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

Comparing AnyEvent/Changes (file contents):
Revision 1.320 by root, Thu Aug 6 13:31:01 2009 UTC vs.
Revision 1.637 by root, Fri Jul 19 16:03:39 2019 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: Danga::Socket backend. 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)
12TODO: inet_aton maybe pack C4? no, add a comment why not
13TODO: socket prepare callbacks, can they easily signal failure as well?
4 14
5TODO: delay tcp_connect 15TODO: Олег Г <verdrehung@gmail.com>, empty A record response causes
64.901 16AnyEvent::resolve_sockaddr to eventually call pack_sockaddr with undef -
17broken dns server, but maybe improve the response.
18
197.16 Fri Jul 19 18:00:21 CEST 2019
20 - add ffdhe group dh parameters from RFC 7919, and use ffdhe3072
21 as new default, instead of schmorp1539.
22 - AnyEvent::Log did not reassess logging status of
23 AnyEvent::Log::logger's when contexts were changed with
24 ->attach/detach/slaves, causing them to not log even though a
25 recent attach should have caused them to log.
26 - added some more logging configuration examples.
27 - mention RFC 8482 in AnyEvent::DNS.
28
297.15 Tue Feb 26 03:07:42 CET 2019
30 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
31 separate major and minor versions, while code passed only a single
32 value. This version follows the documentation and now passes separate
33 major and minor values.
34 - work around Net::SSLeay not having been ported to openssl 1.1, but many
35 distributions compiling it against openssl 1.1, which unfortunately
36 succeeds and results in a very broken module.
37 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
38 in the __ member, to allow decoding of undecodable resouce records
39 containing compressed domian names.
40 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
41 addresses. It now does and converts them to ipv4 mapped addresses.
42 - support CAA records, based on patch by Steve Atkins.
43 - add freenom and cloudflare nameservers as dns fallback.
44 - AnyEvent::Strict would not properly ward against io watchers
45 on files when the handle passed was a file descriptor.
46 - document "internal" variables used by the dns en-/decoder to allow
47 enterprising users to extend them in a semi-official way.
48
497.14 Sat Jun 24 01:44:19 CEST 2017
50 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
51 by Paul Howarth and apparently tracked down by Mike McCauley).
52 - AnyEvent::Handle->keepalive was documented (and defined) twice
53 (found by Matt Merhar).
54 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
55 listening socket unless a guard is used. Change this so that no
56 cleanup will be performed unless a guard is used and document this more
57 clearly.
58 - make tcp_bind/tcp_server error messages more regular.
59 - work around latest perl upstream breakage - this time, perl5porters were
60 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
61 stupidly apply anything that has "security fix" stamped on it, it's likely
62 going to be a problem in practise (and working around it is trivial on
63 windows and unix).
64 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
65 to constants.
66 - update warnings to common::sense 3.74 standards.
67
687.13 Sat Sep 17 04:31:49 CEST 2016
69 - Only call tlsext_host_name for non-empty common names (reported
70 by Maxime Soulé).
71 - log a (single) notice message if SNI is not supported.
72 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
73 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
74 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
75 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
76 testcases and the specification and not in the code - the post-9.0.0
77 unicode files have all known problems fixed, so finally the AnyEvent
78 IDNA implementation can pass the full IDNA testsuite - without needing
79 a single fix :)
80 - guarantee (and document) that condvar callbacks will be removed
81 on invocation - important to avoid circular references.
82
837.12 Wed Jan 27 19:12:26 CET 2016
84 - use common name as hostname for tls connects, if Net::SSLeay
85 supports SNI.
86 - fix documentation of tls_autostart read type in AnyEvent::Handle,
87 analyzed by Felix Ostmann.
88
897.11 Thu Jul 16 14:36:00 CEST 2015
90 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
91 addresses (extra "::" at end and similar cases).
92 - add a more explicit warning to AnyEvent::Handle that it doesn't
93 work on files, people keep getting confused.
94 - new function AnyEvent::Socket::tcp_bind.
95 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
96 - aligned ipv6 address formatting with RFC 5952 (by not shortening
97 a single :0: to ::).
98 - added stability canary support.
99
1007.09 Sat May 2 16:38:53 CEST 2015
101 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
102 that was renamed to AnyEvent::Log:format_time. uinder its old name
103 (analyzed by Michael Stovenour).
104 - update AnyEvent::DNS fallback resolver addresses:
105 seems google effectively killed most other free dns resolvers,
106 so remove them, but add cable and wireless (ecrc) since it was
107 stable for 20 years or so, official or not, and there should be
108 an alternative to google.
109 - perl5porters broke windows error codes in 5.20, and mapped
110 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
111 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
112 because they were so cool. They probably broke other error codes
113 for no reason, too, but I didn't care for checking, it's too
114 depressing. This version only works around the WSAEWOULDBLOCK
115 issue, because I don't have a nice way to work around the
116 WSAEINPROGRESS bug.
117
1187.08 Wed Dec 10 05:27:17 CET 2014
119 - work around a newly introduced bug in Socket 2.011 (an
120 errornous sun_length check) (analyzed by Maxime Soulé).
121 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
122 (analyzed by Ben Magistro).
123 - AnyEvent::Strict will now confess, not croak. This is in line with
124 it being a development/debugging tool.
125 - work around a number of libglib bugs (debug builds of libglib enforce
126 certain undocumented behaviour patterns such as not being able to
127 remove a child watch source after it has fired, which we will try
128 to emulate to avoid "criticals". what where they thinking...).
129 - mention json security issues in AnyEvent::Handle, now that Douglas
130 Crockford has foolishly and incompatibly changed JSON.
131 - changed default dns resolver "max_outstanding" value from 1 to 10,
132 the latter beinfg the intended value all along
133 (reported by Ilya Chesnokov).
134 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
135 (written by Mike Lowell).
136
1377.07 Tue Dec 17 17:45:02 CET 2013
138 - the documentation for custom tls verify schemes was wrong. make it agree
139 with the code (reported by Maxime Soulé).
140 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
141 - work around an API change in openssl that could cause wrong tls connection
142 aborts, likely on windows only (analyzed by sten).
143 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
144 endless loop (reported by Dietrich Rebmann).
145 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
146 (patch by Maxime Soulé).
147 - document AnyEvent::Impl::IOAsync::set_loop and
148 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
149 functionality has _always_ been available.
150 - force a toplevel domain name in t/81_hosts.t (analyzed by
151 David Jack Wange Olrik).
152 - document that AnyEvent::Log uses AnyEvent::IO.
153 - warn about AnyEvent::Filesys::Notify performance.
154 - praise the joys of AnyEvent::Fork::*.
155 - time for an =encoding directive.
156 - do no longer use JSON to create a default json coder, use
157 JSON::XS or JSON::PP directly.
158
1597.05 Wed Aug 21 10:38:08 CEST 2013
160 - uts46data.pl couldn't be found due to wrong naming of the file
161 (reported by Fulko Hew).
162 - handle lone \015's properly in AE::Handle's default line read
163 (reported by various people).
164 - untaint ip addresses found in /etc/hosts (patch by José Micó).
165 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
166 fail on first use (reported and testcase by Andrew Whatson).
167 - expose AnyEvent::Log::format_time, and allow users to redefine it.
168 - expose AnyEvent::Log::default_format, and allow redefinition.
169 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
170 - AnyEvent::Debug shell can now run coro shell commands, if available.
171 - t/63* tests were wrongly in MANIFEST.
172 - kernel.org's finger server went MIA, switch to freebsd.org and
173 icculus.org.
174 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
175 to function asynchronously (reported by Daniel Carosone).
176 - hard-disable $^W in most tests, it generates too much garbage output.
177 - use a (hopefully) more future-proof method to emulate common::sense.
178 - upgrade to UTS-46:6.2.0.
179 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
180 documented. should not affect anything.
181
1827.04 Thu Nov 15 02:23:58 CET 2012
183 - debugging code left in.
184
1857.03 Thu Nov 15 02:16:11 CET 2012
186 - AnyEvent::Socket::inet_aton did not work when dns resolution was
187 used to find the addresses (analyzed and patch by Maxime Soulé).
188 - fix a memory leak in the /etc/hosts lookup code when hosts don't
189 resolve and are not in hosts (analyzed by sten).
190
1917.02 Tue Aug 14 04:27:58 CEST 2012
192 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
193 program by Yorhel).
194 - verified that AnyEvent::Socket follows RFC5952.
195 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
196
1977.01 Sun May 13 01:03:17 CEST 2012
198 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
199 not available, instead of throwing an exception.
200 - use File::Spec to get the tmpdir in t/*, to avoid needless
201 failures on (most, not mine :) windows boxes.
202 - new handle read types: tls_detect and tls_autostart.
203
2047.0 Fri Apr 13 06:33:30 CEST 2012
205 - child watchers are broken in POE 1.352 (also many earlier
206 versions) and there seems to be no way to work around it, as POE
207 itself is inherently racy. Document this regression and add a delay
208 in t/68_poe_03_child.t for the time being.
209 - new module AnyEvent::IO, that is a frontend to either a pure-perl
210 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
211 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
212 - load /etc/hosts only when DNS has no answer.
213 - stat /etc/hosts on every access and reload it if it changed.
214 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
215 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
216 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
217 a side effect, they now use true appending as opposed to libc
218 appending, and the file might not have been opened when the function
219 returns.
220 - the default logging level was not properly documented in a variety of
221 places, this has hopefully been rectified.
222 - updated uts46data.pl for unicode 6.1.0.
223 - made log messages generated by AnyEvent submodules not
224 include the package name anymore, as it will be logged
225 by default already.
226 - upgrade to the trick used by common-sense 3.5 to work
227 around extra warning torture/breakage under perl 5.15.x.
228 - log messages by anyevent are now ucfirst, are usually full
229 sentences and do no longer include the package name.
230 - the storable read type would sometimes throw an exception instead
231 of causing EBADMSG (analyzed by Maxime Soulé).
232 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
233 keysize (reported by Paul Howarth).
234 - documented Rocco fabricating statements about AnyEvent and me.
235
2366.14 Tue Jan 31 20:00:24 CET 2012
237 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
238 probably an editing mistake (reported by Darin McBride).
239
2406.13 Thu Jan 12 07:27:01 CET 2012
241 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
242 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
243 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
244 regex-matching inside the callback would change the parameter, despite
245 $1 being dynamically scoped per-block. this perl bug is now being
246 worked around (testcase by Cindy Wang).
247
2486.12 Mon Dec 12 13:21:10 CET 2011
249 - $! was clobbered by subsequent calls in tcp_connect, due to
250 the postpone not saving/restoring it (analyzed by Richard Garnish).
251
2526.11 Tue Nov 22 10:36:05 CET 2011
253 - Tk cannot create windows when tainted unless you set a title, so
254 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
255 - escape any nonprintable/nonascii characters when stringifying
256 backtraces.
257 - log the reason loading an interface module fails to load at level debug.
258
2596.1 Tue Oct 4 19:44:30 CEST 2011
260 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
261 and above), and some messages inside AnyEvent have been elevated
262 to higher log levels to print by default.
263 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
264 - work around yet another signal race bug, newly introduced by POE. My
265 how that module sucks.
266 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
267 support ipv6-only operation even with defaults (also use ::1 instead
268 of 127.0.0.1).
269 - give better guidelines on how to chose a logging level.
270 - AnyEvent::Log can now cap (limit) log message priority by context.
271 - AE::log will now save and restore $! and $@, so callers don't have to.
272 - AE::Strict now checks that registered fds are still valid at regular
273 intervals.
274 - skip frequently called 'require' calls if the module appears loaded.
275 - increase signal latency test timer considerably for poor little osx,
276 which can't be bothered with basic posix functionality such as signal
277 delivery.
278 - mention some debugging aids in the tutorial.
279 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
280 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
281 of the context as title.
282 - AnyEvent::Strict would not restore read-write access to $_ when a callback
283 died - this is a user bug, but since AE::Strict exists to help dbeugging...
284 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
285 to work with it - the author is well aware of the issues but refuses
286 to fix them. This extreme step has been taken because the module actively
287 attacks the goals of AnyEvent and makes it harder for module authors to use
288 AnyEvent.
289
2906.02 Fri Aug 26 20:08:31 CEST 2011
291 - log_to_syslog now accepts facility strings.
292 - log_to_syslog would not allow facilities - "$facility|$level"
293 *literally* means that string. sheesh.
294 - the CPAN parser got confused and thought we provide Sys::Syslog, try
295 to work around.
296 - renamed AnyEvent::Impl::FLTK2 to FLTK.
297
2986.01 Fri Aug 26 07:04:11 CEST 2011
299 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
300 the TTL as fourth element - this affects ->resolve, ->request
301 and ->dns_unpack indirectly. Convenience functions such as
302 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
303 AnyEvent::Socket.
304 - INCOMPATIBLE CHANGE: backend authors now should not implement
305 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
306 methods.
307 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
308 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
309 - DEPRECIATION: please change
310 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
311 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
312 in your programs, for future compatibility.
313 - *blush* idle watcher emulation was borked.
314 - AnyEvent::Strict now checks the AE::xxx functions as well.
315 - querying too long domain names would barf deep inside AnyEvent::DNS - since
316 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
317 for too long domain names explicitly and passes an undef to the callback
318 instead (reported by James Bromberger).
319 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
320 - new functions: AnyEvent::postpone, AnyEvent::log.
321 - new module: AnyEvent::Log, for simple logging needs.
322 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
323 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
324 a debug shell.
325 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
326 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
327 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
328 /etc/hosts can be found.
329 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
330 configuration.
331 - env variables can now be specified using AE_xyz instead of
332 PERL_ANYEVENT_xyz, subject to some rules.
333 - lots of new AnyEvent::Debug functionality.
334 - AnyEvent::Strict now checks for modifications of $_ by the callback
335 (by setting it to readonly while executing the callback, if possible).
336 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
337 Invest extra overhead to work around this peculiar behaviour.
338 - split the pure perl backend into a loop module and an impl
339 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
340 - some read types (regex, netstring, json, storable) could
341 errornously keep an AE::Handle object alive on their own
342 (reported by Mohammad Toossi).
343 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
344 supports /etc/hosts overrides.
345 - add more workarounds around child watcher bugs in IO::Async
346 and (newly introduced) in POE.
347 - use glib child watchers and try to work around its limitations,
348 if possible (based on a patch by Kevin Ryde).
349 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
350 - many read types in AnyEvent::Handle would malfunction during
351 transitions to TLS because of a "delete $self->{rbuf}".
352 (reported by Gerald Galster).
353 - improve condvar blocking wait performance (probably only noticable
354 with faster backends), also make it easier for event loops that
355 do not support blocking waits.
356 - PERL_ANYEVENT_MODEL now also allows full module names, although
357 the usefulness of this feature is not yet clear.
358 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
359 - add t/80_ssltext.t.
360 - no longer use AUTOLOAD for the initial loading (for no very good reason).
361 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
362 - do not run event loop detection multiple times when programs call their cached
363 copies of AnyEvent::detect.
364 - parse_hostport now accepts absolute paths als unix domain sockets.
365 - suppress the idiotic warning inside POE by patching the POE::Kernel
366 run flags instead of asking the user to jump through contortions.
367 - add Task::Weaken as dependency (as recommends only at this time).
368 - replace PF_UNSPEC by 0, for good old minix.
369
3706.0 (not officially released, but was on CPAN for about 15 minutes :/)
371
3725.34 Thu May 12 09:59:41 CEST 2011
373 - AE::TLS didn't support both cert_file and key at the same time,
374 due to some copy&paste bug (reported and analyzed by vti).
375 - improve AE::Handle timeout documentation slightly.
376
3775.33 Mon Feb 7 21:51:41 CET 2011
378 - never appeared on CPAN, for unknown reasons.
379 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
380 object when it's already gone (analyzed by Ryan Bullock).
381 - update idna algorithm(s) to UTS#46 version 6.0.0.
382 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
383 uts46 says it's an error, and testsuite makes more sense.
384 - remove leading dots in idn_to_ascii - testsuite says to remove,
385 uts46 says it's an error, and testsuite makes more sense.
386 - fixed a bug in idn ignored character processing.
387 - passes the uts46 6.0.0 testsuite except for three cases which
388 apparently are bugs in the testsuite itself.
389 - passes the uts46 6.0.1 testsuite except for eleven cases which
390 apparently are bugs in the testsuite itself.
391
3925.31 Sun Jan 23 11:49:19 CET 2011
393 - work around a POE bug where POE would first poll for new events and
394 invoke handlers BEFORE handling the message queue.
395 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
396 - stop_read now does just that in TLS mode. see its
397 documentation for caveats.
398 - update to common::sense 3.4 warning set (some warnings in perl impose
399 an arbitrary 32kb limit on regex matches, which affects
400 AnyEvent::Handle).
401 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
402 - new experimental AE::Handle::resettls method.
403
4045.3 Fri Dec 31 05:48:13 CET 2010
405 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
406 issues with IO::Async have been solved in that version, including
407 the availabilty of a default loop!
408 - implement new wbuf_max parameter for AnyEvent::Handle.
409 - added Cocoa::EventLoop backend by Daisuke Murase.
410 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
411 - clarify handle on_prepare documentation.
412 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
413 to avoid leaking them to any child processes.
414 - try to force files into the arch-dependend lib directory, for the
415 half-"turly obsessive compulsive person"(s) that depend on it.
416
4175.29 Sun Dec 5 10:49:21 CET 2010
418 - convert EV backend to EV 4.00 API (so better upgrade EV too).
419 - AE::Handle: implement dynamic read_size adjustment, add
420 max_read_size parameter, reduce default read_size to 2048.
421 - add some FAQ entries for servers and/or on_eof vs. on_error.
422 - work around OS bugs (cygwin again) when lingering in AE::Handle.
423 - reduce memory usage slightly when connect was used in AE::Handle.
424 - wrok around more segfaults in Tk.
425 - document tls_ctx => undef as valid.
426 - detect WNOHANG value at compiletime.
427 - include some probably linux-only support for building AnyEvent
428 as part of the perl core.
429 - improve compatibility of t/08_idna.t to perl 5.8.
430 - make AnyEvent::Strict truly optional (in case somebody builds a
431 minimal perl...).
432
4335.28 Wed Oct 13 04:14:23 CEST 2010
434 - due to a glitch, AnyEvent's internal getprotobyname wasn't
435 actually used in tcp_server.
436 - implement AnyEvent::CondVar->new method, which was documented but
437 missing.
438 - support autoloading of read types in unshift_read.
439 - AnyEvent::Strict: do the test for invalid fh arguments differently,
440 resulting in more informative error messages.
441 - new AnyEvent::FAQ manpage.
442 - clarify that you can actually call push_read etc. in on_read.
443 - put keys and certs in /tmp, not into the current directory,
444 and document this in the SECURITY CONSIDERATIONS sections
445 of AnyEvent::TLS.
446 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
447 requested by Ron Isaacson.
448 - document $AnyEvent::DNS::RESOLVER variable.
449 - incorporated major typo patches by Abhijit Menon-Sen.
450 - AnyEvent::Handle now croaks when negative timeouts are passed.
451 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
452 undocumented, but who knows when they might come in handy.
453
4545.271 Tue Jun 8 12:05:46 CEST 2010
455 - backport to perl 5.8.x.
456
4575.27 Sun Jun 6 12:12:05 CEST 2010
458 - postpone differently in AnyEvent::Socket now, as
459 when not, canceling the connection attempt might fail
460 (found by Felix Antonius Wilhelm Ostmann).
461 - explicitly check for non-stream sockets in AE::Handle, too many
462 clueless people fell into the trap of this somehow working.
463 - simplified and reworked the "OTHER MODULES" section.
464 - better/more condvar examples.
465
4665.261 Wed Apr 28 16:13:36 CEST 2010
467 - AF_INET6 was not properly used from Socket6 during configuration
468 time (found by Andy Grundman).
469
4705.26 Mon Apr 12 04:49:35 CEST 2010
471 - don't generate spurious readyness notifications when select
472 returns EINTR in the pure perl backend (analysed and testcase
473 by Bas Denissen).
474 - give same examples in SYNOPSIS sections of AnyEvent and AE.
475 - provider faster implementations for AE::time/now/now_update for
476 EV and Perl backends.
477 - provide faster AE::cv and AE::time implementations for all
478 backends.
479 - no longer support lower/mixed-case signal names (this was
480 never documented nor universally supported).
481 - some more memory µ-optimisations, and somewhat less messy
482 code to cope with both AnyEvent and AE APIs.
483
4845.251 Sat Mar 13 00:58:21 CET 2010
485 - make test failed if it was run before make install on perl <=
486 5.8.8, and older versions of perl make our live very difficult,
487 so write constants.pl during Makefile.PL time.
488
4895.25 Sat Mar 13 00:23:14 CET 2010
490 - fix a race condition in AnyEvent::Handle that would cause
491 a "bio_read: unsupported method" error (in _another_
492 ssl connection) after user code threw an exception.
493 - added AnyEvent::Handle->destroyed method.
494 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
495 - the AnyEvent::Util::fh_nonblocking function worked only
496 by lucky accident on win32.
497 - smaller and faster AnyEvent::Util::fh_nonblocking.
498 - when the (required!) Time::HiRes module is missing, AnyEvent
499 did not fall back to built-in time properly.
500 - do not load Fcntl at runtime, saving memory and loading time.
501 - precompile a number of constants and use them instead of runtime
502 detection and eval.
503 - free detection code after detection and similar memory optimisations.
504 - Perl backend timer interval best effort drift has been improved
505 (same algorithm as EV).
506 - update unicode idna mapping table.
507
5085.24 Tue Jan 5 11:39:43 CET 2010
509 - cygwin never reports errors from failed connects. 1.5 just gives
510 you continous readyness and EAGAIN, 1.7 is even more broken and
511 just hangs. work around both issues in a cygwin-specific hack.
512 - improve idle watcher documentation slightly.
513
5145.23 Sun Dec 20 23:48:00 CET 2009
515 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
516 - implement punycode_encode/decode, idn_nameprep,
517 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
518 - provide $AE::VERSION.
519 - removed traces of "no strict 'refs'".
520
5215.22 Sat Dec 5 03:51:13 CET 2009
522 - downgrade-or-fail in AnyEvent::Handle::push_write, to
523 diagnose encoding failures earlier and more succinctly.
524 (this works around bugs in perl, throwing away encoding info
525 when passing scalar data to extensions).
526 - add more examples to AnyEvent::Socket manpage.
527 - upgrade internal warning set to the same as common::sense 2.03.
528 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
529 (John Beppu).
530 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
531 - add google public dns servers to fallback server set and make sure
532 we load-balance properly between the three sets. also add all
533 fallback dns servers, not just a random one, to each dns config.
534
5355.21 Thu Nov 19 02:48:47 CET 2009
536 - fix a problem where socket constants were called with parameters
537 (spotted by David Friedland).
538 - fork_call never use'd POSIX (reported by Daisuke Maki).
539 - improve perl 5.6 compatibility further (but it still won't work
540 unless you rip out everything but the core).
541 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
542 uses 10MB of RAM, but doesn't open a console window. *sigh*.
543
5445.202 Wed Oct 14 22:35:44 CEST 2009
545 - AnyEvent::DNS would unexpectedly clobber $_ under windows
546 (analysed by Matthias Waldorf).
547 - AnyEvent::Handle::run_cmd can now pass the PID of the
548 newly-created process, which is much less useful than it might
549 sound (based on patch by Yann Kerherve).
550
5515.201 Tue Sep 29 12:09:25 CEST 2009
552 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
553 (reported by Torsten Foertsch).
554 - common::sense 2.0 could cause tcp_server to throw an exception
555 (analysed by elmex).
556
5575.2 Mon Sep 14 07:04:49 CEST 2009
558 - INCOMPATIBLE CHANGE: do no longer support register_read_type
559 and register_write_type in AnyEvent::Handle, instead support
560 package names (the facility was mostly abused).
561 - implement "packagename-as-read/write type" support in
562 AnyEvent::Handle.
563 - AnyEvent::Handle: new options "keepalive" and "oobinline".
564 - oobinline set by default to avoid security issues.
565 - the pure-perl event loop backend wrongly detected times() fallback
566 support (spotted by Pavel Boldin).
567
5685.12 Tue Sep 1 20:26:50 CEST 2009
569 - be more lenient when parsing resolv.conf files, as some people
570 use hashmarks as comment indicator inside directives (reported
571 by Michael S. Fischer).
572 - use same set of warnings as common::sense 2.0.
573 - fix a potential 32 bit overflow issue due to perl having problems
574 with large hex constants in the Perl backend.
575 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
576 by default.
577 - fix some prototypes.
578 - (optionally) require Async::Interrupt 1.02.
579 - replace opendns fallbacks by verizon/level3 public dns servers.
580
5815.112 Fri Aug 21 13:59:04 CEST 2009
582 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
583 using starttls with an illegal/uncreatable tls context.
584 - port to broken win32 perls.
585
5865.111 Sun Aug 16 18:44:36 CEST 2009
587 - perl errornously requires sockaddr_un structures returned by the
588 kernel to have a specific length. work around this bug by padding
589 sockaddr_un structures (found to fail on OS X, as perl only uses
590 the correct code for GNU/Linux, but likely to fail on other
591 OSes as well).
592
5935.11 Wed Aug 12 17:49:37 CEST 2009
594 - fixed a bug in run_cmd with input from filename.
595 - create a json coder object when none is given in write direction
596 to reduce dependency on JSON.
597
5985.1 Tue Aug 11 03:17:32 CEST 2009
599 - tcp_connect now reports the correct error in $!, not an unrelated
600 one (was broken since 4.91).
601 - AE::Handle did not properly clear rtimeout/wtimeout watchers
602 when retrying a tcp_connect.
603 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
604 - fixed wrong documentation in AE::Socket.
605
6065.01 Mon Aug 10 03:16:32 CEST 2009
607 - last release broke EV child watchers :/.
608
6095.0 Sun Aug 9 17:32:09 CEST 2009
610 - officially introduce and document the AE API.
611 - lowering the timeout at runtime did not have immediate effect
612 in AnyEvent::Handle.
613 - AE::Handle now has separate and independent read and write timeout
614 settings.
615 - probe for Guard module when AnyEvent::Util is loaded, not at
616 runtime (Event doesn't like eval "use" from callbacks, and it avoids
617 exporting the probe function).
618 - Impl::IOAsync's io watchers did not use the filehandle emulation
619 code to work around it's watcher limits.
620 - work around IO::Async's problems with overloaded objects.
621 - try to disable wearnings inside IO::Async, it's just too buggy.
622 - do no longer use Event by default, only EV or Perl, Event is too buggy.
623 - hackishly provide fast AE:: interface when EV or Perl is the backend.
624 - inherit VERSION from AnyEvent::VERSION in most modules.
625
6264.91 Thu Aug 6 15:42:45 CEST 2009
7 - AE::Handle::starttls could get out of sync when the read buffer 627 - AE::Handle::starttls could get out of sync when the read buffer
8 already contains some TLS handshake. 628 already contains some TLS handshake.
9 - AE::Handle did not properly free the TLS session, causing bigger 629 - AE::Handle did not properly free the TLS session, causing bigger
10 memory leaks in Net::SSLeay than expected (reported by toaster). 630 memory leaks in Net::SSLeay than expected (reported by toaster).
11 - AE::Socket::tcp_connect will now artificially delay invoking 631 - AE::Socket::tcp_connect will now artificially delay invoking
12 the callback to avoid returning after invoking the callback. 632 the callback to avoid returning after invoking the callback.
633 - convert many internal watcher uses to AE API.
13 634
144.9 Sat Aug 1 11:07:01 CEST 2009 6354.9 Sat Aug 1 11:07:01 CEST 2009
15 - Glib silently fails when registering a timeout with 636 - Glib silently fails when registering a timeout with
16 a negative value, so avoid that. 637 a negative value, so avoid that.
17 - call condvar callback immediately when it is set 638 - call condvar callback immediately when it is set
… …
34 - new function: AnyEvent::Socket::format_hostport. 655 - new function: AnyEvent::Socket::format_hostport.
35 656
364.881 Tue Jul 28 12:51:53 CEST 2009 6574.881 Tue Jul 28 12:51:53 CEST 2009
37 - work around a bug in local in pre-5.10 perls, causing 658 - work around a bug in local in pre-5.10 perls, causing
38 AnyEvent::Handle to recurse when it should not 659 AnyEvent::Handle to recurse when it should not
39 (analyzed by elmex). 660 (analysed by elmex).
40 661
414.88 Tue Jul 28 04:04:37 CEST 2009 6624.88 Tue Jul 28 04:04:37 CEST 2009
42 - re-bless the handle into a dummy package after calling 663 - re-bless the handle into a dummy package after calling
43 AnyEvent::Handle::destroy, so the user does not need to check 664 AnyEvent::Handle::destroy, so the user does not need to check
44 for errors after every push_write etc. 665 for errors after every push_write etc.
… …
132 in the common case. 753 in the common case.
133 - add AnyEvent::Handle->push_shutdown method. 754 - add AnyEvent::Handle->push_shutdown method.
134 - add an additional error message parameter to AnyEvent::Handle's 755 - add an additional error message parameter to AnyEvent::Handle's
135 on_error callback (for TLS, $! is still available). 756 on_error callback (for TLS, $! is still available).
136 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 757 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
137 - make AnyEvent::Handle more robust against early conenction 758 - make AnyEvent::Handle more robust against early connection
138 failures (during new), and return C<undef> in that case 759 failures (during new), and return C<undef> in that case
139 from the constructor. 760 from the constructor.
140 - AnyEvent::Handle will now try to load only JSON::XS first, 761 - AnyEvent::Handle will now try to load only JSON::XS first,
141 then fall back to JSON. 762 then fall back to JSON.
142 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 763 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
2664.3 Fri Oct 3 09:18:43 CEST 2008 8874.3 Fri Oct 3 09:18:43 CEST 2008
267 - AnyEvent will now install a no-op signal handler for SIGPIPE, 888 - AnyEvent will now install a no-op signal handler for SIGPIPE,
268 unless one has been installed already. 889 unless one has been installed already.
269 - warn about not loaidng AnyEvent::Impl::POE early enough 890 - warn about not loaidng AnyEvent::Impl::POE early enough
270 (patch by Adam Rosenstein). 891 (patch by Adam Rosenstein).
271 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 892 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
272 - fix a bug in where in SSL connect mode, the 893 - fix a bug in where in SSL connect mode, the
273 client would first wait for some data by the server and 894 client would first wait for some data by the server and
274 otherwise hang. (reported and analyzed in an absolutely 895 otherwise hang. (reported and analysed in an absolutely
275 exemplary manner by Adam Rosenstein). 896 exemplary manner by Adam Rosenstein).
276 - fix a bug in where SSL EOF would not be treated 897 - fix a bug in where SSL EOF would not be treated
277 as stream EOF, putting the connection into a hung state. 898 as stream EOF, putting the connection into a hung state.
278 (reported and analyzed in an absolutely exemplary manner by Adam 899 (reported and analysed in an absolutely exemplary manner by Adam
279 Rosenstein). 900 Rosenstein).
280 - fix a potential segfault when the TLS context 901 - fix a potential segfault when the TLS context
281 would go missing in a read callback (e.g. due to stoptls) - 902 would go missing in a read callback (e.g. due to stoptls) -
282 Net::SSLeay of course makes no type checking whatsoever. 903 Net::SSLeay of course makes no type checking whatsoever.
283 - AnyEvent::Handle will not stop the read watcher in TLS mode. 904 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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