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

Comparing AnyEvent/Changes (file contents):
Revision 1.115 by root, Mon May 26 17:15:07 2008 UTC vs.
Revision 1.588 by root, Fri Feb 21 16:22:18 2014 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
303:55:22 <Sadrak> Doku zu new: It is possible to push requests on the read and write queues, and modify properties of the stream, even
4 while
503:55:23 <Sadrak> AnyEvent::Handle is connecting.
603:55:39 <Sadrak> doku zu on_connect: It is not allowed to use the read or write queues while the handle object is connecting.
7
8TODO: document TCP_*** constants
9TODO: add debug/trace logging to some modules
10TODO: use the eval-with-cb-call trick also to speed up JSON?
11TODO: FAQ: common pitfalls? invoke-before-return
12TODO: FAQ: scope-ids? hosts file order?
13TODO: AnyEvent::Handle with AnyEvent::IO?
14TODO: invalid. and localhost. specialcasing inside AEDNS and not AESocket (rfc6761)
15TODO: maybe implement env variable to give hosts precedence
16
17 - work around a newly introduced bug in Socket 2.011 (an
18 errornous sun_length check) (analyzed by Maxime Soulé).
19
207.07 Tue Dec 17 17:45:02 CET 2013
21 - the documentation for custom tls verify schemes was wrong. make it agree
22 with the code (reported by Maxime Soulé).
23 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
24 - work around an API change in openssl that could cause wrong tls connection
25 aborts, likely on windows only (analyzed by sten).
26 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
27 endless loop (reported by Dietrich Rebmann).
28 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
29 (patch by Maxime Soulé).
30 - document AnyEvent::Impl::IOAsync::set_loop and
31 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
32 functionality has _always_ been available.
33 - force a toplevel domain name in t/81_hosts.t (analyzed by
34 David Jack Wange Olrik).
35 - document that AnyEvent::Log uses AnyEvent::IO.
36 - warn about AnyEvent::Filesys::Notify performance.
37 - praise the joys of AnyEvent::Fork::*.
38 - time for an =encoding directive.
39 - do no longer use JSON to create a default json coder, use
40 JSON::XS or JSON::PP directly.
41
427.05 Wed Aug 21 10:38:08 CEST 2013
43 - uts46data.pl couldn't be found due to wrong naming of the file
44 (reported by Fulko Hew).
45 - handle lone \015's properly in AE::Handle's default line read
46 (reported by various people).
47 - untaint ip addresses found in /etc/hosts (patch by José Micó).
48 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
49 fail on first use (reported and testcase by Andrew Whatson).
50 - expose AnyEvent::Log::format_time, and allow users to redefine it.
51 - expose AnyEvent::Log::default_format, and allow redefinition.
52 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
53 - AnyEvent::Debug shell can now run coro shell commands, if available.
54 - t/63* tests were wrongly in MANIFEST.
55 - kernel.org's finger server went MIA, switch to freebsd.org and
56 icculus.org.
57 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
58 to function asynchronously (reported by Daniel Carosone).
59 - hard-disable $^W in most tests, it generates too much garbage output.
60 - use a (hopefully) more future-proof method to emulate common::sense.
61 - upgrade to UTS-46:6.2.0.
62 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
63 documented. should not affect anything.
64
657.04 Thu Nov 15 02:23:58 CET 2012
66 - debugging code left in.
67
687.03 Thu Nov 15 02:16:11 CET 2012
69 - AnyEvent::Socket::inet_aton did not work when dns resolution was
70 used to find the addresses (analyzed and patch by Maxime Soulé).
71 - fix a memory leak in the /etc/hosts lookup code when hosts don't
72 resolve and are not in hosts (analyzed by sten).
73
747.02 Tue Aug 14 04:27:58 CEST 2012
75 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
76 program by Yorhel).
77 - verified that AnyEvent::Socket follows RFC5952.
78 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
79
807.01 Sun May 13 01:03:17 CEST 2012
81 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
82 not available, instead of throwing an exception.
83 - use File::Spec to get the tmpdir in t/*, to avoid needless
84 failures on (most, not mine :) windows boxes.
85 - new handle read types: tls_detect and tls_autostart.
86
877.0 Fri Apr 13 06:33:30 CEST 2012
88 - child watchers are broken in POE 1.352 (also many earlier
89 versions) and there seems to be no way to work around it, as POE
90 itself is inherently racy. Document this regression and add a delay
91 in t/68_poe_03_child.t for the time being.
92 - new module AnyEvent::IO, that is a frontend to either a pure-perl
93 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
94 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
95 - load /etc/hosts only when DNS has no answer.
96 - stat /etc/hosts on every access and reload it if it changed.
97 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
98 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
99 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
100 a side effect, they now use true appending as opposed to libc
101 appending, and the file might not have been opened when the function
102 returns.
103 - the default logging level was not properly documented in a variety of
104 places, this has hopefully been rectified.
105 - updated uts46data.pl for unicode 6.1.0.
106 - made log messages generated by AnyEvent submodules not
107 include the package name anymore, as it will be logged
108 by default already.
109 - upgrade to the trick used by common-sense 3.5 to work
110 around extra warning torture/breakage under perl 5.15.x.
111 - log messages by anyevent are now ucfirst, are usually full
112 sentences and do no longer include the package name.
113 - the storable read type would sometimes throw an exception instead
114 of causing EBADMSG (analyzed by Maxime Soulé).
115 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
116 keysize (reported by Paul Howarth).
117 - documented Rocco fabricating statements about AnyEvent and me.
118
1196.14 Tue Jan 31 20:00:24 CET 2012
120 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
121 probably an editing mistake (reported by Darin McBride).
122
1236.13 Thu Jan 12 07:27:01 CET 2012
124 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
125 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
126 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
127 regex-matching inside the callback would change the parameter, despite
128 $1 being dynamically scoped per-block. this perl bug is now being
129 worked around (testcase by Cindy Wang).
130
1316.12 Mon Dec 12 13:21:10 CET 2011
132 - $! was clobbered by subsequent calls in tcp_connect, due to
133 the postpone not saving/restoring it (analyzed by Richard Garnish).
134
1356.11 Tue Nov 22 10:36:05 CET 2011
136 - Tk cannot create windows when tainted unless you set a title, so
137 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
138 - escape any nonprintable/nonascii characters when stringifying
139 backtraces.
140 - log the reason loading an interface module fails to load at level debug.
141
1426.1 Tue Oct 4 19:44:30 CEST 2011
143 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
144 and above), and some messages inside AnyEvent have been elevated
145 to higher log levels to print by default.
146 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
147 - work around yet another signal race bug, newly introduced by POE. My
148 how that module sucks.
149 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
150 support ipv6-only operation even with defaults (also use ::1 instead
151 of 127.0.0.1).
152 - give better guidelines on how to chose a logging level.
153 - AnyEvent::Log can now cap (limit) log message priority by context.
154 - AE::log will now save and restore $! and $@, so callers don't have to.
155 - AE::Strict now checks that registered fds are still valid at regular
156 intervals.
157 - skip frequently called 'require' calls if the module appears loaded.
158 - increase signal latency test timer considerably for poor little osx,
159 which can't be bothered with basic posix functionality such as signal
160 delivery.
161 - mention some debugging aids in the tutorial.
162 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
163 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
164 of the context as title.
165 - AnyEvent::Strict would not restore read-write access to $_ when a callback
166 died - this is a user bug, but since AE::Strict exists to help dbeugging...
167 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
168 to work with it - the author is well aware of the issues but refuses
169 to fix them. This extreme step has been taken because the module actively
170 attacks the goals of AnyEvent and makes it harder for module authors to use
171 AnyEvent.
172
1736.02 Fri Aug 26 20:08:31 CEST 2011
174 - log_to_syslog now accepts facility strings.
175 - log_to_syslog would not allow facilities - "$facility|$level"
176 *literally* means that string. sheesh.
177 - the CPAN parser got confused and thought we provide Sys::Syslog, try
178 to work around.
179 - renamed AnyEvent::Impl::FLTK2 to FLTK.
180
1816.01 Fri Aug 26 07:04:11 CEST 2011
182 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
183 the TTL as fourth element - this affects ->resolve, ->request
184 and ->dns_unpack indirectly. Convenience functions such as
185 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
186 AnyEvent::Socket.
187 - INCOMPATIBLE CHANGE: backend authors now should not implement
188 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
189 methods.
190 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
191 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
192 - DEPRECIATION: please change
193 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
194 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
195 in your programs, for future compatibility.
196 - *blush* idle watcher emulation was borked.
197 - AnyEvent::Strict now checks the AE::xxx functions as well.
198 - querying too long domain names would barf deep inside AnyEvent::DNS - since
199 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
200 for too long domain names explicitly and passes an undef to the callback
201 instead (reported by James Bromberger).
202 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
203 - new functions: AnyEvent::postpone, AnyEvent::log.
204 - new module: AnyEvent::Log, for simple logging needs.
205 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
206 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
207 a debug shell.
208 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
209 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
210 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
211 /etc/hosts can be found.
212 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
213 configuration.
214 - env variables can now be specified using AE_xyz instead of
215 PERL_ANYEVENT_xyz, subject to some rules.
216 - lots of new AnyEvent::Debug functionality.
217 - AnyEvent::Strict now checks for modifications of $_ by the callback
218 (by setting it to readonly while executing the callback, if possible).
219 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
220 Invest extra overhead to work around this peculiar behaviour.
221 - split the pure perl backend into a loop module and an impl
222 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
223 - some read types (regex, netstring, json, storable) could
224 errornously keep an AE::Handle object alive on their own
225 (reported by Mohammad Toossi).
226 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
227 supports /etc/hosts overrides.
228 - add more workarounds around child watcher bugs in IO::Async
229 and (newly introduced) in POE.
230 - use glib child watchers and try to work around its limitations,
231 if possible (based on a patch by Kevin Ryde).
232 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
233 - many read types in AnyEvent::Handle would malfunction during
234 transitions to TLS because of a "delete $self->{rbuf}".
235 (reported by Gerald Galster).
236 - improve condvar blocking wait performance (probably only noticable
237 with faster backends), also make it easier for event loops that
238 do not support blocking waits.
239 - PERL_ANYEVENT_MODEL now also allows full module names, although
240 the usefulness of this feature is not yet clear.
241 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
242 - add t/80_ssltext.t.
243 - no longer use AUTOLOAD for the initial loading (for no very good reason).
244 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
245 - do not run event loop detection multiple times when programs call their cached
246 copies of AnyEvent::detect.
247 - parse_hostport now accepts absolute paths als unix domain sockets.
248 - suppress the idiotic warning inside POE by patching the POE::Kernel
249 run flags instead of asking the user to jump through contortions.
250 - add Task::Weaken as dependency (as recommends only at this time).
251 - replace PF_UNSPEC by 0, for good old minix.
252
2536.0 (not officially released, but was on CPAN for about 15 minutes :/)
254
2555.34 Thu May 12 09:59:41 CEST 2011
256 - AE::TLS didn't support both cert_file and key at the same time,
257 due to some copy&paste bug (reported and analyzed by vti).
258 - improve AE::Handle timeout documentation slightly.
259
2605.33 Mon Feb 7 21:51:41 CET 2011
261 - never appeared on CPAN, for unknown reasons.
262 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
263 object when it's already gone (analyzed by Ryan Bullock).
264 - update idna algorithm(s) to UTS#46 version 6.0.0.
265 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
266 uts46 says it's an error, and testsuite makes more sense.
267 - remove leading dots in idn_to_ascii - testsuite says to remove,
268 uts46 says it's an error, and testsuite makes more sense.
269 - fixed a bug in idn ignored character processing.
270 - passes the uts46 6.0.0 testsuite except for three cases which
271 apparently are bugs in the testsuite itself.
272 - passes the uts46 6.0.1 testsuite except for eleven cases which
273 apparently are bugs in the testsuite itself.
274
2755.31 Sun Jan 23 11:49:19 CET 2011
276 - work around a POE bug where POE would first poll for new events and
277 invoke handlers BEFORE handling the message queue.
278 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
279 - stop_read now does just that in TLS mode. see its
280 documentation for caveats.
281 - update to common::sense 3.4 warning set (some warnings in perl impose
282 an arbitrary 32kb limit on regex matches, which affects
283 AnyEvent::Handle).
284 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
285 - new experimental AE::Handle::resettls method.
286
2875.3 Fri Dec 31 05:48:13 CET 2010
288 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
289 issues with IO::Async have been solved in that version, including
290 the availabilty of a default loop!
291 - implement new wbuf_max parameter for AnyEvent::Handle.
292 - added Cocoa::EventLoop backend by Daisuke Murase.
293 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
294 - clarify handle on_prepare documentation.
295 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
296 to avoid leaking them to any child processes.
297 - try to force files into the arch-dependend lib directory, for the
298 half-"turly obsessive compulsive person"(s) that depend on it.
299
3005.29 Sun Dec 5 10:49:21 CET 2010
301 - convert EV backend to EV 4.00 API (so better upgrade EV too).
302 - AE::Handle: implement dynamic read_size adjustment, add
303 max_read_size parameter, reduce default read_size to 2048.
304 - add some FAQ entries for servers and/or on_eof vs. on_error.
305 - work around OS bugs (cygwin again) when lingering in AE::Handle.
306 - reduce memory usage slightly when connect was used in AE::Handle.
307 - wrok around more segfaults in Tk.
308 - document tls_ctx => undef as valid.
309 - detect WNOHANG value at compiletime.
310 - include some probably linux-only support for building AnyEvent
311 as part of the perl core.
312 - improve compatibility of t/08_idna.t to perl 5.8.
313 - make AnyEvent::Strict truly optional (in case somebody builds a
314 minimal perl...).
315
3165.28 Wed Oct 13 04:14:23 CEST 2010
317 - due to a glitch, AnyEvent's internal getprotobyname wasn't
318 actually used in tcp_server.
319 - implement AnyEvent::CondVar->new method, which was documented but
320 missing.
321 - support autoloading of read types in unshift_read.
322 - AnyEvent::Strict: do the test for invalid fh arguments differently,
323 resulting in more informative error messages.
324 - new AnyEvent::FAQ manpage.
325 - clarify that you can actually call push_read etc. in on_read.
326 - put keys and certs in /tmp, not into the current directory,
327 and document this in the SECURITY CONSIDERATIONS sections
328 of AnyEvent::TLS.
329 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
330 requested by Ron Isaacson.
331 - document $AnyEvent::DNS::RESOLVER variable.
332 - incorporated major typo patches by Abhijit Menon-Sen.
333 - AnyEvent::Handle now croaks when negative timeouts are passed.
334 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
335 undocumented, but who knows when they might come in handy.
336
3375.271 Tue Jun 8 12:05:46 CEST 2010
338 - backport to perl 5.8.x.
339
3405.27 Sun Jun 6 12:12:05 CEST 2010
341 - postpone differently in AnyEvent::Socket now, as
342 when not, canceling the connection attempt might fail
343 (found by Felix Antonius Wilhelm Ostmann).
344 - explicitly check for non-stream sockets in AE::Handle, too many
345 clueless people fell into the trap of this somehow working.
346 - simplified and reworked the "OTHER MODULES" section.
347 - better/more condvar examples.
348
3495.261 Wed Apr 28 16:13:36 CEST 2010
350 - AF_INET6 was not properly used from Socket6 during configuration
351 time (found by Andy Grundman).
352
3535.26 Mon Apr 12 04:49:35 CEST 2010
354 - don't generate spurious readyness notifications when select
355 returns EINTR in the pure perl backend (analysed and testcase
356 by Bas Denissen).
357 - give same examples in SYNOPSIS sections of AnyEvent and AE.
358 - provider faster implementations for AE::time/now/now_update for
359 EV and Perl backends.
360 - provide faster AE::cv and AE::time implementations for all
361 backends.
362 - no longer support lower/mixed-case signal names (this was
363 never documented nor universally supported).
364 - some more memory µ-optimisations, and somewhat less messy
365 code to cope with both AnyEvent and AE APIs.
366
3675.251 Sat Mar 13 00:58:21 CET 2010
368 - make test failed if it was run before make install on perl <=
369 5.8.8, and older versions of perl make our live very difficult,
370 so write constants.pl during Makefile.PL time.
371
3725.25 Sat Mar 13 00:23:14 CET 2010
373 - fix a race condition in AnyEvent::Handle that would cause
374 a "bio_read: unsupported method" error (in _another_
375 ssl connection) after user code threw an exception.
376 - added AnyEvent::Handle->destroyed method.
377 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
378 - the AnyEvent::Util::fh_nonblocking function worked only
379 by lucky accident on win32.
380 - smaller and faster AnyEvent::Util::fh_nonblocking.
381 - when the (required!) Time::HiRes module is missing, AnyEvent
382 did not fall back to built-in time properly.
383 - do not load Fcntl at runtime, saving memory and loading time.
384 - precompile a number of constants and use them instead of runtime
385 detection and eval.
386 - free detection code after detection and similar memory optimisations.
387 - Perl backend timer interval best effort drift has been improved
388 (same algorithm as EV).
389 - update unicode idna mapping table.
390
3915.24 Tue Jan 5 11:39:43 CET 2010
392 - cygwin never reports errors from failed connects. 1.5 just gives
393 you continous readyness and EAGAIN, 1.7 is even more broken and
394 just hangs. work around both issues in a cygwin-specific hack.
395 - improve idle watcher documentation slightly.
396
3975.23 Sun Dec 20 23:48:00 CET 2009
398 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
399 - implement punycode_encode/decode, idn_nameprep,
400 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
401 - provide $AE::VERSION.
402 - removed traces of "no strict 'refs'".
403
4045.22 Sat Dec 5 03:51:13 CET 2009
405 - downgrade-or-fail in AnyEvent::Handle::push_write, to
406 diagnose encoding failures earlier and more succinctly.
407 (this works around bugs in perl, throwing away encoding info
408 when passing scalar data to extensions).
409 - add more examples to AnyEvent::Socket manpage.
410 - upgrade internal warning set to the same as common::sense 2.03.
411 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
412 (John Beppu).
413 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
414 - add google public dns servers to fallback server set and make sure
415 we load-balance properly between the three sets. also add all
416 fallback dns servers, not just a random one, to each dns config.
417
4185.21 Thu Nov 19 02:48:47 CET 2009
419 - fix a problem where socket constants were called with parameters
420 (spotted by David Friedland).
421 - fork_call never use'd POSIX (reported by Daisuke Maki).
422 - improve perl 5.6 compatibility further (but it still won't work
423 unless you rip out everything but the core).
424 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
425 uses 10MB of RAM, but doesn't open a console window. *sigh*.
426
4275.202 Wed Oct 14 22:35:44 CEST 2009
428 - AnyEvent::DNS would unexpectedly clobber $_ under windows
429 (analysed by Matthias Waldorf).
430 - AnyEvent::Handle::run_cmd can now pass the PID of the
431 newly-created process, which is much less useful than it might
432 sound (based on patch by Yann Kerherve).
433
4345.201 Tue Sep 29 12:09:25 CEST 2009
435 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
436 (reported by Torsten Foertsch).
437 - common::sense 2.0 could cause tcp_server to throw an exception
438 (analysed by elmex).
439
4405.2 Mon Sep 14 07:04:49 CEST 2009
441 - INCOMPATIBLE CHANGE: do no longer support register_read_type
442 and register_write_type in AnyEvent::Handle, instead support
443 package names (the facility was mostly abused).
444 - implement "packagename-as-read/write type" support in
445 AnyEvent::Handle.
446 - AnyEvent::Handle: new options "keepalive" and "oobinline".
447 - oobinline set by default to avoid security issues.
448 - the pure-perl event loop backend wrongly detected times() fallback
449 support (spotted by Pavel Boldin).
450
4515.12 Tue Sep 1 20:26:50 CEST 2009
452 - be more lenient when parsing resolv.conf files, as some people
453 use hashmarks as comment indicator inside directives (reported
454 by Michael S. Fischer).
455 - use same set of warnings as common::sense 2.0.
456 - fix a potential 32 bit overflow issue due to perl having problems
457 with large hex constants in the Perl backend.
458 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
459 by default.
460 - fix some prototypes.
461 - (optionally) require Async::Interrupt 1.02.
462 - replace opendns fallbacks by verizon/level3 public dns servers.
463
4645.112 Fri Aug 21 13:59:04 CEST 2009
465 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
466 using starttls with an illegal/uncreatable tls context.
467 - port to broken win32 perls.
468
4695.111 Sun Aug 16 18:44:36 CEST 2009
470 - perl errornously requires sockaddr_un structures returned by the
471 kernel to have a specific length. work around this bug by padding
472 sockaddr_un structures (found to fail on OS X, as perl only uses
473 the correct code for GNU/Linux, but likely to fail on other
474 OSes as well).
475
4765.11 Wed Aug 12 17:49:37 CEST 2009
477 - fixed a bug in run_cmd with input from filename.
478 - create a json coder object when none is given in write direction
479 to reduce dependency on JSON.
480
4815.1 Tue Aug 11 03:17:32 CEST 2009
482 - tcp_connect now reports the correct error in $!, not an unrelated
483 one (was broken since 4.91).
484 - AE::Handle did not properly clear rtimeout/wtimeout watchers
485 when retrying a tcp_connect.
486 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
487 - fixed wrong documentation in AE::Socket.
488
4895.01 Mon Aug 10 03:16:32 CEST 2009
490 - last release broke EV child watchers :/.
491
4925.0 Sun Aug 9 17:32:09 CEST 2009
493 - officially introduce and document the AE API.
494 - lowering the timeout at runtime did not have immediate effect
495 in AnyEvent::Handle.
496 - AE::Handle now has separate and independent read and write timeout
497 settings.
498 - probe for Guard module when AnyEvent::Util is loaded, not at
499 runtime (Event doesn't like eval "use" from callbacks, and it avoids
500 exporting the probe function).
501 - Impl::IOAsync's io watchers did not use the filehandle emulation
502 code to work around it's watcher limits.
503 - work around IO::Async's problems with overloaded objects.
504 - try to disable wearnings inside IO::Async, it's just too buggy.
505 - do no longer use Event by default, only EV or Perl, Event is too buggy.
506 - hackishly provide fast AE:: interface when EV or Perl is the backend.
507 - inherit VERSION from AnyEvent::VERSION in most modules.
508
5094.91 Thu Aug 6 15:42:45 CEST 2009
510 - AE::Handle::starttls could get out of sync when the read buffer
511 already contains some TLS handshake.
512 - AE::Handle did not properly free the TLS session, causing bigger
513 memory leaks in Net::SSLeay than expected (reported by toaster).
514 - AE::Socket::tcp_connect will now artificially delay invoking
515 the callback to avoid returning after invoking the callback.
516 - convert many internal watcher uses to AE API.
517
5184.9 Sat Aug 1 11:07:01 CEST 2009
519 - Glib silently fails when registering a timeout with
520 a negative value, so avoid that.
521 - call condvar callback immediately when it is set
522 after the condition is already signalled.
523 - check rbuf_max condition only after trying to consume
524 data.
525 - normalise signal numbers to names when using the ae's
526 signal handling, but do not document this (yet).
527 - pure perl signal emulation did not properly set nonblocking
528 mode on the signal pipe on win32 (this is unlikely to have caused
529 any issues).
530 - new module: AnyEvent::Debug.
531 - AnyEvent::Strict now checks that a signal was specified
532 by name (not name or number).
533 - reduce memory footprint in typical cases by ~50kb by
534 only compiling signal code when necessary.
535 - add AnyEvent::Handle->rbuf_max.
536 - grab the AE namespace for future new API, implement stubs for
537 said future/alternative API.
538 - new function: AnyEvent::Socket::format_hostport.
539
5404.881 Tue Jul 28 12:51:53 CEST 2009
541 - work around a bug in local in pre-5.10 perls, causing
542 AnyEvent::Handle to recurse when it should not
543 (analysed by elmex).
544
5454.88 Tue Jul 28 04:04:37 CEST 2009
546 - re-bless the handle into a dummy package after calling
547 AnyEvent::Handle::destroy, so the user does not need to check
548 for errors after every push_write etc.
549 - do not attempt to run t/02_signals.t on obviously broken
550 platforms.
551
5524.87 Sun Jul 26 02:06:16 CEST 2009
553 - do not attempt to linger when there is no longer a valid fh.
554 - enforce tls mode to be either accept or connect, do not simply
555 segfault in Net::SSLeay.
556 - AnyEvent::Handle can now call tcp_connect itself (new parameters
557 connect, on_prepare, on_connect and on_connect_error). Updated
558 tutorial accordingly.
559 - add AnyEvent::Impl::Irssi backend.
560
5614.86 Mon Jul 20 23:52:29 CEST 2009
562 - since the verbose warning is not enough, explicitly document
563 that versions before 1.33 of Net::SSLeay are not secure.
564 - work around signal handling races in Event and (...) Event::Lib.
565 - try to align signal-race timer to full-second boundaries.
566 - work around Tk not liking negative timeouts.
567 - don't complain of different grades of environmental unfriendlyness
568 in IO::Async.
569
5704.85 Sat Jul 18 06:16:14 CEST 2009
571 - nail the signal race problem in perl once and for all
572 (see $AnyEvent::MAX_SIGNAL_LATENCY).
573 - take advantage of Async::Interrupt if it is available.
574 - load Time::HiRes and Guard modules on demand only.
575 - add optional/recommended modules section to AnyEvent
576 documentation.
577 - reduce memory usage considerably (and reduce startup penalty)
578 by not using "strict", "warnings" and "overload" modules.
579 - work around buggy windows/openbsd perls and provide EBADMSG
580 and EPROTO ourselves when missing.
581 - improve perl 5.6 compatibility of the core event loop.
582 - made Net::SSLeay version 1.33 a soft requirement.
583
5844.83 Fri Jul 17 16:56:26 CEST 2009
585 - implement AnyEvent::Socket::getprotobyname.
586 - AnyEvent::CondVar's will now detect recursive blocking
587 waits and will croak, as too many people fall into
588 this trap.
589 - AnyEvent::Handle will now call ->destroy on itself after
590 executing the on_error callback, instead of doing some
591 half-baked internal shutdown, for fatal errors.
592 - clarify on_eof behaviour w.r.t. the read queue and
593 on_read callbacks.
594 - ignore some possible spurious wake-ups in tcp_connect.
595
5964.82 Sat Jul 11 00:34:55 CEST 2009
597 - POE and Event backends didn't accept some callable objects as
598 callbacks.
599 - use Config module instead of POSIX module to detect signal names
600 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
601 module doesn't even have all POSIX signals :/.
602 - use more workarounds around the many refcnt/corruption bugs in
603 Event::Lib.
604 - work around a race condition in perl's select, causing t/03_child.t
605 to rarely fail.
606
6074.81 Thu Jul 9 10:30:30 CEST 2009
608 - AnyEvent::Handle didn't properly diagnose write errors
609 (it expected -1 from syswrite, how lame... :).
610 - support file descriptors in addition to file handles
611 in AnyEvent->io.
612 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
613 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
614 and PERL_ANYEVENT_CA_PATH.
615 - provide a sensible synopsis section for AnyEvent::TLS.
616 - add a "supported backends" section to the manpage.
617 - added simple io watcher test to testsuite, using a
618 portable_socketpair.
619 - tried to improve the stability of the Event::Lib backend,
620 YMMV.
621
6224.8 Mon Jul 6 23:45:16 CEST 2009
623 - AnyEvent::DNS did not properly follow CNAME records with
624 uppercase targets.
625 - AnyEvent::DNS would errornously return AAAA records
626 with v4 mapped addresses (a faulty record) as ipv4 addresses,
627 causing AnyEvent::Socket to throw an exception.
628 - added new module AnyEvent::TLS for easier SSL/TLS context
629 creation, with many options including hostname verification,
630 secure default configuration, lots of documentation and,
631 predefined diffie-hellman keys for perfect forward security
632 and much more. get it while it's still fresh!
633 - use AnyEvent::TLS in AnyEvent::Handle for context management.
634 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
635 so AnyEvent::Socket users have smaller memory footprint
636 in the common case.
637 - add AnyEvent::Handle->push_shutdown method.
638 - add an additional error message parameter to AnyEvent::Handle's
639 on_error callback (for TLS, $! is still available).
640 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
641 - make AnyEvent::Handle more robust against early conenction
642 failures (during new), and return C<undef> in that case
643 from the constructor.
644 - AnyEvent::Handle will now try to load only JSON::XS first,
645 then fall back to JSON.
646 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
647 and because it was documented that way.
648
6494.451 Fri Jul 3 00:28:58 CEST 2009
650 - do not clear rbuf when shutting down an AnyEvent::Handle
651 object - doing so breaks AnyEvent::HTTP.
652
6534.45 Mon Jun 29 22:59:26 CEST 2009
654 - a write error could cause AnyEvent::Handle to create
655 an I/O watcher with an undefined $fh.
656 - special-case mapped ipv4 addresses in both
657 AnyEvent::Socket::format_address and parse_address,
658 to treat them just like ipv4 addresses.
659 - updated and overhauled the AnyEvent::Intro doc.
660 - implement AnyEvent::Socket::format_ipv4/ipv6.
661 - slightly speed up portable_pipe/socketpair.
662 - expand condvar begin/end documentation.
663
6644.42 Fri Jun 26 08:32:18 CEST 2009
665 - preliminary and neccesarily incomplete support for IO::Async.
666 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
667 case it was set to IGNORE, to ensure we can catch child statuses
668 even when the calling env acts stupidly.
669 - updated benchmarks with IO::Async, which performs very well.
670
6714.412 Wed Jun 24 01:35:57 CEST 2009
672 - support an "untaint" attribute for AnyEvent::DNS
673 and set it on the default resolver.
674 - implement and document AnyEvent::Impl::Perl::loop.
675 - remove all anyevent-env variables from %ENV
676 when running in tainted mode.
677 - mention and extend the IO::Lambda benchmark.
678
6794.411 Sun Jun 7 18:48:02 CEST 2009
680 - do not try to use F_SETFD on windows.
681
6824.41 Thu May 14 06:40:11 CEST 2009
683 - work around issues in older perls (5.8.0?) when
684 a signal handler is deleted from the %SIG hash.
685 - use POSIX::_exit in child test, to avoid running
686 destructors.
687 - speed up CHLD handling by relying on SIGCHLD being
688 synchronously delivered, even when we roll our own
689 implementation.
690 - AnyEvent::DNS: add the "dname" resource record name
691 for cosmetic reasons.
692
6934.4 Sun Apr 26 20:12:33 CEST 2009
694 - implemented idle watchers, where applicable.
695 - AnyEvent->time died when Event backend was in use.
696 - fix a memleak in the Tk backend.
697 - sped up Tk timer handling.
698 - clip negative "after" values to 0 in AnyEvent::Impl::Event
699 to avoid spamming out warning messages.
700 - fix Qt timers without interval.
701 - avoid Qt zero-timeout special case, in old libqt's.
702
7034.352 Mon Apr 20 16:31:11 CEST 2009
704 - fix AnyEvent::Strict error messages for child watchers.
705 - fix/update Makefile.PL "recommends" versions.
706 - add AnyEvent->now_update.
707
7084.351 Sat Apr 11 07:56:14 CEST 2009
709 - actually make the signal pipe work under win32.
710 - localise $! in signal handler to avoid changing $!, although
711 perl itself does not seem to save/restore errno either.
712 - set the cloexec flag on the signal pipe (normally set by perl too).
713
7144.35 Fri Mar 27 11:48:20 CET 2009
715 - event models relying on AnyEvent's signal watcher emulation did
716 invoke the callback asynchronously, contradicting documentation
717 and causing signals to get lost (this includes AnyEvent's own
718 event loop). AnyEvent now uses the standard pipe trick to make
719 callback execution synchronous to the event loop.
720 - AnyEvent::Handle didn't free TLS context data on DESTROY
721 (patch by Pavel Shaydo).
722 - work around the ever-incompatibly-changing API of MakeMaker.
723 - document that changing global variables without restoring them
724 is a bad idea in Perl (noted by Adam Rosenstein).
725 - AnyEvent::Strict now barfs if ->io is passed a file.
726
7274.34 Thu Feb 12 18:32:45 CET 2009
728 - separately buffer TLS read data, as otherwise the read queue could
729 deadlock as receiving data is not expected while draining the
730 read queue (which cna only happen with TLS).
731 - raise EBADMSG error on JSON decoding errors.
732 - fix some minor manpage bugs (reported by Maximilian Gaß).
733 - speed up select bitmask parsing quite a bit in the pure perl
734 backend.
735 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
736
7374.331 Tue Jan 6 21:07:25 CET 2009
738 - socketpair fails on many vista machines because vista has
739 completely broken accept/getpeername and getsockname functions,
740 so we provide our own socketpair emulation that kind of works
741 (AnyEvent::Util::portable_pipe).
742 - new function: AnyEvent::Util::portable_socketpair.
743 - take advantage of the Guard module if it exists.
744
7454.33 Fri Nov 21 02:35:40 CET 2008
746 - AnyEvent::Strict did errornously flag a fileno of 0 as
747 illegal.
748 - reduce memory usage and slightly speed up the pure perl backend
749 by only storing the file descriptor, not the file handle.
750 - add missing autocork method to AnyEvent::Handle (reported
751 by Adam Rosenstein).
752 - AnyEvent::DNS->resolve errornously documented an $rcode result
753 argument, but there isn't (spotted by Henrik Krohns).
754 - the naked truth about Tk - it's basically unmaintained.
755
7564.32 Mon Nov 3 22:46:32 CET 2008
757 - fix AnyEvent::Socket::resolve_sockaddr to properly support
758 unix sockets again and choose a proper default for
759 the socket type.
760 - fix call to resolve_sockaddr in tcp_connect.
761
7624.31 Thu Oct 30 04:41:48 CET 2008
763 - implemented AnyEvent::Handle->destroy method.
764 - hint about unexpected effects in TLS mode.
765 - speed up AnyEvent::Impl::Perl by using more arrays
766 instead of hashes without the slightest loss of readability :)
767 - work around a perl argument refcounting bug.
768 - with some perl patches applied, AnyEvent now seems leak-free.
769
7704.3 Fri Oct 3 09:18:43 CEST 2008
771 - AnyEvent will now install a no-op signal handler for SIGPIPE,
772 unless one has been installed already.
773 - warn about not loaidng AnyEvent::Impl::POE early enough
774 (patch by Adam Rosenstein).
775 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
776 - fix a bug in where in SSL connect mode, the
777 client would first wait for some data by the server and
778 otherwise hang. (reported and analysed in an absolutely
779 exemplary manner by Adam Rosenstein).
780 - fix a bug in where SSL EOF would not be treated
781 as stream EOF, putting the connection into a hung state.
782 (reported and analysed in an absolutely exemplary manner by Adam
783 Rosenstein).
784 - fix a potential segfault when the TLS context
785 would go missing in a read callback (e.g. due to stoptls) -
786 Net::SSLeay of course makes no type checking whatsoever.
787 - AnyEvent::Handle will not stop the read watcher in TLS mode.
788 - AnyEvent::Handle->stoptls will send a TLS close notify
789 instead of simply destroying the stream now.
790 - fix a bug where large blocks of data written in TLS mode
791 would not be sent unless triggered by receives.
792 - on_drain will now take the tls write buffer into account.
793 - SSL operations have been streamlined a bit and should be faster
794 now, more to come, though.
795 - added a FAQ section, will document nontrivial issues in there.
796
7974.234 Mon Sep 29 04:08:13 CEST 2008
798 - fix child watcher documentation: only child exits will be reported,
799 no trace events.
800 - mention SIGPIPE in AnyEvent::Handle.
801 - perl backend will now use POSIX::times as monotonic clock
802 when available and the monotonic clock option is not.
803 - run even without Time::HiRes in some cases now, but do not
804 provide sub-second accuracy (all *supported* perl versions have
805 Time::HiRes).
806
8074.233 Fri Aug 22 01:48:05 CEST 2008
808 - fix a bug in the testsuite, causing a hang.
809 - clarified lots of AnyEvent::Handle constructor arguments.
810
8114.232 Thu Aug 21 20:44:25 CEST 2008
812 - fix a bug in AnyEvent::Handle that could cause two fatal error
813 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
814 (testcase by Leon Brocard).
815 - clarified on_eof/on_error documentation a bit.
816
8174.231 Tue Jul 29 13:12:15 CEST 2008
818 - remove some debugging code left in AnyEvent::Util::fork_call
819 (and no, it's impossible to implement with the broken windows
820 perls without resource leaks or worse).
821
8224.23 Tue Jul 29 12:19:59 CEST 2008
823 - document the first parameter passed to condvar callbacks to be
824 the callback.
825 - add AnyEvent::Socket::{ntoa,aton} aliases.
826 - optimize the AE::Handle->push_read (line) for the default
827 eol marker.
828 - optimize push_read (packstring|storable) for small packets.
829 - invoke on_error callback when no on_eof callback is set.
830 - fix a bug in push_read (storable) of unknown impact.
831
8324.22 Sun Jul 20 16:34:13 CEST 2008
833 - new function AnyEvent::Socket::parse_hostport.
834 - as the bulkheads at microsoft can't even get getprotobyname reliably
835 working on their shitty fucking broken os we need to hardcode
836 some common protocol numbers in AnyEvent::Socket. How can
837 people even bother with such a pile of shit as windows.
838
8394.21 Thu Jul 17 14:40:05 CEST 2008
840 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
841 which is incompatible to previous invocation syntax.
842 - work around more windows perl bugs in fork_call. I cannot imagine
843 why anybody would bother implementing fork in such an obviously
844 unusable way.
845 - avoid calling $do_retry if the request has already finished in
846 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
847 K. Sharifulin).
848
8494.2 Sat Jul 12 22:42:11 CEST 2008
850 - support an "interval" argument for repeating timers.
851 - fix a bug in the pure perl event loop that caused timers to
852 be delayed under some circumstances.
853 - implement optional argument checking via PERL_ANYEVENT_STRICT.
854 - not importing anything from Event causes it to malfunction
855 and not use Time::HiRes, work around this bug.
856 - more bug workarounds against the endlessly broken Tk module,
857 seems to work now on 32 bit machines, but less so on 64 bit.
858 - minor optimisations applied to most backends.
859
8604.161 Fri Jul 4 14:24:48 CEST 2008
861 - fixed the t/handle/02_write.t, it read after write, with the
862 assumption that all data could be written before read
863 (reported by Jim Razmus).
864
8654.160 Thu Jul 3 04:02:21 CEST 2008
866 - re-registering signal watchers after unregistering one did
867 not work if a backend falls back on AnyEvent's default
868 implementation (also affected child watchers).
869 - new AnyEvent::Handle options: autocork and no_delay.
870
8714.152 Sun Jun 22 14:15:44 CEST 2008
872 - allow for 32-bit perls that implement shifts differently
873 on different architectures in parse_ipv4 (reported and
874 analysed by Keiichi DAIBA).
875
8764.151 Fri Jun 6 17:34:24 CEST 2008
877 - make sure specifying _only_ on_read and never pushing reads
878 works.
879
8804.15 Fri Jun 6 13:00:46 CEST 2008
881 - the pure perl backend would keep some watchers alive when more than
882 one watcher was registered for the same fd.
883 - new "packstring" and "storable" read and write types
884 for AnyEvent::Handle.
885 - allow on_eof handler to be called after on_error with EPIPE returns.
886 - do not immediately call on_read callback in handle constructor.
887
8884.14 Thu Jun 5 20:29:31 CEST 2008
889 - Fixed a bug in DNS SRV priority sorting.
890 - AnyEvent::Util::guard now reports runtime errors while
891 executing the guard block as warnings.
892 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
893
8944.13 Thu Jun 5 00:47:59 CEST 2008
895 - AnyEvent::DNS only followed cname chains with length 2,
896 contrary to documentation. bumped it up to 10, thanks to
897 microsoft, the current broken-dns-config-king.
898 - AnyEvent::DNS didn't check for socket return status
899 but instead relied on perl not creating filehandles in that
900 case - too bad it gives you a bogus file handle (reported
901 and analysed by Vladimir Timofeev).
902 - fix queue management logic in AnyEvent::Handle:
903 when on_read was registered, the queue was empty
904 and no progress could be made AnyEvent::Handle would
905 enter an endless loop.
906 - correctly start reading again when the handle became
907 busy again after idling.
908 - correctly treat tls shut-downs as EOF condition
909 (for the time being, would be nice to have a callback for that).
910 - correctly call eof callback under all conditions
911 (hopefully).
912 - the timeout callback did not expect that $self can go
913 away any time.
914
9154.12 Tue Jun 3 10:58:04 CEST 2008
916 - include AnyEvent::Intro, a tutorial for anyevent,
917 anyevent::socket and anyevent::handle.
918 - allow more options in on_error.
919
9204.11 Fri May 30 23:42:25 CEST 2008
921 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
922 reverse_lookup and reverse_verify to replace it, support
923 v4mapped and v4compat addresses.
924 - provide more documentation for the resolver class.
925 - really replace longest run of :0: by :: in format_address,
926 also properly convert :: and ::1 again.
927 - support NAPTR record name and decode it.
928 - implement random weight sampling for SRV records, as per
929 rfc 2782.
930 - correctly abort on srv-record targets of ".".
931 - added AnyEvent::DNS::wait_for_slot.
932 - in the unlikely event of a virtual circuit connection
933 being invalidated by a delayed udp reply, AnyEvent::DNS
934 could die. this has been fixed.
935 - plug a probable memleak in the DNS vc code.
936 - use configured timeout also for VC request phase.
937 - implement timeout and max_outstanding methods
938 for resolver class.
939 - update version numbers in all modules.
940
9414.1 Thu May 29 05:45:40 CEST 2008
942 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
943 which mirrors their purpose better (old names still available
944 for a while). Also moved AnyEvent::DNS::addr to
945 AnyEvent::Socket::resolve_sockaddr.
946 - implement AnyEvent->time and AnyEvent->now.
947 - fix IPv6 support in pack_sockaddr.
948 - officially un-experimentalise all newly introduced code.
949 - support unix domain sockets everywhere by specifying
950 a host of "unix/" and the pathname as service.
951 - implement an activity timeout in AnyEvent::Handle.
952 - added a regex read type to AnyEvent::Handle.
953 - added a json read/write type to AnyEvent::Handle.
954 - always croak in on_error in AnyEvent::Handle.
955 - document how to subclass AnyEvent::Handle.
956 - implement AnyEvent::Util::fork_call.
957 - add support for IPv6 nameservers and nameserver statements.
958 - work around _yet_ _another_ windows perl bug in where empty
959 select masks cause errors in the pure perl backend (this
960 workaround cannot be implemented for the other event loops,
961 but EV also works around this).
962 - supply AnyEvent::Socket::address_type and make good use of it.
963 - clarify and fix the Handle documentation w.r.t $self vs. $handle
964 vs. callback arguments.
965 - add some recommends to the META.yml.
966
9674.05 Mon May 26 19:44:06 CEST 2008
968 - some platforms ignore the rfc and prepend an extra sa_len member to
969 the sockaddr structure, cater for those. also use sockaddr_family,
970 didn't know it was there, but comes in handy for the workaround.
3 - undo the import WIN32 hack. 971 - undo the import WIN32 hack.
4 - use sockaddr_family instead of relying on structure layout (because,
5 of course, darwin is broken again).
6 972
74.04 Mon May 26 08:03:31 CEST 2008 9734.04 Mon May 26 08:03:31 CEST 2008
8 - try to work around yet another windows bug: failed connects are reported 974 - try to work around yet another windows bug: failed connects are reported
9 as if it were out-of-band data. windows users: you suck. Only 975 as if it were out-of-band data. windows users: you suck. Only
10 EV, Glib, Event and the pure perl backend can handle this condition. 976 EV, Glib, Event and the pure perl backend can handle this condition.
… …
47 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore. 1013 - AnyEvent::Socket::inet_aton will use AnyEvent::DNS now and not fork anymore.
48 - try very hard not to freeze on broken windows perls. 1014 - try very hard not to freeze on broken windows perls.
49 - make on_eof optional in AnyEvent::Handle. 1015 - make on_eof optional in AnyEvent::Handle.
50 - support service names in tcp_connect and tcp_server. 1016 - support service names in tcp_connect and tcp_server.
51 - work around netbsd/openbsd bug causing unix sockets not to 1017 - work around netbsd/openbsd bug causing unix sockets not to
52 become writable untill empty. 1018 become writable until empty.
53 - only make a dummy read for error detection when getpeername 1019 - only make a dummy read for error detection when getpeername
54 returns ENOTCONN. 1020 returns ENOTCONN.
55 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 1021 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
56 override. 1022 override.
57 - new AnyEvent::Guard::cancel. 1023 - new AnyEvent::Guard::cancel.
58 1024
593.5 Sat May 17 23:17:14 CEST 2008 10253.5 Sat May 17 23:17:14 CEST 2008
60 - gracefully deal with systems not defining CLOCK_MONOTONIC 1026 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
74 1040
753.41 Sun May 11 19:53:13 CEST 2008 10413.41 Sun May 11 19:53:13 CEST 2008
76 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 1042 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
77 1043
783.4 Sun May 11 00:29:25 CEST 2008 10443.4 Sun May 11 00:29:25 CEST 2008
79 - complete re-implementation of condvars with a hopefully much mroe useful 1045 - complete re-implementation of condvars with a hopefully much more useful
80 API (backward compatible functions exist but are undocumented). 1046 API (backward compatible functions exist but are undocumented).
81 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 1047 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
82 (still subject to change). 1048 (still subject to change).
83 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 1049 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
84 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 1050 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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