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

Comparing AnyEvent/Changes (file contents):
Revision 1.269 by root, Wed Jul 8 04:49:18 2009 UTC vs.
Revision 1.550 by root, Fri Apr 13 15:24:57 2012 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.801 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
7
87.01
9 - use File::Spec to get the tmpdir in t/*, to avoid needless
10 failures on (most, not mine :) windows boxes.
11
127.0 Fri Apr 13 06:33:30 CEST 2012
13 - child watchers are broken in POE 1.352 (also many earlier
14 versions) and there seems to be no way to work around it, as POE
15 itself is inherently racy. Document this regression and add a delay
16 in t/68_poe_03_child.t for the time being.
17 - new module AnyEvent::IO, that is a frontend to either a pure-perl
18 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
19 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
20 - load /etc/hosts only when DNS has no answer.
21 - stat /etc/hosts on every access and reload it if it changed.
22 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
23 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
24 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
25 a side effect, they now use true appending as opposed to libc
26 appending, and the file might not have been opened when the function
27 returns.
28 - the default logging level was not properly documented in a variety of
29 places, this has hopefully been rectified.
30 - updated uts46data.pl for unicode 6.1.0.
31 - made log messages generated by AnyEvent submodules not
32 include the package name anymore, as it will be logged
33 by default already.
34 - upgrade to the trick used by common-sense 3.5 to work
35 around extra warning torture/breakage under perl 5.15.x.
36 - log messages by anyevent are now ucfirst, are usually full
37 sentences and do no longer include the package name.
38 - the storable read type would sometimes throw an exception instead
39 of causing EBADMSG (analyzed by Maxime Soulé).
40 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
41 keysize (reported by Paul Howarth).
42 - documented Rocco fabricating statements about AnyEvent and me.
43
446.14 Tue Jan 31 20:00:24 CET 2012
45 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
46 probably an editing mistake (reported by Darin McBride).
47
486.13 Thu Jan 12 07:27:01 CET 2012
49 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
50 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
51 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
52 regex-matching inside the callback would change the parameter, despite
53 $1 being dynamically scoped per-block. this perl bug is now being
54 worked around (testcase by Cindy Wang).
55
566.12 Mon Dec 12 13:21:10 CET 2011
57 - $! was clobbered by subsequent calls in tcp_connect, due to
58 the postpone not saving/restoring it (analyzed by Richard Garnish).
59
606.11 Tue Nov 22 10:36:05 CET 2011
61 - Tk cannot create windows when tainted unless you set a title, so
62 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
63 - escape any nonprintable/nonascii characters when stringifying
64 backtraces.
65 - log the reason loading an interface module fails to load at level debug.
66
676.1 Tue Oct 4 19:44:30 CEST 2011
68 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
69 and above), and some messages inside AnyEvent have been elevated
70 to higher log levels to print by default.
71 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
72 - work around yet another signal race bug, newly introduced by POE. My
73 how that module sucks.
74 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
75 support ipv6-only operation even with defaults (also use ::1 instead
76 of 127.0.0.1).
77 - give better guidelines on how to chose a logging level.
78 - AnyEvent::Log can now cap (limit) log message priority by context.
79 - AE::log will now save and restore $! and $@, so callers don't have to.
80 - AE::Strict now checks that registered fds are still valid at regular
81 intervals.
82 - skip frequently called 'require' calls if the module appears loaded.
83 - increase signal latency test timer considerably for poor little osx,
84 which can't be bothered with basic posix functionality such as signal
85 delivery.
86 - mention some debugging aids in the tutorial.
87 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
88 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
89 of the context as title.
90 - AnyEvent::Strict would not restore read-write access to $_ when a callback
91 died - this is a user bug, but since AE::Strict exists to help dbeugging...
92 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
93 to work with it - the author is well aware of the issues but refuses
94 to fix them. This extreme step has been taken because the module actively
95 attacks the goals of AnyEvent and makes it harder for module authors to use
96 AnyEvent.
97
986.02 Fri Aug 26 20:08:31 CEST 2011
99 - log_to_syslog now accepts facility strings.
100 - log_to_syslog would not allow facilities - "$facility|$level"
101 *literally* means that string. sheesh.
102 - the CPAN parser got confused and thought we provide Sys::Syslog, try
103 to work around.
104 - renamed AnyEvent::Impl::FLTK2 to FLTK.
105
1066.01 Fri Aug 26 07:04:11 CEST 2011
107 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
108 the TTL as fourth element - this affects ->resolve, ->request
109 and ->dns_unpack indirectly. Convenience functions such as
110 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
111 AnyEvent::Socket.
112 - INCOMPATIBLE CHANGE: backend authors now should not implement
113 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
114 methods.
115 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
116 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
117 - DEPRECIATION: please change
118 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
119 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
120 in your programs, for future compatibility.
121 - *blush* idle watcher emulation was borked.
122 - AnyEvent::Strict now checks the AE::xxx functions as well.
123 - querying too long domain names would barf deep inside AnyEvent::DNS - since
124 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
125 for too long domain names explicitly and passes an undef to the callback
126 instead (reported by James Bromberger).
127 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
128 - new functions: AnyEvent::postpone, AnyEvent::log.
129 - new module: AnyEvent::Log, for simple logging needs.
130 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
131 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
132 a debug shell.
133 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
134 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
135 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
136 /etc/hosts can be found.
137 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
138 configuration.
139 - env variables can now be specified using AE_xyz instead of
140 PERL_ANYEVENT_xyz, subject to some rules.
141 - lots of new AnyEvent::Debug functionality.
142 - AnyEvent::Strict now checks for modifications of $_ by the callback
143 (by setting it to readonly while executing the callback, if possible).
144 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
145 Invest extra overhead to work around this peculiar behaviour.
146 - split the pure perl backend into a loop module and an impl
147 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
148 - some read types (regex, netstring, json, storable) could
149 errornously keep an AE::Handle object alive on their own
150 (reported by Mohammad Toossi).
151 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
152 supports /etc/hosts overrides.
153 - add more workarounds around child watcher bugs in IO::Async
154 and (newly introduced) in POE.
155 - use glib child watchers and try to work around its limitations,
156 if possible (based on a patch by Kevin Ryde).
157 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
158 - many read types in AnyEvent::Handle would malfunction during
159 transitions to TLS because of a "delete $self->{rbuf}".
160 (reported by Gerald Galster).
161 - improve condvar blocking wait performance (probably only noticable
162 with faster backends), also make it easier for event loops that
163 do not support blocking waits.
164 - PERL_ANYEVENT_MODEL now also allows full module names, although
165 the usefulness of this feature is not yet clear.
166 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
167 - add t/80_ssltext.t.
168 - no longer use AUTOLOAD for the initial loading (for no very good reason).
169 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
170 - do not run event loop detection multiple times when programs call their cached
171 copies of AnyEvent::detect.
172 - parse_hostport now accepts absolute paths als unix domain sockets.
173 - suppress the idiotic warning inside POE by patching the POE::Kernel
174 run flags instead of asking the user to jump through contortions.
175 - add Task::Weaken as dependency (as recommends only at this time).
176 - replace PF_UNSPEC by 0, for good old minix.
177
1786.0 (not officially released, but was on CPAN for about 15 minutes :/)
179
1805.34 Thu May 12 09:59:41 CEST 2011
181 - AE::TLS didn't support both cert_file and key at the same time,
182 due to some copy&paste bug (reported and analyzed by vti).
183 - improve AE::Handle timeout documentation slightly.
184
1855.33 Mon Feb 7 21:51:41 CET 2011
186 - never appeared on CPAN, for unknown reasons.
187 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
188 object when it's already gone (analyzed by Ryan Bullock).
189 - update idna algorithm(s) to UTS#46 version 6.0.0.
190 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
191 uts46 says it's an error, and testsuite makes more sense.
192 - remove leading dots in idn_to_ascii - testsuite says to remove,
193 uts46 says it's an error, and testsuite makes more sense.
194 - fixed a bug in idn ignored character processing.
195 - passes the uts46 6.0.0 testsuite except for three cases which
196 apparently are bugs in the testsuite itself.
197 - passes the uts46 6.0.1 testsuite except for eleven cases which
198 apparently are bugs in the testsuite itself.
199
2005.31 Sun Jan 23 11:49:19 CET 2011
201 - work around a POE bug where POE would first poll for new events and
202 invoke handlers BEFORE handling the message queue.
203 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
204 - stop_read now does just that in TLS mode. see its
205 documentation for caveats.
206 - update to common::sense 3.4 warning set (some warnings in perl impose
207 an arbitrary 32kb limit on regex matches, which affects
208 AnyEvent::Handle).
209 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
210 - new experimental AE::Handle::resettls method.
211
2125.3 Fri Dec 31 05:48:13 CET 2010
213 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
214 issues with IO::Async have been solved in that version, including
215 the availabilty of a default loop!
216 - implement new wbuf_max parameter for AnyEvent::Handle.
217 - added Cocoa::EventLoop backend by Daisuke Murase.
218 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
219 - clarify handle on_prepare documentation.
220 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
221 to avoid leaking them to any child processes.
222 - try to force files into the arch-dependend lib directory, for the
223 half-"turly obsessive compulsive person"(s) that depend on it.
224
2255.29 Sun Dec 5 10:49:21 CET 2010
226 - convert EV backend to EV 4.00 API (so better upgrade EV too).
227 - AE::Handle: implement dynamic read_size adjustment, add
228 max_read_size parameter, reduce default read_size to 2048.
229 - add some FAQ entries for servers and/or on_eof vs. on_error.
230 - work around OS bugs (cygwin again) when lingering in AE::Handle.
231 - reduce memory usage slightly when connect was used in AE::Handle.
232 - wrok around more segfaults in Tk.
233 - document tls_ctx => undef as valid.
234 - detect WNOHANG value at compiletime.
235 - include some probably linux-only support for building AnyEvent
236 as part of the perl core.
237 - improve compatibility of t/08_idna.t to perl 5.8.
238 - make AnyEvent::Strict truly optional (in case somebody builds a
239 minimal perl...).
240
2415.28 Wed Oct 13 04:14:23 CEST 2010
242 - due to a glitch, AnyEvent's internal getprotobyname wasn't
243 actually used in tcp_server.
244 - implement AnyEvent::CondVar->new method, which was documented but
245 missing.
246 - support autoloading of read types in unshift_read.
247 - AnyEvent::Strict: do the test for invalid fh arguments differently,
248 resulting in more informative error messages.
249 - new AnyEvent::FAQ manpage.
250 - clarify that you can actually call push_read etc. in on_read.
251 - put keys and certs in /tmp, not into the current directory,
252 and document this in the SECURITY CONSIDERATIONS sections
253 of AnyEvent::TLS.
254 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
255 requested by Ron Isaacson.
256 - document $AnyEvent::DNS::RESOLVER variable.
257 - incorporated major typo patches by Abhijit Menon-Sen.
258 - AnyEvent::Handle now croaks when negative timeouts are passed.
259 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
260 undocumented, but who knows when they might come in handy.
261
2625.271 Tue Jun 8 12:05:46 CEST 2010
263 - backport to perl 5.8.x.
264
2655.27 Sun Jun 6 12:12:05 CEST 2010
266 - postpone differently in AnyEvent::Socket now, as
267 when not, canceling the connection attempt might fail
268 (found by Felix Antonius Wilhelm Ostmann).
269 - explicitly check for non-stream sockets in AE::Handle, too many
270 clueless people fell into the trap of this somehow working.
271 - simplified and reworked the "OTHER MODULES" section.
272 - better/more condvar examples.
273
2745.261 Wed Apr 28 16:13:36 CEST 2010
275 - AF_INET6 was not properly used from Socket6 during configuration
276 time (found by Andy Grundman).
277
2785.26 Mon Apr 12 04:49:35 CEST 2010
279 - don't generate spurious readyness notifications when select
280 returns EINTR in the pure perl backend (analysed and testcase
281 by Bas Denissen).
282 - give same examples in SYNOPSIS sections of AnyEvent and AE.
283 - provider faster implementations for AE::time/now/now_update for
284 EV and Perl backends.
285 - provide faster AE::cv and AE::time implementations for all
286 backends.
287 - no longer support lower/mixed-case signal names (this was
288 never documented nor universally supported).
289 - some more memory µ-optimisations, and somewhat less messy
290 code to cope with both AnyEvent and AE APIs.
291
2925.251 Sat Mar 13 00:58:21 CET 2010
293 - make test failed if it was run before make install on perl <=
294 5.8.8, and older versions of perl make our live very difficult,
295 so write constants.pl during Makefile.PL time.
296
2975.25 Sat Mar 13 00:23:14 CET 2010
298 - fix a race condition in AnyEvent::Handle that would cause
299 a "bio_read: unsupported method" error (in _another_
300 ssl connection) after user code threw an exception.
301 - added AnyEvent::Handle->destroyed method.
302 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
303 - the AnyEvent::Util::fh_nonblocking function worked only
304 by lucky accident on win32.
305 - smaller and faster AnyEvent::Util::fh_nonblocking.
306 - when the (required!) Time::HiRes module is missing, AnyEvent
307 did not fall back to built-in time properly.
308 - do not load Fcntl at runtime, saving memory and loading time.
309 - precompile a number of constants and use them instead of runtime
310 detection and eval.
311 - free detection code after detection and similar memory optimisations.
312 - Perl backend timer interval best effort drift has been improved
313 (same algorithm as EV).
314 - update unicode idna mapping table.
315
3165.24 Tue Jan 5 11:39:43 CET 2010
317 - cygwin never reports errors from failed connects. 1.5 just gives
318 you continous readyness and EAGAIN, 1.7 is even more broken and
319 just hangs. work around both issues in a cygwin-specific hack.
320 - improve idle watcher documentation slightly.
321
3225.23 Sun Dec 20 23:48:00 CET 2009
323 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
324 - implement punycode_encode/decode, idn_nameprep,
325 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
326 - provide $AE::VERSION.
327 - removed traces of "no strict 'refs'".
328
3295.22 Sat Dec 5 03:51:13 CET 2009
330 - downgrade-or-fail in AnyEvent::Handle::push_write, to
331 diagnose encoding failures earlier and more succinctly.
332 (this works around bugs in perl, throwing away encoding info
333 when passing scalar data to extensions).
334 - add more examples to AnyEvent::Socket manpage.
335 - upgrade internal warning set to the same as common::sense 2.03.
336 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
337 (John Beppu).
338 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
339 - add google public dns servers to fallback server set and make sure
340 we load-balance properly between the three sets. also add all
341 fallback dns servers, not just a random one, to each dns config.
342
3435.21 Thu Nov 19 02:48:47 CET 2009
344 - fix a problem where socket constants were called with parameters
345 (spotted by David Friedland).
346 - fork_call never use'd POSIX (reported by Daisuke Maki).
347 - improve perl 5.6 compatibility further (but it still won't work
348 unless you rip out everything but the core).
349 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
350 uses 10MB of RAM, but doesn't open a console window. *sigh*.
351
3525.202 Wed Oct 14 22:35:44 CEST 2009
353 - AnyEvent::DNS would unexpectedly clobber $_ under windows
354 (analysed by Matthias Waldorf).
355 - AnyEvent::Handle::run_cmd can now pass the PID of the
356 newly-created process, which is much less useful than it might
357 sound (based on patch by Yann Kerherve).
358
3595.201 Tue Sep 29 12:09:25 CEST 2009
360 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
361 (reported by Torsten Foertsch).
362 - common::sense 2.0 could cause tcp_server to throw an exception
363 (analysed by elmex).
364
3655.2 Mon Sep 14 07:04:49 CEST 2009
366 - INCOMPATIBLE CHANGE: do no longer support register_read_type
367 and register_write_type in AnyEvent::Handle, instead support
368 package names (the facility was mostly abused).
369 - implement "packagename-as-read/write type" support in
370 AnyEvent::Handle.
371 - AnyEvent::Handle: new options "keepalive" and "oobinline".
372 - oobinline set by default to avoid security issues.
373 - the pure-perl event loop backend wrongly detected times() fallback
374 support (spotted by Pavel Boldin).
375
3765.12 Tue Sep 1 20:26:50 CEST 2009
377 - be more lenient when parsing resolv.conf files, as some people
378 use hashmarks as comment indicator inside directives (reported
379 by Michael S. Fischer).
380 - use same set of warnings as common::sense 2.0.
381 - fix a potential 32 bit overflow issue due to perl having problems
382 with large hex constants in the Perl backend.
383 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
384 by default.
385 - fix some prototypes.
386 - (optionally) require Async::Interrupt 1.02.
387 - replace opendns fallbacks by verizon/level3 public dns servers.
388
3895.112 Fri Aug 21 13:59:04 CEST 2009
390 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
391 using starttls with an illegal/uncreatable tls context.
392 - port to broken win32 perls.
393
3945.111 Sun Aug 16 18:44:36 CEST 2009
395 - perl errornously requires sockaddr_un structures returned by the
396 kernel to have a specific length. work around this bug by padding
397 sockaddr_un structures (found to fail on OS X, as perl only uses
398 the correct code for GNU/Linux, but likely to fail on other
399 OSes as well).
400
4015.11 Wed Aug 12 17:49:37 CEST 2009
402 - fixed a bug in run_cmd with input from filename.
403 - create a json coder object when none is given in write direction
404 to reduce dependency on JSON.
405
4065.1 Tue Aug 11 03:17:32 CEST 2009
407 - tcp_connect now reports the correct error in $!, not an unrelated
408 one (was broken since 4.91).
409 - AE::Handle did not properly clear rtimeout/wtimeout watchers
410 when retrying a tcp_connect.
411 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
412 - fixed wrong documentation in AE::Socket.
413
4145.01 Mon Aug 10 03:16:32 CEST 2009
415 - last release broke EV child watchers :/.
416
4175.0 Sun Aug 9 17:32:09 CEST 2009
418 - officially introduce and document the AE API.
419 - lowering the timeout at runtime did not have immediate effect
420 in AnyEvent::Handle.
421 - AE::Handle now has separate and independent read and write timeout
422 settings.
423 - probe for Guard module when AnyEvent::Util is loaded, not at
424 runtime (Event doesn't like eval "use" from callbacks, and it avoids
425 exporting the probe function).
426 - Impl::IOAsync's io watchers did not use the filehandle emulation
427 code to work around it's watcher limits.
428 - work around IO::Async's problems with overloaded objects.
429 - try to disable wearnings inside IO::Async, it's just too buggy.
430 - do no longer use Event by default, only EV or Perl, Event is too buggy.
431 - hackishly provide fast AE:: interface when EV or Perl is the backend.
432 - inherit VERSION from AnyEvent::VERSION in most modules.
433
4344.91 Thu Aug 6 15:42:45 CEST 2009
435 - AE::Handle::starttls could get out of sync when the read buffer
436 already contains some TLS handshake.
437 - AE::Handle did not properly free the TLS session, causing bigger
438 memory leaks in Net::SSLeay than expected (reported by toaster).
439 - AE::Socket::tcp_connect will now artificially delay invoking
440 the callback to avoid returning after invoking the callback.
441 - convert many internal watcher uses to AE API.
442
4434.9 Sat Aug 1 11:07:01 CEST 2009
444 - Glib silently fails when registering a timeout with
445 a negative value, so avoid that.
446 - call condvar callback immediately when it is set
447 after the condition is already signalled.
448 - check rbuf_max condition only after trying to consume
449 data.
450 - normalise signal numbers to names when using the ae's
451 signal handling, but do not document this (yet).
452 - pure perl signal emulation did not properly set nonblocking
453 mode on the signal pipe on win32 (this is unlikely to have caused
454 any issues).
455 - new module: AnyEvent::Debug.
456 - AnyEvent::Strict now checks that a signal was specified
457 by name (not name or number).
458 - reduce memory footprint in typical cases by ~50kb by
459 only compiling signal code when necessary.
460 - add AnyEvent::Handle->rbuf_max.
461 - grab the AE namespace for future new API, implement stubs for
462 said future/alternative API.
463 - new function: AnyEvent::Socket::format_hostport.
464
4654.881 Tue Jul 28 12:51:53 CEST 2009
466 - work around a bug in local in pre-5.10 perls, causing
467 AnyEvent::Handle to recurse when it should not
468 (analysed by elmex).
469
4704.88 Tue Jul 28 04:04:37 CEST 2009
471 - re-bless the handle into a dummy package after calling
472 AnyEvent::Handle::destroy, so the user does not need to check
473 for errors after every push_write etc.
474 - do not attempt to run t/02_signals.t on obviously broken
475 platforms.
476
4774.87 Sun Jul 26 02:06:16 CEST 2009
478 - do not attempt to linger when there is no longer a valid fh.
479 - enforce tls mode to be either accept or connect, do not simply
480 segfault in Net::SSLeay.
481 - AnyEvent::Handle can now call tcp_connect itself (new parameters
482 connect, on_prepare, on_connect and on_connect_error). Updated
483 tutorial accordingly.
484 - add AnyEvent::Impl::Irssi backend.
485
4864.86 Mon Jul 20 23:52:29 CEST 2009
487 - since the verbose warning is not enough, explicitly document
488 that versions before 1.33 of Net::SSLeay are not secure.
489 - work around signal handling races in Event and (...) Event::Lib.
490 - try to align signal-race timer to full-second boundaries.
491 - work around Tk not liking negative timeouts.
492 - don't complain of different grades of environmental unfriendlyness
493 in IO::Async.
494
4954.85 Sat Jul 18 06:16:14 CEST 2009
496 - nail the signal race problem in perl once and for all
497 (see $AnyEvent::MAX_SIGNAL_LATENCY).
498 - take advantage of Async::Interrupt if it is available.
499 - load Time::HiRes and Guard modules on demand only.
500 - add optional/recommended modules section to AnyEvent
501 documentation.
502 - reduce memory usage considerably (and reduce startup penalty)
503 by not using "strict", "warnings" and "overload" modules.
504 - work around buggy windows/openbsd perls and provide EBADMSG
505 and EPROTO ourselves when missing.
506 - improve perl 5.6 compatibility of the core event loop.
507 - made Net::SSLeay version 1.33 a soft requirement.
508
5094.83 Fri Jul 17 16:56:26 CEST 2009
510 - implement AnyEvent::Socket::getprotobyname.
511 - AnyEvent::CondVar's will now detect recursive blocking
512 waits and will croak, as too many people fall into
513 this trap.
514 - AnyEvent::Handle will now call ->destroy on itself after
515 executing the on_error callback, instead of doing some
516 half-baked internal shutdown, for fatal errors.
517 - clarify on_eof behaviour w.r.t. the read queue and
518 on_read callbacks.
519 - ignore some possible spurious wake-ups in tcp_connect.
520
5214.82 Sat Jul 11 00:34:55 CEST 2009
522 - POE and Event backends didn't accept some callable objects as
523 callbacks.
524 - use Config module instead of POSIX module to detect signal names
525 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
526 module doesn't even have all POSIX signals :/.
527 - use more workarounds around the many refcnt/corruption bugs in
528 Event::Lib.
529 - work around a race condition in perl's select, causing t/03_child.t
530 to rarely fail.
531
5324.81 Thu Jul 9 10:30:30 CEST 2009
533 - AnyEvent::Handle didn't properly diagnose write errors
534 (it expected -1 from syswrite, how lame... :).
4 - support file descriptors in addition to file handles 535 - support file descriptors in addition to file handles
5 in AnyEvent->io. 536 in AnyEvent->io.
6 - new env variables: PERL_ANYEVENT_RESOLV_CONF, 537 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
7 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE 538 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
8 and PERL_ANYEVENT_CA_PATH. 539 and PERL_ANYEVENT_CA_PATH.
9 - provide a sensible synopsis section for AnyEvent::TLS. 540 - provide a sensible synopsis section for AnyEvent::TLS.
541 - add a "supported backends" section to the manpage.
542 - added simple io watcher test to testsuite, using a
543 portable_socketpair.
544 - tried to improve the stability of the Event::Lib backend,
545 YMMV.
10 546
114.8 Mon Jul 6 23:45:16 CEST 2009 5474.8 Mon Jul 6 23:45:16 CEST 2009
12 - AnyEvent::DNS did not properly follow CNAME records with 548 - AnyEvent::DNS did not properly follow CNAME records with
13 uppercase targets. 549 uppercase targets.
14 - AnyEvent::DNS would errornously return AAAA records 550 - AnyEvent::DNS would errornously return AAAA records
… …
1594.3 Fri Oct 3 09:18:43 CEST 2008 6954.3 Fri Oct 3 09:18:43 CEST 2008
160 - AnyEvent will now install a no-op signal handler for SIGPIPE, 696 - AnyEvent will now install a no-op signal handler for SIGPIPE,
161 unless one has been installed already. 697 unless one has been installed already.
162 - warn about not loaidng AnyEvent::Impl::POE early enough 698 - warn about not loaidng AnyEvent::Impl::POE early enough
163 (patch by Adam Rosenstein). 699 (patch by Adam Rosenstein).
164 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 700 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
165 - fix a bug in where in SSL connect mode, the 701 - fix a bug in where in SSL connect mode, the
166 client would first wait for some data by the server and 702 client would first wait for some data by the server and
167 otherwise hang. (reported and analyzed in an absolutely 703 otherwise hang. (reported and analysed in an absolutely
168 exemplary manner by Adam Rosenstein). 704 exemplary manner by Adam Rosenstein).
169 - fix a bug in where SSL EOF would not be treated 705 - fix a bug in where SSL EOF would not be treated
170 as stream EOF, putting the connection into a hung state. 706 as stream EOF, putting the connection into a hung state.
171 (reported and analyzed in an absolutely exemplary manner by Adam 707 (reported and analysed in an absolutely exemplary manner by Adam
172 Rosenstein). 708 Rosenstein).
173 - fix a potential segfault when the TLS context 709 - fix a potential segfault when the TLS context
174 would go missing in a read callback (e.g. due to stoptls) - 710 would go missing in a read callback (e.g. due to stoptls) -
175 Net::SSLeay of course makes no type checking whatsoever. 711 Net::SSLeay of course makes no type checking whatsoever.
176 - AnyEvent::Handle will not stop the read watcher in TLS mode. 712 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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