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

Comparing AnyEvent/Changes (file contents):
Revision 1.432 by root, Fri Dec 31 04:50:44 2010 UTC vs.
Revision 1.631 by root, Tue Feb 26 02:08:34 2019 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: document TCP_*** constants 3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent strict and anyevent debug 4TODO: add debug/trace logging to some modules
5TODO: anyevent::trace? or debug tracing all watchers? 5TODO: use the eval-with-cb-call trick also to speed up JSON?
6TODO: anyevent strict checking for $_ changes? 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?
14
157.15 Tue Feb 26 03:07:42 CET 2019
16 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
17 separate major and minor versions, while code passed only a single
18 value. This version follows the documentation and now passes separate
19 major and minor values.
20 - work around Net::SSLeay not having been ported to openssl 1.1, but many
21 distributions compiling it against openssl 1.1, which unfortunately
22 succeeds and results in a very broken module.
23 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
24 in the __ member, to allow decoding of undecodable resouce records
25 containing compressed domian names.
26 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
27 addresses. It now does and converts them to ipv4 mapped addresses.
28 - support CAA records, based on patch by Steve Atkins.
29 - add freenom and cloudflare nameservers as dns fallback.
30 - AnyEvent::Strict would not properly ward against io watchers
31 on files when the handle passed was a file descriptor.
32 - document "internal" variables used by the dns en-/decoder to allow
33 enterprising users to extend them in a semi-official way.
34
357.14 Sat Jun 24 01:44:19 CEST 2017
36 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
37 by Paul Howarth and apparently tracked down by Mike McCauley).
38 - AnyEvent::Handle->keepalive was documented (and defined) twice
39 (found by Matt Merhar).
40 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
41 listening socket unless a guard is used. Change this so that no
42 cleanup will be performed unless a guard is used and document this more
43 clearly.
44 - make tcp_bind/tcp_server error messages more regular.
45 - work around latest perl upstream breakage - this time, perl5porters were
46 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
47 stupidly apply anything that has "security fix" stamped on it, it's likely
48 going to be a problem in practise (and working around it is trivial on
49 windows and unix).
50 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
51 to constants.
52 - update warnings to common::sense 3.74 standards.
53
547.13 Sat Sep 17 04:31:49 CEST 2016
55 - Only call tlsext_host_name for non-empty common names (reported
56 by Maxime Soulé).
57 - log a (single) notice message if SNI is not supported.
58 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
59 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
60 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
61 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
62 testcases and the specification and not in the code - the post-9.0.0
63 unicode files have all known problems fixed, so finally the AnyEvent
64 IDNA implementation can pass the full IDNA testsuite - without needing
65 a single fix :)
66 - guarantee (and document) that condvar callbacks will be removed
67 on invocation - important to avoid circular references.
68
697.12 Wed Jan 27 19:12:26 CET 2016
70 - use common name as hostname for tls connects, if Net::SSLeay
71 supports SNI.
72 - fix documentation of tls_autostart read type in AnyEvent::Handle,
73 analyzed by Felix Ostmann.
74
757.11 Thu Jul 16 14:36:00 CEST 2015
76 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
77 addresses (extra "::" at end and similar cases).
78 - add a more explicit warning to AnyEvent::Handle that it doesn't
79 work on files, people keep getting confused.
80 - new function AnyEvent::Socket::tcp_bind.
81 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
82 - aligned ipv6 address formatting with RFC 5952 (by not shortening
83 a single :0: to ::).
84 - added stability canary support.
85
867.09 Sat May 2 16:38:53 CEST 2015
87 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
88 that was renamed to AnyEvent::Log:format_time. uinder its old name
89 (analyzed by Michael Stovenour).
90 - update AnyEvent::DNS fallback resolver addresses:
91 seems google effectively killed most other free dns resolvers,
92 so remove them, but add cable and wireless (ecrc) since it was
93 stable for 20 years or so, official or not, and there should be
94 an alternative to google.
95 - perl5porters broke windows error codes in 5.20, and mapped
96 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
97 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
98 because they were so cool. They probably broke other error codes
99 for no reason, too, but I didn't care for checking, it's too
100 depressing. This version only works around the WSAEWOULDBLOCK
101 issue, because I don't have a nice way to work around the
102 WSAEINPROGRESS bug.
103
1047.08 Wed Dec 10 05:27:17 CET 2014
105 - work around a newly introduced bug in Socket 2.011 (an
106 errornous sun_length check) (analyzed by Maxime Soulé).
107 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
108 (analyzed by Ben Magistro).
109 - AnyEvent::Strict will now confess, not croak. This is in line with
110 it being a development/debugging tool.
111 - work around a number of libglib bugs (debug builds of libglib enforce
112 certain undocumented behaviour patterns such as not being able to
113 remove a child watch source after it has fired, which we will try
114 to emulate to avoid "criticals". what where they thinking...).
115 - mention json security issues in AnyEvent::Handle, now that Douglas
116 Crockford has foolishly and incompatibly changed JSON.
117 - changed default dns resolver "max_outstanding" value from 1 to 10,
118 the latter beinfg the intended value all along
119 (reported by Ilya Chesnokov).
120 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
121 (written by Mike Lowell).
122
1237.07 Tue Dec 17 17:45:02 CET 2013
124 - the documentation for custom tls verify schemes was wrong. make it agree
125 with the code (reported by Maxime Soulé).
126 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
127 - work around an API change in openssl that could cause wrong tls connection
128 aborts, likely on windows only (analyzed by sten).
129 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
130 endless loop (reported by Dietrich Rebmann).
131 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
132 (patch by Maxime Soulé).
133 - document AnyEvent::Impl::IOAsync::set_loop and
134 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
135 functionality has _always_ been available.
136 - force a toplevel domain name in t/81_hosts.t (analyzed by
137 David Jack Wange Olrik).
138 - document that AnyEvent::Log uses AnyEvent::IO.
139 - warn about AnyEvent::Filesys::Notify performance.
140 - praise the joys of AnyEvent::Fork::*.
141 - time for an =encoding directive.
142 - do no longer use JSON to create a default json coder, use
143 JSON::XS or JSON::PP directly.
144
1457.05 Wed Aug 21 10:38:08 CEST 2013
146 - uts46data.pl couldn't be found due to wrong naming of the file
147 (reported by Fulko Hew).
148 - handle lone \015's properly in AE::Handle's default line read
149 (reported by various people).
150 - untaint ip addresses found in /etc/hosts (patch by José Micó).
151 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
152 fail on first use (reported and testcase by Andrew Whatson).
153 - expose AnyEvent::Log::format_time, and allow users to redefine it.
154 - expose AnyEvent::Log::default_format, and allow redefinition.
155 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
156 - AnyEvent::Debug shell can now run coro shell commands, if available.
157 - t/63* tests were wrongly in MANIFEST.
158 - kernel.org's finger server went MIA, switch to freebsd.org and
159 icculus.org.
160 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
161 to function asynchronously (reported by Daniel Carosone).
162 - hard-disable $^W in most tests, it generates too much garbage output.
163 - use a (hopefully) more future-proof method to emulate common::sense.
164 - upgrade to UTS-46:6.2.0.
165 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
166 documented. should not affect anything.
167
1687.04 Thu Nov 15 02:23:58 CET 2012
169 - debugging code left in.
170
1717.03 Thu Nov 15 02:16:11 CET 2012
172 - AnyEvent::Socket::inet_aton did not work when dns resolution was
173 used to find the addresses (analyzed and patch by Maxime Soulé).
174 - fix a memory leak in the /etc/hosts lookup code when hosts don't
175 resolve and are not in hosts (analyzed by sten).
176
1777.02 Tue Aug 14 04:27:58 CEST 2012
178 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
179 program by Yorhel).
180 - verified that AnyEvent::Socket follows RFC5952.
181 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
182
1837.01 Sun May 13 01:03:17 CEST 2012
184 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
185 not available, instead of throwing an exception.
186 - use File::Spec to get the tmpdir in t/*, to avoid needless
187 failures on (most, not mine :) windows boxes.
188 - new handle read types: tls_detect and tls_autostart.
189
1907.0 Fri Apr 13 06:33:30 CEST 2012
191 - child watchers are broken in POE 1.352 (also many earlier
192 versions) and there seems to be no way to work around it, as POE
193 itself is inherently racy. Document this regression and add a delay
194 in t/68_poe_03_child.t for the time being.
195 - new module AnyEvent::IO, that is a frontend to either a pure-perl
196 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
197 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
198 - load /etc/hosts only when DNS has no answer.
199 - stat /etc/hosts on every access and reload it if it changed.
200 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
201 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
202 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
203 a side effect, they now use true appending as opposed to libc
204 appending, and the file might not have been opened when the function
205 returns.
206 - the default logging level was not properly documented in a variety of
207 places, this has hopefully been rectified.
208 - updated uts46data.pl for unicode 6.1.0.
209 - made log messages generated by AnyEvent submodules not
210 include the package name anymore, as it will be logged
211 by default already.
212 - upgrade to the trick used by common-sense 3.5 to work
213 around extra warning torture/breakage under perl 5.15.x.
214 - log messages by anyevent are now ucfirst, are usually full
215 sentences and do no longer include the package name.
216 - the storable read type would sometimes throw an exception instead
217 of causing EBADMSG (analyzed by Maxime Soulé).
218 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
219 keysize (reported by Paul Howarth).
220 - documented Rocco fabricating statements about AnyEvent and me.
221
2226.14 Tue Jan 31 20:00:24 CET 2012
223 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
224 probably an editing mistake (reported by Darin McBride).
225
2266.13 Thu Jan 12 07:27:01 CET 2012
227 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
228 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
229 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
230 regex-matching inside the callback would change the parameter, despite
231 $1 being dynamically scoped per-block. this perl bug is now being
232 worked around (testcase by Cindy Wang).
233
2346.12 Mon Dec 12 13:21:10 CET 2011
235 - $! was clobbered by subsequent calls in tcp_connect, due to
236 the postpone not saving/restoring it (analyzed by Richard Garnish).
237
2386.11 Tue Nov 22 10:36:05 CET 2011
239 - Tk cannot create windows when tainted unless you set a title, so
240 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
241 - escape any nonprintable/nonascii characters when stringifying
242 backtraces.
243 - log the reason loading an interface module fails to load at level debug.
244
2456.1 Tue Oct 4 19:44:30 CEST 2011
246 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
247 and above), and some messages inside AnyEvent have been elevated
248 to higher log levels to print by default.
249 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
250 - work around yet another signal race bug, newly introduced by POE. My
251 how that module sucks.
252 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
253 support ipv6-only operation even with defaults (also use ::1 instead
254 of 127.0.0.1).
255 - give better guidelines on how to chose a logging level.
256 - AnyEvent::Log can now cap (limit) log message priority by context.
257 - AE::log will now save and restore $! and $@, so callers don't have to.
258 - AE::Strict now checks that registered fds are still valid at regular
259 intervals.
260 - skip frequently called 'require' calls if the module appears loaded.
261 - increase signal latency test timer considerably for poor little osx,
262 which can't be bothered with basic posix functionality such as signal
263 delivery.
264 - mention some debugging aids in the tutorial.
265 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
266 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
267 of the context as title.
268 - AnyEvent::Strict would not restore read-write access to $_ when a callback
269 died - this is a user bug, but since AE::Strict exists to help dbeugging...
270 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
271 to work with it - the author is well aware of the issues but refuses
272 to fix them. This extreme step has been taken because the module actively
273 attacks the goals of AnyEvent and makes it harder for module authors to use
274 AnyEvent.
275
2766.02 Fri Aug 26 20:08:31 CEST 2011
277 - log_to_syslog now accepts facility strings.
278 - log_to_syslog would not allow facilities - "$facility|$level"
279 *literally* means that string. sheesh.
280 - the CPAN parser got confused and thought we provide Sys::Syslog, try
281 to work around.
282 - renamed AnyEvent::Impl::FLTK2 to FLTK.
283
2846.01 Fri Aug 26 07:04:11 CEST 2011
285 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
286 the TTL as fourth element - this affects ->resolve, ->request
287 and ->dns_unpack indirectly. Convenience functions such as
288 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
289 AnyEvent::Socket.
290 - INCOMPATIBLE CHANGE: backend authors now should not implement
291 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
292 methods.
293 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
294 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
295 - DEPRECIATION: please change
296 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
297 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
298 in your programs, for future compatibility.
299 - *blush* idle watcher emulation was borked.
300 - AnyEvent::Strict now checks the AE::xxx functions as well.
301 - querying too long domain names would barf deep inside AnyEvent::DNS - since
302 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
303 for too long domain names explicitly and passes an undef to the callback
304 instead (reported by James Bromberger).
305 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
306 - new functions: AnyEvent::postpone, AnyEvent::log.
307 - new module: AnyEvent::Log, for simple logging needs.
308 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
309 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
310 a debug shell.
311 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
312 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
313 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
314 /etc/hosts can be found.
315 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
316 configuration.
317 - env variables can now be specified using AE_xyz instead of
318 PERL_ANYEVENT_xyz, subject to some rules.
319 - lots of new AnyEvent::Debug functionality.
320 - AnyEvent::Strict now checks for modifications of $_ by the callback
321 (by setting it to readonly while executing the callback, if possible).
322 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
323 Invest extra overhead to work around this peculiar behaviour.
324 - split the pure perl backend into a loop module and an impl
325 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
326 - some read types (regex, netstring, json, storable) could
327 errornously keep an AE::Handle object alive on their own
328 (reported by Mohammad Toossi).
329 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
330 supports /etc/hosts overrides.
331 - add more workarounds around child watcher bugs in IO::Async
332 and (newly introduced) in POE.
333 - use glib child watchers and try to work around its limitations,
334 if possible (based on a patch by Kevin Ryde).
335 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
336 - many read types in AnyEvent::Handle would malfunction during
337 transitions to TLS because of a "delete $self->{rbuf}".
338 (reported by Gerald Galster).
339 - improve condvar blocking wait performance (probably only noticable
340 with faster backends), also make it easier for event loops that
341 do not support blocking waits.
342 - PERL_ANYEVENT_MODEL now also allows full module names, although
343 the usefulness of this feature is not yet clear.
344 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
345 - add t/80_ssltext.t.
346 - no longer use AUTOLOAD for the initial loading (for no very good reason).
347 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
348 - do not run event loop detection multiple times when programs call their cached
349 copies of AnyEvent::detect.
350 - parse_hostport now accepts absolute paths als unix domain sockets.
351 - suppress the idiotic warning inside POE by patching the POE::Kernel
352 run flags instead of asking the user to jump through contortions.
353 - add Task::Weaken as dependency (as recommends only at this time).
354 - replace PF_UNSPEC by 0, for good old minix.
355
3566.0 (not officially released, but was on CPAN for about 15 minutes :/)
357
3585.34 Thu May 12 09:59:41 CEST 2011
359 - AE::TLS didn't support both cert_file and key at the same time,
360 due to some copy&paste bug (reported and analyzed by vti).
361 - improve AE::Handle timeout documentation slightly.
362
3635.33 Mon Feb 7 21:51:41 CET 2011
364 - never appeared on CPAN, for unknown reasons.
365 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
366 object when it's already gone (analyzed by Ryan Bullock).
367 - update idna algorithm(s) to UTS#46 version 6.0.0.
368 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
369 uts46 says it's an error, and testsuite makes more sense.
370 - remove leading dots in idn_to_ascii - testsuite says to remove,
371 uts46 says it's an error, and testsuite makes more sense.
372 - fixed a bug in idn ignored character processing.
373 - passes the uts46 6.0.0 testsuite except for three cases which
374 apparently are bugs in the testsuite itself.
375 - passes the uts46 6.0.1 testsuite except for eleven cases which
376 apparently are bugs in the testsuite itself.
377
3785.31 Sun Jan 23 11:49:19 CET 2011
379 - work around a POE bug where POE would first poll for new events and
380 invoke handlers BEFORE handling the message queue.
381 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
382 - stop_read now does just that in TLS mode. see its
383 documentation for caveats.
384 - update to common::sense 3.4 warning set (some warnings in perl impose
385 an arbitrary 32kb limit on regex matches, which affects
386 AnyEvent::Handle).
387 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
388 - new experimental AE::Handle::resettls method.
7 389
85.3 Fri Dec 31 05:48:13 CET 2010 3905.3 Fri Dec 31 05:48:13 CET 2010
9 - major IOAsync patch by Paul Evans - 0.33 is required, as the major 391 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
10 issues with IO::Async have been solved in that version, including 392 issues with IO::Async have been solved in that version, including
11 the availabilty of a default loop! 393 the availabilty of a default loop!
… …
357 in the common case. 739 in the common case.
358 - add AnyEvent::Handle->push_shutdown method. 740 - add AnyEvent::Handle->push_shutdown method.
359 - add an additional error message parameter to AnyEvent::Handle's 741 - add an additional error message parameter to AnyEvent::Handle's
360 on_error callback (for TLS, $! is still available). 742 on_error callback (for TLS, $! is still available).
361 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 743 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
362 - make AnyEvent::Handle more robust against early conenction 744 - make AnyEvent::Handle more robust against early connection
363 failures (during new), and return C<undef> in that case 745 failures (during new), and return C<undef> in that case
364 from the constructor. 746 from the constructor.
365 - AnyEvent::Handle will now try to load only JSON::XS first, 747 - AnyEvent::Handle will now try to load only JSON::XS first,
366 then fall back to JSON. 748 then fall back to JSON.
367 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 749 - format_ipv4/format_ipv6 are now exported by default, for symmetry,

Diff Legend

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