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 1.552 by root, Sat May 12 23:14:29 2012 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
3TODO: document TCP_*** constants
4TODO: add debug/trace logging to some modules
5TODO: use the eval-with-cb-call trick also to speed up JSON?
6TODO: common pitfalls? invoke-before-return
7TODO: AnyEvent::Handle with AnyEvent::IO?
8
97.01 Sun May 13 01:03:17 CEST 2012
10 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
11 not available, instead of throwing an exception.
12 - use File::Spec to get the tmpdir in t/*, to avoid needless
13 failures on (most, not mine :) windows boxes.
14 - new handle read types: tls_detect and tls_autostart.
15
167.0 Fri Apr 13 06:33:30 CEST 2012
17 - child watchers are broken in POE 1.352 (also many earlier
18 versions) and there seems to be no way to work around it, as POE
19 itself is inherently racy. Document this regression and add a delay
20 in t/68_poe_03_child.t for the time being.
21 - new module AnyEvent::IO, that is a frontend to either a pure-perl
22 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
23 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
24 - load /etc/hosts only when DNS has no answer.
25 - stat /etc/hosts on every access and reload it if it changed.
26 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
27 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
28 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
29 a side effect, they now use true appending as opposed to libc
30 appending, and the file might not have been opened when the function
31 returns.
32 - the default logging level was not properly documented in a variety of
33 places, this has hopefully been rectified.
34 - updated uts46data.pl for unicode 6.1.0.
35 - made log messages generated by AnyEvent submodules not
36 include the package name anymore, as it will be logged
37 by default already.
38 - upgrade to the trick used by common-sense 3.5 to work
39 around extra warning torture/breakage under perl 5.15.x.
40 - log messages by anyevent are now ucfirst, are usually full
41 sentences and do no longer include the package name.
42 - the storable read type would sometimes throw an exception instead
43 of causing EBADMSG (analyzed by Maxime Soulé).
44 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
45 keysize (reported by Paul Howarth).
46 - documented Rocco fabricating statements about AnyEvent and me.
47
486.14 Tue Jan 31 20:00:24 CET 2012
49 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
50 probably an editing mistake (reported by Darin McBride).
51
526.13 Thu Jan 12 07:27:01 CET 2012
53 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
54 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
55 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
56 regex-matching inside the callback would change the parameter, despite
57 $1 being dynamically scoped per-block. this perl bug is now being
58 worked around (testcase by Cindy Wang).
59
606.12 Mon Dec 12 13:21:10 CET 2011
61 - $! was clobbered by subsequent calls in tcp_connect, due to
62 the postpone not saving/restoring it (analyzed by Richard Garnish).
63
646.11 Tue Nov 22 10:36:05 CET 2011
65 - Tk cannot create windows when tainted unless you set a title, so
66 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
67 - escape any nonprintable/nonascii characters when stringifying
68 backtraces.
69 - log the reason loading an interface module fails to load at level debug.
70
716.1 Tue Oct 4 19:44:30 CEST 2011
72 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
73 and above), and some messages inside AnyEvent have been elevated
74 to higher log levels to print by default.
75 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
76 - work around yet another signal race bug, newly introduced by POE. My
77 how that module sucks.
78 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
79 support ipv6-only operation even with defaults (also use ::1 instead
80 of 127.0.0.1).
81 - give better guidelines on how to chose a logging level.
82 - AnyEvent::Log can now cap (limit) log message priority by context.
83 - AE::log will now save and restore $! and $@, so callers don't have to.
84 - AE::Strict now checks that registered fds are still valid at regular
85 intervals.
86 - skip frequently called 'require' calls if the module appears loaded.
87 - increase signal latency test timer considerably for poor little osx,
88 which can't be bothered with basic posix functionality such as signal
89 delivery.
90 - mention some debugging aids in the tutorial.
91 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
92 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
93 of the context as title.
94 - AnyEvent::Strict would not restore read-write access to $_ when a callback
95 died - this is a user bug, but since AE::Strict exists to help dbeugging...
96 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
97 to work with it - the author is well aware of the issues but refuses
98 to fix them. This extreme step has been taken because the module actively
99 attacks the goals of AnyEvent and makes it harder for module authors to use
100 AnyEvent.
101
1026.02 Fri Aug 26 20:08:31 CEST 2011
103 - log_to_syslog now accepts facility strings.
104 - log_to_syslog would not allow facilities - "$facility|$level"
105 *literally* means that string. sheesh.
106 - the CPAN parser got confused and thought we provide Sys::Syslog, try
107 to work around.
108 - renamed AnyEvent::Impl::FLTK2 to FLTK.
109
1106.01 Fri Aug 26 07:04:11 CEST 2011
111 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
112 the TTL as fourth element - this affects ->resolve, ->request
113 and ->dns_unpack indirectly. Convenience functions such as
114 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
115 AnyEvent::Socket.
116 - INCOMPATIBLE CHANGE: backend authors now should not implement
117 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
118 methods.
119 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
120 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
121 - DEPRECIATION: please change
122 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
123 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
124 in your programs, for future compatibility.
125 - *blush* idle watcher emulation was borked.
126 - AnyEvent::Strict now checks the AE::xxx functions as well.
127 - querying too long domain names would barf deep inside AnyEvent::DNS - since
128 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
129 for too long domain names explicitly and passes an undef to the callback
130 instead (reported by James Bromberger).
131 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
132 - new functions: AnyEvent::postpone, AnyEvent::log.
133 - new module: AnyEvent::Log, for simple logging needs.
134 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
135 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
136 a debug shell.
137 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
138 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
139 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
140 /etc/hosts can be found.
141 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
142 configuration.
143 - env variables can now be specified using AE_xyz instead of
144 PERL_ANYEVENT_xyz, subject to some rules.
145 - lots of new AnyEvent::Debug functionality.
146 - AnyEvent::Strict now checks for modifications of $_ by the callback
147 (by setting it to readonly while executing the callback, if possible).
148 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
149 Invest extra overhead to work around this peculiar behaviour.
150 - split the pure perl backend into a loop module and an impl
151 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
152 - some read types (regex, netstring, json, storable) could
153 errornously keep an AE::Handle object alive on their own
154 (reported by Mohammad Toossi).
155 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
156 supports /etc/hosts overrides.
157 - add more workarounds around child watcher bugs in IO::Async
158 and (newly introduced) in POE.
159 - use glib child watchers and try to work around its limitations,
160 if possible (based on a patch by Kevin Ryde).
161 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
162 - many read types in AnyEvent::Handle would malfunction during
163 transitions to TLS because of a "delete $self->{rbuf}".
164 (reported by Gerald Galster).
165 - improve condvar blocking wait performance (probably only noticable
166 with faster backends), also make it easier for event loops that
167 do not support blocking waits.
168 - PERL_ANYEVENT_MODEL now also allows full module names, although
169 the usefulness of this feature is not yet clear.
170 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
171 - add t/80_ssltext.t.
172 - no longer use AUTOLOAD for the initial loading (for no very good reason).
173 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
174 - do not run event loop detection multiple times when programs call their cached
175 copies of AnyEvent::detect.
176 - parse_hostport now accepts absolute paths als unix domain sockets.
177 - suppress the idiotic warning inside POE by patching the POE::Kernel
178 run flags instead of asking the user to jump through contortions.
179 - add Task::Weaken as dependency (as recommends only at this time).
180 - replace PF_UNSPEC by 0, for good old minix.
181
1826.0 (not officially released, but was on CPAN for about 15 minutes :/)
183
1845.34 Thu May 12 09:59:41 CEST 2011
185 - AE::TLS didn't support both cert_file and key at the same time,
186 due to some copy&paste bug (reported and analyzed by vti).
187 - improve AE::Handle timeout documentation slightly.
188
1895.33 Mon Feb 7 21:51:41 CET 2011
190 - never appeared on CPAN, for unknown reasons.
191 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
192 object when it's already gone (analyzed by Ryan Bullock).
193 - update idna algorithm(s) to UTS#46 version 6.0.0.
194 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
195 uts46 says it's an error, and testsuite makes more sense.
196 - remove leading dots in idn_to_ascii - testsuite says to remove,
197 uts46 says it's an error, and testsuite makes more sense.
198 - fixed a bug in idn ignored character processing.
199 - passes the uts46 6.0.0 testsuite except for three cases which
200 apparently are bugs in the testsuite itself.
201 - passes the uts46 6.0.1 testsuite except for eleven cases which
202 apparently are bugs in the testsuite itself.
203
2045.31 Sun Jan 23 11:49:19 CET 2011
205 - work around a POE bug where POE would first poll for new events and
206 invoke handlers BEFORE handling the message queue.
207 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
208 - stop_read now does just that in TLS mode. see its
209 documentation for caveats.
210 - update to common::sense 3.4 warning set (some warnings in perl impose
211 an arbitrary 32kb limit on regex matches, which affects
212 AnyEvent::Handle).
213 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
214 - new experimental AE::Handle::resettls method.
215
2165.3 Fri Dec 31 05:48:13 CET 2010
217 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
218 issues with IO::Async have been solved in that version, including
219 the availabilty of a default loop!
220 - implement new wbuf_max parameter for AnyEvent::Handle.
221 - added Cocoa::EventLoop backend by Daisuke Murase.
222 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
223 - clarify handle on_prepare documentation.
224 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
225 to avoid leaking them to any child processes.
226 - try to force files into the arch-dependend lib directory, for the
227 half-"turly obsessive compulsive person"(s) that depend on it.
228
2295.29 Sun Dec 5 10:49:21 CET 2010
230 - convert EV backend to EV 4.00 API (so better upgrade EV too).
231 - AE::Handle: implement dynamic read_size adjustment, add
232 max_read_size parameter, reduce default read_size to 2048.
233 - add some FAQ entries for servers and/or on_eof vs. on_error.
234 - work around OS bugs (cygwin again) when lingering in AE::Handle.
235 - reduce memory usage slightly when connect was used in AE::Handle.
236 - wrok around more segfaults in Tk.
237 - document tls_ctx => undef as valid.
238 - detect WNOHANG value at compiletime.
239 - include some probably linux-only support for building AnyEvent
240 as part of the perl core.
241 - improve compatibility of t/08_idna.t to perl 5.8.
242 - make AnyEvent::Strict truly optional (in case somebody builds a
243 minimal perl...).
244
2455.28 Wed Oct 13 04:14:23 CEST 2010
246 - due to a glitch, AnyEvent's internal getprotobyname wasn't
247 actually used in tcp_server.
248 - implement AnyEvent::CondVar->new method, which was documented but
249 missing.
250 - support autoloading of read types in unshift_read.
251 - AnyEvent::Strict: do the test for invalid fh arguments differently,
252 resulting in more informative error messages.
253 - new AnyEvent::FAQ manpage.
254 - clarify that you can actually call push_read etc. in on_read.
255 - put keys and certs in /tmp, not into the current directory,
256 and document this in the SECURITY CONSIDERATIONS sections
257 of AnyEvent::TLS.
258 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
259 requested by Ron Isaacson.
260 - document $AnyEvent::DNS::RESOLVER variable.
261 - incorporated major typo patches by Abhijit Menon-Sen.
262 - AnyEvent::Handle now croaks when negative timeouts are passed.
263 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
264 undocumented, but who knows when they might come in handy.
265
2665.271 Tue Jun 8 12:05:46 CEST 2010
267 - backport to perl 5.8.x.
268
2695.27 Sun Jun 6 12:12:05 CEST 2010
270 - postpone differently in AnyEvent::Socket now, as
271 when not, canceling the connection attempt might fail
272 (found by Felix Antonius Wilhelm Ostmann).
273 - explicitly check for non-stream sockets in AE::Handle, too many
274 clueless people fell into the trap of this somehow working.
275 - simplified and reworked the "OTHER MODULES" section.
276 - better/more condvar examples.
277
2785.261 Wed Apr 28 16:13:36 CEST 2010
279 - AF_INET6 was not properly used from Socket6 during configuration
280 time (found by Andy Grundman).
281
2825.26 Mon Apr 12 04:49:35 CEST 2010
283 - don't generate spurious readyness notifications when select
284 returns EINTR in the pure perl backend (analysed and testcase
285 by Bas Denissen).
286 - give same examples in SYNOPSIS sections of AnyEvent and AE.
287 - provider faster implementations for AE::time/now/now_update for
288 EV and Perl backends.
289 - provide faster AE::cv and AE::time implementations for all
290 backends.
291 - no longer support lower/mixed-case signal names (this was
292 never documented nor universally supported).
293 - some more memory µ-optimisations, and somewhat less messy
294 code to cope with both AnyEvent and AE APIs.
295
2965.251 Sat Mar 13 00:58:21 CET 2010
297 - make test failed if it was run before make install on perl <=
298 5.8.8, and older versions of perl make our live very difficult,
299 so write constants.pl during Makefile.PL time.
300
3015.25 Sat Mar 13 00:23:14 CET 2010
302 - fix a race condition in AnyEvent::Handle that would cause
303 a "bio_read: unsupported method" error (in _another_
304 ssl connection) after user code threw an exception.
305 - added AnyEvent::Handle->destroyed method.
306 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
307 - the AnyEvent::Util::fh_nonblocking function worked only
308 by lucky accident on win32.
309 - smaller and faster AnyEvent::Util::fh_nonblocking.
310 - when the (required!) Time::HiRes module is missing, AnyEvent
311 did not fall back to built-in time properly.
312 - do not load Fcntl at runtime, saving memory and loading time.
313 - precompile a number of constants and use them instead of runtime
314 detection and eval.
315 - free detection code after detection and similar memory optimisations.
316 - Perl backend timer interval best effort drift has been improved
317 (same algorithm as EV).
318 - update unicode idna mapping table.
319
3205.24 Tue Jan 5 11:39:43 CET 2010
321 - cygwin never reports errors from failed connects. 1.5 just gives
322 you continous readyness and EAGAIN, 1.7 is even more broken and
323 just hangs. work around both issues in a cygwin-specific hack.
324 - improve idle watcher documentation slightly.
325
3265.23 Sun Dec 20 23:48:00 CET 2009
327 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
328 - implement punycode_encode/decode, idn_nameprep,
329 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
330 - provide $AE::VERSION.
331 - removed traces of "no strict 'refs'".
332
3335.22 Sat Dec 5 03:51:13 CET 2009
334 - downgrade-or-fail in AnyEvent::Handle::push_write, to
335 diagnose encoding failures earlier and more succinctly.
336 (this works around bugs in perl, throwing away encoding info
337 when passing scalar data to extensions).
338 - add more examples to AnyEvent::Socket manpage.
339 - upgrade internal warning set to the same as common::sense 2.03.
340 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
341 (John Beppu).
342 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
343 - add google public dns servers to fallback server set and make sure
344 we load-balance properly between the three sets. also add all
345 fallback dns servers, not just a random one, to each dns config.
346
3475.21 Thu Nov 19 02:48:47 CET 2009
348 - fix a problem where socket constants were called with parameters
349 (spotted by David Friedland).
350 - fork_call never use'd POSIX (reported by Daisuke Maki).
351 - improve perl 5.6 compatibility further (but it still won't work
352 unless you rip out everything but the core).
353 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
354 uses 10MB of RAM, but doesn't open a console window. *sigh*.
355
3565.202 Wed Oct 14 22:35:44 CEST 2009
357 - AnyEvent::DNS would unexpectedly clobber $_ under windows
358 (analysed by Matthias Waldorf).
359 - AnyEvent::Handle::run_cmd can now pass the PID of the
360 newly-created process, which is much less useful than it might
361 sound (based on patch by Yann Kerherve).
362
3635.201 Tue Sep 29 12:09:25 CEST 2009
364 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
365 (reported by Torsten Foertsch).
366 - common::sense 2.0 could cause tcp_server to throw an exception
367 (analysed by elmex).
368
3695.2 Mon Sep 14 07:04:49 CEST 2009
370 - INCOMPATIBLE CHANGE: do no longer support register_read_type
371 and register_write_type in AnyEvent::Handle, instead support
372 package names (the facility was mostly abused).
373 - implement "packagename-as-read/write type" support in
374 AnyEvent::Handle.
375 - AnyEvent::Handle: new options "keepalive" and "oobinline".
376 - oobinline set by default to avoid security issues.
377 - the pure-perl event loop backend wrongly detected times() fallback
378 support (spotted by Pavel Boldin).
379
3805.12 Tue Sep 1 20:26:50 CEST 2009
381 - be more lenient when parsing resolv.conf files, as some people
382 use hashmarks as comment indicator inside directives (reported
383 by Michael S. Fischer).
384 - use same set of warnings as common::sense 2.0.
385 - fix a potential 32 bit overflow issue due to perl having problems
386 with large hex constants in the Perl backend.
387 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
388 by default.
389 - fix some prototypes.
390 - (optionally) require Async::Interrupt 1.02.
391 - replace opendns fallbacks by verizon/level3 public dns servers.
392
3935.112 Fri Aug 21 13:59:04 CEST 2009
394 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
395 using starttls with an illegal/uncreatable tls context.
396 - port to broken win32 perls.
397
3985.111 Sun Aug 16 18:44:36 CEST 2009
399 - perl errornously requires sockaddr_un structures returned by the
400 kernel to have a specific length. work around this bug by padding
401 sockaddr_un structures (found to fail on OS X, as perl only uses
402 the correct code for GNU/Linux, but likely to fail on other
403 OSes as well).
404
4055.11 Wed Aug 12 17:49:37 CEST 2009
406 - fixed a bug in run_cmd with input from filename.
407 - create a json coder object when none is given in write direction
408 to reduce dependency on JSON.
409
4105.1 Tue Aug 11 03:17:32 CEST 2009
411 - tcp_connect now reports the correct error in $!, not an unrelated
412 one (was broken since 4.91).
413 - AE::Handle did not properly clear rtimeout/wtimeout watchers
414 when retrying a tcp_connect.
415 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
416 - fixed wrong documentation in AE::Socket.
417
4185.01 Mon Aug 10 03:16:32 CEST 2009
419 - last release broke EV child watchers :/.
420
4215.0 Sun Aug 9 17:32:09 CEST 2009
422 - officially introduce and document the AE API.
423 - lowering the timeout at runtime did not have immediate effect
424 in AnyEvent::Handle.
425 - AE::Handle now has separate and independent read and write timeout
426 settings.
427 - probe for Guard module when AnyEvent::Util is loaded, not at
428 runtime (Event doesn't like eval "use" from callbacks, and it avoids
429 exporting the probe function).
430 - Impl::IOAsync's io watchers did not use the filehandle emulation
431 code to work around it's watcher limits.
432 - work around IO::Async's problems with overloaded objects.
433 - try to disable wearnings inside IO::Async, it's just too buggy.
434 - do no longer use Event by default, only EV or Perl, Event is too buggy.
435 - hackishly provide fast AE:: interface when EV or Perl is the backend.
436 - inherit VERSION from AnyEvent::VERSION in most modules.
437
4384.91 Thu Aug 6 15:42:45 CEST 2009
439 - AE::Handle::starttls could get out of sync when the read buffer
440 already contains some TLS handshake.
441 - AE::Handle did not properly free the TLS session, causing bigger
442 memory leaks in Net::SSLeay than expected (reported by toaster).
443 - AE::Socket::tcp_connect will now artificially delay invoking
444 the callback to avoid returning after invoking the callback.
445 - convert many internal watcher uses to AE API.
446
4474.9 Sat Aug 1 11:07:01 CEST 2009
448 - Glib silently fails when registering a timeout with
449 a negative value, so avoid that.
450 - call condvar callback immediately when it is set
451 after the condition is already signalled.
452 - check rbuf_max condition only after trying to consume
453 data.
454 - normalise signal numbers to names when using the ae's
455 signal handling, but do not document this (yet).
456 - pure perl signal emulation did not properly set nonblocking
457 mode on the signal pipe on win32 (this is unlikely to have caused
458 any issues).
459 - new module: AnyEvent::Debug.
460 - AnyEvent::Strict now checks that a signal was specified
461 by name (not name or number).
462 - reduce memory footprint in typical cases by ~50kb by
463 only compiling signal code when necessary.
464 - add AnyEvent::Handle->rbuf_max.
465 - grab the AE namespace for future new API, implement stubs for
466 said future/alternative API.
467 - new function: AnyEvent::Socket::format_hostport.
468
4694.881 Tue Jul 28 12:51:53 CEST 2009
470 - work around a bug in local in pre-5.10 perls, causing
471 AnyEvent::Handle to recurse when it should not
472 (analysed by elmex).
473
4744.88 Tue Jul 28 04:04:37 CEST 2009
475 - re-bless the handle into a dummy package after calling
476 AnyEvent::Handle::destroy, so the user does not need to check
477 for errors after every push_write etc.
478 - do not attempt to run t/02_signals.t on obviously broken
479 platforms.
480
4814.87 Sun Jul 26 02:06:16 CEST 2009
482 - do not attempt to linger when there is no longer a valid fh.
483 - enforce tls mode to be either accept or connect, do not simply
484 segfault in Net::SSLeay.
485 - AnyEvent::Handle can now call tcp_connect itself (new parameters
486 connect, on_prepare, on_connect and on_connect_error). Updated
487 tutorial accordingly.
488 - add AnyEvent::Impl::Irssi backend.
489
4904.86 Mon Jul 20 23:52:29 CEST 2009
491 - since the verbose warning is not enough, explicitly document
492 that versions before 1.33 of Net::SSLeay are not secure.
493 - work around signal handling races in Event and (...) Event::Lib.
494 - try to align signal-race timer to full-second boundaries.
495 - work around Tk not liking negative timeouts.
496 - don't complain of different grades of environmental unfriendlyness
497 in IO::Async.
498
4994.85 Sat Jul 18 06:16:14 CEST 2009
500 - nail the signal race problem in perl once and for all
501 (see $AnyEvent::MAX_SIGNAL_LATENCY).
502 - take advantage of Async::Interrupt if it is available.
503 - load Time::HiRes and Guard modules on demand only.
504 - add optional/recommended modules section to AnyEvent
505 documentation.
506 - reduce memory usage considerably (and reduce startup penalty)
507 by not using "strict", "warnings" and "overload" modules.
508 - work around buggy windows/openbsd perls and provide EBADMSG
509 and EPROTO ourselves when missing.
510 - improve perl 5.6 compatibility of the core event loop.
511 - made Net::SSLeay version 1.33 a soft requirement.
512
5134.83 Fri Jul 17 16:56:26 CEST 2009
514 - implement AnyEvent::Socket::getprotobyname.
515 - AnyEvent::CondVar's will now detect recursive blocking
516 waits and will croak, as too many people fall into
517 this trap.
518 - AnyEvent::Handle will now call ->destroy on itself after
519 executing the on_error callback, instead of doing some
520 half-baked internal shutdown, for fatal errors.
521 - clarify on_eof behaviour w.r.t. the read queue and
522 on_read callbacks.
523 - ignore some possible spurious wake-ups in tcp_connect.
524
5254.82 Sat Jul 11 00:34:55 CEST 2009
526 - POE and Event backends didn't accept some callable objects as
527 callbacks.
528 - use Config module instead of POSIX module to detect signal names
529 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
530 module doesn't even have all POSIX signals :/.
531 - use more workarounds around the many refcnt/corruption bugs in
532 Event::Lib.
533 - work around a race condition in perl's select, causing t/03_child.t
534 to rarely fail.
535
5364.81 Thu Jul 9 10:30:30 CEST 2009
537 - AnyEvent::Handle didn't properly diagnose write errors
538 (it expected -1 from syswrite, how lame... :).
539 - support file descriptors in addition to file handles
540 in AnyEvent->io.
541 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
542 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
543 and PERL_ANYEVENT_CA_PATH.
544 - provide a sensible synopsis section for AnyEvent::TLS.
545 - add a "supported backends" section to the manpage.
546 - added simple io watcher test to testsuite, using a
547 portable_socketpair.
548 - tried to improve the stability of the Event::Lib backend,
549 YMMV.
550
5514.8 Mon Jul 6 23:45:16 CEST 2009
552 - AnyEvent::DNS did not properly follow CNAME records with
553 uppercase targets.
554 - AnyEvent::DNS would errornously return AAAA records
555 with v4 mapped addresses (a faulty record) as ipv4 addresses,
556 causing AnyEvent::Socket to throw an exception.
557 - added new module AnyEvent::TLS for easier SSL/TLS context
558 creation, with many options including hostname verification,
559 secure default configuration, lots of documentation and,
560 predefined diffie-hellman keys for perfect forward security
561 and much more. get it while it's still fresh!
562 - use AnyEvent::TLS in AnyEvent::Handle for context management.
563 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
564 so AnyEvent::Socket users have smaller memory footprint
565 in the common case.
566 - add AnyEvent::Handle->push_shutdown method.
567 - add an additional error message parameter to AnyEvent::Handle's
568 on_error callback (for TLS, $! is still available).
569 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
570 - make AnyEvent::Handle more robust against early conenction
571 failures (during new), and return C<undef> in that case
572 from the constructor.
573 - AnyEvent::Handle will now try to load only JSON::XS first,
574 then fall back to JSON.
575 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
576 and because it was documented that way.
577
5784.451 Fri Jul 3 00:28:58 CEST 2009
579 - do not clear rbuf when shutting down an AnyEvent::Handle
580 object - doing so breaks AnyEvent::HTTP.
581
5824.45 Mon Jun 29 22:59:26 CEST 2009
583 - a write error could cause AnyEvent::Handle to create
584 an I/O watcher with an undefined $fh.
585 - special-case mapped ipv4 addresses in both
586 AnyEvent::Socket::format_address and parse_address,
587 to treat them just like ipv4 addresses.
588 - updated and overhauled the AnyEvent::Intro doc.
589 - implement AnyEvent::Socket::format_ipv4/ipv6.
590 - slightly speed up portable_pipe/socketpair.
591 - expand condvar begin/end documentation.
592
5934.42 Fri Jun 26 08:32:18 CEST 2009
594 - preliminary and neccesarily incomplete support for IO::Async.
595 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
596 case it was set to IGNORE, to ensure we can catch child statuses
597 even when the calling env acts stupidly.
598 - updated benchmarks with IO::Async, which performs very well.
599
6004.412 Wed Jun 24 01:35:57 CEST 2009
601 - support an "untaint" attribute for AnyEvent::DNS
602 and set it on the default resolver.
603 - implement and document AnyEvent::Impl::Perl::loop.
604 - remove all anyevent-env variables from %ENV
605 when running in tainted mode.
606 - mention and extend the IO::Lambda benchmark.
607
6084.411 Sun Jun 7 18:48:02 CEST 2009
609 - do not try to use F_SETFD on windows.
610
6114.41 Thu May 14 06:40:11 CEST 2009
612 - work around issues in older perls (5.8.0?) when
613 a signal handler is deleted from the %SIG hash.
614 - use POSIX::_exit in child test, to avoid running
615 destructors.
616 - speed up CHLD handling by relying on SIGCHLD being
617 synchronously delivered, even when we roll our own
618 implementation.
619 - AnyEvent::DNS: add the "dname" resource record name
620 for cosmetic reasons.
621
6224.4 Sun Apr 26 20:12:33 CEST 2009
623 - implemented idle watchers, where applicable.
624 - AnyEvent->time died when Event backend was in use.
625 - fix a memleak in the Tk backend.
626 - sped up Tk timer handling.
627 - clip negative "after" values to 0 in AnyEvent::Impl::Event
628 to avoid spamming out warning messages.
629 - fix Qt timers without interval.
630 - avoid Qt zero-timeout special case, in old libqt's.
631
6324.352 Mon Apr 20 16:31:11 CEST 2009
633 - fix AnyEvent::Strict error messages for child watchers.
634 - fix/update Makefile.PL "recommends" versions.
635 - add AnyEvent->now_update.
636
6374.351 Sat Apr 11 07:56:14 CEST 2009
638 - actually make the signal pipe work under win32.
639 - localise $! in signal handler to avoid changing $!, although
640 perl itself does not seem to save/restore errno either.
641 - set the cloexec flag on the signal pipe (normally set by perl too).
642
6434.35 Fri Mar 27 11:48:20 CET 2009
644 - event models relying on AnyEvent's signal watcher emulation did
645 invoke the callback asynchronously, contradicting documentation
646 and causing signals to get lost (this includes AnyEvent's own
647 event loop). AnyEvent now uses the standard pipe trick to make
648 callback execution synchronous to the event loop.
649 - AnyEvent::Handle didn't free TLS context data on DESTROY
650 (patch by Pavel Shaydo).
651 - work around the ever-incompatibly-changing API of MakeMaker.
652 - document that changing global variables without restoring them
653 is a bad idea in Perl (noted by Adam Rosenstein).
654 - AnyEvent::Strict now barfs if ->io is passed a file.
655
6564.34 Thu Feb 12 18:32:45 CET 2009
657 - separately buffer TLS read data, as otherwise the read queue could
658 deadlock as receiving data is not expected while draining the
659 read queue (which cna only happen with TLS).
660 - raise EBADMSG error on JSON decoding errors.
661 - fix some minor manpage bugs (reported by Maximilian Gaß).
662 - speed up select bitmask parsing quite a bit in the pure perl
663 backend.
664 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
665
6664.331 Tue Jan 6 21:07:25 CET 2009
667 - socketpair fails on many vista machines because vista has
668 completely broken accept/getpeername and getsockname functions,
669 so we provide our own socketpair emulation that kind of works
670 (AnyEvent::Util::portable_pipe).
671 - new function: AnyEvent::Util::portable_socketpair.
672 - take advantage of the Guard module if it exists.
673
6744.33 Fri Nov 21 02:35:40 CET 2008
675 - AnyEvent::Strict did errornously flag a fileno of 0 as
676 illegal.
677 - reduce memory usage and slightly speed up the pure perl backend
678 by only storing the file descriptor, not the file handle.
679 - add missing autocork method to AnyEvent::Handle (reported
680 by Adam Rosenstein).
681 - AnyEvent::DNS->resolve errornously documented an $rcode result
682 argument, but there isn't (spotted by Henrik Krohns).
683 - the naked truth about Tk - it's basically unmaintained.
2 684
34.32 Mon Nov 3 22:46:32 CET 2008 6854.32 Mon Nov 3 22:46:32 CET 2008
4 - fix AnyEvent::Socket::resolve_sockaddr to properly support 686 - fix AnyEvent::Socket::resolve_sockaddr to properly support
5 unix sockets again and choose a proper default for 687 unix sockets again and choose a proper default for
6 the socket type. 688 the socket type.
… …
174.3 Fri Oct 3 09:18:43 CEST 2008 6994.3 Fri Oct 3 09:18:43 CEST 2008
18 - AnyEvent will now install a no-op signal handler for SIGPIPE, 700 - AnyEvent will now install a no-op signal handler for SIGPIPE,
19 unless one has been installed already. 701 unless one has been installed already.
20 - warn about not loaidng AnyEvent::Impl::POE early enough 702 - warn about not loaidng AnyEvent::Impl::POE early enough
21 (patch by Adam Rosenstein). 703 (patch by Adam Rosenstein).
22 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 704 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
23 - fix a bug in where in SSL connect mode, the 705 - fix a bug in where in SSL connect mode, the
24 client would first wait for some data by the server and 706 client would first wait for some data by the server and
25 otherwise hang. (reported and analyzed in an absolutely 707 otherwise hang. (reported and analysed in an absolutely
26 exemplary manner by Adam Rosenstein). 708 exemplary manner by Adam Rosenstein).
27 - fix a bug in where SSL EOF would not be treated 709 - fix a bug in where SSL EOF would not be treated
28 as stream EOF, putting the connection into a hung state. 710 as stream EOF, putting the connection into a hung state.
29 (reported and analyzed in an absolutely exemplary manner by Adam 711 (reported and analysed in an absolutely exemplary manner by Adam
30 Rosenstein). 712 Rosenstein).
31 - fix a potential segfault when the TLS context 713 - fix a potential segfault when the TLS context
32 would go missing in a read callback (e.g. due to stoptls) - 714 would go missing in a read callback (e.g. due to stoptls) -
33 Net::SSLeay of course makes no type checking whatsoever. 715 Net::SSLeay of course makes no type checking whatsoever.
34 - AnyEvent::Handle will not stop the read watcher in TLS mode. 716 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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