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

Comparing AnyEvent/Changes (file contents):
Revision 1.327 by root, Sun Aug 9 10:53:33 2009 UTC vs.
Revision 1.632 by root, Mon Mar 11 20:48:19 2019 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?
14
15TODO: Олег Г <verdrehung@gmail.com>, empty A record response causes
16AnyEvent::resolve_sockaddr to eventually call pack_sockaddr with undef -
17broken dns server, but maybe improve the response.
18
197.15 Tue Feb 26 03:07:42 CET 2019
20 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
21 separate major and minor versions, while code passed only a single
22 value. This version follows the documentation and now passes separate
23 major and minor values.
24 - work around Net::SSLeay not having been ported to openssl 1.1, but many
25 distributions compiling it against openssl 1.1, which unfortunately
26 succeeds and results in a very broken module.
27 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
28 in the __ member, to allow decoding of undecodable resouce records
29 containing compressed domian names.
30 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
31 addresses. It now does and converts them to ipv4 mapped addresses.
32 - support CAA records, based on patch by Steve Atkins.
33 - add freenom and cloudflare nameservers as dns fallback.
34 - AnyEvent::Strict would not properly ward against io watchers
35 on files when the handle passed was a file descriptor.
36 - document "internal" variables used by the dns en-/decoder to allow
37 enterprising users to extend them in a semi-official way.
38
397.14 Sat Jun 24 01:44:19 CEST 2017
40 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
41 by Paul Howarth and apparently tracked down by Mike McCauley).
42 - AnyEvent::Handle->keepalive was documented (and defined) twice
43 (found by Matt Merhar).
44 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
45 listening socket unless a guard is used. Change this so that no
46 cleanup will be performed unless a guard is used and document this more
47 clearly.
48 - make tcp_bind/tcp_server error messages more regular.
49 - work around latest perl upstream breakage - this time, perl5porters were
50 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
51 stupidly apply anything that has "security fix" stamped on it, it's likely
52 going to be a problem in practise (and working around it is trivial on
53 windows and unix).
54 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
55 to constants.
56 - update warnings to common::sense 3.74 standards.
57
587.13 Sat Sep 17 04:31:49 CEST 2016
59 - Only call tlsext_host_name for non-empty common names (reported
60 by Maxime Soulé).
61 - log a (single) notice message if SNI is not supported.
62 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
63 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
64 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
65 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
66 testcases and the specification and not in the code - the post-9.0.0
67 unicode files have all known problems fixed, so finally the AnyEvent
68 IDNA implementation can pass the full IDNA testsuite - without needing
69 a single fix :)
70 - guarantee (and document) that condvar callbacks will be removed
71 on invocation - important to avoid circular references.
72
737.12 Wed Jan 27 19:12:26 CET 2016
74 - use common name as hostname for tls connects, if Net::SSLeay
75 supports SNI.
76 - fix documentation of tls_autostart read type in AnyEvent::Handle,
77 analyzed by Felix Ostmann.
78
797.11 Thu Jul 16 14:36:00 CEST 2015
80 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
81 addresses (extra "::" at end and similar cases).
82 - add a more explicit warning to AnyEvent::Handle that it doesn't
83 work on files, people keep getting confused.
84 - new function AnyEvent::Socket::tcp_bind.
85 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
86 - aligned ipv6 address formatting with RFC 5952 (by not shortening
87 a single :0: to ::).
88 - added stability canary support.
89
907.09 Sat May 2 16:38:53 CEST 2015
91 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
92 that was renamed to AnyEvent::Log:format_time. uinder its old name
93 (analyzed by Michael Stovenour).
94 - update AnyEvent::DNS fallback resolver addresses:
95 seems google effectively killed most other free dns resolvers,
96 so remove them, but add cable and wireless (ecrc) since it was
97 stable for 20 years or so, official or not, and there should be
98 an alternative to google.
99 - perl5porters broke windows error codes in 5.20, and mapped
100 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
101 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
102 because they were so cool. They probably broke other error codes
103 for no reason, too, but I didn't care for checking, it's too
104 depressing. This version only works around the WSAEWOULDBLOCK
105 issue, because I don't have a nice way to work around the
106 WSAEINPROGRESS bug.
107
1087.08 Wed Dec 10 05:27:17 CET 2014
109 - work around a newly introduced bug in Socket 2.011 (an
110 errornous sun_length check) (analyzed by Maxime Soulé).
111 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
112 (analyzed by Ben Magistro).
113 - AnyEvent::Strict will now confess, not croak. This is in line with
114 it being a development/debugging tool.
115 - work around a number of libglib bugs (debug builds of libglib enforce
116 certain undocumented behaviour patterns such as not being able to
117 remove a child watch source after it has fired, which we will try
118 to emulate to avoid "criticals". what where they thinking...).
119 - mention json security issues in AnyEvent::Handle, now that Douglas
120 Crockford has foolishly and incompatibly changed JSON.
121 - changed default dns resolver "max_outstanding" value from 1 to 10,
122 the latter beinfg the intended value all along
123 (reported by Ilya Chesnokov).
124 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
125 (written by Mike Lowell).
126
1277.07 Tue Dec 17 17:45:02 CET 2013
128 - the documentation for custom tls verify schemes was wrong. make it agree
129 with the code (reported by Maxime Soulé).
130 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
131 - work around an API change in openssl that could cause wrong tls connection
132 aborts, likely on windows only (analyzed by sten).
133 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
134 endless loop (reported by Dietrich Rebmann).
135 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
136 (patch by Maxime Soulé).
137 - document AnyEvent::Impl::IOAsync::set_loop and
138 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
139 functionality has _always_ been available.
140 - force a toplevel domain name in t/81_hosts.t (analyzed by
141 David Jack Wange Olrik).
142 - document that AnyEvent::Log uses AnyEvent::IO.
143 - warn about AnyEvent::Filesys::Notify performance.
144 - praise the joys of AnyEvent::Fork::*.
145 - time for an =encoding directive.
146 - do no longer use JSON to create a default json coder, use
147 JSON::XS or JSON::PP directly.
148
1497.05 Wed Aug 21 10:38:08 CEST 2013
150 - uts46data.pl couldn't be found due to wrong naming of the file
151 (reported by Fulko Hew).
152 - handle lone \015's properly in AE::Handle's default line read
153 (reported by various people).
154 - untaint ip addresses found in /etc/hosts (patch by José Micó).
155 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
156 fail on first use (reported and testcase by Andrew Whatson).
157 - expose AnyEvent::Log::format_time, and allow users to redefine it.
158 - expose AnyEvent::Log::default_format, and allow redefinition.
159 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
160 - AnyEvent::Debug shell can now run coro shell commands, if available.
161 - t/63* tests were wrongly in MANIFEST.
162 - kernel.org's finger server went MIA, switch to freebsd.org and
163 icculus.org.
164 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
165 to function asynchronously (reported by Daniel Carosone).
166 - hard-disable $^W in most tests, it generates too much garbage output.
167 - use a (hopefully) more future-proof method to emulate common::sense.
168 - upgrade to UTS-46:6.2.0.
169 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
170 documented. should not affect anything.
171
1727.04 Thu Nov 15 02:23:58 CET 2012
173 - debugging code left in.
174
1757.03 Thu Nov 15 02:16:11 CET 2012
176 - AnyEvent::Socket::inet_aton did not work when dns resolution was
177 used to find the addresses (analyzed and patch by Maxime Soulé).
178 - fix a memory leak in the /etc/hosts lookup code when hosts don't
179 resolve and are not in hosts (analyzed by sten).
180
1817.02 Tue Aug 14 04:27:58 CEST 2012
182 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
183 program by Yorhel).
184 - verified that AnyEvent::Socket follows RFC5952.
185 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
186
1877.01 Sun May 13 01:03:17 CEST 2012
188 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
189 not available, instead of throwing an exception.
190 - use File::Spec to get the tmpdir in t/*, to avoid needless
191 failures on (most, not mine :) windows boxes.
192 - new handle read types: tls_detect and tls_autostart.
193
1947.0 Fri Apr 13 06:33:30 CEST 2012
195 - child watchers are broken in POE 1.352 (also many earlier
196 versions) and there seems to be no way to work around it, as POE
197 itself is inherently racy. Document this regression and add a delay
198 in t/68_poe_03_child.t for the time being.
199 - new module AnyEvent::IO, that is a frontend to either a pure-perl
200 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
201 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
202 - load /etc/hosts only when DNS has no answer.
203 - stat /etc/hosts on every access and reload it if it changed.
204 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
205 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
206 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
207 a side effect, they now use true appending as opposed to libc
208 appending, and the file might not have been opened when the function
209 returns.
210 - the default logging level was not properly documented in a variety of
211 places, this has hopefully been rectified.
212 - updated uts46data.pl for unicode 6.1.0.
213 - made log messages generated by AnyEvent submodules not
214 include the package name anymore, as it will be logged
215 by default already.
216 - upgrade to the trick used by common-sense 3.5 to work
217 around extra warning torture/breakage under perl 5.15.x.
218 - log messages by anyevent are now ucfirst, are usually full
219 sentences and do no longer include the package name.
220 - the storable read type would sometimes throw an exception instead
221 of causing EBADMSG (analyzed by Maxime Soulé).
222 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
223 keysize (reported by Paul Howarth).
224 - documented Rocco fabricating statements about AnyEvent and me.
225
2266.14 Tue Jan 31 20:00:24 CET 2012
227 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
228 probably an editing mistake (reported by Darin McBride).
229
2306.13 Thu Jan 12 07:27:01 CET 2012
231 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
232 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
233 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
234 regex-matching inside the callback would change the parameter, despite
235 $1 being dynamically scoped per-block. this perl bug is now being
236 worked around (testcase by Cindy Wang).
237
2386.12 Mon Dec 12 13:21:10 CET 2011
239 - $! was clobbered by subsequent calls in tcp_connect, due to
240 the postpone not saving/restoring it (analyzed by Richard Garnish).
241
2426.11 Tue Nov 22 10:36:05 CET 2011
243 - Tk cannot create windows when tainted unless you set a title, so
244 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
245 - escape any nonprintable/nonascii characters when stringifying
246 backtraces.
247 - log the reason loading an interface module fails to load at level debug.
248
2496.1 Tue Oct 4 19:44:30 CEST 2011
250 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
251 and above), and some messages inside AnyEvent have been elevated
252 to higher log levels to print by default.
253 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
254 - work around yet another signal race bug, newly introduced by POE. My
255 how that module sucks.
256 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
257 support ipv6-only operation even with defaults (also use ::1 instead
258 of 127.0.0.1).
259 - give better guidelines on how to chose a logging level.
260 - AnyEvent::Log can now cap (limit) log message priority by context.
261 - AE::log will now save and restore $! and $@, so callers don't have to.
262 - AE::Strict now checks that registered fds are still valid at regular
263 intervals.
264 - skip frequently called 'require' calls if the module appears loaded.
265 - increase signal latency test timer considerably for poor little osx,
266 which can't be bothered with basic posix functionality such as signal
267 delivery.
268 - mention some debugging aids in the tutorial.
269 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
270 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
271 of the context as title.
272 - AnyEvent::Strict would not restore read-write access to $_ when a callback
273 died - this is a user bug, but since AE::Strict exists to help dbeugging...
274 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
275 to work with it - the author is well aware of the issues but refuses
276 to fix them. This extreme step has been taken because the module actively
277 attacks the goals of AnyEvent and makes it harder for module authors to use
278 AnyEvent.
279
2806.02 Fri Aug 26 20:08:31 CEST 2011
281 - log_to_syslog now accepts facility strings.
282 - log_to_syslog would not allow facilities - "$facility|$level"
283 *literally* means that string. sheesh.
284 - the CPAN parser got confused and thought we provide Sys::Syslog, try
285 to work around.
286 - renamed AnyEvent::Impl::FLTK2 to FLTK.
287
2886.01 Fri Aug 26 07:04:11 CEST 2011
289 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
290 the TTL as fourth element - this affects ->resolve, ->request
291 and ->dns_unpack indirectly. Convenience functions such as
292 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
293 AnyEvent::Socket.
294 - INCOMPATIBLE CHANGE: backend authors now should not implement
295 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
296 methods.
297 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
298 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
299 - DEPRECIATION: please change
300 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
301 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
302 in your programs, for future compatibility.
303 - *blush* idle watcher emulation was borked.
304 - AnyEvent::Strict now checks the AE::xxx functions as well.
305 - querying too long domain names would barf deep inside AnyEvent::DNS - since
306 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
307 for too long domain names explicitly and passes an undef to the callback
308 instead (reported by James Bromberger).
309 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
310 - new functions: AnyEvent::postpone, AnyEvent::log.
311 - new module: AnyEvent::Log, for simple logging needs.
312 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
313 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
314 a debug shell.
315 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
316 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
317 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
318 /etc/hosts can be found.
319 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
320 configuration.
321 - env variables can now be specified using AE_xyz instead of
322 PERL_ANYEVENT_xyz, subject to some rules.
323 - lots of new AnyEvent::Debug functionality.
324 - AnyEvent::Strict now checks for modifications of $_ by the callback
325 (by setting it to readonly while executing the callback, if possible).
326 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
327 Invest extra overhead to work around this peculiar behaviour.
328 - split the pure perl backend into a loop module and an impl
329 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
330 - some read types (regex, netstring, json, storable) could
331 errornously keep an AE::Handle object alive on their own
332 (reported by Mohammad Toossi).
333 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
334 supports /etc/hosts overrides.
335 - add more workarounds around child watcher bugs in IO::Async
336 and (newly introduced) in POE.
337 - use glib child watchers and try to work around its limitations,
338 if possible (based on a patch by Kevin Ryde).
339 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
340 - many read types in AnyEvent::Handle would malfunction during
341 transitions to TLS because of a "delete $self->{rbuf}".
342 (reported by Gerald Galster).
343 - improve condvar blocking wait performance (probably only noticable
344 with faster backends), also make it easier for event loops that
345 do not support blocking waits.
346 - PERL_ANYEVENT_MODEL now also allows full module names, although
347 the usefulness of this feature is not yet clear.
348 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
349 - add t/80_ssltext.t.
350 - no longer use AUTOLOAD for the initial loading (for no very good reason).
351 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
352 - do not run event loop detection multiple times when programs call their cached
353 copies of AnyEvent::detect.
354 - parse_hostport now accepts absolute paths als unix domain sockets.
355 - suppress the idiotic warning inside POE by patching the POE::Kernel
356 run flags instead of asking the user to jump through contortions.
357 - add Task::Weaken as dependency (as recommends only at this time).
358 - replace PF_UNSPEC by 0, for good old minix.
359
3606.0 (not officially released, but was on CPAN for about 15 minutes :/)
361
3625.34 Thu May 12 09:59:41 CEST 2011
363 - AE::TLS didn't support both cert_file and key at the same time,
364 due to some copy&paste bug (reported and analyzed by vti).
365 - improve AE::Handle timeout documentation slightly.
366
3675.33 Mon Feb 7 21:51:41 CET 2011
368 - never appeared on CPAN, for unknown reasons.
369 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
370 object when it's already gone (analyzed by Ryan Bullock).
371 - update idna algorithm(s) to UTS#46 version 6.0.0.
372 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
373 uts46 says it's an error, and testsuite makes more sense.
374 - remove leading dots in idn_to_ascii - testsuite says to remove,
375 uts46 says it's an error, and testsuite makes more sense.
376 - fixed a bug in idn ignored character processing.
377 - passes the uts46 6.0.0 testsuite except for three cases which
378 apparently are bugs in the testsuite itself.
379 - passes the uts46 6.0.1 testsuite except for eleven cases which
380 apparently are bugs in the testsuite itself.
381
3825.31 Sun Jan 23 11:49:19 CET 2011
383 - work around a POE bug where POE would first poll for new events and
384 invoke handlers BEFORE handling the message queue.
385 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
386 - stop_read now does just that in TLS mode. see its
387 documentation for caveats.
388 - update to common::sense 3.4 warning set (some warnings in perl impose
389 an arbitrary 32kb limit on regex matches, which affects
390 AnyEvent::Handle).
391 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
392 - new experimental AE::Handle::resettls method.
393
3945.3 Fri Dec 31 05:48:13 CET 2010
395 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
396 issues with IO::Async have been solved in that version, including
397 the availabilty of a default loop!
398 - implement new wbuf_max parameter for AnyEvent::Handle.
399 - added Cocoa::EventLoop backend by Daisuke Murase.
400 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
401 - clarify handle on_prepare documentation.
402 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
403 to avoid leaking them to any child processes.
404 - try to force files into the arch-dependend lib directory, for the
405 half-"turly obsessive compulsive person"(s) that depend on it.
406
4075.29 Sun Dec 5 10:49:21 CET 2010
408 - convert EV backend to EV 4.00 API (so better upgrade EV too).
409 - AE::Handle: implement dynamic read_size adjustment, add
410 max_read_size parameter, reduce default read_size to 2048.
411 - add some FAQ entries for servers and/or on_eof vs. on_error.
412 - work around OS bugs (cygwin again) when lingering in AE::Handle.
413 - reduce memory usage slightly when connect was used in AE::Handle.
414 - wrok around more segfaults in Tk.
415 - document tls_ctx => undef as valid.
416 - detect WNOHANG value at compiletime.
417 - include some probably linux-only support for building AnyEvent
418 as part of the perl core.
419 - improve compatibility of t/08_idna.t to perl 5.8.
420 - make AnyEvent::Strict truly optional (in case somebody builds a
421 minimal perl...).
422
4235.28 Wed Oct 13 04:14:23 CEST 2010
424 - due to a glitch, AnyEvent's internal getprotobyname wasn't
425 actually used in tcp_server.
426 - implement AnyEvent::CondVar->new method, which was documented but
427 missing.
428 - support autoloading of read types in unshift_read.
429 - AnyEvent::Strict: do the test for invalid fh arguments differently,
430 resulting in more informative error messages.
431 - new AnyEvent::FAQ manpage.
432 - clarify that you can actually call push_read etc. in on_read.
433 - put keys and certs in /tmp, not into the current directory,
434 and document this in the SECURITY CONSIDERATIONS sections
435 of AnyEvent::TLS.
436 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
437 requested by Ron Isaacson.
438 - document $AnyEvent::DNS::RESOLVER variable.
439 - incorporated major typo patches by Abhijit Menon-Sen.
440 - AnyEvent::Handle now croaks when negative timeouts are passed.
441 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
442 undocumented, but who knows when they might come in handy.
443
4445.271 Tue Jun 8 12:05:46 CEST 2010
445 - backport to perl 5.8.x.
446
4475.27 Sun Jun 6 12:12:05 CEST 2010
448 - postpone differently in AnyEvent::Socket now, as
449 when not, canceling the connection attempt might fail
450 (found by Felix Antonius Wilhelm Ostmann).
451 - explicitly check for non-stream sockets in AE::Handle, too many
452 clueless people fell into the trap of this somehow working.
453 - simplified and reworked the "OTHER MODULES" section.
454 - better/more condvar examples.
455
4565.261 Wed Apr 28 16:13:36 CEST 2010
457 - AF_INET6 was not properly used from Socket6 during configuration
458 time (found by Andy Grundman).
459
4605.26 Mon Apr 12 04:49:35 CEST 2010
461 - don't generate spurious readyness notifications when select
462 returns EINTR in the pure perl backend (analysed and testcase
463 by Bas Denissen).
464 - give same examples in SYNOPSIS sections of AnyEvent and AE.
465 - provider faster implementations for AE::time/now/now_update for
466 EV and Perl backends.
467 - provide faster AE::cv and AE::time implementations for all
468 backends.
469 - no longer support lower/mixed-case signal names (this was
470 never documented nor universally supported).
471 - some more memory µ-optimisations, and somewhat less messy
472 code to cope with both AnyEvent and AE APIs.
473
4745.251 Sat Mar 13 00:58:21 CET 2010
475 - make test failed if it was run before make install on perl <=
476 5.8.8, and older versions of perl make our live very difficult,
477 so write constants.pl during Makefile.PL time.
478
4795.25 Sat Mar 13 00:23:14 CET 2010
480 - fix a race condition in AnyEvent::Handle that would cause
481 a "bio_read: unsupported method" error (in _another_
482 ssl connection) after user code threw an exception.
483 - added AnyEvent::Handle->destroyed method.
484 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
485 - the AnyEvent::Util::fh_nonblocking function worked only
486 by lucky accident on win32.
487 - smaller and faster AnyEvent::Util::fh_nonblocking.
488 - when the (required!) Time::HiRes module is missing, AnyEvent
489 did not fall back to built-in time properly.
490 - do not load Fcntl at runtime, saving memory and loading time.
491 - precompile a number of constants and use them instead of runtime
492 detection and eval.
493 - free detection code after detection and similar memory optimisations.
494 - Perl backend timer interval best effort drift has been improved
495 (same algorithm as EV).
496 - update unicode idna mapping table.
497
4985.24 Tue Jan 5 11:39:43 CET 2010
499 - cygwin never reports errors from failed connects. 1.5 just gives
500 you continous readyness and EAGAIN, 1.7 is even more broken and
501 just hangs. work around both issues in a cygwin-specific hack.
502 - improve idle watcher documentation slightly.
503
5045.23 Sun Dec 20 23:48:00 CET 2009
505 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
506 - implement punycode_encode/decode, idn_nameprep,
507 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
508 - provide $AE::VERSION.
509 - removed traces of "no strict 'refs'".
510
5115.22 Sat Dec 5 03:51:13 CET 2009
512 - downgrade-or-fail in AnyEvent::Handle::push_write, to
513 diagnose encoding failures earlier and more succinctly.
514 (this works around bugs in perl, throwing away encoding info
515 when passing scalar data to extensions).
516 - add more examples to AnyEvent::Socket manpage.
517 - upgrade internal warning set to the same as common::sense 2.03.
518 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
519 (John Beppu).
520 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
521 - add google public dns servers to fallback server set and make sure
522 we load-balance properly between the three sets. also add all
523 fallback dns servers, not just a random one, to each dns config.
524
5255.21 Thu Nov 19 02:48:47 CET 2009
526 - fix a problem where socket constants were called with parameters
527 (spotted by David Friedland).
528 - fork_call never use'd POSIX (reported by Daisuke Maki).
529 - improve perl 5.6 compatibility further (but it still won't work
530 unless you rip out everything but the core).
531 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
532 uses 10MB of RAM, but doesn't open a console window. *sigh*.
533
5345.202 Wed Oct 14 22:35:44 CEST 2009
535 - AnyEvent::DNS would unexpectedly clobber $_ under windows
536 (analysed by Matthias Waldorf).
537 - AnyEvent::Handle::run_cmd can now pass the PID of the
538 newly-created process, which is much less useful than it might
539 sound (based on patch by Yann Kerherve).
540
5415.201 Tue Sep 29 12:09:25 CEST 2009
542 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
543 (reported by Torsten Foertsch).
544 - common::sense 2.0 could cause tcp_server to throw an exception
545 (analysed by elmex).
546
5475.2 Mon Sep 14 07:04:49 CEST 2009
548 - INCOMPATIBLE CHANGE: do no longer support register_read_type
549 and register_write_type in AnyEvent::Handle, instead support
550 package names (the facility was mostly abused).
551 - implement "packagename-as-read/write type" support in
552 AnyEvent::Handle.
553 - AnyEvent::Handle: new options "keepalive" and "oobinline".
554 - oobinline set by default to avoid security issues.
555 - the pure-perl event loop backend wrongly detected times() fallback
556 support (spotted by Pavel Boldin).
557
5585.12 Tue Sep 1 20:26:50 CEST 2009
559 - be more lenient when parsing resolv.conf files, as some people
560 use hashmarks as comment indicator inside directives (reported
561 by Michael S. Fischer).
562 - use same set of warnings as common::sense 2.0.
563 - fix a potential 32 bit overflow issue due to perl having problems
564 with large hex constants in the Perl backend.
565 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
566 by default.
567 - fix some prototypes.
568 - (optionally) require Async::Interrupt 1.02.
569 - replace opendns fallbacks by verizon/level3 public dns servers.
570
5715.112 Fri Aug 21 13:59:04 CEST 2009
572 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
573 using starttls with an illegal/uncreatable tls context.
574 - port to broken win32 perls.
575
5765.111 Sun Aug 16 18:44:36 CEST 2009
577 - perl errornously requires sockaddr_un structures returned by the
578 kernel to have a specific length. work around this bug by padding
579 sockaddr_un structures (found to fail on OS X, as perl only uses
580 the correct code for GNU/Linux, but likely to fail on other
581 OSes as well).
582
5835.11 Wed Aug 12 17:49:37 CEST 2009
584 - fixed a bug in run_cmd with input from filename.
585 - create a json coder object when none is given in write direction
586 to reduce dependency on JSON.
587
5885.1 Tue Aug 11 03:17:32 CEST 2009
589 - tcp_connect now reports the correct error in $!, not an unrelated
590 one (was broken since 4.91).
591 - AE::Handle did not properly clear rtimeout/wtimeout watchers
592 when retrying a tcp_connect.
593 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
594 - fixed wrong documentation in AE::Socket.
595
5965.01 Mon Aug 10 03:16:32 CEST 2009
597 - last release broke EV child watchers :/.
598
5995.0 Sun Aug 9 17:32:09 CEST 2009
600 - officially introduce and document the AE API.
4 - lowering the timeout at runtime did not have immediate effect 601 - lowering the timeout at runtime did not have immediate effect
5 in AnyEvent::Handle. 602 in AnyEvent::Handle.
6 - AE::Handle now has separate and independent read and write timeout 603 - AE::Handle now has separate and independent read and write timeout
7 settings. 604 settings.
8 - probe for Guard module when AnyEvent::Util is loaded, not at 605 - probe for Guard module when AnyEvent::Util is loaded, not at
9 runtime (Event doesn't like eval "use" from callbacks, and it avoids 606 runtime (Event doesn't like eval "use" from callbacks, and it avoids
10 exporting the probe function). 607 exporting the probe function).
608 - Impl::IOAsync's io watchers did not use the filehandle emulation
609 code to work around it's watcher limits.
610 - work around IO::Async's problems with overloaded objects.
611 - try to disable wearnings inside IO::Async, it's just too buggy.
11 - do no longer use Event by default, only EV or Perl, Event is too buggy. 612 - do no longer use Event by default, only EV or Perl, Event is too buggy.
12 - hackishly provide fast AE:: interface when EV is the backend. 613 - hackishly provide fast AE:: interface when EV or Perl is the backend.
13 - inherit VERSION from AnyEvent::VERSION in most modules. 614 - inherit VERSION from AnyEvent::VERSION in most modules.
14 615
154.91 Thu Aug 6 15:42:45 CEST 2009 6164.91 Thu Aug 6 15:42:45 CEST 2009
16 - AE::Handle::starttls could get out of sync when the read buffer 617 - AE::Handle::starttls could get out of sync when the read buffer
17 already contains some TLS handshake. 618 already contains some TLS handshake.
… …
44 - new function: AnyEvent::Socket::format_hostport. 645 - new function: AnyEvent::Socket::format_hostport.
45 646
464.881 Tue Jul 28 12:51:53 CEST 2009 6474.881 Tue Jul 28 12:51:53 CEST 2009
47 - work around a bug in local in pre-5.10 perls, causing 648 - work around a bug in local in pre-5.10 perls, causing
48 AnyEvent::Handle to recurse when it should not 649 AnyEvent::Handle to recurse when it should not
49 (analyzed by elmex). 650 (analysed by elmex).
50 651
514.88 Tue Jul 28 04:04:37 CEST 2009 6524.88 Tue Jul 28 04:04:37 CEST 2009
52 - re-bless the handle into a dummy package after calling 653 - re-bless the handle into a dummy package after calling
53 AnyEvent::Handle::destroy, so the user does not need to check 654 AnyEvent::Handle::destroy, so the user does not need to check
54 for errors after every push_write etc. 655 for errors after every push_write etc.
… …
142 in the common case. 743 in the common case.
143 - add AnyEvent::Handle->push_shutdown method. 744 - add AnyEvent::Handle->push_shutdown method.
144 - add an additional error message parameter to AnyEvent::Handle's 745 - add an additional error message parameter to AnyEvent::Handle's
145 on_error callback (for TLS, $! is still available). 746 on_error callback (for TLS, $! is still available).
146 - add AnyEvent::Handle on_starttls/on_stoptls callbacks. 747 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
147 - make AnyEvent::Handle more robust against early conenction 748 - make AnyEvent::Handle more robust against early connection
148 failures (during new), and return C<undef> in that case 749 failures (during new), and return C<undef> in that case
149 from the constructor. 750 from the constructor.
150 - AnyEvent::Handle will now try to load only JSON::XS first, 751 - AnyEvent::Handle will now try to load only JSON::XS first,
151 then fall back to JSON. 752 then fall back to JSON.
152 - format_ipv4/format_ipv6 are now exported by default, for symmetry, 753 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
… …
2764.3 Fri Oct 3 09:18:43 CEST 2008 8774.3 Fri Oct 3 09:18:43 CEST 2008
277 - AnyEvent will now install a no-op signal handler for SIGPIPE, 878 - AnyEvent will now install a no-op signal handler for SIGPIPE,
278 unless one has been installed already. 879 unless one has been installed already.
279 - warn about not loaidng AnyEvent::Impl::POE early enough 880 - warn about not loaidng AnyEvent::Impl::POE early enough
280 (patch by Adam Rosenstein). 881 (patch by Adam Rosenstein).
281 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 882 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
282 - fix a bug in where in SSL connect mode, the 883 - fix a bug in where in SSL connect mode, the
283 client would first wait for some data by the server and 884 client would first wait for some data by the server and
284 otherwise hang. (reported and analyzed in an absolutely 885 otherwise hang. (reported and analysed in an absolutely
285 exemplary manner by Adam Rosenstein). 886 exemplary manner by Adam Rosenstein).
286 - fix a bug in where SSL EOF would not be treated 887 - fix a bug in where SSL EOF would not be treated
287 as stream EOF, putting the connection into a hung state. 888 as stream EOF, putting the connection into a hung state.
288 (reported and analyzed in an absolutely exemplary manner by Adam 889 (reported and analysed in an absolutely exemplary manner by Adam
289 Rosenstein). 890 Rosenstein).
290 - fix a potential segfault when the TLS context 891 - fix a potential segfault when the TLS context
291 would go missing in a read callback (e.g. due to stoptls) - 892 would go missing in a read callback (e.g. due to stoptls) -
292 Net::SSLeay of course makes no type checking whatsoever. 893 Net::SSLeay of course makes no type checking whatsoever.
293 - AnyEvent::Handle will not stop the read watcher in TLS mode. 894 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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