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

Comparing AnyEvent/Changes (file contents):
Revision 1.150 by root, Thu Jun 5 16:28:48 2008 UTC vs.
Revision 1.426 by root, Tue Dec 28 20:56:36 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.14 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
8TODO: wbuf_max for anyevent
9TODO: fix with the rest: POE + IO::Async? No go. Tk + Event? No go. Again: if
10TODO: ioasync stuff
11TODO: install arch-dependen constants.pl in some arch-dependend path. somehow.
12 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
13 to avoid leaking them to any child processes.
14 - try to force files into the arch-dependend lib directory, for the
15 half-"turly obsessive compulsive person"(s) that depend on it.
16
175.29 Sun Dec 5 10:49:21 CET 2010
18 - convert EV backend to EV 4.00 API (so better upgrade EV too).
19 - AE::Handle: implement dynamic read_size adjustment, add
20 max_read_size parameter, reduce default read_size to 2048.
21 - add some FAQ entries for servers and/or on_eof vs. on_error.
22 - work around OS bugs (cygwin again) when lingering in AE::Handle.
23 - reduce memory usage slightly when connect was used in AE::Handle.
24 - wrok around more segfaults in Tk.
25 - document tls_ctx => undef as valid.
26 - detect WNOHANG value at compiletime.
27 - include some probably linux-only support for building AnyEvent
28 as part of the perl core.
29 - improve compatibility of t/08_idna.t to perl 5.8.
30 - make AnyEvent::Strict truly optional (in case somebody builds a
31 minimal perl...).
32
335.28 Wed Oct 13 04:14:23 CEST 2010
34 - due to a glitch, AnyEvent's internal getprotobyname wasn't
35 actually used in tcp_server.
36 - implement AnyEvent::CondVar->new method, which was documented but
37 missing.
38 - support autoloading of read types in unshift_read.
39 - AnyEvent::Strict: do the test for invalid fh arguments differently,
40 resulting in more informative error messages.
41 - new AnyEvent::FAQ manpage.
42 - clarify that you can actually call push_read etc. in on_read.
43 - put keys and certs in /tmp, not into the current directory,
44 and document this in the SECURITY CONSIDERATIONS sections
45 of AnyEvent::TLS.
46 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
47 requested by Ron Isaacson.
48 - document $AnyEvent::DNS::RESOLVER variable.
49 - incorporated major typo patches by Abhijit Menon-Sen.
50 - AnyEvent::Handle now croaks when negative timeouts are passed.
51 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
52 undocumented, but who knows when they might come in handy.
53
545.271 Tue Jun 8 12:05:46 CEST 2010
55 - backport to perl 5.8.x.
56
575.27 Sun Jun 6 12:12:05 CEST 2010
58 - postpone differently in AnyEvent::Socket now, as
59 when not, canceling the connection attempt might fail
60 (found by Felix Antonius Wilhelm Ostmann).
61 - explicitly check for non-stream sockets in AE::Handle, too many
62 clueless people fell into the trap of this somehow working.
63 - simplified and reworked the "OTHER MODULES" section.
64 - better/more condvar examples.
65
665.261 Wed Apr 28 16:13:36 CEST 2010
67 - AF_INET6 was not properly used from Socket6 during configuration
68 time (found by Andy Grundman).
69
705.26 Mon Apr 12 04:49:35 CEST 2010
71 - don't generate spurious readyness notifications when select
72 returns EINTR in the pure perl backend (analysed and testcase
73 by Bas Denissen).
74 - give same examples in SYNOPSIS sections of AnyEvent and AE.
75 - provider faster implementations for AE::time/now/now_update for
76 EV and Perl backends.
77 - provide faster AE::cv and AE::time implementations for all
78 backends.
79 - no longer support lower/mixed-case signal names (this was
80 never documented nor universally supported).
81 - some more memory µ-optimisations, and somewhat less messy
82 code to cope with both AnyEvent and AE APIs.
83
845.251 Sat Mar 13 00:58:21 CET 2010
85 - make test failed if it was run before make install on perl <=
86 5.8.8, and older versions of perl make our live very difficult,
87 so write constants.pl during Makefile.PL time.
88
895.25 Sat Mar 13 00:23:14 CET 2010
90 - fix a race condition in AnyEvent::Handle that would cause
91 a "bio_read: unsupported method" error (in _another_
92 ssl connection) after user code threw an exception.
93 - added AnyEvent::Handle->destroyed method.
94 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
95 - the AnyEvent::Util::fh_nonblocking function worked only
96 by lucky accident on win32.
97 - smaller and faster AnyEvent::Util::fh_nonblocking.
98 - when the (required!) Time::HiRes module is missing, AnyEvent
99 did not fall back to built-in time properly.
100 - do not load Fcntl at runtime, saving memory and loading time.
101 - precompile a number of constants and use them instead of runtime
102 detection and eval.
103 - free detection code after detection and similar memory optimisations.
104 - Perl backend timer interval best effort drift has been improved
105 (same algorithm as EV).
106 - update unicode idna mapping table.
107
1085.24 Tue Jan 5 11:39:43 CET 2010
109 - cygwin never reports errors from failed connects. 1.5 just gives
110 you continous readyness and EAGAIN, 1.7 is even more broken and
111 just hangs. work around both issues in a cygwin-specific hack.
112 - improve idle watcher documentation slightly.
113
1145.23 Sun Dec 20 23:48:00 CET 2009
115 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
116 - implement punycode_encode/decode, idn_nameprep,
117 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
118 - provide $AE::VERSION.
119 - removed traces of "no strict 'refs'".
120
1215.22 Sat Dec 5 03:51:13 CET 2009
122 - downgrade-or-fail in AnyEvent::Handle::push_write, to
123 diagnose encoding failures earlier and more succinctly.
124 (this works around bugs in perl, throwing away encoding info
125 when passing scalar data to extensions).
126 - add more examples to AnyEvent::Socket manpage.
127 - upgrade internal warning set to the same as common::sense 2.03.
128 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
129 (John Beppu).
130 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
131 - add google public dns servers to fallback server set and make sure
132 we load-balance properly between the three sets. also add all
133 fallback dns servers, not just a random one, to each dns config.
134
1355.21 Thu Nov 19 02:48:47 CET 2009
136 - fix a problem where socket constants were called with parameters
137 (spotted by David Friedland).
138 - fork_call never use'd POSIX (reported by Daisuke Maki).
139 - improve perl 5.6 compatibility further (but it still won't work
140 unless you rip out everything but the core).
141 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
142 uses 10MB of RAM, but doesn't open a console window. *sigh*.
143
1445.202 Wed Oct 14 22:35:44 CEST 2009
145 - AnyEvent::DNS would unexpectedly clobber $_ under windows
146 (analysed by Matthias Waldorf).
147 - AnyEvent::Handle::run_cmd can now pass the PID of the
148 newly-created process, which is much less useful than it might
149 sound (based on patch by Yann Kerherve).
150
1515.201 Tue Sep 29 12:09:25 CEST 2009
152 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
153 (reported by Torsten Foertsch).
154 - common::sense 2.0 could cause tcp_server to throw an exception
155 (analysed by elmex).
156
1575.2 Mon Sep 14 07:04:49 CEST 2009
158 - INCOMPATIBLE CHANGE: do no longer support register_read_type
159 and register_write_type in AnyEvent::Handle, instead support
160 package names (the facility was mostly abused).
161 - implement "packagename-as-read/write type" support in
162 AnyEvent::Handle.
163 - AnyEvent::Handle: new options "keepalive" and "oobinline".
164 - oobinline set by default to avoid security issues.
165 - the pure-perl event loop backend wrongly detected times() fallback
166 support (spotted by Pavel Boldin).
167
1685.12 Tue Sep 1 20:26:50 CEST 2009
169 - be more lenient when parsing resolv.conf files, as some people
170 use hashmarks as comment indicator inside directives (reported
171 by Michael S. Fischer).
172 - use same set of warnings as common::sense 2.0.
173 - fix a potential 32 bit overflow issue due to perl having problems
174 with large hex constants in the Perl backend.
175 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
176 by default.
177 - fix some prototypes.
178 - (optionally) require Async::Interrupt 1.02.
179 - replace opendns fallbacks by verizon/level3 public dns servers.
180
1815.112 Fri Aug 21 13:59:04 CEST 2009
182 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
183 using starttls with an illegal/uncreatable tls context.
184 - port to broken win32 perls.
185
1865.111 Sun Aug 16 18:44:36 CEST 2009
187 - perl errornously requires sockaddr_un structures returned by the
188 kernel to have a specific length. work around this bug by padding
189 sockaddr_un structures (found to fail on OS X, as perl only uses
190 the correct code for GNU/Linux, but likely to fail on other
191 OSes as well).
192
1935.11 Wed Aug 12 17:49:37 CEST 2009
194 - fixed a bug in run_cmd with input from filename.
195 - create a json coder object when none is given in write direction
196 to reduce dependency on JSON.
197
1985.1 Tue Aug 11 03:17:32 CEST 2009
199 - tcp_connect now reports the correct error in $!, not an unrelated
200 one (was broken since 4.91).
201 - AE::Handle did not properly clear rtimeout/wtimeout watchers
202 when retrying a tcp_connect.
203 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
204 - fixed wrong documentation in AE::Socket.
205
2065.01 Mon Aug 10 03:16:32 CEST 2009
207 - last release broke EV child watchers :/.
208
2095.0 Sun Aug 9 17:32:09 CEST 2009
210 - officially introduce and document the AE API.
211 - lowering the timeout at runtime did not have immediate effect
212 in AnyEvent::Handle.
213 - AE::Handle now has separate and independent read and write timeout
214 settings.
215 - probe for Guard module when AnyEvent::Util is loaded, not at
216 runtime (Event doesn't like eval "use" from callbacks, and it avoids
217 exporting the probe function).
218 - Impl::IOAsync's io watchers did not use the filehandle emulation
219 code to work around it's watcher limits.
220 - work around IO::Async's problems with overloaded objects.
221 - try to disable wearnings inside IO::Async, it's just too buggy.
222 - do no longer use Event by default, only EV or Perl, Event is too buggy.
223 - hackishly provide fast AE:: interface when EV or Perl is the backend.
224 - inherit VERSION from AnyEvent::VERSION in most modules.
225
2264.91 Thu Aug 6 15:42:45 CEST 2009
227 - AE::Handle::starttls could get out of sync when the read buffer
228 already contains some TLS handshake.
229 - AE::Handle did not properly free the TLS session, causing bigger
230 memory leaks in Net::SSLeay than expected (reported by toaster).
231 - AE::Socket::tcp_connect will now artificially delay invoking
232 the callback to avoid returning after invoking the callback.
233 - convert many internal watcher uses to AE API.
234
2354.9 Sat Aug 1 11:07:01 CEST 2009
236 - Glib silently fails when registering a timeout with
237 a negative value, so avoid that.
238 - call condvar callback immediately when it is set
239 after the condition is already signalled.
240 - check rbuf_max condition only after trying to consume
241 data.
242 - normalise signal numbers to names when using the ae's
243 signal handling, but do not document this (yet).
244 - pure perl signal emulation did not properly set nonblocking
245 mode on the signal pipe on win32 (this is unlikely to have caused
246 any issues).
247 - new module: AnyEvent::Debug.
248 - AnyEvent::Strict now checks that a signal was specified
249 by name (not name or number).
250 - reduce memory footprint in typical cases by ~50kb by
251 only compiling signal code when necessary.
252 - add AnyEvent::Handle->rbuf_max.
253 - grab the AE namespace for future new API, implement stubs for
254 said future/alternative API.
255 - new function: AnyEvent::Socket::format_hostport.
256
2574.881 Tue Jul 28 12:51:53 CEST 2009
258 - work around a bug in local in pre-5.10 perls, causing
259 AnyEvent::Handle to recurse when it should not
260 (analysed by elmex).
261
2624.88 Tue Jul 28 04:04:37 CEST 2009
263 - re-bless the handle into a dummy package after calling
264 AnyEvent::Handle::destroy, so the user does not need to check
265 for errors after every push_write etc.
266 - do not attempt to run t/02_signals.t on obviously broken
267 platforms.
268
2694.87 Sun Jul 26 02:06:16 CEST 2009
270 - do not attempt to linger when there is no longer a valid fh.
271 - enforce tls mode to be either accept or connect, do not simply
272 segfault in Net::SSLeay.
273 - AnyEvent::Handle can now call tcp_connect itself (new parameters
274 connect, on_prepare, on_connect and on_connect_error). Updated
275 tutorial accordingly.
276 - add AnyEvent::Impl::Irssi backend.
277
2784.86 Mon Jul 20 23:52:29 CEST 2009
279 - since the verbose warning is not enough, explicitly document
280 that versions before 1.33 of Net::SSLeay are not secure.
281 - work around signal handling races in Event and (...) Event::Lib.
282 - try to align signal-race timer to full-second boundaries.
283 - work around Tk not liking negative timeouts.
284 - don't complain of different grades of environmental unfriendlyness
285 in IO::Async.
286
2874.85 Sat Jul 18 06:16:14 CEST 2009
288 - nail the signal race problem in perl once and for all
289 (see $AnyEvent::MAX_SIGNAL_LATENCY).
290 - take advantage of Async::Interrupt if it is available.
291 - load Time::HiRes and Guard modules on demand only.
292 - add optional/recommended modules section to AnyEvent
293 documentation.
294 - reduce memory usage considerably (and reduce startup penalty)
295 by not using "strict", "warnings" and "overload" modules.
296 - work around buggy windows/openbsd perls and provide EBADMSG
297 and EPROTO ourselves when missing.
298 - improve perl 5.6 compatibility of the core event loop.
299 - made Net::SSLeay version 1.33 a soft requirement.
300
3014.83 Fri Jul 17 16:56:26 CEST 2009
302 - implement AnyEvent::Socket::getprotobyname.
303 - AnyEvent::CondVar's will now detect recursive blocking
304 waits and will croak, as too many people fall into
305 this trap.
306 - AnyEvent::Handle will now call ->destroy on itself after
307 executing the on_error callback, instead of doing some
308 half-baked internal shutdown, for fatal errors.
309 - clarify on_eof behaviour w.r.t. the read queue and
310 on_read callbacks.
311 - ignore some possible spurious wake-ups in tcp_connect.
312
3134.82 Sat Jul 11 00:34:55 CEST 2009
314 - POE and Event backends didn't accept some callable objects as
315 callbacks.
316 - use Config module instead of POSIX module to detect signal names
317 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
318 module doesn't even have all POSIX signals :/.
319 - use more workarounds around the many refcnt/corruption bugs in
320 Event::Lib.
321 - work around a race condition in perl's select, causing t/03_child.t
322 to rarely fail.
323
3244.81 Thu Jul 9 10:30:30 CEST 2009
325 - AnyEvent::Handle didn't properly diagnose write errors
326 (it expected -1 from syswrite, how lame... :).
327 - support file descriptors in addition to file handles
328 in AnyEvent->io.
329 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
330 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
331 and PERL_ANYEVENT_CA_PATH.
332 - provide a sensible synopsis section for AnyEvent::TLS.
333 - add a "supported backends" section to the manpage.
334 - added simple io watcher test to testsuite, using a
335 portable_socketpair.
336 - tried to improve the stability of the Event::Lib backend,
337 YMMV.
338
3394.8 Mon Jul 6 23:45:16 CEST 2009
340 - AnyEvent::DNS did not properly follow CNAME records with
341 uppercase targets.
342 - AnyEvent::DNS would errornously return AAAA records
343 with v4 mapped addresses (a faulty record) as ipv4 addresses,
344 causing AnyEvent::Socket to throw an exception.
345 - added new module AnyEvent::TLS for easier SSL/TLS context
346 creation, with many options including hostname verification,
347 secure default configuration, lots of documentation and,
348 predefined diffie-hellman keys for perfect forward security
349 and much more. get it while it's still fresh!
350 - use AnyEvent::TLS in AnyEvent::Handle for context management.
351 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
352 so AnyEvent::Socket users have smaller memory footprint
353 in the common case.
354 - add AnyEvent::Handle->push_shutdown method.
355 - add an additional error message parameter to AnyEvent::Handle's
356 on_error callback (for TLS, $! is still available).
357 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
358 - make AnyEvent::Handle more robust against early conenction
359 failures (during new), and return C<undef> in that case
360 from the constructor.
361 - AnyEvent::Handle will now try to load only JSON::XS first,
362 then fall back to JSON.
363 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
364 and because it was documented that way.
365
3664.451 Fri Jul 3 00:28:58 CEST 2009
367 - do not clear rbuf when shutting down an AnyEvent::Handle
368 object - doing so breaks AnyEvent::HTTP.
369
3704.45 Mon Jun 29 22:59:26 CEST 2009
371 - a write error could cause AnyEvent::Handle to create
372 an I/O watcher with an undefined $fh.
373 - special-case mapped ipv4 addresses in both
374 AnyEvent::Socket::format_address and parse_address,
375 to treat them just like ipv4 addresses.
376 - updated and overhauled the AnyEvent::Intro doc.
377 - implement AnyEvent::Socket::format_ipv4/ipv6.
378 - slightly speed up portable_pipe/socketpair.
379 - expand condvar begin/end documentation.
380
3814.42 Fri Jun 26 08:32:18 CEST 2009
382 - preliminary and neccesarily incomplete support for IO::Async.
383 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
384 case it was set to IGNORE, to ensure we can catch child statuses
385 even when the calling env acts stupidly.
386 - updated benchmarks with IO::Async, which performs very well.
387
3884.412 Wed Jun 24 01:35:57 CEST 2009
389 - support an "untaint" attribute for AnyEvent::DNS
390 and set it on the default resolver.
391 - implement and document AnyEvent::Impl::Perl::loop.
392 - remove all anyevent-env variables from %ENV
393 when running in tainted mode.
394 - mention and extend the IO::Lambda benchmark.
395
3964.411 Sun Jun 7 18:48:02 CEST 2009
397 - do not try to use F_SETFD on windows.
398
3994.41 Thu May 14 06:40:11 CEST 2009
400 - work around issues in older perls (5.8.0?) when
401 a signal handler is deleted from the %SIG hash.
402 - use POSIX::_exit in child test, to avoid running
403 destructors.
404 - speed up CHLD handling by relying on SIGCHLD being
405 synchronously delivered, even when we roll our own
406 implementation.
407 - AnyEvent::DNS: add the "dname" resource record name
408 for cosmetic reasons.
409
4104.4 Sun Apr 26 20:12:33 CEST 2009
411 - implemented idle watchers, where applicable.
412 - AnyEvent->time died when Event backend was in use.
413 - fix a memleak in the Tk backend.
414 - sped up Tk timer handling.
415 - clip negative "after" values to 0 in AnyEvent::Impl::Event
416 to avoid spamming out warning messages.
417 - fix Qt timers without interval.
418 - avoid Qt zero-timeout special case, in old libqt's.
419
4204.352 Mon Apr 20 16:31:11 CEST 2009
421 - fix AnyEvent::Strict error messages for child watchers.
422 - fix/update Makefile.PL "recommends" versions.
423 - add AnyEvent->now_update.
424
4254.351 Sat Apr 11 07:56:14 CEST 2009
426 - actually make the signal pipe work under win32.
427 - localise $! in signal handler to avoid changing $!, although
428 perl itself does not seem to save/restore errno either.
429 - set the cloexec flag on the signal pipe (normally set by perl too).
430
4314.35 Fri Mar 27 11:48:20 CET 2009
432 - event models relying on AnyEvent's signal watcher emulation did
433 invoke the callback asynchronously, contradicting documentation
434 and causing signals to get lost (this includes AnyEvent's own
435 event loop). AnyEvent now uses the standard pipe trick to make
436 callback execution synchronous to the event loop.
437 - AnyEvent::Handle didn't free TLS context data on DESTROY
438 (patch by Pavel Shaydo).
439 - work around the ever-incompatibly-changing API of MakeMaker.
440 - document that changing global variables without restoring them
441 is a bad idea in Perl (noted by Adam Rosenstein).
442 - AnyEvent::Strict now barfs if ->io is passed a file.
443
4444.34 Thu Feb 12 18:32:45 CET 2009
445 - separately buffer TLS read data, as otherwise the read queue could
446 deadlock as receiving data is not expected while draining the
447 read queue (which cna only happen with TLS).
448 - raise EBADMSG error on JSON decoding errors.
449 - fix some minor manpage bugs (reported by Maximilian Gaß).
450 - speed up select bitmask parsing quite a bit in the pure perl
451 backend.
452 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
453
4544.331 Tue Jan 6 21:07:25 CET 2009
455 - socketpair fails on many vista machines because vista has
456 completely broken accept/getpeername and getsockname functions,
457 so we provide our own socketpair emulation that kind of works
458 (AnyEvent::Util::portable_pipe).
459 - new function: AnyEvent::Util::portable_socketpair.
460 - take advantage of the Guard module if it exists.
461
4624.33 Fri Nov 21 02:35:40 CET 2008
463 - AnyEvent::Strict did errornously flag a fileno of 0 as
464 illegal.
465 - reduce memory usage and slightly speed up the pure perl backend
466 by only storing the file descriptor, not the file handle.
467 - add missing autocork method to AnyEvent::Handle (reported
468 by Adam Rosenstein).
469 - AnyEvent::DNS->resolve errornously documented an $rcode result
470 argument, but there isn't (spotted by Henrik Krohns).
471 - the naked truth about Tk - it's basically unmaintained.
472
4734.32 Mon Nov 3 22:46:32 CET 2008
474 - fix AnyEvent::Socket::resolve_sockaddr to properly support
475 unix sockets again and choose a proper default for
476 the socket type.
477 - fix call to resolve_sockaddr in tcp_connect.
478
4794.31 Thu Oct 30 04:41:48 CET 2008
480 - implemented AnyEvent::Handle->destroy method.
481 - hint about unexpected effects in TLS mode.
482 - speed up AnyEvent::Impl::Perl by using more arrays
483 instead of hashes without the slightest loss of readability :)
484 - work around a perl argument refcounting bug.
485 - with some perl patches applied, AnyEvent now seems leak-free.
486
4874.3 Fri Oct 3 09:18:43 CEST 2008
488 - AnyEvent will now install a no-op signal handler for SIGPIPE,
489 unless one has been installed already.
490 - warn about not loaidng AnyEvent::Impl::POE early enough
491 (patch by Adam Rosenstein).
492 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
493 - fix a bug in where in SSL connect mode, the
494 client would first wait for some data by the server and
495 otherwise hang. (reported and analysed in an absolutely
496 exemplary manner by Adam Rosenstein).
497 - fix a bug in where SSL EOF would not be treated
498 as stream EOF, putting the connection into a hung state.
499 (reported and analysed in an absolutely exemplary manner by Adam
500 Rosenstein).
501 - fix a potential segfault when the TLS context
502 would go missing in a read callback (e.g. due to stoptls) -
503 Net::SSLeay of course makes no type checking whatsoever.
504 - AnyEvent::Handle will not stop the read watcher in TLS mode.
505 - AnyEvent::Handle->stoptls will send a TLS close notify
506 instead of simply destroying the stream now.
507 - fix a bug where large blocks of data written in TLS mode
508 would not be sent unless triggered by receives.
509 - on_drain will now take the tls write buffer into account.
510 - SSL operations have been streamlined a bit and should be faster
511 now, more to come, though.
512 - added a FAQ section, will document nontrivial issues in there.
513
5144.234 Mon Sep 29 04:08:13 CEST 2008
515 - fix child watcher documentation: only child exits will be reported,
516 no trace events.
517 - mention SIGPIPE in AnyEvent::Handle.
518 - perl backend will now use POSIX::times as monotonic clock
519 when available and the monotonic clock option is not.
520 - run even without Time::HiRes in some cases now, but do not
521 provide sub-second accuracy (all *supported* perl versions have
522 Time::HiRes).
523
5244.233 Fri Aug 22 01:48:05 CEST 2008
525 - fix a bug in the testsuite, causing a hang.
526 - clarified lots of AnyEvent::Handle constructor arguments.
527
5284.232 Thu Aug 21 20:44:25 CEST 2008
529 - fix a bug in AnyEvent::Handle that could cause two fatal error
530 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
531 (testcase by Leon Brocard).
532 - clarified on_eof/on_error documentation a bit.
533
5344.231 Tue Jul 29 13:12:15 CEST 2008
535 - remove some debugging code left in AnyEvent::Util::fork_call
536 (and no, it's impossible to implement with the broken windows
537 perls without resource leaks or worse).
538
5394.23 Tue Jul 29 12:19:59 CEST 2008
540 - document the first parameter passed to condvar callbacks to be
541 the callback.
542 - add AnyEvent::Socket::{ntoa,aton} aliases.
543 - optimize the AE::Handle->push_read (line) for the default
544 eol marker.
545 - optimize push_read (packstring|storable) for small packets.
546 - invoke on_error callback when no on_eof callback is set.
547 - fix a bug in push_read (storable) of unknown impact.
548
5494.22 Sun Jul 20 16:34:13 CEST 2008
550 - new function AnyEvent::Socket::parse_hostport.
551 - as the bulkheads at microsoft can't even get getprotobyname reliably
552 working on their shitty fucking broken os we need to hardcode
553 some common protocol numbers in AnyEvent::Socket. How can
554 people even bother with such a pile of shit as windows.
555
5564.21 Thu Jul 17 14:40:05 CEST 2008
557 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
558 which is incompatible to previous invocation syntax.
559 - work around more windows perl bugs in fork_call. I cannot imagine
560 why anybody would bother implementing fork in such an obviously
561 unusable way.
562 - avoid calling $do_retry if the request has already finished in
563 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
564 K. Sharifulin).
565
5664.2 Sat Jul 12 22:42:11 CEST 2008
567 - support an "interval" argument for repeating timers.
568 - fix a bug in the pure perl event loop that caused timers to
569 be delayed under some circumstances.
570 - implement optional argument checking via PERL_ANYEVENT_STRICT.
571 - not importing anything from Event causes it to malfunction
572 and not use Time::HiRes, work around this bug.
573 - more bug workarounds against the endlessly broken Tk module,
574 seems to work now on 32 bit machines, but less so on 64 bit.
575 - minor optimisations applied to most backends.
576
5774.161 Fri Jul 4 14:24:48 CEST 2008
578 - fixed the t/handle/02_write.t, it read after write, with the
579 assumption that all data could be written before read
580 (reported by Jim Razmus).
581
5824.160 Thu Jul 3 04:02:21 CEST 2008
583 - re-registering signal watchers after unregistering one did
584 not work if a backend falls back on AnyEvent's default
585 implementation (also affected child watchers).
586 - new AnyEvent::Handle options: autocork and no_delay.
587
5884.152 Sun Jun 22 14:15:44 CEST 2008
589 - allow for 32-bit perls that implement shifts differently
590 on different architectures in parse_ipv4 (reported and
591 analysed by Keiichi DAIBA).
592
5934.151 Fri Jun 6 17:34:24 CEST 2008
594 - make sure specifying _only_ on_read and never pushing reads
595 works.
596
5974.15 Fri Jun 6 13:00:46 CEST 2008
598 - the pure perl backend would keep some watchers alive when more than
599 one watcher was registered for the same fd.
600 - new "packstring" and "storable" read and write types
601 for AnyEvent::Handle.
602 - allow on_eof handler to be called after on_error with EPIPE returns.
603 - do not immediately call on_read callback in handle constructor.
604
6054.14 Thu Jun 5 20:29:31 CEST 2008
4 - Fixed a bug in DNS SRV priority sorting. 606 - Fixed a bug in DNS SRV priority sorting.
5 - AnyEvent::Util::guard now reports runtime errors while 607 - AnyEvent::Util::guard now reports runtime errors while
6 executing the guard block as warnings. 608 executing the guard block as warnings.
609 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
7 610
84.13 Thu Jun 5 00:47:59 CEST 2008 6114.13 Thu Jun 5 00:47:59 CEST 2008
9 - AnyEvent::DNS only followed cname chains with length 2, 612 - AnyEvent::DNS only followed cname chains with length 2,
10 contrary to documentation. bumped it up to 10, thanks to 613 contrary to documentation. bumped it up to 10, thanks to
11 microsoft, the current broken-dns-config-king. 614 microsoft, the current broken-dns-config-king.
… …
130 - support service names in tcp_connect and tcp_server. 733 - support service names in tcp_connect and tcp_server.
131 - work around netbsd/openbsd bug causing unix sockets not to 734 - work around netbsd/openbsd bug causing unix sockets not to
132 become writable until empty. 735 become writable until empty.
133 - only make a dummy read for error detection when getpeername 736 - only make a dummy read for error detection when getpeername
134 returns ENOTCONN. 737 returns ENOTCONN.
135 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 738 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
136 override. 739 override.
137 - new AnyEvent::Guard::cancel. 740 - new AnyEvent::Guard::cancel.
138 741
1393.5 Sat May 17 23:17:14 CEST 2008 7423.5 Sat May 17 23:17:14 CEST 2008
140 - gracefully deal with systems not defining CLOCK_MONOTONIC 743 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
154 757
1553.41 Sun May 11 19:53:13 CEST 2008 7583.41 Sun May 11 19:53:13 CEST 2008
156 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 759 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
157 760
1583.4 Sun May 11 00:29:25 CEST 2008 7613.4 Sun May 11 00:29:25 CEST 2008
159 - complete re-implementation of condvars with a hopefully much mroe useful 762 - complete re-implementation of condvars with a hopefully much more useful
160 API (backward compatible functions exist but are undocumented). 763 API (backward compatible functions exist but are undocumented).
161 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 764 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
162 (still subject to change). 765 (still subject to change).
163 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 766 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
164 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 767 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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