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

Comparing AnyEvent/Changes (file contents):
Revision 1.167 by root, Tue Jul 8 23:44:51 2008 UTC vs.
Revision 1.440 by root, Sun Jan 23 11:17:02 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.2 3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent strict and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict checking for $_ changes?
7
85.31 Sun Jan 23 11:49:19 CET 2011
9 - work around a POE bug where POE would first poll for new events and
10 invoke handlers BEFORE handling the message queue.
11 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
12 - stop_read now does just that in TLS mode. see its
13 documentation for caveats.
14 - update to common::sense 3.4 warning set (some warnings in perl impose
15 an arbitrary 32kb limit on regex matches, which affects
16 AnyEvent::Handle).
17 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
18 - new experimental AE::Handle::resettls method.
19
205.3 Fri Dec 31 05:48:13 CET 2010
21 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
22 issues with IO::Async have been solved in that version, including
23 the availabilty of a default loop!
24 - implement new wbuf_max parameter for AnyEvent::Handle.
25 - added Cocoa::EventLoop backend by Daisuke Murase.
26 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
27 - clarify handle on_prepare documentation.
28 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
29 to avoid leaking them to any child processes.
30 - try to force files into the arch-dependend lib directory, for the
31 half-"turly obsessive compulsive person"(s) that depend on it.
32
335.29 Sun Dec 5 10:49:21 CET 2010
34 - convert EV backend to EV 4.00 API (so better upgrade EV too).
35 - AE::Handle: implement dynamic read_size adjustment, add
36 max_read_size parameter, reduce default read_size to 2048.
37 - add some FAQ entries for servers and/or on_eof vs. on_error.
38 - work around OS bugs (cygwin again) when lingering in AE::Handle.
39 - reduce memory usage slightly when connect was used in AE::Handle.
40 - wrok around more segfaults in Tk.
41 - document tls_ctx => undef as valid.
42 - detect WNOHANG value at compiletime.
43 - include some probably linux-only support for building AnyEvent
44 as part of the perl core.
45 - improve compatibility of t/08_idna.t to perl 5.8.
46 - make AnyEvent::Strict truly optional (in case somebody builds a
47 minimal perl...).
48
495.28 Wed Oct 13 04:14:23 CEST 2010
50 - due to a glitch, AnyEvent's internal getprotobyname wasn't
51 actually used in tcp_server.
52 - implement AnyEvent::CondVar->new method, which was documented but
53 missing.
54 - support autoloading of read types in unshift_read.
55 - AnyEvent::Strict: do the test for invalid fh arguments differently,
56 resulting in more informative error messages.
57 - new AnyEvent::FAQ manpage.
58 - clarify that you can actually call push_read etc. in on_read.
59 - put keys and certs in /tmp, not into the current directory,
60 and document this in the SECURITY CONSIDERATIONS sections
61 of AnyEvent::TLS.
62 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
63 requested by Ron Isaacson.
64 - document $AnyEvent::DNS::RESOLVER variable.
65 - incorporated major typo patches by Abhijit Menon-Sen.
66 - AnyEvent::Handle now croaks when negative timeouts are passed.
67 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
68 undocumented, but who knows when they might come in handy.
69
705.271 Tue Jun 8 12:05:46 CEST 2010
71 - backport to perl 5.8.x.
72
735.27 Sun Jun 6 12:12:05 CEST 2010
74 - postpone differently in AnyEvent::Socket now, as
75 when not, canceling the connection attempt might fail
76 (found by Felix Antonius Wilhelm Ostmann).
77 - explicitly check for non-stream sockets in AE::Handle, too many
78 clueless people fell into the trap of this somehow working.
79 - simplified and reworked the "OTHER MODULES" section.
80 - better/more condvar examples.
81
825.261 Wed Apr 28 16:13:36 CEST 2010
83 - AF_INET6 was not properly used from Socket6 during configuration
84 time (found by Andy Grundman).
85
865.26 Mon Apr 12 04:49:35 CEST 2010
87 - don't generate spurious readyness notifications when select
88 returns EINTR in the pure perl backend (analysed and testcase
89 by Bas Denissen).
90 - give same examples in SYNOPSIS sections of AnyEvent and AE.
91 - provider faster implementations for AE::time/now/now_update for
92 EV and Perl backends.
93 - provide faster AE::cv and AE::time implementations for all
94 backends.
95 - no longer support lower/mixed-case signal names (this was
96 never documented nor universally supported).
97 - some more memory µ-optimisations, and somewhat less messy
98 code to cope with both AnyEvent and AE APIs.
99
1005.251 Sat Mar 13 00:58:21 CET 2010
101 - make test failed if it was run before make install on perl <=
102 5.8.8, and older versions of perl make our live very difficult,
103 so write constants.pl during Makefile.PL time.
104
1055.25 Sat Mar 13 00:23:14 CET 2010
106 - fix a race condition in AnyEvent::Handle that would cause
107 a "bio_read: unsupported method" error (in _another_
108 ssl connection) after user code threw an exception.
109 - added AnyEvent::Handle->destroyed method.
110 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
111 - the AnyEvent::Util::fh_nonblocking function worked only
112 by lucky accident on win32.
113 - smaller and faster AnyEvent::Util::fh_nonblocking.
114 - when the (required!) Time::HiRes module is missing, AnyEvent
115 did not fall back to built-in time properly.
116 - do not load Fcntl at runtime, saving memory and loading time.
117 - precompile a number of constants and use them instead of runtime
118 detection and eval.
119 - free detection code after detection and similar memory optimisations.
120 - Perl backend timer interval best effort drift has been improved
121 (same algorithm as EV).
122 - update unicode idna mapping table.
123
1245.24 Tue Jan 5 11:39:43 CET 2010
125 - cygwin never reports errors from failed connects. 1.5 just gives
126 you continous readyness and EAGAIN, 1.7 is even more broken and
127 just hangs. work around both issues in a cygwin-specific hack.
128 - improve idle watcher documentation slightly.
129
1305.23 Sun Dec 20 23:48:00 CET 2009
131 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
132 - implement punycode_encode/decode, idn_nameprep,
133 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
134 - provide $AE::VERSION.
135 - removed traces of "no strict 'refs'".
136
1375.22 Sat Dec 5 03:51:13 CET 2009
138 - downgrade-or-fail in AnyEvent::Handle::push_write, to
139 diagnose encoding failures earlier and more succinctly.
140 (this works around bugs in perl, throwing away encoding info
141 when passing scalar data to extensions).
142 - add more examples to AnyEvent::Socket manpage.
143 - upgrade internal warning set to the same as common::sense 2.03.
144 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
145 (John Beppu).
146 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
147 - add google public dns servers to fallback server set and make sure
148 we load-balance properly between the three sets. also add all
149 fallback dns servers, not just a random one, to each dns config.
150
1515.21 Thu Nov 19 02:48:47 CET 2009
152 - fix a problem where socket constants were called with parameters
153 (spotted by David Friedland).
154 - fork_call never use'd POSIX (reported by Daisuke Maki).
155 - improve perl 5.6 compatibility further (but it still won't work
156 unless you rip out everything but the core).
157 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
158 uses 10MB of RAM, but doesn't open a console window. *sigh*.
159
1605.202 Wed Oct 14 22:35:44 CEST 2009
161 - AnyEvent::DNS would unexpectedly clobber $_ under windows
162 (analysed by Matthias Waldorf).
163 - AnyEvent::Handle::run_cmd can now pass the PID of the
164 newly-created process, which is much less useful than it might
165 sound (based on patch by Yann Kerherve).
166
1675.201 Tue Sep 29 12:09:25 CEST 2009
168 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
169 (reported by Torsten Foertsch).
170 - common::sense 2.0 could cause tcp_server to throw an exception
171 (analysed by elmex).
172
1735.2 Mon Sep 14 07:04:49 CEST 2009
174 - INCOMPATIBLE CHANGE: do no longer support register_read_type
175 and register_write_type in AnyEvent::Handle, instead support
176 package names (the facility was mostly abused).
177 - implement "packagename-as-read/write type" support in
178 AnyEvent::Handle.
179 - AnyEvent::Handle: new options "keepalive" and "oobinline".
180 - oobinline set by default to avoid security issues.
181 - the pure-perl event loop backend wrongly detected times() fallback
182 support (spotted by Pavel Boldin).
183
1845.12 Tue Sep 1 20:26:50 CEST 2009
185 - be more lenient when parsing resolv.conf files, as some people
186 use hashmarks as comment indicator inside directives (reported
187 by Michael S. Fischer).
188 - use same set of warnings as common::sense 2.0.
189 - fix a potential 32 bit overflow issue due to perl having problems
190 with large hex constants in the Perl backend.
191 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
192 by default.
193 - fix some prototypes.
194 - (optionally) require Async::Interrupt 1.02.
195 - replace opendns fallbacks by verizon/level3 public dns servers.
196
1975.112 Fri Aug 21 13:59:04 CEST 2009
198 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
199 using starttls with an illegal/uncreatable tls context.
200 - port to broken win32 perls.
201
2025.111 Sun Aug 16 18:44:36 CEST 2009
203 - perl errornously requires sockaddr_un structures returned by the
204 kernel to have a specific length. work around this bug by padding
205 sockaddr_un structures (found to fail on OS X, as perl only uses
206 the correct code for GNU/Linux, but likely to fail on other
207 OSes as well).
208
2095.11 Wed Aug 12 17:49:37 CEST 2009
210 - fixed a bug in run_cmd with input from filename.
211 - create a json coder object when none is given in write direction
212 to reduce dependency on JSON.
213
2145.1 Tue Aug 11 03:17:32 CEST 2009
215 - tcp_connect now reports the correct error in $!, not an unrelated
216 one (was broken since 4.91).
217 - AE::Handle did not properly clear rtimeout/wtimeout watchers
218 when retrying a tcp_connect.
219 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
220 - fixed wrong documentation in AE::Socket.
221
2225.01 Mon Aug 10 03:16:32 CEST 2009
223 - last release broke EV child watchers :/.
224
2255.0 Sun Aug 9 17:32:09 CEST 2009
226 - officially introduce and document the AE API.
227 - lowering the timeout at runtime did not have immediate effect
228 in AnyEvent::Handle.
229 - AE::Handle now has separate and independent read and write timeout
230 settings.
231 - probe for Guard module when AnyEvent::Util is loaded, not at
232 runtime (Event doesn't like eval "use" from callbacks, and it avoids
233 exporting the probe function).
234 - Impl::IOAsync's io watchers did not use the filehandle emulation
235 code to work around it's watcher limits.
236 - work around IO::Async's problems with overloaded objects.
237 - try to disable wearnings inside IO::Async, it's just too buggy.
238 - do no longer use Event by default, only EV or Perl, Event is too buggy.
239 - hackishly provide fast AE:: interface when EV or Perl is the backend.
240 - inherit VERSION from AnyEvent::VERSION in most modules.
241
2424.91 Thu Aug 6 15:42:45 CEST 2009
243 - AE::Handle::starttls could get out of sync when the read buffer
244 already contains some TLS handshake.
245 - AE::Handle did not properly free the TLS session, causing bigger
246 memory leaks in Net::SSLeay than expected (reported by toaster).
247 - AE::Socket::tcp_connect will now artificially delay invoking
248 the callback to avoid returning after invoking the callback.
249 - convert many internal watcher uses to AE API.
250
2514.9 Sat Aug 1 11:07:01 CEST 2009
252 - Glib silently fails when registering a timeout with
253 a negative value, so avoid that.
254 - call condvar callback immediately when it is set
255 after the condition is already signalled.
256 - check rbuf_max condition only after trying to consume
257 data.
258 - normalise signal numbers to names when using the ae's
259 signal handling, but do not document this (yet).
260 - pure perl signal emulation did not properly set nonblocking
261 mode on the signal pipe on win32 (this is unlikely to have caused
262 any issues).
263 - new module: AnyEvent::Debug.
264 - AnyEvent::Strict now checks that a signal was specified
265 by name (not name or number).
266 - reduce memory footprint in typical cases by ~50kb by
267 only compiling signal code when necessary.
268 - add AnyEvent::Handle->rbuf_max.
269 - grab the AE namespace for future new API, implement stubs for
270 said future/alternative API.
271 - new function: AnyEvent::Socket::format_hostport.
272
2734.881 Tue Jul 28 12:51:53 CEST 2009
274 - work around a bug in local in pre-5.10 perls, causing
275 AnyEvent::Handle to recurse when it should not
276 (analysed by elmex).
277
2784.88 Tue Jul 28 04:04:37 CEST 2009
279 - re-bless the handle into a dummy package after calling
280 AnyEvent::Handle::destroy, so the user does not need to check
281 for errors after every push_write etc.
282 - do not attempt to run t/02_signals.t on obviously broken
283 platforms.
284
2854.87 Sun Jul 26 02:06:16 CEST 2009
286 - do not attempt to linger when there is no longer a valid fh.
287 - enforce tls mode to be either accept or connect, do not simply
288 segfault in Net::SSLeay.
289 - AnyEvent::Handle can now call tcp_connect itself (new parameters
290 connect, on_prepare, on_connect and on_connect_error). Updated
291 tutorial accordingly.
292 - add AnyEvent::Impl::Irssi backend.
293
2944.86 Mon Jul 20 23:52:29 CEST 2009
295 - since the verbose warning is not enough, explicitly document
296 that versions before 1.33 of Net::SSLeay are not secure.
297 - work around signal handling races in Event and (...) Event::Lib.
298 - try to align signal-race timer to full-second boundaries.
299 - work around Tk not liking negative timeouts.
300 - don't complain of different grades of environmental unfriendlyness
301 in IO::Async.
302
3034.85 Sat Jul 18 06:16:14 CEST 2009
304 - nail the signal race problem in perl once and for all
305 (see $AnyEvent::MAX_SIGNAL_LATENCY).
306 - take advantage of Async::Interrupt if it is available.
307 - load Time::HiRes and Guard modules on demand only.
308 - add optional/recommended modules section to AnyEvent
309 documentation.
310 - reduce memory usage considerably (and reduce startup penalty)
311 by not using "strict", "warnings" and "overload" modules.
312 - work around buggy windows/openbsd perls and provide EBADMSG
313 and EPROTO ourselves when missing.
314 - improve perl 5.6 compatibility of the core event loop.
315 - made Net::SSLeay version 1.33 a soft requirement.
316
3174.83 Fri Jul 17 16:56:26 CEST 2009
318 - implement AnyEvent::Socket::getprotobyname.
319 - AnyEvent::CondVar's will now detect recursive blocking
320 waits and will croak, as too many people fall into
321 this trap.
322 - AnyEvent::Handle will now call ->destroy on itself after
323 executing the on_error callback, instead of doing some
324 half-baked internal shutdown, for fatal errors.
325 - clarify on_eof behaviour w.r.t. the read queue and
326 on_read callbacks.
327 - ignore some possible spurious wake-ups in tcp_connect.
328
3294.82 Sat Jul 11 00:34:55 CEST 2009
330 - POE and Event backends didn't accept some callable objects as
331 callbacks.
332 - use Config module instead of POSIX module to detect signal names
333 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
334 module doesn't even have all POSIX signals :/.
335 - use more workarounds around the many refcnt/corruption bugs in
336 Event::Lib.
337 - work around a race condition in perl's select, causing t/03_child.t
338 to rarely fail.
339
3404.81 Thu Jul 9 10:30:30 CEST 2009
341 - AnyEvent::Handle didn't properly diagnose write errors
342 (it expected -1 from syswrite, how lame... :).
343 - support file descriptors in addition to file handles
344 in AnyEvent->io.
345 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
346 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
347 and PERL_ANYEVENT_CA_PATH.
348 - provide a sensible synopsis section for AnyEvent::TLS.
349 - add a "supported backends" section to the manpage.
350 - added simple io watcher test to testsuite, using a
351 portable_socketpair.
352 - tried to improve the stability of the Event::Lib backend,
353 YMMV.
354
3554.8 Mon Jul 6 23:45:16 CEST 2009
356 - AnyEvent::DNS did not properly follow CNAME records with
357 uppercase targets.
358 - AnyEvent::DNS would errornously return AAAA records
359 with v4 mapped addresses (a faulty record) as ipv4 addresses,
360 causing AnyEvent::Socket to throw an exception.
361 - added new module AnyEvent::TLS for easier SSL/TLS context
362 creation, with many options including hostname verification,
363 secure default configuration, lots of documentation and,
364 predefined diffie-hellman keys for perfect forward security
365 and much more. get it while it's still fresh!
366 - use AnyEvent::TLS in AnyEvent::Handle for context management.
367 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
368 so AnyEvent::Socket users have smaller memory footprint
369 in the common case.
370 - add AnyEvent::Handle->push_shutdown method.
371 - add an additional error message parameter to AnyEvent::Handle's
372 on_error callback (for TLS, $! is still available).
373 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
374 - make AnyEvent::Handle more robust against early conenction
375 failures (during new), and return C<undef> in that case
376 from the constructor.
377 - AnyEvent::Handle will now try to load only JSON::XS first,
378 then fall back to JSON.
379 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
380 and because it was documented that way.
381
3824.451 Fri Jul 3 00:28:58 CEST 2009
383 - do not clear rbuf when shutting down an AnyEvent::Handle
384 object - doing so breaks AnyEvent::HTTP.
385
3864.45 Mon Jun 29 22:59:26 CEST 2009
387 - a write error could cause AnyEvent::Handle to create
388 an I/O watcher with an undefined $fh.
389 - special-case mapped ipv4 addresses in both
390 AnyEvent::Socket::format_address and parse_address,
391 to treat them just like ipv4 addresses.
392 - updated and overhauled the AnyEvent::Intro doc.
393 - implement AnyEvent::Socket::format_ipv4/ipv6.
394 - slightly speed up portable_pipe/socketpair.
395 - expand condvar begin/end documentation.
396
3974.42 Fri Jun 26 08:32:18 CEST 2009
398 - preliminary and neccesarily incomplete support for IO::Async.
399 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
400 case it was set to IGNORE, to ensure we can catch child statuses
401 even when the calling env acts stupidly.
402 - updated benchmarks with IO::Async, which performs very well.
403
4044.412 Wed Jun 24 01:35:57 CEST 2009
405 - support an "untaint" attribute for AnyEvent::DNS
406 and set it on the default resolver.
407 - implement and document AnyEvent::Impl::Perl::loop.
408 - remove all anyevent-env variables from %ENV
409 when running in tainted mode.
410 - mention and extend the IO::Lambda benchmark.
411
4124.411 Sun Jun 7 18:48:02 CEST 2009
413 - do not try to use F_SETFD on windows.
414
4154.41 Thu May 14 06:40:11 CEST 2009
416 - work around issues in older perls (5.8.0?) when
417 a signal handler is deleted from the %SIG hash.
418 - use POSIX::_exit in child test, to avoid running
419 destructors.
420 - speed up CHLD handling by relying on SIGCHLD being
421 synchronously delivered, even when we roll our own
422 implementation.
423 - AnyEvent::DNS: add the "dname" resource record name
424 for cosmetic reasons.
425
4264.4 Sun Apr 26 20:12:33 CEST 2009
427 - implemented idle watchers, where applicable.
428 - AnyEvent->time died when Event backend was in use.
429 - fix a memleak in the Tk backend.
430 - sped up Tk timer handling.
431 - clip negative "after" values to 0 in AnyEvent::Impl::Event
432 to avoid spamming out warning messages.
433 - fix Qt timers without interval.
434 - avoid Qt zero-timeout special case, in old libqt's.
435
4364.352 Mon Apr 20 16:31:11 CEST 2009
437 - fix AnyEvent::Strict error messages for child watchers.
438 - fix/update Makefile.PL "recommends" versions.
439 - add AnyEvent->now_update.
440
4414.351 Sat Apr 11 07:56:14 CEST 2009
442 - actually make the signal pipe work under win32.
443 - localise $! in signal handler to avoid changing $!, although
444 perl itself does not seem to save/restore errno either.
445 - set the cloexec flag on the signal pipe (normally set by perl too).
446
4474.35 Fri Mar 27 11:48:20 CET 2009
448 - event models relying on AnyEvent's signal watcher emulation did
449 invoke the callback asynchronously, contradicting documentation
450 and causing signals to get lost (this includes AnyEvent's own
451 event loop). AnyEvent now uses the standard pipe trick to make
452 callback execution synchronous to the event loop.
453 - AnyEvent::Handle didn't free TLS context data on DESTROY
454 (patch by Pavel Shaydo).
455 - work around the ever-incompatibly-changing API of MakeMaker.
456 - document that changing global variables without restoring them
457 is a bad idea in Perl (noted by Adam Rosenstein).
458 - AnyEvent::Strict now barfs if ->io is passed a file.
459
4604.34 Thu Feb 12 18:32:45 CET 2009
461 - separately buffer TLS read data, as otherwise the read queue could
462 deadlock as receiving data is not expected while draining the
463 read queue (which cna only happen with TLS).
464 - raise EBADMSG error on JSON decoding errors.
465 - fix some minor manpage bugs (reported by Maximilian Gaß).
466 - speed up select bitmask parsing quite a bit in the pure perl
467 backend.
468 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
469
4704.331 Tue Jan 6 21:07:25 CET 2009
471 - socketpair fails on many vista machines because vista has
472 completely broken accept/getpeername and getsockname functions,
473 so we provide our own socketpair emulation that kind of works
474 (AnyEvent::Util::portable_pipe).
475 - new function: AnyEvent::Util::portable_socketpair.
476 - take advantage of the Guard module if it exists.
477
4784.33 Fri Nov 21 02:35:40 CET 2008
479 - AnyEvent::Strict did errornously flag a fileno of 0 as
480 illegal.
481 - reduce memory usage and slightly speed up the pure perl backend
482 by only storing the file descriptor, not the file handle.
483 - add missing autocork method to AnyEvent::Handle (reported
484 by Adam Rosenstein).
485 - AnyEvent::DNS->resolve errornously documented an $rcode result
486 argument, but there isn't (spotted by Henrik Krohns).
487 - the naked truth about Tk - it's basically unmaintained.
488
4894.32 Mon Nov 3 22:46:32 CET 2008
490 - fix AnyEvent::Socket::resolve_sockaddr to properly support
491 unix sockets again and choose a proper default for
492 the socket type.
493 - fix call to resolve_sockaddr in tcp_connect.
494
4954.31 Thu Oct 30 04:41:48 CET 2008
496 - implemented AnyEvent::Handle->destroy method.
497 - hint about unexpected effects in TLS mode.
498 - speed up AnyEvent::Impl::Perl by using more arrays
499 instead of hashes without the slightest loss of readability :)
500 - work around a perl argument refcounting bug.
501 - with some perl patches applied, AnyEvent now seems leak-free.
502
5034.3 Fri Oct 3 09:18:43 CEST 2008
504 - AnyEvent will now install a no-op signal handler for SIGPIPE,
505 unless one has been installed already.
506 - warn about not loaidng AnyEvent::Impl::POE early enough
507 (patch by Adam Rosenstein).
508 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
509 - fix a bug in where in SSL connect mode, the
510 client would first wait for some data by the server and
511 otherwise hang. (reported and analysed in an absolutely
512 exemplary manner by Adam Rosenstein).
513 - fix a bug in where SSL EOF would not be treated
514 as stream EOF, putting the connection into a hung state.
515 (reported and analysed in an absolutely exemplary manner by Adam
516 Rosenstein).
517 - fix a potential segfault when the TLS context
518 would go missing in a read callback (e.g. due to stoptls) -
519 Net::SSLeay of course makes no type checking whatsoever.
520 - AnyEvent::Handle will not stop the read watcher in TLS mode.
521 - AnyEvent::Handle->stoptls will send a TLS close notify
522 instead of simply destroying the stream now.
523 - fix a bug where large blocks of data written in TLS mode
524 would not be sent unless triggered by receives.
525 - on_drain will now take the tls write buffer into account.
526 - SSL operations have been streamlined a bit and should be faster
527 now, more to come, though.
528 - added a FAQ section, will document nontrivial issues in there.
529
5304.234 Mon Sep 29 04:08:13 CEST 2008
531 - fix child watcher documentation: only child exits will be reported,
532 no trace events.
533 - mention SIGPIPE in AnyEvent::Handle.
534 - perl backend will now use POSIX::times as monotonic clock
535 when available and the monotonic clock option is not.
536 - run even without Time::HiRes in some cases now, but do not
537 provide sub-second accuracy (all *supported* perl versions have
538 Time::HiRes).
539
5404.233 Fri Aug 22 01:48:05 CEST 2008
541 - fix a bug in the testsuite, causing a hang.
542 - clarified lots of AnyEvent::Handle constructor arguments.
543
5444.232 Thu Aug 21 20:44:25 CEST 2008
545 - fix a bug in AnyEvent::Handle that could cause two fatal error
546 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
547 (testcase by Leon Brocard).
548 - clarified on_eof/on_error documentation a bit.
549
5504.231 Tue Jul 29 13:12:15 CEST 2008
551 - remove some debugging code left in AnyEvent::Util::fork_call
552 (and no, it's impossible to implement with the broken windows
553 perls without resource leaks or worse).
554
5554.23 Tue Jul 29 12:19:59 CEST 2008
556 - document the first parameter passed to condvar callbacks to be
557 the callback.
558 - add AnyEvent::Socket::{ntoa,aton} aliases.
559 - optimize the AE::Handle->push_read (line) for the default
560 eol marker.
561 - optimize push_read (packstring|storable) for small packets.
562 - invoke on_error callback when no on_eof callback is set.
563 - fix a bug in push_read (storable) of unknown impact.
564
5654.22 Sun Jul 20 16:34:13 CEST 2008
566 - new function AnyEvent::Socket::parse_hostport.
567 - as the bulkheads at microsoft can't even get getprotobyname reliably
568 working on their shitty fucking broken os we need to hardcode
569 some common protocol numbers in AnyEvent::Socket. How can
570 people even bother with such a pile of shit as windows.
571
5724.21 Thu Jul 17 14:40:05 CEST 2008
573 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
574 which is incompatible to previous invocation syntax.
575 - work around more windows perl bugs in fork_call. I cannot imagine
576 why anybody would bother implementing fork in such an obviously
577 unusable way.
578 - avoid calling $do_retry if the request has already finished in
579 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
580 K. Sharifulin).
581
5824.2 Sat Jul 12 22:42:11 CEST 2008
4 - support an "interval" argument for repeating timers. 583 - support an "interval" argument for repeating timers.
5 - not importing anything from Event causes it to malfunction
6 and not use Time::HiRes, work around this bug.
7 - fix a bug in the pure perl event loop that caused timers to 584 - fix a bug in the pure perl event loop that caused timers to
8 be delayed under some circumstances. 585 be delayed under some circumstances.
9 - implement optional argument checking via PERL_ANYEVENT_STRICT. 586 - implement optional argument checking via PERL_ANYEVENT_STRICT.
587 - not importing anything from Event causes it to malfunction
588 and not use Time::HiRes, work around this bug.
589 - more bug workarounds against the endlessly broken Tk module,
590 seems to work now on 32 bit machines, but less so on 64 bit.
10 - minor optimisations applied to most backends. 591 - minor optimisations applied to most backends.
11 592
124.161 Fri Jul 4 14:24:48 CEST 2008 5934.161 Fri Jul 4 14:24:48 CEST 2008
13 - fixed the t/handle/02_write.t, it read after write, with the 594 - fixed the t/handle/02_write.t, it read after write, with the
14 assumption that all data could be written before read 595 assumption that all data could be written before read
… …
168 - support service names in tcp_connect and tcp_server. 749 - support service names in tcp_connect and tcp_server.
169 - work around netbsd/openbsd bug causing unix sockets not to 750 - work around netbsd/openbsd bug causing unix sockets not to
170 become writable until empty. 751 become writable until empty.
171 - only make a dummy read for error detection when getpeername 752 - only make a dummy read for error detection when getpeername
172 returns ENOTCONN. 753 returns ENOTCONN.
173 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 754 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
174 override. 755 override.
175 - new AnyEvent::Guard::cancel. 756 - new AnyEvent::Guard::cancel.
176 757
1773.5 Sat May 17 23:17:14 CEST 2008 7583.5 Sat May 17 23:17:14 CEST 2008
178 - gracefully deal with systems not defining CLOCK_MONOTONIC 759 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
192 773
1933.41 Sun May 11 19:53:13 CEST 2008 7743.41 Sun May 11 19:53:13 CEST 2008
194 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 775 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
195 776
1963.4 Sun May 11 00:29:25 CEST 2008 7773.4 Sun May 11 00:29:25 CEST 2008
197 - complete re-implementation of condvars with a hopefully much mroe useful 778 - complete re-implementation of condvars with a hopefully much more useful
198 API (backward compatible functions exist but are undocumented). 779 API (backward compatible functions exist but are undocumented).
199 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 780 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
200 (still subject to change). 781 (still subject to change).
201 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 782 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
202 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 783 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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