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

Comparing AnyEvent/Changes (file contents):
Revision 1.338 by root, Thu Aug 20 22:58:35 2009 UTC vs.
Revision 1.614 by root, Sat Sep 17 02:33:54 2016 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: more formats for parse_ipv6 (single ipv6 address without port, ...p80, ...#80)
12
13TODO: inet_aton maybe pack C4? no, add a comment why not
14TODO: hosts read multiple times for initial concurrent tcp_connect's?
15
167.13 Sat Sep 17 04:31:49 CEST 2016
17 - Only call tlsext_host_name for non-empty common names (reported
18 by Maxime Soulé).
19 - log a (single) notice message if SNI is not supported.
20 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
21 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
22 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
23 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
24 testcases and the specification and not in the code - the post-9.0.0
25 unicode files have all known problems fixed, so finally the AnyEvent
26 IDNA implementation can pass the full IDNA testsuite - without needing
27 a single fix :)
28 - guarantee (and document) that condvar callbacks will be removed
29 on invocation - important to avoid circular references.
30
317.12 Wed Jan 27 19:12:26 CET 2016
32 - use common name as hostname for tls connects, if Net::SSLeay
33 supports SNI.
34 - fix documentation of tls_autostart read type in AnyEvent::Handle,
35 analyzed by Felix Ostmann.
36
377.11 Thu Jul 16 14:36:00 CEST 2015
38 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
39 addresses (extra "::" at end and similar cases).
40 - add a more explicit warning to AnyEvent::Handle that it doesn't
41 work on files, people keep getting confused.
42 - new function AnyEvent::Socket::tcp_bind.
43 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
44 - aligned ipv6 address formatting with RFC 5952 (by not shortening
45 a single :0: to ::).
46 - added stability canary support.
47
487.09 Sat May 2 16:38:53 CEST 2015
49 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
50 that was renamed to AnyEvent::Log:format_time. uinder its old name
51 (analyzed by Michael Stovenour).
52 - update AnyEvent::DNS fallback resolver addresses:
53 seems google effectively killed most other free dns resolvers,
54 so remove them, but add cable and wireless (ecrc) since it was
55 stable for 20 years or so, official or not, and there should be
56 an alternative to google.
57 - perl5porters broke windows error codes in 5.20, and mapped
58 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
59 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
60 because they were so cool. They probably broke other error codes
61 for no reason, too, but I didn't care for checking, it's too
62 depressing. This version only works around the WSAEWOULDBLOCK
63 issue, because I don't have a nice way to work around the
64 WSAEINPROGRESS bug.
65
667.08 Wed Dec 10 05:27:17 CET 2014
67 - work around a newly introduced bug in Socket 2.011 (an
68 errornous sun_length check) (analyzed by Maxime Soulé).
69 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
70 (analyzed by Ben Magistro).
71 - AnyEvent::Strict will now confess, not croak. This is in line with
72 it being a development/debugging tool.
73 - work around a number of libglib bugs (debug builds of libglib enforce
74 certain undocumented behaviour patterns such as not being able to
75 remove a child watch source after it has fired, which we will try
76 to emulate to avoid "criticals". what where they thinking...).
77 - mention json security issues in AnyEvent::Handle, now that Douglas
78 Crockford has foolishly and incompatibly changed JSON.
79 - changed default dns resolver "max_outstanding" value from 1 to 10,
80 the latter beinfg the intended value all along
81 (reported by Ilya Chesnokov).
82 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
83 (written by Mike Lowell).
84
857.07 Tue Dec 17 17:45:02 CET 2013
86 - the documentation for custom tls verify schemes was wrong. make it agree
87 with the code (reported by Maxime Soulé).
88 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
89 - work around an API change in openssl that could cause wrong tls connection
90 aborts, likely on windows only (analyzed by sten).
91 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
92 endless loop (reported by Dietrich Rebmann).
93 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
94 (patch by Maxime Soulé).
95 - document AnyEvent::Impl::IOAsync::set_loop and
96 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
97 functionality has _always_ been available.
98 - force a toplevel domain name in t/81_hosts.t (analyzed by
99 David Jack Wange Olrik).
100 - document that AnyEvent::Log uses AnyEvent::IO.
101 - warn about AnyEvent::Filesys::Notify performance.
102 - praise the joys of AnyEvent::Fork::*.
103 - time for an =encoding directive.
104 - do no longer use JSON to create a default json coder, use
105 JSON::XS or JSON::PP directly.
106
1077.05 Wed Aug 21 10:38:08 CEST 2013
108 - uts46data.pl couldn't be found due to wrong naming of the file
109 (reported by Fulko Hew).
110 - handle lone \015's properly in AE::Handle's default line read
111 (reported by various people).
112 - untaint ip addresses found in /etc/hosts (patch by José Micó).
113 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
114 fail on first use (reported and testcase by Andrew Whatson).
115 - expose AnyEvent::Log::format_time, and allow users to redefine it.
116 - expose AnyEvent::Log::default_format, and allow redefinition.
117 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
118 - AnyEvent::Debug shell can now run coro shell commands, if available.
119 - t/63* tests were wrongly in MANIFEST.
120 - kernel.org's finger server went MIA, switch to freebsd.org and
121 icculus.org.
122 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
123 to function asynchronously (reported by Daniel Carosone).
124 - hard-disable $^W in most tests, it generates too much garbage output.
125 - use a (hopefully) more future-proof method to emulate common::sense.
126 - upgrade to UTS-46:6.2.0.
127 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
128 documented. should not affect anything.
129
1307.04 Thu Nov 15 02:23:58 CET 2012
131 - debugging code left in.
132
1337.03 Thu Nov 15 02:16:11 CET 2012
134 - AnyEvent::Socket::inet_aton did not work when dns resolution was
135 used to find the addresses (analyzed and patch by Maxime Soulé).
136 - fix a memory leak in the /etc/hosts lookup code when hosts don't
137 resolve and are not in hosts (analyzed by sten).
138
1397.02 Tue Aug 14 04:27:58 CEST 2012
140 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
141 program by Yorhel).
142 - verified that AnyEvent::Socket follows RFC5952.
143 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
144
1457.01 Sun May 13 01:03:17 CEST 2012
146 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
147 not available, instead of throwing an exception.
148 - use File::Spec to get the tmpdir in t/*, to avoid needless
149 failures on (most, not mine :) windows boxes.
150 - new handle read types: tls_detect and tls_autostart.
151
1527.0 Fri Apr 13 06:33:30 CEST 2012
153 - child watchers are broken in POE 1.352 (also many earlier
154 versions) and there seems to be no way to work around it, as POE
155 itself is inherently racy. Document this regression and add a delay
156 in t/68_poe_03_child.t for the time being.
157 - new module AnyEvent::IO, that is a frontend to either a pure-perl
158 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
159 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
160 - load /etc/hosts only when DNS has no answer.
161 - stat /etc/hosts on every access and reload it if it changed.
162 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
163 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
164 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
165 a side effect, they now use true appending as opposed to libc
166 appending, and the file might not have been opened when the function
167 returns.
168 - the default logging level was not properly documented in a variety of
169 places, this has hopefully been rectified.
170 - updated uts46data.pl for unicode 6.1.0.
171 - made log messages generated by AnyEvent submodules not
172 include the package name anymore, as it will be logged
173 by default already.
174 - upgrade to the trick used by common-sense 3.5 to work
175 around extra warning torture/breakage under perl 5.15.x.
176 - log messages by anyevent are now ucfirst, are usually full
177 sentences and do no longer include the package name.
178 - the storable read type would sometimes throw an exception instead
179 of causing EBADMSG (analyzed by Maxime Soulé).
180 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
181 keysize (reported by Paul Howarth).
182 - documented Rocco fabricating statements about AnyEvent and me.
183
1846.14 Tue Jan 31 20:00:24 CET 2012
185 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
186 probably an editing mistake (reported by Darin McBride).
187
1886.13 Thu Jan 12 07:27:01 CET 2012
189 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
190 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
191 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
192 regex-matching inside the callback would change the parameter, despite
193 $1 being dynamically scoped per-block. this perl bug is now being
194 worked around (testcase by Cindy Wang).
195
1966.12 Mon Dec 12 13:21:10 CET 2011
197 - $! was clobbered by subsequent calls in tcp_connect, due to
198 the postpone not saving/restoring it (analyzed by Richard Garnish).
199
2006.11 Tue Nov 22 10:36:05 CET 2011
201 - Tk cannot create windows when tainted unless you set a title, so
202 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
203 - escape any nonprintable/nonascii characters when stringifying
204 backtraces.
205 - log the reason loading an interface module fails to load at level debug.
206
2076.1 Tue Oct 4 19:44:30 CEST 2011
208 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
209 and above), and some messages inside AnyEvent have been elevated
210 to higher log levels to print by default.
211 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
212 - work around yet another signal race bug, newly introduced by POE. My
213 how that module sucks.
214 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
215 support ipv6-only operation even with defaults (also use ::1 instead
216 of 127.0.0.1).
217 - give better guidelines on how to chose a logging level.
218 - AnyEvent::Log can now cap (limit) log message priority by context.
219 - AE::log will now save and restore $! and $@, so callers don't have to.
220 - AE::Strict now checks that registered fds are still valid at regular
221 intervals.
222 - skip frequently called 'require' calls if the module appears loaded.
223 - increase signal latency test timer considerably for poor little osx,
224 which can't be bothered with basic posix functionality such as signal
225 delivery.
226 - mention some debugging aids in the tutorial.
227 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
228 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
229 of the context as title.
230 - AnyEvent::Strict would not restore read-write access to $_ when a callback
231 died - this is a user bug, but since AE::Strict exists to help dbeugging...
232 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
233 to work with it - the author is well aware of the issues but refuses
234 to fix them. This extreme step has been taken because the module actively
235 attacks the goals of AnyEvent and makes it harder for module authors to use
236 AnyEvent.
237
2386.02 Fri Aug 26 20:08:31 CEST 2011
239 - log_to_syslog now accepts facility strings.
240 - log_to_syslog would not allow facilities - "$facility|$level"
241 *literally* means that string. sheesh.
242 - the CPAN parser got confused and thought we provide Sys::Syslog, try
243 to work around.
244 - renamed AnyEvent::Impl::FLTK2 to FLTK.
245
2466.01 Fri Aug 26 07:04:11 CEST 2011
247 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
248 the TTL as fourth element - this affects ->resolve, ->request
249 and ->dns_unpack indirectly. Convenience functions such as
250 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
251 AnyEvent::Socket.
252 - INCOMPATIBLE CHANGE: backend authors now should not implement
253 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
254 methods.
255 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
256 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
257 - DEPRECIATION: please change
258 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
259 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
260 in your programs, for future compatibility.
261 - *blush* idle watcher emulation was borked.
262 - AnyEvent::Strict now checks the AE::xxx functions as well.
263 - querying too long domain names would barf deep inside AnyEvent::DNS - since
264 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
265 for too long domain names explicitly and passes an undef to the callback
266 instead (reported by James Bromberger).
267 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
268 - new functions: AnyEvent::postpone, AnyEvent::log.
269 - new module: AnyEvent::Log, for simple logging needs.
270 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
271 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
272 a debug shell.
273 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
274 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
275 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
276 /etc/hosts can be found.
277 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
278 configuration.
279 - env variables can now be specified using AE_xyz instead of
280 PERL_ANYEVENT_xyz, subject to some rules.
281 - lots of new AnyEvent::Debug functionality.
282 - AnyEvent::Strict now checks for modifications of $_ by the callback
283 (by setting it to readonly while executing the callback, if possible).
284 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
285 Invest extra overhead to work around this peculiar behaviour.
286 - split the pure perl backend into a loop module and an impl
287 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
288 - some read types (regex, netstring, json, storable) could
289 errornously keep an AE::Handle object alive on their own
290 (reported by Mohammad Toossi).
291 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
292 supports /etc/hosts overrides.
293 - add more workarounds around child watcher bugs in IO::Async
294 and (newly introduced) in POE.
295 - use glib child watchers and try to work around its limitations,
296 if possible (based on a patch by Kevin Ryde).
297 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
298 - many read types in AnyEvent::Handle would malfunction during
299 transitions to TLS because of a "delete $self->{rbuf}".
300 (reported by Gerald Galster).
301 - improve condvar blocking wait performance (probably only noticable
302 with faster backends), also make it easier for event loops that
303 do not support blocking waits.
304 - PERL_ANYEVENT_MODEL now also allows full module names, although
305 the usefulness of this feature is not yet clear.
306 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
307 - add t/80_ssltext.t.
308 - no longer use AUTOLOAD for the initial loading (for no very good reason).
309 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
310 - do not run event loop detection multiple times when programs call their cached
311 copies of AnyEvent::detect.
312 - parse_hostport now accepts absolute paths als unix domain sockets.
313 - suppress the idiotic warning inside POE by patching the POE::Kernel
314 run flags instead of asking the user to jump through contortions.
315 - add Task::Weaken as dependency (as recommends only at this time).
316 - replace PF_UNSPEC by 0, for good old minix.
317
3186.0 (not officially released, but was on CPAN for about 15 minutes :/)
319
3205.34 Thu May 12 09:59:41 CEST 2011
321 - AE::TLS didn't support both cert_file and key at the same time,
322 due to some copy&paste bug (reported and analyzed by vti).
323 - improve AE::Handle timeout documentation slightly.
324
3255.33 Mon Feb 7 21:51:41 CET 2011
326 - never appeared on CPAN, for unknown reasons.
327 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
328 object when it's already gone (analyzed by Ryan Bullock).
329 - update idna algorithm(s) to UTS#46 version 6.0.0.
330 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
331 uts46 says it's an error, and testsuite makes more sense.
332 - remove leading dots in idn_to_ascii - testsuite says to remove,
333 uts46 says it's an error, and testsuite makes more sense.
334 - fixed a bug in idn ignored character processing.
335 - passes the uts46 6.0.0 testsuite except for three cases which
336 apparently are bugs in the testsuite itself.
337 - passes the uts46 6.0.1 testsuite except for eleven cases which
338 apparently are bugs in the testsuite itself.
339
3405.31 Sun Jan 23 11:49:19 CET 2011
341 - work around a POE bug where POE would first poll for new events and
342 invoke handlers BEFORE handling the message queue.
343 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
344 - stop_read now does just that in TLS mode. see its
345 documentation for caveats.
346 - update to common::sense 3.4 warning set (some warnings in perl impose
347 an arbitrary 32kb limit on regex matches, which affects
348 AnyEvent::Handle).
349 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
350 - new experimental AE::Handle::resettls method.
351
3525.3 Fri Dec 31 05:48:13 CET 2010
353 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
354 issues with IO::Async have been solved in that version, including
355 the availabilty of a default loop!
356 - implement new wbuf_max parameter for AnyEvent::Handle.
357 - added Cocoa::EventLoop backend by Daisuke Murase.
358 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
359 - clarify handle on_prepare documentation.
360 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
361 to avoid leaking them to any child processes.
362 - try to force files into the arch-dependend lib directory, for the
363 half-"turly obsessive compulsive person"(s) that depend on it.
364
3655.29 Sun Dec 5 10:49:21 CET 2010
366 - convert EV backend to EV 4.00 API (so better upgrade EV too).
367 - AE::Handle: implement dynamic read_size adjustment, add
368 max_read_size parameter, reduce default read_size to 2048.
369 - add some FAQ entries for servers and/or on_eof vs. on_error.
370 - work around OS bugs (cygwin again) when lingering in AE::Handle.
371 - reduce memory usage slightly when connect was used in AE::Handle.
372 - wrok around more segfaults in Tk.
373 - document tls_ctx => undef as valid.
374 - detect WNOHANG value at compiletime.
375 - include some probably linux-only support for building AnyEvent
376 as part of the perl core.
377 - improve compatibility of t/08_idna.t to perl 5.8.
378 - make AnyEvent::Strict truly optional (in case somebody builds a
379 minimal perl...).
380
3815.28 Wed Oct 13 04:14:23 CEST 2010
382 - due to a glitch, AnyEvent's internal getprotobyname wasn't
383 actually used in tcp_server.
384 - implement AnyEvent::CondVar->new method, which was documented but
385 missing.
386 - support autoloading of read types in unshift_read.
387 - AnyEvent::Strict: do the test for invalid fh arguments differently,
388 resulting in more informative error messages.
389 - new AnyEvent::FAQ manpage.
390 - clarify that you can actually call push_read etc. in on_read.
391 - put keys and certs in /tmp, not into the current directory,
392 and document this in the SECURITY CONSIDERATIONS sections
393 of AnyEvent::TLS.
394 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
395 requested by Ron Isaacson.
396 - document $AnyEvent::DNS::RESOLVER variable.
397 - incorporated major typo patches by Abhijit Menon-Sen.
398 - AnyEvent::Handle now croaks when negative timeouts are passed.
399 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
400 undocumented, but who knows when they might come in handy.
401
4025.271 Tue Jun 8 12:05:46 CEST 2010
403 - backport to perl 5.8.x.
404
4055.27 Sun Jun 6 12:12:05 CEST 2010
406 - postpone differently in AnyEvent::Socket now, as
407 when not, canceling the connection attempt might fail
408 (found by Felix Antonius Wilhelm Ostmann).
409 - explicitly check for non-stream sockets in AE::Handle, too many
410 clueless people fell into the trap of this somehow working.
411 - simplified and reworked the "OTHER MODULES" section.
412 - better/more condvar examples.
413
4145.261 Wed Apr 28 16:13:36 CEST 2010
415 - AF_INET6 was not properly used from Socket6 during configuration
416 time (found by Andy Grundman).
417
4185.26 Mon Apr 12 04:49:35 CEST 2010
419 - don't generate spurious readyness notifications when select
420 returns EINTR in the pure perl backend (analysed and testcase
421 by Bas Denissen).
422 - give same examples in SYNOPSIS sections of AnyEvent and AE.
423 - provider faster implementations for AE::time/now/now_update for
424 EV and Perl backends.
425 - provide faster AE::cv and AE::time implementations for all
426 backends.
427 - no longer support lower/mixed-case signal names (this was
428 never documented nor universally supported).
429 - some more memory µ-optimisations, and somewhat less messy
430 code to cope with both AnyEvent and AE APIs.
431
4325.251 Sat Mar 13 00:58:21 CET 2010
433 - make test failed if it was run before make install on perl <=
434 5.8.8, and older versions of perl make our live very difficult,
435 so write constants.pl during Makefile.PL time.
436
4375.25 Sat Mar 13 00:23:14 CET 2010
438 - fix a race condition in AnyEvent::Handle that would cause
439 a "bio_read: unsupported method" error (in _another_
440 ssl connection) after user code threw an exception.
441 - added AnyEvent::Handle->destroyed method.
442 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
443 - the AnyEvent::Util::fh_nonblocking function worked only
444 by lucky accident on win32.
445 - smaller and faster AnyEvent::Util::fh_nonblocking.
446 - when the (required!) Time::HiRes module is missing, AnyEvent
447 did not fall back to built-in time properly.
448 - do not load Fcntl at runtime, saving memory and loading time.
449 - precompile a number of constants and use them instead of runtime
450 detection and eval.
451 - free detection code after detection and similar memory optimisations.
452 - Perl backend timer interval best effort drift has been improved
453 (same algorithm as EV).
454 - update unicode idna mapping table.
455
4565.24 Tue Jan 5 11:39:43 CET 2010
457 - cygwin never reports errors from failed connects. 1.5 just gives
458 you continous readyness and EAGAIN, 1.7 is even more broken and
459 just hangs. work around both issues in a cygwin-specific hack.
460 - improve idle watcher documentation slightly.
461
4625.23 Sun Dec 20 23:48:00 CET 2009
463 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
464 - implement punycode_encode/decode, idn_nameprep,
465 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
466 - provide $AE::VERSION.
467 - removed traces of "no strict 'refs'".
468
4695.22 Sat Dec 5 03:51:13 CET 2009
470 - downgrade-or-fail in AnyEvent::Handle::push_write, to
471 diagnose encoding failures earlier and more succinctly.
472 (this works around bugs in perl, throwing away encoding info
473 when passing scalar data to extensions).
474 - add more examples to AnyEvent::Socket manpage.
475 - upgrade internal warning set to the same as common::sense 2.03.
476 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
477 (John Beppu).
478 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
479 - add google public dns servers to fallback server set and make sure
480 we load-balance properly between the three sets. also add all
481 fallback dns servers, not just a random one, to each dns config.
482
4835.21 Thu Nov 19 02:48:47 CET 2009
484 - fix a problem where socket constants were called with parameters
485 (spotted by David Friedland).
486 - fork_call never use'd POSIX (reported by Daisuke Maki).
487 - improve perl 5.6 compatibility further (but it still won't work
488 unless you rip out everything but the core).
489 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
490 uses 10MB of RAM, but doesn't open a console window. *sigh*.
491
4925.202 Wed Oct 14 22:35:44 CEST 2009
493 - AnyEvent::DNS would unexpectedly clobber $_ under windows
494 (analysed by Matthias Waldorf).
495 - AnyEvent::Handle::run_cmd can now pass the PID of the
496 newly-created process, which is much less useful than it might
497 sound (based on patch by Yann Kerherve).
498
4995.201 Tue Sep 29 12:09:25 CEST 2009
500 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
501 (reported by Torsten Foertsch).
502 - common::sense 2.0 could cause tcp_server to throw an exception
503 (analysed by elmex).
504
5055.2 Mon Sep 14 07:04:49 CEST 2009
506 - INCOMPATIBLE CHANGE: do no longer support register_read_type
507 and register_write_type in AnyEvent::Handle, instead support
508 package names (the facility was mostly abused).
509 - implement "packagename-as-read/write type" support in
510 AnyEvent::Handle.
511 - AnyEvent::Handle: new options "keepalive" and "oobinline".
512 - oobinline set by default to avoid security issues.
513 - the pure-perl event loop backend wrongly detected times() fallback
514 support (spotted by Pavel Boldin).
515
5165.12 Tue Sep 1 20:26:50 CEST 2009
517 - be more lenient when parsing resolv.conf files, as some people
518 use hashmarks as comment indicator inside directives (reported
519 by Michael S. Fischer).
520 - use same set of warnings as common::sense 2.0.
521 - fix a potential 32 bit overflow issue due to perl having problems
522 with large hex constants in the Perl backend.
523 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
524 by default.
525 - fix some prototypes.
526 - (optionally) require Async::Interrupt 1.02.
527 - replace opendns fallbacks by verizon/level3 public dns servers.
528
5295.112 Fri Aug 21 13:59:04 CEST 2009
3 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when 530 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
4 using starttls with an illegal/uncreatable tls context. 531 using starttls with an illegal/uncreatable tls context.
532 - port to broken win32 perls.
5 533
65.111 Sun Aug 16 18:44:36 CEST 2009 5345.111 Sun Aug 16 18:44:36 CEST 2009
7 - perl errornously requires sockaddr_un structures returned by the 535 - perl errornously requires sockaddr_un structures returned by the
8 kernel to have a specific length. work around this bug by padding 536 kernel to have a specific length. work around this bug by padding
9 sockaddr_un structures (found to fail on OS X, as perl only uses 537 sockaddr_un structures (found to fail on OS X, as perl only uses
… …
75 - new function: AnyEvent::Socket::format_hostport. 603 - new function: AnyEvent::Socket::format_hostport.
76 604
774.881 Tue Jul 28 12:51:53 CEST 2009 6054.881 Tue Jul 28 12:51:53 CEST 2009
78 - work around a bug in local in pre-5.10 perls, causing 606 - work around a bug in local in pre-5.10 perls, causing
79 AnyEvent::Handle to recurse when it should not 607 AnyEvent::Handle to recurse when it should not
80 (analyzed by elmex). 608 (analysed by elmex).
81 609
824.88 Tue Jul 28 04:04:37 CEST 2009 6104.88 Tue Jul 28 04:04:37 CEST 2009
83 - re-bless the handle into a dummy package after calling 611 - re-bless the handle into a dummy package after calling
84 AnyEvent::Handle::destroy, so the user does not need to check 612 AnyEvent::Handle::destroy, so the user does not need to check
85 for errors after every push_write etc. 613 for errors after every push_write etc.
… …
173 in the common case. 701 in the common case.
174 - add AnyEvent::Handle->push_shutdown method. 702 - add AnyEvent::Handle->push_shutdown method.
175 - add an additional error message parameter to AnyEvent::Handle's 703 - add an additional error message parameter to AnyEvent::Handle's
176 on_error callback (for TLS, $! is still available). 704 on_error callback (for TLS, $! is still available).
177 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 705 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
178 - make AnyEvent::Handle more robust against early conenction 706 - make AnyEvent::Handle more robust against early connection
179 failures (during new), and return C<undef> in that case 707 failures (during new), and return C<undef> in that case
180 from the constructor. 708 from the constructor.
181 - AnyEvent::Handle will now try to load only JSON::XS first, 709 - AnyEvent::Handle will now try to load only JSON::XS first,
182 then fall back to JSON. 710 then fall back to JSON.
183 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 711 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
3074.3 Fri Oct 3 09:18:43 CEST 2008 8354.3 Fri Oct 3 09:18:43 CEST 2008
308 - AnyEvent will now install a no-op signal handler for SIGPIPE, 836 - AnyEvent will now install a no-op signal handler for SIGPIPE,
309 unless one has been installed already. 837 unless one has been installed already.
310 - warn about not loaidng AnyEvent::Impl::POE early enough 838 - warn about not loaidng AnyEvent::Impl::POE early enough
311 (patch by Adam Rosenstein). 839 (patch by Adam Rosenstein).
312 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 840 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
313 - fix a bug in where in SSL connect mode, the 841 - fix a bug in where in SSL connect mode, the
314 client would first wait for some data by the server and 842 client would first wait for some data by the server and
315 otherwise hang. (reported and analyzed in an absolutely 843 otherwise hang. (reported and analysed in an absolutely
316 exemplary manner by Adam Rosenstein). 844 exemplary manner by Adam Rosenstein).
317 - fix a bug in where SSL EOF would not be treated 845 - fix a bug in where SSL EOF would not be treated
318 as stream EOF, putting the connection into a hung state. 846 as stream EOF, putting the connection into a hung state.
319 (reported and analyzed in an absolutely exemplary manner by Adam 847 (reported and analysed in an absolutely exemplary manner by Adam
320 Rosenstein). 848 Rosenstein).
321 - fix a potential segfault when the TLS context 849 - fix a potential segfault when the TLS context
322 would go missing in a read callback (e.g. due to stoptls) - 850 would go missing in a read callback (e.g. due to stoptls) -
323 Net::SSLeay of course makes no type checking whatsoever. 851 Net::SSLeay of course makes no type checking whatsoever.
324 - AnyEvent::Handle will not stop the read watcher in TLS mode. 852 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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