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

Comparing AnyEvent/Changes (file contents):
Revision 1.166 by root, Tue Jul 8 19:50:25 2008 UTC vs.
Revision 1.589 by root, Sun Apr 13 11:33:02 2014 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.2 303:55:22 <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
227.07 Tue Dec 17 17:45:02 CET 2013
23 - the documentation for custom tls verify schemes was wrong. make it agree
24 with the code (reported by Maxime Soulé).
25 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
26 - work around an API change in openssl that could cause wrong tls connection
27 aborts, likely on windows only (analyzed by sten).
28 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
29 endless loop (reported by Dietrich Rebmann).
30 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
31 (patch by Maxime Soulé).
32 - document AnyEvent::Impl::IOAsync::set_loop and
33 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
34 functionality has _always_ been available.
35 - force a toplevel domain name in t/81_hosts.t (analyzed by
36 David Jack Wange Olrik).
37 - document that AnyEvent::Log uses AnyEvent::IO.
38 - warn about AnyEvent::Filesys::Notify performance.
39 - praise the joys of AnyEvent::Fork::*.
40 - time for an =encoding directive.
41 - do no longer use JSON to create a default json coder, use
42 JSON::XS or JSON::PP directly.
43
447.05 Wed Aug 21 10:38:08 CEST 2013
45 - uts46data.pl couldn't be found due to wrong naming of the file
46 (reported by Fulko Hew).
47 - handle lone \015's properly in AE::Handle's default line read
48 (reported by various people).
49 - untaint ip addresses found in /etc/hosts (patch by José Micó).
50 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
51 fail on first use (reported and testcase by Andrew Whatson).
52 - expose AnyEvent::Log::format_time, and allow users to redefine it.
53 - expose AnyEvent::Log::default_format, and allow redefinition.
54 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
55 - AnyEvent::Debug shell can now run coro shell commands, if available.
56 - t/63* tests were wrongly in MANIFEST.
57 - kernel.org's finger server went MIA, switch to freebsd.org and
58 icculus.org.
59 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
60 to function asynchronously (reported by Daniel Carosone).
61 - hard-disable $^W in most tests, it generates too much garbage output.
62 - use a (hopefully) more future-proof method to emulate common::sense.
63 - upgrade to UTS-46:6.2.0.
64 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
65 documented. should not affect anything.
66
677.04 Thu Nov 15 02:23:58 CET 2012
68 - debugging code left in.
69
707.03 Thu Nov 15 02:16:11 CET 2012
71 - AnyEvent::Socket::inet_aton did not work when dns resolution was
72 used to find the addresses (analyzed and patch by Maxime Soulé).
73 - fix a memory leak in the /etc/hosts lookup code when hosts don't
74 resolve and are not in hosts (analyzed by sten).
75
767.02 Tue Aug 14 04:27:58 CEST 2012
77 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
78 program by Yorhel).
79 - verified that AnyEvent::Socket follows RFC5952.
80 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
81
827.01 Sun May 13 01:03:17 CEST 2012
83 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
84 not available, instead of throwing an exception.
85 - use File::Spec to get the tmpdir in t/*, to avoid needless
86 failures on (most, not mine :) windows boxes.
87 - new handle read types: tls_detect and tls_autostart.
88
897.0 Fri Apr 13 06:33:30 CEST 2012
90 - child watchers are broken in POE 1.352 (also many earlier
91 versions) and there seems to be no way to work around it, as POE
92 itself is inherently racy. Document this regression and add a delay
93 in t/68_poe_03_child.t for the time being.
94 - new module AnyEvent::IO, that is a frontend to either a pure-perl
95 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
96 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
97 - load /etc/hosts only when DNS has no answer.
98 - stat /etc/hosts on every access and reload it if it changed.
99 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
100 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
101 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
102 a side effect, they now use true appending as opposed to libc
103 appending, and the file might not have been opened when the function
104 returns.
105 - the default logging level was not properly documented in a variety of
106 places, this has hopefully been rectified.
107 - updated uts46data.pl for unicode 6.1.0.
108 - made log messages generated by AnyEvent submodules not
109 include the package name anymore, as it will be logged
110 by default already.
111 - upgrade to the trick used by common-sense 3.5 to work
112 around extra warning torture/breakage under perl 5.15.x.
113 - log messages by anyevent are now ucfirst, are usually full
114 sentences and do no longer include the package name.
115 - the storable read type would sometimes throw an exception instead
116 of causing EBADMSG (analyzed by Maxime Soulé).
117 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
118 keysize (reported by Paul Howarth).
119 - documented Rocco fabricating statements about AnyEvent and me.
120
1216.14 Tue Jan 31 20:00:24 CET 2012
122 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
123 probably an editing mistake (reported by Darin McBride).
124
1256.13 Thu Jan 12 07:27:01 CET 2012
126 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
127 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
128 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
129 regex-matching inside the callback would change the parameter, despite
130 $1 being dynamically scoped per-block. this perl bug is now being
131 worked around (testcase by Cindy Wang).
132
1336.12 Mon Dec 12 13:21:10 CET 2011
134 - $! was clobbered by subsequent calls in tcp_connect, due to
135 the postpone not saving/restoring it (analyzed by Richard Garnish).
136
1376.11 Tue Nov 22 10:36:05 CET 2011
138 - Tk cannot create windows when tainted unless you set a title, so
139 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
140 - escape any nonprintable/nonascii characters when stringifying
141 backtraces.
142 - log the reason loading an interface module fails to load at level debug.
143
1446.1 Tue Oct 4 19:44:30 CEST 2011
145 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
146 and above), and some messages inside AnyEvent have been elevated
147 to higher log levels to print by default.
148 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
149 - work around yet another signal race bug, newly introduced by POE. My
150 how that module sucks.
151 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
152 support ipv6-only operation even with defaults (also use ::1 instead
153 of 127.0.0.1).
154 - give better guidelines on how to chose a logging level.
155 - AnyEvent::Log can now cap (limit) log message priority by context.
156 - AE::log will now save and restore $! and $@, so callers don't have to.
157 - AE::Strict now checks that registered fds are still valid at regular
158 intervals.
159 - skip frequently called 'require' calls if the module appears loaded.
160 - increase signal latency test timer considerably for poor little osx,
161 which can't be bothered with basic posix functionality such as signal
162 delivery.
163 - mention some debugging aids in the tutorial.
164 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
165 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
166 of the context as title.
167 - AnyEvent::Strict would not restore read-write access to $_ when a callback
168 died - this is a user bug, but since AE::Strict exists to help dbeugging...
169 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
170 to work with it - the author is well aware of the issues but refuses
171 to fix them. This extreme step has been taken because the module actively
172 attacks the goals of AnyEvent and makes it harder for module authors to use
173 AnyEvent.
174
1756.02 Fri Aug 26 20:08:31 CEST 2011
176 - log_to_syslog now accepts facility strings.
177 - log_to_syslog would not allow facilities - "$facility|$level"
178 *literally* means that string. sheesh.
179 - the CPAN parser got confused and thought we provide Sys::Syslog, try
180 to work around.
181 - renamed AnyEvent::Impl::FLTK2 to FLTK.
182
1836.01 Fri Aug 26 07:04:11 CEST 2011
184 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
185 the TTL as fourth element - this affects ->resolve, ->request
186 and ->dns_unpack indirectly. Convenience functions such as
187 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
188 AnyEvent::Socket.
189 - INCOMPATIBLE CHANGE: backend authors now should not implement
190 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
191 methods.
192 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
193 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
194 - DEPRECIATION: please change
195 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
196 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
197 in your programs, for future compatibility.
198 - *blush* idle watcher emulation was borked.
199 - AnyEvent::Strict now checks the AE::xxx functions as well.
200 - querying too long domain names would barf deep inside AnyEvent::DNS - since
201 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
202 for too long domain names explicitly and passes an undef to the callback
203 instead (reported by James Bromberger).
204 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
205 - new functions: AnyEvent::postpone, AnyEvent::log.
206 - new module: AnyEvent::Log, for simple logging needs.
207 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
208 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
209 a debug shell.
210 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
211 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
212 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
213 /etc/hosts can be found.
214 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
215 configuration.
216 - env variables can now be specified using AE_xyz instead of
217 PERL_ANYEVENT_xyz, subject to some rules.
218 - lots of new AnyEvent::Debug functionality.
219 - AnyEvent::Strict now checks for modifications of $_ by the callback
220 (by setting it to readonly while executing the callback, if possible).
221 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
222 Invest extra overhead to work around this peculiar behaviour.
223 - split the pure perl backend into a loop module and an impl
224 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
225 - some read types (regex, netstring, json, storable) could
226 errornously keep an AE::Handle object alive on their own
227 (reported by Mohammad Toossi).
228 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
229 supports /etc/hosts overrides.
230 - add more workarounds around child watcher bugs in IO::Async
231 and (newly introduced) in POE.
232 - use glib child watchers and try to work around its limitations,
233 if possible (based on a patch by Kevin Ryde).
234 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
235 - many read types in AnyEvent::Handle would malfunction during
236 transitions to TLS because of a "delete $self->{rbuf}".
237 (reported by Gerald Galster).
238 - improve condvar blocking wait performance (probably only noticable
239 with faster backends), also make it easier for event loops that
240 do not support blocking waits.
241 - PERL_ANYEVENT_MODEL now also allows full module names, although
242 the usefulness of this feature is not yet clear.
243 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
244 - add t/80_ssltext.t.
245 - no longer use AUTOLOAD for the initial loading (for no very good reason).
246 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
247 - do not run event loop detection multiple times when programs call their cached
248 copies of AnyEvent::detect.
249 - parse_hostport now accepts absolute paths als unix domain sockets.
250 - suppress the idiotic warning inside POE by patching the POE::Kernel
251 run flags instead of asking the user to jump through contortions.
252 - add Task::Weaken as dependency (as recommends only at this time).
253 - replace PF_UNSPEC by 0, for good old minix.
254
2556.0 (not officially released, but was on CPAN for about 15 minutes :/)
256
2575.34 Thu May 12 09:59:41 CEST 2011
258 - AE::TLS didn't support both cert_file and key at the same time,
259 due to some copy&paste bug (reported and analyzed by vti).
260 - improve AE::Handle timeout documentation slightly.
261
2625.33 Mon Feb 7 21:51:41 CET 2011
263 - never appeared on CPAN, for unknown reasons.
264 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
265 object when it's already gone (analyzed by Ryan Bullock).
266 - update idna algorithm(s) to UTS#46 version 6.0.0.
267 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
268 uts46 says it's an error, and testsuite makes more sense.
269 - remove leading dots in idn_to_ascii - testsuite says to remove,
270 uts46 says it's an error, and testsuite makes more sense.
271 - fixed a bug in idn ignored character processing.
272 - passes the uts46 6.0.0 testsuite except for three cases which
273 apparently are bugs in the testsuite itself.
274 - passes the uts46 6.0.1 testsuite except for eleven cases which
275 apparently are bugs in the testsuite itself.
276
2775.31 Sun Jan 23 11:49:19 CET 2011
278 - work around a POE bug where POE would first poll for new events and
279 invoke handlers BEFORE handling the message queue.
280 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
281 - stop_read now does just that in TLS mode. see its
282 documentation for caveats.
283 - update to common::sense 3.4 warning set (some warnings in perl impose
284 an arbitrary 32kb limit on regex matches, which affects
285 AnyEvent::Handle).
286 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
287 - new experimental AE::Handle::resettls method.
288
2895.3 Fri Dec 31 05:48:13 CET 2010
290 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
291 issues with IO::Async have been solved in that version, including
292 the availabilty of a default loop!
293 - implement new wbuf_max parameter for AnyEvent::Handle.
294 - added Cocoa::EventLoop backend by Daisuke Murase.
295 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
296 - clarify handle on_prepare documentation.
297 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
298 to avoid leaking them to any child processes.
299 - try to force files into the arch-dependend lib directory, for the
300 half-"turly obsessive compulsive person"(s) that depend on it.
301
3025.29 Sun Dec 5 10:49:21 CET 2010
303 - convert EV backend to EV 4.00 API (so better upgrade EV too).
304 - AE::Handle: implement dynamic read_size adjustment, add
305 max_read_size parameter, reduce default read_size to 2048.
306 - add some FAQ entries for servers and/or on_eof vs. on_error.
307 - work around OS bugs (cygwin again) when lingering in AE::Handle.
308 - reduce memory usage slightly when connect was used in AE::Handle.
309 - wrok around more segfaults in Tk.
310 - document tls_ctx => undef as valid.
311 - detect WNOHANG value at compiletime.
312 - include some probably linux-only support for building AnyEvent
313 as part of the perl core.
314 - improve compatibility of t/08_idna.t to perl 5.8.
315 - make AnyEvent::Strict truly optional (in case somebody builds a
316 minimal perl...).
317
3185.28 Wed Oct 13 04:14:23 CEST 2010
319 - due to a glitch, AnyEvent's internal getprotobyname wasn't
320 actually used in tcp_server.
321 - implement AnyEvent::CondVar->new method, which was documented but
322 missing.
323 - support autoloading of read types in unshift_read.
324 - AnyEvent::Strict: do the test for invalid fh arguments differently,
325 resulting in more informative error messages.
326 - new AnyEvent::FAQ manpage.
327 - clarify that you can actually call push_read etc. in on_read.
328 - put keys and certs in /tmp, not into the current directory,
329 and document this in the SECURITY CONSIDERATIONS sections
330 of AnyEvent::TLS.
331 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
332 requested by Ron Isaacson.
333 - document $AnyEvent::DNS::RESOLVER variable.
334 - incorporated major typo patches by Abhijit Menon-Sen.
335 - AnyEvent::Handle now croaks when negative timeouts are passed.
336 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
337 undocumented, but who knows when they might come in handy.
338
3395.271 Tue Jun 8 12:05:46 CEST 2010
340 - backport to perl 5.8.x.
341
3425.27 Sun Jun 6 12:12:05 CEST 2010
343 - postpone differently in AnyEvent::Socket now, as
344 when not, canceling the connection attempt might fail
345 (found by Felix Antonius Wilhelm Ostmann).
346 - explicitly check for non-stream sockets in AE::Handle, too many
347 clueless people fell into the trap of this somehow working.
348 - simplified and reworked the "OTHER MODULES" section.
349 - better/more condvar examples.
350
3515.261 Wed Apr 28 16:13:36 CEST 2010
352 - AF_INET6 was not properly used from Socket6 during configuration
353 time (found by Andy Grundman).
354
3555.26 Mon Apr 12 04:49:35 CEST 2010
356 - don't generate spurious readyness notifications when select
357 returns EINTR in the pure perl backend (analysed and testcase
358 by Bas Denissen).
359 - give same examples in SYNOPSIS sections of AnyEvent and AE.
360 - provider faster implementations for AE::time/now/now_update for
361 EV and Perl backends.
362 - provide faster AE::cv and AE::time implementations for all
363 backends.
364 - no longer support lower/mixed-case signal names (this was
365 never documented nor universally supported).
366 - some more memory µ-optimisations, and somewhat less messy
367 code to cope with both AnyEvent and AE APIs.
368
3695.251 Sat Mar 13 00:58:21 CET 2010
370 - make test failed if it was run before make install on perl <=
371 5.8.8, and older versions of perl make our live very difficult,
372 so write constants.pl during Makefile.PL time.
373
3745.25 Sat Mar 13 00:23:14 CET 2010
375 - fix a race condition in AnyEvent::Handle that would cause
376 a "bio_read: unsupported method" error (in _another_
377 ssl connection) after user code threw an exception.
378 - added AnyEvent::Handle->destroyed method.
379 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
380 - the AnyEvent::Util::fh_nonblocking function worked only
381 by lucky accident on win32.
382 - smaller and faster AnyEvent::Util::fh_nonblocking.
383 - when the (required!) Time::HiRes module is missing, AnyEvent
384 did not fall back to built-in time properly.
385 - do not load Fcntl at runtime, saving memory and loading time.
386 - precompile a number of constants and use them instead of runtime
387 detection and eval.
388 - free detection code after detection and similar memory optimisations.
389 - Perl backend timer interval best effort drift has been improved
390 (same algorithm as EV).
391 - update unicode idna mapping table.
392
3935.24 Tue Jan 5 11:39:43 CET 2010
394 - cygwin never reports errors from failed connects. 1.5 just gives
395 you continous readyness and EAGAIN, 1.7 is even more broken and
396 just hangs. work around both issues in a cygwin-specific hack.
397 - improve idle watcher documentation slightly.
398
3995.23 Sun Dec 20 23:48:00 CET 2009
400 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
401 - implement punycode_encode/decode, idn_nameprep,
402 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
403 - provide $AE::VERSION.
404 - removed traces of "no strict 'refs'".
405
4065.22 Sat Dec 5 03:51:13 CET 2009
407 - downgrade-or-fail in AnyEvent::Handle::push_write, to
408 diagnose encoding failures earlier and more succinctly.
409 (this works around bugs in perl, throwing away encoding info
410 when passing scalar data to extensions).
411 - add more examples to AnyEvent::Socket manpage.
412 - upgrade internal warning set to the same as common::sense 2.03.
413 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
414 (John Beppu).
415 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
416 - add google public dns servers to fallback server set and make sure
417 we load-balance properly between the three sets. also add all
418 fallback dns servers, not just a random one, to each dns config.
419
4205.21 Thu Nov 19 02:48:47 CET 2009
421 - fix a problem where socket constants were called with parameters
422 (spotted by David Friedland).
423 - fork_call never use'd POSIX (reported by Daisuke Maki).
424 - improve perl 5.6 compatibility further (but it still won't work
425 unless you rip out everything but the core).
426 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
427 uses 10MB of RAM, but doesn't open a console window. *sigh*.
428
4295.202 Wed Oct 14 22:35:44 CEST 2009
430 - AnyEvent::DNS would unexpectedly clobber $_ under windows
431 (analysed by Matthias Waldorf).
432 - AnyEvent::Handle::run_cmd can now pass the PID of the
433 newly-created process, which is much less useful than it might
434 sound (based on patch by Yann Kerherve).
435
4365.201 Tue Sep 29 12:09:25 CEST 2009
437 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
438 (reported by Torsten Foertsch).
439 - common::sense 2.0 could cause tcp_server to throw an exception
440 (analysed by elmex).
441
4425.2 Mon Sep 14 07:04:49 CEST 2009
443 - INCOMPATIBLE CHANGE: do no longer support register_read_type
444 and register_write_type in AnyEvent::Handle, instead support
445 package names (the facility was mostly abused).
446 - implement "packagename-as-read/write type" support in
447 AnyEvent::Handle.
448 - AnyEvent::Handle: new options "keepalive" and "oobinline".
449 - oobinline set by default to avoid security issues.
450 - the pure-perl event loop backend wrongly detected times() fallback
451 support (spotted by Pavel Boldin).
452
4535.12 Tue Sep 1 20:26:50 CEST 2009
454 - be more lenient when parsing resolv.conf files, as some people
455 use hashmarks as comment indicator inside directives (reported
456 by Michael S. Fischer).
457 - use same set of warnings as common::sense 2.0.
458 - fix a potential 32 bit overflow issue due to perl having problems
459 with large hex constants in the Perl backend.
460 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
461 by default.
462 - fix some prototypes.
463 - (optionally) require Async::Interrupt 1.02.
464 - replace opendns fallbacks by verizon/level3 public dns servers.
465
4665.112 Fri Aug 21 13:59:04 CEST 2009
467 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
468 using starttls with an illegal/uncreatable tls context.
469 - port to broken win32 perls.
470
4715.111 Sun Aug 16 18:44:36 CEST 2009
472 - perl errornously requires sockaddr_un structures returned by the
473 kernel to have a specific length. work around this bug by padding
474 sockaddr_un structures (found to fail on OS X, as perl only uses
475 the correct code for GNU/Linux, but likely to fail on other
476 OSes as well).
477
4785.11 Wed Aug 12 17:49:37 CEST 2009
479 - fixed a bug in run_cmd with input from filename.
480 - create a json coder object when none is given in write direction
481 to reduce dependency on JSON.
482
4835.1 Tue Aug 11 03:17:32 CEST 2009
484 - tcp_connect now reports the correct error in $!, not an unrelated
485 one (was broken since 4.91).
486 - AE::Handle did not properly clear rtimeout/wtimeout watchers
487 when retrying a tcp_connect.
488 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
489 - fixed wrong documentation in AE::Socket.
490
4915.01 Mon Aug 10 03:16:32 CEST 2009
492 - last release broke EV child watchers :/.
493
4945.0 Sun Aug 9 17:32:09 CEST 2009
495 - officially introduce and document the AE API.
496 - lowering the timeout at runtime did not have immediate effect
497 in AnyEvent::Handle.
498 - AE::Handle now has separate and independent read and write timeout
499 settings.
500 - probe for Guard module when AnyEvent::Util is loaded, not at
501 runtime (Event doesn't like eval "use" from callbacks, and it avoids
502 exporting the probe function).
503 - Impl::IOAsync's io watchers did not use the filehandle emulation
504 code to work around it's watcher limits.
505 - work around IO::Async's problems with overloaded objects.
506 - try to disable wearnings inside IO::Async, it's just too buggy.
507 - do no longer use Event by default, only EV or Perl, Event is too buggy.
508 - hackishly provide fast AE:: interface when EV or Perl is the backend.
509 - inherit VERSION from AnyEvent::VERSION in most modules.
510
5114.91 Thu Aug 6 15:42:45 CEST 2009
512 - AE::Handle::starttls could get out of sync when the read buffer
513 already contains some TLS handshake.
514 - AE::Handle did not properly free the TLS session, causing bigger
515 memory leaks in Net::SSLeay than expected (reported by toaster).
516 - AE::Socket::tcp_connect will now artificially delay invoking
517 the callback to avoid returning after invoking the callback.
518 - convert many internal watcher uses to AE API.
519
5204.9 Sat Aug 1 11:07:01 CEST 2009
521 - Glib silently fails when registering a timeout with
522 a negative value, so avoid that.
523 - call condvar callback immediately when it is set
524 after the condition is already signalled.
525 - check rbuf_max condition only after trying to consume
526 data.
527 - normalise signal numbers to names when using the ae's
528 signal handling, but do not document this (yet).
529 - pure perl signal emulation did not properly set nonblocking
530 mode on the signal pipe on win32 (this is unlikely to have caused
531 any issues).
532 - new module: AnyEvent::Debug.
533 - AnyEvent::Strict now checks that a signal was specified
534 by name (not name or number).
535 - reduce memory footprint in typical cases by ~50kb by
536 only compiling signal code when necessary.
537 - add AnyEvent::Handle->rbuf_max.
538 - grab the AE namespace for future new API, implement stubs for
539 said future/alternative API.
540 - new function: AnyEvent::Socket::format_hostport.
541
5424.881 Tue Jul 28 12:51:53 CEST 2009
543 - work around a bug in local in pre-5.10 perls, causing
544 AnyEvent::Handle to recurse when it should not
545 (analysed by elmex).
546
5474.88 Tue Jul 28 04:04:37 CEST 2009
548 - re-bless the handle into a dummy package after calling
549 AnyEvent::Handle::destroy, so the user does not need to check
550 for errors after every push_write etc.
551 - do not attempt to run t/02_signals.t on obviously broken
552 platforms.
553
5544.87 Sun Jul 26 02:06:16 CEST 2009
555 - do not attempt to linger when there is no longer a valid fh.
556 - enforce tls mode to be either accept or connect, do not simply
557 segfault in Net::SSLeay.
558 - AnyEvent::Handle can now call tcp_connect itself (new parameters
559 connect, on_prepare, on_connect and on_connect_error). Updated
560 tutorial accordingly.
561 - add AnyEvent::Impl::Irssi backend.
562
5634.86 Mon Jul 20 23:52:29 CEST 2009
564 - since the verbose warning is not enough, explicitly document
565 that versions before 1.33 of Net::SSLeay are not secure.
566 - work around signal handling races in Event and (...) Event::Lib.
567 - try to align signal-race timer to full-second boundaries.
568 - work around Tk not liking negative timeouts.
569 - don't complain of different grades of environmental unfriendlyness
570 in IO::Async.
571
5724.85 Sat Jul 18 06:16:14 CEST 2009
573 - nail the signal race problem in perl once and for all
574 (see $AnyEvent::MAX_SIGNAL_LATENCY).
575 - take advantage of Async::Interrupt if it is available.
576 - load Time::HiRes and Guard modules on demand only.
577 - add optional/recommended modules section to AnyEvent
578 documentation.
579 - reduce memory usage considerably (and reduce startup penalty)
580 by not using "strict", "warnings" and "overload" modules.
581 - work around buggy windows/openbsd perls and provide EBADMSG
582 and EPROTO ourselves when missing.
583 - improve perl 5.6 compatibility of the core event loop.
584 - made Net::SSLeay version 1.33 a soft requirement.
585
5864.83 Fri Jul 17 16:56:26 CEST 2009
587 - implement AnyEvent::Socket::getprotobyname.
588 - AnyEvent::CondVar's will now detect recursive blocking
589 waits and will croak, as too many people fall into
590 this trap.
591 - AnyEvent::Handle will now call ->destroy on itself after
592 executing the on_error callback, instead of doing some
593 half-baked internal shutdown, for fatal errors.
594 - clarify on_eof behaviour w.r.t. the read queue and
595 on_read callbacks.
596 - ignore some possible spurious wake-ups in tcp_connect.
597
5984.82 Sat Jul 11 00:34:55 CEST 2009
599 - POE and Event backends didn't accept some callable objects as
600 callbacks.
601 - use Config module instead of POSIX module to detect signal names
602 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
603 module doesn't even have all POSIX signals :/.
604 - use more workarounds around the many refcnt/corruption bugs in
605 Event::Lib.
606 - work around a race condition in perl's select, causing t/03_child.t
607 to rarely fail.
608
6094.81 Thu Jul 9 10:30:30 CEST 2009
610 - AnyEvent::Handle didn't properly diagnose write errors
611 (it expected -1 from syswrite, how lame... :).
612 - support file descriptors in addition to file handles
613 in AnyEvent->io.
614 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
615 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
616 and PERL_ANYEVENT_CA_PATH.
617 - provide a sensible synopsis section for AnyEvent::TLS.
618 - add a "supported backends" section to the manpage.
619 - added simple io watcher test to testsuite, using a
620 portable_socketpair.
621 - tried to improve the stability of the Event::Lib backend,
622 YMMV.
623
6244.8 Mon Jul 6 23:45:16 CEST 2009
625 - AnyEvent::DNS did not properly follow CNAME records with
626 uppercase targets.
627 - AnyEvent::DNS would errornously return AAAA records
628 with v4 mapped addresses (a faulty record) as ipv4 addresses,
629 causing AnyEvent::Socket to throw an exception.
630 - added new module AnyEvent::TLS for easier SSL/TLS context
631 creation, with many options including hostname verification,
632 secure default configuration, lots of documentation and,
633 predefined diffie-hellman keys for perfect forward security
634 and much more. get it while it's still fresh!
635 - use AnyEvent::TLS in AnyEvent::Handle for context management.
636 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
637 so AnyEvent::Socket users have smaller memory footprint
638 in the common case.
639 - add AnyEvent::Handle->push_shutdown method.
640 - add an additional error message parameter to AnyEvent::Handle's
641 on_error callback (for TLS, $! is still available).
642 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
643 - make AnyEvent::Handle more robust against early conenction
644 failures (during new), and return C<undef> in that case
645 from the constructor.
646 - AnyEvent::Handle will now try to load only JSON::XS first,
647 then fall back to JSON.
648 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
649 and because it was documented that way.
650
6514.451 Fri Jul 3 00:28:58 CEST 2009
652 - do not clear rbuf when shutting down an AnyEvent::Handle
653 object - doing so breaks AnyEvent::HTTP.
654
6554.45 Mon Jun 29 22:59:26 CEST 2009
656 - a write error could cause AnyEvent::Handle to create
657 an I/O watcher with an undefined $fh.
658 - special-case mapped ipv4 addresses in both
659 AnyEvent::Socket::format_address and parse_address,
660 to treat them just like ipv4 addresses.
661 - updated and overhauled the AnyEvent::Intro doc.
662 - implement AnyEvent::Socket::format_ipv4/ipv6.
663 - slightly speed up portable_pipe/socketpair.
664 - expand condvar begin/end documentation.
665
6664.42 Fri Jun 26 08:32:18 CEST 2009
667 - preliminary and neccesarily incomplete support for IO::Async.
668 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
669 case it was set to IGNORE, to ensure we can catch child statuses
670 even when the calling env acts stupidly.
671 - updated benchmarks with IO::Async, which performs very well.
672
6734.412 Wed Jun 24 01:35:57 CEST 2009
674 - support an "untaint" attribute for AnyEvent::DNS
675 and set it on the default resolver.
676 - implement and document AnyEvent::Impl::Perl::loop.
677 - remove all anyevent-env variables from %ENV
678 when running in tainted mode.
679 - mention and extend the IO::Lambda benchmark.
680
6814.411 Sun Jun 7 18:48:02 CEST 2009
682 - do not try to use F_SETFD on windows.
683
6844.41 Thu May 14 06:40:11 CEST 2009
685 - work around issues in older perls (5.8.0?) when
686 a signal handler is deleted from the %SIG hash.
687 - use POSIX::_exit in child test, to avoid running
688 destructors.
689 - speed up CHLD handling by relying on SIGCHLD being
690 synchronously delivered, even when we roll our own
691 implementation.
692 - AnyEvent::DNS: add the "dname" resource record name
693 for cosmetic reasons.
694
6954.4 Sun Apr 26 20:12:33 CEST 2009
696 - implemented idle watchers, where applicable.
697 - AnyEvent->time died when Event backend was in use.
698 - fix a memleak in the Tk backend.
699 - sped up Tk timer handling.
700 - clip negative "after" values to 0 in AnyEvent::Impl::Event
701 to avoid spamming out warning messages.
702 - fix Qt timers without interval.
703 - avoid Qt zero-timeout special case, in old libqt's.
704
7054.352 Mon Apr 20 16:31:11 CEST 2009
706 - fix AnyEvent::Strict error messages for child watchers.
707 - fix/update Makefile.PL "recommends" versions.
708 - add AnyEvent->now_update.
709
7104.351 Sat Apr 11 07:56:14 CEST 2009
711 - actually make the signal pipe work under win32.
712 - localise $! in signal handler to avoid changing $!, although
713 perl itself does not seem to save/restore errno either.
714 - set the cloexec flag on the signal pipe (normally set by perl too).
715
7164.35 Fri Mar 27 11:48:20 CET 2009
717 - event models relying on AnyEvent's signal watcher emulation did
718 invoke the callback asynchronously, contradicting documentation
719 and causing signals to get lost (this includes AnyEvent's own
720 event loop). AnyEvent now uses the standard pipe trick to make
721 callback execution synchronous to the event loop.
722 - AnyEvent::Handle didn't free TLS context data on DESTROY
723 (patch by Pavel Shaydo).
724 - work around the ever-incompatibly-changing API of MakeMaker.
725 - document that changing global variables without restoring them
726 is a bad idea in Perl (noted by Adam Rosenstein).
727 - AnyEvent::Strict now barfs if ->io is passed a file.
728
7294.34 Thu Feb 12 18:32:45 CET 2009
730 - separately buffer TLS read data, as otherwise the read queue could
731 deadlock as receiving data is not expected while draining the
732 read queue (which cna only happen with TLS).
733 - raise EBADMSG error on JSON decoding errors.
734 - fix some minor manpage bugs (reported by Maximilian Gaß).
735 - speed up select bitmask parsing quite a bit in the pure perl
736 backend.
737 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
738
7394.331 Tue Jan 6 21:07:25 CET 2009
740 - socketpair fails on many vista machines because vista has
741 completely broken accept/getpeername and getsockname functions,
742 so we provide our own socketpair emulation that kind of works
743 (AnyEvent::Util::portable_pipe).
744 - new function: AnyEvent::Util::portable_socketpair.
745 - take advantage of the Guard module if it exists.
746
7474.33 Fri Nov 21 02:35:40 CET 2008
748 - AnyEvent::Strict did errornously flag a fileno of 0 as
749 illegal.
750 - reduce memory usage and slightly speed up the pure perl backend
751 by only storing the file descriptor, not the file handle.
752 - add missing autocork method to AnyEvent::Handle (reported
753 by Adam Rosenstein).
754 - AnyEvent::DNS->resolve errornously documented an $rcode result
755 argument, but there isn't (spotted by Henrik Krohns).
756 - the naked truth about Tk - it's basically unmaintained.
757
7584.32 Mon Nov 3 22:46:32 CET 2008
759 - fix AnyEvent::Socket::resolve_sockaddr to properly support
760 unix sockets again and choose a proper default for
761 the socket type.
762 - fix call to resolve_sockaddr in tcp_connect.
763
7644.31 Thu Oct 30 04:41:48 CET 2008
765 - implemented AnyEvent::Handle->destroy method.
766 - hint about unexpected effects in TLS mode.
767 - speed up AnyEvent::Impl::Perl by using more arrays
768 instead of hashes without the slightest loss of readability :)
769 - work around a perl argument refcounting bug.
770 - with some perl patches applied, AnyEvent now seems leak-free.
771
7724.3 Fri Oct 3 09:18:43 CEST 2008
773 - AnyEvent will now install a no-op signal handler for SIGPIPE,
774 unless one has been installed already.
775 - warn about not loaidng AnyEvent::Impl::POE early enough
776 (patch by Adam Rosenstein).
777 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
778 - fix a bug in where in SSL connect mode, the
779 client would first wait for some data by the server and
780 otherwise hang. (reported and analysed in an absolutely
781 exemplary manner by Adam Rosenstein).
782 - fix a bug in where SSL EOF would not be treated
783 as stream EOF, putting the connection into a hung state.
784 (reported and analysed in an absolutely exemplary manner by Adam
785 Rosenstein).
786 - fix a potential segfault when the TLS context
787 would go missing in a read callback (e.g. due to stoptls) -
788 Net::SSLeay of course makes no type checking whatsoever.
789 - AnyEvent::Handle will not stop the read watcher in TLS mode.
790 - AnyEvent::Handle->stoptls will send a TLS close notify
791 instead of simply destroying the stream now.
792 - fix a bug where large blocks of data written in TLS mode
793 would not be sent unless triggered by receives.
794 - on_drain will now take the tls write buffer into account.
795 - SSL operations have been streamlined a bit and should be faster
796 now, more to come, though.
797 - added a FAQ section, will document nontrivial issues in there.
798
7994.234 Mon Sep 29 04:08:13 CEST 2008
800 - fix child watcher documentation: only child exits will be reported,
801 no trace events.
802 - mention SIGPIPE in AnyEvent::Handle.
803 - perl backend will now use POSIX::times as monotonic clock
804 when available and the monotonic clock option is not.
805 - run even without Time::HiRes in some cases now, but do not
806 provide sub-second accuracy (all *supported* perl versions have
807 Time::HiRes).
808
8094.233 Fri Aug 22 01:48:05 CEST 2008
810 - fix a bug in the testsuite, causing a hang.
811 - clarified lots of AnyEvent::Handle constructor arguments.
812
8134.232 Thu Aug 21 20:44:25 CEST 2008
814 - fix a bug in AnyEvent::Handle that could cause two fatal error
815 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
816 (testcase by Leon Brocard).
817 - clarified on_eof/on_error documentation a bit.
818
8194.231 Tue Jul 29 13:12:15 CEST 2008
820 - remove some debugging code left in AnyEvent::Util::fork_call
821 (and no, it's impossible to implement with the broken windows
822 perls without resource leaks or worse).
823
8244.23 Tue Jul 29 12:19:59 CEST 2008
825 - document the first parameter passed to condvar callbacks to be
826 the callback.
827 - add AnyEvent::Socket::{ntoa,aton} aliases.
828 - optimize the AE::Handle->push_read (line) for the default
829 eol marker.
830 - optimize push_read (packstring|storable) for small packets.
831 - invoke on_error callback when no on_eof callback is set.
832 - fix a bug in push_read (storable) of unknown impact.
833
8344.22 Sun Jul 20 16:34:13 CEST 2008
835 - new function AnyEvent::Socket::parse_hostport.
836 - as the bulkheads at microsoft can't even get getprotobyname reliably
837 working on their shitty fucking broken os we need to hardcode
838 some common protocol numbers in AnyEvent::Socket. How can
839 people even bother with such a pile of shit as windows.
840
8414.21 Thu Jul 17 14:40:05 CEST 2008
842 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
843 which is incompatible to previous invocation syntax.
844 - work around more windows perl bugs in fork_call. I cannot imagine
845 why anybody would bother implementing fork in such an obviously
846 unusable way.
847 - avoid calling $do_retry if the request has already finished in
848 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
849 K. Sharifulin).
850
8514.2 Sat Jul 12 22:42:11 CEST 2008
4 - support an "interval" argument for repeating timers. 852 - support an "interval" argument for repeating timers.
853 - fix a bug in the pure perl event loop that caused timers to
854 be delayed under some circumstances.
855 - implement optional argument checking via PERL_ANYEVENT_STRICT.
5 - not importing anything from Event causes it to malfunction 856 - not importing anything from Event causes it to malfunction
6 and not use Time::HiRes, work around this bug. 857 and not use Time::HiRes, work around this bug.
7 - fix a bug in the pure perl event loop that caused timers to 858 - more bug workarounds against the endlessly broken Tk module,
8 be delayed under some circumstances. 859 seems to work now on 32 bit machines, but less so on 64 bit.
9 - minor optimisations applied to most backends. 860 - minor optimisations applied to most backends.
10 861
114.161 Fri Jul 4 14:24:48 CEST 2008 8624.161 Fri Jul 4 14:24:48 CEST 2008
12 - fixed the t/handle/02_write.t, it read after write, with the 863 - fixed the t/handle/02_write.t, it read after write, with the
13 assumption that all data could be written before read 864 assumption that all data could be written before read
… …
167 - support service names in tcp_connect and tcp_server. 1018 - support service names in tcp_connect and tcp_server.
168 - work around netbsd/openbsd bug causing unix sockets not to 1019 - work around netbsd/openbsd bug causing unix sockets not to
169 become writable until empty. 1020 become writable until empty.
170 - only make a dummy read for error detection when getpeername 1021 - only make a dummy read for error detection when getpeername
171 returns ENOTCONN. 1022 returns ENOTCONN.
172 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 1023 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
173 override. 1024 override.
174 - new AnyEvent::Guard::cancel. 1025 - new AnyEvent::Guard::cancel.
175 1026
1763.5 Sat May 17 23:17:14 CEST 2008 10273.5 Sat May 17 23:17:14 CEST 2008
177 - gracefully deal with systems not defining CLOCK_MONOTONIC 1028 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
191 1042
1923.41 Sun May 11 19:53:13 CEST 2008 10433.41 Sun May 11 19:53:13 CEST 2008
193 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 1044 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
194 1045
1953.4 Sun May 11 00:29:25 CEST 2008 10463.4 Sun May 11 00:29:25 CEST 2008
196 - complete re-implementation of condvars with a hopefully much mroe useful 1047 - complete re-implementation of condvars with a hopefully much more useful
197 API (backward compatible functions exist but are undocumented). 1048 API (backward compatible functions exist but are undocumented).
198 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 1049 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
199 (still subject to change). 1050 (still subject to change).
200 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 1051 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
201 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 1052 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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