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

Comparing AnyEvent/Changes (file contents):
Revision 1.179 by root, Sun Jul 27 03:28:36 2008 UTC vs.
Revision 1.412 by root, Wed Oct 13 01:15:57 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: check for -w? somehow?
4TODO: document TCP_*** constants
5 - due to a glitch, AnyEvent's internal getprotobyname wasn't
6 actually used in tcp_server.
7 - implement AnyEvent::CondVar->new method, which was documented but
8 missing.
9 - support autoloading of read types in unshift_read.
10 - AnyEvent::Strict: do the test for invalid fh arguments differently,
11 resulting in more informative error messages.
12 - new AnyEvent::FAQ manpage.
13 - clarify that you can actually call push_read etc. in on_read.
14 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
15 requested by Ron Isaacson.
16 - document $AnyEvent::DNS::RESOLVER variable.
17 - incorporated major typo patches by Abhijit Menon-Sen.
18 - AnyEvent::Handle now croaks when negative timeouts are passed.
19 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
20 undocumented, but who knows when they might come in handy.
21
225.271 Tue Jun 8 12:05:46 CEST 2010
23 - backport to perl 5.8.x.
24
255.27 Sun Jun 6 12:12:05 CEST 2010
26 - postpone differently in AnyEvent::Socket now, as
27 when not, canceling the connection attempt might fail
28 (found by Felix Antonius Wilhelm Ostmann).
29 - explicitly check for non-stream sockets in AE::Handle, too many
30 clueless people fell into the trap of this somehow working.
31 - simplified and reworked the "OTHER MODULES" section.
32 - better/more condvar examples.
33
345.261 Wed Apr 28 16:13:36 CEST 2010
35 - AF_INET6 was not properly used from Socket6 during configuration
36 time (found by Andy Grundman).
37
385.26 Mon Apr 12 04:49:35 CEST 2010
39 - don't generate spurious readyness notifications when select
40 returns EINTR in the pure perl backend (analysed and testcase
41 by Bas Denissen).
42 - give same examples in SYNOPSIS sections of AnyEvent and AE.
43 - provider faster implementations for AE::time/now/now_update for
44 EV and Perl backends.
45 - provide faster AE::cv and AE::time implementations for all
46 backends.
47 - no longer support lower/mixed-case signal names (this was
48 never documented nor universally supported).
49 - some more memory µ-optimisations, and somewhat less messy
50 code to cope with both AnyEvent and AE APIs.
51
525.251 Sat Mar 13 00:58:21 CET 2010
53 - make test failed if it was run before make install on perl <=
54 5.8.8, and older versions of perl make our live very difficult,
55 so write constants.pl during Makefile.PL time.
56
575.25 Sat Mar 13 00:23:14 CET 2010
58 - fix a race condition in AnyEvent::Handle that would cause
59 a "bio_read: unsupported method" error (in _another_
60 ssl connection) after user code threw an exception.
61 - added AnyEvent::Handle->destroyed method.
62 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
63 - the AnyEvent::Util::fh_nonblocking function worked only
64 by lucky accident on win32.
65 - smaller and faster AnyEvent::Util::fh_nonblocking.
66 - when the (required!) Time::HiRes module is missing, AnyEvent
67 did not fall back to built-in time properly.
68 - do not load Fcntl at runtime, saving memory and loading time.
69 - precompile a number of constants and use them instead of runtime
70 detection and eval.
71 - free detection code after detection and similar memory optimisations.
72 - Perl backend timer interval best effort drift has been improved
73 (same algorithm as EV).
74 - update unicode idna mapping table.
75
765.24 Tue Jan 5 11:39:43 CET 2010
77 - cygwin never reports errors from failed connects. 1.5 just gives
78 you continous readyness and EAGAIN, 1.7 is even more broken and
79 just hangs. work around both issues in a cygwin-specific hack.
80 - improve idle watcher documentation slightly.
81
825.23 Sun Dec 20 23:48:00 CET 2009
83 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
84 - implement punycode_encode/decode, idn_nameprep,
85 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
86 - provide $AE::VERSION.
87 - removed traces of "no strict 'refs'".
88
895.22 Sat Dec 5 03:51:13 CET 2009
90 - downgrade-or-fail in AnyEvent::Handle::push_write, to
91 diagnose encoding failures earlier and more succinctly.
92 (this works around bugs in perl, throwing away encoding info
93 when passing scalar data to extensions).
94 - add more examples to AnyEvent::Socket manpage.
95 - upgrade internal warning set to the same as common::sense 2.03.
96 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
97 (John Beppu).
98 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
99 - add google public dns servers to fallback server set and make sure
100 we load-balance properly between the three sets. also add all
101 fallback dns servers, not just a random one, to each dns config.
102
1035.21 Thu Nov 19 02:48:47 CET 2009
104 - fix a problem where socket constants were called with parameters
105 (spotted by David Friedland).
106 - fork_call never use'd POSIX (reported by Daisuke Maki).
107 - improve perl 5.6 compatibility further (but it still won't work
108 unless you rip out everything but the core).
109 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
110 uses 10MB of RAM, but doesn't open a console window. *sigh*.
111
1125.202 Wed Oct 14 22:35:44 CEST 2009
113 - AnyEvent::DNS would unexpectedly clobber $_ under windows
114 (analysed by Matthias Waldorf).
115 - AnyEvent::Handle::run_cmd can now pass the PID of the
116 newly-created process, which is much less useful than it might
117 sound (based on patch by Yann Kerherve).
118
1195.201 Tue Sep 29 12:09:25 CEST 2009
120 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
121 (reported by Torsten Foertsch).
122 - common::sense 2.0 could cause tcp_server to throw an exception
123 (analysed by elmex).
124
1255.2 Mon Sep 14 07:04:49 CEST 2009
126 - INCOMPATIBLE CHANGE: do no longer support register_read_type
127 and register_write_type in AnyEvent::Handle, instead support
128 package names (the facility was mostly abused).
129 - implement "packagename-as-read/write type" support in
130 AnyEvent::Handle.
131 - AnyEvent::Handle: new options "keepalive" and "oobinline".
132 - oobinline set by default to avoid security issues.
133 - the pure-perl event loop backend wrongly detected times() fallback
134 support (spotted by Pavel Boldin).
135
1365.12 Tue Sep 1 20:26:50 CEST 2009
137 - be more lenient when parsing resolv.conf files, as some people
138 use hashmarks as comment indicator inside directives (reported
139 by Michael S. Fischer).
140 - use same set of warnings as common::sense 2.0.
141 - fix a potential 32 bit overflow issue due to perl having problems
142 with large hex constants in the Perl backend.
143 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
144 by default.
145 - fix some prototypes.
146 - (optionally) require Async::Interrupt 1.02.
147 - replace opendns fallbacks by verizon/level3 public dns servers.
148
1495.112 Fri Aug 21 13:59:04 CEST 2009
150 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
151 using starttls with an illegal/uncreatable tls context.
152 - port to broken win32 perls.
153
1545.111 Sun Aug 16 18:44:36 CEST 2009
155 - perl errornously requires sockaddr_un structures returned by the
156 kernel to have a specific length. work around this bug by padding
157 sockaddr_un structures (found to fail on OS X, as perl only uses
158 the correct code for GNU/Linux, but likely to fail on other
159 OSes as well).
160
1615.11 Wed Aug 12 17:49:37 CEST 2009
162 - fixed a bug in run_cmd with input from filename.
163 - create a json coder object when none is given in write direction
164 to reduce dependency on JSON.
165
1665.1 Tue Aug 11 03:17:32 CEST 2009
167 - tcp_connect now reports the correct error in $!, not an unrelated
168 one (was broken since 4.91).
169 - AE::Handle did not properly clear rtimeout/wtimeout watchers
170 when retrying a tcp_connect.
171 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
172 - fixed wrong documentation in AE::Socket.
173
1745.01 Mon Aug 10 03:16:32 CEST 2009
175 - last release broke EV child watchers :/.
176
1775.0 Sun Aug 9 17:32:09 CEST 2009
178 - officially introduce and document the AE API.
179 - lowering the timeout at runtime did not have immediate effect
180 in AnyEvent::Handle.
181 - AE::Handle now has separate and independent read and write timeout
182 settings.
183 - probe for Guard module when AnyEvent::Util is loaded, not at
184 runtime (Event doesn't like eval "use" from callbacks, and it avoids
185 exporting the probe function).
186 - Impl::IOAsync's io watchers did not use the filehandle emulation
187 code to work around it's watcher limits.
188 - work around IO::Async's problems with overloaded objects.
189 - try to disable wearnings inside IO::Async, it's just too buggy.
190 - do no longer use Event by default, only EV or Perl, Event is too buggy.
191 - hackishly provide fast AE:: interface when EV or Perl is the backend.
192 - inherit VERSION from AnyEvent::VERSION in most modules.
193
1944.91 Thu Aug 6 15:42:45 CEST 2009
195 - AE::Handle::starttls could get out of sync when the read buffer
196 already contains some TLS handshake.
197 - AE::Handle did not properly free the TLS session, causing bigger
198 memory leaks in Net::SSLeay than expected (reported by toaster).
199 - AE::Socket::tcp_connect will now artificially delay invoking
200 the callback to avoid returning after invoking the callback.
201 - convert many internal watcher uses to AE API.
202
2034.9 Sat Aug 1 11:07:01 CEST 2009
204 - Glib silently fails when registering a timeout with
205 a negative value, so avoid that.
206 - call condvar callback immediately when it is set
207 after the condition is already signalled.
208 - check rbuf_max condition only after trying to consume
209 data.
210 - normalise signal numbers to names when using the ae's
211 signal handling, but do not document this (yet).
212 - pure perl signal emulation did not properly set nonblocking
213 mode on the signal pipe on win32 (this is unlikely to have caused
214 any issues).
215 - new module: AnyEvent::Debug.
216 - AnyEvent::Strict now checks that a signal was specified
217 by name (not name or number).
218 - reduce memory footprint in typical cases by ~50kb by
219 only compiling signal code when necessary.
220 - add AnyEvent::Handle->rbuf_max.
221 - grab the AE namespace for future new API, implement stubs for
222 said future/alternative API.
223 - new function: AnyEvent::Socket::format_hostport.
224
2254.881 Tue Jul 28 12:51:53 CEST 2009
226 - work around a bug in local in pre-5.10 perls, causing
227 AnyEvent::Handle to recurse when it should not
228 (analysed by elmex).
229
2304.88 Tue Jul 28 04:04:37 CEST 2009
231 - re-bless the handle into a dummy package after calling
232 AnyEvent::Handle::destroy, so the user does not need to check
233 for errors after every push_write etc.
234 - do not attempt to run t/02_signals.t on obviously broken
235 platforms.
236
2374.87 Sun Jul 26 02:06:16 CEST 2009
238 - do not attempt to linger when there is no longer a valid fh.
239 - enforce tls mode to be either accept or connect, do not simply
240 segfault in Net::SSLeay.
241 - AnyEvent::Handle can now call tcp_connect itself (new parameters
242 connect, on_prepare, on_connect and on_connect_error). Updated
243 tutorial accordingly.
244 - add AnyEvent::Impl::Irssi backend.
245
2464.86 Mon Jul 20 23:52:29 CEST 2009
247 - since the verbose warning is not enough, explicitly document
248 that versions before 1.33 of Net::SSLeay are not secure.
249 - work around signal handling races in Event and (...) Event::Lib.
250 - try to align signal-race timer to full-second boundaries.
251 - work around Tk not liking negative timeouts.
252 - don't complain of different grades of environmental unfriendlyness
253 in IO::Async.
254
2554.85 Sat Jul 18 06:16:14 CEST 2009
256 - nail the signal race problem in perl once and for all
257 (see $AnyEvent::MAX_SIGNAL_LATENCY).
258 - take advantage of Async::Interrupt if it is available.
259 - load Time::HiRes and Guard modules on demand only.
260 - add optional/recommended modules section to AnyEvent
261 documentation.
262 - reduce memory usage considerably (and reduce startup penalty)
263 by not using "strict", "warnings" and "overload" modules.
264 - work around buggy windows/openbsd perls and provide EBADMSG
265 and EPROTO ourselves when missing.
266 - improve perl 5.6 compatibility of the core event loop.
267 - made Net::SSLeay version 1.33 a soft requirement.
268
2694.83 Fri Jul 17 16:56:26 CEST 2009
270 - implement AnyEvent::Socket::getprotobyname.
271 - AnyEvent::CondVar's will now detect recursive blocking
272 waits and will croak, as too many people fall into
273 this trap.
274 - AnyEvent::Handle will now call ->destroy on itself after
275 executing the on_error callback, instead of doing some
276 half-baked internal shutdown, for fatal errors.
277 - clarify on_eof behaviour w.r.t. the read queue and
278 on_read callbacks.
279 - ignore some possible spurious wake-ups in tcp_connect.
280
2814.82 Sat Jul 11 00:34:55 CEST 2009
282 - POE and Event backends didn't accept some callable objects as
283 callbacks.
284 - use Config module instead of POSIX module to detect signal names
285 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
286 module doesn't even have all POSIX signals :/.
287 - use more workarounds around the many refcnt/corruption bugs in
288 Event::Lib.
289 - work around a race condition in perl's select, causing t/03_child.t
290 to rarely fail.
291
2924.81 Thu Jul 9 10:30:30 CEST 2009
293 - AnyEvent::Handle didn't properly diagnose write errors
294 (it expected -1 from syswrite, how lame... :).
295 - support file descriptors in addition to file handles
296 in AnyEvent->io.
297 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
298 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
299 and PERL_ANYEVENT_CA_PATH.
300 - provide a sensible synopsis section for AnyEvent::TLS.
301 - add a "supported backends" section to the manpage.
302 - added simple io watcher test to testsuite, using a
303 portable_socketpair.
304 - tried to improve the stability of the Event::Lib backend,
305 YMMV.
306
3074.8 Mon Jul 6 23:45:16 CEST 2009
308 - AnyEvent::DNS did not properly follow CNAME records with
309 uppercase targets.
310 - AnyEvent::DNS would errornously return AAAA records
311 with v4 mapped addresses (a faulty record) as ipv4 addresses,
312 causing AnyEvent::Socket to throw an exception.
313 - added new module AnyEvent::TLS for easier SSL/TLS context
314 creation, with many options including hostname verification,
315 secure default configuration, lots of documentation and,
316 predefined diffie-hellman keys for perfect forward security
317 and much more. get it while it's still fresh!
318 - use AnyEvent::TLS in AnyEvent::Handle for context management.
319 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
320 so AnyEvent::Socket users have smaller memory footprint
321 in the common case.
322 - add AnyEvent::Handle->push_shutdown method.
323 - add an additional error message parameter to AnyEvent::Handle's
324 on_error callback (for TLS, $! is still available).
325 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
326 - make AnyEvent::Handle more robust against early conenction
327 failures (during new), and return C<undef> in that case
328 from the constructor.
329 - AnyEvent::Handle will now try to load only JSON::XS first,
330 then fall back to JSON.
331 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
332 and because it was documented that way.
333
3344.451 Fri Jul 3 00:28:58 CEST 2009
335 - do not clear rbuf when shutting down an AnyEvent::Handle
336 object - doing so breaks AnyEvent::HTTP.
337
3384.45 Mon Jun 29 22:59:26 CEST 2009
339 - a write error could cause AnyEvent::Handle to create
340 an I/O watcher with an undefined $fh.
341 - special-case mapped ipv4 addresses in both
342 AnyEvent::Socket::format_address and parse_address,
343 to treat them just like ipv4 addresses.
344 - updated and overhauled the AnyEvent::Intro doc.
345 - implement AnyEvent::Socket::format_ipv4/ipv6.
346 - slightly speed up portable_pipe/socketpair.
347 - expand condvar begin/end documentation.
348
3494.42 Fri Jun 26 08:32:18 CEST 2009
350 - preliminary and neccesarily incomplete support for IO::Async.
351 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
352 case it was set to IGNORE, to ensure we can catch child statuses
353 even when the calling env acts stupidly.
354 - updated benchmarks with IO::Async, which performs very well.
355
3564.412 Wed Jun 24 01:35:57 CEST 2009
357 - support an "untaint" attribute for AnyEvent::DNS
358 and set it on the default resolver.
359 - implement and document AnyEvent::Impl::Perl::loop.
360 - remove all anyevent-env variables from %ENV
361 when running in tainted mode.
362 - mention and extend the IO::Lambda benchmark.
363
3644.411 Sun Jun 7 18:48:02 CEST 2009
365 - do not try to use F_SETFD on windows.
366
3674.41 Thu May 14 06:40:11 CEST 2009
368 - work around issues in older perls (5.8.0?) when
369 a signal handler is deleted from the %SIG hash.
370 - use POSIX::_exit in child test, to avoid running
371 destructors.
372 - speed up CHLD handling by relying on SIGCHLD being
373 synchronously delivered, even when we roll our own
374 implementation.
375 - AnyEvent::DNS: add the "dname" resource record name
376 for cosmetic reasons.
377
3784.4 Sun Apr 26 20:12:33 CEST 2009
379 - implemented idle watchers, where applicable.
380 - AnyEvent->time died when Event backend was in use.
381 - fix a memleak in the Tk backend.
382 - sped up Tk timer handling.
383 - clip negative "after" values to 0 in AnyEvent::Impl::Event
384 to avoid spamming out warning messages.
385 - fix Qt timers without interval.
386 - avoid Qt zero-timeout special case, in old libqt's.
387
3884.352 Mon Apr 20 16:31:11 CEST 2009
389 - fix AnyEvent::Strict error messages for child watchers.
390 - fix/update Makefile.PL "recommends" versions.
391 - add AnyEvent->now_update.
392
3934.351 Sat Apr 11 07:56:14 CEST 2009
394 - actually make the signal pipe work under win32.
395 - localise $! in signal handler to avoid changing $!, although
396 perl itself does not seem to save/restore errno either.
397 - set the cloexec flag on the signal pipe (normally set by perl too).
398
3994.35 Fri Mar 27 11:48:20 CET 2009
400 - event models relying on AnyEvent's signal watcher emulation did
401 invoke the callback asynchronously, contradicting documentation
402 and causing signals to get lost (this includes AnyEvent's own
403 event loop). AnyEvent now uses the standard pipe trick to make
404 callback execution synchronous to the event loop.
405 - AnyEvent::Handle didn't free TLS context data on DESTROY
406 (patch by Pavel Shaydo).
407 - work around the ever-incompatibly-changing API of MakeMaker.
408 - document that changing global variables without restoring them
409 is a bad idea in Perl (noted by Adam Rosenstein).
410 - AnyEvent::Strict now barfs if ->io is passed a file.
411
4124.34 Thu Feb 12 18:32:45 CET 2009
413 - separately buffer TLS read data, as otherwise the read queue could
414 deadlock as receiving data is not expected while draining the
415 read queue (which cna only happen with TLS).
416 - raise EBADMSG error on JSON decoding errors.
417 - fix some minor manpage bugs (reported by Maximilian Gaß).
418 - speed up select bitmask parsing quite a bit in the pure perl
419 backend.
420 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
421
4224.331 Tue Jan 6 21:07:25 CET 2009
423 - socketpair fails on many vista machines because vista has
424 completely broken accept/getpeername and getsockname functions,
425 so we provide our own socketpair emulation that kind of works
426 (AnyEvent::Util::portable_pipe).
427 - new function: AnyEvent::Util::portable_socketpair.
428 - take advantage of the Guard module if it exists.
429
4304.33 Fri Nov 21 02:35:40 CET 2008
431 - AnyEvent::Strict did errornously flag a fileno of 0 as
432 illegal.
433 - reduce memory usage and slightly speed up the pure perl backend
434 by only storing the file descriptor, not the file handle.
435 - add missing autocork method to AnyEvent::Handle (reported
436 by Adam Rosenstein).
437 - AnyEvent::DNS->resolve errornously documented an $rcode result
438 argument, but there isn't (spotted by Henrik Krohns).
439 - the naked truth about Tk - it's basically unmaintained.
440
4414.32 Mon Nov 3 22:46:32 CET 2008
442 - fix AnyEvent::Socket::resolve_sockaddr to properly support
443 unix sockets again and choose a proper default for
444 the socket type.
445 - fix call to resolve_sockaddr in tcp_connect.
446
4474.31 Thu Oct 30 04:41:48 CET 2008
448 - implemented AnyEvent::Handle->destroy method.
449 - hint about unexpected effects in TLS mode.
450 - speed up AnyEvent::Impl::Perl by using more arrays
451 instead of hashes without the slightest loss of readability :)
452 - work around a perl argument refcounting bug.
453 - with some perl patches applied, AnyEvent now seems leak-free.
454
4554.3 Fri Oct 3 09:18:43 CEST 2008
456 - AnyEvent will now install a no-op signal handler for SIGPIPE,
457 unless one has been installed already.
458 - warn about not loaidng AnyEvent::Impl::POE early enough
459 (patch by Adam Rosenstein).
460 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
461 - fix a bug in where in SSL connect mode, the
462 client would first wait for some data by the server and
463 otherwise hang. (reported and analysed in an absolutely
464 exemplary manner by Adam Rosenstein).
465 - fix a bug in where SSL EOF would not be treated
466 as stream EOF, putting the connection into a hung state.
467 (reported and analysed in an absolutely exemplary manner by Adam
468 Rosenstein).
469 - fix a potential segfault when the TLS context
470 would go missing in a read callback (e.g. due to stoptls) -
471 Net::SSLeay of course makes no type checking whatsoever.
472 - AnyEvent::Handle will not stop the read watcher in TLS mode.
473 - AnyEvent::Handle->stoptls will send a TLS close notify
474 instead of simply destroying the stream now.
475 - fix a bug where large blocks of data written in TLS mode
476 would not be sent unless triggered by receives.
477 - on_drain will now take the tls write buffer into account.
478 - SSL operations have been streamlined a bit and should be faster
479 now, more to come, though.
480 - added a FAQ section, will document nontrivial issues in there.
481
4824.234 Mon Sep 29 04:08:13 CEST 2008
483 - fix child watcher documentation: only child exits will be reported,
484 no trace events.
485 - mention SIGPIPE in AnyEvent::Handle.
486 - perl backend will now use POSIX::times as monotonic clock
487 when available and the monotonic clock option is not.
488 - run even without Time::HiRes in some cases now, but do not
489 provide sub-second accuracy (all *supported* perl versions have
490 Time::HiRes).
491
4924.233 Fri Aug 22 01:48:05 CEST 2008
493 - fix a bug in the testsuite, causing a hang.
494 - clarified lots of AnyEvent::Handle constructor arguments.
495
4964.232 Thu Aug 21 20:44:25 CEST 2008
497 - fix a bug in AnyEvent::Handle that could cause two fatal error
498 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
499 (testcase by Leon Brocard).
500 - clarified on_eof/on_error documentation a bit.
501
5024.231 Tue Jul 29 13:12:15 CEST 2008
503 - remove some debugging code left in AnyEvent::Util::fork_call
504 (and no, it's impossible to implement with the broken windows
505 perls without resource leaks or worse).
506
5074.23 Tue Jul 29 12:19:59 CEST 2008
3 - document the first parameter passed to condvar callbacks to be 508 - document the first parameter passed to condvar callbacks to be
4 the callback. 509 the callback.
5 - add AnyEvent::Socket::{ntoa,aton} aliases. 510 - add AnyEvent::Socket::{ntoa,aton} aliases.
6 - optimize the AE::Handle->push_read (line ..) for the default 511 - optimize the AE::Handle->push_read (line) for the default
7 eol marker. 512 eol marker.
513 - optimize push_read (packstring|storable) for small packets.
514 - invoke on_error callback when no on_eof callback is set.
515 - fix a bug in push_read (storable) of unknown impact.
8 516
94.22 Sun Jul 20 16:34:13 CEST 2008 5174.22 Sun Jul 20 16:34:13 CEST 2008
10 - new function AnyEvent::Socket::parse_hostport. 518 - new function AnyEvent::Socket::parse_hostport.
11 - as the bulkheads at microsoft can't even get getprotobyname reliably 519 - as the bulkheads at microsoft can't even get getprotobyname reliably
12 working on their shitty fucking broken os we need to hardcode 520 working on their shitty fucking broken os we need to hardcode
… …
193 - support service names in tcp_connect and tcp_server. 701 - support service names in tcp_connect and tcp_server.
194 - work around netbsd/openbsd bug causing unix sockets not to 702 - work around netbsd/openbsd bug causing unix sockets not to
195 become writable until empty. 703 become writable until empty.
196 - only make a dummy read for error detection when getpeername 704 - only make a dummy read for error detection when getpeername
197 returns ENOTCONN. 705 returns ENOTCONN.
198 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 706 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
199 override. 707 override.
200 - new AnyEvent::Guard::cancel. 708 - new AnyEvent::Guard::cancel.
201 709
2023.5 Sat May 17 23:17:14 CEST 2008 7103.5 Sat May 17 23:17:14 CEST 2008
203 - gracefully deal with systems not defining CLOCK_MONOTONIC 711 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
217 725
2183.41 Sun May 11 19:53:13 CEST 2008 7263.41 Sun May 11 19:53:13 CEST 2008
219 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 727 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
220 728
2213.4 Sun May 11 00:29:25 CEST 2008 7293.4 Sun May 11 00:29:25 CEST 2008
222 - complete re-implementation of condvars with a hopefully much mroe useful 730 - complete re-implementation of condvars with a hopefully much more useful
223 API (backward compatible functions exist but are undocumented). 731 API (backward compatible functions exist but are undocumented).
224 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 732 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
225 (still subject to change). 733 (still subject to change).
226 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 734 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
227 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 735 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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