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

Comparing AnyEvent/Changes (file contents):
Revision 1.180 by root, Sun Jul 27 07:25:39 2008 UTC vs.
Revision 1.500 by root, Sat Aug 27 15:59:12 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: document TCP_*** constants
4
5 - increase signal latency test timer considerably for poor little osx.
6
76.02 Fri Aug 26 20:08:31 CEST 2011
8 - log_to_syslog now accepts facility strings.
9 - log_to_syslog would not allow facilities - "$facility|$level"
10 *literally* means that string. sheesh.
11 - the CPAN parser got confused and thought we provide Sys::Syslog, try
12 to work around.
13 - renamed AnyEvent::Impl::FLTK2 to FLTK.
14
156.01 Fri Aug 26 07:04:11 CEST 2011
16 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
17 the TTL as fourth element - this affects ->resolve, ->request
18 and ->dns_unpack indirectly. Convenience functions such as
19 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
20 AnyEvent::Socket.
21 - INCOMPATIBLE CHANGE: backend authors now should not implement
22 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
23 methods.
24 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
25 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
26 - DEPRECIATION: please change
27 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
28 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
29 in your programs, for future compatibility.
30 - *blush* idle watcher emulation was borked.
31 - AnyEvent::Strict now checks the AE::xxx functions as well.
32 - querying too long domain names would barf deep inside AnyEvent::DNS - since
33 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
34 for too long domain names explicitly and passes an undef to the callback
35 instead (reported by James Bromberger).
36 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
37 - new functions: AnyEvent::postpone, AnyEvent::log.
38 - new module: AnyEvent::Log, for simple logging needs.
39 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
40 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
41 a debug shell.
42 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
43 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
44 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
45 F</etc/hosts> can be found.
46 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
47 configuration.
48 - env variables can now be specified using AE_xyz instead of
49 PERL_ANYEVENT_xyz, subject to some rules.
50 - lots of new AnyEvent::Debug functionality.
51 - AnyEvent::Strict now checks for modifications $_ by the callback
52 (by setting it to readonly while executing the callback, if possible).
53 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
54 Invest extra overhead to work around this peculiar behaviour.
55 - split the pure perl backend into a loop module and an impl
56 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
57 - some read types (regex, netstring, json, storable) could
58 errornously keep an AE::Handle object alive on their own
59 (reported by Mohammad Toossi).
60 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
61 supports /etc/hosts overrides.
62 - add more workarounds around child watcher bugs in IO::Async
63 and (newly introduced) in POE.
64 - use glib child watchers and try to work around its limitations,
65 if possible (based on a patch by Kevin Ryde).
66 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
67 - many read types in AnyEvent::Handle would malfunction during
68 transitions to TLS because of a "delete $self->{rbuf}".
69 (reported by Gerald Galster).
70 - improve condvar blocking wait performance (probably only noticable
71 with faster backends), also make it easier for event loops that
72 do not support blocking waits.
73 - PERL_ANYEVENT_MODEL now also allows full module names, although
74 the usefulness of this feature is not yet clear.
75 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
76 - add t/80_ssltext.t.
77 - no longer use AUTOLOAD for the initial loading (for no very good reason).
78 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
79 - do not run event loop detection multiple times when programs call their cached
80 copies of AnyEvent::detect.
81 - parse_hostport now accepts absolute paths als unix domain sockets.
82 - suppress the idiotic warning inside POE by patching the POE::Kernel
83 run flags instead of asking the user to jump through contortions.
84 - add Task::Weaken as dependency (as recommends only at this time).
85 - replace PF_UNSPEC by 0, for good old minix.
86
876.0 (not officially released, but was on CPAN for about 15 minutes :/)
88
895.34 Thu May 12 09:59:41 CEST 2011
90 - AE::TLS didn't support both cert_file and key at the same time,
91 due to some copy&paste bug (reported and analyzed by vti).
92 - improve AE::Handle timeout documentation slightly.
93
945.33 Mon Feb 7 21:51:41 CET 2011
95 - never appeared on CPAN, for unknown reasons.
96 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
97 object when it's already gone (analyzed by Ryan Bullock).
98 - update idna algorithm(s) to UTS#46 version 6.0.0.
99 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
100 uts46 says it's an error, and testsuite makes more sense.
101 - remove leading dots in idn_to_ascii - testsuite says to remove,
102 uts46 says it's an error, and testsuite makes more sense.
103 - fixed a bug in idn ignored character processing.
104 - passes the uts46 6.0.0 testsuite except for three cases which
105 apparently are bugs in the testsuite itself.
106 - passes the uts46 6.0.1 testsuite except for eleven cases which
107 apparently are bugs in the testsuite itself.
108
1095.31 Sun Jan 23 11:49:19 CET 2011
110 - work around a POE bug where POE would first poll for new events and
111 invoke handlers BEFORE handling the message queue.
112 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
113 - stop_read now does just that in TLS mode. see its
114 documentation for caveats.
115 - update to common::sense 3.4 warning set (some warnings in perl impose
116 an arbitrary 32kb limit on regex matches, which affects
117 AnyEvent::Handle).
118 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
119 - new experimental AE::Handle::resettls method.
120
1215.3 Fri Dec 31 05:48:13 CET 2010
122 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
123 issues with IO::Async have been solved in that version, including
124 the availabilty of a default loop!
125 - implement new wbuf_max parameter for AnyEvent::Handle.
126 - added Cocoa::EventLoop backend by Daisuke Murase.
127 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
128 - clarify handle on_prepare documentation.
129 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
130 to avoid leaking them to any child processes.
131 - try to force files into the arch-dependend lib directory, for the
132 half-"turly obsessive compulsive person"(s) that depend on it.
133
1345.29 Sun Dec 5 10:49:21 CET 2010
135 - convert EV backend to EV 4.00 API (so better upgrade EV too).
136 - AE::Handle: implement dynamic read_size adjustment, add
137 max_read_size parameter, reduce default read_size to 2048.
138 - add some FAQ entries for servers and/or on_eof vs. on_error.
139 - work around OS bugs (cygwin again) when lingering in AE::Handle.
140 - reduce memory usage slightly when connect was used in AE::Handle.
141 - wrok around more segfaults in Tk.
142 - document tls_ctx => undef as valid.
143 - detect WNOHANG value at compiletime.
144 - include some probably linux-only support for building AnyEvent
145 as part of the perl core.
146 - improve compatibility of t/08_idna.t to perl 5.8.
147 - make AnyEvent::Strict truly optional (in case somebody builds a
148 minimal perl...).
149
1505.28 Wed Oct 13 04:14:23 CEST 2010
151 - due to a glitch, AnyEvent's internal getprotobyname wasn't
152 actually used in tcp_server.
153 - implement AnyEvent::CondVar->new method, which was documented but
154 missing.
155 - support autoloading of read types in unshift_read.
156 - AnyEvent::Strict: do the test for invalid fh arguments differently,
157 resulting in more informative error messages.
158 - new AnyEvent::FAQ manpage.
159 - clarify that you can actually call push_read etc. in on_read.
160 - put keys and certs in /tmp, not into the current directory,
161 and document this in the SECURITY CONSIDERATIONS sections
162 of AnyEvent::TLS.
163 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
164 requested by Ron Isaacson.
165 - document $AnyEvent::DNS::RESOLVER variable.
166 - incorporated major typo patches by Abhijit Menon-Sen.
167 - AnyEvent::Handle now croaks when negative timeouts are passed.
168 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
169 undocumented, but who knows when they might come in handy.
170
1715.271 Tue Jun 8 12:05:46 CEST 2010
172 - backport to perl 5.8.x.
173
1745.27 Sun Jun 6 12:12:05 CEST 2010
175 - postpone differently in AnyEvent::Socket now, as
176 when not, canceling the connection attempt might fail
177 (found by Felix Antonius Wilhelm Ostmann).
178 - explicitly check for non-stream sockets in AE::Handle, too many
179 clueless people fell into the trap of this somehow working.
180 - simplified and reworked the "OTHER MODULES" section.
181 - better/more condvar examples.
182
1835.261 Wed Apr 28 16:13:36 CEST 2010
184 - AF_INET6 was not properly used from Socket6 during configuration
185 time (found by Andy Grundman).
186
1875.26 Mon Apr 12 04:49:35 CEST 2010
188 - don't generate spurious readyness notifications when select
189 returns EINTR in the pure perl backend (analysed and testcase
190 by Bas Denissen).
191 - give same examples in SYNOPSIS sections of AnyEvent and AE.
192 - provider faster implementations for AE::time/now/now_update for
193 EV and Perl backends.
194 - provide faster AE::cv and AE::time implementations for all
195 backends.
196 - no longer support lower/mixed-case signal names (this was
197 never documented nor universally supported).
198 - some more memory µ-optimisations, and somewhat less messy
199 code to cope with both AnyEvent and AE APIs.
200
2015.251 Sat Mar 13 00:58:21 CET 2010
202 - make test failed if it was run before make install on perl <=
203 5.8.8, and older versions of perl make our live very difficult,
204 so write constants.pl during Makefile.PL time.
205
2065.25 Sat Mar 13 00:23:14 CET 2010
207 - fix a race condition in AnyEvent::Handle that would cause
208 a "bio_read: unsupported method" error (in _another_
209 ssl connection) after user code threw an exception.
210 - added AnyEvent::Handle->destroyed method.
211 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
212 - the AnyEvent::Util::fh_nonblocking function worked only
213 by lucky accident on win32.
214 - smaller and faster AnyEvent::Util::fh_nonblocking.
215 - when the (required!) Time::HiRes module is missing, AnyEvent
216 did not fall back to built-in time properly.
217 - do not load Fcntl at runtime, saving memory and loading time.
218 - precompile a number of constants and use them instead of runtime
219 detection and eval.
220 - free detection code after detection and similar memory optimisations.
221 - Perl backend timer interval best effort drift has been improved
222 (same algorithm as EV).
223 - update unicode idna mapping table.
224
2255.24 Tue Jan 5 11:39:43 CET 2010
226 - cygwin never reports errors from failed connects. 1.5 just gives
227 you continous readyness and EAGAIN, 1.7 is even more broken and
228 just hangs. work around both issues in a cygwin-specific hack.
229 - improve idle watcher documentation slightly.
230
2315.23 Sun Dec 20 23:48:00 CET 2009
232 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
233 - implement punycode_encode/decode, idn_nameprep,
234 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
235 - provide $AE::VERSION.
236 - removed traces of "no strict 'refs'".
237
2385.22 Sat Dec 5 03:51:13 CET 2009
239 - downgrade-or-fail in AnyEvent::Handle::push_write, to
240 diagnose encoding failures earlier and more succinctly.
241 (this works around bugs in perl, throwing away encoding info
242 when passing scalar data to extensions).
243 - add more examples to AnyEvent::Socket manpage.
244 - upgrade internal warning set to the same as common::sense 2.03.
245 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
246 (John Beppu).
247 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
248 - add google public dns servers to fallback server set and make sure
249 we load-balance properly between the three sets. also add all
250 fallback dns servers, not just a random one, to each dns config.
251
2525.21 Thu Nov 19 02:48:47 CET 2009
253 - fix a problem where socket constants were called with parameters
254 (spotted by David Friedland).
255 - fork_call never use'd POSIX (reported by Daisuke Maki).
256 - improve perl 5.6 compatibility further (but it still won't work
257 unless you rip out everything but the core).
258 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
259 uses 10MB of RAM, but doesn't open a console window. *sigh*.
260
2615.202 Wed Oct 14 22:35:44 CEST 2009
262 - AnyEvent::DNS would unexpectedly clobber $_ under windows
263 (analysed by Matthias Waldorf).
264 - AnyEvent::Handle::run_cmd can now pass the PID of the
265 newly-created process, which is much less useful than it might
266 sound (based on patch by Yann Kerherve).
267
2685.201 Tue Sep 29 12:09:25 CEST 2009
269 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
270 (reported by Torsten Foertsch).
271 - common::sense 2.0 could cause tcp_server to throw an exception
272 (analysed by elmex).
273
2745.2 Mon Sep 14 07:04:49 CEST 2009
275 - INCOMPATIBLE CHANGE: do no longer support register_read_type
276 and register_write_type in AnyEvent::Handle, instead support
277 package names (the facility was mostly abused).
278 - implement "packagename-as-read/write type" support in
279 AnyEvent::Handle.
280 - AnyEvent::Handle: new options "keepalive" and "oobinline".
281 - oobinline set by default to avoid security issues.
282 - the pure-perl event loop backend wrongly detected times() fallback
283 support (spotted by Pavel Boldin).
284
2855.12 Tue Sep 1 20:26:50 CEST 2009
286 - be more lenient when parsing resolv.conf files, as some people
287 use hashmarks as comment indicator inside directives (reported
288 by Michael S. Fischer).
289 - use same set of warnings as common::sense 2.0.
290 - fix a potential 32 bit overflow issue due to perl having problems
291 with large hex constants in the Perl backend.
292 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
293 by default.
294 - fix some prototypes.
295 - (optionally) require Async::Interrupt 1.02.
296 - replace opendns fallbacks by verizon/level3 public dns servers.
297
2985.112 Fri Aug 21 13:59:04 CEST 2009
299 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
300 using starttls with an illegal/uncreatable tls context.
301 - port to broken win32 perls.
302
3035.111 Sun Aug 16 18:44:36 CEST 2009
304 - perl errornously requires sockaddr_un structures returned by the
305 kernel to have a specific length. work around this bug by padding
306 sockaddr_un structures (found to fail on OS X, as perl only uses
307 the correct code for GNU/Linux, but likely to fail on other
308 OSes as well).
309
3105.11 Wed Aug 12 17:49:37 CEST 2009
311 - fixed a bug in run_cmd with input from filename.
312 - create a json coder object when none is given in write direction
313 to reduce dependency on JSON.
314
3155.1 Tue Aug 11 03:17:32 CEST 2009
316 - tcp_connect now reports the correct error in $!, not an unrelated
317 one (was broken since 4.91).
318 - AE::Handle did not properly clear rtimeout/wtimeout watchers
319 when retrying a tcp_connect.
320 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
321 - fixed wrong documentation in AE::Socket.
322
3235.01 Mon Aug 10 03:16:32 CEST 2009
324 - last release broke EV child watchers :/.
325
3265.0 Sun Aug 9 17:32:09 CEST 2009
327 - officially introduce and document the AE API.
328 - lowering the timeout at runtime did not have immediate effect
329 in AnyEvent::Handle.
330 - AE::Handle now has separate and independent read and write timeout
331 settings.
332 - probe for Guard module when AnyEvent::Util is loaded, not at
333 runtime (Event doesn't like eval "use" from callbacks, and it avoids
334 exporting the probe function).
335 - Impl::IOAsync's io watchers did not use the filehandle emulation
336 code to work around it's watcher limits.
337 - work around IO::Async's problems with overloaded objects.
338 - try to disable wearnings inside IO::Async, it's just too buggy.
339 - do no longer use Event by default, only EV or Perl, Event is too buggy.
340 - hackishly provide fast AE:: interface when EV or Perl is the backend.
341 - inherit VERSION from AnyEvent::VERSION in most modules.
342
3434.91 Thu Aug 6 15:42:45 CEST 2009
344 - AE::Handle::starttls could get out of sync when the read buffer
345 already contains some TLS handshake.
346 - AE::Handle did not properly free the TLS session, causing bigger
347 memory leaks in Net::SSLeay than expected (reported by toaster).
348 - AE::Socket::tcp_connect will now artificially delay invoking
349 the callback to avoid returning after invoking the callback.
350 - convert many internal watcher uses to AE API.
351
3524.9 Sat Aug 1 11:07:01 CEST 2009
353 - Glib silently fails when registering a timeout with
354 a negative value, so avoid that.
355 - call condvar callback immediately when it is set
356 after the condition is already signalled.
357 - check rbuf_max condition only after trying to consume
358 data.
359 - normalise signal numbers to names when using the ae's
360 signal handling, but do not document this (yet).
361 - pure perl signal emulation did not properly set nonblocking
362 mode on the signal pipe on win32 (this is unlikely to have caused
363 any issues).
364 - new module: AnyEvent::Debug.
365 - AnyEvent::Strict now checks that a signal was specified
366 by name (not name or number).
367 - reduce memory footprint in typical cases by ~50kb by
368 only compiling signal code when necessary.
369 - add AnyEvent::Handle->rbuf_max.
370 - grab the AE namespace for future new API, implement stubs for
371 said future/alternative API.
372 - new function: AnyEvent::Socket::format_hostport.
373
3744.881 Tue Jul 28 12:51:53 CEST 2009
375 - work around a bug in local in pre-5.10 perls, causing
376 AnyEvent::Handle to recurse when it should not
377 (analysed by elmex).
378
3794.88 Tue Jul 28 04:04:37 CEST 2009
380 - re-bless the handle into a dummy package after calling
381 AnyEvent::Handle::destroy, so the user does not need to check
382 for errors after every push_write etc.
383 - do not attempt to run t/02_signals.t on obviously broken
384 platforms.
385
3864.87 Sun Jul 26 02:06:16 CEST 2009
387 - do not attempt to linger when there is no longer a valid fh.
388 - enforce tls mode to be either accept or connect, do not simply
389 segfault in Net::SSLeay.
390 - AnyEvent::Handle can now call tcp_connect itself (new parameters
391 connect, on_prepare, on_connect and on_connect_error). Updated
392 tutorial accordingly.
393 - add AnyEvent::Impl::Irssi backend.
394
3954.86 Mon Jul 20 23:52:29 CEST 2009
396 - since the verbose warning is not enough, explicitly document
397 that versions before 1.33 of Net::SSLeay are not secure.
398 - work around signal handling races in Event and (...) Event::Lib.
399 - try to align signal-race timer to full-second boundaries.
400 - work around Tk not liking negative timeouts.
401 - don't complain of different grades of environmental unfriendlyness
402 in IO::Async.
403
4044.85 Sat Jul 18 06:16:14 CEST 2009
405 - nail the signal race problem in perl once and for all
406 (see $AnyEvent::MAX_SIGNAL_LATENCY).
407 - take advantage of Async::Interrupt if it is available.
408 - load Time::HiRes and Guard modules on demand only.
409 - add optional/recommended modules section to AnyEvent
410 documentation.
411 - reduce memory usage considerably (and reduce startup penalty)
412 by not using "strict", "warnings" and "overload" modules.
413 - work around buggy windows/openbsd perls and provide EBADMSG
414 and EPROTO ourselves when missing.
415 - improve perl 5.6 compatibility of the core event loop.
416 - made Net::SSLeay version 1.33 a soft requirement.
417
4184.83 Fri Jul 17 16:56:26 CEST 2009
419 - implement AnyEvent::Socket::getprotobyname.
420 - AnyEvent::CondVar's will now detect recursive blocking
421 waits and will croak, as too many people fall into
422 this trap.
423 - AnyEvent::Handle will now call ->destroy on itself after
424 executing the on_error callback, instead of doing some
425 half-baked internal shutdown, for fatal errors.
426 - clarify on_eof behaviour w.r.t. the read queue and
427 on_read callbacks.
428 - ignore some possible spurious wake-ups in tcp_connect.
429
4304.82 Sat Jul 11 00:34:55 CEST 2009
431 - POE and Event backends didn't accept some callable objects as
432 callbacks.
433 - use Config module instead of POSIX module to detect signal names
434 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
435 module doesn't even have all POSIX signals :/.
436 - use more workarounds around the many refcnt/corruption bugs in
437 Event::Lib.
438 - work around a race condition in perl's select, causing t/03_child.t
439 to rarely fail.
440
4414.81 Thu Jul 9 10:30:30 CEST 2009
442 - AnyEvent::Handle didn't properly diagnose write errors
443 (it expected -1 from syswrite, how lame... :).
444 - support file descriptors in addition to file handles
445 in AnyEvent->io.
446 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
447 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
448 and PERL_ANYEVENT_CA_PATH.
449 - provide a sensible synopsis section for AnyEvent::TLS.
450 - add a "supported backends" section to the manpage.
451 - added simple io watcher test to testsuite, using a
452 portable_socketpair.
453 - tried to improve the stability of the Event::Lib backend,
454 YMMV.
455
4564.8 Mon Jul 6 23:45:16 CEST 2009
457 - AnyEvent::DNS did not properly follow CNAME records with
458 uppercase targets.
459 - AnyEvent::DNS would errornously return AAAA records
460 with v4 mapped addresses (a faulty record) as ipv4 addresses,
461 causing AnyEvent::Socket to throw an exception.
462 - added new module AnyEvent::TLS for easier SSL/TLS context
463 creation, with many options including hostname verification,
464 secure default configuration, lots of documentation and,
465 predefined diffie-hellman keys for perfect forward security
466 and much more. get it while it's still fresh!
467 - use AnyEvent::TLS in AnyEvent::Handle for context management.
468 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
469 so AnyEvent::Socket users have smaller memory footprint
470 in the common case.
471 - add AnyEvent::Handle->push_shutdown method.
472 - add an additional error message parameter to AnyEvent::Handle's
473 on_error callback (for TLS, $! is still available).
474 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
475 - make AnyEvent::Handle more robust against early conenction
476 failures (during new), and return C<undef> in that case
477 from the constructor.
478 - AnyEvent::Handle will now try to load only JSON::XS first,
479 then fall back to JSON.
480 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
481 and because it was documented that way.
482
4834.451 Fri Jul 3 00:28:58 CEST 2009
484 - do not clear rbuf when shutting down an AnyEvent::Handle
485 object - doing so breaks AnyEvent::HTTP.
486
4874.45 Mon Jun 29 22:59:26 CEST 2009
488 - a write error could cause AnyEvent::Handle to create
489 an I/O watcher with an undefined $fh.
490 - special-case mapped ipv4 addresses in both
491 AnyEvent::Socket::format_address and parse_address,
492 to treat them just like ipv4 addresses.
493 - updated and overhauled the AnyEvent::Intro doc.
494 - implement AnyEvent::Socket::format_ipv4/ipv6.
495 - slightly speed up portable_pipe/socketpair.
496 - expand condvar begin/end documentation.
497
4984.42 Fri Jun 26 08:32:18 CEST 2009
499 - preliminary and neccesarily incomplete support for IO::Async.
500 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
501 case it was set to IGNORE, to ensure we can catch child statuses
502 even when the calling env acts stupidly.
503 - updated benchmarks with IO::Async, which performs very well.
504
5054.412 Wed Jun 24 01:35:57 CEST 2009
506 - support an "untaint" attribute for AnyEvent::DNS
507 and set it on the default resolver.
508 - implement and document AnyEvent::Impl::Perl::loop.
509 - remove all anyevent-env variables from %ENV
510 when running in tainted mode.
511 - mention and extend the IO::Lambda benchmark.
512
5134.411 Sun Jun 7 18:48:02 CEST 2009
514 - do not try to use F_SETFD on windows.
515
5164.41 Thu May 14 06:40:11 CEST 2009
517 - work around issues in older perls (5.8.0?) when
518 a signal handler is deleted from the %SIG hash.
519 - use POSIX::_exit in child test, to avoid running
520 destructors.
521 - speed up CHLD handling by relying on SIGCHLD being
522 synchronously delivered, even when we roll our own
523 implementation.
524 - AnyEvent::DNS: add the "dname" resource record name
525 for cosmetic reasons.
526
5274.4 Sun Apr 26 20:12:33 CEST 2009
528 - implemented idle watchers, where applicable.
529 - AnyEvent->time died when Event backend was in use.
530 - fix a memleak in the Tk backend.
531 - sped up Tk timer handling.
532 - clip negative "after" values to 0 in AnyEvent::Impl::Event
533 to avoid spamming out warning messages.
534 - fix Qt timers without interval.
535 - avoid Qt zero-timeout special case, in old libqt's.
536
5374.352 Mon Apr 20 16:31:11 CEST 2009
538 - fix AnyEvent::Strict error messages for child watchers.
539 - fix/update Makefile.PL "recommends" versions.
540 - add AnyEvent->now_update.
541
5424.351 Sat Apr 11 07:56:14 CEST 2009
543 - actually make the signal pipe work under win32.
544 - localise $! in signal handler to avoid changing $!, although
545 perl itself does not seem to save/restore errno either.
546 - set the cloexec flag on the signal pipe (normally set by perl too).
547
5484.35 Fri Mar 27 11:48:20 CET 2009
549 - event models relying on AnyEvent's signal watcher emulation did
550 invoke the callback asynchronously, contradicting documentation
551 and causing signals to get lost (this includes AnyEvent's own
552 event loop). AnyEvent now uses the standard pipe trick to make
553 callback execution synchronous to the event loop.
554 - AnyEvent::Handle didn't free TLS context data on DESTROY
555 (patch by Pavel Shaydo).
556 - work around the ever-incompatibly-changing API of MakeMaker.
557 - document that changing global variables without restoring them
558 is a bad idea in Perl (noted by Adam Rosenstein).
559 - AnyEvent::Strict now barfs if ->io is passed a file.
560
5614.34 Thu Feb 12 18:32:45 CET 2009
562 - separately buffer TLS read data, as otherwise the read queue could
563 deadlock as receiving data is not expected while draining the
564 read queue (which cna only happen with TLS).
565 - raise EBADMSG error on JSON decoding errors.
566 - fix some minor manpage bugs (reported by Maximilian Gaß).
567 - speed up select bitmask parsing quite a bit in the pure perl
568 backend.
569 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
570
5714.331 Tue Jan 6 21:07:25 CET 2009
572 - socketpair fails on many vista machines because vista has
573 completely broken accept/getpeername and getsockname functions,
574 so we provide our own socketpair emulation that kind of works
575 (AnyEvent::Util::portable_pipe).
576 - new function: AnyEvent::Util::portable_socketpair.
577 - take advantage of the Guard module if it exists.
578
5794.33 Fri Nov 21 02:35:40 CET 2008
580 - AnyEvent::Strict did errornously flag a fileno of 0 as
581 illegal.
582 - reduce memory usage and slightly speed up the pure perl backend
583 by only storing the file descriptor, not the file handle.
584 - add missing autocork method to AnyEvent::Handle (reported
585 by Adam Rosenstein).
586 - AnyEvent::DNS->resolve errornously documented an $rcode result
587 argument, but there isn't (spotted by Henrik Krohns).
588 - the naked truth about Tk - it's basically unmaintained.
589
5904.32 Mon Nov 3 22:46:32 CET 2008
591 - fix AnyEvent::Socket::resolve_sockaddr to properly support
592 unix sockets again and choose a proper default for
593 the socket type.
594 - fix call to resolve_sockaddr in tcp_connect.
595
5964.31 Thu Oct 30 04:41:48 CET 2008
597 - implemented AnyEvent::Handle->destroy method.
598 - hint about unexpected effects in TLS mode.
599 - speed up AnyEvent::Impl::Perl by using more arrays
600 instead of hashes without the slightest loss of readability :)
601 - work around a perl argument refcounting bug.
602 - with some perl patches applied, AnyEvent now seems leak-free.
603
6044.3 Fri Oct 3 09:18:43 CEST 2008
605 - AnyEvent will now install a no-op signal handler for SIGPIPE,
606 unless one has been installed already.
607 - warn about not loaidng AnyEvent::Impl::POE early enough
608 (patch by Adam Rosenstein).
609 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
610 - fix a bug in where in SSL connect mode, the
611 client would first wait for some data by the server and
612 otherwise hang. (reported and analysed in an absolutely
613 exemplary manner by Adam Rosenstein).
614 - fix a bug in where SSL EOF would not be treated
615 as stream EOF, putting the connection into a hung state.
616 (reported and analysed in an absolutely exemplary manner by Adam
617 Rosenstein).
618 - fix a potential segfault when the TLS context
619 would go missing in a read callback (e.g. due to stoptls) -
620 Net::SSLeay of course makes no type checking whatsoever.
621 - AnyEvent::Handle will not stop the read watcher in TLS mode.
622 - AnyEvent::Handle->stoptls will send a TLS close notify
623 instead of simply destroying the stream now.
624 - fix a bug where large blocks of data written in TLS mode
625 would not be sent unless triggered by receives.
626 - on_drain will now take the tls write buffer into account.
627 - SSL operations have been streamlined a bit and should be faster
628 now, more to come, though.
629 - added a FAQ section, will document nontrivial issues in there.
630
6314.234 Mon Sep 29 04:08:13 CEST 2008
632 - fix child watcher documentation: only child exits will be reported,
633 no trace events.
634 - mention SIGPIPE in AnyEvent::Handle.
635 - perl backend will now use POSIX::times as monotonic clock
636 when available and the monotonic clock option is not.
637 - run even without Time::HiRes in some cases now, but do not
638 provide sub-second accuracy (all *supported* perl versions have
639 Time::HiRes).
640
6414.233 Fri Aug 22 01:48:05 CEST 2008
642 - fix a bug in the testsuite, causing a hang.
643 - clarified lots of AnyEvent::Handle constructor arguments.
644
6454.232 Thu Aug 21 20:44:25 CEST 2008
646 - fix a bug in AnyEvent::Handle that could cause two fatal error
647 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
648 (testcase by Leon Brocard).
649 - clarified on_eof/on_error documentation a bit.
650
6514.231 Tue Jul 29 13:12:15 CEST 2008
652 - remove some debugging code left in AnyEvent::Util::fork_call
653 (and no, it's impossible to implement with the broken windows
654 perls without resource leaks or worse).
655
6564.23 Tue Jul 29 12:19:59 CEST 2008
3 - document the first parameter passed to condvar callbacks to be 657 - document the first parameter passed to condvar callbacks to be
4 the callback. 658 the callback.
5 - add AnyEvent::Socket::{ntoa,aton} aliases. 659 - add AnyEvent::Socket::{ntoa,aton} aliases.
6 - optimize the AE::Handle->push_read (line) for the default 660 - optimize the AE::Handle->push_read (line) for the default
7 eol marker. 661 eol marker.
8 - optimize push_read (packstring|storable) for small packets. 662 - optimize push_read (packstring|storable) for small packets.
663 - invoke on_error callback when no on_eof callback is set.
9 - fix a bug in push_read (storable) of unknown impact. 664 - fix a bug in push_read (storable) of unknown impact.
10 665
114.22 Sun Jul 20 16:34:13 CEST 2008 6664.22 Sun Jul 20 16:34:13 CEST 2008
12 - new function AnyEvent::Socket::parse_hostport. 667 - new function AnyEvent::Socket::parse_hostport.
13 - as the bulkheads at microsoft can't even get getprotobyname reliably 668 - as the bulkheads at microsoft can't even get getprotobyname reliably
… …
195 - support service names in tcp_connect and tcp_server. 850 - support service names in tcp_connect and tcp_server.
196 - work around netbsd/openbsd bug causing unix sockets not to 851 - work around netbsd/openbsd bug causing unix sockets not to
197 become writable until empty. 852 become writable until empty.
198 - only make a dummy read for error detection when getpeername 853 - only make a dummy read for error detection when getpeername
199 returns ENOTCONN. 854 returns ENOTCONN.
200 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 855 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
201 override. 856 override.
202 - new AnyEvent::Guard::cancel. 857 - new AnyEvent::Guard::cancel.
203 858
2043.5 Sat May 17 23:17:14 CEST 2008 8593.5 Sat May 17 23:17:14 CEST 2008
205 - gracefully deal with systems not defining CLOCK_MONOTONIC 860 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
219 874
2203.41 Sun May 11 19:53:13 CEST 2008 8753.41 Sun May 11 19:53:13 CEST 2008
221 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 876 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
222 877
2233.4 Sun May 11 00:29:25 CEST 2008 8783.4 Sun May 11 00:29:25 CEST 2008
224 - complete re-implementation of condvars with a hopefully much mroe useful 879 - complete re-implementation of condvars with a hopefully much more useful
225 API (backward compatible functions exist but are undocumented). 880 API (backward compatible functions exist but are undocumented).
226 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 881 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
227 (still subject to change). 882 (still subject to change).
228 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 883 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
229 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 884 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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