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

Comparing AnyEvent/Changes (file contents):
Revision 1.163 by root, Fri Jul 4 12:25:44 2008 UTC vs.
Revision 1.418 by root, Mon Nov 15 17:11:00 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent struct and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict etc. for AE::
7TODO: anyevent strict checking for $_ changes?
8
9 - convert EV backend to EV 4.00 API.
10 - AE::Handle: implement dynamic read_size adjustment, add
11 max_read_size parameter, reduce default read_size to 2048.
12 - work around OS bugs (cygwin again) when lingering in AE::Handle.
13 - reduce memory usage slightly when connect was used in AE::Handle.
14
155.28 Wed Oct 13 04:14:23 CEST 2010
16 - due to a glitch, AnyEvent's internal getprotobyname wasn't
17 actually used in tcp_server.
18 - implement AnyEvent::CondVar->new method, which was documented but
19 missing.
20 - support autoloading of read types in unshift_read.
21 - AnyEvent::Strict: do the test for invalid fh arguments differently,
22 resulting in more informative error messages.
23 - new AnyEvent::FAQ manpage.
24 - clarify that you can actually call push_read etc. in on_read.
25 - put keys and certs in /tmp, not into the current directory,
26 and document this in the SECURITY CONSIDERATIONS sections
27 of AnyEvent::TLS.
28 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
29 requested by Ron Isaacson.
30 - document $AnyEvent::DNS::RESOLVER variable.
31 - incorporated major typo patches by Abhijit Menon-Sen.
32 - AnyEvent::Handle now croaks when negative timeouts are passed.
33 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
34 undocumented, but who knows when they might come in handy.
35
365.271 Tue Jun 8 12:05:46 CEST 2010
37 - backport to perl 5.8.x.
38
395.27 Sun Jun 6 12:12:05 CEST 2010
40 - postpone differently in AnyEvent::Socket now, as
41 when not, canceling the connection attempt might fail
42 (found by Felix Antonius Wilhelm Ostmann).
43 - explicitly check for non-stream sockets in AE::Handle, too many
44 clueless people fell into the trap of this somehow working.
45 - simplified and reworked the "OTHER MODULES" section.
46 - better/more condvar examples.
47
485.261 Wed Apr 28 16:13:36 CEST 2010
49 - AF_INET6 was not properly used from Socket6 during configuration
50 time (found by Andy Grundman).
51
525.26 Mon Apr 12 04:49:35 CEST 2010
53 - don't generate spurious readyness notifications when select
54 returns EINTR in the pure perl backend (analysed and testcase
55 by Bas Denissen).
56 - give same examples in SYNOPSIS sections of AnyEvent and AE.
57 - provider faster implementations for AE::time/now/now_update for
58 EV and Perl backends.
59 - provide faster AE::cv and AE::time implementations for all
60 backends.
61 - no longer support lower/mixed-case signal names (this was
62 never documented nor universally supported).
63 - some more memory µ-optimisations, and somewhat less messy
64 code to cope with both AnyEvent and AE APIs.
65
665.251 Sat Mar 13 00:58:21 CET 2010
67 - make test failed if it was run before make install on perl <=
68 5.8.8, and older versions of perl make our live very difficult,
69 so write constants.pl during Makefile.PL time.
70
715.25 Sat Mar 13 00:23:14 CET 2010
72 - fix a race condition in AnyEvent::Handle that would cause
73 a "bio_read: unsupported method" error (in _another_
74 ssl connection) after user code threw an exception.
75 - added AnyEvent::Handle->destroyed method.
76 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
77 - the AnyEvent::Util::fh_nonblocking function worked only
78 by lucky accident on win32.
79 - smaller and faster AnyEvent::Util::fh_nonblocking.
80 - when the (required!) Time::HiRes module is missing, AnyEvent
81 did not fall back to built-in time properly.
82 - do not load Fcntl at runtime, saving memory and loading time.
83 - precompile a number of constants and use them instead of runtime
84 detection and eval.
85 - free detection code after detection and similar memory optimisations.
86 - Perl backend timer interval best effort drift has been improved
87 (same algorithm as EV).
88 - update unicode idna mapping table.
89
905.24 Tue Jan 5 11:39:43 CET 2010
91 - cygwin never reports errors from failed connects. 1.5 just gives
92 you continous readyness and EAGAIN, 1.7 is even more broken and
93 just hangs. work around both issues in a cygwin-specific hack.
94 - improve idle watcher documentation slightly.
95
965.23 Sun Dec 20 23:48:00 CET 2009
97 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
98 - implement punycode_encode/decode, idn_nameprep,
99 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
100 - provide $AE::VERSION.
101 - removed traces of "no strict 'refs'".
102
1035.22 Sat Dec 5 03:51:13 CET 2009
104 - downgrade-or-fail in AnyEvent::Handle::push_write, to
105 diagnose encoding failures earlier and more succinctly.
106 (this works around bugs in perl, throwing away encoding info
107 when passing scalar data to extensions).
108 - add more examples to AnyEvent::Socket manpage.
109 - upgrade internal warning set to the same as common::sense 2.03.
110 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
111 (John Beppu).
112 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
113 - add google public dns servers to fallback server set and make sure
114 we load-balance properly between the three sets. also add all
115 fallback dns servers, not just a random one, to each dns config.
116
1175.21 Thu Nov 19 02:48:47 CET 2009
118 - fix a problem where socket constants were called with parameters
119 (spotted by David Friedland).
120 - fork_call never use'd POSIX (reported by Daisuke Maki).
121 - improve perl 5.6 compatibility further (but it still won't work
122 unless you rip out everything but the core).
123 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
124 uses 10MB of RAM, but doesn't open a console window. *sigh*.
125
1265.202 Wed Oct 14 22:35:44 CEST 2009
127 - AnyEvent::DNS would unexpectedly clobber $_ under windows
128 (analysed by Matthias Waldorf).
129 - AnyEvent::Handle::run_cmd can now pass the PID of the
130 newly-created process, which is much less useful than it might
131 sound (based on patch by Yann Kerherve).
132
1335.201 Tue Sep 29 12:09:25 CEST 2009
134 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
135 (reported by Torsten Foertsch).
136 - common::sense 2.0 could cause tcp_server to throw an exception
137 (analysed by elmex).
138
1395.2 Mon Sep 14 07:04:49 CEST 2009
140 - INCOMPATIBLE CHANGE: do no longer support register_read_type
141 and register_write_type in AnyEvent::Handle, instead support
142 package names (the facility was mostly abused).
143 - implement "packagename-as-read/write type" support in
144 AnyEvent::Handle.
145 - AnyEvent::Handle: new options "keepalive" and "oobinline".
146 - oobinline set by default to avoid security issues.
147 - the pure-perl event loop backend wrongly detected times() fallback
148 support (spotted by Pavel Boldin).
149
1505.12 Tue Sep 1 20:26:50 CEST 2009
151 - be more lenient when parsing resolv.conf files, as some people
152 use hashmarks as comment indicator inside directives (reported
153 by Michael S. Fischer).
154 - use same set of warnings as common::sense 2.0.
155 - fix a potential 32 bit overflow issue due to perl having problems
156 with large hex constants in the Perl backend.
157 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
158 by default.
159 - fix some prototypes.
160 - (optionally) require Async::Interrupt 1.02.
161 - replace opendns fallbacks by verizon/level3 public dns servers.
162
1635.112 Fri Aug 21 13:59:04 CEST 2009
164 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
165 using starttls with an illegal/uncreatable tls context.
166 - port to broken win32 perls.
167
1685.111 Sun Aug 16 18:44:36 CEST 2009
169 - perl errornously requires sockaddr_un structures returned by the
170 kernel to have a specific length. work around this bug by padding
171 sockaddr_un structures (found to fail on OS X, as perl only uses
172 the correct code for GNU/Linux, but likely to fail on other
173 OSes as well).
174
1755.11 Wed Aug 12 17:49:37 CEST 2009
176 - fixed a bug in run_cmd with input from filename.
177 - create a json coder object when none is given in write direction
178 to reduce dependency on JSON.
179
1805.1 Tue Aug 11 03:17:32 CEST 2009
181 - tcp_connect now reports the correct error in $!, not an unrelated
182 one (was broken since 4.91).
183 - AE::Handle did not properly clear rtimeout/wtimeout watchers
184 when retrying a tcp_connect.
185 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
186 - fixed wrong documentation in AE::Socket.
187
1885.01 Mon Aug 10 03:16:32 CEST 2009
189 - last release broke EV child watchers :/.
190
1915.0 Sun Aug 9 17:32:09 CEST 2009
192 - officially introduce and document the AE API.
193 - lowering the timeout at runtime did not have immediate effect
194 in AnyEvent::Handle.
195 - AE::Handle now has separate and independent read and write timeout
196 settings.
197 - probe for Guard module when AnyEvent::Util is loaded, not at
198 runtime (Event doesn't like eval "use" from callbacks, and it avoids
199 exporting the probe function).
200 - Impl::IOAsync's io watchers did not use the filehandle emulation
201 code to work around it's watcher limits.
202 - work around IO::Async's problems with overloaded objects.
203 - try to disable wearnings inside IO::Async, it's just too buggy.
204 - do no longer use Event by default, only EV or Perl, Event is too buggy.
205 - hackishly provide fast AE:: interface when EV or Perl is the backend.
206 - inherit VERSION from AnyEvent::VERSION in most modules.
207
2084.91 Thu Aug 6 15:42:45 CEST 2009
209 - AE::Handle::starttls could get out of sync when the read buffer
210 already contains some TLS handshake.
211 - AE::Handle did not properly free the TLS session, causing bigger
212 memory leaks in Net::SSLeay than expected (reported by toaster).
213 - AE::Socket::tcp_connect will now artificially delay invoking
214 the callback to avoid returning after invoking the callback.
215 - convert many internal watcher uses to AE API.
216
2174.9 Sat Aug 1 11:07:01 CEST 2009
218 - Glib silently fails when registering a timeout with
219 a negative value, so avoid that.
220 - call condvar callback immediately when it is set
221 after the condition is already signalled.
222 - check rbuf_max condition only after trying to consume
223 data.
224 - normalise signal numbers to names when using the ae's
225 signal handling, but do not document this (yet).
226 - pure perl signal emulation did not properly set nonblocking
227 mode on the signal pipe on win32 (this is unlikely to have caused
228 any issues).
229 - new module: AnyEvent::Debug.
230 - AnyEvent::Strict now checks that a signal was specified
231 by name (not name or number).
232 - reduce memory footprint in typical cases by ~50kb by
233 only compiling signal code when necessary.
234 - add AnyEvent::Handle->rbuf_max.
235 - grab the AE namespace for future new API, implement stubs for
236 said future/alternative API.
237 - new function: AnyEvent::Socket::format_hostport.
238
2394.881 Tue Jul 28 12:51:53 CEST 2009
240 - work around a bug in local in pre-5.10 perls, causing
241 AnyEvent::Handle to recurse when it should not
242 (analysed by elmex).
243
2444.88 Tue Jul 28 04:04:37 CEST 2009
245 - re-bless the handle into a dummy package after calling
246 AnyEvent::Handle::destroy, so the user does not need to check
247 for errors after every push_write etc.
248 - do not attempt to run t/02_signals.t on obviously broken
249 platforms.
250
2514.87 Sun Jul 26 02:06:16 CEST 2009
252 - do not attempt to linger when there is no longer a valid fh.
253 - enforce tls mode to be either accept or connect, do not simply
254 segfault in Net::SSLeay.
255 - AnyEvent::Handle can now call tcp_connect itself (new parameters
256 connect, on_prepare, on_connect and on_connect_error). Updated
257 tutorial accordingly.
258 - add AnyEvent::Impl::Irssi backend.
259
2604.86 Mon Jul 20 23:52:29 CEST 2009
261 - since the verbose warning is not enough, explicitly document
262 that versions before 1.33 of Net::SSLeay are not secure.
263 - work around signal handling races in Event and (...) Event::Lib.
264 - try to align signal-race timer to full-second boundaries.
265 - work around Tk not liking negative timeouts.
266 - don't complain of different grades of environmental unfriendlyness
267 in IO::Async.
268
2694.85 Sat Jul 18 06:16:14 CEST 2009
270 - nail the signal race problem in perl once and for all
271 (see $AnyEvent::MAX_SIGNAL_LATENCY).
272 - take advantage of Async::Interrupt if it is available.
273 - load Time::HiRes and Guard modules on demand only.
274 - add optional/recommended modules section to AnyEvent
275 documentation.
276 - reduce memory usage considerably (and reduce startup penalty)
277 by not using "strict", "warnings" and "overload" modules.
278 - work around buggy windows/openbsd perls and provide EBADMSG
279 and EPROTO ourselves when missing.
280 - improve perl 5.6 compatibility of the core event loop.
281 - made Net::SSLeay version 1.33 a soft requirement.
282
2834.83 Fri Jul 17 16:56:26 CEST 2009
284 - implement AnyEvent::Socket::getprotobyname.
285 - AnyEvent::CondVar's will now detect recursive blocking
286 waits and will croak, as too many people fall into
287 this trap.
288 - AnyEvent::Handle will now call ->destroy on itself after
289 executing the on_error callback, instead of doing some
290 half-baked internal shutdown, for fatal errors.
291 - clarify on_eof behaviour w.r.t. the read queue and
292 on_read callbacks.
293 - ignore some possible spurious wake-ups in tcp_connect.
294
2954.82 Sat Jul 11 00:34:55 CEST 2009
296 - POE and Event backends didn't accept some callable objects as
297 callbacks.
298 - use Config module instead of POSIX module to detect signal names
299 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
300 module doesn't even have all POSIX signals :/.
301 - use more workarounds around the many refcnt/corruption bugs in
302 Event::Lib.
303 - work around a race condition in perl's select, causing t/03_child.t
304 to rarely fail.
305
3064.81 Thu Jul 9 10:30:30 CEST 2009
307 - AnyEvent::Handle didn't properly diagnose write errors
308 (it expected -1 from syswrite, how lame... :).
309 - support file descriptors in addition to file handles
310 in AnyEvent->io.
311 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
312 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
313 and PERL_ANYEVENT_CA_PATH.
314 - provide a sensible synopsis section for AnyEvent::TLS.
315 - add a "supported backends" section to the manpage.
316 - added simple io watcher test to testsuite, using a
317 portable_socketpair.
318 - tried to improve the stability of the Event::Lib backend,
319 YMMV.
320
3214.8 Mon Jul 6 23:45:16 CEST 2009
322 - AnyEvent::DNS did not properly follow CNAME records with
323 uppercase targets.
324 - AnyEvent::DNS would errornously return AAAA records
325 with v4 mapped addresses (a faulty record) as ipv4 addresses,
326 causing AnyEvent::Socket to throw an exception.
327 - added new module AnyEvent::TLS for easier SSL/TLS context
328 creation, with many options including hostname verification,
329 secure default configuration, lots of documentation and,
330 predefined diffie-hellman keys for perfect forward security
331 and much more. get it while it's still fresh!
332 - use AnyEvent::TLS in AnyEvent::Handle for context management.
333 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
334 so AnyEvent::Socket users have smaller memory footprint
335 in the common case.
336 - add AnyEvent::Handle->push_shutdown method.
337 - add an additional error message parameter to AnyEvent::Handle's
338 on_error callback (for TLS, $! is still available).
339 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
340 - make AnyEvent::Handle more robust against early conenction
341 failures (during new), and return C<undef> in that case
342 from the constructor.
343 - AnyEvent::Handle will now try to load only JSON::XS first,
344 then fall back to JSON.
345 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
346 and because it was documented that way.
347
3484.451 Fri Jul 3 00:28:58 CEST 2009
349 - do not clear rbuf when shutting down an AnyEvent::Handle
350 object - doing so breaks AnyEvent::HTTP.
351
3524.45 Mon Jun 29 22:59:26 CEST 2009
353 - a write error could cause AnyEvent::Handle to create
354 an I/O watcher with an undefined $fh.
355 - special-case mapped ipv4 addresses in both
356 AnyEvent::Socket::format_address and parse_address,
357 to treat them just like ipv4 addresses.
358 - updated and overhauled the AnyEvent::Intro doc.
359 - implement AnyEvent::Socket::format_ipv4/ipv6.
360 - slightly speed up portable_pipe/socketpair.
361 - expand condvar begin/end documentation.
362
3634.42 Fri Jun 26 08:32:18 CEST 2009
364 - preliminary and neccesarily incomplete support for IO::Async.
365 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
366 case it was set to IGNORE, to ensure we can catch child statuses
367 even when the calling env acts stupidly.
368 - updated benchmarks with IO::Async, which performs very well.
369
3704.412 Wed Jun 24 01:35:57 CEST 2009
371 - support an "untaint" attribute for AnyEvent::DNS
372 and set it on the default resolver.
373 - implement and document AnyEvent::Impl::Perl::loop.
374 - remove all anyevent-env variables from %ENV
375 when running in tainted mode.
376 - mention and extend the IO::Lambda benchmark.
377
3784.411 Sun Jun 7 18:48:02 CEST 2009
379 - do not try to use F_SETFD on windows.
380
3814.41 Thu May 14 06:40:11 CEST 2009
382 - work around issues in older perls (5.8.0?) when
383 a signal handler is deleted from the %SIG hash.
384 - use POSIX::_exit in child test, to avoid running
385 destructors.
386 - speed up CHLD handling by relying on SIGCHLD being
387 synchronously delivered, even when we roll our own
388 implementation.
389 - AnyEvent::DNS: add the "dname" resource record name
390 for cosmetic reasons.
391
3924.4 Sun Apr 26 20:12:33 CEST 2009
393 - implemented idle watchers, where applicable.
394 - AnyEvent->time died when Event backend was in use.
395 - fix a memleak in the Tk backend.
396 - sped up Tk timer handling.
397 - clip negative "after" values to 0 in AnyEvent::Impl::Event
398 to avoid spamming out warning messages.
399 - fix Qt timers without interval.
400 - avoid Qt zero-timeout special case, in old libqt's.
401
4024.352 Mon Apr 20 16:31:11 CEST 2009
403 - fix AnyEvent::Strict error messages for child watchers.
404 - fix/update Makefile.PL "recommends" versions.
405 - add AnyEvent->now_update.
406
4074.351 Sat Apr 11 07:56:14 CEST 2009
408 - actually make the signal pipe work under win32.
409 - localise $! in signal handler to avoid changing $!, although
410 perl itself does not seem to save/restore errno either.
411 - set the cloexec flag on the signal pipe (normally set by perl too).
412
4134.35 Fri Mar 27 11:48:20 CET 2009
414 - event models relying on AnyEvent's signal watcher emulation did
415 invoke the callback asynchronously, contradicting documentation
416 and causing signals to get lost (this includes AnyEvent's own
417 event loop). AnyEvent now uses the standard pipe trick to make
418 callback execution synchronous to the event loop.
419 - AnyEvent::Handle didn't free TLS context data on DESTROY
420 (patch by Pavel Shaydo).
421 - work around the ever-incompatibly-changing API of MakeMaker.
422 - document that changing global variables without restoring them
423 is a bad idea in Perl (noted by Adam Rosenstein).
424 - AnyEvent::Strict now barfs if ->io is passed a file.
425
4264.34 Thu Feb 12 18:32:45 CET 2009
427 - separately buffer TLS read data, as otherwise the read queue could
428 deadlock as receiving data is not expected while draining the
429 read queue (which cna only happen with TLS).
430 - raise EBADMSG error on JSON decoding errors.
431 - fix some minor manpage bugs (reported by Maximilian Gaß).
432 - speed up select bitmask parsing quite a bit in the pure perl
433 backend.
434 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
435
4364.331 Tue Jan 6 21:07:25 CET 2009
437 - socketpair fails on many vista machines because vista has
438 completely broken accept/getpeername and getsockname functions,
439 so we provide our own socketpair emulation that kind of works
440 (AnyEvent::Util::portable_pipe).
441 - new function: AnyEvent::Util::portable_socketpair.
442 - take advantage of the Guard module if it exists.
443
4444.33 Fri Nov 21 02:35:40 CET 2008
445 - AnyEvent::Strict did errornously flag a fileno of 0 as
446 illegal.
447 - reduce memory usage and slightly speed up the pure perl backend
448 by only storing the file descriptor, not the file handle.
449 - add missing autocork method to AnyEvent::Handle (reported
450 by Adam Rosenstein).
451 - AnyEvent::DNS->resolve errornously documented an $rcode result
452 argument, but there isn't (spotted by Henrik Krohns).
453 - the naked truth about Tk - it's basically unmaintained.
454
4554.32 Mon Nov 3 22:46:32 CET 2008
456 - fix AnyEvent::Socket::resolve_sockaddr to properly support
457 unix sockets again and choose a proper default for
458 the socket type.
459 - fix call to resolve_sockaddr in tcp_connect.
460
4614.31 Thu Oct 30 04:41:48 CET 2008
462 - implemented AnyEvent::Handle->destroy method.
463 - hint about unexpected effects in TLS mode.
464 - speed up AnyEvent::Impl::Perl by using more arrays
465 instead of hashes without the slightest loss of readability :)
466 - work around a perl argument refcounting bug.
467 - with some perl patches applied, AnyEvent now seems leak-free.
468
4694.3 Fri Oct 3 09:18:43 CEST 2008
470 - AnyEvent will now install a no-op signal handler for SIGPIPE,
471 unless one has been installed already.
472 - warn about not loaidng AnyEvent::Impl::POE early enough
473 (patch by Adam Rosenstein).
474 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
475 - fix a bug in where in SSL connect mode, the
476 client would first wait for some data by the server and
477 otherwise hang. (reported and analysed in an absolutely
478 exemplary manner by Adam Rosenstein).
479 - fix a bug in where SSL EOF would not be treated
480 as stream EOF, putting the connection into a hung state.
481 (reported and analysed in an absolutely exemplary manner by Adam
482 Rosenstein).
483 - fix a potential segfault when the TLS context
484 would go missing in a read callback (e.g. due to stoptls) -
485 Net::SSLeay of course makes no type checking whatsoever.
486 - AnyEvent::Handle will not stop the read watcher in TLS mode.
487 - AnyEvent::Handle->stoptls will send a TLS close notify
488 instead of simply destroying the stream now.
489 - fix a bug where large blocks of data written in TLS mode
490 would not be sent unless triggered by receives.
491 - on_drain will now take the tls write buffer into account.
492 - SSL operations have been streamlined a bit and should be faster
493 now, more to come, though.
494 - added a FAQ section, will document nontrivial issues in there.
495
4964.234 Mon Sep 29 04:08:13 CEST 2008
497 - fix child watcher documentation: only child exits will be reported,
498 no trace events.
499 - mention SIGPIPE in AnyEvent::Handle.
500 - perl backend will now use POSIX::times as monotonic clock
501 when available and the monotonic clock option is not.
502 - run even without Time::HiRes in some cases now, but do not
503 provide sub-second accuracy (all *supported* perl versions have
504 Time::HiRes).
505
5064.233 Fri Aug 22 01:48:05 CEST 2008
507 - fix a bug in the testsuite, causing a hang.
508 - clarified lots of AnyEvent::Handle constructor arguments.
509
5104.232 Thu Aug 21 20:44:25 CEST 2008
511 - fix a bug in AnyEvent::Handle that could cause two fatal error
512 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
513 (testcase by Leon Brocard).
514 - clarified on_eof/on_error documentation a bit.
515
5164.231 Tue Jul 29 13:12:15 CEST 2008
517 - remove some debugging code left in AnyEvent::Util::fork_call
518 (and no, it's impossible to implement with the broken windows
519 perls without resource leaks or worse).
520
5214.23 Tue Jul 29 12:19:59 CEST 2008
522 - document the first parameter passed to condvar callbacks to be
523 the callback.
524 - add AnyEvent::Socket::{ntoa,aton} aliases.
525 - optimize the AE::Handle->push_read (line) for the default
526 eol marker.
527 - optimize push_read (packstring|storable) for small packets.
528 - invoke on_error callback when no on_eof callback is set.
529 - fix a bug in push_read (storable) of unknown impact.
530
5314.22 Sun Jul 20 16:34:13 CEST 2008
532 - new function AnyEvent::Socket::parse_hostport.
533 - as the bulkheads at microsoft can't even get getprotobyname reliably
534 working on their shitty fucking broken os we need to hardcode
535 some common protocol numbers in AnyEvent::Socket. How can
536 people even bother with such a pile of shit as windows.
537
5384.21 Thu Jul 17 14:40:05 CEST 2008
539 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
540 which is incompatible to previous invocation syntax.
541 - work around more windows perl bugs in fork_call. I cannot imagine
542 why anybody would bother implementing fork in such an obviously
543 unusable way.
544 - avoid calling $do_retry if the request has already finished in
545 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
546 K. Sharifulin).
547
5484.2 Sat Jul 12 22:42:11 CEST 2008
549 - support an "interval" argument for repeating timers.
550 - fix a bug in the pure perl event loop that caused timers to
551 be delayed under some circumstances.
552 - implement optional argument checking via PERL_ANYEVENT_STRICT.
553 - not importing anything from Event causes it to malfunction
554 and not use Time::HiRes, work around this bug.
555 - more bug workarounds against the endlessly broken Tk module,
556 seems to work now on 32 bit machines, but less so on 64 bit.
557 - minor optimisations applied to most backends.
2 558
34.161 Fri Jul 4 14:24:48 CEST 2008 5594.161 Fri Jul 4 14:24:48 CEST 2008
4 - fixed the t/handle/02_write.t, it read after write, with the 560 - fixed the t/handle/02_write.t, it read after write, with the
5 assumption that all data could be written before read 561 assumption that all data could be written before read
6 (reported by Jim Razmus). 562 (reported by Jim Razmus).
… …
159 - support service names in tcp_connect and tcp_server. 715 - support service names in tcp_connect and tcp_server.
160 - work around netbsd/openbsd bug causing unix sockets not to 716 - work around netbsd/openbsd bug causing unix sockets not to
161 become writable until empty. 717 become writable until empty.
162 - only make a dummy read for error detection when getpeername 718 - only make a dummy read for error detection when getpeername
163 returns ENOTCONN. 719 returns ENOTCONN.
164 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 720 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
165 override. 721 override.
166 - new AnyEvent::Guard::cancel. 722 - new AnyEvent::Guard::cancel.
167 723
1683.5 Sat May 17 23:17:14 CEST 2008 7243.5 Sat May 17 23:17:14 CEST 2008
169 - gracefully deal with systems not defining CLOCK_MONOTONIC 725 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
183 739
1843.41 Sun May 11 19:53:13 CEST 2008 7403.41 Sun May 11 19:53:13 CEST 2008
185 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 741 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
186 742
1873.4 Sun May 11 00:29:25 CEST 2008 7433.4 Sun May 11 00:29:25 CEST 2008
188 - complete re-implementation of condvars with a hopefully much mroe useful 744 - complete re-implementation of condvars with a hopefully much more useful
189 API (backward compatible functions exist but are undocumented). 745 API (backward compatible functions exist but are undocumented).
190 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 746 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
191 (still subject to change). 747 (still subject to change).
192 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 748 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
193 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 749 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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