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

Comparing AnyEvent/Changes (file contents):
Revision 1.235 by root, Wed May 6 03:45:27 2009 UTC vs.
Revision 1.522 by root, Tue Jan 10 13:43:22 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
5
66.13
7 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
8 regex-matching inside the callback would change the parameter, despite
9 $1 being dynamically scoped per-block. this perl bug is now being
10 worked around (testcase by Cindy Wang).
11
126.12 Mon Dec 12 13:21:10 CET 2011
13 - $! was clobbered by subsequent calls in tcp_connect, due to
14 the postpone not saving/restoring it (analyzed by Richard Garnish).
15
166.11 Tue Nov 22 10:36:05 CET 2011
17 - Tk cannot create windows when tainted unless you set a title, so
18 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
19 - escape any nonprintable/nonascii characters when stringifying
20 backtraces.
21 - log the reason loading an interface module fails to load at level debug.
22
236.1 Tue Oct 4 19:44:30 CEST 2011
24 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
25 and above), and some messages inside AnyEvent have been elevated
26 to higher log levels to print by default.
27 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
28 - work around yet another signal race bug, newly introduced by POE. My
29 how that module sucks.
30 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
31 support ipv6-only operation even with defaults (also use ::1 instead
32 of 127.0.0.1).
33 - give better guidelines on how to chose a logging level.
34 - AnyEvent::Log can now cap (limit) log message priority by context.
35 - AE::log will now save and restore $! and $@, so callers don't have to.
36 - AE::Strict now checks that registered fds are still valid at regular
37 intervals.
38 - skip frequently called 'require' calls if the module appears loaded.
39 - increase signal latency test timer considerably for poor little osx,
40 which can't be bothered with basic posix functionality such as signal
41 delivery.
42 - mention some debugging aids in the tutorial.
43 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
44 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
45 of the context as title.
46 - AnyEvent::Strict would not restore read-write access to $_ when a callback
47 died - this is a user bug, but since AE::Strict exists to help dbeugging...
48 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
49 to work with it - the author is well aware of the issues but refuses
50 to fix them. This extreme step has been taken because the module actively
51 attacks the goals of AnyEvent and makes it harder for module authors to use
52 AnyEvent.
53
546.02 Fri Aug 26 20:08:31 CEST 2011
55 - log_to_syslog now accepts facility strings.
56 - log_to_syslog would not allow facilities - "$facility|$level"
57 *literally* means that string. sheesh.
58 - the CPAN parser got confused and thought we provide Sys::Syslog, try
59 to work around.
60 - renamed AnyEvent::Impl::FLTK2 to FLTK.
61
626.01 Fri Aug 26 07:04:11 CEST 2011
63 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
64 the TTL as fourth element - this affects ->resolve, ->request
65 and ->dns_unpack indirectly. Convenience functions such as
66 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
67 AnyEvent::Socket.
68 - INCOMPATIBLE CHANGE: backend authors now should not implement
69 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
70 methods.
71 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
72 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
73 - DEPRECIATION: please change
74 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
75 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
76 in your programs, for future compatibility.
77 - *blush* idle watcher emulation was borked.
78 - AnyEvent::Strict now checks the AE::xxx functions as well.
79 - querying too long domain names would barf deep inside AnyEvent::DNS - since
80 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
81 for too long domain names explicitly and passes an undef to the callback
82 instead (reported by James Bromberger).
83 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
84 - new functions: AnyEvent::postpone, AnyEvent::log.
85 - new module: AnyEvent::Log, for simple logging needs.
86 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
87 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
88 a debug shell.
89 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
90 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
91 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
92 /etc/hosts can be found.
93 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
94 configuration.
95 - env variables can now be specified using AE_xyz instead of
96 PERL_ANYEVENT_xyz, subject to some rules.
97 - lots of new AnyEvent::Debug functionality.
98 - AnyEvent::Strict now checks for modifications $_ by the callback
99 (by setting it to readonly while executing the callback, if possible).
100 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
101 Invest extra overhead to work around this peculiar behaviour.
102 - split the pure perl backend into a loop module and an impl
103 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
104 - some read types (regex, netstring, json, storable) could
105 errornously keep an AE::Handle object alive on their own
106 (reported by Mohammad Toossi).
107 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
108 supports /etc/hosts overrides.
109 - add more workarounds around child watcher bugs in IO::Async
110 and (newly introduced) in POE.
111 - use glib child watchers and try to work around its limitations,
112 if possible (based on a patch by Kevin Ryde).
113 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
114 - many read types in AnyEvent::Handle would malfunction during
115 transitions to TLS because of a "delete $self->{rbuf}".
116 (reported by Gerald Galster).
117 - improve condvar blocking wait performance (probably only noticable
118 with faster backends), also make it easier for event loops that
119 do not support blocking waits.
120 - PERL_ANYEVENT_MODEL now also allows full module names, although
121 the usefulness of this feature is not yet clear.
122 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
123 - add t/80_ssltext.t.
124 - no longer use AUTOLOAD for the initial loading (for no very good reason).
125 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
126 - do not run event loop detection multiple times when programs call their cached
127 copies of AnyEvent::detect.
128 - parse_hostport now accepts absolute paths als unix domain sockets.
129 - suppress the idiotic warning inside POE by patching the POE::Kernel
130 run flags instead of asking the user to jump through contortions.
131 - add Task::Weaken as dependency (as recommends only at this time).
132 - replace PF_UNSPEC by 0, for good old minix.
133
1346.0 (not officially released, but was on CPAN for about 15 minutes :/)
135
1365.34 Thu May 12 09:59:41 CEST 2011
137 - AE::TLS didn't support both cert_file and key at the same time,
138 due to some copy&paste bug (reported and analyzed by vti).
139 - improve AE::Handle timeout documentation slightly.
140
1415.33 Mon Feb 7 21:51:41 CET 2011
142 - never appeared on CPAN, for unknown reasons.
143 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
144 object when it's already gone (analyzed by Ryan Bullock).
145 - update idna algorithm(s) to UTS#46 version 6.0.0.
146 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
147 uts46 says it's an error, and testsuite makes more sense.
148 - remove leading dots in idn_to_ascii - testsuite says to remove,
149 uts46 says it's an error, and testsuite makes more sense.
150 - fixed a bug in idn ignored character processing.
151 - passes the uts46 6.0.0 testsuite except for three cases which
152 apparently are bugs in the testsuite itself.
153 - passes the uts46 6.0.1 testsuite except for eleven cases which
154 apparently are bugs in the testsuite itself.
155
1565.31 Sun Jan 23 11:49:19 CET 2011
157 - work around a POE bug where POE would first poll for new events and
158 invoke handlers BEFORE handling the message queue.
159 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
160 - stop_read now does just that in TLS mode. see its
161 documentation for caveats.
162 - update to common::sense 3.4 warning set (some warnings in perl impose
163 an arbitrary 32kb limit on regex matches, which affects
164 AnyEvent::Handle).
165 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
166 - new experimental AE::Handle::resettls method.
167
1685.3 Fri Dec 31 05:48:13 CET 2010
169 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
170 issues with IO::Async have been solved in that version, including
171 the availabilty of a default loop!
172 - implement new wbuf_max parameter for AnyEvent::Handle.
173 - added Cocoa::EventLoop backend by Daisuke Murase.
174 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
175 - clarify handle on_prepare documentation.
176 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
177 to avoid leaking them to any child processes.
178 - try to force files into the arch-dependend lib directory, for the
179 half-"turly obsessive compulsive person"(s) that depend on it.
180
1815.29 Sun Dec 5 10:49:21 CET 2010
182 - convert EV backend to EV 4.00 API (so better upgrade EV too).
183 - AE::Handle: implement dynamic read_size adjustment, add
184 max_read_size parameter, reduce default read_size to 2048.
185 - add some FAQ entries for servers and/or on_eof vs. on_error.
186 - work around OS bugs (cygwin again) when lingering in AE::Handle.
187 - reduce memory usage slightly when connect was used in AE::Handle.
188 - wrok around more segfaults in Tk.
189 - document tls_ctx => undef as valid.
190 - detect WNOHANG value at compiletime.
191 - include some probably linux-only support for building AnyEvent
192 as part of the perl core.
193 - improve compatibility of t/08_idna.t to perl 5.8.
194 - make AnyEvent::Strict truly optional (in case somebody builds a
195 minimal perl...).
196
1975.28 Wed Oct 13 04:14:23 CEST 2010
198 - due to a glitch, AnyEvent's internal getprotobyname wasn't
199 actually used in tcp_server.
200 - implement AnyEvent::CondVar->new method, which was documented but
201 missing.
202 - support autoloading of read types in unshift_read.
203 - AnyEvent::Strict: do the test for invalid fh arguments differently,
204 resulting in more informative error messages.
205 - new AnyEvent::FAQ manpage.
206 - clarify that you can actually call push_read etc. in on_read.
207 - put keys and certs in /tmp, not into the current directory,
208 and document this in the SECURITY CONSIDERATIONS sections
209 of AnyEvent::TLS.
210 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
211 requested by Ron Isaacson.
212 - document $AnyEvent::DNS::RESOLVER variable.
213 - incorporated major typo patches by Abhijit Menon-Sen.
214 - AnyEvent::Handle now croaks when negative timeouts are passed.
215 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
216 undocumented, but who knows when they might come in handy.
217
2185.271 Tue Jun 8 12:05:46 CEST 2010
219 - backport to perl 5.8.x.
220
2215.27 Sun Jun 6 12:12:05 CEST 2010
222 - postpone differently in AnyEvent::Socket now, as
223 when not, canceling the connection attempt might fail
224 (found by Felix Antonius Wilhelm Ostmann).
225 - explicitly check for non-stream sockets in AE::Handle, too many
226 clueless people fell into the trap of this somehow working.
227 - simplified and reworked the "OTHER MODULES" section.
228 - better/more condvar examples.
229
2305.261 Wed Apr 28 16:13:36 CEST 2010
231 - AF_INET6 was not properly used from Socket6 during configuration
232 time (found by Andy Grundman).
233
2345.26 Mon Apr 12 04:49:35 CEST 2010
235 - don't generate spurious readyness notifications when select
236 returns EINTR in the pure perl backend (analysed and testcase
237 by Bas Denissen).
238 - give same examples in SYNOPSIS sections of AnyEvent and AE.
239 - provider faster implementations for AE::time/now/now_update for
240 EV and Perl backends.
241 - provide faster AE::cv and AE::time implementations for all
242 backends.
243 - no longer support lower/mixed-case signal names (this was
244 never documented nor universally supported).
245 - some more memory ยต-optimisations, and somewhat less messy
246 code to cope with both AnyEvent and AE APIs.
247
2485.251 Sat Mar 13 00:58:21 CET 2010
249 - make test failed if it was run before make install on perl <=
250 5.8.8, and older versions of perl make our live very difficult,
251 so write constants.pl during Makefile.PL time.
252
2535.25 Sat Mar 13 00:23:14 CET 2010
254 - fix a race condition in AnyEvent::Handle that would cause
255 a "bio_read: unsupported method" error (in _another_
256 ssl connection) after user code threw an exception.
257 - added AnyEvent::Handle->destroyed method.
258 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
259 - the AnyEvent::Util::fh_nonblocking function worked only
260 by lucky accident on win32.
261 - smaller and faster AnyEvent::Util::fh_nonblocking.
262 - when the (required!) Time::HiRes module is missing, AnyEvent
263 did not fall back to built-in time properly.
264 - do not load Fcntl at runtime, saving memory and loading time.
265 - precompile a number of constants and use them instead of runtime
266 detection and eval.
267 - free detection code after detection and similar memory optimisations.
268 - Perl backend timer interval best effort drift has been improved
269 (same algorithm as EV).
270 - update unicode idna mapping table.
271
2725.24 Tue Jan 5 11:39:43 CET 2010
273 - cygwin never reports errors from failed connects. 1.5 just gives
274 you continous readyness and EAGAIN, 1.7 is even more broken and
275 just hangs. work around both issues in a cygwin-specific hack.
276 - improve idle watcher documentation slightly.
277
2785.23 Sun Dec 20 23:48:00 CET 2009
279 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
280 - implement punycode_encode/decode, idn_nameprep,
281 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
282 - provide $AE::VERSION.
283 - removed traces of "no strict 'refs'".
284
2855.22 Sat Dec 5 03:51:13 CET 2009
286 - downgrade-or-fail in AnyEvent::Handle::push_write, to
287 diagnose encoding failures earlier and more succinctly.
288 (this works around bugs in perl, throwing away encoding info
289 when passing scalar data to extensions).
290 - add more examples to AnyEvent::Socket manpage.
291 - upgrade internal warning set to the same as common::sense 2.03.
292 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
293 (John Beppu).
294 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
295 - add google public dns servers to fallback server set and make sure
296 we load-balance properly between the three sets. also add all
297 fallback dns servers, not just a random one, to each dns config.
298
2995.21 Thu Nov 19 02:48:47 CET 2009
300 - fix a problem where socket constants were called with parameters
301 (spotted by David Friedland).
302 - fork_call never use'd POSIX (reported by Daisuke Maki).
303 - improve perl 5.6 compatibility further (but it still won't work
304 unless you rip out everything but the core).
305 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
306 uses 10MB of RAM, but doesn't open a console window. *sigh*.
307
3085.202 Wed Oct 14 22:35:44 CEST 2009
309 - AnyEvent::DNS would unexpectedly clobber $_ under windows
310 (analysed by Matthias Waldorf).
311 - AnyEvent::Handle::run_cmd can now pass the PID of the
312 newly-created process, which is much less useful than it might
313 sound (based on patch by Yann Kerherve).
314
3155.201 Tue Sep 29 12:09:25 CEST 2009
316 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
317 (reported by Torsten Foertsch).
318 - common::sense 2.0 could cause tcp_server to throw an exception
319 (analysed by elmex).
320
3215.2 Mon Sep 14 07:04:49 CEST 2009
322 - INCOMPATIBLE CHANGE: do no longer support register_read_type
323 and register_write_type in AnyEvent::Handle, instead support
324 package names (the facility was mostly abused).
325 - implement "packagename-as-read/write type" support in
326 AnyEvent::Handle.
327 - AnyEvent::Handle: new options "keepalive" and "oobinline".
328 - oobinline set by default to avoid security issues.
329 - the pure-perl event loop backend wrongly detected times() fallback
330 support (spotted by Pavel Boldin).
331
3325.12 Tue Sep 1 20:26:50 CEST 2009
333 - be more lenient when parsing resolv.conf files, as some people
334 use hashmarks as comment indicator inside directives (reported
335 by Michael S. Fischer).
336 - use same set of warnings as common::sense 2.0.
337 - fix a potential 32 bit overflow issue due to perl having problems
338 with large hex constants in the Perl backend.
339 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
340 by default.
341 - fix some prototypes.
342 - (optionally) require Async::Interrupt 1.02.
343 - replace opendns fallbacks by verizon/level3 public dns servers.
344
3455.112 Fri Aug 21 13:59:04 CEST 2009
346 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
347 using starttls with an illegal/uncreatable tls context.
348 - port to broken win32 perls.
349
3505.111 Sun Aug 16 18:44:36 CEST 2009
351 - perl errornously requires sockaddr_un structures returned by the
352 kernel to have a specific length. work around this bug by padding
353 sockaddr_un structures (found to fail on OS X, as perl only uses
354 the correct code for GNU/Linux, but likely to fail on other
355 OSes as well).
356
3575.11 Wed Aug 12 17:49:37 CEST 2009
358 - fixed a bug in run_cmd with input from filename.
359 - create a json coder object when none is given in write direction
360 to reduce dependency on JSON.
361
3625.1 Tue Aug 11 03:17:32 CEST 2009
363 - tcp_connect now reports the correct error in $!, not an unrelated
364 one (was broken since 4.91).
365 - AE::Handle did not properly clear rtimeout/wtimeout watchers
366 when retrying a tcp_connect.
367 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
368 - fixed wrong documentation in AE::Socket.
369
3705.01 Mon Aug 10 03:16:32 CEST 2009
371 - last release broke EV child watchers :/.
372
3735.0 Sun Aug 9 17:32:09 CEST 2009
374 - officially introduce and document the AE API.
375 - lowering the timeout at runtime did not have immediate effect
376 in AnyEvent::Handle.
377 - AE::Handle now has separate and independent read and write timeout
378 settings.
379 - probe for Guard module when AnyEvent::Util is loaded, not at
380 runtime (Event doesn't like eval "use" from callbacks, and it avoids
381 exporting the probe function).
382 - Impl::IOAsync's io watchers did not use the filehandle emulation
383 code to work around it's watcher limits.
384 - work around IO::Async's problems with overloaded objects.
385 - try to disable wearnings inside IO::Async, it's just too buggy.
386 - do no longer use Event by default, only EV or Perl, Event is too buggy.
387 - hackishly provide fast AE:: interface when EV or Perl is the backend.
388 - inherit VERSION from AnyEvent::VERSION in most modules.
389
3904.91 Thu Aug 6 15:42:45 CEST 2009
391 - AE::Handle::starttls could get out of sync when the read buffer
392 already contains some TLS handshake.
393 - AE::Handle did not properly free the TLS session, causing bigger
394 memory leaks in Net::SSLeay than expected (reported by toaster).
395 - AE::Socket::tcp_connect will now artificially delay invoking
396 the callback to avoid returning after invoking the callback.
397 - convert many internal watcher uses to AE API.
398
3994.9 Sat Aug 1 11:07:01 CEST 2009
400 - Glib silently fails when registering a timeout with
401 a negative value, so avoid that.
402 - call condvar callback immediately when it is set
403 after the condition is already signalled.
404 - check rbuf_max condition only after trying to consume
405 data.
406 - normalise signal numbers to names when using the ae's
407 signal handling, but do not document this (yet).
408 - pure perl signal emulation did not properly set nonblocking
409 mode on the signal pipe on win32 (this is unlikely to have caused
410 any issues).
411 - new module: AnyEvent::Debug.
412 - AnyEvent::Strict now checks that a signal was specified
413 by name (not name or number).
414 - reduce memory footprint in typical cases by ~50kb by
415 only compiling signal code when necessary.
416 - add AnyEvent::Handle->rbuf_max.
417 - grab the AE namespace for future new API, implement stubs for
418 said future/alternative API.
419 - new function: AnyEvent::Socket::format_hostport.
420
4214.881 Tue Jul 28 12:51:53 CEST 2009
422 - work around a bug in local in pre-5.10 perls, causing
423 AnyEvent::Handle to recurse when it should not
424 (analysed by elmex).
425
4264.88 Tue Jul 28 04:04:37 CEST 2009
427 - re-bless the handle into a dummy package after calling
428 AnyEvent::Handle::destroy, so the user does not need to check
429 for errors after every push_write etc.
430 - do not attempt to run t/02_signals.t on obviously broken
431 platforms.
432
4334.87 Sun Jul 26 02:06:16 CEST 2009
434 - do not attempt to linger when there is no longer a valid fh.
435 - enforce tls mode to be either accept or connect, do not simply
436 segfault in Net::SSLeay.
437 - AnyEvent::Handle can now call tcp_connect itself (new parameters
438 connect, on_prepare, on_connect and on_connect_error). Updated
439 tutorial accordingly.
440 - add AnyEvent::Impl::Irssi backend.
441
4424.86 Mon Jul 20 23:52:29 CEST 2009
443 - since the verbose warning is not enough, explicitly document
444 that versions before 1.33 of Net::SSLeay are not secure.
445 - work around signal handling races in Event and (...) Event::Lib.
446 - try to align signal-race timer to full-second boundaries.
447 - work around Tk not liking negative timeouts.
448 - don't complain of different grades of environmental unfriendlyness
449 in IO::Async.
450
4514.85 Sat Jul 18 06:16:14 CEST 2009
452 - nail the signal race problem in perl once and for all
453 (see $AnyEvent::MAX_SIGNAL_LATENCY).
454 - take advantage of Async::Interrupt if it is available.
455 - load Time::HiRes and Guard modules on demand only.
456 - add optional/recommended modules section to AnyEvent
457 documentation.
458 - reduce memory usage considerably (and reduce startup penalty)
459 by not using "strict", "warnings" and "overload" modules.
460 - work around buggy windows/openbsd perls and provide EBADMSG
461 and EPROTO ourselves when missing.
462 - improve perl 5.6 compatibility of the core event loop.
463 - made Net::SSLeay version 1.33 a soft requirement.
464
4654.83 Fri Jul 17 16:56:26 CEST 2009
466 - implement AnyEvent::Socket::getprotobyname.
467 - AnyEvent::CondVar's will now detect recursive blocking
468 waits and will croak, as too many people fall into
469 this trap.
470 - AnyEvent::Handle will now call ->destroy on itself after
471 executing the on_error callback, instead of doing some
472 half-baked internal shutdown, for fatal errors.
473 - clarify on_eof behaviour w.r.t. the read queue and
474 on_read callbacks.
475 - ignore some possible spurious wake-ups in tcp_connect.
476
4774.82 Sat Jul 11 00:34:55 CEST 2009
478 - POE and Event backends didn't accept some callable objects as
479 callbacks.
480 - use Config module instead of POSIX module to detect signal names
481 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
482 module doesn't even have all POSIX signals :/.
483 - use more workarounds around the many refcnt/corruption bugs in
484 Event::Lib.
485 - work around a race condition in perl's select, causing t/03_child.t
486 to rarely fail.
487
4884.81 Thu Jul 9 10:30:30 CEST 2009
489 - AnyEvent::Handle didn't properly diagnose write errors
490 (it expected -1 from syswrite, how lame... :).
491 - support file descriptors in addition to file handles
492 in AnyEvent->io.
493 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
494 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
495 and PERL_ANYEVENT_CA_PATH.
496 - provide a sensible synopsis section for AnyEvent::TLS.
497 - add a "supported backends" section to the manpage.
498 - added simple io watcher test to testsuite, using a
499 portable_socketpair.
500 - tried to improve the stability of the Event::Lib backend,
501 YMMV.
502
5034.8 Mon Jul 6 23:45:16 CEST 2009
504 - AnyEvent::DNS did not properly follow CNAME records with
505 uppercase targets.
506 - AnyEvent::DNS would errornously return AAAA records
507 with v4 mapped addresses (a faulty record) as ipv4 addresses,
508 causing AnyEvent::Socket to throw an exception.
509 - added new module AnyEvent::TLS for easier SSL/TLS context
510 creation, with many options including hostname verification,
511 secure default configuration, lots of documentation and,
512 predefined diffie-hellman keys for perfect forward security
513 and much more. get it while it's still fresh!
514 - use AnyEvent::TLS in AnyEvent::Handle for context management.
515 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
516 so AnyEvent::Socket users have smaller memory footprint
517 in the common case.
518 - add AnyEvent::Handle->push_shutdown method.
519 - add an additional error message parameter to AnyEvent::Handle's
520 on_error callback (for TLS, $! is still available).
521 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
522 - make AnyEvent::Handle more robust against early conenction
523 failures (during new), and return C<undef> in that case
524 from the constructor.
525 - AnyEvent::Handle will now try to load only JSON::XS first,
526 then fall back to JSON.
527 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
528 and because it was documented that way.
529
5304.451 Fri Jul 3 00:28:58 CEST 2009
531 - do not clear rbuf when shutting down an AnyEvent::Handle
532 object - doing so breaks AnyEvent::HTTP.
533
5344.45 Mon Jun 29 22:59:26 CEST 2009
535 - a write error could cause AnyEvent::Handle to create
536 an I/O watcher with an undefined $fh.
537 - special-case mapped ipv4 addresses in both
538 AnyEvent::Socket::format_address and parse_address,
539 to treat them just like ipv4 addresses.
540 - updated and overhauled the AnyEvent::Intro doc.
541 - implement AnyEvent::Socket::format_ipv4/ipv6.
542 - slightly speed up portable_pipe/socketpair.
543 - expand condvar begin/end documentation.
544
5454.42 Fri Jun 26 08:32:18 CEST 2009
546 - preliminary and neccesarily incomplete support for IO::Async.
547 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
548 case it was set to IGNORE, to ensure we can catch child statuses
549 even when the calling env acts stupidly.
550 - updated benchmarks with IO::Async, which performs very well.
551
5524.412 Wed Jun 24 01:35:57 CEST 2009
553 - support an "untaint" attribute for AnyEvent::DNS
554 and set it on the default resolver.
555 - implement and document AnyEvent::Impl::Perl::loop.
556 - remove all anyevent-env variables from %ENV
557 when running in tainted mode.
558 - mention and extend the IO::Lambda benchmark.
559
5604.411 Sun Jun 7 18:48:02 CEST 2009
561 - do not try to use F_SETFD on windows.
562
5634.41 Thu May 14 06:40:11 CEST 2009
564 - work around issues in older perls (5.8.0?) when
565 a signal handler is deleted from the %SIG hash.
566 - use POSIX::_exit in child test, to avoid running
567 destructors.
568 - speed up CHLD handling by relying on SIGCHLD being
569 synchronously delivered, even when we roll our own
570 implementation.
3 - AnyEvent::DNS: add the "dname" resource record name. 571 - AnyEvent::DNS: add the "dname" resource record name
572 for cosmetic reasons.
4 573
54.4 Sun Apr 26 20:12:33 CEST 2009 5744.4 Sun Apr 26 20:12:33 CEST 2009
6 - implemented idle watchers, where applicable. 575 - implemented idle watchers, where applicable.
7 - AnyEvent->time died when Event backend was in use. 576 - AnyEvent->time died when Event backend was in use.
8 - fix a memleak in the Tk backend. 577 - fix a memleak in the Tk backend.
9 - sped up Tk timer handling. 578 - sped up Tk timer handling.
10 - clip negative "after" values to 0 in AnyEvent::Impl::Event 579 - clip negative "after" values to 0 in AnyEvent::Impl::Event
11 to avoid spamming out warning messages. 580 to avoid spamming out warning messages.
12 - fix Qt timers without interval. 581 - fix Qt timers without interval.
13 - avoid Qt zero-timeout specialcase, in old libqt's. 582 - avoid Qt zero-timeout special case, in old libqt's.
14 583
154.352 Mon Apr 20 16:31:11 CEST 2009 5844.352 Mon Apr 20 16:31:11 CEST 2009
16 - fix AnyEvent::Strict error messages for child watchers. 585 - fix AnyEvent::Strict error messages for child watchers.
17 - fix/update Makefile.PL "recommends" versions. 586 - fix/update Makefile.PL "recommends" versions.
18 - add AnyEvent->now_update. 587 - add AnyEvent->now_update.
… …
824.3 Fri Oct 3 09:18:43 CEST 2008 6514.3 Fri Oct 3 09:18:43 CEST 2008
83 - AnyEvent will now install a no-op signal handler for SIGPIPE, 652 - AnyEvent will now install a no-op signal handler for SIGPIPE,
84 unless one has been installed already. 653 unless one has been installed already.
85 - warn about not loaidng AnyEvent::Impl::POE early enough 654 - warn about not loaidng AnyEvent::Impl::POE early enough
86 (patch by Adam Rosenstein). 655 (patch by Adam Rosenstein).
87 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 656 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
88 - fix a bug in where in SSL connect mode, the 657 - fix a bug in where in SSL connect mode, the
89 client would first wait for some data by the server and 658 client would first wait for some data by the server and
90 otherwise hang. (reported and analyzed in an absolutely 659 otherwise hang. (reported and analysed in an absolutely
91 exemplary manner by Adam Rosenstein). 660 exemplary manner by Adam Rosenstein).
92 - fix a bug in where SSL EOF would not be treated 661 - fix a bug in where SSL EOF would not be treated
93 as stream EOF, putting the connection into a hung state. 662 as stream EOF, putting the connection into a hung state.
94 (reported and analyzed in an absolutely exemplary manner by Adam 663 (reported and analysed in an absolutely exemplary manner by Adam
95 Rosenstein). 664 Rosenstein).
96 - fix a potential segfault when the TLS context 665 - fix a potential segfault when the TLS context
97 would go missing in a read callback (e.g. due to stoptls) - 666 would go missing in a read callback (e.g. due to stoptls) -
98 Net::SSLeay of course makes no type checking whatsoever. 667 Net::SSLeay of course makes no type checking whatsoever.
99 - AnyEvent::Handle will not stop the read watcher in TLS mode. 668 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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