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

Comparing AnyEvent/Changes (file contents):
Revision 1.258 by root, Thu Jul 2 23:22:34 2009 UTC vs.
Revision 1.602 by root, Tue Jun 16 03:04:26 2015 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
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 - new function AnyEvent::Socket::tcp_bind.
15 - aligned ipv6 address formatting with RFC 5952 (by not shortening
16 a single :0: to ::).
17 - added stability canary support.
18
197.09 Sat May 2 16:38:53 CEST 2015
20 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
21 that was renamed to AnyEvent::Log:format_time. uinder its old name
22 (analyzed by Michael Stovenour).
23 - update AnyEvent::DNS fallback resolver addresses:
24 seems google effectively killed most other free dns resolvers,
25 so remove them, but add cable and wireless (ecrc) since it was
26 stable for 20 years or so, official or not, and there should be
27 an alternative to google.
28 - perl5porters broke windows error codes in 5.20, and mapped
29 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
30 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
31 because they were so cool. They probably broke other error codes
32 for no reason, too, but I didn't care for checking, it's too
33 depressing. This version only works around the WSAEWOULDBLOCK
34 issue, because I don't have a nice way to work around the
35 WSAEINPROGRESS bug.
36
377.08 Wed Dec 10 05:27:17 CET 2014
38 - work around a newly introduced bug in Socket 2.011 (an
39 errornous sun_length check) (analyzed by Maxime Soulé).
40 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
41 (analyzed by Ben Magistro).
42 - AnyEvent::Strict will now confess, not croak. This is in line with
43 it being a development/debugging tool.
44 - work around a number of libglib bugs (debug builds of libglib enforce
45 certain undocumented behaviour patterns such as not being able to
46 remove a child watch source after it has fired, which we will try
47 to emulate to avoid "criticals". what where they thinking...).
48 - mention json security issues in AnyEvent::Handle, now that Douglas
49 Crockford has foolishly and incompatibly changed JSON.
50 - changed default dns resolver "max_outstanding" value from 1 to 10,
51 the latter beinfg the intended value all along
52 (reported by Ilya Chesnokov).
53 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
54 (written by Mike Lowell).
55
567.07 Tue Dec 17 17:45:02 CET 2013
57 - the documentation for custom tls verify schemes was wrong. make it agree
58 with the code (reported by Maxime Soulé).
59 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
60 - work around an API change in openssl that could cause wrong tls connection
61 aborts, likely on windows only (analyzed by sten).
62 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
63 endless loop (reported by Dietrich Rebmann).
64 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
65 (patch by Maxime Soulé).
66 - document AnyEvent::Impl::IOAsync::set_loop and
67 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
68 functionality has _always_ been available.
69 - force a toplevel domain name in t/81_hosts.t (analyzed by
70 David Jack Wange Olrik).
71 - document that AnyEvent::Log uses AnyEvent::IO.
72 - warn about AnyEvent::Filesys::Notify performance.
73 - praise the joys of AnyEvent::Fork::*.
74 - time for an =encoding directive.
75 - do no longer use JSON to create a default json coder, use
76 JSON::XS or JSON::PP directly.
77
787.05 Wed Aug 21 10:38:08 CEST 2013
79 - uts46data.pl couldn't be found due to wrong naming of the file
80 (reported by Fulko Hew).
81 - handle lone \015's properly in AE::Handle's default line read
82 (reported by various people).
83 - untaint ip addresses found in /etc/hosts (patch by José Micó).
84 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
85 fail on first use (reported and testcase by Andrew Whatson).
86 - expose AnyEvent::Log::format_time, and allow users to redefine it.
87 - expose AnyEvent::Log::default_format, and allow redefinition.
88 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
89 - AnyEvent::Debug shell can now run coro shell commands, if available.
90 - t/63* tests were wrongly in MANIFEST.
91 - kernel.org's finger server went MIA, switch to freebsd.org and
92 icculus.org.
93 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
94 to function asynchronously (reported by Daniel Carosone).
95 - hard-disable $^W in most tests, it generates too much garbage output.
96 - use a (hopefully) more future-proof method to emulate common::sense.
97 - upgrade to UTS-46:6.2.0.
98 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
99 documented. should not affect anything.
100
1017.04 Thu Nov 15 02:23:58 CET 2012
102 - debugging code left in.
103
1047.03 Thu Nov 15 02:16:11 CET 2012
105 - AnyEvent::Socket::inet_aton did not work when dns resolution was
106 used to find the addresses (analyzed and patch by Maxime Soulé).
107 - fix a memory leak in the /etc/hosts lookup code when hosts don't
108 resolve and are not in hosts (analyzed by sten).
109
1107.02 Tue Aug 14 04:27:58 CEST 2012
111 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
112 program by Yorhel).
113 - verified that AnyEvent::Socket follows RFC5952.
114 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
115
1167.01 Sun May 13 01:03:17 CEST 2012
117 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
118 not available, instead of throwing an exception.
119 - use File::Spec to get the tmpdir in t/*, to avoid needless
120 failures on (most, not mine :) windows boxes.
121 - new handle read types: tls_detect and tls_autostart.
122
1237.0 Fri Apr 13 06:33:30 CEST 2012
124 - child watchers are broken in POE 1.352 (also many earlier
125 versions) and there seems to be no way to work around it, as POE
126 itself is inherently racy. Document this regression and add a delay
127 in t/68_poe_03_child.t for the time being.
128 - new module AnyEvent::IO, that is a frontend to either a pure-perl
129 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
130 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
131 - load /etc/hosts only when DNS has no answer.
132 - stat /etc/hosts on every access and reload it if it changed.
133 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
134 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
135 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
136 a side effect, they now use true appending as opposed to libc
137 appending, and the file might not have been opened when the function
138 returns.
139 - the default logging level was not properly documented in a variety of
140 places, this has hopefully been rectified.
141 - updated uts46data.pl for unicode 6.1.0.
142 - made log messages generated by AnyEvent submodules not
143 include the package name anymore, as it will be logged
144 by default already.
145 - upgrade to the trick used by common-sense 3.5 to work
146 around extra warning torture/breakage under perl 5.15.x.
147 - log messages by anyevent are now ucfirst, are usually full
148 sentences and do no longer include the package name.
149 - the storable read type would sometimes throw an exception instead
150 of causing EBADMSG (analyzed by Maxime Soulé).
151 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
152 keysize (reported by Paul Howarth).
153 - documented Rocco fabricating statements about AnyEvent and me.
154
1556.14 Tue Jan 31 20:00:24 CET 2012
156 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
157 probably an editing mistake (reported by Darin McBride).
158
1596.13 Thu Jan 12 07:27:01 CET 2012
160 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
161 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
162 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
163 regex-matching inside the callback would change the parameter, despite
164 $1 being dynamically scoped per-block. this perl bug is now being
165 worked around (testcase by Cindy Wang).
166
1676.12 Mon Dec 12 13:21:10 CET 2011
168 - $! was clobbered by subsequent calls in tcp_connect, due to
169 the postpone not saving/restoring it (analyzed by Richard Garnish).
170
1716.11 Tue Nov 22 10:36:05 CET 2011
172 - Tk cannot create windows when tainted unless you set a title, so
173 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
174 - escape any nonprintable/nonascii characters when stringifying
175 backtraces.
176 - log the reason loading an interface module fails to load at level debug.
177
1786.1 Tue Oct 4 19:44:30 CEST 2011
179 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
180 and above), and some messages inside AnyEvent have been elevated
181 to higher log levels to print by default.
182 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
183 - work around yet another signal race bug, newly introduced by POE. My
184 how that module sucks.
185 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
186 support ipv6-only operation even with defaults (also use ::1 instead
187 of 127.0.0.1).
188 - give better guidelines on how to chose a logging level.
189 - AnyEvent::Log can now cap (limit) log message priority by context.
190 - AE::log will now save and restore $! and $@, so callers don't have to.
191 - AE::Strict now checks that registered fds are still valid at regular
192 intervals.
193 - skip frequently called 'require' calls if the module appears loaded.
194 - increase signal latency test timer considerably for poor little osx,
195 which can't be bothered with basic posix functionality such as signal
196 delivery.
197 - mention some debugging aids in the tutorial.
198 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
199 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
200 of the context as title.
201 - AnyEvent::Strict would not restore read-write access to $_ when a callback
202 died - this is a user bug, but since AE::Strict exists to help dbeugging...
203 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
204 to work with it - the author is well aware of the issues but refuses
205 to fix them. This extreme step has been taken because the module actively
206 attacks the goals of AnyEvent and makes it harder for module authors to use
207 AnyEvent.
208
2096.02 Fri Aug 26 20:08:31 CEST 2011
210 - log_to_syslog now accepts facility strings.
211 - log_to_syslog would not allow facilities - "$facility|$level"
212 *literally* means that string. sheesh.
213 - the CPAN parser got confused and thought we provide Sys::Syslog, try
214 to work around.
215 - renamed AnyEvent::Impl::FLTK2 to FLTK.
216
2176.01 Fri Aug 26 07:04:11 CEST 2011
218 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
219 the TTL as fourth element - this affects ->resolve, ->request
220 and ->dns_unpack indirectly. Convenience functions such as
221 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
222 AnyEvent::Socket.
223 - INCOMPATIBLE CHANGE: backend authors now should not implement
224 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
225 methods.
226 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
227 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
228 - DEPRECIATION: please change
229 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
230 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
231 in your programs, for future compatibility.
232 - *blush* idle watcher emulation was borked.
233 - AnyEvent::Strict now checks the AE::xxx functions as well.
234 - querying too long domain names would barf deep inside AnyEvent::DNS - since
235 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
236 for too long domain names explicitly and passes an undef to the callback
237 instead (reported by James Bromberger).
238 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
239 - new functions: AnyEvent::postpone, AnyEvent::log.
240 - new module: AnyEvent::Log, for simple logging needs.
241 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
242 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
243 a debug shell.
244 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
245 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
246 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
247 /etc/hosts can be found.
248 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
249 configuration.
250 - env variables can now be specified using AE_xyz instead of
251 PERL_ANYEVENT_xyz, subject to some rules.
252 - lots of new AnyEvent::Debug functionality.
253 - AnyEvent::Strict now checks for modifications of $_ by the callback
254 (by setting it to readonly while executing the callback, if possible).
255 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
256 Invest extra overhead to work around this peculiar behaviour.
257 - split the pure perl backend into a loop module and an impl
258 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
259 - some read types (regex, netstring, json, storable) could
260 errornously keep an AE::Handle object alive on their own
261 (reported by Mohammad Toossi).
262 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
263 supports /etc/hosts overrides.
264 - add more workarounds around child watcher bugs in IO::Async
265 and (newly introduced) in POE.
266 - use glib child watchers and try to work around its limitations,
267 if possible (based on a patch by Kevin Ryde).
268 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
269 - many read types in AnyEvent::Handle would malfunction during
270 transitions to TLS because of a "delete $self->{rbuf}".
271 (reported by Gerald Galster).
272 - improve condvar blocking wait performance (probably only noticable
273 with faster backends), also make it easier for event loops that
274 do not support blocking waits.
275 - PERL_ANYEVENT_MODEL now also allows full module names, although
276 the usefulness of this feature is not yet clear.
277 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
278 - add t/80_ssltext.t.
279 - no longer use AUTOLOAD for the initial loading (for no very good reason).
280 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
281 - do not run event loop detection multiple times when programs call their cached
282 copies of AnyEvent::detect.
283 - parse_hostport now accepts absolute paths als unix domain sockets.
284 - suppress the idiotic warning inside POE by patching the POE::Kernel
285 run flags instead of asking the user to jump through contortions.
286 - add Task::Weaken as dependency (as recommends only at this time).
287 - replace PF_UNSPEC by 0, for good old minix.
288
2896.0 (not officially released, but was on CPAN for about 15 minutes :/)
290
2915.34 Thu May 12 09:59:41 CEST 2011
292 - AE::TLS didn't support both cert_file and key at the same time,
293 due to some copy&paste bug (reported and analyzed by vti).
294 - improve AE::Handle timeout documentation slightly.
295
2965.33 Mon Feb 7 21:51:41 CET 2011
297 - never appeared on CPAN, for unknown reasons.
298 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
299 object when it's already gone (analyzed by Ryan Bullock).
300 - update idna algorithm(s) to UTS#46 version 6.0.0.
301 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
302 uts46 says it's an error, and testsuite makes more sense.
303 - remove leading dots in idn_to_ascii - testsuite says to remove,
304 uts46 says it's an error, and testsuite makes more sense.
305 - fixed a bug in idn ignored character processing.
306 - passes the uts46 6.0.0 testsuite except for three cases which
307 apparently are bugs in the testsuite itself.
308 - passes the uts46 6.0.1 testsuite except for eleven cases which
309 apparently are bugs in the testsuite itself.
310
3115.31 Sun Jan 23 11:49:19 CET 2011
312 - work around a POE bug where POE would first poll for new events and
313 invoke handlers BEFORE handling the message queue.
314 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
315 - stop_read now does just that in TLS mode. see its
316 documentation for caveats.
317 - update to common::sense 3.4 warning set (some warnings in perl impose
318 an arbitrary 32kb limit on regex matches, which affects
319 AnyEvent::Handle).
320 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
321 - new experimental AE::Handle::resettls method.
322
3235.3 Fri Dec 31 05:48:13 CET 2010
324 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
325 issues with IO::Async have been solved in that version, including
326 the availabilty of a default loop!
327 - implement new wbuf_max parameter for AnyEvent::Handle.
328 - added Cocoa::EventLoop backend by Daisuke Murase.
329 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
330 - clarify handle on_prepare documentation.
331 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
332 to avoid leaking them to any child processes.
333 - try to force files into the arch-dependend lib directory, for the
334 half-"turly obsessive compulsive person"(s) that depend on it.
335
3365.29 Sun Dec 5 10:49:21 CET 2010
337 - convert EV backend to EV 4.00 API (so better upgrade EV too).
338 - AE::Handle: implement dynamic read_size adjustment, add
339 max_read_size parameter, reduce default read_size to 2048.
340 - add some FAQ entries for servers and/or on_eof vs. on_error.
341 - work around OS bugs (cygwin again) when lingering in AE::Handle.
342 - reduce memory usage slightly when connect was used in AE::Handle.
343 - wrok around more segfaults in Tk.
344 - document tls_ctx => undef as valid.
345 - detect WNOHANG value at compiletime.
346 - include some probably linux-only support for building AnyEvent
347 as part of the perl core.
348 - improve compatibility of t/08_idna.t to perl 5.8.
349 - make AnyEvent::Strict truly optional (in case somebody builds a
350 minimal perl...).
351
3525.28 Wed Oct 13 04:14:23 CEST 2010
353 - due to a glitch, AnyEvent's internal getprotobyname wasn't
354 actually used in tcp_server.
355 - implement AnyEvent::CondVar->new method, which was documented but
356 missing.
357 - support autoloading of read types in unshift_read.
358 - AnyEvent::Strict: do the test for invalid fh arguments differently,
359 resulting in more informative error messages.
360 - new AnyEvent::FAQ manpage.
361 - clarify that you can actually call push_read etc. in on_read.
362 - put keys and certs in /tmp, not into the current directory,
363 and document this in the SECURITY CONSIDERATIONS sections
364 of AnyEvent::TLS.
365 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
366 requested by Ron Isaacson.
367 - document $AnyEvent::DNS::RESOLVER variable.
368 - incorporated major typo patches by Abhijit Menon-Sen.
369 - AnyEvent::Handle now croaks when negative timeouts are passed.
370 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
371 undocumented, but who knows when they might come in handy.
372
3735.271 Tue Jun 8 12:05:46 CEST 2010
374 - backport to perl 5.8.x.
375
3765.27 Sun Jun 6 12:12:05 CEST 2010
377 - postpone differently in AnyEvent::Socket now, as
378 when not, canceling the connection attempt might fail
379 (found by Felix Antonius Wilhelm Ostmann).
380 - explicitly check for non-stream sockets in AE::Handle, too many
381 clueless people fell into the trap of this somehow working.
382 - simplified and reworked the "OTHER MODULES" section.
383 - better/more condvar examples.
384
3855.261 Wed Apr 28 16:13:36 CEST 2010
386 - AF_INET6 was not properly used from Socket6 during configuration
387 time (found by Andy Grundman).
388
3895.26 Mon Apr 12 04:49:35 CEST 2010
390 - don't generate spurious readyness notifications when select
391 returns EINTR in the pure perl backend (analysed and testcase
392 by Bas Denissen).
393 - give same examples in SYNOPSIS sections of AnyEvent and AE.
394 - provider faster implementations for AE::time/now/now_update for
395 EV and Perl backends.
396 - provide faster AE::cv and AE::time implementations for all
397 backends.
398 - no longer support lower/mixed-case signal names (this was
399 never documented nor universally supported).
400 - some more memory µ-optimisations, and somewhat less messy
401 code to cope with both AnyEvent and AE APIs.
402
4035.251 Sat Mar 13 00:58:21 CET 2010
404 - make test failed if it was run before make install on perl <=
405 5.8.8, and older versions of perl make our live very difficult,
406 so write constants.pl during Makefile.PL time.
407
4085.25 Sat Mar 13 00:23:14 CET 2010
409 - fix a race condition in AnyEvent::Handle that would cause
410 a "bio_read: unsupported method" error (in _another_
411 ssl connection) after user code threw an exception.
412 - added AnyEvent::Handle->destroyed method.
413 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
414 - the AnyEvent::Util::fh_nonblocking function worked only
415 by lucky accident on win32.
416 - smaller and faster AnyEvent::Util::fh_nonblocking.
417 - when the (required!) Time::HiRes module is missing, AnyEvent
418 did not fall back to built-in time properly.
419 - do not load Fcntl at runtime, saving memory and loading time.
420 - precompile a number of constants and use them instead of runtime
421 detection and eval.
422 - free detection code after detection and similar memory optimisations.
423 - Perl backend timer interval best effort drift has been improved
424 (same algorithm as EV).
425 - update unicode idna mapping table.
426
4275.24 Tue Jan 5 11:39:43 CET 2010
428 - cygwin never reports errors from failed connects. 1.5 just gives
429 you continous readyness and EAGAIN, 1.7 is even more broken and
430 just hangs. work around both issues in a cygwin-specific hack.
431 - improve idle watcher documentation slightly.
432
4335.23 Sun Dec 20 23:48:00 CET 2009
434 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
435 - implement punycode_encode/decode, idn_nameprep,
436 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
437 - provide $AE::VERSION.
438 - removed traces of "no strict 'refs'".
439
4405.22 Sat Dec 5 03:51:13 CET 2009
441 - downgrade-or-fail in AnyEvent::Handle::push_write, to
442 diagnose encoding failures earlier and more succinctly.
443 (this works around bugs in perl, throwing away encoding info
444 when passing scalar data to extensions).
445 - add more examples to AnyEvent::Socket manpage.
446 - upgrade internal warning set to the same as common::sense 2.03.
447 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
448 (John Beppu).
449 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
450 - add google public dns servers to fallback server set and make sure
451 we load-balance properly between the three sets. also add all
452 fallback dns servers, not just a random one, to each dns config.
453
4545.21 Thu Nov 19 02:48:47 CET 2009
455 - fix a problem where socket constants were called with parameters
456 (spotted by David Friedland).
457 - fork_call never use'd POSIX (reported by Daisuke Maki).
458 - improve perl 5.6 compatibility further (but it still won't work
459 unless you rip out everything but the core).
460 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
461 uses 10MB of RAM, but doesn't open a console window. *sigh*.
462
4635.202 Wed Oct 14 22:35:44 CEST 2009
464 - AnyEvent::DNS would unexpectedly clobber $_ under windows
465 (analysed by Matthias Waldorf).
466 - AnyEvent::Handle::run_cmd can now pass the PID of the
467 newly-created process, which is much less useful than it might
468 sound (based on patch by Yann Kerherve).
469
4705.201 Tue Sep 29 12:09:25 CEST 2009
471 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
472 (reported by Torsten Foertsch).
473 - common::sense 2.0 could cause tcp_server to throw an exception
474 (analysed by elmex).
475
4765.2 Mon Sep 14 07:04:49 CEST 2009
477 - INCOMPATIBLE CHANGE: do no longer support register_read_type
478 and register_write_type in AnyEvent::Handle, instead support
479 package names (the facility was mostly abused).
480 - implement "packagename-as-read/write type" support in
481 AnyEvent::Handle.
482 - AnyEvent::Handle: new options "keepalive" and "oobinline".
483 - oobinline set by default to avoid security issues.
484 - the pure-perl event loop backend wrongly detected times() fallback
485 support (spotted by Pavel Boldin).
486
4875.12 Tue Sep 1 20:26:50 CEST 2009
488 - be more lenient when parsing resolv.conf files, as some people
489 use hashmarks as comment indicator inside directives (reported
490 by Michael S. Fischer).
491 - use same set of warnings as common::sense 2.0.
492 - fix a potential 32 bit overflow issue due to perl having problems
493 with large hex constants in the Perl backend.
494 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
495 by default.
496 - fix some prototypes.
497 - (optionally) require Async::Interrupt 1.02.
498 - replace opendns fallbacks by verizon/level3 public dns servers.
499
5005.112 Fri Aug 21 13:59:04 CEST 2009
501 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
502 using starttls with an illegal/uncreatable tls context.
503 - port to broken win32 perls.
504
5055.111 Sun Aug 16 18:44:36 CEST 2009
506 - perl errornously requires sockaddr_un structures returned by the
507 kernel to have a specific length. work around this bug by padding
508 sockaddr_un structures (found to fail on OS X, as perl only uses
509 the correct code for GNU/Linux, but likely to fail on other
510 OSes as well).
511
5125.11 Wed Aug 12 17:49:37 CEST 2009
513 - fixed a bug in run_cmd with input from filename.
514 - create a json coder object when none is given in write direction
515 to reduce dependency on JSON.
516
5175.1 Tue Aug 11 03:17:32 CEST 2009
518 - tcp_connect now reports the correct error in $!, not an unrelated
519 one (was broken since 4.91).
520 - AE::Handle did not properly clear rtimeout/wtimeout watchers
521 when retrying a tcp_connect.
522 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
523 - fixed wrong documentation in AE::Socket.
524
5255.01 Mon Aug 10 03:16:32 CEST 2009
526 - last release broke EV child watchers :/.
527
5285.0 Sun Aug 9 17:32:09 CEST 2009
529 - officially introduce and document the AE API.
530 - lowering the timeout at runtime did not have immediate effect
531 in AnyEvent::Handle.
532 - AE::Handle now has separate and independent read and write timeout
533 settings.
534 - probe for Guard module when AnyEvent::Util is loaded, not at
535 runtime (Event doesn't like eval "use" from callbacks, and it avoids
536 exporting the probe function).
537 - Impl::IOAsync's io watchers did not use the filehandle emulation
538 code to work around it's watcher limits.
539 - work around IO::Async's problems with overloaded objects.
540 - try to disable wearnings inside IO::Async, it's just too buggy.
541 - do no longer use Event by default, only EV or Perl, Event is too buggy.
542 - hackishly provide fast AE:: interface when EV or Perl is the backend.
543 - inherit VERSION from AnyEvent::VERSION in most modules.
544
5454.91 Thu Aug 6 15:42:45 CEST 2009
546 - AE::Handle::starttls could get out of sync when the read buffer
547 already contains some TLS handshake.
548 - AE::Handle did not properly free the TLS session, causing bigger
549 memory leaks in Net::SSLeay than expected (reported by toaster).
550 - AE::Socket::tcp_connect will now artificially delay invoking
551 the callback to avoid returning after invoking the callback.
552 - convert many internal watcher uses to AE API.
553
5544.9 Sat Aug 1 11:07:01 CEST 2009
555 - Glib silently fails when registering a timeout with
556 a negative value, so avoid that.
557 - call condvar callback immediately when it is set
558 after the condition is already signalled.
559 - check rbuf_max condition only after trying to consume
560 data.
561 - normalise signal numbers to names when using the ae's
562 signal handling, but do not document this (yet).
563 - pure perl signal emulation did not properly set nonblocking
564 mode on the signal pipe on win32 (this is unlikely to have caused
565 any issues).
566 - new module: AnyEvent::Debug.
567 - AnyEvent::Strict now checks that a signal was specified
568 by name (not name or number).
569 - reduce memory footprint in typical cases by ~50kb by
570 only compiling signal code when necessary.
571 - add AnyEvent::Handle->rbuf_max.
572 - grab the AE namespace for future new API, implement stubs for
573 said future/alternative API.
574 - new function: AnyEvent::Socket::format_hostport.
575
5764.881 Tue Jul 28 12:51:53 CEST 2009
577 - work around a bug in local in pre-5.10 perls, causing
578 AnyEvent::Handle to recurse when it should not
579 (analysed by elmex).
580
5814.88 Tue Jul 28 04:04:37 CEST 2009
582 - re-bless the handle into a dummy package after calling
583 AnyEvent::Handle::destroy, so the user does not need to check
584 for errors after every push_write etc.
585 - do not attempt to run t/02_signals.t on obviously broken
586 platforms.
587
5884.87 Sun Jul 26 02:06:16 CEST 2009
589 - do not attempt to linger when there is no longer a valid fh.
590 - enforce tls mode to be either accept or connect, do not simply
591 segfault in Net::SSLeay.
592 - AnyEvent::Handle can now call tcp_connect itself (new parameters
593 connect, on_prepare, on_connect and on_connect_error). Updated
594 tutorial accordingly.
595 - add AnyEvent::Impl::Irssi backend.
596
5974.86 Mon Jul 20 23:52:29 CEST 2009
598 - since the verbose warning is not enough, explicitly document
599 that versions before 1.33 of Net::SSLeay are not secure.
600 - work around signal handling races in Event and (...) Event::Lib.
601 - try to align signal-race timer to full-second boundaries.
602 - work around Tk not liking negative timeouts.
603 - don't complain of different grades of environmental unfriendlyness
604 in IO::Async.
605
6064.85 Sat Jul 18 06:16:14 CEST 2009
607 - nail the signal race problem in perl once and for all
608 (see $AnyEvent::MAX_SIGNAL_LATENCY).
609 - take advantage of Async::Interrupt if it is available.
610 - load Time::HiRes and Guard modules on demand only.
611 - add optional/recommended modules section to AnyEvent
612 documentation.
613 - reduce memory usage considerably (and reduce startup penalty)
614 by not using "strict", "warnings" and "overload" modules.
615 - work around buggy windows/openbsd perls and provide EBADMSG
616 and EPROTO ourselves when missing.
617 - improve perl 5.6 compatibility of the core event loop.
618 - made Net::SSLeay version 1.33 a soft requirement.
619
6204.83 Fri Jul 17 16:56:26 CEST 2009
621 - implement AnyEvent::Socket::getprotobyname.
622 - AnyEvent::CondVar's will now detect recursive blocking
623 waits and will croak, as too many people fall into
624 this trap.
625 - AnyEvent::Handle will now call ->destroy on itself after
626 executing the on_error callback, instead of doing some
627 half-baked internal shutdown, for fatal errors.
628 - clarify on_eof behaviour w.r.t. the read queue and
629 on_read callbacks.
630 - ignore some possible spurious wake-ups in tcp_connect.
631
6324.82 Sat Jul 11 00:34:55 CEST 2009
633 - POE and Event backends didn't accept some callable objects as
634 callbacks.
635 - use Config module instead of POSIX module to detect signal names
636 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
637 module doesn't even have all POSIX signals :/.
638 - use more workarounds around the many refcnt/corruption bugs in
639 Event::Lib.
640 - work around a race condition in perl's select, causing t/03_child.t
641 to rarely fail.
642
6434.81 Thu Jul 9 10:30:30 CEST 2009
644 - AnyEvent::Handle didn't properly diagnose write errors
645 (it expected -1 from syswrite, how lame... :).
646 - support file descriptors in addition to file handles
647 in AnyEvent->io.
648 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
649 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
650 and PERL_ANYEVENT_CA_PATH.
651 - provide a sensible synopsis section for AnyEvent::TLS.
652 - add a "supported backends" section to the manpage.
653 - added simple io watcher test to testsuite, using a
654 portable_socketpair.
655 - tried to improve the stability of the Event::Lib backend,
656 YMMV.
657
6584.8 Mon Jul 6 23:45:16 CEST 2009
659 - AnyEvent::DNS did not properly follow CNAME records with
660 uppercase targets.
661 - AnyEvent::DNS would errornously return AAAA records
662 with v4 mapped addresses (a faulty record) as ipv4 addresses,
663 causing AnyEvent::Socket to throw an exception.
664 - added new module AnyEvent::TLS for easier SSL/TLS context
665 creation, with many options including hostname verification,
666 secure default configuration, lots of documentation and,
667 predefined diffie-hellman keys for perfect forward security
668 and much more. get it while it's still fresh!
669 - use AnyEvent::TLS in AnyEvent::Handle for context management.
3 - load AnyEvent::Handle only on demand in AnyEvent::DNS, 670 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
4 so AnyEvent::Socket users have smaller memory footprint 671 so AnyEvent::Socket users have smaller memory footprint
5 in the common case. 672 in the common case.
6 - add AnyEvent::Handle->push_shutdown method. 673 - add AnyEvent::Handle->push_shutdown method.
7 - add an additional error message parameter to AnyEvent::Handle's 674 - add an additional error message parameter to AnyEvent::Handle's
8 on_error callback (for TLS, $! is still available). 675 on_error callback (for TLS, $! is still available).
676 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
9 - make AnyEvent::Handle more robust against early conenction 677 - make AnyEvent::Handle more robust against early connection
10 failures (during new), and return C<undef> in that case 678 failures (during new), and return C<undef> in that case
11 from the constructor. 679 from the constructor.
680 - AnyEvent::Handle will now try to load only JSON::XS first,
681 then fall back to JSON.
682 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
683 and because it was documented that way.
12 684
134.451 Fri Jul 3 00:28:58 CEST 2009 6854.451 Fri Jul 3 00:28:58 CEST 2009
14 - do not clear rbuf when shutting down an AnyEvent::Handle 686 - do not clear rbuf when shutting down an AnyEvent::Handle
15 object - doing so breaks AnyEvent::HTTP. 687 object - doing so breaks AnyEvent::HTTP.
16 688
… …
1344.3 Fri Oct 3 09:18:43 CEST 2008 8064.3 Fri Oct 3 09:18:43 CEST 2008
135 - AnyEvent will now install a no-op signal handler for SIGPIPE, 807 - AnyEvent will now install a no-op signal handler for SIGPIPE,
136 unless one has been installed already. 808 unless one has been installed already.
137 - warn about not loaidng AnyEvent::Impl::POE early enough 809 - warn about not loaidng AnyEvent::Impl::POE early enough
138 (patch by Adam Rosenstein). 810 (patch by Adam Rosenstein).
139 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 811 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
140 - fix a bug in where in SSL connect mode, the 812 - fix a bug in where in SSL connect mode, the
141 client would first wait for some data by the server and 813 client would first wait for some data by the server and
142 otherwise hang. (reported and analyzed in an absolutely 814 otherwise hang. (reported and analysed in an absolutely
143 exemplary manner by Adam Rosenstein). 815 exemplary manner by Adam Rosenstein).
144 - fix a bug in where SSL EOF would not be treated 816 - fix a bug in where SSL EOF would not be treated
145 as stream EOF, putting the connection into a hung state. 817 as stream EOF, putting the connection into a hung state.
146 (reported and analyzed in an absolutely exemplary manner by Adam 818 (reported and analysed in an absolutely exemplary manner by Adam
147 Rosenstein). 819 Rosenstein).
148 - fix a potential segfault when the TLS context 820 - fix a potential segfault when the TLS context
149 would go missing in a read callback (e.g. due to stoptls) - 821 would go missing in a read callback (e.g. due to stoptls) -
150 Net::SSLeay of course makes no type checking whatsoever. 822 Net::SSLeay of course makes no type checking whatsoever.
151 - AnyEvent::Handle will not stop the read watcher in TLS mode. 823 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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