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

Comparing AnyEvent/Changes (file contents):
Revision 1.334 by root, Tue Aug 11 01:18:27 2009 UTC vs.
Revision 1.615 by root, Sun Oct 30 23:03:36 2016 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2
3TODO: document TCP_*** constants
4TODO: add debug/trace logging to some modules
5TODO: use the eval-with-cb-call trick also to speed up JSON?
6TODO: FAQ: common pitfalls? invoke-before-return
7TODO: FAQ: scope-ids? hosts file order?
8TODO: AnyEvent::Handle with AnyEvent::IO?
9TODO: invalid. and localhost. specialcasing inside AEDNS and not AESocket (rfc6761)
10TODO: maybe implement env variable to give hosts precedence
11TODO: more formats for parse_ipv6 (single ipv6 address without port, ...p80, ...#80)
12
13TODO: inet_aton maybe pack C4? no, add a comment why not
14TODO: hosts read multiple times for initial concurrent tcp_connect's?
15
16 - AnyEvent::Handle->keepalive was documented (and defined) twice
17 (found by Matt Merhar).
18
197.13 Sat Sep 17 04:31:49 CEST 2016
20 - Only call tlsext_host_name for non-empty common names (reported
21 by Maxime Soulé).
22 - log a (single) notice message if SNI is not supported.
23 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
24 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
25 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
26 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
27 testcases and the specification and not in the code - the post-9.0.0
28 unicode files have all known problems fixed, so finally the AnyEvent
29 IDNA implementation can pass the full IDNA testsuite - without needing
30 a single fix :)
31 - guarantee (and document) that condvar callbacks will be removed
32 on invocation - important to avoid circular references.
33
347.12 Wed Jan 27 19:12:26 CET 2016
35 - use common name as hostname for tls connects, if Net::SSLeay
36 supports SNI.
37 - fix documentation of tls_autostart read type in AnyEvent::Handle,
38 analyzed by Felix Ostmann.
39
407.11 Thu Jul 16 14:36:00 CEST 2015
41 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
42 addresses (extra "::" at end and similar cases).
43 - add a more explicit warning to AnyEvent::Handle that it doesn't
44 work on files, people keep getting confused.
45 - new function AnyEvent::Socket::tcp_bind.
46 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
47 - aligned ipv6 address formatting with RFC 5952 (by not shortening
48 a single :0: to ::).
49 - added stability canary support.
50
517.09 Sat May 2 16:38:53 CEST 2015
52 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
53 that was renamed to AnyEvent::Log:format_time. uinder its old name
54 (analyzed by Michael Stovenour).
55 - update AnyEvent::DNS fallback resolver addresses:
56 seems google effectively killed most other free dns resolvers,
57 so remove them, but add cable and wireless (ecrc) since it was
58 stable for 20 years or so, official or not, and there should be
59 an alternative to google.
60 - perl5porters broke windows error codes in 5.20, and mapped
61 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
62 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
63 because they were so cool. They probably broke other error codes
64 for no reason, too, but I didn't care for checking, it's too
65 depressing. This version only works around the WSAEWOULDBLOCK
66 issue, because I don't have a nice way to work around the
67 WSAEINPROGRESS bug.
68
697.08 Wed Dec 10 05:27:17 CET 2014
70 - work around a newly introduced bug in Socket 2.011 (an
71 errornous sun_length check) (analyzed by Maxime Soulé).
72 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
73 (analyzed by Ben Magistro).
74 - AnyEvent::Strict will now confess, not croak. This is in line with
75 it being a development/debugging tool.
76 - work around a number of libglib bugs (debug builds of libglib enforce
77 certain undocumented behaviour patterns such as not being able to
78 remove a child watch source after it has fired, which we will try
79 to emulate to avoid "criticals". what where they thinking...).
80 - mention json security issues in AnyEvent::Handle, now that Douglas
81 Crockford has foolishly and incompatibly changed JSON.
82 - changed default dns resolver "max_outstanding" value from 1 to 10,
83 the latter beinfg the intended value all along
84 (reported by Ilya Chesnokov).
85 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
86 (written by Mike Lowell).
87
887.07 Tue Dec 17 17:45:02 CET 2013
89 - the documentation for custom tls verify schemes was wrong. make it agree
90 with the code (reported by Maxime Soulé).
91 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
92 - work around an API change in openssl that could cause wrong tls connection
93 aborts, likely on windows only (analyzed by sten).
94 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
95 endless loop (reported by Dietrich Rebmann).
96 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
97 (patch by Maxime Soulé).
98 - document AnyEvent::Impl::IOAsync::set_loop and
99 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
100 functionality has _always_ been available.
101 - force a toplevel domain name in t/81_hosts.t (analyzed by
102 David Jack Wange Olrik).
103 - document that AnyEvent::Log uses AnyEvent::IO.
104 - warn about AnyEvent::Filesys::Notify performance.
105 - praise the joys of AnyEvent::Fork::*.
106 - time for an =encoding directive.
107 - do no longer use JSON to create a default json coder, use
108 JSON::XS or JSON::PP directly.
109
1107.05 Wed Aug 21 10:38:08 CEST 2013
111 - uts46data.pl couldn't be found due to wrong naming of the file
112 (reported by Fulko Hew).
113 - handle lone \015's properly in AE::Handle's default line read
114 (reported by various people).
115 - untaint ip addresses found in /etc/hosts (patch by José Micó).
116 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
117 fail on first use (reported and testcase by Andrew Whatson).
118 - expose AnyEvent::Log::format_time, and allow users to redefine it.
119 - expose AnyEvent::Log::default_format, and allow redefinition.
120 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
121 - AnyEvent::Debug shell can now run coro shell commands, if available.
122 - t/63* tests were wrongly in MANIFEST.
123 - kernel.org's finger server went MIA, switch to freebsd.org and
124 icculus.org.
125 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
126 to function asynchronously (reported by Daniel Carosone).
127 - hard-disable $^W in most tests, it generates too much garbage output.
128 - use a (hopefully) more future-proof method to emulate common::sense.
129 - upgrade to UTS-46:6.2.0.
130 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
131 documented. should not affect anything.
132
1337.04 Thu Nov 15 02:23:58 CET 2012
134 - debugging code left in.
135
1367.03 Thu Nov 15 02:16:11 CET 2012
137 - AnyEvent::Socket::inet_aton did not work when dns resolution was
138 used to find the addresses (analyzed and patch by Maxime Soulé).
139 - fix a memory leak in the /etc/hosts lookup code when hosts don't
140 resolve and are not in hosts (analyzed by sten).
141
1427.02 Tue Aug 14 04:27:58 CEST 2012
143 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
144 program by Yorhel).
145 - verified that AnyEvent::Socket follows RFC5952.
146 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
147
1487.01 Sun May 13 01:03:17 CEST 2012
149 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
150 not available, instead of throwing an exception.
151 - use File::Spec to get the tmpdir in t/*, to avoid needless
152 failures on (most, not mine :) windows boxes.
153 - new handle read types: tls_detect and tls_autostart.
154
1557.0 Fri Apr 13 06:33:30 CEST 2012
156 - child watchers are broken in POE 1.352 (also many earlier
157 versions) and there seems to be no way to work around it, as POE
158 itself is inherently racy. Document this regression and add a delay
159 in t/68_poe_03_child.t for the time being.
160 - new module AnyEvent::IO, that is a frontend to either a pure-perl
161 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
162 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
163 - load /etc/hosts only when DNS has no answer.
164 - stat /etc/hosts on every access and reload it if it changed.
165 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
166 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
167 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
168 a side effect, they now use true appending as opposed to libc
169 appending, and the file might not have been opened when the function
170 returns.
171 - the default logging level was not properly documented in a variety of
172 places, this has hopefully been rectified.
173 - updated uts46data.pl for unicode 6.1.0.
174 - made log messages generated by AnyEvent submodules not
175 include the package name anymore, as it will be logged
176 by default already.
177 - upgrade to the trick used by common-sense 3.5 to work
178 around extra warning torture/breakage under perl 5.15.x.
179 - log messages by anyevent are now ucfirst, are usually full
180 sentences and do no longer include the package name.
181 - the storable read type would sometimes throw an exception instead
182 of causing EBADMSG (analyzed by Maxime Soulé).
183 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
184 keysize (reported by Paul Howarth).
185 - documented Rocco fabricating statements about AnyEvent and me.
186
1876.14 Tue Jan 31 20:00:24 CET 2012
188 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
189 probably an editing mistake (reported by Darin McBride).
190
1916.13 Thu Jan 12 07:27:01 CET 2012
192 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
193 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
194 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
195 regex-matching inside the callback would change the parameter, despite
196 $1 being dynamically scoped per-block. this perl bug is now being
197 worked around (testcase by Cindy Wang).
198
1996.12 Mon Dec 12 13:21:10 CET 2011
200 - $! was clobbered by subsequent calls in tcp_connect, due to
201 the postpone not saving/restoring it (analyzed by Richard Garnish).
202
2036.11 Tue Nov 22 10:36:05 CET 2011
204 - Tk cannot create windows when tainted unless you set a title, so
205 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
206 - escape any nonprintable/nonascii characters when stringifying
207 backtraces.
208 - log the reason loading an interface module fails to load at level debug.
209
2106.1 Tue Oct 4 19:44:30 CEST 2011
211 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
212 and above), and some messages inside AnyEvent have been elevated
213 to higher log levels to print by default.
214 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
215 - work around yet another signal race bug, newly introduced by POE. My
216 how that module sucks.
217 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
218 support ipv6-only operation even with defaults (also use ::1 instead
219 of 127.0.0.1).
220 - give better guidelines on how to chose a logging level.
221 - AnyEvent::Log can now cap (limit) log message priority by context.
222 - AE::log will now save and restore $! and $@, so callers don't have to.
223 - AE::Strict now checks that registered fds are still valid at regular
224 intervals.
225 - skip frequently called 'require' calls if the module appears loaded.
226 - increase signal latency test timer considerably for poor little osx,
227 which can't be bothered with basic posix functionality such as signal
228 delivery.
229 - mention some debugging aids in the tutorial.
230 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
231 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
232 of the context as title.
233 - AnyEvent::Strict would not restore read-write access to $_ when a callback
234 died - this is a user bug, but since AE::Strict exists to help dbeugging...
235 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
236 to work with it - the author is well aware of the issues but refuses
237 to fix them. This extreme step has been taken because the module actively
238 attacks the goals of AnyEvent and makes it harder for module authors to use
239 AnyEvent.
240
2416.02 Fri Aug 26 20:08:31 CEST 2011
242 - log_to_syslog now accepts facility strings.
243 - log_to_syslog would not allow facilities - "$facility|$level"
244 *literally* means that string. sheesh.
245 - the CPAN parser got confused and thought we provide Sys::Syslog, try
246 to work around.
247 - renamed AnyEvent::Impl::FLTK2 to FLTK.
248
2496.01 Fri Aug 26 07:04:11 CEST 2011
250 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
251 the TTL as fourth element - this affects ->resolve, ->request
252 and ->dns_unpack indirectly. Convenience functions such as
253 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
254 AnyEvent::Socket.
255 - INCOMPATIBLE CHANGE: backend authors now should not implement
256 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
257 methods.
258 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
259 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
260 - DEPRECIATION: please change
261 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
262 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
263 in your programs, for future compatibility.
264 - *blush* idle watcher emulation was borked.
265 - AnyEvent::Strict now checks the AE::xxx functions as well.
266 - querying too long domain names would barf deep inside AnyEvent::DNS - since
267 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
268 for too long domain names explicitly and passes an undef to the callback
269 instead (reported by James Bromberger).
270 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
271 - new functions: AnyEvent::postpone, AnyEvent::log.
272 - new module: AnyEvent::Log, for simple logging needs.
273 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
274 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
275 a debug shell.
276 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
277 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
278 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
279 /etc/hosts can be found.
280 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
281 configuration.
282 - env variables can now be specified using AE_xyz instead of
283 PERL_ANYEVENT_xyz, subject to some rules.
284 - lots of new AnyEvent::Debug functionality.
285 - AnyEvent::Strict now checks for modifications of $_ by the callback
286 (by setting it to readonly while executing the callback, if possible).
287 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
288 Invest extra overhead to work around this peculiar behaviour.
289 - split the pure perl backend into a loop module and an impl
290 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
291 - some read types (regex, netstring, json, storable) could
292 errornously keep an AE::Handle object alive on their own
293 (reported by Mohammad Toossi).
294 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
295 supports /etc/hosts overrides.
296 - add more workarounds around child watcher bugs in IO::Async
297 and (newly introduced) in POE.
298 - use glib child watchers and try to work around its limitations,
299 if possible (based on a patch by Kevin Ryde).
300 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
301 - many read types in AnyEvent::Handle would malfunction during
302 transitions to TLS because of a "delete $self->{rbuf}".
303 (reported by Gerald Galster).
304 - improve condvar blocking wait performance (probably only noticable
305 with faster backends), also make it easier for event loops that
306 do not support blocking waits.
307 - PERL_ANYEVENT_MODEL now also allows full module names, although
308 the usefulness of this feature is not yet clear.
309 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
310 - add t/80_ssltext.t.
311 - no longer use AUTOLOAD for the initial loading (for no very good reason).
312 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
313 - do not run event loop detection multiple times when programs call their cached
314 copies of AnyEvent::detect.
315 - parse_hostport now accepts absolute paths als unix domain sockets.
316 - suppress the idiotic warning inside POE by patching the POE::Kernel
317 run flags instead of asking the user to jump through contortions.
318 - add Task::Weaken as dependency (as recommends only at this time).
319 - replace PF_UNSPEC by 0, for good old minix.
320
3216.0 (not officially released, but was on CPAN for about 15 minutes :/)
322
3235.34 Thu May 12 09:59:41 CEST 2011
324 - AE::TLS didn't support both cert_file and key at the same time,
325 due to some copy&paste bug (reported and analyzed by vti).
326 - improve AE::Handle timeout documentation slightly.
327
3285.33 Mon Feb 7 21:51:41 CET 2011
329 - never appeared on CPAN, for unknown reasons.
330 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
331 object when it's already gone (analyzed by Ryan Bullock).
332 - update idna algorithm(s) to UTS#46 version 6.0.0.
333 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
334 uts46 says it's an error, and testsuite makes more sense.
335 - remove leading dots in idn_to_ascii - testsuite says to remove,
336 uts46 says it's an error, and testsuite makes more sense.
337 - fixed a bug in idn ignored character processing.
338 - passes the uts46 6.0.0 testsuite except for three cases which
339 apparently are bugs in the testsuite itself.
340 - passes the uts46 6.0.1 testsuite except for eleven cases which
341 apparently are bugs in the testsuite itself.
342
3435.31 Sun Jan 23 11:49:19 CET 2011
344 - work around a POE bug where POE would first poll for new events and
345 invoke handlers BEFORE handling the message queue.
346 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
347 - stop_read now does just that in TLS mode. see its
348 documentation for caveats.
349 - update to common::sense 3.4 warning set (some warnings in perl impose
350 an arbitrary 32kb limit on regex matches, which affects
351 AnyEvent::Handle).
352 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
353 - new experimental AE::Handle::resettls method.
354
3555.3 Fri Dec 31 05:48:13 CET 2010
356 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
357 issues with IO::Async have been solved in that version, including
358 the availabilty of a default loop!
359 - implement new wbuf_max parameter for AnyEvent::Handle.
360 - added Cocoa::EventLoop backend by Daisuke Murase.
361 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
362 - clarify handle on_prepare documentation.
363 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
364 to avoid leaking them to any child processes.
365 - try to force files into the arch-dependend lib directory, for the
366 half-"turly obsessive compulsive person"(s) that depend on it.
367
3685.29 Sun Dec 5 10:49:21 CET 2010
369 - convert EV backend to EV 4.00 API (so better upgrade EV too).
370 - AE::Handle: implement dynamic read_size adjustment, add
371 max_read_size parameter, reduce default read_size to 2048.
372 - add some FAQ entries for servers and/or on_eof vs. on_error.
373 - work around OS bugs (cygwin again) when lingering in AE::Handle.
374 - reduce memory usage slightly when connect was used in AE::Handle.
375 - wrok around more segfaults in Tk.
376 - document tls_ctx => undef as valid.
377 - detect WNOHANG value at compiletime.
378 - include some probably linux-only support for building AnyEvent
379 as part of the perl core.
380 - improve compatibility of t/08_idna.t to perl 5.8.
381 - make AnyEvent::Strict truly optional (in case somebody builds a
382 minimal perl...).
383
3845.28 Wed Oct 13 04:14:23 CEST 2010
385 - due to a glitch, AnyEvent's internal getprotobyname wasn't
386 actually used in tcp_server.
387 - implement AnyEvent::CondVar->new method, which was documented but
388 missing.
389 - support autoloading of read types in unshift_read.
390 - AnyEvent::Strict: do the test for invalid fh arguments differently,
391 resulting in more informative error messages.
392 - new AnyEvent::FAQ manpage.
393 - clarify that you can actually call push_read etc. in on_read.
394 - put keys and certs in /tmp, not into the current directory,
395 and document this in the SECURITY CONSIDERATIONS sections
396 of AnyEvent::TLS.
397 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
398 requested by Ron Isaacson.
399 - document $AnyEvent::DNS::RESOLVER variable.
400 - incorporated major typo patches by Abhijit Menon-Sen.
401 - AnyEvent::Handle now croaks when negative timeouts are passed.
402 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
403 undocumented, but who knows when they might come in handy.
404
4055.271 Tue Jun 8 12:05:46 CEST 2010
406 - backport to perl 5.8.x.
407
4085.27 Sun Jun 6 12:12:05 CEST 2010
409 - postpone differently in AnyEvent::Socket now, as
410 when not, canceling the connection attempt might fail
411 (found by Felix Antonius Wilhelm Ostmann).
412 - explicitly check for non-stream sockets in AE::Handle, too many
413 clueless people fell into the trap of this somehow working.
414 - simplified and reworked the "OTHER MODULES" section.
415 - better/more condvar examples.
416
4175.261 Wed Apr 28 16:13:36 CEST 2010
418 - AF_INET6 was not properly used from Socket6 during configuration
419 time (found by Andy Grundman).
420
4215.26 Mon Apr 12 04:49:35 CEST 2010
422 - don't generate spurious readyness notifications when select
423 returns EINTR in the pure perl backend (analysed and testcase
424 by Bas Denissen).
425 - give same examples in SYNOPSIS sections of AnyEvent and AE.
426 - provider faster implementations for AE::time/now/now_update for
427 EV and Perl backends.
428 - provide faster AE::cv and AE::time implementations for all
429 backends.
430 - no longer support lower/mixed-case signal names (this was
431 never documented nor universally supported).
432 - some more memory µ-optimisations, and somewhat less messy
433 code to cope with both AnyEvent and AE APIs.
434
4355.251 Sat Mar 13 00:58:21 CET 2010
436 - make test failed if it was run before make install on perl <=
437 5.8.8, and older versions of perl make our live very difficult,
438 so write constants.pl during Makefile.PL time.
439
4405.25 Sat Mar 13 00:23:14 CET 2010
441 - fix a race condition in AnyEvent::Handle that would cause
442 a "bio_read: unsupported method" error (in _another_
443 ssl connection) after user code threw an exception.
444 - added AnyEvent::Handle->destroyed method.
445 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
446 - the AnyEvent::Util::fh_nonblocking function worked only
447 by lucky accident on win32.
448 - smaller and faster AnyEvent::Util::fh_nonblocking.
449 - when the (required!) Time::HiRes module is missing, AnyEvent
450 did not fall back to built-in time properly.
451 - do not load Fcntl at runtime, saving memory and loading time.
452 - precompile a number of constants and use them instead of runtime
453 detection and eval.
454 - free detection code after detection and similar memory optimisations.
455 - Perl backend timer interval best effort drift has been improved
456 (same algorithm as EV).
457 - update unicode idna mapping table.
458
4595.24 Tue Jan 5 11:39:43 CET 2010
460 - cygwin never reports errors from failed connects. 1.5 just gives
461 you continous readyness and EAGAIN, 1.7 is even more broken and
462 just hangs. work around both issues in a cygwin-specific hack.
463 - improve idle watcher documentation slightly.
464
4655.23 Sun Dec 20 23:48:00 CET 2009
466 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
467 - implement punycode_encode/decode, idn_nameprep,
468 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
469 - provide $AE::VERSION.
470 - removed traces of "no strict 'refs'".
471
4725.22 Sat Dec 5 03:51:13 CET 2009
473 - downgrade-or-fail in AnyEvent::Handle::push_write, to
474 diagnose encoding failures earlier and more succinctly.
475 (this works around bugs in perl, throwing away encoding info
476 when passing scalar data to extensions).
477 - add more examples to AnyEvent::Socket manpage.
478 - upgrade internal warning set to the same as common::sense 2.03.
479 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
480 (John Beppu).
481 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
482 - add google public dns servers to fallback server set and make sure
483 we load-balance properly between the three sets. also add all
484 fallback dns servers, not just a random one, to each dns config.
485
4865.21 Thu Nov 19 02:48:47 CET 2009
487 - fix a problem where socket constants were called with parameters
488 (spotted by David Friedland).
489 - fork_call never use'd POSIX (reported by Daisuke Maki).
490 - improve perl 5.6 compatibility further (but it still won't work
491 unless you rip out everything but the core).
492 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
493 uses 10MB of RAM, but doesn't open a console window. *sigh*.
494
4955.202 Wed Oct 14 22:35:44 CEST 2009
496 - AnyEvent::DNS would unexpectedly clobber $_ under windows
497 (analysed by Matthias Waldorf).
498 - AnyEvent::Handle::run_cmd can now pass the PID of the
499 newly-created process, which is much less useful than it might
500 sound (based on patch by Yann Kerherve).
501
5025.201 Tue Sep 29 12:09:25 CEST 2009
503 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
504 (reported by Torsten Foertsch).
505 - common::sense 2.0 could cause tcp_server to throw an exception
506 (analysed by elmex).
507
5085.2 Mon Sep 14 07:04:49 CEST 2009
509 - INCOMPATIBLE CHANGE: do no longer support register_read_type
510 and register_write_type in AnyEvent::Handle, instead support
511 package names (the facility was mostly abused).
512 - implement "packagename-as-read/write type" support in
513 AnyEvent::Handle.
514 - AnyEvent::Handle: new options "keepalive" and "oobinline".
515 - oobinline set by default to avoid security issues.
516 - the pure-perl event loop backend wrongly detected times() fallback
517 support (spotted by Pavel Boldin).
518
5195.12 Tue Sep 1 20:26:50 CEST 2009
520 - be more lenient when parsing resolv.conf files, as some people
521 use hashmarks as comment indicator inside directives (reported
522 by Michael S. Fischer).
523 - use same set of warnings as common::sense 2.0.
524 - fix a potential 32 bit overflow issue due to perl having problems
525 with large hex constants in the Perl backend.
526 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
527 by default.
528 - fix some prototypes.
529 - (optionally) require Async::Interrupt 1.02.
530 - replace opendns fallbacks by verizon/level3 public dns servers.
531
5325.112 Fri Aug 21 13:59:04 CEST 2009
533 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
534 using starttls with an illegal/uncreatable tls context.
535 - port to broken win32 perls.
536
5375.111 Sun Aug 16 18:44:36 CEST 2009
538 - perl errornously requires sockaddr_un structures returned by the
539 kernel to have a specific length. work around this bug by padding
540 sockaddr_un structures (found to fail on OS X, as perl only uses
541 the correct code for GNU/Linux, but likely to fail on other
542 OSes as well).
543
5445.11 Wed Aug 12 17:49:37 CEST 2009
545 - fixed a bug in run_cmd with input from filename.
546 - create a json coder object when none is given in write direction
547 to reduce dependency on JSON.
2 548
35.1 Tue Aug 11 03:17:32 CEST 2009 5495.1 Tue Aug 11 03:17:32 CEST 2009
4 - tcp_connect now reports the correct error in $!, not an unrelated 550 - tcp_connect now reports the correct error in $!, not an unrelated
5 one (was broken since 4.91). 551 one (was broken since 4.91).
6 - AE::Handle did not properly clear rtimeout/wtimeout watchers 552 - AE::Handle did not properly clear rtimeout/wtimeout watchers
… …
60 - new function: AnyEvent::Socket::format_hostport. 606 - new function: AnyEvent::Socket::format_hostport.
61 607
624.881 Tue Jul 28 12:51:53 CEST 2009 6084.881 Tue Jul 28 12:51:53 CEST 2009
63 - work around a bug in local in pre-5.10 perls, causing 609 - work around a bug in local in pre-5.10 perls, causing
64 AnyEvent::Handle to recurse when it should not 610 AnyEvent::Handle to recurse when it should not
65 (analyzed by elmex). 611 (analysed by elmex).
66 612
674.88 Tue Jul 28 04:04:37 CEST 2009 6134.88 Tue Jul 28 04:04:37 CEST 2009
68 - re-bless the handle into a dummy package after calling 614 - re-bless the handle into a dummy package after calling
69 AnyEvent::Handle::destroy, so the user does not need to check 615 AnyEvent::Handle::destroy, so the user does not need to check
70 for errors after every push_write etc. 616 for errors after every push_write etc.
… …
158 in the common case. 704 in the common case.
159 - add AnyEvent::Handle->push_shutdown method. 705 - add AnyEvent::Handle->push_shutdown method.
160 - add an additional error message parameter to AnyEvent::Handle's 706 - add an additional error message parameter to AnyEvent::Handle's
161 on_error callback (for TLS, $! is still available). 707 on_error callback (for TLS, $! is still available).
162 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 708 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
163 - make AnyEvent::Handle more robust against early conenction 709 - make AnyEvent::Handle more robust against early connection
164 failures (during new), and return C<undef> in that case 710 failures (during new), and return C<undef> in that case
165 from the constructor. 711 from the constructor.
166 - AnyEvent::Handle will now try to load only JSON::XS first, 712 - AnyEvent::Handle will now try to load only JSON::XS first,
167 then fall back to JSON. 713 then fall back to JSON.
168 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 714 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
2924.3 Fri Oct 3 09:18:43 CEST 2008 8384.3 Fri Oct 3 09:18:43 CEST 2008
293 - AnyEvent will now install a no-op signal handler for SIGPIPE, 839 - AnyEvent will now install a no-op signal handler for SIGPIPE,
294 unless one has been installed already. 840 unless one has been installed already.
295 - warn about not loaidng AnyEvent::Impl::POE early enough 841 - warn about not loaidng AnyEvent::Impl::POE early enough
296 (patch by Adam Rosenstein). 842 (patch by Adam Rosenstein).
297 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 843 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
298 - fix a bug in where in SSL connect mode, the 844 - fix a bug in where in SSL connect mode, the
299 client would first wait for some data by the server and 845 client would first wait for some data by the server and
300 otherwise hang. (reported and analyzed in an absolutely 846 otherwise hang. (reported and analysed in an absolutely
301 exemplary manner by Adam Rosenstein). 847 exemplary manner by Adam Rosenstein).
302 - fix a bug in where SSL EOF would not be treated 848 - fix a bug in where SSL EOF would not be treated
303 as stream EOF, putting the connection into a hung state. 849 as stream EOF, putting the connection into a hung state.
304 (reported and analyzed in an absolutely exemplary manner by Adam 850 (reported and analysed in an absolutely exemplary manner by Adam
305 Rosenstein). 851 Rosenstein).
306 - fix a potential segfault when the TLS context 852 - fix a potential segfault when the TLS context
307 would go missing in a read callback (e.g. due to stoptls) - 853 would go missing in a read callback (e.g. due to stoptls) -
308 Net::SSLeay of course makes no type checking whatsoever. 854 Net::SSLeay of course makes no type checking whatsoever.
309 - AnyEvent::Handle will not stop the read watcher in TLS mode. 855 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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