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

Comparing AnyEvent/Changes (file contents):
Revision 1.328 by root, Sun Aug 9 13:27:23 2009 UTC vs.
Revision 1.644 by root, Thu Mar 12 13:43:19 2020 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.92 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?
14TODO: Олег Г <verdrehung@gmail.com>, empty A record response causes
15AnyEvent::resolve_sockaddr to eventually call pack_sockaddr with undef -
16broken dns server, but maybe improve the response.
17TODO: AnyEvent::Loop should not use POSIIX when ti isn't used.
18
19TODO: AnyEvent::Log contexts should have a way to inject log messages received form elsewhere, using timestamp etc.
20TODO: add some cheta-sheet to log synopsis regarding AE_LOG=AnyEvent::MP=+
21TODO: tcp_connect option for AnyEvent::Handle?
22 - AnyEvent::Loop no longer loads the POSIX module if Time::HiRes
23 has monotonic clock support (suggested by Felipe Gasper).
24 - since Net::SSLeay (up to 1.88-1) is _still not ported to openssl 1.1
25 and many distributions therefore will have a completely broken
26 Net::SSLeay module for the forseeable future, AnyEvent::TLS
27 will now log a warning (that is not seen by default unless you
28 increase log verbosity).
29 - add tlsv1_3 disable switch to AnyEvent::TLS.
30 - add experimental "tls" method support for AnyEvent::TLS.
31 - tls methods were lowercase, but documented to be uppercase.
32 fix documentation.
33
347.17 Wed Sep 18 03:04:49 CEST 2019
35 - work around antique openssl version in RHEL 7 by formatting
36 dh parameters differently (reported by several people).
37 - add t/13_weaken.t.
38
397.16 Fri Jul 19 18:00:21 CEST 2019
40 - add ffdhe group dh parameters from RFC 7919, and use ffdhe3072
41 as new default, instead of schmorp1539.
42 - AnyEvent::Log did not reassess logging status of
43 AnyEvent::Log::logger's when contexts were changed with
44 ->attach/detach/slaves, causing them to not log even though a
45 recent attach should have caused them to log.
46 - added some more logging configuration examples.
47 - mention RFC 8482 in AnyEvent::DNS.
48
497.15 Tue Feb 26 03:07:42 CET 2019
50 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
51 separate major and minor versions, while code passed only a single
52 value. This version follows the documentation and now passes separate
53 major and minor values.
54 - work around Net::SSLeay not having been ported to openssl 1.1, but many
55 distributions compiling it against openssl 1.1, which unfortunately
56 succeeds and results in a very broken module.
57 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
58 in the __ member, to allow decoding of undecodable resouce records
59 containing compressed domian names.
60 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
61 addresses. It now does and converts them to ipv4 mapped addresses.
62 - support CAA records, based on patch by Steve Atkins.
63 - add freenom and cloudflare nameservers as dns fallback.
64 - AnyEvent::Strict would not properly ward against io watchers
65 on files when the handle passed was a file descriptor.
66 - document "internal" variables used by the dns en-/decoder to allow
67 enterprising users to extend them in a semi-official way.
68
697.14 Sat Jun 24 01:44:19 CEST 2017
70 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
71 by Paul Howarth and apparently tracked down by Mike McCauley).
72 - AnyEvent::Handle->keepalive was documented (and defined) twice
73 (found by Matt Merhar).
74 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
75 listening socket unless a guard is used. Change this so that no
76 cleanup will be performed unless a guard is used and document this more
77 clearly.
78 - make tcp_bind/tcp_server error messages more regular.
79 - work around latest perl upstream breakage - this time, perl5porters were
80 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
81 stupidly apply anything that has "security fix" stamped on it, it's likely
82 going to be a problem in practise (and working around it is trivial on
83 windows and unix).
84 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
85 to constants.
86 - update warnings to common::sense 3.74 standards.
87
887.13 Sat Sep 17 04:31:49 CEST 2016
89 - Only call tlsext_host_name for non-empty common names (reported
90 by Maxime Soulé).
91 - log a (single) notice message if SNI is not supported.
92 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
93 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
94 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
95 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
96 testcases and the specification and not in the code - the post-9.0.0
97 unicode files have all known problems fixed, so finally the AnyEvent
98 IDNA implementation can pass the full IDNA testsuite - without needing
99 a single fix :)
100 - guarantee (and document) that condvar callbacks will be removed
101 on invocation - important to avoid circular references.
102
1037.12 Wed Jan 27 19:12:26 CET 2016
104 - use common name as hostname for tls connects, if Net::SSLeay
105 supports SNI.
106 - fix documentation of tls_autostart read type in AnyEvent::Handle,
107 analyzed by Felix Ostmann.
108
1097.11 Thu Jul 16 14:36:00 CEST 2015
110 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
111 addresses (extra "::" at end and similar cases).
112 - add a more explicit warning to AnyEvent::Handle that it doesn't
113 work on files, people keep getting confused.
114 - new function AnyEvent::Socket::tcp_bind.
115 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
116 - aligned ipv6 address formatting with RFC 5952 (by not shortening
117 a single :0: to ::).
118 - added stability canary support.
119
1207.09 Sat May 2 16:38:53 CEST 2015
121 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
122 that was renamed to AnyEvent::Log:format_time. uinder its old name
123 (analyzed by Michael Stovenour).
124 - update AnyEvent::DNS fallback resolver addresses:
125 seems google effectively killed most other free dns resolvers,
126 so remove them, but add cable and wireless (ecrc) since it was
127 stable for 20 years or so, official or not, and there should be
128 an alternative to google.
129 - perl5porters broke windows error codes in 5.20, and mapped
130 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
131 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
132 because they were so cool. They probably broke other error codes
133 for no reason, too, but I didn't care for checking, it's too
134 depressing. This version only works around the WSAEWOULDBLOCK
135 issue, because I don't have a nice way to work around the
136 WSAEINPROGRESS bug.
137
1387.08 Wed Dec 10 05:27:17 CET 2014
139 - work around a newly introduced bug in Socket 2.011 (an
140 errornous sun_length check) (analyzed by Maxime Soulé).
141 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
142 (analyzed by Ben Magistro).
143 - AnyEvent::Strict will now confess, not croak. This is in line with
144 it being a development/debugging tool.
145 - work around a number of libglib bugs (debug builds of libglib enforce
146 certain undocumented behaviour patterns such as not being able to
147 remove a child watch source after it has fired, which we will try
148 to emulate to avoid "criticals". what where they thinking...).
149 - mention json security issues in AnyEvent::Handle, now that Douglas
150 Crockford has foolishly and incompatibly changed JSON.
151 - changed default dns resolver "max_outstanding" value from 1 to 10,
152 the latter beinfg the intended value all along
153 (reported by Ilya Chesnokov).
154 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
155 (written by Mike Lowell).
156
1577.07 Tue Dec 17 17:45:02 CET 2013
158 - the documentation for custom tls verify schemes was wrong. make it agree
159 with the code (reported by Maxime Soulé).
160 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
161 - work around an API change in openssl that could cause wrong tls connection
162 aborts, likely on windows only (analyzed by sten).
163 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
164 endless loop (reported by Dietrich Rebmann).
165 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
166 (patch by Maxime Soulé).
167 - document AnyEvent::Impl::IOAsync::set_loop and
168 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
169 functionality has _always_ been available.
170 - force a toplevel domain name in t/81_hosts.t (analyzed by
171 David Jack Wange Olrik).
172 - document that AnyEvent::Log uses AnyEvent::IO.
173 - warn about AnyEvent::Filesys::Notify performance.
174 - praise the joys of AnyEvent::Fork::*.
175 - time for an =encoding directive.
176 - do no longer use JSON to create a default json coder, use
177 JSON::XS or JSON::PP directly.
178
1797.05 Wed Aug 21 10:38:08 CEST 2013
180 - uts46data.pl couldn't be found due to wrong naming of the file
181 (reported by Fulko Hew).
182 - handle lone \015's properly in AE::Handle's default line read
183 (reported by various people).
184 - untaint ip addresses found in /etc/hosts (patch by José Micó).
185 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
186 fail on first use (reported and testcase by Andrew Whatson).
187 - expose AnyEvent::Log::format_time, and allow users to redefine it.
188 - expose AnyEvent::Log::default_format, and allow redefinition.
189 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
190 - AnyEvent::Debug shell can now run coro shell commands, if available.
191 - t/63* tests were wrongly in MANIFEST.
192 - kernel.org's finger server went MIA, switch to freebsd.org and
193 icculus.org.
194 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
195 to function asynchronously (reported by Daniel Carosone).
196 - hard-disable $^W in most tests, it generates too much garbage output.
197 - use a (hopefully) more future-proof method to emulate common::sense.
198 - upgrade to UTS-46:6.2.0.
199 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
200 documented. should not affect anything.
201
2027.04 Thu Nov 15 02:23:58 CET 2012
203 - debugging code left in.
204
2057.03 Thu Nov 15 02:16:11 CET 2012
206 - AnyEvent::Socket::inet_aton did not work when dns resolution was
207 used to find the addresses (analyzed and patch by Maxime Soulé).
208 - fix a memory leak in the /etc/hosts lookup code when hosts don't
209 resolve and are not in hosts (analyzed by sten).
210
2117.02 Tue Aug 14 04:27:58 CEST 2012
212 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
213 program by Yorhel).
214 - verified that AnyEvent::Socket follows RFC5952.
215 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
216
2177.01 Sun May 13 01:03:17 CEST 2012
218 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
219 not available, instead of throwing an exception.
220 - use File::Spec to get the tmpdir in t/*, to avoid needless
221 failures on (most, not mine :) windows boxes.
222 - new handle read types: tls_detect and tls_autostart.
223
2247.0 Fri Apr 13 06:33:30 CEST 2012
225 - child watchers are broken in POE 1.352 (also many earlier
226 versions) and there seems to be no way to work around it, as POE
227 itself is inherently racy. Document this regression and add a delay
228 in t/68_poe_03_child.t for the time being.
229 - new module AnyEvent::IO, that is a frontend to either a pure-perl
230 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
231 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
232 - load /etc/hosts only when DNS has no answer.
233 - stat /etc/hosts on every access and reload it if it changed.
234 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
235 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
236 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
237 a side effect, they now use true appending as opposed to libc
238 appending, and the file might not have been opened when the function
239 returns.
240 - the default logging level was not properly documented in a variety of
241 places, this has hopefully been rectified.
242 - updated uts46data.pl for unicode 6.1.0.
243 - made log messages generated by AnyEvent submodules not
244 include the package name anymore, as it will be logged
245 by default already.
246 - upgrade to the trick used by common-sense 3.5 to work
247 around extra warning torture/breakage under perl 5.15.x.
248 - log messages by anyevent are now ucfirst, are usually full
249 sentences and do no longer include the package name.
250 - the storable read type would sometimes throw an exception instead
251 of causing EBADMSG (analyzed by Maxime Soulé).
252 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
253 keysize (reported by Paul Howarth).
254 - documented Rocco fabricating statements about AnyEvent and me.
255
2566.14 Tue Jan 31 20:00:24 CET 2012
257 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
258 probably an editing mistake (reported by Darin McBride).
259
2606.13 Thu Jan 12 07:27:01 CET 2012
261 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
262 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
263 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
264 regex-matching inside the callback would change the parameter, despite
265 $1 being dynamically scoped per-block. this perl bug is now being
266 worked around (testcase by Cindy Wang).
267
2686.12 Mon Dec 12 13:21:10 CET 2011
269 - $! was clobbered by subsequent calls in tcp_connect, due to
270 the postpone not saving/restoring it (analyzed by Richard Garnish).
271
2726.11 Tue Nov 22 10:36:05 CET 2011
273 - Tk cannot create windows when tainted unless you set a title, so
274 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
275 - escape any nonprintable/nonascii characters when stringifying
276 backtraces.
277 - log the reason loading an interface module fails to load at level debug.
278
2796.1 Tue Oct 4 19:44:30 CEST 2011
280 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
281 and above), and some messages inside AnyEvent have been elevated
282 to higher log levels to print by default.
283 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
284 - work around yet another signal race bug, newly introduced by POE. My
285 how that module sucks.
286 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
287 support ipv6-only operation even with defaults (also use ::1 instead
288 of 127.0.0.1).
289 - give better guidelines on how to chose a logging level.
290 - AnyEvent::Log can now cap (limit) log message priority by context.
291 - AE::log will now save and restore $! and $@, so callers don't have to.
292 - AE::Strict now checks that registered fds are still valid at regular
293 intervals.
294 - skip frequently called 'require' calls if the module appears loaded.
295 - increase signal latency test timer considerably for poor little osx,
296 which can't be bothered with basic posix functionality such as signal
297 delivery.
298 - mention some debugging aids in the tutorial.
299 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
300 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
301 of the context as title.
302 - AnyEvent::Strict would not restore read-write access to $_ when a callback
303 died - this is a user bug, but since AE::Strict exists to help dbeugging...
304 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
305 to work with it - the author is well aware of the issues but refuses
306 to fix them. This extreme step has been taken because the module actively
307 attacks the goals of AnyEvent and makes it harder for module authors to use
308 AnyEvent.
309
3106.02 Fri Aug 26 20:08:31 CEST 2011
311 - log_to_syslog now accepts facility strings.
312 - log_to_syslog would not allow facilities - "$facility|$level"
313 *literally* means that string. sheesh.
314 - the CPAN parser got confused and thought we provide Sys::Syslog, try
315 to work around.
316 - renamed AnyEvent::Impl::FLTK2 to FLTK.
317
3186.01 Fri Aug 26 07:04:11 CEST 2011
319 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
320 the TTL as fourth element - this affects ->resolve, ->request
321 and ->dns_unpack indirectly. Convenience functions such as
322 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
323 AnyEvent::Socket.
324 - INCOMPATIBLE CHANGE: backend authors now should not implement
325 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
326 methods.
327 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
328 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
329 - DEPRECIATION: please change
330 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
331 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
332 in your programs, for future compatibility.
333 - *blush* idle watcher emulation was borked.
334 - AnyEvent::Strict now checks the AE::xxx functions as well.
335 - querying too long domain names would barf deep inside AnyEvent::DNS - since
336 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
337 for too long domain names explicitly and passes an undef to the callback
338 instead (reported by James Bromberger).
339 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
340 - new functions: AnyEvent::postpone, AnyEvent::log.
341 - new module: AnyEvent::Log, for simple logging needs.
342 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
343 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
344 a debug shell.
345 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
346 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
347 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
348 /etc/hosts can be found.
349 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
350 configuration.
351 - env variables can now be specified using AE_xyz instead of
352 PERL_ANYEVENT_xyz, subject to some rules.
353 - lots of new AnyEvent::Debug functionality.
354 - AnyEvent::Strict now checks for modifications of $_ by the callback
355 (by setting it to readonly while executing the callback, if possible).
356 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
357 Invest extra overhead to work around this peculiar behaviour.
358 - split the pure perl backend into a loop module and an impl
359 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
360 - some read types (regex, netstring, json, storable) could
361 errornously keep an AE::Handle object alive on their own
362 (reported by Mohammad Toossi).
363 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
364 supports /etc/hosts overrides.
365 - add more workarounds around child watcher bugs in IO::Async
366 and (newly introduced) in POE.
367 - use glib child watchers and try to work around its limitations,
368 if possible (based on a patch by Kevin Ryde).
369 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
370 - many read types in AnyEvent::Handle would malfunction during
371 transitions to TLS because of a "delete $self->{rbuf}".
372 (reported by Gerald Galster).
373 - improve condvar blocking wait performance (probably only noticable
374 with faster backends), also make it easier for event loops that
375 do not support blocking waits.
376 - PERL_ANYEVENT_MODEL now also allows full module names, although
377 the usefulness of this feature is not yet clear.
378 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
379 - add t/80_ssltext.t.
380 - no longer use AUTOLOAD for the initial loading (for no very good reason).
381 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
382 - do not run event loop detection multiple times when programs call their cached
383 copies of AnyEvent::detect.
384 - parse_hostport now accepts absolute paths als unix domain sockets.
385 - suppress the idiotic warning inside POE by patching the POE::Kernel
386 run flags instead of asking the user to jump through contortions.
387 - add Task::Weaken as dependency (as recommends only at this time).
388 - replace PF_UNSPEC by 0, for good old minix.
389
3906.0 (not officially released, but was on CPAN for about 15 minutes :/)
391
3925.34 Thu May 12 09:59:41 CEST 2011
393 - AE::TLS didn't support both cert_file and key at the same time,
394 due to some copy&paste bug (reported and analyzed by vti).
395 - improve AE::Handle timeout documentation slightly.
396
3975.33 Mon Feb 7 21:51:41 CET 2011
398 - never appeared on CPAN, for unknown reasons.
399 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
400 object when it's already gone (analyzed by Ryan Bullock).
401 - update idna algorithm(s) to UTS#46 version 6.0.0.
402 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
403 uts46 says it's an error, and testsuite makes more sense.
404 - remove leading dots in idn_to_ascii - testsuite says to remove,
405 uts46 says it's an error, and testsuite makes more sense.
406 - fixed a bug in idn ignored character processing.
407 - passes the uts46 6.0.0 testsuite except for three cases which
408 apparently are bugs in the testsuite itself.
409 - passes the uts46 6.0.1 testsuite except for eleven cases which
410 apparently are bugs in the testsuite itself.
411
4125.31 Sun Jan 23 11:49:19 CET 2011
413 - work around a POE bug where POE would first poll for new events and
414 invoke handlers BEFORE handling the message queue.
415 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
416 - stop_read now does just that in TLS mode. see its
417 documentation for caveats.
418 - update to common::sense 3.4 warning set (some warnings in perl impose
419 an arbitrary 32kb limit on regex matches, which affects
420 AnyEvent::Handle).
421 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
422 - new experimental AE::Handle::resettls method.
423
4245.3 Fri Dec 31 05:48:13 CET 2010
425 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
426 issues with IO::Async have been solved in that version, including
427 the availabilty of a default loop!
428 - implement new wbuf_max parameter for AnyEvent::Handle.
429 - added Cocoa::EventLoop backend by Daisuke Murase.
430 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
431 - clarify handle on_prepare documentation.
432 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
433 to avoid leaking them to any child processes.
434 - try to force files into the arch-dependend lib directory, for the
435 half-"turly obsessive compulsive person"(s) that depend on it.
436
4375.29 Sun Dec 5 10:49:21 CET 2010
438 - convert EV backend to EV 4.00 API (so better upgrade EV too).
439 - AE::Handle: implement dynamic read_size adjustment, add
440 max_read_size parameter, reduce default read_size to 2048.
441 - add some FAQ entries for servers and/or on_eof vs. on_error.
442 - work around OS bugs (cygwin again) when lingering in AE::Handle.
443 - reduce memory usage slightly when connect was used in AE::Handle.
444 - wrok around more segfaults in Tk.
445 - document tls_ctx => undef as valid.
446 - detect WNOHANG value at compiletime.
447 - include some probably linux-only support for building AnyEvent
448 as part of the perl core.
449 - improve compatibility of t/08_idna.t to perl 5.8.
450 - make AnyEvent::Strict truly optional (in case somebody builds a
451 minimal perl...).
452
4535.28 Wed Oct 13 04:14:23 CEST 2010
454 - due to a glitch, AnyEvent's internal getprotobyname wasn't
455 actually used in tcp_server.
456 - implement AnyEvent::CondVar->new method, which was documented but
457 missing.
458 - support autoloading of read types in unshift_read.
459 - AnyEvent::Strict: do the test for invalid fh arguments differently,
460 resulting in more informative error messages.
461 - new AnyEvent::FAQ manpage.
462 - clarify that you can actually call push_read etc. in on_read.
463 - put keys and certs in /tmp, not into the current directory,
464 and document this in the SECURITY CONSIDERATIONS sections
465 of AnyEvent::TLS.
466 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
467 requested by Ron Isaacson.
468 - document $AnyEvent::DNS::RESOLVER variable.
469 - incorporated major typo patches by Abhijit Menon-Sen.
470 - AnyEvent::Handle now croaks when negative timeouts are passed.
471 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
472 undocumented, but who knows when they might come in handy.
473
4745.271 Tue Jun 8 12:05:46 CEST 2010
475 - backport to perl 5.8.x.
476
4775.27 Sun Jun 6 12:12:05 CEST 2010
478 - postpone differently in AnyEvent::Socket now, as
479 when not, canceling the connection attempt might fail
480 (found by Felix Antonius Wilhelm Ostmann).
481 - explicitly check for non-stream sockets in AE::Handle, too many
482 clueless people fell into the trap of this somehow working.
483 - simplified and reworked the "OTHER MODULES" section.
484 - better/more condvar examples.
485
4865.261 Wed Apr 28 16:13:36 CEST 2010
487 - AF_INET6 was not properly used from Socket6 during configuration
488 time (found by Andy Grundman).
489
4905.26 Mon Apr 12 04:49:35 CEST 2010
491 - don't generate spurious readyness notifications when select
492 returns EINTR in the pure perl backend (analysed and testcase
493 by Bas Denissen).
494 - give same examples in SYNOPSIS sections of AnyEvent and AE.
495 - provider faster implementations for AE::time/now/now_update for
496 EV and Perl backends.
497 - provide faster AE::cv and AE::time implementations for all
498 backends.
499 - no longer support lower/mixed-case signal names (this was
500 never documented nor universally supported).
501 - some more memory µ-optimisations, and somewhat less messy
502 code to cope with both AnyEvent and AE APIs.
503
5045.251 Sat Mar 13 00:58:21 CET 2010
505 - make test failed if it was run before make install on perl <=
506 5.8.8, and older versions of perl make our live very difficult,
507 so write constants.pl during Makefile.PL time.
508
5095.25 Sat Mar 13 00:23:14 CET 2010
510 - fix a race condition in AnyEvent::Handle that would cause
511 a "bio_read: unsupported method" error (in _another_
512 ssl connection) after user code threw an exception.
513 - added AnyEvent::Handle->destroyed method.
514 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
515 - the AnyEvent::Util::fh_nonblocking function worked only
516 by lucky accident on win32.
517 - smaller and faster AnyEvent::Util::fh_nonblocking.
518 - when the (required!) Time::HiRes module is missing, AnyEvent
519 did not fall back to built-in time properly.
520 - do not load Fcntl at runtime, saving memory and loading time.
521 - precompile a number of constants and use them instead of runtime
522 detection and eval.
523 - free detection code after detection and similar memory optimisations.
524 - Perl backend timer interval best effort drift has been improved
525 (same algorithm as EV).
526 - update unicode idna mapping table.
527
5285.24 Tue Jan 5 11:39:43 CET 2010
529 - cygwin never reports errors from failed connects. 1.5 just gives
530 you continous readyness and EAGAIN, 1.7 is even more broken and
531 just hangs. work around both issues in a cygwin-specific hack.
532 - improve idle watcher documentation slightly.
533
5345.23 Sun Dec 20 23:48:00 CET 2009
535 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
536 - implement punycode_encode/decode, idn_nameprep,
537 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
538 - provide $AE::VERSION.
539 - removed traces of "no strict 'refs'".
540
5415.22 Sat Dec 5 03:51:13 CET 2009
542 - downgrade-or-fail in AnyEvent::Handle::push_write, to
543 diagnose encoding failures earlier and more succinctly.
544 (this works around bugs in perl, throwing away encoding info
545 when passing scalar data to extensions).
546 - add more examples to AnyEvent::Socket manpage.
547 - upgrade internal warning set to the same as common::sense 2.03.
548 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
549 (John Beppu).
550 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
551 - add google public dns servers to fallback server set and make sure
552 we load-balance properly between the three sets. also add all
553 fallback dns servers, not just a random one, to each dns config.
554
5555.21 Thu Nov 19 02:48:47 CET 2009
556 - fix a problem where socket constants were called with parameters
557 (spotted by David Friedland).
558 - fork_call never use'd POSIX (reported by Daisuke Maki).
559 - improve perl 5.6 compatibility further (but it still won't work
560 unless you rip out everything but the core).
561 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
562 uses 10MB of RAM, but doesn't open a console window. *sigh*.
563
5645.202 Wed Oct 14 22:35:44 CEST 2009
565 - AnyEvent::DNS would unexpectedly clobber $_ under windows
566 (analysed by Matthias Waldorf).
567 - AnyEvent::Handle::run_cmd can now pass the PID of the
568 newly-created process, which is much less useful than it might
569 sound (based on patch by Yann Kerherve).
570
5715.201 Tue Sep 29 12:09:25 CEST 2009
572 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
573 (reported by Torsten Foertsch).
574 - common::sense 2.0 could cause tcp_server to throw an exception
575 (analysed by elmex).
576
5775.2 Mon Sep 14 07:04:49 CEST 2009
578 - INCOMPATIBLE CHANGE: do no longer support register_read_type
579 and register_write_type in AnyEvent::Handle, instead support
580 package names (the facility was mostly abused).
581 - implement "packagename-as-read/write type" support in
582 AnyEvent::Handle.
583 - AnyEvent::Handle: new options "keepalive" and "oobinline".
584 - oobinline set by default to avoid security issues.
585 - the pure-perl event loop backend wrongly detected times() fallback
586 support (spotted by Pavel Boldin).
587
5885.12 Tue Sep 1 20:26:50 CEST 2009
589 - be more lenient when parsing resolv.conf files, as some people
590 use hashmarks as comment indicator inside directives (reported
591 by Michael S. Fischer).
592 - use same set of warnings as common::sense 2.0.
593 - fix a potential 32 bit overflow issue due to perl having problems
594 with large hex constants in the Perl backend.
595 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
596 by default.
597 - fix some prototypes.
598 - (optionally) require Async::Interrupt 1.02.
599 - replace opendns fallbacks by verizon/level3 public dns servers.
600
6015.112 Fri Aug 21 13:59:04 CEST 2009
602 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
603 using starttls with an illegal/uncreatable tls context.
604 - port to broken win32 perls.
605
6065.111 Sun Aug 16 18:44:36 CEST 2009
607 - perl errornously requires sockaddr_un structures returned by the
608 kernel to have a specific length. work around this bug by padding
609 sockaddr_un structures (found to fail on OS X, as perl only uses
610 the correct code for GNU/Linux, but likely to fail on other
611 OSes as well).
612
6135.11 Wed Aug 12 17:49:37 CEST 2009
614 - fixed a bug in run_cmd with input from filename.
615 - create a json coder object when none is given in write direction
616 to reduce dependency on JSON.
617
6185.1 Tue Aug 11 03:17:32 CEST 2009
619 - tcp_connect now reports the correct error in $!, not an unrelated
620 one (was broken since 4.91).
621 - AE::Handle did not properly clear rtimeout/wtimeout watchers
622 when retrying a tcp_connect.
623 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
624 - fixed wrong documentation in AE::Socket.
625
6265.01 Mon Aug 10 03:16:32 CEST 2009
627 - last release broke EV child watchers :/.
628
6295.0 Sun Aug 9 17:32:09 CEST 2009
630 - officially introduce and document the AE API.
4 - lowering the timeout at runtime did not have immediate effect 631 - lowering the timeout at runtime did not have immediate effect
5 in AnyEvent::Handle. 632 in AnyEvent::Handle.
6 - AE::Handle now has separate and independent read and write timeout 633 - AE::Handle now has separate and independent read and write timeout
7 settings. 634 settings.
8 - probe for Guard module when AnyEvent::Util is loaded, not at 635 - probe for Guard module when AnyEvent::Util is loaded, not at
… …
11 - Impl::IOAsync's io watchers did not use the filehandle emulation 638 - Impl::IOAsync's io watchers did not use the filehandle emulation
12 code to work around it's watcher limits. 639 code to work around it's watcher limits.
13 - work around IO::Async's problems with overloaded objects. 640 - work around IO::Async's problems with overloaded objects.
14 - try to disable wearnings inside IO::Async, it's just too buggy. 641 - try to disable wearnings inside IO::Async, it's just too buggy.
15 - do no longer use Event by default, only EV or Perl, Event is too buggy. 642 - do no longer use Event by default, only EV or Perl, Event is too buggy.
16 - hackishly provide fast AE:: interface when EV is the backend. 643 - hackishly provide fast AE:: interface when EV or Perl is the backend.
17 - inherit VERSION from AnyEvent::VERSION in most modules. 644 - inherit VERSION from AnyEvent::VERSION in most modules.
18 645
194.91 Thu Aug 6 15:42:45 CEST 2009 6464.91 Thu Aug 6 15:42:45 CEST 2009
20 - AE::Handle::starttls could get out of sync when the read buffer 647 - AE::Handle::starttls could get out of sync when the read buffer
21 already contains some TLS handshake. 648 already contains some TLS handshake.
… …
48 - new function: AnyEvent::Socket::format_hostport. 675 - new function: AnyEvent::Socket::format_hostport.
49 676
504.881 Tue Jul 28 12:51:53 CEST 2009 6774.881 Tue Jul 28 12:51:53 CEST 2009
51 - work around a bug in local in pre-5.10 perls, causing 678 - work around a bug in local in pre-5.10 perls, causing
52 AnyEvent::Handle to recurse when it should not 679 AnyEvent::Handle to recurse when it should not
53 (analyzed by elmex). 680 (analysed by elmex).
54 681
554.88 Tue Jul 28 04:04:37 CEST 2009 6824.88 Tue Jul 28 04:04:37 CEST 2009
56 - re-bless the handle into a dummy package after calling 683 - re-bless the handle into a dummy package after calling
57 AnyEvent::Handle::destroy, so the user does not need to check 684 AnyEvent::Handle::destroy, so the user does not need to check
58 for errors after every push_write etc. 685 for errors after every push_write etc.
… …
146 in the common case. 773 in the common case.
147 - add AnyEvent::Handle->push_shutdown method. 774 - add AnyEvent::Handle->push_shutdown method.
148 - add an additional error message parameter to AnyEvent::Handle's 775 - add an additional error message parameter to AnyEvent::Handle's
149 on_error callback (for TLS, $! is still available). 776 on_error callback (for TLS, $! is still available).
150 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 777 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
151 - make AnyEvent::Handle more robust against early conenction 778 - make AnyEvent::Handle more robust against early connection
152 failures (during new), and return C<undef> in that case 779 failures (during new), and return C<undef> in that case
153 from the constructor. 780 from the constructor.
154 - AnyEvent::Handle will now try to load only JSON::XS first, 781 - AnyEvent::Handle will now try to load only JSON::XS first,
155 then fall back to JSON. 782 then fall back to JSON.
156 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 783 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
2804.3 Fri Oct 3 09:18:43 CEST 2008 9074.3 Fri Oct 3 09:18:43 CEST 2008
281 - AnyEvent will now install a no-op signal handler for SIGPIPE, 908 - AnyEvent will now install a no-op signal handler for SIGPIPE,
282 unless one has been installed already. 909 unless one has been installed already.
283 - warn about not loaidng AnyEvent::Impl::POE early enough 910 - warn about not loaidng AnyEvent::Impl::POE early enough
284 (patch by Adam Rosenstein). 911 (patch by Adam Rosenstein).
285 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 912 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
286 - fix a bug in where in SSL connect mode, the 913 - fix a bug in where in SSL connect mode, the
287 client would first wait for some data by the server and 914 client would first wait for some data by the server and
288 otherwise hang. (reported and analyzed in an absolutely 915 otherwise hang. (reported and analysed in an absolutely
289 exemplary manner by Adam Rosenstein). 916 exemplary manner by Adam Rosenstein).
290 - fix a bug in where SSL EOF would not be treated 917 - fix a bug in where SSL EOF would not be treated
291 as stream EOF, putting the connection into a hung state. 918 as stream EOF, putting the connection into a hung state.
292 (reported and analyzed in an absolutely exemplary manner by Adam 919 (reported and analysed in an absolutely exemplary manner by Adam
293 Rosenstein). 920 Rosenstein).
294 - fix a potential segfault when the TLS context 921 - fix a potential segfault when the TLS context
295 would go missing in a read callback (e.g. due to stoptls) - 922 would go missing in a read callback (e.g. due to stoptls) -
296 Net::SSLeay of course makes no type checking whatsoever. 923 Net::SSLeay of course makes no type checking whatsoever.
297 - AnyEvent::Handle will not stop the read watcher in TLS mode. 924 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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