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

Comparing AnyEvent/Changes (file contents):
Revision 1.168 by root, Wed Jul 9 11:32:49 2008 UTC vs.
Revision 1.641 by root, Fri Feb 7 15:06:01 2020 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
34.2 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)
12TODO: inet_aton maybe pack C4? no, add a comment why not
13TODO: socket prepare callbacks, can they easily signal failure as well?
14TODO: Олег Г <verdrehung@gmail.com>, empty A record response causes
15AnyEvent::resolve_sockaddr to eventually call pack_sockaddr with undef -
16broken dns server, but maybe improve the response.
17TODO: AnyEvent::Loop should not use POSIIX when ti isn't used.
18
19TODO: AnyEvent::Log contexts should have a way to inject log messages received form elsewhere, using timestamp etc.
20TODO: add some cheta-sheet to log synopsis regarding AE_LOG=AnyEvent::MP=+
21TODO: tcp_connect option for AnyEvent::Handle?
22 - AnyEvent::Loop no longer loads the POSIX module if Time::HiRes
23 has monotonic clock support (suggested by Felipe Gasper).
24 - since Net::SSLeay (up to 1.88-1) is _still not ported to openssl 1.1
25 and many distributions therefore will have a completely broken
26 Net::SSLeay module for the forseeable future, AnyEvent::TLS
27 will now log a warning (that is not seen by default unless you
28 increase log verbosity).
29
307.17 Wed Sep 18 03:04:49 CEST 2019
31 - work around antique openssl version in RHEL 7 by formatting
32 dh parameters differently (reported by several people).
33 - add t/13_weaken.t.
34
357.16 Fri Jul 19 18:00:21 CEST 2019
36 - add ffdhe group dh parameters from RFC 7919, and use ffdhe3072
37 as new default, instead of schmorp1539.
38 - AnyEvent::Log did not reassess logging status of
39 AnyEvent::Log::logger's when contexts were changed with
40 ->attach/detach/slaves, causing them to not log even though a
41 recent attach should have caused them to log.
42 - added some more logging configuration examples.
43 - mention RFC 8482 in AnyEvent::DNS.
44
457.15 Tue Feb 26 03:07:42 CET 2019
46 - INCOMPATIBLE CHANGE: AnyEvent::Handle's tls_detect documentation gave
47 separate major and minor versions, while code passed only a single
48 value. This version follows the documentation and now passes separate
49 major and minor values.
50 - work around Net::SSLeay not having been ported to openssl 1.1, but many
51 distributions compiling it against openssl 1.1, which unfortunately
52 succeeds and results in a very broken module.
53 - AnyEvent::DNS::dns_unpack now stores the original DNS packet
54 in the __ member, to allow decoding of undecodable resouce records
55 containing compressed domian names.
56 - AnyEvent::Socket::parse_ipv6 would NOT, as advertised, accept ipv4
57 addresses. It now does and converts them to ipv4 mapped addresses.
58 - support CAA records, based on patch by Steve Atkins.
59 - add freenom and cloudflare nameservers as dns fallback.
60 - AnyEvent::Strict would not properly ward against io watchers
61 on files when the handle passed was a file descriptor.
62 - document "internal" variables used by the dns en-/decoder to allow
63 enterprising users to extend them in a semi-official way.
64
657.14 Sat Jun 24 01:44:19 CEST 2017
66 - fix a crash bug in AnyEvent::Handle with openssl 1.1.0 (patched
67 by Paul Howarth and apparently tracked down by Mike McCauley).
68 - AnyEvent::Handle->keepalive was documented (and defined) twice
69 (found by Matt Merhar).
70 - AnyEvent::Socket::tcp_bind/tcp_server would immediately unlink a unix
71 listening socket unless a guard is used. Change this so that no
72 cleanup will be performed unless a guard is used and document this more
73 clearly.
74 - make tcp_bind/tcp_server error messages more regular.
75 - work around latest perl upstream breakage - this time, perl5porters were
76 apparently bullied by a bogus CVE (CVE-2016-1238), and since distros
77 stupidly apply anything that has "security fix" stamped on it, it's likely
78 going to be a problem in practise (and working around it is trivial on
79 windows and unix).
80 - add TCP_FASTOPEN/MSG_FASTOPEN and MSG_NOSIGNAL/MSG_MORE/MSG_DONTWAIT
81 to constants.
82 - update warnings to common::sense 3.74 standards.
83
847.13 Sat Sep 17 04:31:49 CEST 2016
85 - Only call tlsext_host_name for non-empty common names (reported
86 by Maxime Soulé).
87 - log a (single) notice message if SNI is not supported.
88 - upgrade to UTS-46:9.0.0 draft and switch to non-transitional
89 behaviour, beating thunderbird, ie, edge, chrome and safari to it :)
90 (see also https://bugzilla.mozilla.org/show_bug.cgi?id=1218179)
91 - turns out the UTS-46 IDNA testcase failures were indeed bugs in the
92 testcases and the specification and not in the code - the post-9.0.0
93 unicode files have all known problems fixed, so finally the AnyEvent
94 IDNA implementation can pass the full IDNA testsuite - without needing
95 a single fix :)
96 - guarantee (and document) that condvar callbacks will be removed
97 on invocation - important to avoid circular references.
98
997.12 Wed Jan 27 19:12:26 CET 2016
100 - use common name as hostname for tls connects, if Net::SSLeay
101 supports SNI.
102 - fix documentation of tls_autostart read type in AnyEvent::Handle,
103 analyzed by Felix Ostmann.
104
1057.11 Thu Jul 16 14:36:00 CEST 2015
106 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
107 addresses (extra "::" at end and similar cases).
108 - add a more explicit warning to AnyEvent::Handle that it doesn't
109 work on files, people keep getting confused.
110 - new function AnyEvent::Socket::tcp_bind.
111 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
112 - aligned ipv6 address formatting with RFC 5952 (by not shortening
113 a single :0: to ::).
114 - added stability canary support.
115
1167.09 Sat May 2 16:38:53 CEST 2015
117 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
118 that was renamed to AnyEvent::Log:format_time. uinder its old name
119 (analyzed by Michael Stovenour).
120 - update AnyEvent::DNS fallback resolver addresses:
121 seems google effectively killed most other free dns resolvers,
122 so remove them, but add cable and wireless (ecrc) since it was
123 stable for 20 years or so, official or not, and there should be
124 an alternative to google.
125 - perl5porters broke windows error codes in 5.20, and mapped
126 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
127 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
128 because they were so cool. They probably broke other error codes
129 for no reason, too, but I didn't care for checking, it's too
130 depressing. This version only works around the WSAEWOULDBLOCK
131 issue, because I don't have a nice way to work around the
132 WSAEINPROGRESS bug.
133
1347.08 Wed Dec 10 05:27:17 CET 2014
135 - work around a newly introduced bug in Socket 2.011 (an
136 errornous sun_length check) (analyzed by Maxime Soulé).
137 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
138 (analyzed by Ben Magistro).
139 - AnyEvent::Strict will now confess, not croak. This is in line with
140 it being a development/debugging tool.
141 - work around a number of libglib bugs (debug builds of libglib enforce
142 certain undocumented behaviour patterns such as not being able to
143 remove a child watch source after it has fired, which we will try
144 to emulate to avoid "criticals". what where they thinking...).
145 - mention json security issues in AnyEvent::Handle, now that Douglas
146 Crockford has foolishly and incompatibly changed JSON.
147 - changed default dns resolver "max_outstanding" value from 1 to 10,
148 the latter beinfg the intended value all along
149 (reported by Ilya Chesnokov).
150 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
151 (written by Mike Lowell).
152
1537.07 Tue Dec 17 17:45:02 CET 2013
154 - the documentation for custom tls verify schemes was wrong. make it agree
155 with the code (reported by Maxime Soulé).
156 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
157 - work around an API change in openssl that could cause wrong tls connection
158 aborts, likely on windows only (analyzed by sten).
159 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
160 endless loop (reported by Dietrich Rebmann).
161 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
162 (patch by Maxime Soulé).
163 - document AnyEvent::Impl::IOAsync::set_loop and
164 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
165 functionality has _always_ been available.
166 - force a toplevel domain name in t/81_hosts.t (analyzed by
167 David Jack Wange Olrik).
168 - document that AnyEvent::Log uses AnyEvent::IO.
169 - warn about AnyEvent::Filesys::Notify performance.
170 - praise the joys of AnyEvent::Fork::*.
171 - time for an =encoding directive.
172 - do no longer use JSON to create a default json coder, use
173 JSON::XS or JSON::PP directly.
174
1757.05 Wed Aug 21 10:38:08 CEST 2013
176 - uts46data.pl couldn't be found due to wrong naming of the file
177 (reported by Fulko Hew).
178 - handle lone \015's properly in AE::Handle's default line read
179 (reported by various people).
180 - untaint ip addresses found in /etc/hosts (patch by José Micó).
181 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
182 fail on first use (reported and testcase by Andrew Whatson).
183 - expose AnyEvent::Log::format_time, and allow users to redefine it.
184 - expose AnyEvent::Log::default_format, and allow redefinition.
185 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
186 - AnyEvent::Debug shell can now run coro shell commands, if available.
187 - t/63* tests were wrongly in MANIFEST.
188 - kernel.org's finger server went MIA, switch to freebsd.org and
189 icculus.org.
190 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
191 to function asynchronously (reported by Daniel Carosone).
192 - hard-disable $^W in most tests, it generates too much garbage output.
193 - use a (hopefully) more future-proof method to emulate common::sense.
194 - upgrade to UTS-46:6.2.0.
195 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
196 documented. should not affect anything.
197
1987.04 Thu Nov 15 02:23:58 CET 2012
199 - debugging code left in.
200
2017.03 Thu Nov 15 02:16:11 CET 2012
202 - AnyEvent::Socket::inet_aton did not work when dns resolution was
203 used to find the addresses (analyzed and patch by Maxime Soulé).
204 - fix a memory leak in the /etc/hosts lookup code when hosts don't
205 resolve and are not in hosts (analyzed by sten).
206
2077.02 Tue Aug 14 04:27:58 CEST 2012
208 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
209 program by Yorhel).
210 - verified that AnyEvent::Socket follows RFC5952.
211 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
212
2137.01 Sun May 13 01:03:17 CEST 2012
214 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
215 not available, instead of throwing an exception.
216 - use File::Spec to get the tmpdir in t/*, to avoid needless
217 failures on (most, not mine :) windows boxes.
218 - new handle read types: tls_detect and tls_autostart.
219
2207.0 Fri Apr 13 06:33:30 CEST 2012
221 - child watchers are broken in POE 1.352 (also many earlier
222 versions) and there seems to be no way to work around it, as POE
223 itself is inherently racy. Document this regression and add a delay
224 in t/68_poe_03_child.t for the time being.
225 - new module AnyEvent::IO, that is a frontend to either a pure-perl
226 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
227 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
228 - load /etc/hosts only when DNS has no answer.
229 - stat /etc/hosts on every access and reload it if it changed.
230 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
231 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
232 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
233 a side effect, they now use true appending as opposed to libc
234 appending, and the file might not have been opened when the function
235 returns.
236 - the default logging level was not properly documented in a variety of
237 places, this has hopefully been rectified.
238 - updated uts46data.pl for unicode 6.1.0.
239 - made log messages generated by AnyEvent submodules not
240 include the package name anymore, as it will be logged
241 by default already.
242 - upgrade to the trick used by common-sense 3.5 to work
243 around extra warning torture/breakage under perl 5.15.x.
244 - log messages by anyevent are now ucfirst, are usually full
245 sentences and do no longer include the package name.
246 - the storable read type would sometimes throw an exception instead
247 of causing EBADMSG (analyzed by Maxime Soulé).
248 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
249 keysize (reported by Paul Howarth).
250 - documented Rocco fabricating statements about AnyEvent and me.
251
2526.14 Tue Jan 31 20:00:24 CET 2012
253 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
254 probably an editing mistake (reported by Darin McBride).
255
2566.13 Thu Jan 12 07:27:01 CET 2012
257 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
258 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
259 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
260 regex-matching inside the callback would change the parameter, despite
261 $1 being dynamically scoped per-block. this perl bug is now being
262 worked around (testcase by Cindy Wang).
263
2646.12 Mon Dec 12 13:21:10 CET 2011
265 - $! was clobbered by subsequent calls in tcp_connect, due to
266 the postpone not saving/restoring it (analyzed by Richard Garnish).
267
2686.11 Tue Nov 22 10:36:05 CET 2011
269 - Tk cannot create windows when tainted unless you set a title, so
270 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
271 - escape any nonprintable/nonascii characters when stringifying
272 backtraces.
273 - log the reason loading an interface module fails to load at level debug.
274
2756.1 Tue Oct 4 19:44:30 CEST 2011
276 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
277 and above), and some messages inside AnyEvent have been elevated
278 to higher log levels to print by default.
279 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
280 - work around yet another signal race bug, newly introduced by POE. My
281 how that module sucks.
282 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
283 support ipv6-only operation even with defaults (also use ::1 instead
284 of 127.0.0.1).
285 - give better guidelines on how to chose a logging level.
286 - AnyEvent::Log can now cap (limit) log message priority by context.
287 - AE::log will now save and restore $! and $@, so callers don't have to.
288 - AE::Strict now checks that registered fds are still valid at regular
289 intervals.
290 - skip frequently called 'require' calls if the module appears loaded.
291 - increase signal latency test timer considerably for poor little osx,
292 which can't be bothered with basic posix functionality such as signal
293 delivery.
294 - mention some debugging aids in the tutorial.
295 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
296 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
297 of the context as title.
298 - AnyEvent::Strict would not restore read-write access to $_ when a callback
299 died - this is a user bug, but since AE::Strict exists to help dbeugging...
300 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
301 to work with it - the author is well aware of the issues but refuses
302 to fix them. This extreme step has been taken because the module actively
303 attacks the goals of AnyEvent and makes it harder for module authors to use
304 AnyEvent.
305
3066.02 Fri Aug 26 20:08:31 CEST 2011
307 - log_to_syslog now accepts facility strings.
308 - log_to_syslog would not allow facilities - "$facility|$level"
309 *literally* means that string. sheesh.
310 - the CPAN parser got confused and thought we provide Sys::Syslog, try
311 to work around.
312 - renamed AnyEvent::Impl::FLTK2 to FLTK.
313
3146.01 Fri Aug 26 07:04:11 CEST 2011
315 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
316 the TTL as fourth element - this affects ->resolve, ->request
317 and ->dns_unpack indirectly. Convenience functions such as
318 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
319 AnyEvent::Socket.
320 - INCOMPATIBLE CHANGE: backend authors now should not implement
321 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
322 methods.
323 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
324 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
325 - DEPRECIATION: please change
326 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
327 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
328 in your programs, for future compatibility.
329 - *blush* idle watcher emulation was borked.
330 - AnyEvent::Strict now checks the AE::xxx functions as well.
331 - querying too long domain names would barf deep inside AnyEvent::DNS - since
332 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
333 for too long domain names explicitly and passes an undef to the callback
334 instead (reported by James Bromberger).
335 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
336 - new functions: AnyEvent::postpone, AnyEvent::log.
337 - new module: AnyEvent::Log, for simple logging needs.
338 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
339 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
340 a debug shell.
341 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
342 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
343 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
344 /etc/hosts can be found.
345 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
346 configuration.
347 - env variables can now be specified using AE_xyz instead of
348 PERL_ANYEVENT_xyz, subject to some rules.
349 - lots of new AnyEvent::Debug functionality.
350 - AnyEvent::Strict now checks for modifications of $_ by the callback
351 (by setting it to readonly while executing the callback, if possible).
352 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
353 Invest extra overhead to work around this peculiar behaviour.
354 - split the pure perl backend into a loop module and an impl
355 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
356 - some read types (regex, netstring, json, storable) could
357 errornously keep an AE::Handle object alive on their own
358 (reported by Mohammad Toossi).
359 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
360 supports /etc/hosts overrides.
361 - add more workarounds around child watcher bugs in IO::Async
362 and (newly introduced) in POE.
363 - use glib child watchers and try to work around its limitations,
364 if possible (based on a patch by Kevin Ryde).
365 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
366 - many read types in AnyEvent::Handle would malfunction during
367 transitions to TLS because of a "delete $self->{rbuf}".
368 (reported by Gerald Galster).
369 - improve condvar blocking wait performance (probably only noticable
370 with faster backends), also make it easier for event loops that
371 do not support blocking waits.
372 - PERL_ANYEVENT_MODEL now also allows full module names, although
373 the usefulness of this feature is not yet clear.
374 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
375 - add t/80_ssltext.t.
376 - no longer use AUTOLOAD for the initial loading (for no very good reason).
377 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
378 - do not run event loop detection multiple times when programs call their cached
379 copies of AnyEvent::detect.
380 - parse_hostport now accepts absolute paths als unix domain sockets.
381 - suppress the idiotic warning inside POE by patching the POE::Kernel
382 run flags instead of asking the user to jump through contortions.
383 - add Task::Weaken as dependency (as recommends only at this time).
384 - replace PF_UNSPEC by 0, for good old minix.
385
3866.0 (not officially released, but was on CPAN for about 15 minutes :/)
387
3885.34 Thu May 12 09:59:41 CEST 2011
389 - AE::TLS didn't support both cert_file and key at the same time,
390 due to some copy&paste bug (reported and analyzed by vti).
391 - improve AE::Handle timeout documentation slightly.
392
3935.33 Mon Feb 7 21:51:41 CET 2011
394 - never appeared on CPAN, for unknown reasons.
395 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
396 object when it's already gone (analyzed by Ryan Bullock).
397 - update idna algorithm(s) to UTS#46 version 6.0.0.
398 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
399 uts46 says it's an error, and testsuite makes more sense.
400 - remove leading dots in idn_to_ascii - testsuite says to remove,
401 uts46 says it's an error, and testsuite makes more sense.
402 - fixed a bug in idn ignored character processing.
403 - passes the uts46 6.0.0 testsuite except for three cases which
404 apparently are bugs in the testsuite itself.
405 - passes the uts46 6.0.1 testsuite except for eleven cases which
406 apparently are bugs in the testsuite itself.
407
4085.31 Sun Jan 23 11:49:19 CET 2011
409 - work around a POE bug where POE would first poll for new events and
410 invoke handlers BEFORE handling the message queue.
411 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
412 - stop_read now does just that in TLS mode. see its
413 documentation for caveats.
414 - update to common::sense 3.4 warning set (some warnings in perl impose
415 an arbitrary 32kb limit on regex matches, which affects
416 AnyEvent::Handle).
417 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
418 - new experimental AE::Handle::resettls method.
419
4205.3 Fri Dec 31 05:48:13 CET 2010
421 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
422 issues with IO::Async have been solved in that version, including
423 the availabilty of a default loop!
424 - implement new wbuf_max parameter for AnyEvent::Handle.
425 - added Cocoa::EventLoop backend by Daisuke Murase.
426 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
427 - clarify handle on_prepare documentation.
428 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
429 to avoid leaking them to any child processes.
430 - try to force files into the arch-dependend lib directory, for the
431 half-"turly obsessive compulsive person"(s) that depend on it.
432
4335.29 Sun Dec 5 10:49:21 CET 2010
434 - convert EV backend to EV 4.00 API (so better upgrade EV too).
435 - AE::Handle: implement dynamic read_size adjustment, add
436 max_read_size parameter, reduce default read_size to 2048.
437 - add some FAQ entries for servers and/or on_eof vs. on_error.
438 - work around OS bugs (cygwin again) when lingering in AE::Handle.
439 - reduce memory usage slightly when connect was used in AE::Handle.
440 - wrok around more segfaults in Tk.
441 - document tls_ctx => undef as valid.
442 - detect WNOHANG value at compiletime.
443 - include some probably linux-only support for building AnyEvent
444 as part of the perl core.
445 - improve compatibility of t/08_idna.t to perl 5.8.
446 - make AnyEvent::Strict truly optional (in case somebody builds a
447 minimal perl...).
448
4495.28 Wed Oct 13 04:14:23 CEST 2010
450 - due to a glitch, AnyEvent's internal getprotobyname wasn't
451 actually used in tcp_server.
452 - implement AnyEvent::CondVar->new method, which was documented but
453 missing.
454 - support autoloading of read types in unshift_read.
455 - AnyEvent::Strict: do the test for invalid fh arguments differently,
456 resulting in more informative error messages.
457 - new AnyEvent::FAQ manpage.
458 - clarify that you can actually call push_read etc. in on_read.
459 - put keys and certs in /tmp, not into the current directory,
460 and document this in the SECURITY CONSIDERATIONS sections
461 of AnyEvent::TLS.
462 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
463 requested by Ron Isaacson.
464 - document $AnyEvent::DNS::RESOLVER variable.
465 - incorporated major typo patches by Abhijit Menon-Sen.
466 - AnyEvent::Handle now croaks when negative timeouts are passed.
467 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
468 undocumented, but who knows when they might come in handy.
469
4705.271 Tue Jun 8 12:05:46 CEST 2010
471 - backport to perl 5.8.x.
472
4735.27 Sun Jun 6 12:12:05 CEST 2010
474 - postpone differently in AnyEvent::Socket now, as
475 when not, canceling the connection attempt might fail
476 (found by Felix Antonius Wilhelm Ostmann).
477 - explicitly check for non-stream sockets in AE::Handle, too many
478 clueless people fell into the trap of this somehow working.
479 - simplified and reworked the "OTHER MODULES" section.
480 - better/more condvar examples.
481
4825.261 Wed Apr 28 16:13:36 CEST 2010
483 - AF_INET6 was not properly used from Socket6 during configuration
484 time (found by Andy Grundman).
485
4865.26 Mon Apr 12 04:49:35 CEST 2010
487 - don't generate spurious readyness notifications when select
488 returns EINTR in the pure perl backend (analysed and testcase
489 by Bas Denissen).
490 - give same examples in SYNOPSIS sections of AnyEvent and AE.
491 - provider faster implementations for AE::time/now/now_update for
492 EV and Perl backends.
493 - provide faster AE::cv and AE::time implementations for all
494 backends.
495 - no longer support lower/mixed-case signal names (this was
496 never documented nor universally supported).
497 - some more memory µ-optimisations, and somewhat less messy
498 code to cope with both AnyEvent and AE APIs.
499
5005.251 Sat Mar 13 00:58:21 CET 2010
501 - make test failed if it was run before make install on perl <=
502 5.8.8, and older versions of perl make our live very difficult,
503 so write constants.pl during Makefile.PL time.
504
5055.25 Sat Mar 13 00:23:14 CET 2010
506 - fix a race condition in AnyEvent::Handle that would cause
507 a "bio_read: unsupported method" error (in _another_
508 ssl connection) after user code threw an exception.
509 - added AnyEvent::Handle->destroyed method.
510 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
511 - the AnyEvent::Util::fh_nonblocking function worked only
512 by lucky accident on win32.
513 - smaller and faster AnyEvent::Util::fh_nonblocking.
514 - when the (required!) Time::HiRes module is missing, AnyEvent
515 did not fall back to built-in time properly.
516 - do not load Fcntl at runtime, saving memory and loading time.
517 - precompile a number of constants and use them instead of runtime
518 detection and eval.
519 - free detection code after detection and similar memory optimisations.
520 - Perl backend timer interval best effort drift has been improved
521 (same algorithm as EV).
522 - update unicode idna mapping table.
523
5245.24 Tue Jan 5 11:39:43 CET 2010
525 - cygwin never reports errors from failed connects. 1.5 just gives
526 you continous readyness and EAGAIN, 1.7 is even more broken and
527 just hangs. work around both issues in a cygwin-specific hack.
528 - improve idle watcher documentation slightly.
529
5305.23 Sun Dec 20 23:48:00 CET 2009
531 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
532 - implement punycode_encode/decode, idn_nameprep,
533 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
534 - provide $AE::VERSION.
535 - removed traces of "no strict 'refs'".
536
5375.22 Sat Dec 5 03:51:13 CET 2009
538 - downgrade-or-fail in AnyEvent::Handle::push_write, to
539 diagnose encoding failures earlier and more succinctly.
540 (this works around bugs in perl, throwing away encoding info
541 when passing scalar data to extensions).
542 - add more examples to AnyEvent::Socket manpage.
543 - upgrade internal warning set to the same as common::sense 2.03.
544 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
545 (John Beppu).
546 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
547 - add google public dns servers to fallback server set and make sure
548 we load-balance properly between the three sets. also add all
549 fallback dns servers, not just a random one, to each dns config.
550
5515.21 Thu Nov 19 02:48:47 CET 2009
552 - fix a problem where socket constants were called with parameters
553 (spotted by David Friedland).
554 - fork_call never use'd POSIX (reported by Daisuke Maki).
555 - improve perl 5.6 compatibility further (but it still won't work
556 unless you rip out everything but the core).
557 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
558 uses 10MB of RAM, but doesn't open a console window. *sigh*.
559
5605.202 Wed Oct 14 22:35:44 CEST 2009
561 - AnyEvent::DNS would unexpectedly clobber $_ under windows
562 (analysed by Matthias Waldorf).
563 - AnyEvent::Handle::run_cmd can now pass the PID of the
564 newly-created process, which is much less useful than it might
565 sound (based on patch by Yann Kerherve).
566
5675.201 Tue Sep 29 12:09:25 CEST 2009
568 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
569 (reported by Torsten Foertsch).
570 - common::sense 2.0 could cause tcp_server to throw an exception
571 (analysed by elmex).
572
5735.2 Mon Sep 14 07:04:49 CEST 2009
574 - INCOMPATIBLE CHANGE: do no longer support register_read_type
575 and register_write_type in AnyEvent::Handle, instead support
576 package names (the facility was mostly abused).
577 - implement "packagename-as-read/write type" support in
578 AnyEvent::Handle.
579 - AnyEvent::Handle: new options "keepalive" and "oobinline".
580 - oobinline set by default to avoid security issues.
581 - the pure-perl event loop backend wrongly detected times() fallback
582 support (spotted by Pavel Boldin).
583
5845.12 Tue Sep 1 20:26:50 CEST 2009
585 - be more lenient when parsing resolv.conf files, as some people
586 use hashmarks as comment indicator inside directives (reported
587 by Michael S. Fischer).
588 - use same set of warnings as common::sense 2.0.
589 - fix a potential 32 bit overflow issue due to perl having problems
590 with large hex constants in the Perl backend.
591 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
592 by default.
593 - fix some prototypes.
594 - (optionally) require Async::Interrupt 1.02.
595 - replace opendns fallbacks by verizon/level3 public dns servers.
596
5975.112 Fri Aug 21 13:59:04 CEST 2009
598 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
599 using starttls with an illegal/uncreatable tls context.
600 - port to broken win32 perls.
601
6025.111 Sun Aug 16 18:44:36 CEST 2009
603 - perl errornously requires sockaddr_un structures returned by the
604 kernel to have a specific length. work around this bug by padding
605 sockaddr_un structures (found to fail on OS X, as perl only uses
606 the correct code for GNU/Linux, but likely to fail on other
607 OSes as well).
608
6095.11 Wed Aug 12 17:49:37 CEST 2009
610 - fixed a bug in run_cmd with input from filename.
611 - create a json coder object when none is given in write direction
612 to reduce dependency on JSON.
613
6145.1 Tue Aug 11 03:17:32 CEST 2009
615 - tcp_connect now reports the correct error in $!, not an unrelated
616 one (was broken since 4.91).
617 - AE::Handle did not properly clear rtimeout/wtimeout watchers
618 when retrying a tcp_connect.
619 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
620 - fixed wrong documentation in AE::Socket.
621
6225.01 Mon Aug 10 03:16:32 CEST 2009
623 - last release broke EV child watchers :/.
624
6255.0 Sun Aug 9 17:32:09 CEST 2009
626 - officially introduce and document the AE API.
627 - lowering the timeout at runtime did not have immediate effect
628 in AnyEvent::Handle.
629 - AE::Handle now has separate and independent read and write timeout
630 settings.
631 - probe for Guard module when AnyEvent::Util is loaded, not at
632 runtime (Event doesn't like eval "use" from callbacks, and it avoids
633 exporting the probe function).
634 - Impl::IOAsync's io watchers did not use the filehandle emulation
635 code to work around it's watcher limits.
636 - work around IO::Async's problems with overloaded objects.
637 - try to disable wearnings inside IO::Async, it's just too buggy.
638 - do no longer use Event by default, only EV or Perl, Event is too buggy.
639 - hackishly provide fast AE:: interface when EV or Perl is the backend.
640 - inherit VERSION from AnyEvent::VERSION in most modules.
641
6424.91 Thu Aug 6 15:42:45 CEST 2009
643 - AE::Handle::starttls could get out of sync when the read buffer
644 already contains some TLS handshake.
645 - AE::Handle did not properly free the TLS session, causing bigger
646 memory leaks in Net::SSLeay than expected (reported by toaster).
647 - AE::Socket::tcp_connect will now artificially delay invoking
648 the callback to avoid returning after invoking the callback.
649 - convert many internal watcher uses to AE API.
650
6514.9 Sat Aug 1 11:07:01 CEST 2009
652 - Glib silently fails when registering a timeout with
653 a negative value, so avoid that.
654 - call condvar callback immediately when it is set
655 after the condition is already signalled.
656 - check rbuf_max condition only after trying to consume
657 data.
658 - normalise signal numbers to names when using the ae's
659 signal handling, but do not document this (yet).
660 - pure perl signal emulation did not properly set nonblocking
661 mode on the signal pipe on win32 (this is unlikely to have caused
662 any issues).
663 - new module: AnyEvent::Debug.
664 - AnyEvent::Strict now checks that a signal was specified
665 by name (not name or number).
666 - reduce memory footprint in typical cases by ~50kb by
667 only compiling signal code when necessary.
668 - add AnyEvent::Handle->rbuf_max.
669 - grab the AE namespace for future new API, implement stubs for
670 said future/alternative API.
671 - new function: AnyEvent::Socket::format_hostport.
672
6734.881 Tue Jul 28 12:51:53 CEST 2009
674 - work around a bug in local in pre-5.10 perls, causing
675 AnyEvent::Handle to recurse when it should not
676 (analysed by elmex).
677
6784.88 Tue Jul 28 04:04:37 CEST 2009
679 - re-bless the handle into a dummy package after calling
680 AnyEvent::Handle::destroy, so the user does not need to check
681 for errors after every push_write etc.
682 - do not attempt to run t/02_signals.t on obviously broken
683 platforms.
684
6854.87 Sun Jul 26 02:06:16 CEST 2009
686 - do not attempt to linger when there is no longer a valid fh.
687 - enforce tls mode to be either accept or connect, do not simply
688 segfault in Net::SSLeay.
689 - AnyEvent::Handle can now call tcp_connect itself (new parameters
690 connect, on_prepare, on_connect and on_connect_error). Updated
691 tutorial accordingly.
692 - add AnyEvent::Impl::Irssi backend.
693
6944.86 Mon Jul 20 23:52:29 CEST 2009
695 - since the verbose warning is not enough, explicitly document
696 that versions before 1.33 of Net::SSLeay are not secure.
697 - work around signal handling races in Event and (...) Event::Lib.
698 - try to align signal-race timer to full-second boundaries.
699 - work around Tk not liking negative timeouts.
700 - don't complain of different grades of environmental unfriendlyness
701 in IO::Async.
702
7034.85 Sat Jul 18 06:16:14 CEST 2009
704 - nail the signal race problem in perl once and for all
705 (see $AnyEvent::MAX_SIGNAL_LATENCY).
706 - take advantage of Async::Interrupt if it is available.
707 - load Time::HiRes and Guard modules on demand only.
708 - add optional/recommended modules section to AnyEvent
709 documentation.
710 - reduce memory usage considerably (and reduce startup penalty)
711 by not using "strict", "warnings" and "overload" modules.
712 - work around buggy windows/openbsd perls and provide EBADMSG
713 and EPROTO ourselves when missing.
714 - improve perl 5.6 compatibility of the core event loop.
715 - made Net::SSLeay version 1.33 a soft requirement.
716
7174.83 Fri Jul 17 16:56:26 CEST 2009
718 - implement AnyEvent::Socket::getprotobyname.
719 - AnyEvent::CondVar's will now detect recursive blocking
720 waits and will croak, as too many people fall into
721 this trap.
722 - AnyEvent::Handle will now call ->destroy on itself after
723 executing the on_error callback, instead of doing some
724 half-baked internal shutdown, for fatal errors.
725 - clarify on_eof behaviour w.r.t. the read queue and
726 on_read callbacks.
727 - ignore some possible spurious wake-ups in tcp_connect.
728
7294.82 Sat Jul 11 00:34:55 CEST 2009
730 - POE and Event backends didn't accept some callable objects as
731 callbacks.
732 - use Config module instead of POSIX module to detect signal names
733 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
734 module doesn't even have all POSIX signals :/.
735 - use more workarounds around the many refcnt/corruption bugs in
736 Event::Lib.
737 - work around a race condition in perl's select, causing t/03_child.t
738 to rarely fail.
739
7404.81 Thu Jul 9 10:30:30 CEST 2009
741 - AnyEvent::Handle didn't properly diagnose write errors
742 (it expected -1 from syswrite, how lame... :).
743 - support file descriptors in addition to file handles
744 in AnyEvent->io.
745 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
746 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
747 and PERL_ANYEVENT_CA_PATH.
748 - provide a sensible synopsis section for AnyEvent::TLS.
749 - add a "supported backends" section to the manpage.
750 - added simple io watcher test to testsuite, using a
751 portable_socketpair.
752 - tried to improve the stability of the Event::Lib backend,
753 YMMV.
754
7554.8 Mon Jul 6 23:45:16 CEST 2009
756 - AnyEvent::DNS did not properly follow CNAME records with
757 uppercase targets.
758 - AnyEvent::DNS would errornously return AAAA records
759 with v4 mapped addresses (a faulty record) as ipv4 addresses,
760 causing AnyEvent::Socket to throw an exception.
761 - added new module AnyEvent::TLS for easier SSL/TLS context
762 creation, with many options including hostname verification,
763 secure default configuration, lots of documentation and,
764 predefined diffie-hellman keys for perfect forward security
765 and much more. get it while it's still fresh!
766 - use AnyEvent::TLS in AnyEvent::Handle for context management.
767 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
768 so AnyEvent::Socket users have smaller memory footprint
769 in the common case.
770 - add AnyEvent::Handle->push_shutdown method.
771 - add an additional error message parameter to AnyEvent::Handle's
772 on_error callback (for TLS, $! is still available).
773 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
774 - make AnyEvent::Handle more robust against early connection
775 failures (during new), and return C<undef> in that case
776 from the constructor.
777 - AnyEvent::Handle will now try to load only JSON::XS first,
778 then fall back to JSON.
779 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
780 and because it was documented that way.
781
7824.451 Fri Jul 3 00:28:58 CEST 2009
783 - do not clear rbuf when shutting down an AnyEvent::Handle
784 object - doing so breaks AnyEvent::HTTP.
785
7864.45 Mon Jun 29 22:59:26 CEST 2009
787 - a write error could cause AnyEvent::Handle to create
788 an I/O watcher with an undefined $fh.
789 - special-case mapped ipv4 addresses in both
790 AnyEvent::Socket::format_address and parse_address,
791 to treat them just like ipv4 addresses.
792 - updated and overhauled the AnyEvent::Intro doc.
793 - implement AnyEvent::Socket::format_ipv4/ipv6.
794 - slightly speed up portable_pipe/socketpair.
795 - expand condvar begin/end documentation.
796
7974.42 Fri Jun 26 08:32:18 CEST 2009
798 - preliminary and neccesarily incomplete support for IO::Async.
799 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
800 case it was set to IGNORE, to ensure we can catch child statuses
801 even when the calling env acts stupidly.
802 - updated benchmarks with IO::Async, which performs very well.
803
8044.412 Wed Jun 24 01:35:57 CEST 2009
805 - support an "untaint" attribute for AnyEvent::DNS
806 and set it on the default resolver.
807 - implement and document AnyEvent::Impl::Perl::loop.
808 - remove all anyevent-env variables from %ENV
809 when running in tainted mode.
810 - mention and extend the IO::Lambda benchmark.
811
8124.411 Sun Jun 7 18:48:02 CEST 2009
813 - do not try to use F_SETFD on windows.
814
8154.41 Thu May 14 06:40:11 CEST 2009
816 - work around issues in older perls (5.8.0?) when
817 a signal handler is deleted from the %SIG hash.
818 - use POSIX::_exit in child test, to avoid running
819 destructors.
820 - speed up CHLD handling by relying on SIGCHLD being
821 synchronously delivered, even when we roll our own
822 implementation.
823 - AnyEvent::DNS: add the "dname" resource record name
824 for cosmetic reasons.
825
8264.4 Sun Apr 26 20:12:33 CEST 2009
827 - implemented idle watchers, where applicable.
828 - AnyEvent->time died when Event backend was in use.
829 - fix a memleak in the Tk backend.
830 - sped up Tk timer handling.
831 - clip negative "after" values to 0 in AnyEvent::Impl::Event
832 to avoid spamming out warning messages.
833 - fix Qt timers without interval.
834 - avoid Qt zero-timeout special case, in old libqt's.
835
8364.352 Mon Apr 20 16:31:11 CEST 2009
837 - fix AnyEvent::Strict error messages for child watchers.
838 - fix/update Makefile.PL "recommends" versions.
839 - add AnyEvent->now_update.
840
8414.351 Sat Apr 11 07:56:14 CEST 2009
842 - actually make the signal pipe work under win32.
843 - localise $! in signal handler to avoid changing $!, although
844 perl itself does not seem to save/restore errno either.
845 - set the cloexec flag on the signal pipe (normally set by perl too).
846
8474.35 Fri Mar 27 11:48:20 CET 2009
848 - event models relying on AnyEvent's signal watcher emulation did
849 invoke the callback asynchronously, contradicting documentation
850 and causing signals to get lost (this includes AnyEvent's own
851 event loop). AnyEvent now uses the standard pipe trick to make
852 callback execution synchronous to the event loop.
853 - AnyEvent::Handle didn't free TLS context data on DESTROY
854 (patch by Pavel Shaydo).
855 - work around the ever-incompatibly-changing API of MakeMaker.
856 - document that changing global variables without restoring them
857 is a bad idea in Perl (noted by Adam Rosenstein).
858 - AnyEvent::Strict now barfs if ->io is passed a file.
859
8604.34 Thu Feb 12 18:32:45 CET 2009
861 - separately buffer TLS read data, as otherwise the read queue could
862 deadlock as receiving data is not expected while draining the
863 read queue (which cna only happen with TLS).
864 - raise EBADMSG error on JSON decoding errors.
865 - fix some minor manpage bugs (reported by Maximilian Gaß).
866 - speed up select bitmask parsing quite a bit in the pure perl
867 backend.
868 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
869
8704.331 Tue Jan 6 21:07:25 CET 2009
871 - socketpair fails on many vista machines because vista has
872 completely broken accept/getpeername and getsockname functions,
873 so we provide our own socketpair emulation that kind of works
874 (AnyEvent::Util::portable_pipe).
875 - new function: AnyEvent::Util::portable_socketpair.
876 - take advantage of the Guard module if it exists.
877
8784.33 Fri Nov 21 02:35:40 CET 2008
879 - AnyEvent::Strict did errornously flag a fileno of 0 as
880 illegal.
881 - reduce memory usage and slightly speed up the pure perl backend
882 by only storing the file descriptor, not the file handle.
883 - add missing autocork method to AnyEvent::Handle (reported
884 by Adam Rosenstein).
885 - AnyEvent::DNS->resolve errornously documented an $rcode result
886 argument, but there isn't (spotted by Henrik Krohns).
887 - the naked truth about Tk - it's basically unmaintained.
888
8894.32 Mon Nov 3 22:46:32 CET 2008
890 - fix AnyEvent::Socket::resolve_sockaddr to properly support
891 unix sockets again and choose a proper default for
892 the socket type.
893 - fix call to resolve_sockaddr in tcp_connect.
894
8954.31 Thu Oct 30 04:41:48 CET 2008
896 - implemented AnyEvent::Handle->destroy method.
897 - hint about unexpected effects in TLS mode.
898 - speed up AnyEvent::Impl::Perl by using more arrays
899 instead of hashes without the slightest loss of readability :)
900 - work around a perl argument refcounting bug.
901 - with some perl patches applied, AnyEvent now seems leak-free.
902
9034.3 Fri Oct 3 09:18:43 CEST 2008
904 - AnyEvent will now install a no-op signal handler for SIGPIPE,
905 unless one has been installed already.
906 - warn about not loaidng AnyEvent::Impl::POE early enough
907 (patch by Adam Rosenstein).
908 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
909 - fix a bug in where in SSL connect mode, the
910 client would first wait for some data by the server and
911 otherwise hang. (reported and analysed in an absolutely
912 exemplary manner by Adam Rosenstein).
913 - fix a bug in where SSL EOF would not be treated
914 as stream EOF, putting the connection into a hung state.
915 (reported and analysed in an absolutely exemplary manner by Adam
916 Rosenstein).
917 - fix a potential segfault when the TLS context
918 would go missing in a read callback (e.g. due to stoptls) -
919 Net::SSLeay of course makes no type checking whatsoever.
920 - AnyEvent::Handle will not stop the read watcher in TLS mode.
921 - AnyEvent::Handle->stoptls will send a TLS close notify
922 instead of simply destroying the stream now.
923 - fix a bug where large blocks of data written in TLS mode
924 would not be sent unless triggered by receives.
925 - on_drain will now take the tls write buffer into account.
926 - SSL operations have been streamlined a bit and should be faster
927 now, more to come, though.
928 - added a FAQ section, will document nontrivial issues in there.
929
9304.234 Mon Sep 29 04:08:13 CEST 2008
931 - fix child watcher documentation: only child exits will be reported,
932 no trace events.
933 - mention SIGPIPE in AnyEvent::Handle.
934 - perl backend will now use POSIX::times as monotonic clock
935 when available and the monotonic clock option is not.
936 - run even without Time::HiRes in some cases now, but do not
937 provide sub-second accuracy (all *supported* perl versions have
938 Time::HiRes).
939
9404.233 Fri Aug 22 01:48:05 CEST 2008
941 - fix a bug in the testsuite, causing a hang.
942 - clarified lots of AnyEvent::Handle constructor arguments.
943
9444.232 Thu Aug 21 20:44:25 CEST 2008
945 - fix a bug in AnyEvent::Handle that could cause two fatal error
946 callbacks on EOF, which caused AnyEvent::HTTP to signal success twice
947 (testcase by Leon Brocard).
948 - clarified on_eof/on_error documentation a bit.
949
9504.231 Tue Jul 29 13:12:15 CEST 2008
951 - remove some debugging code left in AnyEvent::Util::fork_call
952 (and no, it's impossible to implement with the broken windows
953 perls without resource leaks or worse).
954
9554.23 Tue Jul 29 12:19:59 CEST 2008
956 - document the first parameter passed to condvar callbacks to be
957 the callback.
958 - add AnyEvent::Socket::{ntoa,aton} aliases.
959 - optimize the AE::Handle->push_read (line) for the default
960 eol marker.
961 - optimize push_read (packstring|storable) for small packets.
962 - invoke on_error callback when no on_eof callback is set.
963 - fix a bug in push_read (storable) of unknown impact.
964
9654.22 Sun Jul 20 16:34:13 CEST 2008
966 - new function AnyEvent::Socket::parse_hostport.
967 - as the bulkheads at microsoft can't even get getprotobyname reliably
968 working on their shitty fucking broken os we need to hardcode
969 some common protocol numbers in AnyEvent::Socket. How can
970 people even bother with such a pile of shit as windows.
971
9724.21 Thu Jul 17 14:40:05 CEST 2008
973 - INCOMPATIBLE CHANGE: AnyEvent::Util::fork_call now has a prototype
974 which is incompatible to previous invocation syntax.
975 - work around more windows perl bugs in fork_call. I cannot imagine
976 why anybody would bother implementing fork in such an obviously
977 unusable way.
978 - avoid calling $do_retry if the request has already finished in
979 AnyEvent::DNS, thus avoiding an exception (reported by Anatoly
980 K. Sharifulin).
981
9824.2 Sat Jul 12 22:42:11 CEST 2008
4 - support an "interval" argument for repeating timers. 983 - support an "interval" argument for repeating timers.
5 - not importing anything from Event causes it to malfunction
6 and not use Time::HiRes, work around this bug.
7 - fix a bug in the pure perl event loop that caused timers to 984 - fix a bug in the pure perl event loop that caused timers to
8 be delayed under some circumstances. 985 be delayed under some circumstances.
9 - implement optional argument checking via PERL_ANYEVENT_STRICT. 986 - implement optional argument checking via PERL_ANYEVENT_STRICT.
987 - not importing anything from Event causes it to malfunction
988 and not use Time::HiRes, work around this bug.
10 - more bug workaroudns against the endlessly broken Tk module, 989 - more bug workarounds against the endlessly broken Tk module,
11 seems to work now on 32 bit machines, but less so on 64 bit. 990 seems to work now on 32 bit machines, but less so on 64 bit.
12 - minor optimisations applied to most backends. 991 - minor optimisations applied to most backends.
13 992
144.161 Fri Jul 4 14:24:48 CEST 2008 9934.161 Fri Jul 4 14:24:48 CEST 2008
15 - fixed the t/handle/02_write.t, it read after write, with the 994 - fixed the t/handle/02_write.t, it read after write, with the
… …
170 - support service names in tcp_connect and tcp_server. 1149 - support service names in tcp_connect and tcp_server.
171 - work around netbsd/openbsd bug causing unix sockets not to 1150 - work around netbsd/openbsd bug causing unix sockets not to
172 become writable until empty. 1151 become writable until empty.
173 - only make a dummy read for error detection when getpeername 1152 - only make a dummy read for error detection when getpeername
174 returns ENOTCONN. 1153 returns ENOTCONN.
175 - defend AnyEvent::Util::inet_aton against Coro::LWP brutal 1154 - defend AnyEvent::Util::inet_aton against Coro::LWP's brutal
176 override. 1155 override.
177 - new AnyEvent::Guard::cancel. 1156 - new AnyEvent::Guard::cancel.
178 1157
1793.5 Sat May 17 23:17:14 CEST 2008 11583.5 Sat May 17 23:17:14 CEST 2008
180 - gracefully deal with systems not defining CLOCK_MONOTONIC 1159 - gracefully deal with systems not defining CLOCK_MONOTONIC
… …
194 1173
1953.41 Sun May 11 19:53:13 CEST 2008 11743.41 Sun May 11 19:53:13 CEST 2008
196 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao. 1175 - work around a bug in perl 5.8.8's local. Reported by Yi Ma Mao.
197 1176
1983.4 Sun May 11 00:29:25 CEST 2008 11773.4 Sun May 11 00:29:25 CEST 2008
199 - complete re-implementation of condvars with a hopefully much mroe useful 1178 - complete re-implementation of condvars with a hopefully much more useful
200 API (backward compatible functions exist but are undocumented). 1179 API (backward compatible functions exist but are undocumented).
201 - AnyEvent::Handle fully rewritten with a hopefully more powerful API 1180 - AnyEvent::Handle fully rewritten with a hopefully more powerful API
202 (still subject to change). 1181 (still subject to change).
203 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl. 1182 - take advantage of CLOCK_MONOTONIC, if available, in AnyEvent::Impl::Perl.
204 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which 1183 - provide AnyEvent::post_detect and @AnyEvent::post_detect, which

Diff Legend

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