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

Comparing AnyEvent/Changes (file contents):
Revision 1.169 by root, Sat Jul 12 20:45:27 2008 UTC vs.
Revision 1.437 by root, Fri Jan 21 13:20:38 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
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
8TODO: resetttls?
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.
15
165.3 Fri Dec 31 05:48:13 CET 2010
17 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
18 issues with IO::Async have been solved in that version, including
19 the availabilty of a default loop!
20 - implement new wbuf_max parameter for AnyEvent::Handle.
21 - added Cocoa::EventLoop backend by Daisuke Murase.
22 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
23 - clarify handle on_prepare documentation.
24 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
25 to avoid leaking them to any child processes.
26 - try to force files into the arch-dependend lib directory, for the
27 half-"turly obsessive compulsive person"(s) that depend on it.
28
295.29 Sun Dec 5 10:49:21 CET 2010
30 - convert EV backend to EV 4.00 API (so better upgrade EV too).
31 - AE::Handle: implement dynamic read_size adjustment, add
32 max_read_size parameter, reduce default read_size to 2048.
33 - add some FAQ entries for servers and/or on_eof vs. on_error.
34 - work around OS bugs (cygwin again) when lingering in AE::Handle.
35 - reduce memory usage slightly when connect was used in AE::Handle.
36 - wrok around more segfaults in Tk.
37 - document tls_ctx => undef as valid.
38 - detect WNOHANG value at compiletime.
39 - include some probably linux-only support for building AnyEvent
40 as part of the perl core.
41 - improve compatibility of t/08_idna.t to perl 5.8.
42 - make AnyEvent::Strict truly optional (in case somebody builds a
43 minimal perl...).
44
455.28 Wed Oct 13 04:14:23 CEST 2010
46 - due to a glitch, AnyEvent's internal getprotobyname wasn't
47 actually used in tcp_server.
48 - implement AnyEvent::CondVar->new method, which was documented but
49 missing.
50 - support autoloading of read types in unshift_read.
51 - AnyEvent::Strict: do the test for invalid fh arguments differently,
52 resulting in more informative error messages.
53 - new AnyEvent::FAQ manpage.
54 - clarify that you can actually call push_read etc. in on_read.
55 - put keys and certs in /tmp, not into the current directory,
56 and document this in the SECURITY CONSIDERATIONS sections
57 of AnyEvent::TLS.
58 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
59 requested by Ron Isaacson.
60 - document $AnyEvent::DNS::RESOLVER variable.
61 - incorporated major typo patches by Abhijit Menon-Sen.
62 - AnyEvent::Handle now croaks when negative timeouts are passed.
63 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
64 undocumented, but who knows when they might come in handy.
65
665.271 Tue Jun 8 12:05:46 CEST 2010
67 - backport to perl 5.8.x.
68
695.27 Sun Jun 6 12:12:05 CEST 2010
70 - postpone differently in AnyEvent::Socket now, as
71 when not, canceling the connection attempt might fail
72 (found by Felix Antonius Wilhelm Ostmann).
73 - explicitly check for non-stream sockets in AE::Handle, too many
74 clueless people fell into the trap of this somehow working.
75 - simplified and reworked the "OTHER MODULES" section.
76 - better/more condvar examples.
77
785.261 Wed Apr 28 16:13:36 CEST 2010
79 - AF_INET6 was not properly used from Socket6 during configuration
80 time (found by Andy Grundman).
81
825.26 Mon Apr 12 04:49:35 CEST 2010
83 - don't generate spurious readyness notifications when select
84 returns EINTR in the pure perl backend (analysed and testcase
85 by Bas Denissen).
86 - give same examples in SYNOPSIS sections of AnyEvent and AE.
87 - provider faster implementations for AE::time/now/now_update for
88 EV and Perl backends.
89 - provide faster AE::cv and AE::time implementations for all
90 backends.
91 - no longer support lower/mixed-case signal names (this was
92 never documented nor universally supported).
93 - some more memory µ-optimisations, and somewhat less messy
94 code to cope with both AnyEvent and AE APIs.
95
965.251 Sat Mar 13 00:58:21 CET 2010
97 - make test failed if it was run before make install on perl <=
98 5.8.8, and older versions of perl make our live very difficult,
99 so write constants.pl during Makefile.PL time.
100
1015.25 Sat Mar 13 00:23:14 CET 2010
102 - fix a race condition in AnyEvent::Handle that would cause
103 a "bio_read: unsupported method" error (in _another_
104 ssl connection) after user code threw an exception.
105 - added AnyEvent::Handle->destroyed method.
106 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
107 - the AnyEvent::Util::fh_nonblocking function worked only
108 by lucky accident on win32.
109 - smaller and faster AnyEvent::Util::fh_nonblocking.
110 - when the (required!) Time::HiRes module is missing, AnyEvent
111 did not fall back to built-in time properly.
112 - do not load Fcntl at runtime, saving memory and loading time.
113 - precompile a number of constants and use them instead of runtime
114 detection and eval.
115 - free detection code after detection and similar memory optimisations.
116 - Perl backend timer interval best effort drift has been improved
117 (same algorithm as EV).
118 - update unicode idna mapping table.
119
1205.24 Tue Jan 5 11:39:43 CET 2010
121 - cygwin never reports errors from failed connects. 1.5 just gives
122 you continous readyness and EAGAIN, 1.7 is even more broken and
123 just hangs. work around both issues in a cygwin-specific hack.
124 - improve idle watcher documentation slightly.
125
1265.23 Sun Dec 20 23:48:00 CET 2009
127 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
128 - implement punycode_encode/decode, idn_nameprep,
129 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
130 - provide $AE::VERSION.
131 - removed traces of "no strict 'refs'".
132
1335.22 Sat Dec 5 03:51:13 CET 2009
134 - downgrade-or-fail in AnyEvent::Handle::push_write, to
135 diagnose encoding failures earlier and more succinctly.
136 (this works around bugs in perl, throwing away encoding info
137 when passing scalar data to extensions).
138 - add more examples to AnyEvent::Socket manpage.
139 - upgrade internal warning set to the same as common::sense 2.03.
140 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
141 (John Beppu).
142 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
143 - add google public dns servers to fallback server set and make sure
144 we load-balance properly between the three sets. also add all
145 fallback dns servers, not just a random one, to each dns config.
146
1475.21 Thu Nov 19 02:48:47 CET 2009
148 - fix a problem where socket constants were called with parameters
149 (spotted by David Friedland).
150 - fork_call never use'd POSIX (reported by Daisuke Maki).
151 - improve perl 5.6 compatibility further (but it still won't work
152 unless you rip out everything but the core).
153 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
154 uses 10MB of RAM, but doesn't open a console window. *sigh*.
155
1565.202 Wed Oct 14 22:35:44 CEST 2009
157 - AnyEvent::DNS would unexpectedly clobber $_ under windows
158 (analysed by Matthias Waldorf).
159 - AnyEvent::Handle::run_cmd can now pass the PID of the
160 newly-created process, which is much less useful than it might
161 sound (based on patch by Yann Kerherve).
162
1635.201 Tue Sep 29 12:09:25 CEST 2009
164 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
165 (reported by Torsten Foertsch).
166 - common::sense 2.0 could cause tcp_server to throw an exception
167 (analysed by elmex).
168
1695.2 Mon Sep 14 07:04:49 CEST 2009
170 - INCOMPATIBLE CHANGE: do no longer support register_read_type
171 and register_write_type in AnyEvent::Handle, instead support
172 package names (the facility was mostly abused).
173 - implement "packagename-as-read/write type" support in
174 AnyEvent::Handle.
175 - AnyEvent::Handle: new options "keepalive" and "oobinline".
176 - oobinline set by default to avoid security issues.
177 - the pure-perl event loop backend wrongly detected times() fallback
178 support (spotted by Pavel Boldin).
179
1805.12 Tue Sep 1 20:26:50 CEST 2009
181 - be more lenient when parsing resolv.conf files, as some people
182 use hashmarks as comment indicator inside directives (reported
183 by Michael S. Fischer).
184 - use same set of warnings as common::sense 2.0.
185 - fix a potential 32 bit overflow issue due to perl having problems
186 with large hex constants in the Perl backend.
187 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
188 by default.
189 - fix some prototypes.
190 - (optionally) require Async::Interrupt 1.02.
191 - replace opendns fallbacks by verizon/level3 public dns servers.
192
1935.112 Fri Aug 21 13:59:04 CEST 2009
194 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
195 using starttls with an illegal/uncreatable tls context.
196 - port to broken win32 perls.
197
1985.111 Sun Aug 16 18:44:36 CEST 2009
199 - perl errornously requires sockaddr_un structures returned by the
200 kernel to have a specific length. work around this bug by padding
201 sockaddr_un structures (found to fail on OS X, as perl only uses
202 the correct code for GNU/Linux, but likely to fail on other
203 OSes as well).
204
2055.11 Wed Aug 12 17:49:37 CEST 2009
206 - fixed a bug in run_cmd with input from filename.
207 - create a json coder object when none is given in write direction
208 to reduce dependency on JSON.
209
2105.1 Tue Aug 11 03:17:32 CEST 2009
211 - tcp_connect now reports the correct error in $!, not an unrelated
212 one (was broken since 4.91).
213 - AE::Handle did not properly clear rtimeout/wtimeout watchers
214 when retrying a tcp_connect.
215 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
216 - fixed wrong documentation in AE::Socket.
217
2185.01 Mon Aug 10 03:16:32 CEST 2009
219 - last release broke EV child watchers :/.
220
2215.0 Sun Aug 9 17:32:09 CEST 2009
222 - officially introduce and document the AE API.
223 - lowering the timeout at runtime did not have immediate effect
224 in AnyEvent::Handle.
225 - AE::Handle now has separate and independent read and write timeout
226 settings.
227 - probe for Guard module when AnyEvent::Util is loaded, not at
228 runtime (Event doesn't like eval "use" from callbacks, and it avoids
229 exporting the probe function).
230 - Impl::IOAsync's io watchers did not use the filehandle emulation
231 code to work around it's watcher limits.
232 - work around IO::Async's problems with overloaded objects.
233 - try to disable wearnings inside IO::Async, it's just too buggy.
234 - do no longer use Event by default, only EV or Perl, Event is too buggy.
235 - hackishly provide fast AE:: interface when EV or Perl is the backend.
236 - inherit VERSION from AnyEvent::VERSION in most modules.
237
2384.91 Thu Aug 6 15:42:45 CEST 2009
239 - AE::Handle::starttls could get out of sync when the read buffer
240 already contains some TLS handshake.
241 - AE::Handle did not properly free the TLS session, causing bigger
242 memory leaks in Net::SSLeay than expected (reported by toaster).
243 - AE::Socket::tcp_connect will now artificially delay invoking
244 the callback to avoid returning after invoking the callback.
245 - convert many internal watcher uses to AE API.
246
2474.9 Sat Aug 1 11:07:01 CEST 2009
248 - Glib silently fails when registering a timeout with
249 a negative value, so avoid that.
250 - call condvar callback immediately when it is set
251 after the condition is already signalled.
252 - check rbuf_max condition only after trying to consume
253 data.
254 - normalise signal numbers to names when using the ae's
255 signal handling, but do not document this (yet).
256 - pure perl signal emulation did not properly set nonblocking
257 mode on the signal pipe on win32 (this is unlikely to have caused
258 any issues).
259 - new module: AnyEvent::Debug.
260 - AnyEvent::Strict now checks that a signal was specified
261 by name (not name or number).
262 - reduce memory footprint in typical cases by ~50kb by
263 only compiling signal code when necessary.
264 - add AnyEvent::Handle->rbuf_max.
265 - grab the AE namespace for future new API, implement stubs for
266 said future/alternative API.
267 - new function: AnyEvent::Socket::format_hostport.
268
2694.881 Tue Jul 28 12:51:53 CEST 2009
270 - work around a bug in local in pre-5.10 perls, causing
271 AnyEvent::Handle to recurse when it should not
272 (analysed by elmex).
273
2744.88 Tue Jul 28 04:04:37 CEST 2009
275 - re-bless the handle into a dummy package after calling
276 AnyEvent::Handle::destroy, so the user does not need to check
277 for errors after every push_write etc.
278 - do not attempt to run t/02_signals.t on obviously broken
279 platforms.
280
2814.87 Sun Jul 26 02:06:16 CEST 2009
282 - do not attempt to linger when there is no longer a valid fh.
283 - enforce tls mode to be either accept or connect, do not simply
284 segfault in Net::SSLeay.
285 - AnyEvent::Handle can now call tcp_connect itself (new parameters
286 connect, on_prepare, on_connect and on_connect_error). Updated
287 tutorial accordingly.
288 - add AnyEvent::Impl::Irssi backend.
289
2904.86 Mon Jul 20 23:52:29 CEST 2009
291 - since the verbose warning is not enough, explicitly document
292 that versions before 1.33 of Net::SSLeay are not secure.
293 - work around signal handling races in Event and (...) Event::Lib.
294 - try to align signal-race timer to full-second boundaries.
295 - work around Tk not liking negative timeouts.
296 - don't complain of different grades of environmental unfriendlyness
297 in IO::Async.
298
2994.85 Sat Jul 18 06:16:14 CEST 2009
300 - nail the signal race problem in perl once and for all
301 (see $AnyEvent::MAX_SIGNAL_LATENCY).
302 - take advantage of Async::Interrupt if it is available.
303 - load Time::HiRes and Guard modules on demand only.
304 - add optional/recommended modules section to AnyEvent
305 documentation.
306 - reduce memory usage considerably (and reduce startup penalty)
307 by not using "strict", "warnings" and "overload" modules.
308 - work around buggy windows/openbsd perls and provide EBADMSG
309 and EPROTO ourselves when missing.
310 - improve perl 5.6 compatibility of the core event loop.
311 - made Net::SSLeay version 1.33 a soft requirement.
312
3134.83 Fri Jul 17 16:56:26 CEST 2009
314 - implement AnyEvent::Socket::getprotobyname.
315 - AnyEvent::CondVar's will now detect recursive blocking
316 waits and will croak, as too many people fall into
317 this trap.
318 - AnyEvent::Handle will now call ->destroy on itself after
319 executing the on_error callback, instead of doing some
320 half-baked internal shutdown, for fatal errors.
321 - clarify on_eof behaviour w.r.t. the read queue and
322 on_read callbacks.
323 - ignore some possible spurious wake-ups in tcp_connect.
324
3254.82 Sat Jul 11 00:34:55 CEST 2009
326 - POE and Event backends didn't accept some callable objects as
327 callbacks.
328 - use Config module instead of POSIX module to detect signal names
329 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
330 module doesn't even have all POSIX signals :/.
331 - use more workarounds around the many refcnt/corruption bugs in
332 Event::Lib.
333 - work around a race condition in perl's select, causing t/03_child.t
334 to rarely fail.
335
3364.81 Thu Jul 9 10:30:30 CEST 2009
337 - AnyEvent::Handle didn't properly diagnose write errors
338 (it expected -1 from syswrite, how lame... :).
339 - support file descriptors in addition to file handles
340 in AnyEvent->io.
341 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
342 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
343 and PERL_ANYEVENT_CA_PATH.
344 - provide a sensible synopsis section for AnyEvent::TLS.
345 - add a "supported backends" section to the manpage.
346 - added simple io watcher test to testsuite, using a
347 portable_socketpair.
348 - tried to improve the stability of the Event::Lib backend,
349 YMMV.
350
3514.8 Mon Jul 6 23:45:16 CEST 2009
352 - AnyEvent::DNS did not properly follow CNAME records with
353 uppercase targets.
354 - AnyEvent::DNS would errornously return AAAA records
355 with v4 mapped addresses (a faulty record) as ipv4 addresses,
356 causing AnyEvent::Socket to throw an exception.
357 - added new module AnyEvent::TLS for easier SSL/TLS context
358 creation, with many options including hostname verification,
359 secure default configuration, lots of documentation and,
360 predefined diffie-hellman keys for perfect forward security
361 and much more. get it while it's still fresh!
362 - use AnyEvent::TLS in AnyEvent::Handle for context management.
363 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
364 so AnyEvent::Socket users have smaller memory footprint
365 in the common case.
366 - add AnyEvent::Handle->push_shutdown method.
367 - add an additional error message parameter to AnyEvent::Handle's
368 on_error callback (for TLS, $! is still available).
369 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
370 - make AnyEvent::Handle more robust against early conenction
371 failures (during new), and return C<undef> in that case
372 from the constructor.
373 - AnyEvent::Handle will now try to load only JSON::XS first,
374 then fall back to JSON.
375 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
376 and because it was documented that way.
377
3784.451 Fri Jul 3 00:28:58 CEST 2009
379 - do not clear rbuf when shutting down an AnyEvent::Handle
380 object - doing so breaks AnyEvent::HTTP.
381
3824.45 Mon Jun 29 22:59:26 CEST 2009
383 - a write error could cause AnyEvent::Handle to create
384 an I/O watcher with an undefined $fh.
385 - special-case mapped ipv4 addresses in both
386 AnyEvent::Socket::format_address and parse_address,
387 to treat them just like ipv4 addresses.
388 - updated and overhauled the AnyEvent::Intro doc.
389 - implement AnyEvent::Socket::format_ipv4/ipv6.
390 - slightly speed up portable_pipe/socketpair.
391 - expand condvar begin/end documentation.
392
3934.42 Fri Jun 26 08:32:18 CEST 2009
394 - preliminary and neccesarily incomplete support for IO::Async.
395 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
396 case it was set to IGNORE, to ensure we can catch child statuses
397 even when the calling env acts stupidly.
398 - updated benchmarks with IO::Async, which performs very well.
399
4004.412 Wed Jun 24 01:35:57 CEST 2009
401 - support an "untaint" attribute for AnyEvent::DNS
402 and set it on the default resolver.
403 - implement and document AnyEvent::Impl::Perl::loop.
404 - remove all anyevent-env variables from %ENV
405 when running in tainted mode.
406 - mention and extend the IO::Lambda benchmark.
407
4084.411 Sun Jun 7 18:48:02 CEST 2009
409 - do not try to use F_SETFD on windows.
410
4114.41 Thu May 14 06:40:11 CEST 2009
412 - work around issues in older perls (5.8.0?) when
413 a signal handler is deleted from the %SIG hash.
414 - use POSIX::_exit in child test, to avoid running
415 destructors.
416 - speed up CHLD handling by relying on SIGCHLD being
417 synchronously delivered, even when we roll our own
418 implementation.
419 - AnyEvent::DNS: add the "dname" resource record name
420 for cosmetic reasons.
421
4224.4 Sun Apr 26 20:12:33 CEST 2009
423 - implemented idle watchers, where applicable.
424 - AnyEvent->time died when Event backend was in use.
425 - fix a memleak in the Tk backend.
426 - sped up Tk timer handling.
427 - clip negative "after" values to 0 in AnyEvent::Impl::Event
428 to avoid spamming out warning messages.
429 - fix Qt timers without interval.
430 - avoid Qt zero-timeout special case, in old libqt's.
431
4324.352 Mon Apr 20 16:31:11 CEST 2009
433 - fix AnyEvent::Strict error messages for child watchers.
434 - fix/update Makefile.PL "recommends" versions.
435 - add AnyEvent->now_update.
436
4374.351 Sat Apr 11 07:56:14 CEST 2009
438 - actually make the signal pipe work under win32.
439 - localise $! in signal handler to avoid changing $!, although
440 perl itself does not seem to save/restore errno either.
441 - set the cloexec flag on the signal pipe (normally set by perl too).
442
4434.35 Fri Mar 27 11:48:20 CET 2009
444 - event models relying on AnyEvent's signal watcher emulation did
445 invoke the callback asynchronously, contradicting documentation
446 and causing signals to get lost (this includes AnyEvent's own
447 event loop). AnyEvent now uses the standard pipe trick to make
448 callback execution synchronous to the event loop.
449 - AnyEvent::Handle didn't free TLS context data on DESTROY
450 (patch by Pavel Shaydo).
451 - work around the ever-incompatibly-changing API of MakeMaker.
452 - document that changing global variables without restoring them
453 is a bad idea in Perl (noted by Adam Rosenstein).
454 - AnyEvent::Strict now barfs if ->io is passed a file.
455
4564.34 Thu Feb 12 18:32:45 CET 2009
457 - separately buffer TLS read data, as otherwise the read queue could
458 deadlock as receiving data is not expected while draining the
459 read queue (which cna only happen with TLS).
460 - raise EBADMSG error on JSON decoding errors.
461 - fix some minor manpage bugs (reported by Maximilian Gaß).
462 - speed up select bitmask parsing quite a bit in the pure perl
463 backend.
464 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
465
4664.331 Tue Jan 6 21:07:25 CET 2009
467 - socketpair fails on many vista machines because vista has
468 completely broken accept/getpeername and getsockname functions,
469 so we provide our own socketpair emulation that kind of works
470 (AnyEvent::Util::portable_pipe).
471 - new function: AnyEvent::Util::portable_socketpair.
472 - take advantage of the Guard module if it exists.
473
4744.33 Fri Nov 21 02:35:40 CET 2008
475 - AnyEvent::Strict did errornously flag a fileno of 0 as
476 illegal.
477 - reduce memory usage and slightly speed up the pure perl backend
478 by only storing the file descriptor, not the file handle.
479 - add missing autocork method to AnyEvent::Handle (reported
480 by Adam Rosenstein).
481 - AnyEvent::DNS->resolve errornously documented an $rcode result
482 argument, but there isn't (spotted by Henrik Krohns).
483 - the naked truth about Tk - it's basically unmaintained.
484
4854.32 Mon Nov 3 22:46:32 CET 2008
486 - fix AnyEvent::Socket::resolve_sockaddr to properly support
487 unix sockets again and choose a proper default for
488 the socket type.
489 - fix call to resolve_sockaddr in tcp_connect.
490
4914.31 Thu Oct 30 04:41:48 CET 2008
492 - implemented AnyEvent::Handle->destroy method.
493 - hint about unexpected effects in TLS mode.
494 - speed up AnyEvent::Impl::Perl by using more arrays
495 instead of hashes without the slightest loss of readability :)
496 - work around a perl argument refcounting bug.
497 - with some perl patches applied, AnyEvent now seems leak-free.
498
4994.3 Fri Oct 3 09:18:43 CEST 2008
500 - AnyEvent will now install a no-op signal handler for SIGPIPE,
501 unless one has been installed already.
502 - warn about not loaidng AnyEvent::Impl::POE early enough
503 (patch by Adam Rosenstein).
504 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
505 - fix a bug in where in SSL connect mode, the
506 client would first wait for some data by the server and
507 otherwise hang. (reported and analysed in an absolutely
508 exemplary manner by Adam Rosenstein).
509 - fix a bug in where SSL EOF would not be treated
510 as stream EOF, putting the connection into a hung state.
511 (reported and analysed in an absolutely exemplary manner by Adam
512 Rosenstein).
513 - fix a potential segfault when the TLS context
514 would go missing in a read callback (e.g. due to stoptls) -
515 Net::SSLeay of course makes no type checking whatsoever.
516 - AnyEvent::Handle will not stop the read watcher in TLS mode.
517 - AnyEvent::Handle->stoptls will send a TLS close notify
518 instead of simply destroying the stream now.
519 - fix a bug where large blocks of data written in TLS mode
520 would not be sent unless triggered by receives.
521 - on_drain will now take the tls write buffer into account.
522 - SSL operations have been streamlined a bit and should be faster
523 now, more to come, though.
524 - added a FAQ section, will document nontrivial issues in there.
525
5264.234 Mon Sep 29 04:08:13 CEST 2008
527 - fix child watcher documentation: only child exits will be reported,
528 no trace events.
529 - mention SIGPIPE in AnyEvent::Handle.
530 - perl backend will now use POSIX::times as monotonic clock
531 when available and the monotonic clock option is not.
532 - run even without Time::HiRes in some cases now, but do not
533 provide sub-second accuracy (all *supported* perl versions have
534 Time::HiRes).
535
5364.233 Fri Aug 22 01:48:05 CEST 2008
537 - fix a bug in the testsuite, causing a hang.
538 - clarified lots of AnyEvent::Handle constructor arguments.
539
5404.232 Thu Aug 21 20:44:25 CEST 2008
541 - fix a bug in AnyEvent::Handle that could cause two fatal error
542 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
543 (testcase by Leon Brocard).
544 - clarified on_eof/on_error documentation a bit.
545
5464.231 Tue Jul 29 13:12:15 CEST 2008
547 - remove some debugging code left in AnyEvent::Util::fork_call
548 (and no, it's impossible to implement with the broken windows
549 perls without resource leaks or worse).
550
5514.23 Tue Jul 29 12:19:59 CEST 2008
552 - document the first parameter passed to condvar callbacks to be
553 the callback.
554 - add AnyEvent::Socket::{ntoa,aton} aliases.
555 - optimize the AE::Handle->push_read (line) for the default
556 eol marker.
557 - optimize push_read (packstring|storable) for small packets.
558 - invoke on_error callback when no on_eof callback is set.
559 - fix a bug in push_read (storable) of unknown impact.
560
5614.22 Sun Jul 20 16:34:13 CEST 2008
562 - new function AnyEvent::Socket::parse_hostport.
563 - as the bulkheads at microsoft can't even get getprotobyname reliably
564 working on their shitty fucking broken os we need to hardcode
565 some common protocol numbers in AnyEvent::Socket. How can
566 people even bother with such a pile of shit as windows.
567
5684.21 Thu Jul 17 14:40:05 CEST 2008
569 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
570 which is incompatible to previous invocation syntax.
571 - work around more windows perl bugs in fork_call. I cannot imagine
572 why anybody would bother implementing fork in such an obviously
573 unusable way.
574 - avoid calling $do_retry if the request has already finished in
575 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
576 K. Sharifulin).
2 577
34.2 Sat Jul 12 22:42:11 CEST 2008 5784.2 Sat Jul 12 22:42:11 CEST 2008
4 - support an "interval" argument for repeating timers. 579 - support an "interval" argument for repeating timers.
5 - fix a bug in the pure perl event loop that caused timers to 580 - fix a bug in the pure perl event loop that caused timers to
6 be delayed under some circumstances. 581 be delayed under some circumstances.
… …
170 - support service names in tcp_connect and tcp_server. 745 - support service names in tcp_connect and tcp_server.
171 - work around netbsd/openbsd bug causing unix sockets not to 746 - work around netbsd/openbsd bug causing unix sockets not to
172 become writable until empty. 747 become writable until empty.
173 - only make a dummy read for error detection when getpeername 748 - only make a dummy read for error detection when getpeername
174 returns ENOTCONN. 749 returns ENOTCONN.
175 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 750 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
176 override. 751 override.
177 - new AnyEvent::Guard::cancel. 752 - new AnyEvent::Guard::cancel.
178 753
1793.5 Sat May 17 23:17:14 CEST 2008 7543.5 Sat May 17 23:17:14 CEST 2008
180 - gracefully deal with systems not defining CLOCK_MONOTONIC 755 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
194 769
1953.41 Sun May 11 19:53:13 CEST 2008 7703.41 Sun May 11 19:53:13 CEST 2008
196 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 771 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
197 772
1983.4 Sun May 11 00:29:25 CEST 2008 7733.4 Sun May 11 00:29:25 CEST 2008
199 - complete re-implementation of condvars with a hopefully much mroe useful 774 - complete re-implementation of condvars with a hopefully much more useful
200 API (backward compatible functions exist but are undocumented). 775 API (backward compatible functions exist but are undocumented).
201 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 776 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
202 (still subject to change). 777 (still subject to change).
203 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 778 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
204 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 779 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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