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

Comparing AnyEvent/Changes (file contents):
Revision 1.216 by root, Sun Feb 8 00:20:53 2009 UTC vs.
Revision 1.530 by root, Tue Mar 27 16:21:11 2012 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: document TCP_*** constants
4TODO: add debug/trace logging to some modules
5TODO: AnyEvent::IO
6
7 - documented Rocco fabricating statements about AnyEvent and me.
8 - the default logging level was not properly documented in a variety of
9 places, this has hopefully been rectified.
10 - updated uts46data.pl for unicode 6.1.0.
11 - made log messages genearted by AnyEvent submodules not
12 include the package name anymore, as it will be logged
13 by default already.
14 - upgrade to the trick used by common-sense 3.5 to work
15 around extra warning torture/breakage under perl 5.15.x.
16
176.14 Tue Jan 31 20:00:24 CET 2012
18 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
19 probably an editing mistake (reported by Darin McBride).
20
216.13 Thu Jan 12 07:27:01 CET 2012
22 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
23 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
24 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
25 regex-matching inside the callback would change the parameter, despite
26 $1 being dynamically scoped per-block. this perl bug is now being
27 worked around (testcase by Cindy Wang).
28
296.12 Mon Dec 12 13:21:10 CET 2011
30 - $! was clobbered by subsequent calls in tcp_connect, due to
31 the postpone not saving/restoring it (analyzed by Richard Garnish).
32
336.11 Tue Nov 22 10:36:05 CET 2011
34 - Tk cannot create windows when tainted unless you set a title, so
35 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
36 - escape any nonprintable/nonascii characters when stringifying
37 backtraces.
38 - log the reason loading an interface module fails to load at level debug.
39
406.1 Tue Oct 4 19:44:30 CEST 2011
41 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
42 and above), and some messages inside AnyEvent have been elevated
43 to higher log levels to print by default.
44 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
45 - work around yet another signal race bug, newly introduced by POE. My
46 how that module sucks.
47 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
48 support ipv6-only operation even with defaults (also use ::1 instead
49 of 127.0.0.1).
50 - give better guidelines on how to chose a logging level.
51 - AnyEvent::Log can now cap (limit) log message priority by context.
52 - AE::log will now save and restore $! and $@, so callers don't have to.
53 - AE::Strict now checks that registered fds are still valid at regular
54 intervals.
55 - skip frequently called 'require' calls if the module appears loaded.
56 - increase signal latency test timer considerably for poor little osx,
57 which can't be bothered with basic posix functionality such as signal
58 delivery.
59 - mention some debugging aids in the tutorial.
60 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
61 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
62 of the context as title.
63 - AnyEvent::Strict would not restore read-write access to $_ when a callback
64 died - this is a user bug, but since AE::Strict exists to help dbeugging...
65 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
66 to work with it - the author is well aware of the issues but refuses
67 to fix them. This extreme step has been taken because the module actively
68 attacks the goals of AnyEvent and makes it harder for module authors to use
69 AnyEvent.
70
716.02 Fri Aug 26 20:08:31 CEST 2011
72 - log_to_syslog now accepts facility strings.
73 - log_to_syslog would not allow facilities - "$facility|$level"
74 *literally* means that string. sheesh.
75 - the CPAN parser got confused and thought we provide Sys::Syslog, try
76 to work around.
77 - renamed AnyEvent::Impl::FLTK2 to FLTK.
78
796.01 Fri Aug 26 07:04:11 CEST 2011
80 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
81 the TTL as fourth element - this affects ->resolve, ->request
82 and ->dns_unpack indirectly. Convenience functions such as
83 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
84 AnyEvent::Socket.
85 - INCOMPATIBLE CHANGE: backend authors now should not implement
86 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
87 methods.
88 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
89 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
90 - DEPRECIATION: please change
91 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
92 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
93 in your programs, for future compatibility.
94 - *blush* idle watcher emulation was borked.
95 - AnyEvent::Strict now checks the AE::xxx functions as well.
96 - querying too long domain names would barf deep inside AnyEvent::DNS - since
97 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
98 for too long domain names explicitly and passes an undef to the callback
99 instead (reported by James Bromberger).
100 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
101 - new functions: AnyEvent::postpone, AnyEvent::log.
102 - new module: AnyEvent::Log, for simple logging needs.
103 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
104 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
105 a debug shell.
106 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
107 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
108 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
109 /etc/hosts can be found.
110 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
111 configuration.
112 - env variables can now be specified using AE_xyz instead of
113 PERL_ANYEVENT_xyz, subject to some rules.
114 - lots of new AnyEvent::Debug functionality.
115 - AnyEvent::Strict now checks for modifications $_ by the callback
116 (by setting it to readonly while executing the callback, if possible).
117 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
118 Invest extra overhead to work around this peculiar behaviour.
119 - split the pure perl backend into a loop module and an impl
120 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
121 - some read types (regex, netstring, json, storable) could
122 errornously keep an AE::Handle object alive on their own
123 (reported by Mohammad Toossi).
124 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
125 supports /etc/hosts overrides.
126 - add more workarounds around child watcher bugs in IO::Async
127 and (newly introduced) in POE.
128 - use glib child watchers and try to work around its limitations,
129 if possible (based on a patch by Kevin Ryde).
130 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
131 - many read types in AnyEvent::Handle would malfunction during
132 transitions to TLS because of a "delete $self->{rbuf}".
133 (reported by Gerald Galster).
134 - improve condvar blocking wait performance (probably only noticable
135 with faster backends), also make it easier for event loops that
136 do not support blocking waits.
137 - PERL_ANYEVENT_MODEL now also allows full module names, although
138 the usefulness of this feature is not yet clear.
139 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
140 - add t/80_ssltext.t.
141 - no longer use AUTOLOAD for the initial loading (for no very good reason).
142 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
143 - do not run event loop detection multiple times when programs call their cached
144 copies of AnyEvent::detect.
145 - parse_hostport now accepts absolute paths als unix domain sockets.
146 - suppress the idiotic warning inside POE by patching the POE::Kernel
147 run flags instead of asking the user to jump through contortions.
148 - add Task::Weaken as dependency (as recommends only at this time).
149 - replace PF_UNSPEC by 0, for good old minix.
150
1516.0 (not officially released, but was on CPAN for about 15 minutes :/)
152
1535.34 Thu May 12 09:59:41 CEST 2011
154 - AE::TLS didn't support both cert_file and key at the same time,
155 due to some copy&paste bug (reported and analyzed by vti).
156 - improve AE::Handle timeout documentation slightly.
157
1585.33 Mon Feb 7 21:51:41 CET 2011
159 - never appeared on CPAN, for unknown reasons.
160 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
161 object when it's already gone (analyzed by Ryan Bullock).
162 - update idna algorithm(s) to UTS#46 version 6.0.0.
163 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
164 uts46 says it's an error, and testsuite makes more sense.
165 - remove leading dots in idn_to_ascii - testsuite says to remove,
166 uts46 says it's an error, and testsuite makes more sense.
167 - fixed a bug in idn ignored character processing.
168 - passes the uts46 6.0.0 testsuite except for three cases which
169 apparently are bugs in the testsuite itself.
170 - passes the uts46 6.0.1 testsuite except for eleven cases which
171 apparently are bugs in the testsuite itself.
172
1735.31 Sun Jan 23 11:49:19 CET 2011
174 - work around a POE bug where POE would first poll for new events and
175 invoke handlers BEFORE handling the message queue.
176 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
177 - stop_read now does just that in TLS mode. see its
178 documentation for caveats.
179 - update to common::sense 3.4 warning set (some warnings in perl impose
180 an arbitrary 32kb limit on regex matches, which affects
181 AnyEvent::Handle).
182 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
183 - new experimental AE::Handle::resettls method.
184
1855.3 Fri Dec 31 05:48:13 CET 2010
186 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
187 issues with IO::Async have been solved in that version, including
188 the availabilty of a default loop!
189 - implement new wbuf_max parameter for AnyEvent::Handle.
190 - added Cocoa::EventLoop backend by Daisuke Murase.
191 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
192 - clarify handle on_prepare documentation.
193 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
194 to avoid leaking them to any child processes.
195 - try to force files into the arch-dependend lib directory, for the
196 half-"turly obsessive compulsive person"(s) that depend on it.
197
1985.29 Sun Dec 5 10:49:21 CET 2010
199 - convert EV backend to EV 4.00 API (so better upgrade EV too).
200 - AE::Handle: implement dynamic read_size adjustment, add
201 max_read_size parameter, reduce default read_size to 2048.
202 - add some FAQ entries for servers and/or on_eof vs. on_error.
203 - work around OS bugs (cygwin again) when lingering in AE::Handle.
204 - reduce memory usage slightly when connect was used in AE::Handle.
205 - wrok around more segfaults in Tk.
206 - document tls_ctx => undef as valid.
207 - detect WNOHANG value at compiletime.
208 - include some probably linux-only support for building AnyEvent
209 as part of the perl core.
210 - improve compatibility of t/08_idna.t to perl 5.8.
211 - make AnyEvent::Strict truly optional (in case somebody builds a
212 minimal perl...).
213
2145.28 Wed Oct 13 04:14:23 CEST 2010
215 - due to a glitch, AnyEvent's internal getprotobyname wasn't
216 actually used in tcp_server.
217 - implement AnyEvent::CondVar->new method, which was documented but
218 missing.
219 - support autoloading of read types in unshift_read.
220 - AnyEvent::Strict: do the test for invalid fh arguments differently,
221 resulting in more informative error messages.
222 - new AnyEvent::FAQ manpage.
223 - clarify that you can actually call push_read etc. in on_read.
224 - put keys and certs in /tmp, not into the current directory,
225 and document this in the SECURITY CONSIDERATIONS sections
226 of AnyEvent::TLS.
227 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
228 requested by Ron Isaacson.
229 - document $AnyEvent::DNS::RESOLVER variable.
230 - incorporated major typo patches by Abhijit Menon-Sen.
231 - AnyEvent::Handle now croaks when negative timeouts are passed.
232 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
233 undocumented, but who knows when they might come in handy.
234
2355.271 Tue Jun 8 12:05:46 CEST 2010
236 - backport to perl 5.8.x.
237
2385.27 Sun Jun 6 12:12:05 CEST 2010
239 - postpone differently in AnyEvent::Socket now, as
240 when not, canceling the connection attempt might fail
241 (found by Felix Antonius Wilhelm Ostmann).
242 - explicitly check for non-stream sockets in AE::Handle, too many
243 clueless people fell into the trap of this somehow working.
244 - simplified and reworked the "OTHER MODULES" section.
245 - better/more condvar examples.
246
2475.261 Wed Apr 28 16:13:36 CEST 2010
248 - AF_INET6 was not properly used from Socket6 during configuration
249 time (found by Andy Grundman).
250
2515.26 Mon Apr 12 04:49:35 CEST 2010
252 - don't generate spurious readyness notifications when select
253 returns EINTR in the pure perl backend (analysed and testcase
254 by Bas Denissen).
255 - give same examples in SYNOPSIS sections of AnyEvent and AE.
256 - provider faster implementations for AE::time/now/now_update for
257 EV and Perl backends.
258 - provide faster AE::cv and AE::time implementations for all
259 backends.
260 - no longer support lower/mixed-case signal names (this was
261 never documented nor universally supported).
262 - some more memory µ-optimisations, and somewhat less messy
263 code to cope with both AnyEvent and AE APIs.
264
2655.251 Sat Mar 13 00:58:21 CET 2010
266 - make test failed if it was run before make install on perl <=
267 5.8.8, and older versions of perl make our live very difficult,
268 so write constants.pl during Makefile.PL time.
269
2705.25 Sat Mar 13 00:23:14 CET 2010
271 - fix a race condition in AnyEvent::Handle that would cause
272 a "bio_read: unsupported method" error (in _another_
273 ssl connection) after user code threw an exception.
274 - added AnyEvent::Handle->destroyed method.
275 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
276 - the AnyEvent::Util::fh_nonblocking function worked only
277 by lucky accident on win32.
278 - smaller and faster AnyEvent::Util::fh_nonblocking.
279 - when the (required!) Time::HiRes module is missing, AnyEvent
280 did not fall back to built-in time properly.
281 - do not load Fcntl at runtime, saving memory and loading time.
282 - precompile a number of constants and use them instead of runtime
283 detection and eval.
284 - free detection code after detection and similar memory optimisations.
285 - Perl backend timer interval best effort drift has been improved
286 (same algorithm as EV).
287 - update unicode idna mapping table.
288
2895.24 Tue Jan 5 11:39:43 CET 2010
290 - cygwin never reports errors from failed connects. 1.5 just gives
291 you continous readyness and EAGAIN, 1.7 is even more broken and
292 just hangs. work around both issues in a cygwin-specific hack.
293 - improve idle watcher documentation slightly.
294
2955.23 Sun Dec 20 23:48:00 CET 2009
296 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
297 - implement punycode_encode/decode, idn_nameprep,
298 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
299 - provide $AE::VERSION.
300 - removed traces of "no strict 'refs'".
301
3025.22 Sat Dec 5 03:51:13 CET 2009
303 - downgrade-or-fail in AnyEvent::Handle::push_write, to
304 diagnose encoding failures earlier and more succinctly.
305 (this works around bugs in perl, throwing away encoding info
306 when passing scalar data to extensions).
307 - add more examples to AnyEvent::Socket manpage.
308 - upgrade internal warning set to the same as common::sense 2.03.
309 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
310 (John Beppu).
311 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
312 - add google public dns servers to fallback server set and make sure
313 we load-balance properly between the three sets. also add all
314 fallback dns servers, not just a random one, to each dns config.
315
3165.21 Thu Nov 19 02:48:47 CET 2009
317 - fix a problem where socket constants were called with parameters
318 (spotted by David Friedland).
319 - fork_call never use'd POSIX (reported by Daisuke Maki).
320 - improve perl 5.6 compatibility further (but it still won't work
321 unless you rip out everything but the core).
322 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
323 uses 10MB of RAM, but doesn't open a console window. *sigh*.
324
3255.202 Wed Oct 14 22:35:44 CEST 2009
326 - AnyEvent::DNS would unexpectedly clobber $_ under windows
327 (analysed by Matthias Waldorf).
328 - AnyEvent::Handle::run_cmd can now pass the PID of the
329 newly-created process, which is much less useful than it might
330 sound (based on patch by Yann Kerherve).
331
3325.201 Tue Sep 29 12:09:25 CEST 2009
333 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
334 (reported by Torsten Foertsch).
335 - common::sense 2.0 could cause tcp_server to throw an exception
336 (analysed by elmex).
337
3385.2 Mon Sep 14 07:04:49 CEST 2009
339 - INCOMPATIBLE CHANGE: do no longer support register_read_type
340 and register_write_type in AnyEvent::Handle, instead support
341 package names (the facility was mostly abused).
342 - implement "packagename-as-read/write type" support in
343 AnyEvent::Handle.
344 - AnyEvent::Handle: new options "keepalive" and "oobinline".
345 - oobinline set by default to avoid security issues.
346 - the pure-perl event loop backend wrongly detected times() fallback
347 support (spotted by Pavel Boldin).
348
3495.12 Tue Sep 1 20:26:50 CEST 2009
350 - be more lenient when parsing resolv.conf files, as some people
351 use hashmarks as comment indicator inside directives (reported
352 by Michael S. Fischer).
353 - use same set of warnings as common::sense 2.0.
354 - fix a potential 32 bit overflow issue due to perl having problems
355 with large hex constants in the Perl backend.
356 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
357 by default.
358 - fix some prototypes.
359 - (optionally) require Async::Interrupt 1.02.
360 - replace opendns fallbacks by verizon/level3 public dns servers.
361
3625.112 Fri Aug 21 13:59:04 CEST 2009
363 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
364 using starttls with an illegal/uncreatable tls context.
365 - port to broken win32 perls.
366
3675.111 Sun Aug 16 18:44:36 CEST 2009
368 - perl errornously requires sockaddr_un structures returned by the
369 kernel to have a specific length. work around this bug by padding
370 sockaddr_un structures (found to fail on OS X, as perl only uses
371 the correct code for GNU/Linux, but likely to fail on other
372 OSes as well).
373
3745.11 Wed Aug 12 17:49:37 CEST 2009
375 - fixed a bug in run_cmd with input from filename.
376 - create a json coder object when none is given in write direction
377 to reduce dependency on JSON.
378
3795.1 Tue Aug 11 03:17:32 CEST 2009
380 - tcp_connect now reports the correct error in $!, not an unrelated
381 one (was broken since 4.91).
382 - AE::Handle did not properly clear rtimeout/wtimeout watchers
383 when retrying a tcp_connect.
384 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
385 - fixed wrong documentation in AE::Socket.
386
3875.01 Mon Aug 10 03:16:32 CEST 2009
388 - last release broke EV child watchers :/.
389
3905.0 Sun Aug 9 17:32:09 CEST 2009
391 - officially introduce and document the AE API.
392 - lowering the timeout at runtime did not have immediate effect
393 in AnyEvent::Handle.
394 - AE::Handle now has separate and independent read and write timeout
395 settings.
396 - probe for Guard module when AnyEvent::Util is loaded, not at
397 runtime (Event doesn't like eval "use" from callbacks, and it avoids
398 exporting the probe function).
399 - Impl::IOAsync's io watchers did not use the filehandle emulation
400 code to work around it's watcher limits.
401 - work around IO::Async's problems with overloaded objects.
402 - try to disable wearnings inside IO::Async, it's just too buggy.
403 - do no longer use Event by default, only EV or Perl, Event is too buggy.
404 - hackishly provide fast AE:: interface when EV or Perl is the backend.
405 - inherit VERSION from AnyEvent::VERSION in most modules.
406
4074.91 Thu Aug 6 15:42:45 CEST 2009
408 - AE::Handle::starttls could get out of sync when the read buffer
409 already contains some TLS handshake.
410 - AE::Handle did not properly free the TLS session, causing bigger
411 memory leaks in Net::SSLeay than expected (reported by toaster).
412 - AE::Socket::tcp_connect will now artificially delay invoking
413 the callback to avoid returning after invoking the callback.
414 - convert many internal watcher uses to AE API.
415
4164.9 Sat Aug 1 11:07:01 CEST 2009
417 - Glib silently fails when registering a timeout with
418 a negative value, so avoid that.
419 - call condvar callback immediately when it is set
420 after the condition is already signalled.
421 - check rbuf_max condition only after trying to consume
422 data.
423 - normalise signal numbers to names when using the ae's
424 signal handling, but do not document this (yet).
425 - pure perl signal emulation did not properly set nonblocking
426 mode on the signal pipe on win32 (this is unlikely to have caused
427 any issues).
428 - new module: AnyEvent::Debug.
429 - AnyEvent::Strict now checks that a signal was specified
430 by name (not name or number).
431 - reduce memory footprint in typical cases by ~50kb by
432 only compiling signal code when necessary.
433 - add AnyEvent::Handle->rbuf_max.
434 - grab the AE namespace for future new API, implement stubs for
435 said future/alternative API.
436 - new function: AnyEvent::Socket::format_hostport.
437
4384.881 Tue Jul 28 12:51:53 CEST 2009
439 - work around a bug in local in pre-5.10 perls, causing
440 AnyEvent::Handle to recurse when it should not
441 (analysed by elmex).
442
4434.88 Tue Jul 28 04:04:37 CEST 2009
444 - re-bless the handle into a dummy package after calling
445 AnyEvent::Handle::destroy, so the user does not need to check
446 for errors after every push_write etc.
447 - do not attempt to run t/02_signals.t on obviously broken
448 platforms.
449
4504.87 Sun Jul 26 02:06:16 CEST 2009
451 - do not attempt to linger when there is no longer a valid fh.
452 - enforce tls mode to be either accept or connect, do not simply
453 segfault in Net::SSLeay.
454 - AnyEvent::Handle can now call tcp_connect itself (new parameters
455 connect, on_prepare, on_connect and on_connect_error). Updated
456 tutorial accordingly.
457 - add AnyEvent::Impl::Irssi backend.
458
4594.86 Mon Jul 20 23:52:29 CEST 2009
460 - since the verbose warning is not enough, explicitly document
461 that versions before 1.33 of Net::SSLeay are not secure.
462 - work around signal handling races in Event and (...) Event::Lib.
463 - try to align signal-race timer to full-second boundaries.
464 - work around Tk not liking negative timeouts.
465 - don't complain of different grades of environmental unfriendlyness
466 in IO::Async.
467
4684.85 Sat Jul 18 06:16:14 CEST 2009
469 - nail the signal race problem in perl once and for all
470 (see $AnyEvent::MAX_SIGNAL_LATENCY).
471 - take advantage of Async::Interrupt if it is available.
472 - load Time::HiRes and Guard modules on demand only.
473 - add optional/recommended modules section to AnyEvent
474 documentation.
475 - reduce memory usage considerably (and reduce startup penalty)
476 by not using "strict", "warnings" and "overload" modules.
477 - work around buggy windows/openbsd perls and provide EBADMSG
478 and EPROTO ourselves when missing.
479 - improve perl 5.6 compatibility of the core event loop.
480 - made Net::SSLeay version 1.33 a soft requirement.
481
4824.83 Fri Jul 17 16:56:26 CEST 2009
483 - implement AnyEvent::Socket::getprotobyname.
484 - AnyEvent::CondVar's will now detect recursive blocking
485 waits and will croak, as too many people fall into
486 this trap.
487 - AnyEvent::Handle will now call ->destroy on itself after
488 executing the on_error callback, instead of doing some
489 half-baked internal shutdown, for fatal errors.
490 - clarify on_eof behaviour w.r.t. the read queue and
491 on_read callbacks.
492 - ignore some possible spurious wake-ups in tcp_connect.
493
4944.82 Sat Jul 11 00:34:55 CEST 2009
495 - POE and Event backends didn't accept some callable objects as
496 callbacks.
497 - use Config module instead of POSIX module to detect signal names
498 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
499 module doesn't even have all POSIX signals :/.
500 - use more workarounds around the many refcnt/corruption bugs in
501 Event::Lib.
502 - work around a race condition in perl's select, causing t/03_child.t
503 to rarely fail.
504
5054.81 Thu Jul 9 10:30:30 CEST 2009
506 - AnyEvent::Handle didn't properly diagnose write errors
507 (it expected -1 from syswrite, how lame... :).
508 - support file descriptors in addition to file handles
509 in AnyEvent->io.
510 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
511 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
512 and PERL_ANYEVENT_CA_PATH.
513 - provide a sensible synopsis section for AnyEvent::TLS.
514 - add a "supported backends" section to the manpage.
515 - added simple io watcher test to testsuite, using a
516 portable_socketpair.
517 - tried to improve the stability of the Event::Lib backend,
518 YMMV.
519
5204.8 Mon Jul 6 23:45:16 CEST 2009
521 - AnyEvent::DNS did not properly follow CNAME records with
522 uppercase targets.
523 - AnyEvent::DNS would errornously return AAAA records
524 with v4 mapped addresses (a faulty record) as ipv4 addresses,
525 causing AnyEvent::Socket to throw an exception.
526 - added new module AnyEvent::TLS for easier SSL/TLS context
527 creation, with many options including hostname verification,
528 secure default configuration, lots of documentation and,
529 predefined diffie-hellman keys for perfect forward security
530 and much more. get it while it's still fresh!
531 - use AnyEvent::TLS in AnyEvent::Handle for context management.
532 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
533 so AnyEvent::Socket users have smaller memory footprint
534 in the common case.
535 - add AnyEvent::Handle->push_shutdown method.
536 - add an additional error message parameter to AnyEvent::Handle's
537 on_error callback (for TLS, $! is still available).
538 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
539 - make AnyEvent::Handle more robust against early conenction
540 failures (during new), and return C<undef> in that case
541 from the constructor.
542 - AnyEvent::Handle will now try to load only JSON::XS first,
543 then fall back to JSON.
544 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
545 and because it was documented that way.
546
5474.451 Fri Jul 3 00:28:58 CEST 2009
548 - do not clear rbuf when shutting down an AnyEvent::Handle
549 object - doing so breaks AnyEvent::HTTP.
550
5514.45 Mon Jun 29 22:59:26 CEST 2009
552 - a write error could cause AnyEvent::Handle to create
553 an I/O watcher with an undefined $fh.
554 - special-case mapped ipv4 addresses in both
555 AnyEvent::Socket::format_address and parse_address,
556 to treat them just like ipv4 addresses.
557 - updated and overhauled the AnyEvent::Intro doc.
558 - implement AnyEvent::Socket::format_ipv4/ipv6.
559 - slightly speed up portable_pipe/socketpair.
560 - expand condvar begin/end documentation.
561
5624.42 Fri Jun 26 08:32:18 CEST 2009
563 - preliminary and neccesarily incomplete support for IO::Async.
564 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
565 case it was set to IGNORE, to ensure we can catch child statuses
566 even when the calling env acts stupidly.
567 - updated benchmarks with IO::Async, which performs very well.
568
5694.412 Wed Jun 24 01:35:57 CEST 2009
570 - support an "untaint" attribute for AnyEvent::DNS
571 and set it on the default resolver.
572 - implement and document AnyEvent::Impl::Perl::loop.
573 - remove all anyevent-env variables from %ENV
574 when running in tainted mode.
575 - mention and extend the IO::Lambda benchmark.
576
5774.411 Sun Jun 7 18:48:02 CEST 2009
578 - do not try to use F_SETFD on windows.
579
5804.41 Thu May 14 06:40:11 CEST 2009
581 - work around issues in older perls (5.8.0?) when
582 a signal handler is deleted from the %SIG hash.
583 - use POSIX::_exit in child test, to avoid running
584 destructors.
585 - speed up CHLD handling by relying on SIGCHLD being
586 synchronously delivered, even when we roll our own
587 implementation.
588 - AnyEvent::DNS: add the "dname" resource record name
589 for cosmetic reasons.
590
5914.4 Sun Apr 26 20:12:33 CEST 2009
592 - implemented idle watchers, where applicable.
593 - AnyEvent->time died when Event backend was in use.
594 - fix a memleak in the Tk backend.
595 - sped up Tk timer handling.
596 - clip negative "after" values to 0 in AnyEvent::Impl::Event
597 to avoid spamming out warning messages.
598 - fix Qt timers without interval.
599 - avoid Qt zero-timeout special case, in old libqt's.
600
6014.352 Mon Apr 20 16:31:11 CEST 2009
602 - fix AnyEvent::Strict error messages for child watchers.
603 - fix/update Makefile.PL "recommends" versions.
604 - add AnyEvent->now_update.
605
6064.351 Sat Apr 11 07:56:14 CEST 2009
607 - actually make the signal pipe work under win32.
608 - localise $! in signal handler to avoid changing $!, although
609 perl itself does not seem to save/restore errno either.
610 - set the cloexec flag on the signal pipe (normally set by perl too).
611
6124.35 Fri Mar 27 11:48:20 CET 2009
613 - event models relying on AnyEvent's signal watcher emulation did
614 invoke the callback asynchronously, contradicting documentation
615 and causing signals to get lost (this includes AnyEvent's own
616 event loop). AnyEvent now uses the standard pipe trick to make
617 callback execution synchronous to the event loop.
618 - AnyEvent::Handle didn't free TLS context data on DESTROY
619 (patch by Pavel Shaydo).
620 - work around the ever-incompatibly-changing API of MakeMaker.
621 - document that changing global variables without restoring them
622 is a bad idea in Perl (noted by Adam Rosenstein).
623 - AnyEvent::Strict now barfs if ->io is passed a file.
624
6254.34 Thu Feb 12 18:32:45 CET 2009
626 - separately buffer TLS read data, as otherwise the read queue could
627 deadlock as receiving data is not expected while draining the
628 read queue (which cna only happen with TLS).
3 - raise EBADMSG error on JSON decoding errors. 629 - raise EBADMSG error on JSON decoding errors.
4 - fix some minor manpage bugs (reported by Maximilian Gaß). 630 - fix some minor manpage bugs (reported by Maximilian Gaß).
5 - speed up select bitmask parsing quite a bit in the pure perl 631 - speed up select bitmask parsing quite a bit in the pure perl
6 backend. 632 backend.
7 - use CORE::select instead of just select in AnyEvent::Impl::Perl. 633 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
… …
424.3 Fri Oct 3 09:18:43 CEST 2008 6684.3 Fri Oct 3 09:18:43 CEST 2008
43 - AnyEvent will now install a no-op signal handler for SIGPIPE, 669 - AnyEvent will now install a no-op signal handler for SIGPIPE,
44 unless one has been installed already. 670 unless one has been installed already.
45 - warn about not loaidng AnyEvent::Impl::POE early enough 671 - warn about not loaidng AnyEvent::Impl::POE early enough
46 (patch by Adam Rosenstein). 672 (patch by Adam Rosenstein).
47 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 673 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
48 - fix a bug in where in SSL connect mode, the 674 - fix a bug in where in SSL connect mode, the
49 client would first wait for some data by the server and 675 client would first wait for some data by the server and
50 otherwise hang. (reported and analyzed in an absolutely 676 otherwise hang. (reported and analysed in an absolutely
51 exemplary manner by Adam Rosenstein). 677 exemplary manner by Adam Rosenstein).
52 - fix a bug in where SSL EOF would not be treated 678 - fix a bug in where SSL EOF would not be treated
53 as stream EOF, putting the connection into a hung state. 679 as stream EOF, putting the connection into a hung state.
54 (reported and analyzed in an absolutely exemplary manner by Adam 680 (reported and analysed in an absolutely exemplary manner by Adam
55 Rosenstein). 681 Rosenstein).
56 - fix a potential segfault when the TLS context 682 - fix a potential segfault when the TLS context
57 would go missing in a read callback (e.g. due to stoptls) - 683 would go missing in a read callback (e.g. due to stoptls) -
58 Net::SSLeay of course makes no type checking whatsoever. 684 Net::SSLeay of course makes no type checking whatsoever.
59 - AnyEvent::Handle will not stop the read watcher in TLS mode. 685 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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