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

Comparing AnyEvent/Changes (file contents):
Revision 1.205 by root, Mon Nov 3 21:49:25 2008 UTC vs.
Revision 1.498 by root, Fri Aug 26 18:08:07 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
3TODO: document TCP_*** constants
4
5 - log_to_syslog now accepts facility strings.
6 - log_to_syslog would not allow facilities - "$facility|$level"
7 *literally* means that string. sheesh.
8 - the CPAN parser got confused and thought we provide Sys::Syslog, try
9 to work around.
10 - renamed AnyEvent::Impl::FLTK2 to FLTK.
11
126.01 Fri Aug 26 07:04:11 CEST 2011
13 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
14 the TTL as fourth element - this affects ->resolve, ->request
15 and ->dns_unpack indirectly. Convenience functions such as
16 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
17 AnyEvent::Socket.
18 - INCOMPATIBLE CHANGE: backend authors now should not implement
19 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
20 methods.
21 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
22 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
23 - DEPRECIATION: please change
24 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
25 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
26 in your programs, for future compatibility.
27 - *blush* idle watcher emulation was borked.
28 - AnyEvent::Strict now checks the AE::xxx functions as well.
29 - querying too long domain names would barf deep inside AnyEvent::DNS - since
30 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
31 for too long domain names explicitly and passes an undef to the callback
32 instead (reported by James Bromberger).
33 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
34 - new functions: AnyEvent::postpone, AnyEvent::log.
35 - new module: AnyEvent::Log, for simple logging needs.
36 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
37 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
38 a debug shell.
39 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
40 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
41 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
42 F</etc/hosts> can be found.
43 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
44 configuration.
45 - env variables can now be specified using AE_xyz instead of
46 PERL_ANYEVENT_xyz, subject to some rules.
47 - lots of new AnyEvent::Debug functionality.
48 - AnyEvent::Strict now checks for modifications $_ by the callback
49 (by setting it to readonly while executing the callback, if possible).
50 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
51 Invest extra overhead to work around this peculiar behaviour.
52 - split the pure perl backend into a loop module and an impl
53 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
54 - some read types (regex, netstring, json, storable) could
55 errornously keep an AE::Handle object alive on their own
56 (reported by Mohammad Toossi).
57 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
58 supports /etc/hosts overrides.
59 - add more workarounds around child watcher bugs in IO::Async
60 and (newly introduced) in POE.
61 - use glib child watchers and try to work around its limitations,
62 if possible (based on a patch by Kevin Ryde).
63 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
64 - many read types in AnyEvent::Handle would malfunction during
65 transitions to TLS because of a "delete $self->{rbuf}".
66 (reported by Gerald Galster).
67 - improve condvar blocking wait performance (probably only noticable
68 with faster backends), also make it easier for event loops that
69 do not support blocking waits.
70 - PERL_ANYEVENT_MODEL now also allows full module names, although
71 the usefulness of this feature is not yet clear.
72 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
73 - add t/80_ssltext.t.
74 - no longer use AUTOLOAD for the initial loading (for no very good reason).
75 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
76 - do not run event loop detection multiple times when programs call their cached
77 copies of AnyEvent::detect.
78 - parse_hostport now accepts absolute paths als unix domain sockets.
79 - suppress the idiotic warning inside POE by patching the POE::Kernel
80 run flags instead of asking the user to jump through contortions.
81 - add Task::Weaken as dependency (as recommends only at this time).
82 - replace PF_UNSPEC by 0, for good old minix.
83
846.0 (not officially released, but was on CPAN for about 15 minutes :/)
85
865.34 Thu May 12 09:59:41 CEST 2011
87 - AE::TLS didn't support both cert_file and key at the same time,
88 due to some copy&paste bug (reported and analyzed by vti).
89 - improve AE::Handle timeout documentation slightly.
90
915.33 Mon Feb 7 21:51:41 CET 2011
92 - never appeared on CPAN, for unknown reasons.
93 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
94 object when it's already gone (analyzed by Ryan Bullock).
95 - update idna algorithm(s) to UTS#46 version 6.0.0.
96 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
97 uts46 says it's an error, and testsuite makes more sense.
98 - remove leading dots in idn_to_ascii - testsuite says to remove,
99 uts46 says it's an error, and testsuite makes more sense.
100 - fixed a bug in idn ignored character processing.
101 - passes the uts46 6.0.0 testsuite except for three cases which
102 apparently are bugs in the testsuite itself.
103 - passes the uts46 6.0.1 testsuite except for eleven cases which
104 apparently are bugs in the testsuite itself.
105
1065.31 Sun Jan 23 11:49:19 CET 2011
107 - work around a POE bug where POE would first poll for new events and
108 invoke handlers BEFORE handling the message queue.
109 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
110 - stop_read now does just that in TLS mode. see its
111 documentation for caveats.
112 - update to common::sense 3.4 warning set (some warnings in perl impose
113 an arbitrary 32kb limit on regex matches, which affects
114 AnyEvent::Handle).
115 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
116 - new experimental AE::Handle::resettls method.
117
1185.3 Fri Dec 31 05:48:13 CET 2010
119 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
120 issues with IO::Async have been solved in that version, including
121 the availabilty of a default loop!
122 - implement new wbuf_max parameter for AnyEvent::Handle.
123 - added Cocoa::EventLoop backend by Daisuke Murase.
124 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
125 - clarify handle on_prepare documentation.
126 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
127 to avoid leaking them to any child processes.
128 - try to force files into the arch-dependend lib directory, for the
129 half-"turly obsessive compulsive person"(s) that depend on it.
130
1315.29 Sun Dec 5 10:49:21 CET 2010
132 - convert EV backend to EV 4.00 API (so better upgrade EV too).
133 - AE::Handle: implement dynamic read_size adjustment, add
134 max_read_size parameter, reduce default read_size to 2048.
135 - add some FAQ entries for servers and/or on_eof vs. on_error.
136 - work around OS bugs (cygwin again) when lingering in AE::Handle.
137 - reduce memory usage slightly when connect was used in AE::Handle.
138 - wrok around more segfaults in Tk.
139 - document tls_ctx => undef as valid.
140 - detect WNOHANG value at compiletime.
141 - include some probably linux-only support for building AnyEvent
142 as part of the perl core.
143 - improve compatibility of t/08_idna.t to perl 5.8.
144 - make AnyEvent::Strict truly optional (in case somebody builds a
145 minimal perl...).
146
1475.28 Wed Oct 13 04:14:23 CEST 2010
148 - due to a glitch, AnyEvent's internal getprotobyname wasn't
149 actually used in tcp_server.
150 - implement AnyEvent::CondVar->new method, which was documented but
151 missing.
152 - support autoloading of read types in unshift_read.
153 - AnyEvent::Strict: do the test for invalid fh arguments differently,
154 resulting in more informative error messages.
155 - new AnyEvent::FAQ manpage.
156 - clarify that you can actually call push_read etc. in on_read.
157 - put keys and certs in /tmp, not into the current directory,
158 and document this in the SECURITY CONSIDERATIONS sections
159 of AnyEvent::TLS.
160 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
161 requested by Ron Isaacson.
162 - document $AnyEvent::DNS::RESOLVER variable.
163 - incorporated major typo patches by Abhijit Menon-Sen.
164 - AnyEvent::Handle now croaks when negative timeouts are passed.
165 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
166 undocumented, but who knows when they might come in handy.
167
1685.271 Tue Jun 8 12:05:46 CEST 2010
169 - backport to perl 5.8.x.
170
1715.27 Sun Jun 6 12:12:05 CEST 2010
172 - postpone differently in AnyEvent::Socket now, as
173 when not, canceling the connection attempt might fail
174 (found by Felix Antonius Wilhelm Ostmann).
175 - explicitly check for non-stream sockets in AE::Handle, too many
176 clueless people fell into the trap of this somehow working.
177 - simplified and reworked the "OTHER MODULES" section.
178 - better/more condvar examples.
179
1805.261 Wed Apr 28 16:13:36 CEST 2010
181 - AF_INET6 was not properly used from Socket6 during configuration
182 time (found by Andy Grundman).
183
1845.26 Mon Apr 12 04:49:35 CEST 2010
185 - don't generate spurious readyness notifications when select
186 returns EINTR in the pure perl backend (analysed and testcase
187 by Bas Denissen).
188 - give same examples in SYNOPSIS sections of AnyEvent and AE.
189 - provider faster implementations for AE::time/now/now_update for
190 EV and Perl backends.
191 - provide faster AE::cv and AE::time implementations for all
192 backends.
193 - no longer support lower/mixed-case signal names (this was
194 never documented nor universally supported).
195 - some more memory µ-optimisations, and somewhat less messy
196 code to cope with both AnyEvent and AE APIs.
197
1985.251 Sat Mar 13 00:58:21 CET 2010
199 - make test failed if it was run before make install on perl <=
200 5.8.8, and older versions of perl make our live very difficult,
201 so write constants.pl during Makefile.PL time.
202
2035.25 Sat Mar 13 00:23:14 CET 2010
204 - fix a race condition in AnyEvent::Handle that would cause
205 a "bio_read: unsupported method" error (in _another_
206 ssl connection) after user code threw an exception.
207 - added AnyEvent::Handle->destroyed method.
208 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
209 - the AnyEvent::Util::fh_nonblocking function worked only
210 by lucky accident on win32.
211 - smaller and faster AnyEvent::Util::fh_nonblocking.
212 - when the (required!) Time::HiRes module is missing, AnyEvent
213 did not fall back to built-in time properly.
214 - do not load Fcntl at runtime, saving memory and loading time.
215 - precompile a number of constants and use them instead of runtime
216 detection and eval.
217 - free detection code after detection and similar memory optimisations.
218 - Perl backend timer interval best effort drift has been improved
219 (same algorithm as EV).
220 - update unicode idna mapping table.
221
2225.24 Tue Jan 5 11:39:43 CET 2010
223 - cygwin never reports errors from failed connects. 1.5 just gives
224 you continous readyness and EAGAIN, 1.7 is even more broken and
225 just hangs. work around both issues in a cygwin-specific hack.
226 - improve idle watcher documentation slightly.
227
2285.23 Sun Dec 20 23:48:00 CET 2009
229 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
230 - implement punycode_encode/decode, idn_nameprep,
231 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
232 - provide $AE::VERSION.
233 - removed traces of "no strict 'refs'".
234
2355.22 Sat Dec 5 03:51:13 CET 2009
236 - downgrade-or-fail in AnyEvent::Handle::push_write, to
237 diagnose encoding failures earlier and more succinctly.
238 (this works around bugs in perl, throwing away encoding info
239 when passing scalar data to extensions).
240 - add more examples to AnyEvent::Socket manpage.
241 - upgrade internal warning set to the same as common::sense 2.03.
242 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
243 (John Beppu).
244 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
245 - add google public dns servers to fallback server set and make sure
246 we load-balance properly between the three sets. also add all
247 fallback dns servers, not just a random one, to each dns config.
248
2495.21 Thu Nov 19 02:48:47 CET 2009
250 - fix a problem where socket constants were called with parameters
251 (spotted by David Friedland).
252 - fork_call never use'd POSIX (reported by Daisuke Maki).
253 - improve perl 5.6 compatibility further (but it still won't work
254 unless you rip out everything but the core).
255 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
256 uses 10MB of RAM, but doesn't open a console window. *sigh*.
257
2585.202 Wed Oct 14 22:35:44 CEST 2009
259 - AnyEvent::DNS would unexpectedly clobber $_ under windows
260 (analysed by Matthias Waldorf).
261 - AnyEvent::Handle::run_cmd can now pass the PID of the
262 newly-created process, which is much less useful than it might
263 sound (based on patch by Yann Kerherve).
264
2655.201 Tue Sep 29 12:09:25 CEST 2009
266 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
267 (reported by Torsten Foertsch).
268 - common::sense 2.0 could cause tcp_server to throw an exception
269 (analysed by elmex).
270
2715.2 Mon Sep 14 07:04:49 CEST 2009
272 - INCOMPATIBLE CHANGE: do no longer support register_read_type
273 and register_write_type in AnyEvent::Handle, instead support
274 package names (the facility was mostly abused).
275 - implement "packagename-as-read/write type" support in
276 AnyEvent::Handle.
277 - AnyEvent::Handle: new options "keepalive" and "oobinline".
278 - oobinline set by default to avoid security issues.
279 - the pure-perl event loop backend wrongly detected times() fallback
280 support (spotted by Pavel Boldin).
281
2825.12 Tue Sep 1 20:26:50 CEST 2009
283 - be more lenient when parsing resolv.conf files, as some people
284 use hashmarks as comment indicator inside directives (reported
285 by Michael S. Fischer).
286 - use same set of warnings as common::sense 2.0.
287 - fix a potential 32 bit overflow issue due to perl having problems
288 with large hex constants in the Perl backend.
289 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
290 by default.
291 - fix some prototypes.
292 - (optionally) require Async::Interrupt 1.02.
293 - replace opendns fallbacks by verizon/level3 public dns servers.
294
2955.112 Fri Aug 21 13:59:04 CEST 2009
296 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
297 using starttls with an illegal/uncreatable tls context.
298 - port to broken win32 perls.
299
3005.111 Sun Aug 16 18:44:36 CEST 2009
301 - perl errornously requires sockaddr_un structures returned by the
302 kernel to have a specific length. work around this bug by padding
303 sockaddr_un structures (found to fail on OS X, as perl only uses
304 the correct code for GNU/Linux, but likely to fail on other
305 OSes as well).
306
3075.11 Wed Aug 12 17:49:37 CEST 2009
308 - fixed a bug in run_cmd with input from filename.
309 - create a json coder object when none is given in write direction
310 to reduce dependency on JSON.
311
3125.1 Tue Aug 11 03:17:32 CEST 2009
313 - tcp_connect now reports the correct error in $!, not an unrelated
314 one (was broken since 4.91).
315 - AE::Handle did not properly clear rtimeout/wtimeout watchers
316 when retrying a tcp_connect.
317 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
318 - fixed wrong documentation in AE::Socket.
319
3205.01 Mon Aug 10 03:16:32 CEST 2009
321 - last release broke EV child watchers :/.
322
3235.0 Sun Aug 9 17:32:09 CEST 2009
324 - officially introduce and document the AE API.
325 - lowering the timeout at runtime did not have immediate effect
326 in AnyEvent::Handle.
327 - AE::Handle now has separate and independent read and write timeout
328 settings.
329 - probe for Guard module when AnyEvent::Util is loaded, not at
330 runtime (Event doesn't like eval "use" from callbacks, and it avoids
331 exporting the probe function).
332 - Impl::IOAsync's io watchers did not use the filehandle emulation
333 code to work around it's watcher limits.
334 - work around IO::Async's problems with overloaded objects.
335 - try to disable wearnings inside IO::Async, it's just too buggy.
336 - do no longer use Event by default, only EV or Perl, Event is too buggy.
337 - hackishly provide fast AE:: interface when EV or Perl is the backend.
338 - inherit VERSION from AnyEvent::VERSION in most modules.
339
3404.91 Thu Aug 6 15:42:45 CEST 2009
341 - AE::Handle::starttls could get out of sync when the read buffer
342 already contains some TLS handshake.
343 - AE::Handle did not properly free the TLS session, causing bigger
344 memory leaks in Net::SSLeay than expected (reported by toaster).
345 - AE::Socket::tcp_connect will now artificially delay invoking
346 the callback to avoid returning after invoking the callback.
347 - convert many internal watcher uses to AE API.
348
3494.9 Sat Aug 1 11:07:01 CEST 2009
350 - Glib silently fails when registering a timeout with
351 a negative value, so avoid that.
352 - call condvar callback immediately when it is set
353 after the condition is already signalled.
354 - check rbuf_max condition only after trying to consume
355 data.
356 - normalise signal numbers to names when using the ae's
357 signal handling, but do not document this (yet).
358 - pure perl signal emulation did not properly set nonblocking
359 mode on the signal pipe on win32 (this is unlikely to have caused
360 any issues).
361 - new module: AnyEvent::Debug.
362 - AnyEvent::Strict now checks that a signal was specified
363 by name (not name or number).
364 - reduce memory footprint in typical cases by ~50kb by
365 only compiling signal code when necessary.
366 - add AnyEvent::Handle->rbuf_max.
367 - grab the AE namespace for future new API, implement stubs for
368 said future/alternative API.
369 - new function: AnyEvent::Socket::format_hostport.
370
3714.881 Tue Jul 28 12:51:53 CEST 2009
372 - work around a bug in local in pre-5.10 perls, causing
373 AnyEvent::Handle to recurse when it should not
374 (analysed by elmex).
375
3764.88 Tue Jul 28 04:04:37 CEST 2009
377 - re-bless the handle into a dummy package after calling
378 AnyEvent::Handle::destroy, so the user does not need to check
379 for errors after every push_write etc.
380 - do not attempt to run t/02_signals.t on obviously broken
381 platforms.
382
3834.87 Sun Jul 26 02:06:16 CEST 2009
384 - do not attempt to linger when there is no longer a valid fh.
385 - enforce tls mode to be either accept or connect, do not simply
386 segfault in Net::SSLeay.
387 - AnyEvent::Handle can now call tcp_connect itself (new parameters
388 connect, on_prepare, on_connect and on_connect_error). Updated
389 tutorial accordingly.
390 - add AnyEvent::Impl::Irssi backend.
391
3924.86 Mon Jul 20 23:52:29 CEST 2009
393 - since the verbose warning is not enough, explicitly document
394 that versions before 1.33 of Net::SSLeay are not secure.
395 - work around signal handling races in Event and (...) Event::Lib.
396 - try to align signal-race timer to full-second boundaries.
397 - work around Tk not liking negative timeouts.
398 - don't complain of different grades of environmental unfriendlyness
399 in IO::Async.
400
4014.85 Sat Jul 18 06:16:14 CEST 2009
402 - nail the signal race problem in perl once and for all
403 (see $AnyEvent::MAX_SIGNAL_LATENCY).
404 - take advantage of Async::Interrupt if it is available.
405 - load Time::HiRes and Guard modules on demand only.
406 - add optional/recommended modules section to AnyEvent
407 documentation.
408 - reduce memory usage considerably (and reduce startup penalty)
409 by not using "strict", "warnings" and "overload" modules.
410 - work around buggy windows/openbsd perls and provide EBADMSG
411 and EPROTO ourselves when missing.
412 - improve perl 5.6 compatibility of the core event loop.
413 - made Net::SSLeay version 1.33 a soft requirement.
414
4154.83 Fri Jul 17 16:56:26 CEST 2009
416 - implement AnyEvent::Socket::getprotobyname.
417 - AnyEvent::CondVar's will now detect recursive blocking
418 waits and will croak, as too many people fall into
419 this trap.
420 - AnyEvent::Handle will now call ->destroy on itself after
421 executing the on_error callback, instead of doing some
422 half-baked internal shutdown, for fatal errors.
423 - clarify on_eof behaviour w.r.t. the read queue and
424 on_read callbacks.
425 - ignore some possible spurious wake-ups in tcp_connect.
426
4274.82 Sat Jul 11 00:34:55 CEST 2009
428 - POE and Event backends didn't accept some callable objects as
429 callbacks.
430 - use Config module instead of POSIX module to detect signal names
431 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
432 module doesn't even have all POSIX signals :/.
433 - use more workarounds around the many refcnt/corruption bugs in
434 Event::Lib.
435 - work around a race condition in perl's select, causing t/03_child.t
436 to rarely fail.
437
4384.81 Thu Jul 9 10:30:30 CEST 2009
439 - AnyEvent::Handle didn't properly diagnose write errors
440 (it expected -1 from syswrite, how lame... :).
441 - support file descriptors in addition to file handles
442 in AnyEvent->io.
443 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
444 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
445 and PERL_ANYEVENT_CA_PATH.
446 - provide a sensible synopsis section for AnyEvent::TLS.
447 - add a "supported backends" section to the manpage.
448 - added simple io watcher test to testsuite, using a
449 portable_socketpair.
450 - tried to improve the stability of the Event::Lib backend,
451 YMMV.
452
4534.8 Mon Jul 6 23:45:16 CEST 2009
454 - AnyEvent::DNS did not properly follow CNAME records with
455 uppercase targets.
456 - AnyEvent::DNS would errornously return AAAA records
457 with v4 mapped addresses (a faulty record) as ipv4 addresses,
458 causing AnyEvent::Socket to throw an exception.
459 - added new module AnyEvent::TLS for easier SSL/TLS context
460 creation, with many options including hostname verification,
461 secure default configuration, lots of documentation and,
462 predefined diffie-hellman keys for perfect forward security
463 and much more. get it while it's still fresh!
464 - use AnyEvent::TLS in AnyEvent::Handle for context management.
465 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
466 so AnyEvent::Socket users have smaller memory footprint
467 in the common case.
468 - add AnyEvent::Handle->push_shutdown method.
469 - add an additional error message parameter to AnyEvent::Handle's
470 on_error callback (for TLS, $! is still available).
471 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
472 - make AnyEvent::Handle more robust against early conenction
473 failures (during new), and return C<undef> in that case
474 from the constructor.
475 - AnyEvent::Handle will now try to load only JSON::XS first,
476 then fall back to JSON.
477 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
478 and because it was documented that way.
479
4804.451 Fri Jul 3 00:28:58 CEST 2009
481 - do not clear rbuf when shutting down an AnyEvent::Handle
482 object - doing so breaks AnyEvent::HTTP.
483
4844.45 Mon Jun 29 22:59:26 CEST 2009
485 - a write error could cause AnyEvent::Handle to create
486 an I/O watcher with an undefined $fh.
487 - special-case mapped ipv4 addresses in both
488 AnyEvent::Socket::format_address and parse_address,
489 to treat them just like ipv4 addresses.
490 - updated and overhauled the AnyEvent::Intro doc.
491 - implement AnyEvent::Socket::format_ipv4/ipv6.
492 - slightly speed up portable_pipe/socketpair.
493 - expand condvar begin/end documentation.
494
4954.42 Fri Jun 26 08:32:18 CEST 2009
496 - preliminary and neccesarily incomplete support for IO::Async.
497 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
498 case it was set to IGNORE, to ensure we can catch child statuses
499 even when the calling env acts stupidly.
500 - updated benchmarks with IO::Async, which performs very well.
501
5024.412 Wed Jun 24 01:35:57 CEST 2009
503 - support an "untaint" attribute for AnyEvent::DNS
504 and set it on the default resolver.
505 - implement and document AnyEvent::Impl::Perl::loop.
506 - remove all anyevent-env variables from %ENV
507 when running in tainted mode.
508 - mention and extend the IO::Lambda benchmark.
509
5104.411 Sun Jun 7 18:48:02 CEST 2009
511 - do not try to use F_SETFD on windows.
512
5134.41 Thu May 14 06:40:11 CEST 2009
514 - work around issues in older perls (5.8.0?) when
515 a signal handler is deleted from the %SIG hash.
516 - use POSIX::_exit in child test, to avoid running
517 destructors.
518 - speed up CHLD handling by relying on SIGCHLD being
519 synchronously delivered, even when we roll our own
520 implementation.
521 - AnyEvent::DNS: add the "dname" resource record name
522 for cosmetic reasons.
523
5244.4 Sun Apr 26 20:12:33 CEST 2009
525 - implemented idle watchers, where applicable.
526 - AnyEvent->time died when Event backend was in use.
527 - fix a memleak in the Tk backend.
528 - sped up Tk timer handling.
529 - clip negative "after" values to 0 in AnyEvent::Impl::Event
530 to avoid spamming out warning messages.
531 - fix Qt timers without interval.
532 - avoid Qt zero-timeout special case, in old libqt's.
533
5344.352 Mon Apr 20 16:31:11 CEST 2009
535 - fix AnyEvent::Strict error messages for child watchers.
536 - fix/update Makefile.PL "recommends" versions.
537 - add AnyEvent->now_update.
538
5394.351 Sat Apr 11 07:56:14 CEST 2009
540 - actually make the signal pipe work under win32.
541 - localise $! in signal handler to avoid changing $!, although
542 perl itself does not seem to save/restore errno either.
543 - set the cloexec flag on the signal pipe (normally set by perl too).
544
5454.35 Fri Mar 27 11:48:20 CET 2009
546 - event models relying on AnyEvent's signal watcher emulation did
547 invoke the callback asynchronously, contradicting documentation
548 and causing signals to get lost (this includes AnyEvent's own
549 event loop). AnyEvent now uses the standard pipe trick to make
550 callback execution synchronous to the event loop.
551 - AnyEvent::Handle didn't free TLS context data on DESTROY
552 (patch by Pavel Shaydo).
553 - work around the ever-incompatibly-changing API of MakeMaker.
554 - document that changing global variables without restoring them
555 is a bad idea in Perl (noted by Adam Rosenstein).
556 - AnyEvent::Strict now barfs if ->io is passed a file.
557
5584.34 Thu Feb 12 18:32:45 CET 2009
559 - separately buffer TLS read data, as otherwise the read queue could
560 deadlock as receiving data is not expected while draining the
561 read queue (which cna only happen with TLS).
562 - raise EBADMSG error on JSON decoding errors.
563 - fix some minor manpage bugs (reported by Maximilian Gaß).
564 - speed up select bitmask parsing quite a bit in the pure perl
565 backend.
566 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
567
5684.331 Tue Jan 6 21:07:25 CET 2009
569 - socketpair fails on many vista machines because vista has
570 completely broken accept/getpeername and getsockname functions,
571 so we provide our own socketpair emulation that kind of works
572 (AnyEvent::Util::portable_pipe).
573 - new function: AnyEvent::Util::portable_socketpair.
574 - take advantage of the Guard module if it exists.
575
5764.33 Fri Nov 21 02:35:40 CET 2008
577 - AnyEvent::Strict did errornously flag a fileno of 0 as
578 illegal.
579 - reduce memory usage and slightly speed up the pure perl backend
580 by only storing the file descriptor, not the file handle.
581 - add missing autocork method to AnyEvent::Handle (reported
582 by Adam Rosenstein).
583 - AnyEvent::DNS->resolve errornously documented an $rcode result
584 argument, but there isn't (spotted by Henrik Krohns).
585 - the naked truth about Tk - it's basically unmaintained.
2 586
34.32 Mon Nov 3 22:46:32 CET 2008 5874.32 Mon Nov 3 22:46:32 CET 2008
4 - fix AnyEvent::Socket::resolve_sockaddr to properly support 588 - fix AnyEvent::Socket::resolve_sockaddr to properly support
5 unix sockets again and choose a proper default for 589 unix sockets again and choose a proper default for
6 the socket type. 590 the socket type.
… …
174.3 Fri Oct 3 09:18:43 CEST 2008 6014.3 Fri Oct 3 09:18:43 CEST 2008
18 - AnyEvent will now install a no-op signal handler for SIGPIPE, 602 - AnyEvent will now install a no-op signal handler for SIGPIPE,
19 unless one has been installed already. 603 unless one has been installed already.
20 - warn about not loaidng AnyEvent::Impl::POE early enough 604 - warn about not loaidng AnyEvent::Impl::POE early enough
21 (patch by Adam Rosenstein). 605 (patch by Adam Rosenstein).
22 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 606 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
23 - fix a bug in where in SSL connect mode, the 607 - fix a bug in where in SSL connect mode, the
24 client would first wait for some data by the server and 608 client would first wait for some data by the server and
25 otherwise hang. (reported and analyzed in an absolutely 609 otherwise hang. (reported and analysed in an absolutely
26 exemplary manner by Adam Rosenstein). 610 exemplary manner by Adam Rosenstein).
27 - fix a bug in where SSL EOF would not be treated 611 - fix a bug in where SSL EOF would not be treated
28 as stream EOF, putting the connection into a hung state. 612 as stream EOF, putting the connection into a hung state.
29 (reported and analyzed in an absolutely exemplary manner by Adam 613 (reported and analysed in an absolutely exemplary manner by Adam
30 Rosenstein). 614 Rosenstein).
31 - fix a potential segfault when the TLS context 615 - fix a potential segfault when the TLS context
32 would go missing in a read callback (e.g. due to stoptls) - 616 would go missing in a read callback (e.g. due to stoptls) -
33 Net::SSLeay of course makes no type checking whatsoever. 617 Net::SSLeay of course makes no type checking whatsoever.
34 - AnyEvent::Handle will not stop the read watcher in TLS mode. 618 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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