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

Comparing AnyEvent/Changes (file contents):
Revision 1.192 by root, Wed Oct 1 14:49:23 2008 UTC vs.
Revision 1.447 by root, Mon Jul 18 01:19:43 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.3 3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent strict and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict checking for $_ changes?
7
8TODO: break the API and put the rr ttl where it belongs.
9TODO: Kevin Ryde (3.8K) 1 inbox Glib child processes
10 - add support for (currently third-party) AnyEvent::Impl::FLTK.
11 - many read types in AnyEvent::Handle would malfunction during
12 transitions to TLS because of a "delete $self->{rbuf}".
13 (reported by Gerald Galster).
14
155.34 Thu May 12 09:59:41 CEST 2011
16 - AE::TLS didn't support both cert_file and key at the same time,
17 due to some copy&paste bug (reported and analyzed by vti).
18 - improve AE::Handle timeout documentation slightly.
19
205.33 Mon Feb 7 21:51:41 CET 2011
21 - never appeared on CPAN, for unknown reasons.
22 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
23 object when it's already gone (analyzed by Ryan Bullock).
24 - update idna algorithm(s) to UTS#46 version 6.0.0.
25 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
26 uts46 says it's an error, and testsuite makes more sense.
27 - remove leading dots in idn_to_ascii - testsuite says to remove,
28 uts46 says it's an error, and testsuite makes more sense.
29 - fixed a bug in idn ignored character processing.
30 - passes the uts46 6.0.0 testsuite except for three cases which
31 apparently are bugs in the testsuite itself.
32 - passes the uts46 6.0.1 testsuite except for eleven cases which
33 apparently are bugs in the testsuite itself.
34
355.31 Sun Jan 23 11:49:19 CET 2011
36 - work around a POE bug where POE would first poll for new events and
37 invoke handlers BEFORE handling the message queue.
38 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
39 - stop_read now does just that in TLS mode. see its
40 documentation for caveats.
41 - update to common::sense 3.4 warning set (some warnings in perl impose
42 an arbitrary 32kb limit on regex matches, which affects
43 AnyEvent::Handle).
44 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
45 - new experimental AE::Handle::resettls method.
46
475.3 Fri Dec 31 05:48:13 CET 2010
48 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
49 issues with IO::Async have been solved in that version, including
50 the availabilty of a default loop!
51 - implement new wbuf_max parameter for AnyEvent::Handle.
52 - added Cocoa::EventLoop backend by Daisuke Murase.
53 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
54 - clarify handle on_prepare documentation.
55 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
56 to avoid leaking them to any child processes.
57 - try to force files into the arch-dependend lib directory, for the
58 half-"turly obsessive compulsive person"(s) that depend on it.
59
605.29 Sun Dec 5 10:49:21 CET 2010
61 - convert EV backend to EV 4.00 API (so better upgrade EV too).
62 - AE::Handle: implement dynamic read_size adjustment, add
63 max_read_size parameter, reduce default read_size to 2048.
64 - add some FAQ entries for servers and/or on_eof vs. on_error.
65 - work around OS bugs (cygwin again) when lingering in AE::Handle.
66 - reduce memory usage slightly when connect was used in AE::Handle.
67 - wrok around more segfaults in Tk.
68 - document tls_ctx => undef as valid.
69 - detect WNOHANG value at compiletime.
70 - include some probably linux-only support for building AnyEvent
71 as part of the perl core.
72 - improve compatibility of t/08_idna.t to perl 5.8.
73 - make AnyEvent::Strict truly optional (in case somebody builds a
74 minimal perl...).
75
765.28 Wed Oct 13 04:14:23 CEST 2010
77 - due to a glitch, AnyEvent's internal getprotobyname wasn't
78 actually used in tcp_server.
79 - implement AnyEvent::CondVar->new method, which was documented but
80 missing.
81 - support autoloading of read types in unshift_read.
82 - AnyEvent::Strict: do the test for invalid fh arguments differently,
83 resulting in more informative error messages.
84 - new AnyEvent::FAQ manpage.
85 - clarify that you can actually call push_read etc. in on_read.
86 - put keys and certs in /tmp, not into the current directory,
87 and document this in the SECURITY CONSIDERATIONS sections
88 of AnyEvent::TLS.
89 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
90 requested by Ron Isaacson.
91 - document $AnyEvent::DNS::RESOLVER variable.
92 - incorporated major typo patches by Abhijit Menon-Sen.
93 - AnyEvent::Handle now croaks when negative timeouts are passed.
94 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
95 undocumented, but who knows when they might come in handy.
96
975.271 Tue Jun 8 12:05:46 CEST 2010
98 - backport to perl 5.8.x.
99
1005.27 Sun Jun 6 12:12:05 CEST 2010
101 - postpone differently in AnyEvent::Socket now, as
102 when not, canceling the connection attempt might fail
103 (found by Felix Antonius Wilhelm Ostmann).
104 - explicitly check for non-stream sockets in AE::Handle, too many
105 clueless people fell into the trap of this somehow working.
106 - simplified and reworked the "OTHER MODULES" section.
107 - better/more condvar examples.
108
1095.261 Wed Apr 28 16:13:36 CEST 2010
110 - AF_INET6 was not properly used from Socket6 during configuration
111 time (found by Andy Grundman).
112
1135.26 Mon Apr 12 04:49:35 CEST 2010
114 - don't generate spurious readyness notifications when select
115 returns EINTR in the pure perl backend (analysed and testcase
116 by Bas Denissen).
117 - give same examples in SYNOPSIS sections of AnyEvent and AE.
118 - provider faster implementations for AE::time/now/now_update for
119 EV and Perl backends.
120 - provide faster AE::cv and AE::time implementations for all
121 backends.
122 - no longer support lower/mixed-case signal names (this was
123 never documented nor universally supported).
124 - some more memory µ-optimisations, and somewhat less messy
125 code to cope with both AnyEvent and AE APIs.
126
1275.251 Sat Mar 13 00:58:21 CET 2010
128 - make test failed if it was run before make install on perl <=
129 5.8.8, and older versions of perl make our live very difficult,
130 so write constants.pl during Makefile.PL time.
131
1325.25 Sat Mar 13 00:23:14 CET 2010
133 - fix a race condition in AnyEvent::Handle that would cause
134 a "bio_read: unsupported method" error (in _another_
135 ssl connection) after user code threw an exception.
136 - added AnyEvent::Handle->destroyed method.
137 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
138 - the AnyEvent::Util::fh_nonblocking function worked only
139 by lucky accident on win32.
140 - smaller and faster AnyEvent::Util::fh_nonblocking.
141 - when the (required!) Time::HiRes module is missing, AnyEvent
142 did not fall back to built-in time properly.
143 - do not load Fcntl at runtime, saving memory and loading time.
144 - precompile a number of constants and use them instead of runtime
145 detection and eval.
146 - free detection code after detection and similar memory optimisations.
147 - Perl backend timer interval best effort drift has been improved
148 (same algorithm as EV).
149 - update unicode idna mapping table.
150
1515.24 Tue Jan 5 11:39:43 CET 2010
152 - cygwin never reports errors from failed connects. 1.5 just gives
153 you continous readyness and EAGAIN, 1.7 is even more broken and
154 just hangs. work around both issues in a cygwin-specific hack.
155 - improve idle watcher documentation slightly.
156
1575.23 Sun Dec 20 23:48:00 CET 2009
158 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
159 - implement punycode_encode/decode, idn_nameprep,
160 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
161 - provide $AE::VERSION.
162 - removed traces of "no strict 'refs'".
163
1645.22 Sat Dec 5 03:51:13 CET 2009
165 - downgrade-or-fail in AnyEvent::Handle::push_write, to
166 diagnose encoding failures earlier and more succinctly.
167 (this works around bugs in perl, throwing away encoding info
168 when passing scalar data to extensions).
169 - add more examples to AnyEvent::Socket manpage.
170 - upgrade internal warning set to the same as common::sense 2.03.
171 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
172 (John Beppu).
173 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
174 - add google public dns servers to fallback server set and make sure
175 we load-balance properly between the three sets. also add all
176 fallback dns servers, not just a random one, to each dns config.
177
1785.21 Thu Nov 19 02:48:47 CET 2009
179 - fix a problem where socket constants were called with parameters
180 (spotted by David Friedland).
181 - fork_call never use'd POSIX (reported by Daisuke Maki).
182 - improve perl 5.6 compatibility further (but it still won't work
183 unless you rip out everything but the core).
184 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
185 uses 10MB of RAM, but doesn't open a console window. *sigh*.
186
1875.202 Wed Oct 14 22:35:44 CEST 2009
188 - AnyEvent::DNS would unexpectedly clobber $_ under windows
189 (analysed by Matthias Waldorf).
190 - AnyEvent::Handle::run_cmd can now pass the PID of the
191 newly-created process, which is much less useful than it might
192 sound (based on patch by Yann Kerherve).
193
1945.201 Tue Sep 29 12:09:25 CEST 2009
195 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
196 (reported by Torsten Foertsch).
197 - common::sense 2.0 could cause tcp_server to throw an exception
198 (analysed by elmex).
199
2005.2 Mon Sep 14 07:04:49 CEST 2009
201 - INCOMPATIBLE CHANGE: do no longer support register_read_type
202 and register_write_type in AnyEvent::Handle, instead support
203 package names (the facility was mostly abused).
204 - implement "packagename-as-read/write type" support in
205 AnyEvent::Handle.
206 - AnyEvent::Handle: new options "keepalive" and "oobinline".
207 - oobinline set by default to avoid security issues.
208 - the pure-perl event loop backend wrongly detected times() fallback
209 support (spotted by Pavel Boldin).
210
2115.12 Tue Sep 1 20:26:50 CEST 2009
212 - be more lenient when parsing resolv.conf files, as some people
213 use hashmarks as comment indicator inside directives (reported
214 by Michael S. Fischer).
215 - use same set of warnings as common::sense 2.0.
216 - fix a potential 32 bit overflow issue due to perl having problems
217 with large hex constants in the Perl backend.
218 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
219 by default.
220 - fix some prototypes.
221 - (optionally) require Async::Interrupt 1.02.
222 - replace opendns fallbacks by verizon/level3 public dns servers.
223
2245.112 Fri Aug 21 13:59:04 CEST 2009
225 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
226 using starttls with an illegal/uncreatable tls context.
227 - port to broken win32 perls.
228
2295.111 Sun Aug 16 18:44:36 CEST 2009
230 - perl errornously requires sockaddr_un structures returned by the
231 kernel to have a specific length. work around this bug by padding
232 sockaddr_un structures (found to fail on OS X, as perl only uses
233 the correct code for GNU/Linux, but likely to fail on other
234 OSes as well).
235
2365.11 Wed Aug 12 17:49:37 CEST 2009
237 - fixed a bug in run_cmd with input from filename.
238 - create a json coder object when none is given in write direction
239 to reduce dependency on JSON.
240
2415.1 Tue Aug 11 03:17:32 CEST 2009
242 - tcp_connect now reports the correct error in $!, not an unrelated
243 one (was broken since 4.91).
244 - AE::Handle did not properly clear rtimeout/wtimeout watchers
245 when retrying a tcp_connect.
246 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
247 - fixed wrong documentation in AE::Socket.
248
2495.01 Mon Aug 10 03:16:32 CEST 2009
250 - last release broke EV child watchers :/.
251
2525.0 Sun Aug 9 17:32:09 CEST 2009
253 - officially introduce and document the AE API.
254 - lowering the timeout at runtime did not have immediate effect
255 in AnyEvent::Handle.
256 - AE::Handle now has separate and independent read and write timeout
257 settings.
258 - probe for Guard module when AnyEvent::Util is loaded, not at
259 runtime (Event doesn't like eval "use" from callbacks, and it avoids
260 exporting the probe function).
261 - Impl::IOAsync's io watchers did not use the filehandle emulation
262 code to work around it's watcher limits.
263 - work around IO::Async's problems with overloaded objects.
264 - try to disable wearnings inside IO::Async, it's just too buggy.
265 - do no longer use Event by default, only EV or Perl, Event is too buggy.
266 - hackishly provide fast AE:: interface when EV or Perl is the backend.
267 - inherit VERSION from AnyEvent::VERSION in most modules.
268
2694.91 Thu Aug 6 15:42:45 CEST 2009
270 - AE::Handle::starttls could get out of sync when the read buffer
271 already contains some TLS handshake.
272 - AE::Handle did not properly free the TLS session, causing bigger
273 memory leaks in Net::SSLeay than expected (reported by toaster).
274 - AE::Socket::tcp_connect will now artificially delay invoking
275 the callback to avoid returning after invoking the callback.
276 - convert many internal watcher uses to AE API.
277
2784.9 Sat Aug 1 11:07:01 CEST 2009
279 - Glib silently fails when registering a timeout with
280 a negative value, so avoid that.
281 - call condvar callback immediately when it is set
282 after the condition is already signalled.
283 - check rbuf_max condition only after trying to consume
284 data.
285 - normalise signal numbers to names when using the ae's
286 signal handling, but do not document this (yet).
287 - pure perl signal emulation did not properly set nonblocking
288 mode on the signal pipe on win32 (this is unlikely to have caused
289 any issues).
290 - new module: AnyEvent::Debug.
291 - AnyEvent::Strict now checks that a signal was specified
292 by name (not name or number).
293 - reduce memory footprint in typical cases by ~50kb by
294 only compiling signal code when necessary.
295 - add AnyEvent::Handle->rbuf_max.
296 - grab the AE namespace for future new API, implement stubs for
297 said future/alternative API.
298 - new function: AnyEvent::Socket::format_hostport.
299
3004.881 Tue Jul 28 12:51:53 CEST 2009
301 - work around a bug in local in pre-5.10 perls, causing
302 AnyEvent::Handle to recurse when it should not
303 (analysed by elmex).
304
3054.88 Tue Jul 28 04:04:37 CEST 2009
306 - re-bless the handle into a dummy package after calling
307 AnyEvent::Handle::destroy, so the user does not need to check
308 for errors after every push_write etc.
309 - do not attempt to run t/02_signals.t on obviously broken
310 platforms.
311
3124.87 Sun Jul 26 02:06:16 CEST 2009
313 - do not attempt to linger when there is no longer a valid fh.
314 - enforce tls mode to be either accept or connect, do not simply
315 segfault in Net::SSLeay.
316 - AnyEvent::Handle can now call tcp_connect itself (new parameters
317 connect, on_prepare, on_connect and on_connect_error). Updated
318 tutorial accordingly.
319 - add AnyEvent::Impl::Irssi backend.
320
3214.86 Mon Jul 20 23:52:29 CEST 2009
322 - since the verbose warning is not enough, explicitly document
323 that versions before 1.33 of Net::SSLeay are not secure.
324 - work around signal handling races in Event and (...) Event::Lib.
325 - try to align signal-race timer to full-second boundaries.
326 - work around Tk not liking negative timeouts.
327 - don't complain of different grades of environmental unfriendlyness
328 in IO::Async.
329
3304.85 Sat Jul 18 06:16:14 CEST 2009
331 - nail the signal race problem in perl once and for all
332 (see $AnyEvent::MAX_SIGNAL_LATENCY).
333 - take advantage of Async::Interrupt if it is available.
334 - load Time::HiRes and Guard modules on demand only.
335 - add optional/recommended modules section to AnyEvent
336 documentation.
337 - reduce memory usage considerably (and reduce startup penalty)
338 by not using "strict", "warnings" and "overload" modules.
339 - work around buggy windows/openbsd perls and provide EBADMSG
340 and EPROTO ourselves when missing.
341 - improve perl 5.6 compatibility of the core event loop.
342 - made Net::SSLeay version 1.33 a soft requirement.
343
3444.83 Fri Jul 17 16:56:26 CEST 2009
345 - implement AnyEvent::Socket::getprotobyname.
346 - AnyEvent::CondVar's will now detect recursive blocking
347 waits and will croak, as too many people fall into
348 this trap.
349 - AnyEvent::Handle will now call ->destroy on itself after
350 executing the on_error callback, instead of doing some
351 half-baked internal shutdown, for fatal errors.
352 - clarify on_eof behaviour w.r.t. the read queue and
353 on_read callbacks.
354 - ignore some possible spurious wake-ups in tcp_connect.
355
3564.82 Sat Jul 11 00:34:55 CEST 2009
357 - POE and Event backends didn't accept some callable objects as
358 callbacks.
359 - use Config module instead of POSIX module to detect signal names
360 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
361 module doesn't even have all POSIX signals :/.
362 - use more workarounds around the many refcnt/corruption bugs in
363 Event::Lib.
364 - work around a race condition in perl's select, causing t/03_child.t
365 to rarely fail.
366
3674.81 Thu Jul 9 10:30:30 CEST 2009
368 - AnyEvent::Handle didn't properly diagnose write errors
369 (it expected -1 from syswrite, how lame... :).
370 - support file descriptors in addition to file handles
371 in AnyEvent->io.
372 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
373 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
374 and PERL_ANYEVENT_CA_PATH.
375 - provide a sensible synopsis section for AnyEvent::TLS.
376 - add a "supported backends" section to the manpage.
377 - added simple io watcher test to testsuite, using a
378 portable_socketpair.
379 - tried to improve the stability of the Event::Lib backend,
380 YMMV.
381
3824.8 Mon Jul 6 23:45:16 CEST 2009
383 - AnyEvent::DNS did not properly follow CNAME records with
384 uppercase targets.
385 - AnyEvent::DNS would errornously return AAAA records
386 with v4 mapped addresses (a faulty record) as ipv4 addresses,
387 causing AnyEvent::Socket to throw an exception.
388 - added new module AnyEvent::TLS for easier SSL/TLS context
389 creation, with many options including hostname verification,
390 secure default configuration, lots of documentation and,
391 predefined diffie-hellman keys for perfect forward security
392 and much more. get it while it's still fresh!
393 - use AnyEvent::TLS in AnyEvent::Handle for context management.
394 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
395 so AnyEvent::Socket users have smaller memory footprint
396 in the common case.
397 - add AnyEvent::Handle->push_shutdown method.
398 - add an additional error message parameter to AnyEvent::Handle's
399 on_error callback (for TLS, $! is still available).
400 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
401 - make AnyEvent::Handle more robust against early conenction
402 failures (during new), and return C<undef> in that case
403 from the constructor.
404 - AnyEvent::Handle will now try to load only JSON::XS first,
405 then fall back to JSON.
406 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
407 and because it was documented that way.
408
4094.451 Fri Jul 3 00:28:58 CEST 2009
410 - do not clear rbuf when shutting down an AnyEvent::Handle
411 object - doing so breaks AnyEvent::HTTP.
412
4134.45 Mon Jun 29 22:59:26 CEST 2009
414 - a write error could cause AnyEvent::Handle to create
415 an I/O watcher with an undefined $fh.
416 - special-case mapped ipv4 addresses in both
417 AnyEvent::Socket::format_address and parse_address,
418 to treat them just like ipv4 addresses.
419 - updated and overhauled the AnyEvent::Intro doc.
420 - implement AnyEvent::Socket::format_ipv4/ipv6.
421 - slightly speed up portable_pipe/socketpair.
422 - expand condvar begin/end documentation.
423
4244.42 Fri Jun 26 08:32:18 CEST 2009
425 - preliminary and neccesarily incomplete support for IO::Async.
426 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
427 case it was set to IGNORE, to ensure we can catch child statuses
428 even when the calling env acts stupidly.
429 - updated benchmarks with IO::Async, which performs very well.
430
4314.412 Wed Jun 24 01:35:57 CEST 2009
432 - support an "untaint" attribute for AnyEvent::DNS
433 and set it on the default resolver.
434 - implement and document AnyEvent::Impl::Perl::loop.
435 - remove all anyevent-env variables from %ENV
436 when running in tainted mode.
437 - mention and extend the IO::Lambda benchmark.
438
4394.411 Sun Jun 7 18:48:02 CEST 2009
440 - do not try to use F_SETFD on windows.
441
4424.41 Thu May 14 06:40:11 CEST 2009
443 - work around issues in older perls (5.8.0?) when
444 a signal handler is deleted from the %SIG hash.
445 - use POSIX::_exit in child test, to avoid running
446 destructors.
447 - speed up CHLD handling by relying on SIGCHLD being
448 synchronously delivered, even when we roll our own
449 implementation.
450 - AnyEvent::DNS: add the "dname" resource record name
451 for cosmetic reasons.
452
4534.4 Sun Apr 26 20:12:33 CEST 2009
454 - implemented idle watchers, where applicable.
455 - AnyEvent->time died when Event backend was in use.
456 - fix a memleak in the Tk backend.
457 - sped up Tk timer handling.
458 - clip negative "after" values to 0 in AnyEvent::Impl::Event
459 to avoid spamming out warning messages.
460 - fix Qt timers without interval.
461 - avoid Qt zero-timeout special case, in old libqt's.
462
4634.352 Mon Apr 20 16:31:11 CEST 2009
464 - fix AnyEvent::Strict error messages for child watchers.
465 - fix/update Makefile.PL "recommends" versions.
466 - add AnyEvent->now_update.
467
4684.351 Sat Apr 11 07:56:14 CEST 2009
469 - actually make the signal pipe work under win32.
470 - localise $! in signal handler to avoid changing $!, although
471 perl itself does not seem to save/restore errno either.
472 - set the cloexec flag on the signal pipe (normally set by perl too).
473
4744.35 Fri Mar 27 11:48:20 CET 2009
475 - event models relying on AnyEvent's signal watcher emulation did
476 invoke the callback asynchronously, contradicting documentation
477 and causing signals to get lost (this includes AnyEvent's own
478 event loop). AnyEvent now uses the standard pipe trick to make
479 callback execution synchronous to the event loop.
480 - AnyEvent::Handle didn't free TLS context data on DESTROY
481 (patch by Pavel Shaydo).
482 - work around the ever-incompatibly-changing API of MakeMaker.
483 - document that changing global variables without restoring them
484 is a bad idea in Perl (noted by Adam Rosenstein).
485 - AnyEvent::Strict now barfs if ->io is passed a file.
486
4874.34 Thu Feb 12 18:32:45 CET 2009
488 - separately buffer TLS read data, as otherwise the read queue could
489 deadlock as receiving data is not expected while draining the
490 read queue (which cna only happen with TLS).
491 - raise EBADMSG error on JSON decoding errors.
492 - fix some minor manpage bugs (reported by Maximilian Gaß).
493 - speed up select bitmask parsing quite a bit in the pure perl
494 backend.
495 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
496
4974.331 Tue Jan 6 21:07:25 CET 2009
498 - socketpair fails on many vista machines because vista has
499 completely broken accept/getpeername and getsockname functions,
500 so we provide our own socketpair emulation that kind of works
501 (AnyEvent::Util::portable_pipe).
502 - new function: AnyEvent::Util::portable_socketpair.
503 - take advantage of the Guard module if it exists.
504
5054.33 Fri Nov 21 02:35:40 CET 2008
506 - AnyEvent::Strict did errornously flag a fileno of 0 as
507 illegal.
508 - reduce memory usage and slightly speed up the pure perl backend
509 by only storing the file descriptor, not the file handle.
510 - add missing autocork method to AnyEvent::Handle (reported
511 by Adam Rosenstein).
512 - AnyEvent::DNS->resolve errornously documented an $rcode result
513 argument, but there isn't (spotted by Henrik Krohns).
514 - the naked truth about Tk - it's basically unmaintained.
515
5164.32 Mon Nov 3 22:46:32 CET 2008
517 - fix AnyEvent::Socket::resolve_sockaddr to properly support
518 unix sockets again and choose a proper default for
519 the socket type.
520 - fix call to resolve_sockaddr in tcp_connect.
521
5224.31 Thu Oct 30 04:41:48 CET 2008
523 - implemented AnyEvent::Handle->destroy method.
524 - hint about unexpected effects in TLS mode.
525 - speed up AnyEvent::Impl::Perl by using more arrays
526 instead of hashes without the slightest loss of readability :)
527 - work around a perl argument refcounting bug.
528 - with some perl patches applied, AnyEvent now seems leak-free.
529
5304.3 Fri Oct 3 09:18:43 CEST 2008
531 - AnyEvent will now install a no-op signal handler for SIGPIPE,
532 unless one has been installed already.
533 - warn about not loaidng AnyEvent::Impl::POE early enough
534 (patch by Adam Rosenstein).
535 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
4 - fix a bug in AnyEvent::Handle where in SSL connect mode, the 536 - fix a bug in where in SSL connect mode, the
5 client would first wait for some data by the server and 537 client would first wait for some data by the server and
6 otherwise hang. (Reported and analyzed in an absolutely 538 otherwise hang. (reported and analysed in an absolutely
7 exemplary manner by Adam Rosenstein). 539 exemplary manner by Adam Rosenstein).
8 - fix a bug in AnyEvent::Handle where SSL EOF would not be treated 540 - fix a bug in where SSL EOF would not be treated
9 as stream EOF, putting the connection into a hung state. 541 as stream EOF, putting the connection into a hung state.
10 (Reported and analyzed in an absolutely exemplary manner by Adam 542 (reported and analysed in an absolutely exemplary manner by Adam
11 Rosenstein). 543 Rosenstein).
12 - fix a potential segfault in AnyEvent::Handle when the TLS context 544 - fix a potential segfault when the TLS context
13 would go missing in a read callback (e.g. due to stoptls) - 545 would go missing in a read callback (e.g. due to stoptls) -
14 Net::SSLeay of course makes no type checking whatsoever. 546 Net::SSLeay of course makes no type checking whatsoever.
15 - AnyEvent::Handle will not stop the read watcher in TLS mode. 547 - AnyEvent::Handle will not stop the read watcher in TLS mode.
16 - AnyEvent::Handle->stoptls will send a TLS close notify 548 - AnyEvent::Handle->stoptls will send a TLS close notify
17 instead of simply destroying the stream now. 549 instead of simply destroying the stream now.
550 - fix a bug where large blocks of data written in TLS mode
551 would not be sent unless triggered by receives.
552 - on_drain will now take the tls write buffer into account.
553 - SSL operations have been streamlined a bit and should be faster
554 now, more to come, though.
555 - added a FAQ section, will document nontrivial issues in there.
18 556
194.234 Mon Sep 29 04:08:13 CEST 2008 5574.234 Mon Sep 29 04:08:13 CEST 2008
20 - fix child watcher documentation: only child exits will be reported, 558 - fix child watcher documentation: only child exits will be reported,
21 no trace events. 559 no trace events.
22 - mention SIGPIPE in AnyEvent::Handle. 560 - mention SIGPIPE in AnyEvent::Handle.

Diff Legend

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