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

Comparing AnyEvent/Changes (file contents):
Revision 1.270 by root, Wed Jul 8 05:30:55 2009 UTC vs.
Revision 1.601 by root, Fri May 8 22:47:13 2015 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.801 3TODO: document TCP_*** constants
4TODO: add debug/trace logging to some modules
5TODO: use the eval-with-cb-call trick also to speed up JSON?
6TODO: FAQ: common pitfalls? invoke-before-return
7TODO: FAQ: scope-ids? hosts file order?
8TODO: AnyEvent::Handle with AnyEvent::IO?
9TODO: invalid. and localhost. specialcasing inside AEDNS and not AESocket (rfc6761)
10TODO: maybe implement env variable to give hosts precedence
11TODO: hosts always read? that's not expected
12TODO: more formats for parse_ipv6 (single ipv6 address without port, ...p80, ...#80)
13
14 - aligned ipv6 address formatting with RFC 5952 (by not shortening
15 a single :0: to ::).
16
177.09 Sat May 2 16:38:53 CEST 2015
18 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
19 that was renamed to AnyEvent::Log:format_time. uinder its old name
20 (analyzed by Michael Stovenour).
21 - update AnyEvent::DNS fallback resolver addresses:
22 seems google effectively killed most other free dns resolvers,
23 so remove them, but add cable and wireless (ecrc) since it was
24 stable for 20 years or so, official or not, and there should be
25 an alternative to google.
26 - perl5porters broke windows error codes in 5.20, and mapped
27 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
28 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
29 because they were so cool. They probably broke other error codes
30 for no reason, too, but I didn't care for checking, it's too
31 depressing. This version only works around the WSAEWOULDBLOCK
32 issue, because I don't have a nice way to work around the
33 WSAEINPROGRESS bug.
34
357.08 Wed Dec 10 05:27:17 CET 2014
36 - work around a newly introduced bug in Socket 2.011 (an
37 errornous sun_length check) (analyzed by Maxime Soulé).
38 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
39 (analyzed by Ben Magistro).
40 - AnyEvent::Strict will now confess, not croak. This is in line with
41 it being a development/debugging tool.
42 - work around a number of libglib bugs (debug builds of libglib enforce
43 certain undocumented behaviour patterns such as not being able to
44 remove a child watch source after it has fired, which we will try
45 to emulate to avoid "criticals". what where they thinking...).
46 - mention json security issues in AnyEvent::Handle, now that Douglas
47 Crockford has foolishly and incompatibly changed JSON.
48 - changed default dns resolver "max_outstanding" value from 1 to 10,
49 the latter beinfg the intended value all along
50 (reported by Ilya Chesnokov).
51 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
52 (written by Mike Lowell).
53
547.07 Tue Dec 17 17:45:02 CET 2013
55 - the documentation for custom tls verify schemes was wrong. make it agree
56 with the code (reported by Maxime Soulé).
57 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
58 - work around an API change in openssl that could cause wrong tls connection
59 aborts, likely on windows only (analyzed by sten).
60 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
61 endless loop (reported by Dietrich Rebmann).
62 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
63 (patch by Maxime Soulé).
64 - document AnyEvent::Impl::IOAsync::set_loop and
65 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
66 functionality has _always_ been available.
67 - force a toplevel domain name in t/81_hosts.t (analyzed by
68 David Jack Wange Olrik).
69 - document that AnyEvent::Log uses AnyEvent::IO.
70 - warn about AnyEvent::Filesys::Notify performance.
71 - praise the joys of AnyEvent::Fork::*.
72 - time for an =encoding directive.
73 - do no longer use JSON to create a default json coder, use
74 JSON::XS or JSON::PP directly.
75
767.05 Wed Aug 21 10:38:08 CEST 2013
77 - uts46data.pl couldn't be found due to wrong naming of the file
78 (reported by Fulko Hew).
79 - handle lone \015's properly in AE::Handle's default line read
80 (reported by various people).
81 - untaint ip addresses found in /etc/hosts (patch by José Micó).
82 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
83 fail on first use (reported and testcase by Andrew Whatson).
84 - expose AnyEvent::Log::format_time, and allow users to redefine it.
85 - expose AnyEvent::Log::default_format, and allow redefinition.
86 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
87 - AnyEvent::Debug shell can now run coro shell commands, if available.
88 - t/63* tests were wrongly in MANIFEST.
89 - kernel.org's finger server went MIA, switch to freebsd.org and
90 icculus.org.
91 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
92 to function asynchronously (reported by Daniel Carosone).
93 - hard-disable $^W in most tests, it generates too much garbage output.
94 - use a (hopefully) more future-proof method to emulate common::sense.
95 - upgrade to UTS-46:6.2.0.
96 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
97 documented. should not affect anything.
98
997.04 Thu Nov 15 02:23:58 CET 2012
100 - debugging code left in.
101
1027.03 Thu Nov 15 02:16:11 CET 2012
103 - AnyEvent::Socket::inet_aton did not work when dns resolution was
104 used to find the addresses (analyzed and patch by Maxime Soulé).
105 - fix a memory leak in the /etc/hosts lookup code when hosts don't
106 resolve and are not in hosts (analyzed by sten).
107
1087.02 Tue Aug 14 04:27:58 CEST 2012
109 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
110 program by Yorhel).
111 - verified that AnyEvent::Socket follows RFC5952.
112 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
113
1147.01 Sun May 13 01:03:17 CEST 2012
115 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
116 not available, instead of throwing an exception.
117 - use File::Spec to get the tmpdir in t/*, to avoid needless
118 failures on (most, not mine :) windows boxes.
119 - new handle read types: tls_detect and tls_autostart.
120
1217.0 Fri Apr 13 06:33:30 CEST 2012
122 - child watchers are broken in POE 1.352 (also many earlier
123 versions) and there seems to be no way to work around it, as POE
124 itself is inherently racy. Document this regression and add a delay
125 in t/68_poe_03_child.t for the time being.
126 - new module AnyEvent::IO, that is a frontend to either a pure-perl
127 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
128 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
129 - load /etc/hosts only when DNS has no answer.
130 - stat /etc/hosts on every access and reload it if it changed.
131 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
132 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
133 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
134 a side effect, they now use true appending as opposed to libc
135 appending, and the file might not have been opened when the function
136 returns.
137 - the default logging level was not properly documented in a variety of
138 places, this has hopefully been rectified.
139 - updated uts46data.pl for unicode 6.1.0.
140 - made log messages generated by AnyEvent submodules not
141 include the package name anymore, as it will be logged
142 by default already.
143 - upgrade to the trick used by common-sense 3.5 to work
144 around extra warning torture/breakage under perl 5.15.x.
145 - log messages by anyevent are now ucfirst, are usually full
146 sentences and do no longer include the package name.
147 - the storable read type would sometimes throw an exception instead
148 of causing EBADMSG (analyzed by Maxime Soulé).
149 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
150 keysize (reported by Paul Howarth).
151 - documented Rocco fabricating statements about AnyEvent and me.
152
1536.14 Tue Jan 31 20:00:24 CET 2012
154 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
155 probably an editing mistake (reported by Darin McBride).
156
1576.13 Thu Jan 12 07:27:01 CET 2012
158 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
159 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
160 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
161 regex-matching inside the callback would change the parameter, despite
162 $1 being dynamically scoped per-block. this perl bug is now being
163 worked around (testcase by Cindy Wang).
164
1656.12 Mon Dec 12 13:21:10 CET 2011
166 - $! was clobbered by subsequent calls in tcp_connect, due to
167 the postpone not saving/restoring it (analyzed by Richard Garnish).
168
1696.11 Tue Nov 22 10:36:05 CET 2011
170 - Tk cannot create windows when tainted unless you set a title, so
171 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
172 - escape any nonprintable/nonascii characters when stringifying
173 backtraces.
174 - log the reason loading an interface module fails to load at level debug.
175
1766.1 Tue Oct 4 19:44:30 CEST 2011
177 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
178 and above), and some messages inside AnyEvent have been elevated
179 to higher log levels to print by default.
180 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
181 - work around yet another signal race bug, newly introduced by POE. My
182 how that module sucks.
183 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
184 support ipv6-only operation even with defaults (also use ::1 instead
185 of 127.0.0.1).
186 - give better guidelines on how to chose a logging level.
187 - AnyEvent::Log can now cap (limit) log message priority by context.
188 - AE::log will now save and restore $! and $@, so callers don't have to.
189 - AE::Strict now checks that registered fds are still valid at regular
190 intervals.
191 - skip frequently called 'require' calls if the module appears loaded.
192 - increase signal latency test timer considerably for poor little osx,
193 which can't be bothered with basic posix functionality such as signal
194 delivery.
195 - mention some debugging aids in the tutorial.
196 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
197 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
198 of the context as title.
199 - AnyEvent::Strict would not restore read-write access to $_ when a callback
200 died - this is a user bug, but since AE::Strict exists to help dbeugging...
201 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
202 to work with it - the author is well aware of the issues but refuses
203 to fix them. This extreme step has been taken because the module actively
204 attacks the goals of AnyEvent and makes it harder for module authors to use
205 AnyEvent.
206
2076.02 Fri Aug 26 20:08:31 CEST 2011
208 - log_to_syslog now accepts facility strings.
209 - log_to_syslog would not allow facilities - "$facility|$level"
210 *literally* means that string. sheesh.
211 - the CPAN parser got confused and thought we provide Sys::Syslog, try
212 to work around.
213 - renamed AnyEvent::Impl::FLTK2 to FLTK.
214
2156.01 Fri Aug 26 07:04:11 CEST 2011
216 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
217 the TTL as fourth element - this affects ->resolve, ->request
218 and ->dns_unpack indirectly. Convenience functions such as
219 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
220 AnyEvent::Socket.
221 - INCOMPATIBLE CHANGE: backend authors now should not implement
222 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
223 methods.
224 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
225 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
226 - DEPRECIATION: please change
227 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
228 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
229 in your programs, for future compatibility.
230 - *blush* idle watcher emulation was borked.
231 - AnyEvent::Strict now checks the AE::xxx functions as well.
232 - querying too long domain names would barf deep inside AnyEvent::DNS - since
233 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
234 for too long domain names explicitly and passes an undef to the callback
235 instead (reported by James Bromberger).
236 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
237 - new functions: AnyEvent::postpone, AnyEvent::log.
238 - new module: AnyEvent::Log, for simple logging needs.
239 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
240 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
241 a debug shell.
242 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
243 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
244 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
245 /etc/hosts can be found.
246 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
247 configuration.
248 - env variables can now be specified using AE_xyz instead of
249 PERL_ANYEVENT_xyz, subject to some rules.
250 - lots of new AnyEvent::Debug functionality.
251 - AnyEvent::Strict now checks for modifications of $_ by the callback
252 (by setting it to readonly while executing the callback, if possible).
253 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
254 Invest extra overhead to work around this peculiar behaviour.
255 - split the pure perl backend into a loop module and an impl
256 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
257 - some read types (regex, netstring, json, storable) could
258 errornously keep an AE::Handle object alive on their own
259 (reported by Mohammad Toossi).
260 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
261 supports /etc/hosts overrides.
262 - add more workarounds around child watcher bugs in IO::Async
263 and (newly introduced) in POE.
264 - use glib child watchers and try to work around its limitations,
265 if possible (based on a patch by Kevin Ryde).
266 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
267 - many read types in AnyEvent::Handle would malfunction during
268 transitions to TLS because of a "delete $self->{rbuf}".
269 (reported by Gerald Galster).
270 - improve condvar blocking wait performance (probably only noticable
271 with faster backends), also make it easier for event loops that
272 do not support blocking waits.
273 - PERL_ANYEVENT_MODEL now also allows full module names, although
274 the usefulness of this feature is not yet clear.
275 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
276 - add t/80_ssltext.t.
277 - no longer use AUTOLOAD for the initial loading (for no very good reason).
278 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
279 - do not run event loop detection multiple times when programs call their cached
280 copies of AnyEvent::detect.
281 - parse_hostport now accepts absolute paths als unix domain sockets.
282 - suppress the idiotic warning inside POE by patching the POE::Kernel
283 run flags instead of asking the user to jump through contortions.
284 - add Task::Weaken as dependency (as recommends only at this time).
285 - replace PF_UNSPEC by 0, for good old minix.
286
2876.0 (not officially released, but was on CPAN for about 15 minutes :/)
288
2895.34 Thu May 12 09:59:41 CEST 2011
290 - AE::TLS didn't support both cert_file and key at the same time,
291 due to some copy&paste bug (reported and analyzed by vti).
292 - improve AE::Handle timeout documentation slightly.
293
2945.33 Mon Feb 7 21:51:41 CET 2011
295 - never appeared on CPAN, for unknown reasons.
296 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
297 object when it's already gone (analyzed by Ryan Bullock).
298 - update idna algorithm(s) to UTS#46 version 6.0.0.
299 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
300 uts46 says it's an error, and testsuite makes more sense.
301 - remove leading dots in idn_to_ascii - testsuite says to remove,
302 uts46 says it's an error, and testsuite makes more sense.
303 - fixed a bug in idn ignored character processing.
304 - passes the uts46 6.0.0 testsuite except for three cases which
305 apparently are bugs in the testsuite itself.
306 - passes the uts46 6.0.1 testsuite except for eleven cases which
307 apparently are bugs in the testsuite itself.
308
3095.31 Sun Jan 23 11:49:19 CET 2011
310 - work around a POE bug where POE would first poll for new events and
311 invoke handlers BEFORE handling the message queue.
312 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
313 - stop_read now does just that in TLS mode. see its
314 documentation for caveats.
315 - update to common::sense 3.4 warning set (some warnings in perl impose
316 an arbitrary 32kb limit on regex matches, which affects
317 AnyEvent::Handle).
318 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
319 - new experimental AE::Handle::resettls method.
320
3215.3 Fri Dec 31 05:48:13 CET 2010
322 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
323 issues with IO::Async have been solved in that version, including
324 the availabilty of a default loop!
325 - implement new wbuf_max parameter for AnyEvent::Handle.
326 - added Cocoa::EventLoop backend by Daisuke Murase.
327 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
328 - clarify handle on_prepare documentation.
329 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
330 to avoid leaking them to any child processes.
331 - try to force files into the arch-dependend lib directory, for the
332 half-"turly obsessive compulsive person"(s) that depend on it.
333
3345.29 Sun Dec 5 10:49:21 CET 2010
335 - convert EV backend to EV 4.00 API (so better upgrade EV too).
336 - AE::Handle: implement dynamic read_size adjustment, add
337 max_read_size parameter, reduce default read_size to 2048.
338 - add some FAQ entries for servers and/or on_eof vs. on_error.
339 - work around OS bugs (cygwin again) when lingering in AE::Handle.
340 - reduce memory usage slightly when connect was used in AE::Handle.
341 - wrok around more segfaults in Tk.
342 - document tls_ctx => undef as valid.
343 - detect WNOHANG value at compiletime.
344 - include some probably linux-only support for building AnyEvent
345 as part of the perl core.
346 - improve compatibility of t/08_idna.t to perl 5.8.
347 - make AnyEvent::Strict truly optional (in case somebody builds a
348 minimal perl...).
349
3505.28 Wed Oct 13 04:14:23 CEST 2010
351 - due to a glitch, AnyEvent's internal getprotobyname wasn't
352 actually used in tcp_server.
353 - implement AnyEvent::CondVar->new method, which was documented but
354 missing.
355 - support autoloading of read types in unshift_read.
356 - AnyEvent::Strict: do the test for invalid fh arguments differently,
357 resulting in more informative error messages.
358 - new AnyEvent::FAQ manpage.
359 - clarify that you can actually call push_read etc. in on_read.
360 - put keys and certs in /tmp, not into the current directory,
361 and document this in the SECURITY CONSIDERATIONS sections
362 of AnyEvent::TLS.
363 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
364 requested by Ron Isaacson.
365 - document $AnyEvent::DNS::RESOLVER variable.
366 - incorporated major typo patches by Abhijit Menon-Sen.
367 - AnyEvent::Handle now croaks when negative timeouts are passed.
368 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
369 undocumented, but who knows when they might come in handy.
370
3715.271 Tue Jun 8 12:05:46 CEST 2010
372 - backport to perl 5.8.x.
373
3745.27 Sun Jun 6 12:12:05 CEST 2010
375 - postpone differently in AnyEvent::Socket now, as
376 when not, canceling the connection attempt might fail
377 (found by Felix Antonius Wilhelm Ostmann).
378 - explicitly check for non-stream sockets in AE::Handle, too many
379 clueless people fell into the trap of this somehow working.
380 - simplified and reworked the "OTHER MODULES" section.
381 - better/more condvar examples.
382
3835.261 Wed Apr 28 16:13:36 CEST 2010
384 - AF_INET6 was not properly used from Socket6 during configuration
385 time (found by Andy Grundman).
386
3875.26 Mon Apr 12 04:49:35 CEST 2010
388 - don't generate spurious readyness notifications when select
389 returns EINTR in the pure perl backend (analysed and testcase
390 by Bas Denissen).
391 - give same examples in SYNOPSIS sections of AnyEvent and AE.
392 - provider faster implementations for AE::time/now/now_update for
393 EV and Perl backends.
394 - provide faster AE::cv and AE::time implementations for all
395 backends.
396 - no longer support lower/mixed-case signal names (this was
397 never documented nor universally supported).
398 - some more memory µ-optimisations, and somewhat less messy
399 code to cope with both AnyEvent and AE APIs.
400
4015.251 Sat Mar 13 00:58:21 CET 2010
402 - make test failed if it was run before make install on perl <=
403 5.8.8, and older versions of perl make our live very difficult,
404 so write constants.pl during Makefile.PL time.
405
4065.25 Sat Mar 13 00:23:14 CET 2010
407 - fix a race condition in AnyEvent::Handle that would cause
408 a "bio_read: unsupported method" error (in _another_
409 ssl connection) after user code threw an exception.
410 - added AnyEvent::Handle->destroyed method.
411 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
412 - the AnyEvent::Util::fh_nonblocking function worked only
413 by lucky accident on win32.
414 - smaller and faster AnyEvent::Util::fh_nonblocking.
415 - when the (required!) Time::HiRes module is missing, AnyEvent
416 did not fall back to built-in time properly.
417 - do not load Fcntl at runtime, saving memory and loading time.
418 - precompile a number of constants and use them instead of runtime
419 detection and eval.
420 - free detection code after detection and similar memory optimisations.
421 - Perl backend timer interval best effort drift has been improved
422 (same algorithm as EV).
423 - update unicode idna mapping table.
424
4255.24 Tue Jan 5 11:39:43 CET 2010
426 - cygwin never reports errors from failed connects. 1.5 just gives
427 you continous readyness and EAGAIN, 1.7 is even more broken and
428 just hangs. work around both issues in a cygwin-specific hack.
429 - improve idle watcher documentation slightly.
430
4315.23 Sun Dec 20 23:48:00 CET 2009
432 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
433 - implement punycode_encode/decode, idn_nameprep,
434 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
435 - provide $AE::VERSION.
436 - removed traces of "no strict 'refs'".
437
4385.22 Sat Dec 5 03:51:13 CET 2009
439 - downgrade-or-fail in AnyEvent::Handle::push_write, to
440 diagnose encoding failures earlier and more succinctly.
441 (this works around bugs in perl, throwing away encoding info
442 when passing scalar data to extensions).
443 - add more examples to AnyEvent::Socket manpage.
444 - upgrade internal warning set to the same as common::sense 2.03.
445 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
446 (John Beppu).
447 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
448 - add google public dns servers to fallback server set and make sure
449 we load-balance properly between the three sets. also add all
450 fallback dns servers, not just a random one, to each dns config.
451
4525.21 Thu Nov 19 02:48:47 CET 2009
453 - fix a problem where socket constants were called with parameters
454 (spotted by David Friedland).
455 - fork_call never use'd POSIX (reported by Daisuke Maki).
456 - improve perl 5.6 compatibility further (but it still won't work
457 unless you rip out everything but the core).
458 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
459 uses 10MB of RAM, but doesn't open a console window. *sigh*.
460
4615.202 Wed Oct 14 22:35:44 CEST 2009
462 - AnyEvent::DNS would unexpectedly clobber $_ under windows
463 (analysed by Matthias Waldorf).
464 - AnyEvent::Handle::run_cmd can now pass the PID of the
465 newly-created process, which is much less useful than it might
466 sound (based on patch by Yann Kerherve).
467
4685.201 Tue Sep 29 12:09:25 CEST 2009
469 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
470 (reported by Torsten Foertsch).
471 - common::sense 2.0 could cause tcp_server to throw an exception
472 (analysed by elmex).
473
4745.2 Mon Sep 14 07:04:49 CEST 2009
475 - INCOMPATIBLE CHANGE: do no longer support register_read_type
476 and register_write_type in AnyEvent::Handle, instead support
477 package names (the facility was mostly abused).
478 - implement "packagename-as-read/write type" support in
479 AnyEvent::Handle.
480 - AnyEvent::Handle: new options "keepalive" and "oobinline".
481 - oobinline set by default to avoid security issues.
482 - the pure-perl event loop backend wrongly detected times() fallback
483 support (spotted by Pavel Boldin).
484
4855.12 Tue Sep 1 20:26:50 CEST 2009
486 - be more lenient when parsing resolv.conf files, as some people
487 use hashmarks as comment indicator inside directives (reported
488 by Michael S. Fischer).
489 - use same set of warnings as common::sense 2.0.
490 - fix a potential 32 bit overflow issue due to perl having problems
491 with large hex constants in the Perl backend.
492 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
493 by default.
494 - fix some prototypes.
495 - (optionally) require Async::Interrupt 1.02.
496 - replace opendns fallbacks by verizon/level3 public dns servers.
497
4985.112 Fri Aug 21 13:59:04 CEST 2009
499 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
500 using starttls with an illegal/uncreatable tls context.
501 - port to broken win32 perls.
502
5035.111 Sun Aug 16 18:44:36 CEST 2009
504 - perl errornously requires sockaddr_un structures returned by the
505 kernel to have a specific length. work around this bug by padding
506 sockaddr_un structures (found to fail on OS X, as perl only uses
507 the correct code for GNU/Linux, but likely to fail on other
508 OSes as well).
509
5105.11 Wed Aug 12 17:49:37 CEST 2009
511 - fixed a bug in run_cmd with input from filename.
512 - create a json coder object when none is given in write direction
513 to reduce dependency on JSON.
514
5155.1 Tue Aug 11 03:17:32 CEST 2009
516 - tcp_connect now reports the correct error in $!, not an unrelated
517 one (was broken since 4.91).
518 - AE::Handle did not properly clear rtimeout/wtimeout watchers
519 when retrying a tcp_connect.
520 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
521 - fixed wrong documentation in AE::Socket.
522
5235.01 Mon Aug 10 03:16:32 CEST 2009
524 - last release broke EV child watchers :/.
525
5265.0 Sun Aug 9 17:32:09 CEST 2009
527 - officially introduce and document the AE API.
528 - lowering the timeout at runtime did not have immediate effect
529 in AnyEvent::Handle.
530 - AE::Handle now has separate and independent read and write timeout
531 settings.
532 - probe for Guard module when AnyEvent::Util is loaded, not at
533 runtime (Event doesn't like eval "use" from callbacks, and it avoids
534 exporting the probe function).
535 - Impl::IOAsync's io watchers did not use the filehandle emulation
536 code to work around it's watcher limits.
537 - work around IO::Async's problems with overloaded objects.
538 - try to disable wearnings inside IO::Async, it's just too buggy.
539 - do no longer use Event by default, only EV or Perl, Event is too buggy.
540 - hackishly provide fast AE:: interface when EV or Perl is the backend.
541 - inherit VERSION from AnyEvent::VERSION in most modules.
542
5434.91 Thu Aug 6 15:42:45 CEST 2009
544 - AE::Handle::starttls could get out of sync when the read buffer
545 already contains some TLS handshake.
546 - AE::Handle did not properly free the TLS session, causing bigger
547 memory leaks in Net::SSLeay than expected (reported by toaster).
548 - AE::Socket::tcp_connect will now artificially delay invoking
549 the callback to avoid returning after invoking the callback.
550 - convert many internal watcher uses to AE API.
551
5524.9 Sat Aug 1 11:07:01 CEST 2009
553 - Glib silently fails when registering a timeout with
554 a negative value, so avoid that.
555 - call condvar callback immediately when it is set
556 after the condition is already signalled.
557 - check rbuf_max condition only after trying to consume
558 data.
559 - normalise signal numbers to names when using the ae's
560 signal handling, but do not document this (yet).
561 - pure perl signal emulation did not properly set nonblocking
562 mode on the signal pipe on win32 (this is unlikely to have caused
563 any issues).
564 - new module: AnyEvent::Debug.
565 - AnyEvent::Strict now checks that a signal was specified
566 by name (not name or number).
567 - reduce memory footprint in typical cases by ~50kb by
568 only compiling signal code when necessary.
569 - add AnyEvent::Handle->rbuf_max.
570 - grab the AE namespace for future new API, implement stubs for
571 said future/alternative API.
572 - new function: AnyEvent::Socket::format_hostport.
573
5744.881 Tue Jul 28 12:51:53 CEST 2009
575 - work around a bug in local in pre-5.10 perls, causing
576 AnyEvent::Handle to recurse when it should not
577 (analysed by elmex).
578
5794.88 Tue Jul 28 04:04:37 CEST 2009
580 - re-bless the handle into a dummy package after calling
581 AnyEvent::Handle::destroy, so the user does not need to check
582 for errors after every push_write etc.
583 - do not attempt to run t/02_signals.t on obviously broken
584 platforms.
585
5864.87 Sun Jul 26 02:06:16 CEST 2009
587 - do not attempt to linger when there is no longer a valid fh.
588 - enforce tls mode to be either accept or connect, do not simply
589 segfault in Net::SSLeay.
590 - AnyEvent::Handle can now call tcp_connect itself (new parameters
591 connect, on_prepare, on_connect and on_connect_error). Updated
592 tutorial accordingly.
593 - add AnyEvent::Impl::Irssi backend.
594
5954.86 Mon Jul 20 23:52:29 CEST 2009
596 - since the verbose warning is not enough, explicitly document
597 that versions before 1.33 of Net::SSLeay are not secure.
598 - work around signal handling races in Event and (...) Event::Lib.
599 - try to align signal-race timer to full-second boundaries.
600 - work around Tk not liking negative timeouts.
601 - don't complain of different grades of environmental unfriendlyness
602 in IO::Async.
603
6044.85 Sat Jul 18 06:16:14 CEST 2009
605 - nail the signal race problem in perl once and for all
606 (see $AnyEvent::MAX_SIGNAL_LATENCY).
607 - take advantage of Async::Interrupt if it is available.
608 - load Time::HiRes and Guard modules on demand only.
609 - add optional/recommended modules section to AnyEvent
610 documentation.
611 - reduce memory usage considerably (and reduce startup penalty)
612 by not using "strict", "warnings" and "overload" modules.
613 - work around buggy windows/openbsd perls and provide EBADMSG
614 and EPROTO ourselves when missing.
615 - improve perl 5.6 compatibility of the core event loop.
616 - made Net::SSLeay version 1.33 a soft requirement.
617
6184.83 Fri Jul 17 16:56:26 CEST 2009
619 - implement AnyEvent::Socket::getprotobyname.
620 - AnyEvent::CondVar's will now detect recursive blocking
621 waits and will croak, as too many people fall into
622 this trap.
623 - AnyEvent::Handle will now call ->destroy on itself after
624 executing the on_error callback, instead of doing some
625 half-baked internal shutdown, for fatal errors.
626 - clarify on_eof behaviour w.r.t. the read queue and
627 on_read callbacks.
628 - ignore some possible spurious wake-ups in tcp_connect.
629
6304.82 Sat Jul 11 00:34:55 CEST 2009
631 - POE and Event backends didn't accept some callable objects as
632 callbacks.
633 - use Config module instead of POSIX module to detect signal names
634 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
635 module doesn't even have all POSIX signals :/.
636 - use more workarounds around the many refcnt/corruption bugs in
637 Event::Lib.
638 - work around a race condition in perl's select, causing t/03_child.t
639 to rarely fail.
640
6414.81 Thu Jul 9 10:30:30 CEST 2009
642 - AnyEvent::Handle didn't properly diagnose write errors
643 (it expected -1 from syswrite, how lame... :).
4 - support file descriptors in addition to file handles 644 - support file descriptors in addition to file handles
5 in AnyEvent->io. 645 in AnyEvent->io.
6 - new env variables: PERL_ANYEVENT_RESOLV_CONF, 646 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
7 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE 647 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
8 and PERL_ANYEVENT_CA_PATH. 648 and PERL_ANYEVENT_CA_PATH.
9 - provide a sensible synopsis section for AnyEvent::TLS. 649 - provide a sensible synopsis section for AnyEvent::TLS.
650 - add a "supported backends" section to the manpage.
10 - added simple io watcher test to testsuite, using a 651 - added simple io watcher test to testsuite, using a
11 portable_socketpair. 652 portable_socketpair.
653 - tried to improve the stability of the Event::Lib backend,
654 YMMV.
12 655
134.8 Mon Jul 6 23:45:16 CEST 2009 6564.8 Mon Jul 6 23:45:16 CEST 2009
14 - AnyEvent::DNS did not properly follow CNAME records with 657 - AnyEvent::DNS did not properly follow CNAME records with
15 uppercase targets. 658 uppercase targets.
16 - AnyEvent::DNS would errornously return AAAA records 659 - AnyEvent::DNS would errornously return AAAA records
… …
27 in the common case. 670 in the common case.
28 - add AnyEvent::Handle->push_shutdown method. 671 - add AnyEvent::Handle->push_shutdown method.
29 - add an additional error message parameter to AnyEvent::Handle's 672 - add an additional error message parameter to AnyEvent::Handle's
30 on_error callback (for TLS, $! is still available). 673 on_error callback (for TLS, $! is still available).
31 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 674 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
32 - make AnyEvent::Handle more robust against early conenction 675 - make AnyEvent::Handle more robust against early connection
33 failures (during new), and return C<undef> in that case 676 failures (during new), and return C<undef> in that case
34 from the constructor. 677 from the constructor.
35 - AnyEvent::Handle will now try to load only JSON::XS first, 678 - AnyEvent::Handle will now try to load only JSON::XS first,
36 then fall back to JSON. 679 then fall back to JSON.
37 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 680 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
1614.3 Fri Oct 3 09:18:43 CEST 2008 8044.3 Fri Oct 3 09:18:43 CEST 2008
162 - AnyEvent will now install a no-op signal handler for SIGPIPE, 805 - AnyEvent will now install a no-op signal handler for SIGPIPE,
163 unless one has been installed already. 806 unless one has been installed already.
164 - warn about not loaidng AnyEvent::Impl::POE early enough 807 - warn about not loaidng AnyEvent::Impl::POE early enough
165 (patch by Adam Rosenstein). 808 (patch by Adam Rosenstein).
166 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 809 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
167 - fix a bug in where in SSL connect mode, the 810 - fix a bug in where in SSL connect mode, the
168 client would first wait for some data by the server and 811 client would first wait for some data by the server and
169 otherwise hang. (reported and analyzed in an absolutely 812 otherwise hang. (reported and analysed in an absolutely
170 exemplary manner by Adam Rosenstein). 813 exemplary manner by Adam Rosenstein).
171 - fix a bug in where SSL EOF would not be treated 814 - fix a bug in where SSL EOF would not be treated
172 as stream EOF, putting the connection into a hung state. 815 as stream EOF, putting the connection into a hung state.
173 (reported and analyzed in an absolutely exemplary manner by Adam 816 (reported and analysed in an absolutely exemplary manner by Adam
174 Rosenstein). 817 Rosenstein).
175 - fix a potential segfault when the TLS context 818 - fix a potential segfault when the TLS context
176 would go missing in a read callback (e.g. due to stoptls) - 819 would go missing in a read callback (e.g. due to stoptls) -
177 Net::SSLeay of course makes no type checking whatsoever. 820 Net::SSLeay of course makes no type checking whatsoever.
178 - AnyEvent::Handle will not stop the read watcher in TLS mode. 821 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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