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

Comparing AnyEvent/Changes (file contents):
Revision 1.414 by root, Sat Oct 16 02:01:54 2010 UTC vs.
Revision 1.636 by root, Fri Jul 19 16:01:47 2019 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: document TCP_*** constants 37.16 Fri Jul 19 18:00:21 CEST 2019
4TODO: catch AE::xxx for anyevent struct and anyevent debug 4 - add ffdhe group dh parameters from RFC 7919, and use ffdhe3072
5TODO: anyevent::trace? or debug tracing all watchers? 5 as new default, instead of schmorp1539.
6 - AnyEvent::Log did not reassess logging status of
7 AnyEvent::Log::logger's when contexts were changed with
8 ->attach/detach/slaves, causing them to not log even though a
9 recent attach should have caused them to log.
10 - added some more logging configuration examples.
11 - mention RFC 8482 in AnyEvent::DNS.
6 12
137.15 Tue Feb 26 03:07:42 CET 2019
14 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
15 separate major and minor versions, while code passed only a single
16 value. This version follows the documentation and now passes separate
17 major and minor values.
18 - work around Net::SSLeay not having been ported to openssl 1.1, but many
19 distributions compiling it against openssl 1.1, which unfortunately
20 succeeds and results in a very broken module.
21 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
22 in the __ member, to allow decoding of undecodable resouce records
23 containing compressed domian names.
24 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
25 addresses. It now does and converts them to ipv4 mapped addresses.
26 - support CAA records, based on patch by Steve Atkins.
27 - add freenom and cloudflare nameservers as dns fallback.
28 - AnyEvent::Strict would not properly ward against io watchers
29 on files when the handle passed was a file descriptor.
30 - document "internal" variables used by the dns en-/decoder to allow
31 enterprising users to extend them in a semi-official way.
32
337.14 Sat Jun 24 01:44:19 CEST 2017
34 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
35 by Paul Howarth and apparently tracked down by Mike McCauley).
36 - AnyEvent::Handle->keepalive was documented (and defined) twice
37 (found by Matt Merhar).
38 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
39 listening socket unless a guard is used. Change this so that no
40 cleanup will be performed unless a guard is used and document this more
41 clearly.
42 - make tcp_bind/tcp_server error messages more regular.
43 - work around latest perl upstream breakage - this time, perl5porters were
44 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
45 stupidly apply anything that has "security fix" stamped on it, it's likely
46 going to be a problem in practise (and working around it is trivial on
47 windows and unix).
48 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
49 to constants.
50 - update warnings to common::sense 3.74 standards.
51
527.13 Sat Sep 17 04:31:49 CEST 2016
53 - Only call tlsext_host_name for non-empty common names (reported
54 by Maxime Soulé).
55 - log a (single) notice message if SNI is not supported.
56 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
57 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
58 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
59 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
60 testcases and the specification and not in the code - the post-9.0.0
61 unicode files have all known problems fixed, so finally the AnyEvent
62 IDNA implementation can pass the full IDNA testsuite - without needing
63 a single fix :)
64 - guarantee (and document) that condvar callbacks will be removed
65 on invocation - important to avoid circular references.
66
677.12 Wed Jan 27 19:12:26 CET 2016
68 - use common name as hostname for tls connects, if Net::SSLeay
69 supports SNI.
70 - fix documentation of tls_autostart read type in AnyEvent::Handle,
71 analyzed by Felix Ostmann.
72
737.11 Thu Jul 16 14:36:00 CEST 2015
74 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
75 addresses (extra "::" at end and similar cases).
76 - add a more explicit warning to AnyEvent::Handle that it doesn't
77 work on files, people keep getting confused.
78 - new function AnyEvent::Socket::tcp_bind.
79 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
80 - aligned ipv6 address formatting with RFC 5952 (by not shortening
81 a single :0: to ::).
82 - added stability canary support.
83
847.09 Sat May 2 16:38:53 CEST 2015
85 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
86 that was renamed to AnyEvent::Log:format_time. uinder its old name
87 (analyzed by Michael Stovenour).
88 - update AnyEvent::DNS fallback resolver addresses:
89 seems google effectively killed most other free dns resolvers,
90 so remove them, but add cable and wireless (ecrc) since it was
91 stable for 20 years or so, official or not, and there should be
92 an alternative to google.
93 - perl5porters broke windows error codes in 5.20, and mapped
94 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
95 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
96 because they were so cool. They probably broke other error codes
97 for no reason, too, but I didn't care for checking, it's too
98 depressing. This version only works around the WSAEWOULDBLOCK
99 issue, because I don't have a nice way to work around the
100 WSAEINPROGRESS bug.
101
1027.08 Wed Dec 10 05:27:17 CET 2014
103 - work around a newly introduced bug in Socket 2.011 (an
104 errornous sun_length check) (analyzed by Maxime Soulé).
105 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
106 (analyzed by Ben Magistro).
107 - AnyEvent::Strict will now confess, not croak. This is in line with
108 it being a development/debugging tool.
109 - work around a number of libglib bugs (debug builds of libglib enforce
110 certain undocumented behaviour patterns such as not being able to
111 remove a child watch source after it has fired, which we will try
112 to emulate to avoid "criticals". what where they thinking...).
113 - mention json security issues in AnyEvent::Handle, now that Douglas
114 Crockford has foolishly and incompatibly changed JSON.
115 - changed default dns resolver "max_outstanding" value from 1 to 10,
116 the latter beinfg the intended value all along
117 (reported by Ilya Chesnokov).
118 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
119 (written by Mike Lowell).
120
1217.07 Tue Dec 17 17:45:02 CET 2013
122 - the documentation for custom tls verify schemes was wrong. make it agree
123 with the code (reported by Maxime Soulé).
124 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
125 - work around an API change in openssl that could cause wrong tls connection
126 aborts, likely on windows only (analyzed by sten).
127 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
128 endless loop (reported by Dietrich Rebmann).
129 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
130 (patch by Maxime Soulé).
131 - document AnyEvent::Impl::IOAsync::set_loop and
132 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
133 functionality has _always_ been available.
134 - force a toplevel domain name in t/81_hosts.t (analyzed by
135 David Jack Wange Olrik).
136 - document that AnyEvent::Log uses AnyEvent::IO.
137 - warn about AnyEvent::Filesys::Notify performance.
138 - praise the joys of AnyEvent::Fork::*.
139 - time for an =encoding directive.
140 - do no longer use JSON to create a default json coder, use
141 JSON::XS or JSON::PP directly.
142
1437.05 Wed Aug 21 10:38:08 CEST 2013
144 - uts46data.pl couldn't be found due to wrong naming of the file
145 (reported by Fulko Hew).
146 - handle lone \015's properly in AE::Handle's default line read
147 (reported by various people).
148 - untaint ip addresses found in /etc/hosts (patch by José Micó).
149 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
150 fail on first use (reported and testcase by Andrew Whatson).
151 - expose AnyEvent::Log::format_time, and allow users to redefine it.
152 - expose AnyEvent::Log::default_format, and allow redefinition.
153 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
154 - AnyEvent::Debug shell can now run coro shell commands, if available.
155 - t/63* tests were wrongly in MANIFEST.
156 - kernel.org's finger server went MIA, switch to freebsd.org and
157 icculus.org.
158 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
159 to function asynchronously (reported by Daniel Carosone).
160 - hard-disable $^W in most tests, it generates too much garbage output.
161 - use a (hopefully) more future-proof method to emulate common::sense.
162 - upgrade to UTS-46:6.2.0.
163 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
164 documented. should not affect anything.
165
1667.04 Thu Nov 15 02:23:58 CET 2012
167 - debugging code left in.
168
1697.03 Thu Nov 15 02:16:11 CET 2012
170 - AnyEvent::Socket::inet_aton did not work when dns resolution was
171 used to find the addresses (analyzed and patch by Maxime Soulé).
172 - fix a memory leak in the /etc/hosts lookup code when hosts don't
173 resolve and are not in hosts (analyzed by sten).
174
1757.02 Tue Aug 14 04:27:58 CEST 2012
176 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
177 program by Yorhel).
178 - verified that AnyEvent::Socket follows RFC5952.
179 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
180
1817.01 Sun May 13 01:03:17 CEST 2012
182 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
183 not available, instead of throwing an exception.
184 - use File::Spec to get the tmpdir in t/*, to avoid needless
185 failures on (most, not mine :) windows boxes.
186 - new handle read types: tls_detect and tls_autostart.
187
1887.0 Fri Apr 13 06:33:30 CEST 2012
189 - child watchers are broken in POE 1.352 (also many earlier
190 versions) and there seems to be no way to work around it, as POE
191 itself is inherently racy. Document this regression and add a delay
192 in t/68_poe_03_child.t for the time being.
193 - new module AnyEvent::IO, that is a frontend to either a pure-perl
194 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
195 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
196 - load /etc/hosts only when DNS has no answer.
197 - stat /etc/hosts on every access and reload it if it changed.
198 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
199 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
200 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
201 a side effect, they now use true appending as opposed to libc
202 appending, and the file might not have been opened when the function
203 returns.
204 - the default logging level was not properly documented in a variety of
205 places, this has hopefully been rectified.
206 - updated uts46data.pl for unicode 6.1.0.
207 - made log messages generated by AnyEvent submodules not
208 include the package name anymore, as it will be logged
209 by default already.
210 - upgrade to the trick used by common-sense 3.5 to work
211 around extra warning torture/breakage under perl 5.15.x.
212 - log messages by anyevent are now ucfirst, are usually full
213 sentences and do no longer include the package name.
214 - the storable read type would sometimes throw an exception instead
215 of causing EBADMSG (analyzed by Maxime Soulé).
216 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
217 keysize (reported by Paul Howarth).
218 - documented Rocco fabricating statements about AnyEvent and me.
219
2206.14 Tue Jan 31 20:00:24 CET 2012
221 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
222 probably an editing mistake (reported by Darin McBride).
223
2246.13 Thu Jan 12 07:27:01 CET 2012
225 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
226 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
227 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
228 regex-matching inside the callback would change the parameter, despite
229 $1 being dynamically scoped per-block. this perl bug is now being
230 worked around (testcase by Cindy Wang).
231
2326.12 Mon Dec 12 13:21:10 CET 2011
233 - $! was clobbered by subsequent calls in tcp_connect, due to
234 the postpone not saving/restoring it (analyzed by Richard Garnish).
235
2366.11 Tue Nov 22 10:36:05 CET 2011
237 - Tk cannot create windows when tainted unless you set a title, so
238 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
239 - escape any nonprintable/nonascii characters when stringifying
240 backtraces.
241 - log the reason loading an interface module fails to load at level debug.
242
2436.1 Tue Oct 4 19:44:30 CEST 2011
244 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
245 and above), and some messages inside AnyEvent have been elevated
246 to higher log levels to print by default.
247 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
248 - work around yet another signal race bug, newly introduced by POE. My
249 how that module sucks.
250 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
251 support ipv6-only operation even with defaults (also use ::1 instead
252 of 127.0.0.1).
253 - give better guidelines on how to chose a logging level.
254 - AnyEvent::Log can now cap (limit) log message priority by context.
255 - AE::log will now save and restore $! and $@, so callers don't have to.
256 - AE::Strict now checks that registered fds are still valid at regular
257 intervals.
258 - skip frequently called 'require' calls if the module appears loaded.
259 - increase signal latency test timer considerably for poor little osx,
260 which can't be bothered with basic posix functionality such as signal
261 delivery.
262 - mention some debugging aids in the tutorial.
263 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
264 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
265 of the context as title.
266 - AnyEvent::Strict would not restore read-write access to $_ when a callback
267 died - this is a user bug, but since AE::Strict exists to help dbeugging...
268 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
269 to work with it - the author is well aware of the issues but refuses
270 to fix them. This extreme step has been taken because the module actively
271 attacks the goals of AnyEvent and makes it harder for module authors to use
272 AnyEvent.
273
2746.02 Fri Aug 26 20:08:31 CEST 2011
275 - log_to_syslog now accepts facility strings.
276 - log_to_syslog would not allow facilities - "$facility|$level"
277 *literally* means that string. sheesh.
278 - the CPAN parser got confused and thought we provide Sys::Syslog, try
279 to work around.
280 - renamed AnyEvent::Impl::FLTK2 to FLTK.
281
2826.01 Fri Aug 26 07:04:11 CEST 2011
283 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
284 the TTL as fourth element - this affects ->resolve, ->request
285 and ->dns_unpack indirectly. Convenience functions such as
286 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
287 AnyEvent::Socket.
288 - INCOMPATIBLE CHANGE: backend authors now should not implement
289 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
290 methods.
291 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
292 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
293 - DEPRECIATION: please change
294 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
295 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
296 in your programs, for future compatibility.
297 - *blush* idle watcher emulation was borked.
298 - AnyEvent::Strict now checks the AE::xxx functions as well.
299 - querying too long domain names would barf deep inside AnyEvent::DNS - since
300 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
301 for too long domain names explicitly and passes an undef to the callback
302 instead (reported by James Bromberger).
303 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
304 - new functions: AnyEvent::postpone, AnyEvent::log.
305 - new module: AnyEvent::Log, for simple logging needs.
306 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
307 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
308 a debug shell.
309 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
310 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
311 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
312 /etc/hosts can be found.
313 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
314 configuration.
315 - env variables can now be specified using AE_xyz instead of
316 PERL_ANYEVENT_xyz, subject to some rules.
317 - lots of new AnyEvent::Debug functionality.
318 - AnyEvent::Strict now checks for modifications of $_ by the callback
319 (by setting it to readonly while executing the callback, if possible).
320 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
321 Invest extra overhead to work around this peculiar behaviour.
322 - split the pure perl backend into a loop module and an impl
323 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
324 - some read types (regex, netstring, json, storable) could
325 errornously keep an AE::Handle object alive on their own
326 (reported by Mohammad Toossi).
327 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
328 supports /etc/hosts overrides.
329 - add more workarounds around child watcher bugs in IO::Async
330 and (newly introduced) in POE.
331 - use glib child watchers and try to work around its limitations,
332 if possible (based on a patch by Kevin Ryde).
333 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
334 - many read types in AnyEvent::Handle would malfunction during
335 transitions to TLS because of a "delete $self->{rbuf}".
336 (reported by Gerald Galster).
337 - improve condvar blocking wait performance (probably only noticable
338 with faster backends), also make it easier for event loops that
339 do not support blocking waits.
340 - PERL_ANYEVENT_MODEL now also allows full module names, although
341 the usefulness of this feature is not yet clear.
342 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
343 - add t/80_ssltext.t.
344 - no longer use AUTOLOAD for the initial loading (for no very good reason).
345 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
346 - do not run event loop detection multiple times when programs call their cached
347 copies of AnyEvent::detect.
348 - parse_hostport now accepts absolute paths als unix domain sockets.
349 - suppress the idiotic warning inside POE by patching the POE::Kernel
350 run flags instead of asking the user to jump through contortions.
351 - add Task::Weaken as dependency (as recommends only at this time).
352 - replace PF_UNSPEC by 0, for good old minix.
353
3546.0 (not officially released, but was on CPAN for about 15 minutes :/)
355
3565.34 Thu May 12 09:59:41 CEST 2011
357 - AE::TLS didn't support both cert_file and key at the same time,
358 due to some copy&paste bug (reported and analyzed by vti).
359 - improve AE::Handle timeout documentation slightly.
360
3615.33 Mon Feb 7 21:51:41 CET 2011
362 - never appeared on CPAN, for unknown reasons.
363 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
364 object when it's already gone (analyzed by Ryan Bullock).
365 - update idna algorithm(s) to UTS#46 version 6.0.0.
366 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
367 uts46 says it's an error, and testsuite makes more sense.
368 - remove leading dots in idn_to_ascii - testsuite says to remove,
369 uts46 says it's an error, and testsuite makes more sense.
370 - fixed a bug in idn ignored character processing.
371 - passes the uts46 6.0.0 testsuite except for three cases which
372 apparently are bugs in the testsuite itself.
373 - passes the uts46 6.0.1 testsuite except for eleven cases which
374 apparently are bugs in the testsuite itself.
375
3765.31 Sun Jan 23 11:49:19 CET 2011
377 - work around a POE bug where POE would first poll for new events and
378 invoke handlers BEFORE handling the message queue.
379 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
380 - stop_read now does just that in TLS mode. see its
381 documentation for caveats.
382 - update to common::sense 3.4 warning set (some warnings in perl impose
383 an arbitrary 32kb limit on regex matches, which affects
384 AnyEvent::Handle).
385 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
386 - new experimental AE::Handle::resettls method.
387
3885.3 Fri Dec 31 05:48:13 CET 2010
389 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
390 issues with IO::Async have been solved in that version, including
391 the availabilty of a default loop!
392 - implement new wbuf_max parameter for AnyEvent::Handle.
393 - added Cocoa::EventLoop backend by Daisuke Murase.
394 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
395 - clarify handle on_prepare documentation.
396 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
397 to avoid leaking them to any child processes.
398 - try to force files into the arch-dependend lib directory, for the
399 half-"turly obsessive compulsive person"(s) that depend on it.
400
4015.29 Sun Dec 5 10:49:21 CET 2010
402 - convert EV backend to EV 4.00 API (so better upgrade EV too).
403 - AE::Handle: implement dynamic read_size adjustment, add
404 max_read_size parameter, reduce default read_size to 2048.
405 - add some FAQ entries for servers and/or on_eof vs. on_error.
7 - work around OS bugs (cygwin again) when lingering in AE::Handle. 406 - work around OS bugs (cygwin again) when lingering in AE::Handle.
407 - reduce memory usage slightly when connect was used in AE::Handle.
408 - wrok around more segfaults in Tk.
409 - document tls_ctx => undef as valid.
410 - detect WNOHANG value at compiletime.
411 - include some probably linux-only support for building AnyEvent
412 as part of the perl core.
413 - improve compatibility of t/08_idna.t to perl 5.8.
414 - make AnyEvent::Strict truly optional (in case somebody builds a
415 minimal perl...).
8 416
95.28 Wed Oct 13 04:14:23 CEST 2010 4175.28 Wed Oct 13 04:14:23 CEST 2010
10 - due to a glitch, AnyEvent's internal getprotobyname wasn't 418 - due to a glitch, AnyEvent's internal getprotobyname wasn't
11 actually used in tcp_server. 419 actually used in tcp_server.
12 - implement AnyEvent::CondVar->new method, which was documented but 420 - implement AnyEvent::CondVar->new method, which was documented but
… …
329 in the common case. 737 in the common case.
330 - add AnyEvent::Handle->push_shutdown method. 738 - add AnyEvent::Handle->push_shutdown method.
331 - add an additional error message parameter to AnyEvent::Handle's 739 - add an additional error message parameter to AnyEvent::Handle's
332 on_error callback (for TLS, $! is still available). 740 on_error callback (for TLS, $! is still available).
333 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 741 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
334 - make AnyEvent::Handle more robust against early conenction 742 - make AnyEvent::Handle more robust against early connection
335 failures (during new), and return C<undef> in that case 743 failures (during new), and return C<undef> in that case
336 from the constructor. 744 from the constructor.
337 - AnyEvent::Handle will now try to load only JSON::XS first, 745 - AnyEvent::Handle will now try to load only JSON::XS first,
338 then fall back to JSON. 746 then fall back to JSON.
339 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 747 - format_ipv4/format_ipv6 are now exported by default, for symmetry,

Diff Legend

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