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

Comparing AnyEvent/Changes (file contents):
Revision 1.134 by root, Thu May 29 23:15:20 2008 UTC vs.
Revision 1.594 by root, Fri Sep 5 22:17:26 2014 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
303:55:22 <Sadrak> Doku zu new: It is possible to push requests on the read and write queues, and modify properties of the stream, even
4 while
503:55:23 <Sadrak> AnyEvent::Handle is connecting.
603:55:39 <Sadrak> doku zu on_connect: It is not allowed to use the read or write queues while the handle object is connecting.
7
8TODO: document TCP_*** constants
9TODO: add debug/trace logging to some modules
10TODO: use the eval-with-cb-call trick also to speed up JSON?
11TODO: FAQ: common pitfalls? invoke-before-return
12TODO: FAQ: scope-ids? hosts file order?
13TODO: AnyEvent::Handle with AnyEvent::IO?
14TODO: invalid. and localhost. specialcasing inside AEDNS and not AESocket (rfc6761)
15TODO: maybe implement env variable to give hosts precedence
16
17 - work around a newly introduced bug in Socket 2.011 (an
18 errornous sun_length check) (analyzed by Maxime Soulé).
19 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
20 (analyzed by Ben Magistro).
21 - AnyEvent::Strict will now confess, not croak. This is in line with
22 it being a development/debugging tool.
23 - work around a number of libglib bugs (debug builds of libglib enforce
24 certain undocumented behaviour patterns such as not being able to
25 remove a child watch source after it has fired, which we will try
26 to emulate to avoid "criticals". what where they thinking...).
27 - mention json security issues in AnyEvent::Handle, now that Douglas
28 Crockford has foolishly and incompatibly changed JSON.
29 - changed default dns resolver "max_outstanding" value from 1 to 10,
30 the latter beinfg the intended value all along
31 (reported by Ilya Chesnokov).
32 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
33 (written by Mike Lowell).
34
357.07 Tue Dec 17 17:45:02 CET 2013
36 - the documentation for custom tls verify schemes was wrong. make it agree
37 with the code (reported by Maxime Soulé).
38 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
39 - work around an API change in openssl that could cause wrong tls connection
40 aborts, likely on windows only (analyzed by sten).
41 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
42 endless loop (reported by Dietrich Rebmann).
43 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
44 (patch by Maxime Soulé).
45 - document AnyEvent::Impl::IOAsync::set_loop and
46 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
47 functionality has _always_ been available.
48 - force a toplevel domain name in t/81_hosts.t (analyzed by
49 David Jack Wange Olrik).
50 - document that AnyEvent::Log uses AnyEvent::IO.
51 - warn about AnyEvent::Filesys::Notify performance.
52 - praise the joys of AnyEvent::Fork::*.
53 - time for an =encoding directive.
54 - do no longer use JSON to create a default json coder, use
55 JSON::XS or JSON::PP directly.
56
577.05 Wed Aug 21 10:38:08 CEST 2013
58 - uts46data.pl couldn't be found due to wrong naming of the file
59 (reported by Fulko Hew).
60 - handle lone \015's properly in AE::Handle's default line read
61 (reported by various people).
62 - untaint ip addresses found in /etc/hosts (patch by José Micó).
63 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
64 fail on first use (reported and testcase by Andrew Whatson).
65 - expose AnyEvent::Log::format_time, and allow users to redefine it.
66 - expose AnyEvent::Log::default_format, and allow redefinition.
67 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
68 - AnyEvent::Debug shell can now run coro shell commands, if available.
69 - t/63* tests were wrongly in MANIFEST.
70 - kernel.org's finger server went MIA, switch to freebsd.org and
71 icculus.org.
72 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
73 to function asynchronously (reported by Daniel Carosone).
74 - hard-disable $^W in most tests, it generates too much garbage output.
75 - use a (hopefully) more future-proof method to emulate common::sense.
76 - upgrade to UTS-46:6.2.0.
77 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
78 documented. should not affect anything.
79
807.04 Thu Nov 15 02:23:58 CET 2012
81 - debugging code left in.
82
837.03 Thu Nov 15 02:16:11 CET 2012
84 - AnyEvent::Socket::inet_aton did not work when dns resolution was
85 used to find the addresses (analyzed and patch by Maxime Soulé).
86 - fix a memory leak in the /etc/hosts lookup code when hosts don't
87 resolve and are not in hosts (analyzed by sten).
88
897.02 Tue Aug 14 04:27:58 CEST 2012
90 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
91 program by Yorhel).
92 - verified that AnyEvent::Socket follows RFC5952.
93 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
94
957.01 Sun May 13 01:03:17 CEST 2012
96 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
97 not available, instead of throwing an exception.
98 - use File::Spec to get the tmpdir in t/*, to avoid needless
99 failures on (most, not mine :) windows boxes.
100 - new handle read types: tls_detect and tls_autostart.
101
1027.0 Fri Apr 13 06:33:30 CEST 2012
103 - child watchers are broken in POE 1.352 (also many earlier
104 versions) and there seems to be no way to work around it, as POE
105 itself is inherently racy. Document this regression and add a delay
106 in t/68_poe_03_child.t for the time being.
107 - new module AnyEvent::IO, that is a frontend to either a pure-perl
108 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
109 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
110 - load /etc/hosts only when DNS has no answer.
111 - stat /etc/hosts on every access and reload it if it changed.
112 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
113 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
114 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
115 a side effect, they now use true appending as opposed to libc
116 appending, and the file might not have been opened when the function
117 returns.
118 - the default logging level was not properly documented in a variety of
119 places, this has hopefully been rectified.
120 - updated uts46data.pl for unicode 6.1.0.
121 - made log messages generated by AnyEvent submodules not
122 include the package name anymore, as it will be logged
123 by default already.
124 - upgrade to the trick used by common-sense 3.5 to work
125 around extra warning torture/breakage under perl 5.15.x.
126 - log messages by anyevent are now ucfirst, are usually full
127 sentences and do no longer include the package name.
128 - the storable read type would sometimes throw an exception instead
129 of causing EBADMSG (analyzed by Maxime Soulé).
130 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
131 keysize (reported by Paul Howarth).
132 - documented Rocco fabricating statements about AnyEvent and me.
133
1346.14 Tue Jan 31 20:00:24 CET 2012
135 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
136 probably an editing mistake (reported by Darin McBride).
137
1386.13 Thu Jan 12 07:27:01 CET 2012
139 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
140 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
141 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
142 regex-matching inside the callback would change the parameter, despite
143 $1 being dynamically scoped per-block. this perl bug is now being
144 worked around (testcase by Cindy Wang).
145
1466.12 Mon Dec 12 13:21:10 CET 2011
147 - $! was clobbered by subsequent calls in tcp_connect, due to
148 the postpone not saving/restoring it (analyzed by Richard Garnish).
149
1506.11 Tue Nov 22 10:36:05 CET 2011
151 - Tk cannot create windows when tainted unless you set a title, so
152 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
153 - escape any nonprintable/nonascii characters when stringifying
154 backtraces.
155 - log the reason loading an interface module fails to load at level debug.
156
1576.1 Tue Oct 4 19:44:30 CEST 2011
158 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
159 and above), and some messages inside AnyEvent have been elevated
160 to higher log levels to print by default.
161 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
162 - work around yet another signal race bug, newly introduced by POE. My
163 how that module sucks.
164 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
165 support ipv6-only operation even with defaults (also use ::1 instead
166 of 127.0.0.1).
167 - give better guidelines on how to chose a logging level.
168 - AnyEvent::Log can now cap (limit) log message priority by context.
169 - AE::log will now save and restore $! and $@, so callers don't have to.
170 - AE::Strict now checks that registered fds are still valid at regular
171 intervals.
172 - skip frequently called 'require' calls if the module appears loaded.
173 - increase signal latency test timer considerably for poor little osx,
174 which can't be bothered with basic posix functionality such as signal
175 delivery.
176 - mention some debugging aids in the tutorial.
177 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
178 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
179 of the context as title.
180 - AnyEvent::Strict would not restore read-write access to $_ when a callback
181 died - this is a user bug, but since AE::Strict exists to help dbeugging...
182 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
183 to work with it - the author is well aware of the issues but refuses
184 to fix them. This extreme step has been taken because the module actively
185 attacks the goals of AnyEvent and makes it harder for module authors to use
186 AnyEvent.
187
1886.02 Fri Aug 26 20:08:31 CEST 2011
189 - log_to_syslog now accepts facility strings.
190 - log_to_syslog would not allow facilities - "$facility|$level"
191 *literally* means that string. sheesh.
192 - the CPAN parser got confused and thought we provide Sys::Syslog, try
193 to work around.
194 - renamed AnyEvent::Impl::FLTK2 to FLTK.
195
1966.01 Fri Aug 26 07:04:11 CEST 2011
197 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
198 the TTL as fourth element - this affects ->resolve, ->request
199 and ->dns_unpack indirectly. Convenience functions such as
200 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
201 AnyEvent::Socket.
202 - INCOMPATIBLE CHANGE: backend authors now should not implement
203 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
204 methods.
205 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
206 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
207 - DEPRECIATION: please change
208 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
209 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
210 in your programs, for future compatibility.
211 - *blush* idle watcher emulation was borked.
212 - AnyEvent::Strict now checks the AE::xxx functions as well.
213 - querying too long domain names would barf deep inside AnyEvent::DNS - since
214 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
215 for too long domain names explicitly and passes an undef to the callback
216 instead (reported by James Bromberger).
217 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
218 - new functions: AnyEvent::postpone, AnyEvent::log.
219 - new module: AnyEvent::Log, for simple logging needs.
220 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
221 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
222 a debug shell.
223 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
224 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
225 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
226 /etc/hosts can be found.
227 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
228 configuration.
229 - env variables can now be specified using AE_xyz instead of
230 PERL_ANYEVENT_xyz, subject to some rules.
231 - lots of new AnyEvent::Debug functionality.
232 - AnyEvent::Strict now checks for modifications of $_ by the callback
233 (by setting it to readonly while executing the callback, if possible).
234 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
235 Invest extra overhead to work around this peculiar behaviour.
236 - split the pure perl backend into a loop module and an impl
237 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
238 - some read types (regex, netstring, json, storable) could
239 errornously keep an AE::Handle object alive on their own
240 (reported by Mohammad Toossi).
241 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
242 supports /etc/hosts overrides.
243 - add more workarounds around child watcher bugs in IO::Async
244 and (newly introduced) in POE.
245 - use glib child watchers and try to work around its limitations,
246 if possible (based on a patch by Kevin Ryde).
247 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
248 - many read types in AnyEvent::Handle would malfunction during
249 transitions to TLS because of a "delete $self->{rbuf}".
250 (reported by Gerald Galster).
251 - improve condvar blocking wait performance (probably only noticable
252 with faster backends), also make it easier for event loops that
253 do not support blocking waits.
254 - PERL_ANYEVENT_MODEL now also allows full module names, although
255 the usefulness of this feature is not yet clear.
256 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
257 - add t/80_ssltext.t.
258 - no longer use AUTOLOAD for the initial loading (for no very good reason).
259 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
260 - do not run event loop detection multiple times when programs call their cached
261 copies of AnyEvent::detect.
262 - parse_hostport now accepts absolute paths als unix domain sockets.
263 - suppress the idiotic warning inside POE by patching the POE::Kernel
264 run flags instead of asking the user to jump through contortions.
265 - add Task::Weaken as dependency (as recommends only at this time).
266 - replace PF_UNSPEC by 0, for good old minix.
267
2686.0 (not officially released, but was on CPAN for about 15 minutes :/)
269
2705.34 Thu May 12 09:59:41 CEST 2011
271 - AE::TLS didn't support both cert_file and key at the same time,
272 due to some copy&paste bug (reported and analyzed by vti).
273 - improve AE::Handle timeout documentation slightly.
274
2755.33 Mon Feb 7 21:51:41 CET 2011
276 - never appeared on CPAN, for unknown reasons.
277 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
278 object when it's already gone (analyzed by Ryan Bullock).
279 - update idna algorithm(s) to UTS#46 version 6.0.0.
280 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
281 uts46 says it's an error, and testsuite makes more sense.
282 - remove leading dots in idn_to_ascii - testsuite says to remove,
283 uts46 says it's an error, and testsuite makes more sense.
284 - fixed a bug in idn ignored character processing.
285 - passes the uts46 6.0.0 testsuite except for three cases which
286 apparently are bugs in the testsuite itself.
287 - passes the uts46 6.0.1 testsuite except for eleven cases which
288 apparently are bugs in the testsuite itself.
289
2905.31 Sun Jan 23 11:49:19 CET 2011
291 - work around a POE bug where POE would first poll for new events and
292 invoke handlers BEFORE handling the message queue.
293 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
294 - stop_read now does just that in TLS mode. see its
295 documentation for caveats.
296 - update to common::sense 3.4 warning set (some warnings in perl impose
297 an arbitrary 32kb limit on regex matches, which affects
298 AnyEvent::Handle).
299 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
300 - new experimental AE::Handle::resettls method.
301
3025.3 Fri Dec 31 05:48:13 CET 2010
303 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
304 issues with IO::Async have been solved in that version, including
305 the availabilty of a default loop!
306 - implement new wbuf_max parameter for AnyEvent::Handle.
307 - added Cocoa::EventLoop backend by Daisuke Murase.
308 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
309 - clarify handle on_prepare documentation.
310 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
311 to avoid leaking them to any child processes.
312 - try to force files into the arch-dependend lib directory, for the
313 half-"turly obsessive compulsive person"(s) that depend on it.
314
3155.29 Sun Dec 5 10:49:21 CET 2010
316 - convert EV backend to EV 4.00 API (so better upgrade EV too).
317 - AE::Handle: implement dynamic read_size adjustment, add
318 max_read_size parameter, reduce default read_size to 2048.
319 - add some FAQ entries for servers and/or on_eof vs. on_error.
320 - work around OS bugs (cygwin again) when lingering in AE::Handle.
321 - reduce memory usage slightly when connect was used in AE::Handle.
322 - wrok around more segfaults in Tk.
323 - document tls_ctx => undef as valid.
324 - detect WNOHANG value at compiletime.
325 - include some probably linux-only support for building AnyEvent
326 as part of the perl core.
327 - improve compatibility of t/08_idna.t to perl 5.8.
328 - make AnyEvent::Strict truly optional (in case somebody builds a
329 minimal perl...).
330
3315.28 Wed Oct 13 04:14:23 CEST 2010
332 - due to a glitch, AnyEvent's internal getprotobyname wasn't
333 actually used in tcp_server.
334 - implement AnyEvent::CondVar->new method, which was documented but
335 missing.
336 - support autoloading of read types in unshift_read.
337 - AnyEvent::Strict: do the test for invalid fh arguments differently,
338 resulting in more informative error messages.
339 - new AnyEvent::FAQ manpage.
340 - clarify that you can actually call push_read etc. in on_read.
341 - put keys and certs in /tmp, not into the current directory,
342 and document this in the SECURITY CONSIDERATIONS sections
343 of AnyEvent::TLS.
344 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
345 requested by Ron Isaacson.
346 - document $AnyEvent::DNS::RESOLVER variable.
347 - incorporated major typo patches by Abhijit Menon-Sen.
348 - AnyEvent::Handle now croaks when negative timeouts are passed.
349 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
350 undocumented, but who knows when they might come in handy.
351
3525.271 Tue Jun 8 12:05:46 CEST 2010
353 - backport to perl 5.8.x.
354
3555.27 Sun Jun 6 12:12:05 CEST 2010
356 - postpone differently in AnyEvent::Socket now, as
357 when not, canceling the connection attempt might fail
358 (found by Felix Antonius Wilhelm Ostmann).
359 - explicitly check for non-stream sockets in AE::Handle, too many
360 clueless people fell into the trap of this somehow working.
361 - simplified and reworked the "OTHER MODULES" section.
362 - better/more condvar examples.
363
3645.261 Wed Apr 28 16:13:36 CEST 2010
365 - AF_INET6 was not properly used from Socket6 during configuration
366 time (found by Andy Grundman).
367
3685.26 Mon Apr 12 04:49:35 CEST 2010
369 - don't generate spurious readyness notifications when select
370 returns EINTR in the pure perl backend (analysed and testcase
371 by Bas Denissen).
372 - give same examples in SYNOPSIS sections of AnyEvent and AE.
373 - provider faster implementations for AE::time/now/now_update for
374 EV and Perl backends.
375 - provide faster AE::cv and AE::time implementations for all
376 backends.
377 - no longer support lower/mixed-case signal names (this was
378 never documented nor universally supported).
379 - some more memory µ-optimisations, and somewhat less messy
380 code to cope with both AnyEvent and AE APIs.
381
3825.251 Sat Mar 13 00:58:21 CET 2010
383 - make test failed if it was run before make install on perl <=
384 5.8.8, and older versions of perl make our live very difficult,
385 so write constants.pl during Makefile.PL time.
386
3875.25 Sat Mar 13 00:23:14 CET 2010
388 - fix a race condition in AnyEvent::Handle that would cause
389 a "bio_read: unsupported method" error (in _another_
390 ssl connection) after user code threw an exception.
391 - added AnyEvent::Handle->destroyed method.
392 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
393 - the AnyEvent::Util::fh_nonblocking function worked only
394 by lucky accident on win32.
395 - smaller and faster AnyEvent::Util::fh_nonblocking.
396 - when the (required!) Time::HiRes module is missing, AnyEvent
397 did not fall back to built-in time properly.
398 - do not load Fcntl at runtime, saving memory and loading time.
399 - precompile a number of constants and use them instead of runtime
400 detection and eval.
401 - free detection code after detection and similar memory optimisations.
402 - Perl backend timer interval best effort drift has been improved
403 (same algorithm as EV).
404 - update unicode idna mapping table.
405
4065.24 Tue Jan 5 11:39:43 CET 2010
407 - cygwin never reports errors from failed connects. 1.5 just gives
408 you continous readyness and EAGAIN, 1.7 is even more broken and
409 just hangs. work around both issues in a cygwin-specific hack.
410 - improve idle watcher documentation slightly.
411
4125.23 Sun Dec 20 23:48:00 CET 2009
413 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
414 - implement punycode_encode/decode, idn_nameprep,
415 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
416 - provide $AE::VERSION.
417 - removed traces of "no strict 'refs'".
418
4195.22 Sat Dec 5 03:51:13 CET 2009
420 - downgrade-or-fail in AnyEvent::Handle::push_write, to
421 diagnose encoding failures earlier and more succinctly.
422 (this works around bugs in perl, throwing away encoding info
423 when passing scalar data to extensions).
424 - add more examples to AnyEvent::Socket manpage.
425 - upgrade internal warning set to the same as common::sense 2.03.
426 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
427 (John Beppu).
428 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
429 - add google public dns servers to fallback server set and make sure
430 we load-balance properly between the three sets. also add all
431 fallback dns servers, not just a random one, to each dns config.
432
4335.21 Thu Nov 19 02:48:47 CET 2009
434 - fix a problem where socket constants were called with parameters
435 (spotted by David Friedland).
436 - fork_call never use'd POSIX (reported by Daisuke Maki).
437 - improve perl 5.6 compatibility further (but it still won't work
438 unless you rip out everything but the core).
439 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
440 uses 10MB of RAM, but doesn't open a console window. *sigh*.
441
4425.202 Wed Oct 14 22:35:44 CEST 2009
443 - AnyEvent::DNS would unexpectedly clobber $_ under windows
444 (analysed by Matthias Waldorf).
445 - AnyEvent::Handle::run_cmd can now pass the PID of the
446 newly-created process, which is much less useful than it might
447 sound (based on patch by Yann Kerherve).
448
4495.201 Tue Sep 29 12:09:25 CEST 2009
450 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
451 (reported by Torsten Foertsch).
452 - common::sense 2.0 could cause tcp_server to throw an exception
453 (analysed by elmex).
454
4555.2 Mon Sep 14 07:04:49 CEST 2009
456 - INCOMPATIBLE CHANGE: do no longer support register_read_type
457 and register_write_type in AnyEvent::Handle, instead support
458 package names (the facility was mostly abused).
459 - implement "packagename-as-read/write type" support in
460 AnyEvent::Handle.
461 - AnyEvent::Handle: new options "keepalive" and "oobinline".
462 - oobinline set by default to avoid security issues.
463 - the pure-perl event loop backend wrongly detected times() fallback
464 support (spotted by Pavel Boldin).
465
4665.12 Tue Sep 1 20:26:50 CEST 2009
467 - be more lenient when parsing resolv.conf files, as some people
468 use hashmarks as comment indicator inside directives (reported
469 by Michael S. Fischer).
470 - use same set of warnings as common::sense 2.0.
471 - fix a potential 32 bit overflow issue due to perl having problems
472 with large hex constants in the Perl backend.
473 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
474 by default.
475 - fix some prototypes.
476 - (optionally) require Async::Interrupt 1.02.
477 - replace opendns fallbacks by verizon/level3 public dns servers.
478
4795.112 Fri Aug 21 13:59:04 CEST 2009
480 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
481 using starttls with an illegal/uncreatable tls context.
482 - port to broken win32 perls.
483
4845.111 Sun Aug 16 18:44:36 CEST 2009
485 - perl errornously requires sockaddr_un structures returned by the
486 kernel to have a specific length. work around this bug by padding
487 sockaddr_un structures (found to fail on OS X, as perl only uses
488 the correct code for GNU/Linux, but likely to fail on other
489 OSes as well).
490
4915.11 Wed Aug 12 17:49:37 CEST 2009
492 - fixed a bug in run_cmd with input from filename.
493 - create a json coder object when none is given in write direction
494 to reduce dependency on JSON.
495
4965.1 Tue Aug 11 03:17:32 CEST 2009
497 - tcp_connect now reports the correct error in $!, not an unrelated
498 one (was broken since 4.91).
499 - AE::Handle did not properly clear rtimeout/wtimeout watchers
500 when retrying a tcp_connect.
501 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
502 - fixed wrong documentation in AE::Socket.
503
5045.01 Mon Aug 10 03:16:32 CEST 2009
505 - last release broke EV child watchers :/.
506
5075.0 Sun Aug 9 17:32:09 CEST 2009
508 - officially introduce and document the AE API.
509 - lowering the timeout at runtime did not have immediate effect
510 in AnyEvent::Handle.
511 - AE::Handle now has separate and independent read and write timeout
512 settings.
513 - probe for Guard module when AnyEvent::Util is loaded, not at
514 runtime (Event doesn't like eval "use" from callbacks, and it avoids
515 exporting the probe function).
516 - Impl::IOAsync's io watchers did not use the filehandle emulation
517 code to work around it's watcher limits.
518 - work around IO::Async's problems with overloaded objects.
519 - try to disable wearnings inside IO::Async, it's just too buggy.
520 - do no longer use Event by default, only EV or Perl, Event is too buggy.
521 - hackishly provide fast AE:: interface when EV or Perl is the backend.
522 - inherit VERSION from AnyEvent::VERSION in most modules.
523
5244.91 Thu Aug 6 15:42:45 CEST 2009
525 - AE::Handle::starttls could get out of sync when the read buffer
526 already contains some TLS handshake.
527 - AE::Handle did not properly free the TLS session, causing bigger
528 memory leaks in Net::SSLeay than expected (reported by toaster).
529 - AE::Socket::tcp_connect will now artificially delay invoking
530 the callback to avoid returning after invoking the callback.
531 - convert many internal watcher uses to AE API.
532
5334.9 Sat Aug 1 11:07:01 CEST 2009
534 - Glib silently fails when registering a timeout with
535 a negative value, so avoid that.
536 - call condvar callback immediately when it is set
537 after the condition is already signalled.
538 - check rbuf_max condition only after trying to consume
539 data.
540 - normalise signal numbers to names when using the ae's
541 signal handling, but do not document this (yet).
542 - pure perl signal emulation did not properly set nonblocking
543 mode on the signal pipe on win32 (this is unlikely to have caused
544 any issues).
545 - new module: AnyEvent::Debug.
546 - AnyEvent::Strict now checks that a signal was specified
547 by name (not name or number).
548 - reduce memory footprint in typical cases by ~50kb by
549 only compiling signal code when necessary.
550 - add AnyEvent::Handle->rbuf_max.
551 - grab the AE namespace for future new API, implement stubs for
552 said future/alternative API.
553 - new function: AnyEvent::Socket::format_hostport.
554
5554.881 Tue Jul 28 12:51:53 CEST 2009
556 - work around a bug in local in pre-5.10 perls, causing
557 AnyEvent::Handle to recurse when it should not
558 (analysed by elmex).
559
5604.88 Tue Jul 28 04:04:37 CEST 2009
561 - re-bless the handle into a dummy package after calling
562 AnyEvent::Handle::destroy, so the user does not need to check
563 for errors after every push_write etc.
564 - do not attempt to run t/02_signals.t on obviously broken
565 platforms.
566
5674.87 Sun Jul 26 02:06:16 CEST 2009
568 - do not attempt to linger when there is no longer a valid fh.
569 - enforce tls mode to be either accept or connect, do not simply
570 segfault in Net::SSLeay.
571 - AnyEvent::Handle can now call tcp_connect itself (new parameters
572 connect, on_prepare, on_connect and on_connect_error). Updated
573 tutorial accordingly.
574 - add AnyEvent::Impl::Irssi backend.
575
5764.86 Mon Jul 20 23:52:29 CEST 2009
577 - since the verbose warning is not enough, explicitly document
578 that versions before 1.33 of Net::SSLeay are not secure.
579 - work around signal handling races in Event and (...) Event::Lib.
580 - try to align signal-race timer to full-second boundaries.
581 - work around Tk not liking negative timeouts.
582 - don't complain of different grades of environmental unfriendlyness
583 in IO::Async.
584
5854.85 Sat Jul 18 06:16:14 CEST 2009
586 - nail the signal race problem in perl once and for all
587 (see $AnyEvent::MAX_SIGNAL_LATENCY).
588 - take advantage of Async::Interrupt if it is available.
589 - load Time::HiRes and Guard modules on demand only.
590 - add optional/recommended modules section to AnyEvent
591 documentation.
592 - reduce memory usage considerably (and reduce startup penalty)
593 by not using "strict", "warnings" and "overload" modules.
594 - work around buggy windows/openbsd perls and provide EBADMSG
595 and EPROTO ourselves when missing.
596 - improve perl 5.6 compatibility of the core event loop.
597 - made Net::SSLeay version 1.33 a soft requirement.
598
5994.83 Fri Jul 17 16:56:26 CEST 2009
600 - implement AnyEvent::Socket::getprotobyname.
601 - AnyEvent::CondVar's will now detect recursive blocking
602 waits and will croak, as too many people fall into
603 this trap.
604 - AnyEvent::Handle will now call ->destroy on itself after
605 executing the on_error callback, instead of doing some
606 half-baked internal shutdown, for fatal errors.
607 - clarify on_eof behaviour w.r.t. the read queue and
608 on_read callbacks.
609 - ignore some possible spurious wake-ups in tcp_connect.
610
6114.82 Sat Jul 11 00:34:55 CEST 2009
612 - POE and Event backends didn't accept some callable objects as
613 callbacks.
614 - use Config module instead of POSIX module to detect signal names
615 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
616 module doesn't even have all POSIX signals :/.
617 - use more workarounds around the many refcnt/corruption bugs in
618 Event::Lib.
619 - work around a race condition in perl's select, causing t/03_child.t
620 to rarely fail.
621
6224.81 Thu Jul 9 10:30:30 CEST 2009
623 - AnyEvent::Handle didn't properly diagnose write errors
624 (it expected -1 from syswrite, how lame... :).
625 - support file descriptors in addition to file handles
626 in AnyEvent->io.
627 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
628 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
629 and PERL_ANYEVENT_CA_PATH.
630 - provide a sensible synopsis section for AnyEvent::TLS.
631 - add a "supported backends" section to the manpage.
632 - added simple io watcher test to testsuite, using a
633 portable_socketpair.
634 - tried to improve the stability of the Event::Lib backend,
635 YMMV.
636
6374.8 Mon Jul 6 23:45:16 CEST 2009
638 - AnyEvent::DNS did not properly follow CNAME records with
639 uppercase targets.
640 - AnyEvent::DNS would errornously return AAAA records
641 with v4 mapped addresses (a faulty record) as ipv4 addresses,
642 causing AnyEvent::Socket to throw an exception.
643 - added new module AnyEvent::TLS for easier SSL/TLS context
644 creation, with many options including hostname verification,
645 secure default configuration, lots of documentation and,
646 predefined diffie-hellman keys for perfect forward security
647 and much more. get it while it's still fresh!
648 - use AnyEvent::TLS in AnyEvent::Handle for context management.
649 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
650 so AnyEvent::Socket users have smaller memory footprint
651 in the common case.
652 - add AnyEvent::Handle->push_shutdown method.
653 - add an additional error message parameter to AnyEvent::Handle's
654 on_error callback (for TLS, $! is still available).
655 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
656 - make AnyEvent::Handle more robust against early conenction
657 failures (during new), and return C<undef> in that case
658 from the constructor.
659 - AnyEvent::Handle will now try to load only JSON::XS first,
660 then fall back to JSON.
661 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
662 and because it was documented that way.
663
6644.451 Fri Jul 3 00:28:58 CEST 2009
665 - do not clear rbuf when shutting down an AnyEvent::Handle
666 object - doing so breaks AnyEvent::HTTP.
667
6684.45 Mon Jun 29 22:59:26 CEST 2009
669 - a write error could cause AnyEvent::Handle to create
670 an I/O watcher with an undefined $fh.
671 - special-case mapped ipv4 addresses in both
672 AnyEvent::Socket::format_address and parse_address,
673 to treat them just like ipv4 addresses.
674 - updated and overhauled the AnyEvent::Intro doc.
675 - implement AnyEvent::Socket::format_ipv4/ipv6.
676 - slightly speed up portable_pipe/socketpair.
677 - expand condvar begin/end documentation.
678
6794.42 Fri Jun 26 08:32:18 CEST 2009
680 - preliminary and neccesarily incomplete support for IO::Async.
681 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
682 case it was set to IGNORE, to ensure we can catch child statuses
683 even when the calling env acts stupidly.
684 - updated benchmarks with IO::Async, which performs very well.
685
6864.412 Wed Jun 24 01:35:57 CEST 2009
687 - support an "untaint" attribute for AnyEvent::DNS
688 and set it on the default resolver.
689 - implement and document AnyEvent::Impl::Perl::loop.
690 - remove all anyevent-env variables from %ENV
691 when running in tainted mode.
692 - mention and extend the IO::Lambda benchmark.
693
6944.411 Sun Jun 7 18:48:02 CEST 2009
695 - do not try to use F_SETFD on windows.
696
6974.41 Thu May 14 06:40:11 CEST 2009
698 - work around issues in older perls (5.8.0?) when
699 a signal handler is deleted from the %SIG hash.
700 - use POSIX::_exit in child test, to avoid running
701 destructors.
702 - speed up CHLD handling by relying on SIGCHLD being
703 synchronously delivered, even when we roll our own
704 implementation.
705 - AnyEvent::DNS: add the "dname" resource record name
706 for cosmetic reasons.
707
7084.4 Sun Apr 26 20:12:33 CEST 2009
709 - implemented idle watchers, where applicable.
710 - AnyEvent->time died when Event backend was in use.
711 - fix a memleak in the Tk backend.
712 - sped up Tk timer handling.
713 - clip negative "after" values to 0 in AnyEvent::Impl::Event
714 to avoid spamming out warning messages.
715 - fix Qt timers without interval.
716 - avoid Qt zero-timeout special case, in old libqt's.
717
7184.352 Mon Apr 20 16:31:11 CEST 2009
719 - fix AnyEvent::Strict error messages for child watchers.
720 - fix/update Makefile.PL "recommends" versions.
721 - add AnyEvent->now_update.
722
7234.351 Sat Apr 11 07:56:14 CEST 2009
724 - actually make the signal pipe work under win32.
725 - localise $! in signal handler to avoid changing $!, although
726 perl itself does not seem to save/restore errno either.
727 - set the cloexec flag on the signal pipe (normally set by perl too).
728
7294.35 Fri Mar 27 11:48:20 CET 2009
730 - event models relying on AnyEvent's signal watcher emulation did
731 invoke the callback asynchronously, contradicting documentation
732 and causing signals to get lost (this includes AnyEvent's own
733 event loop). AnyEvent now uses the standard pipe trick to make
734 callback execution synchronous to the event loop.
735 - AnyEvent::Handle didn't free TLS context data on DESTROY
736 (patch by Pavel Shaydo).
737 - work around the ever-incompatibly-changing API of MakeMaker.
738 - document that changing global variables without restoring them
739 is a bad idea in Perl (noted by Adam Rosenstein).
740 - AnyEvent::Strict now barfs if ->io is passed a file.
741
7424.34 Thu Feb 12 18:32:45 CET 2009
743 - separately buffer TLS read data, as otherwise the read queue could
744 deadlock as receiving data is not expected while draining the
745 read queue (which cna only happen with TLS).
746 - raise EBADMSG error on JSON decoding errors.
747 - fix some minor manpage bugs (reported by Maximilian Gaß).
748 - speed up select bitmask parsing quite a bit in the pure perl
749 backend.
750 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
751
7524.331 Tue Jan 6 21:07:25 CET 2009
753 - socketpair fails on many vista machines because vista has
754 completely broken accept/getpeername and getsockname functions,
755 so we provide our own socketpair emulation that kind of works
756 (AnyEvent::Util::portable_pipe).
757 - new function: AnyEvent::Util::portable_socketpair.
758 - take advantage of the Guard module if it exists.
759
7604.33 Fri Nov 21 02:35:40 CET 2008
761 - AnyEvent::Strict did errornously flag a fileno of 0 as
762 illegal.
763 - reduce memory usage and slightly speed up the pure perl backend
764 by only storing the file descriptor, not the file handle.
765 - add missing autocork method to AnyEvent::Handle (reported
766 by Adam Rosenstein).
767 - AnyEvent::DNS->resolve errornously documented an $rcode result
768 argument, but there isn't (spotted by Henrik Krohns).
769 - the naked truth about Tk - it's basically unmaintained.
770
7714.32 Mon Nov 3 22:46:32 CET 2008
772 - fix AnyEvent::Socket::resolve_sockaddr to properly support
773 unix sockets again and choose a proper default for
774 the socket type.
775 - fix call to resolve_sockaddr in tcp_connect.
776
7774.31 Thu Oct 30 04:41:48 CET 2008
778 - implemented AnyEvent::Handle->destroy method.
779 - hint about unexpected effects in TLS mode.
780 - speed up AnyEvent::Impl::Perl by using more arrays
781 instead of hashes without the slightest loss of readability :)
782 - work around a perl argument refcounting bug.
783 - with some perl patches applied, AnyEvent now seems leak-free.
784
7854.3 Fri Oct 3 09:18:43 CEST 2008
786 - AnyEvent will now install a no-op signal handler for SIGPIPE,
787 unless one has been installed already.
788 - warn about not loaidng AnyEvent::Impl::POE early enough
789 (patch by Adam Rosenstein).
790 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
791 - fix a bug in where in SSL connect mode, the
792 client would first wait for some data by the server and
793 otherwise hang. (reported and analysed in an absolutely
794 exemplary manner by Adam Rosenstein).
795 - fix a bug in where SSL EOF would not be treated
796 as stream EOF, putting the connection into a hung state.
797 (reported and analysed in an absolutely exemplary manner by Adam
798 Rosenstein).
799 - fix a potential segfault when the TLS context
800 would go missing in a read callback (e.g. due to stoptls) -
801 Net::SSLeay of course makes no type checking whatsoever.
802 - AnyEvent::Handle will not stop the read watcher in TLS mode.
803 - AnyEvent::Handle->stoptls will send a TLS close notify
804 instead of simply destroying the stream now.
805 - fix a bug where large blocks of data written in TLS mode
806 would not be sent unless triggered by receives.
807 - on_drain will now take the tls write buffer into account.
808 - SSL operations have been streamlined a bit and should be faster
809 now, more to come, though.
810 - added a FAQ section, will document nontrivial issues in there.
811
8124.234 Mon Sep 29 04:08:13 CEST 2008
813 - fix child watcher documentation: only child exits will be reported,
814 no trace events.
815 - mention SIGPIPE in AnyEvent::Handle.
816 - perl backend will now use POSIX::times as monotonic clock
817 when available and the monotonic clock option is not.
818 - run even without Time::HiRes in some cases now, but do not
819 provide sub-second accuracy (all *supported* perl versions have
820 Time::HiRes).
821
8224.233 Fri Aug 22 01:48:05 CEST 2008
823 - fix a bug in the testsuite, causing a hang.
824 - clarified lots of AnyEvent::Handle constructor arguments.
825
8264.232 Thu Aug 21 20:44:25 CEST 2008
827 - fix a bug in AnyEvent::Handle that could cause two fatal error
828 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
829 (testcase by Leon Brocard).
830 - clarified on_eof/on_error documentation a bit.
831
8324.231 Tue Jul 29 13:12:15 CEST 2008
833 - remove some debugging code left in AnyEvent::Util::fork_call
834 (and no, it's impossible to implement with the broken windows
835 perls without resource leaks or worse).
836
8374.23 Tue Jul 29 12:19:59 CEST 2008
838 - document the first parameter passed to condvar callbacks to be
839 the callback.
840 - add AnyEvent::Socket::{ntoa,aton} aliases.
841 - optimize the AE::Handle->push_read (line) for the default
842 eol marker.
843 - optimize push_read (packstring|storable) for small packets.
844 - invoke on_error callback when no on_eof callback is set.
845 - fix a bug in push_read (storable) of unknown impact.
846
8474.22 Sun Jul 20 16:34:13 CEST 2008
848 - new function AnyEvent::Socket::parse_hostport.
849 - as the bulkheads at microsoft can't even get getprotobyname reliably
850 working on their shitty fucking broken os we need to hardcode
851 some common protocol numbers in AnyEvent::Socket. How can
852 people even bother with such a pile of shit as windows.
853
8544.21 Thu Jul 17 14:40:05 CEST 2008
855 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
856 which is incompatible to previous invocation syntax.
857 - work around more windows perl bugs in fork_call. I cannot imagine
858 why anybody would bother implementing fork in such an obviously
859 unusable way.
860 - avoid calling $do_retry if the request has already finished in
861 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
862 K. Sharifulin).
863
8644.2 Sat Jul 12 22:42:11 CEST 2008
865 - support an "interval" argument for repeating timers.
866 - fix a bug in the pure perl event loop that caused timers to
867 be delayed under some circumstances.
868 - implement optional argument checking via PERL_ANYEVENT_STRICT.
869 - not importing anything from Event causes it to malfunction
870 and not use Time::HiRes, work around this bug.
871 - more bug workarounds against the endlessly broken Tk module,
872 seems to work now on 32 bit machines, but less so on 64 bit.
873 - minor optimisations applied to most backends.
874
8754.161 Fri Jul 4 14:24:48 CEST 2008
876 - fixed the t/handle/02_write.t, it read after write, with the
877 assumption that all data could be written before read
878 (reported by Jim Razmus).
879
8804.160 Thu Jul 3 04:02:21 CEST 2008
881 - re-registering signal watchers after unregistering one did
882 not work if a backend falls back on AnyEvent's default
883 implementation (also affected child watchers).
884 - new AnyEvent::Handle options: autocork and no_delay.
885
8864.152 Sun Jun 22 14:15:44 CEST 2008
887 - allow for 32-bit perls that implement shifts differently
888 on different architectures in parse_ipv4 (reported and
889 analysed by Keiichi DAIBA).
890
8914.151 Fri Jun 6 17:34:24 CEST 2008
892 - make sure specifying _only_ on_read and never pushing reads
893 works.
894
8954.15 Fri Jun 6 13:00:46 CEST 2008
896 - the pure perl backend would keep some watchers alive when more than
897 one watcher was registered for the same fd.
898 - new "packstring" and "storable" read and write types
899 for AnyEvent::Handle.
900 - allow on_eof handler to be called after on_error with EPIPE returns.
901 - do not immediately call on_read callback in handle constructor.
902
9034.14 Thu Jun 5 20:29:31 CEST 2008
904 - Fixed a bug in DNS SRV priority sorting.
905 - AnyEvent::Util::guard now reports runtime errors while
906 executing the guard block as warnings.
907 - handle 0-byte-reads just before EOF correctly in AnyEvent::Handle.
908
9094.13 Thu Jun 5 00:47:59 CEST 2008
910 - AnyEvent::DNS only followed cname chains with length 2,
911 contrary to documentation. bumped it up to 10, thanks to
912 microsoft, the current broken-dns-config-king.
913 - AnyEvent::DNS didn't check for socket return status
914 but instead relied on perl not creating filehandles in that
915 case - too bad it gives you a bogus file handle (reported
916 and analysed by Vladimir Timofeev).
917 - fix queue management logic in AnyEvent::Handle:
918 when on_read was registered, the queue was empty
919 and no progress could be made AnyEvent::Handle would
920 enter an endless loop.
921 - correctly start reading again when the handle became
922 busy again after idling.
923 - correctly treat tls shut-downs as EOF condition
924 (for the time being, would be nice to have a callback for that).
925 - correctly call eof callback under all conditions
926 (hopefully).
927 - the timeout callback did not expect that $self can go
928 away any time.
929
9304.12 Tue Jun 3 10:58:04 CEST 2008
931 - include AnyEvent::Intro, a tutorial for anyevent,
932 anyevent::socket and anyevent::handle.
933 - allow more options in on_error.
934
9354.11 Fri May 30 23:42:25 CEST 2008
936 - INCOMPATIBLE CHANGE: replace ptr by real PTR lookup, provide
937 reverse_lookup and reverse_verify to replace it, support
938 v4mapped and v4compat addresses.
3 - provide more documentation for the resolver class. 939 - provide more documentation for the resolver class.
4 - really replace longest run of :0: by :: in format_address, 940 - really replace longest run of :0: by :: in format_address,
5 also properly convert :: and ::1 again. 941 also properly convert :: and ::1 again.
6 - support NAPTR record name and decode it. 942 - support NAPTR record name and decode it.
7 - implement random weight sampling for SRV records, as per 943 - implement random weight sampling for SRV records, as per
8 rfc 2782. 944 rfc 2782.
9 - replace ptr by real PTR lookup, provide reverse_lookup 945 - correctly abort on srv-record targets of ".".
10 and reverse_verify to replace it, support v4mapped and 946 - added AnyEvent::DNS::wait_for_slot.
11 v4compat addresses. 947 - in the unlikely event of a virtual circuit connection
948 being invalidated by a delayed udp reply, AnyEvent::DNS
949 could die. this has been fixed.
950 - plug a probable memleak in the DNS vc code.
951 - use configured timeout also for VC request phase.
952 - implement timeout and max_outstanding methods
953 for resolver class.
954 - update version numbers in all modules.
12 955
134.1 Thu May 29 05:45:40 CEST 2008 9564.1 Thu May 29 05:45:40 CEST 2008
14 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address 957 - INCOMPATIBLE CHANGE: renamed xxx_ip functions to xxx_address
15 which mirrors their purpose better (old names still available 958 which mirrors their purpose better (old names still available
16 for a while). Also moved AnyEvent::DNS::addr to 959 for a while). Also moved AnyEvent::DNS::addr to
… …
88 - support service names in tcp_connect and tcp_server. 1031 - support service names in tcp_connect and tcp_server.
89 - work around netbsd/openbsd bug causing unix sockets not to 1032 - work around netbsd/openbsd bug causing unix sockets not to
90 become writable until empty. 1033 become writable until empty.
91 - only make a dummy read for error detection when getpeername 1034 - only make a dummy read for error detection when getpeername
92 returns ENOTCONN. 1035 returns ENOTCONN.
93 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 1036 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
94 override. 1037 override.
95 - new AnyEvent::Guard::cancel. 1038 - new AnyEvent::Guard::cancel.
96 1039
973.5 Sat May 17 23:17:14 CEST 2008 10403.5 Sat May 17 23:17:14 CEST 2008
98 - gracefully deal with systems not defining CLOCK_MONOTONIC 1041 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
112 1055
1133.41 Sun May 11 19:53:13 CEST 2008 10563.41 Sun May 11 19:53:13 CEST 2008
114 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 1057 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
115 1058
1163.4 Sun May 11 00:29:25 CEST 2008 10593.4 Sun May 11 00:29:25 CEST 2008
117 - complete re-implementation of condvars with a hopefully much mroe useful 1060 - complete re-implementation of condvars with a hopefully much more useful
118 API (backward compatible functions exist but are undocumented). 1061 API (backward compatible functions exist but are undocumented).
119 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 1062 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
120 (still subject to change). 1063 (still subject to change).
121 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 1064 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
122 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 1065 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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