSL EOF would not be treated
320 as stream EOF, putting the connection into a hung state.
321 (reported and analyzed in an absolutely exemplary manner by Adam
322 Rosenstein).
323 - fix a potential segfault when the TLS context
324 would go missing in a read callback (e.g. due to stoptls) -
325 Net::SSLeay of course makes no type checking whatsoever.
326 - AnyEvent::Handle will not stop the read watcher in TLS mode.
327 - AnyEvent::Handle->stoptls will send a TLS close notify
328 instead of simply destroying the stream now.
329 - fix a bug where large blocks of data written in TLS mode
330 would not be sent unless triggered by receives.
331 - on_drain will now take the tls write buffer into account.
332 - SSL operations have been streamlined a bit and should be faster
333 now, more to come, though.
334 - added a FAQ section, will document nontrivial issues in there.
335
3364.234 Mon Sep 29 04:08:13 CEST 2008
337 - fix child watcher documentation: only child exits will be reported,
338 no trace events.
339 - mention SIGPIPE in AnyEvent::Handle.
340 - perl backend will now use POSIX::times as monotonic clock
341 when available and the monotonic clock option is not.
342 - run even without Time::HiRes in some cases now, but do not
343 provide sub-second accuracy (all *supported* perl versions have
344 Time::HiRes).
345
3464.233 Fri Aug 22 01:48:05 CEST 2008
347 - fix a bug in the testsuite, causing a hang.
348 - clarified lots of AnyEvent::Handle constructor arguments.
349
3504.232 Thu Aug 21 20:44:25 CEST 2008
351 - fix a bug in AnyEvent::Handle that could cause two fatal error
352 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
353 (testcase by Leon Brocard).
354 - clarified on_eof/on_error documentation a bit.
355
3564.231 Tue Jul 29 13:12:15 CEST 2008
357 - remove some debugging code left in AnyEvent::Util::fork_call
358 (and no, it's impossible to implement with the broken windows
359 perls without resource leaks or worse).
360
3614.23 Tue Jul 29 12:19:59 CEST 2008
362 - document the first parameter passed to condvar callbacks to be
363 the callback.
364 - add AnyEvent::Socket::{ntoa,aton} aliases.
365 - optimize the AE::Handle->push_read (line) for the default
366 eol marker.
367 - optimize push_read (packstring|storable) for small packets.
368 - invoke on_error callback when no on_eof callback is set.
369 - fix a bug in push_read (storable) of unknown impact.
370
3714.22 Sun Jul 20 16:34:13 CEST 2008
372 - new function AnyEvent::Socket::parse_hostport.
373 - as the bulkheads at microsoft can't even get getprotobyname reliably
374 working on their shitty fucking broken os we need to hardcode
375 some common protocol numbers in AnyEvent::Socket. How can
376 people even bother with such a pile of shit as windows.
377
3784.21 Thu Jul 17 14:40:05 CEST 2008
379 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
380 which is incompatible to previous invocation syntax.
381 - work around more windows perl bugs in fork_call. I cannot imagine
382 why anybody would bother implementing fork in such an obviously
383 unusable way.
384 - avoid calling $do_retry if the request has already finished in
385 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
386 K. Sharifulin).
387
3884.2 Sat Jul 12 22:42:11 CEST 2008
389 - support an "interval" argument for repeating timers.
390 - fix a bug in the pure perl event loop that caused timers to
391 be delayed under some circumstances.
392 - implement optional argument checking via PERL_ANYEVENT_STRICT.
393 - not importing anything from Event causes it to malfunction
394 and not use Time::HiRes, work around this bug.
395 - more bug workarounds against the endlessly broken Tk module,
396 seems to work now on 32 bit machines, but less so on 64 bit.
397 - minor optimisations applied to most backends.
398
3994.161 Fri Jul 4 14:24:48 CEST 2008
400 - fixed the t/handle/02_write.t, it read after write, with the
401 assumption that all data could be written before read
402 (reported by Jim Razmus).
403
4044.160 Thu Jul 3 04:02:21 CEST 2008
405 - re-registering signal watchers after unregistering one did
406 not work if a backend falls back on AnyEvent's default
407 implementation (also affected child watchers).
408 - new AnyEvent::Handle options: autocork and no_delay.
409
4104.152 Sun Jun 22 14:15:44 CEST 2008
411 - allow for 32-bit perls that implement shifts differently
412 on different architectures in parse_ipv4 (reported and
413 analysed by Keiichi DAIBA).
414
4154.151 Fri Jun 6 17:34:24 CEST 2008
416 - make sure specifying _only_ on_read and never pushing reads
417 works.
418
4194.15 Fri Jun 6 13:00:46 CEST 2008
420 - the pure perl backend would keep some watchers alive when more than
421 one watcher was registered for the same fd.
422 - new "packstring" and "storable" read and write types
423 for AnyEvent::Handle.
424 - allow on_eof handler to be called after on_error with EPIPE returns.
425 - do not immediately call on_read callback in handle constructor.
426
4274.14 Thu Jun 5 20:29:31 CEST 2008
428 - Fixed a bug in DNS SRV priority sorting.
429 - AnyEvent::Util::guard now reports runtime errors while
430 executing the guard block as warnings.
431 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
432
4334.13 Thu Jun 5 00:47:59 CEST 2008
434 - AnyEvent::DNS only followed cname chains with length 2,
435 contrary to documentation. bumped it up to 10, thanks to
436 microsoft, the current broken-dns-config-king.
437 - AnyEvent::DNS didn't check for socket return status
438 but instead relied on perl not creating filehandles in that
439 case - too bad it gives you a bogus file handle (reported
440 and analysed by Vladimir Timofeev).
441 - fix queue management logic in AnyEvent::Handle:
442 when on_read was registered, the queue was empty
443 and no progress could be made AnyEvent::Handle would
444 enter an endless loop.
445 - correctly start reading again when the handle became
446 busy again after idling.
447 - correctly treat tls shut-downs as EOF condition
448 (for the time being, would be nice to have a callback for that).
449 - correctly call eof callback under all conditions
450 (hopefully).
451 - the timeout callback did not expect that $self can go
452 away any time.
453
4544.12 Tue Jun 3 10:58:04 CEST 2008
455 - include AnyEvent::Intro, a tutorial for anyevent,
456 anyevent::socket and anyevent::handle.
457 - allow more options in on_error.
458
4594.11 Fri May 30 23:42:25 CEST 2008
460 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
461 reverse_lookup and reverse_verify to replace it, support
462 v4mapped and v4compat addresses.
463 - provide more documentation for the resolver class.
464 - really replace longest run of :0: by :: in format_address,
465 also properly convert :: and ::1 again.
466 - support NAPTR record name and decode it.
467 - implement random weight sampling for SRV records, as per
468 rfc 2782.
469 - correctly abort on srv-record targets of ".".
470 - added AnyEvent::DNS::wait_for_slot.
471 - in the unlikely event of a virtual circuit connection
472 being invalidated by a delayed udp reply, AnyEvent::DNS
473 could die. this has been fixed.
474 - plug a probable memleak in the DNS vc code.
475 - use configured timeout also for VC request phase.
476 - implement timeout and max_outstanding methods
477 for resolver class.
478 - update version numbers in all modules.
479
4804.1 Thu May 29 05:45:40 CEST 2008
481 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
482 which mirrors their purpose better (old names still available
483 for a while). Also moved AnyEvent::DNS::addr to
484 AnyEvent::Socket::resolve_sockaddr.
485 - implement AnyEvent->time and AnyEvent->now.
486 - fix IPv6 support in pack_sockaddr.
487 - officially un-experimentalise all newly introduced code.
488 - support unix domain sockets everywhere by specifying
489 a host of "unix/" and the pathname as service.
490 - implement an activity timeout in AnyEvent::Handle.
491 - added a regex read type to AnyEvent::Handle.
492 - added a json read/write type to AnyEvent::Handle.
493 - always croak in on_error in AnyEvent::Handle.
494 - document how to subclass AnyEvent::Handle.
495 - implement AnyEvent::Util::fork_call.
496 - add support for IPv6 nameservers and nameserver statements.
497 - work around _yet_ _another_ windows perl bug in where empty
498 select masks cause errors in the pure perl backend (this
499 workaround cannot be implemented for the other event loops,
500 but EV also works around this).
501 - supply AnyEvent::Socket::address_type and make good use of it.
502 - clarify and fix the Handle documentation w.r.t $self vs. $handle
503 vs. callback arguments.
504 - add some recommends to the META.yml.
505
5064.05 Mon May 26 19:44:06 CEST 2008
507 - some platforms ignore the rfc and prepend an extra sa_len member to
508 the sockaddr structure, cater for those. also use sockaddr_family,
509 didn't know it was there, but comes in handy for the workaround.
510 - undo the import WIN32 hack.
511
5124.04 Mon May 26 08:03:31 CEST 2008
513 - try to work around yet another windows bug: failed connects are reported
514 as if it were out-of-band data. windows users: you suck. Only
515 EV, Glib, Event and the pure perl backend can handle this condition.
516 - optimize write algorithm in AnyEvent::Handle.
517 - properly parse PERL_ANYEVENT_PROTOCOLS and use the info correctly.
518 - tcp_connect now properly iterates through all targets.
519 - check wether IPv6 sockets can actually be created, otherwise, disable ipv6
520 support entirely. also hardcode some AF_INET6 constants for the
521 sake of perl 5.8.
522 - vastly improved nameserver/suffix detection algorithm for windows
523 (still a hack...)
524 - try to use a fallback dns server if no dns servers could be found.
525 - splatter around some local $SIG{__DIE__}'s to work around bad
526 die handlers.
527
5284.03 Sun May 25 05:05:57 CEST 2008
529 - work around perl bugs on windows, where perl returns undocumented
530 error codes for sysread, syswrite etc.
531 - fix AnyEvent::Handle::unshift_read to really unshift and not push.
532 - AnyEvent::Handle could sometimes stumble over 0-byte writes.
533 - fix unpack_sockaddr, now works for AF_INET6 sockaddr's.
534 - updated documentation.
535 - lots of workarounds for perl 5.8 quirks.
536 - implement a more flexible type system for both reads and writes
537 in AnyEvent::Handle.
538 - support IPv6 in tcp_server.
539 - pass local hostname and port to the prepare callback in tcp_server.
540 - make types extendable in AnyEvent::Handle.
541 - croak, not die, in AnyEvent::Handle;
542
5434.0 Sat May 24 19:58:08 CEST 2008
4 - added (experimental) AnyEvent::DNS module. 544 - added (experimental) AnyEvent::DNS module.
5 - added (experimental) AnyEvent::Socket module. 545 - added (experimental) AnyEvent::Socket module.
6 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle. 546 - added (experiemntal) support for TLS/SSL in AnyEvent::Handle.
7 - make condvar->end call ->send as documented. 547 - make condvar->end call ->send as documented.
8 - make condvar constructor support the documented cb => argument. 548 - make condvar constructor support the documented cb => argument.
… …
12 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. 552 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
13 - try very hard not to freeze on broken windows perls. 553 - try very hard not to freeze on broken windows perls.
14 - make on_eof optional in AnyEvent::Handle. 554 - make on_eof optional in AnyEvent::Handle.
15 - support service names in tcp_connect and tcp_server. 555 - support service names in tcp_connect and tcp_server.
16 - work around netbsd/openbsd bug causing unix sockets not to 556 - work around netbsd/openbsd bug causing unix sockets not to
17 become writable untill empty. 557 become writable until empty.
18 - only make a dummy read for error detection when getpeername 558 - only make a dummy read for error detection when getpeername
19 returns ENOTCONN. 559 returns ENOTCONN.
20 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 560 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
21 override. 561 override.
22 - new AnyEvent::Guard::cancel. 562 - new AnyEvent::Guard::cancel.
23 563
243.5 Sat May 17 23:17:14 CEST 2008 5643.5 Sat May 17 23:17:14 CEST 2008
25 - gracefully deal with systems not defining CLOCK_MONOTONIC 565 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
39 579
403.41 Sun May 11 19:53:13 CEST 2008 5803.41 Sun May 11 19:53:13 CEST 2008
41 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 581 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
42 582
433.4 Sun May 11 00:29:25 CEST 2008 5833.4 Sun May 11 00:29:25 CEST 2008
44 - complete re-implementation of condvars with a hopefully much mroe useful 584 - complete re-implementation of condvars with a hopefully much more useful
45 API (backward compatible functions exist but are undocumented). 585 API (backward compatible functions exist but are undocumented).
46 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 586 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
47 (still subject to change). 587 (still subject to change).
48 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 588 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
49 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 589 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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