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

Comparing AnyEvent/Changes (file contents):
Revision 1.235 by root, Wed May 6 03:45:27 2009 UTC vs.
Revision 1.609 by root, Sun Mar 6 03:04:28 2016 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 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?
15TODO: AnyEvent::TLS eval NEt::SSLeay gt versiobn test incorrect syntax
16
177.12 Wed Jan 27 19:12:26 CET 2016
18 - use common name as hostname for tls connects, if Net::SSLeay
19 supports SNI.
20 - fix documentation of tls_autostart read type in AnyEvent::Handle,
21 analyzed by Felix Ostmann.
22
237.11 Thu Jul 16 14:36:00 CEST 2015
24 - AnyEvent::Socket::parse_ipv6 could accept malformed ipv6
25 addresses (extra "::" at end and similar cases).
26 - add a more explicit warning to AnyEvent::Handle that it doesn't
27 work on files, people keep getting confused.
28 - new function AnyEvent::Socket::tcp_bind.
29 - new functions AnyEvent::fh_block and AnyEvent::fh_unblock.
30 - aligned ipv6 address formatting with RFC 5952 (by not shortening
31 a single :0: to ::).
32 - added stability canary support.
33
347.09 Sat May 2 16:38:53 CEST 2015
35 - AnyEvent::Debug called an internal function (AnyEvent::Log::ft)
36 that was renamed to AnyEvent::Log:format_time. uinder its old name
37 (analyzed by Michael Stovenour).
38 - update AnyEvent::DNS fallback resolver addresses:
39 seems google effectively killed most other free dns resolvers,
40 so remove them, but add cable and wireless (ecrc) since it was
41 stable for 20 years or so, official or not, and there should be
42 an alternative to google.
43 - perl5porters broke windows error codes in 5.20, and mapped
44 WSAEWOULDBLOCK on the (different) EWOULDBLOCK error code, and
45 WSAEINPROGRESS into the incompatible ERINPROGRESS code, probably
46 because they were so cool. They probably broke other error codes
47 for no reason, too, but I didn't care for checking, it's too
48 depressing. This version only works around the WSAEWOULDBLOCK
49 issue, because I don't have a nice way to work around the
50 WSAEINPROGRESS bug.
51
527.08 Wed Dec 10 05:27:17 CET 2014
53 - work around a newly introduced bug in Socket 2.011 (an
54 errornous sun_length check) (analyzed by Maxime Soulé).
55 - AnyEvent::TLS didn't load (but refer to) AnyEvent::Socket
56 (analyzed by Ben Magistro).
57 - AnyEvent::Strict will now confess, not croak. This is in line with
58 it being a development/debugging tool.
59 - work around a number of libglib bugs (debug builds of libglib enforce
60 certain undocumented behaviour patterns such as not being able to
61 remove a child watch source after it has fired, which we will try
62 to emulate to avoid "criticals". what where they thinking...).
63 - mention json security issues in AnyEvent::Handle, now that Douglas
64 Crockford has foolishly and incompatibly changed JSON.
65 - changed default dns resolver "max_outstanding" value from 1 to 10,
66 the latter beinfg the intended value all along
67 (reported by Ilya Chesnokov).
68 - added new "AnyEvent::Impl::UV" interface module to the UV event lib
69 (written by Mike Lowell).
70
717.07 Tue Dec 17 17:45:02 CET 2013
72 - the documentation for custom tls verify schemes was wrong. make it agree
73 with the code (reported by Maxime Soulé).
74 - added cbor read and write types to AnyEvent::Handle (using CBOR::XS).
75 - work around an API change in openssl that could cause wrong tls connection
76 aborts, likely on windows only (analyzed by sten).
77 - calling AnyEvent->now_update with AnyEvent::Impl::Perl caused an
78 endless loop (reported by Dietrich Rebmann).
79 - add tlsv1_1 and tlsv1_2 protocols to AnyEvent::TLS
80 (patch by Maxime Soulé).
81 - document AnyEvent::Impl::IOAsync::set_loop and
82 $AnyEvent::Impl::IOAsync::LOOP. Though only documented now, this
83 functionality has _always_ been available.
84 - force a toplevel domain name in t/81_hosts.t (analyzed by
85 David Jack Wange Olrik).
86 - document that AnyEvent::Log uses AnyEvent::IO.
87 - warn about AnyEvent::Filesys::Notify performance.
88 - praise the joys of AnyEvent::Fork::*.
89 - time for an =encoding directive.
90 - do no longer use JSON to create a default json coder, use
91 JSON::XS or JSON::PP directly.
92
937.05 Wed Aug 21 10:38:08 CEST 2013
94 - uts46data.pl couldn't be found due to wrong naming of the file
95 (reported by Fulko Hew).
96 - handle lone \015's properly in AE::Handle's default line read
97 (reported by various people).
98 - untaint ip addresses found in /etc/hosts (patch by José Micó).
99 - the memleak fix in 7.03 caused resolving via /etc/hosts to always
100 fail on first use (reported and testcase by Andrew Whatson).
101 - expose AnyEvent::Log::format_time, and allow users to redefine it.
102 - expose AnyEvent::Log::default_format, and allow redefinition.
103 - expose AnyEvent::Log::fatal_exit, to allow redefinition.
104 - AnyEvent::Debug shell can now run coro shell commands, if available.
105 - t/63* tests were wrongly in MANIFEST.
106 - kernel.org's finger server went MIA, switch to freebsd.org and
107 icculus.org.
108 - clarify that IO::AIO and AnyEvent::AIO are needed for AnyEvent::IO
109 to function asynchronously (reported by Daniel Carosone).
110 - hard-disable $^W in most tests, it generates too much garbage output.
111 - use a (hopefully) more future-proof method to emulate common::sense.
112 - upgrade to UTS-46:6.2.0.
113 - switch to INSTLIB from INSTLIBDIR, as INSTLIBDIR was wrongly
114 documented. should not affect anything.
115
1167.04 Thu Nov 15 02:23:58 CET 2012
117 - debugging code left in.
118
1197.03 Thu Nov 15 02:16:11 CET 2012
120 - AnyEvent::Socket::inet_aton did not work when dns resolution was
121 used to find the addresses (analyzed and patch by Maxime Soulé).
122 - fix a memory leak in the /etc/hosts lookup code when hosts don't
123 resolve and are not in hosts (analyzed by sten).
124
1257.02 Tue Aug 14 04:27:58 CEST 2012
126 - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test
127 program by Yorhel).
128 - verified that AnyEvent::Socket follows RFC5952.
129 - try to parse "ADDR#PORT" in addition to "ADDR PORT".
130
1317.01 Sun May 13 01:03:17 CEST 2012
132 - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but
133 not available, instead of throwing an exception.
134 - use File::Spec to get the tmpdir in t/*, to avoid needless
135 failures on (most, not mine :) windows boxes.
136 - new handle read types: tls_detect and tls_autostart.
137
1387.0 Fri Apr 13 06:33:30 CEST 2012
139 - child watchers are broken in POE 1.352 (also many earlier
140 versions) and there seems to be no way to work around it, as POE
141 itself is inherently racy. Document this regression and add a delay
142 in t/68_poe_03_child.t for the time being.
143 - new module AnyEvent::IO, that is a frontend to either a pure-perl
144 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
145 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
146 - load /etc/hosts only when DNS has no answer.
147 - stat /etc/hosts on every access and reload it if it changed.
148 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
149 - fix a buggy croak in the dh parameter reading in AnyEvent::TLS.
150 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
151 a side effect, they now use true appending as opposed to libc
152 appending, and the file might not have been opened when the function
153 returns.
154 - the default logging level was not properly documented in a variety of
155 places, this has hopefully been rectified.
156 - updated uts46data.pl for unicode 6.1.0.
157 - made log messages generated by AnyEvent submodules not
158 include the package name anymore, as it will be logged
159 by default already.
160 - upgrade to the trick used by common-sense 3.5 to work
161 around extra warning torture/breakage under perl 5.15.x.
162 - log messages by anyevent are now ucfirst, are usually full
163 sentences and do no longer include the package name.
164 - the storable read type would sometimes throw an exception instead
165 of causing EBADMSG (analyzed by Maxime Soulé).
166 - work around a bug in openssl 1.0.1 which enforces a minimum rsa
167 keysize (reported by Paul Howarth).
168 - documented Rocco fabricating statements about AnyEvent and me.
169
1706.14 Tue Jan 31 20:00:24 CET 2012
171 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
172 probably an editing mistake (reported by Darin McBride).
173
1746.13 Thu Jan 12 07:27:01 CET 2012
175 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
176 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
177 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
178 regex-matching inside the callback would change the parameter, despite
179 $1 being dynamically scoped per-block. this perl bug is now being
180 worked around (testcase by Cindy Wang).
181
1826.12 Mon Dec 12 13:21:10 CET 2011
183 - $! was clobbered by subsequent calls in tcp_connect, due to
184 the postpone not saving/restoring it (analyzed by Richard Garnish).
185
1866.11 Tue Nov 22 10:36:05 CET 2011
187 - Tk cannot create windows when tainted unless you set a title, so
188 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
189 - escape any nonprintable/nonascii characters when stringifying
190 backtraces.
191 - log the reason loading an interface module fails to load at level debug.
192
1936.1 Tue Oct 4 19:44:30 CEST 2011
194 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
195 and above), and some messages inside AnyEvent have been elevated
196 to higher log levels to print by default.
197 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
198 - work around yet another signal race bug, newly introduced by POE. My
199 how that module sucks.
200 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
201 support ipv6-only operation even with defaults (also use ::1 instead
202 of 127.0.0.1).
203 - give better guidelines on how to chose a logging level.
204 - AnyEvent::Log can now cap (limit) log message priority by context.
205 - AE::log will now save and restore $! and $@, so callers don't have to.
206 - AE::Strict now checks that registered fds are still valid at regular
207 intervals.
208 - skip frequently called 'require' calls if the module appears loaded.
209 - increase signal latency test timer considerably for poor little osx,
210 which can't be bothered with basic posix functionality such as signal
211 delivery.
212 - mention some debugging aids in the tutorial.
213 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
214 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
215 of the context as title.
216 - AnyEvent::Strict would not restore read-write access to $_ when a callback
217 died - this is a user bug, but since AE::Strict exists to help dbeugging...
218 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
219 to work with it - the author is well aware of the issues but refuses
220 to fix them. This extreme step has been taken because the module actively
221 attacks the goals of AnyEvent and makes it harder for module authors to use
222 AnyEvent.
223
2246.02 Fri Aug 26 20:08:31 CEST 2011
225 - log_to_syslog now accepts facility strings.
226 - log_to_syslog would not allow facilities - "$facility|$level"
227 *literally* means that string. sheesh.
228 - the CPAN parser got confused and thought we provide Sys::Syslog, try
229 to work around.
230 - renamed AnyEvent::Impl::FLTK2 to FLTK.
231
2326.01 Fri Aug 26 07:04:11 CEST 2011
233 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
234 the TTL as fourth element - this affects ->resolve, ->request
235 and ->dns_unpack indirectly. Convenience functions such as
236 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
237 AnyEvent::Socket.
238 - INCOMPATIBLE CHANGE: backend authors now should not implement
239 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
240 methods.
241 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
242 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
243 - DEPRECIATION: please change
244 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
245 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
246 in your programs, for future compatibility.
247 - *blush* idle watcher emulation was borked.
248 - AnyEvent::Strict now checks the AE::xxx functions as well.
249 - querying too long domain names would barf deep inside AnyEvent::DNS - since
250 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
251 for too long domain names explicitly and passes an undef to the callback
252 instead (reported by James Bromberger).
253 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
254 - new functions: AnyEvent::postpone, AnyEvent::log.
255 - new module: AnyEvent::Log, for simple logging needs.
256 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
257 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
258 a debug shell.
259 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
260 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
261 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
262 /etc/hosts can be found.
263 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
264 configuration.
265 - env variables can now be specified using AE_xyz instead of
266 PERL_ANYEVENT_xyz, subject to some rules.
267 - lots of new AnyEvent::Debug functionality.
268 - AnyEvent::Strict now checks for modifications of $_ by the callback
269 (by setting it to readonly while executing the callback, if possible).
270 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
271 Invest extra overhead to work around this peculiar behaviour.
272 - split the pure perl backend into a loop module and an impl
273 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
274 - some read types (regex, netstring, json, storable) could
275 errornously keep an AE::Handle object alive on their own
276 (reported by Mohammad Toossi).
277 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
278 supports /etc/hosts overrides.
279 - add more workarounds around child watcher bugs in IO::Async
280 and (newly introduced) in POE.
281 - use glib child watchers and try to work around its limitations,
282 if possible (based on a patch by Kevin Ryde).
283 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
284 - many read types in AnyEvent::Handle would malfunction during
285 transitions to TLS because of a "delete $self->{rbuf}".
286 (reported by Gerald Galster).
287 - improve condvar blocking wait performance (probably only noticable
288 with faster backends), also make it easier for event loops that
289 do not support blocking waits.
290 - PERL_ANYEVENT_MODEL now also allows full module names, although
291 the usefulness of this feature is not yet clear.
292 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
293 - add t/80_ssltext.t.
294 - no longer use AUTOLOAD for the initial loading (for no very good reason).
295 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
296 - do not run event loop detection multiple times when programs call their cached
297 copies of AnyEvent::detect.
298 - parse_hostport now accepts absolute paths als unix domain sockets.
299 - suppress the idiotic warning inside POE by patching the POE::Kernel
300 run flags instead of asking the user to jump through contortions.
301 - add Task::Weaken as dependency (as recommends only at this time).
302 - replace PF_UNSPEC by 0, for good old minix.
303
3046.0 (not officially released, but was on CPAN for about 15 minutes :/)
305
3065.34 Thu May 12 09:59:41 CEST 2011
307 - AE::TLS didn't support both cert_file and key at the same time,
308 due to some copy&paste bug (reported and analyzed by vti).
309 - improve AE::Handle timeout documentation slightly.
310
3115.33 Mon Feb 7 21:51:41 CET 2011
312 - never appeared on CPAN, for unknown reasons.
313 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
314 object when it's already gone (analyzed by Ryan Bullock).
315 - update idna algorithm(s) to UTS#46 version 6.0.0.
316 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
317 uts46 says it's an error, and testsuite makes more sense.
318 - remove leading dots in idn_to_ascii - testsuite says to remove,
319 uts46 says it's an error, and testsuite makes more sense.
320 - fixed a bug in idn ignored character processing.
321 - passes the uts46 6.0.0 testsuite except for three cases which
322 apparently are bugs in the testsuite itself.
323 - passes the uts46 6.0.1 testsuite except for eleven cases which
324 apparently are bugs in the testsuite itself.
325
3265.31 Sun Jan 23 11:49:19 CET 2011
327 - work around a POE bug where POE would first poll for new events and
328 invoke handlers BEFORE handling the message queue.
329 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
330 - stop_read now does just that in TLS mode. see its
331 documentation for caveats.
332 - update to common::sense 3.4 warning set (some warnings in perl impose
333 an arbitrary 32kb limit on regex matches, which affects
334 AnyEvent::Handle).
335 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
336 - new experimental AE::Handle::resettls method.
337
3385.3 Fri Dec 31 05:48:13 CET 2010
339 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
340 issues with IO::Async have been solved in that version, including
341 the availabilty of a default loop!
342 - implement new wbuf_max parameter for AnyEvent::Handle.
343 - added Cocoa::EventLoop backend by Daisuke Murase.
344 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
345 - clarify handle on_prepare documentation.
346 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
347 to avoid leaking them to any child processes.
348 - try to force files into the arch-dependend lib directory, for the
349 half-"turly obsessive compulsive person"(s) that depend on it.
350
3515.29 Sun Dec 5 10:49:21 CET 2010
352 - convert EV backend to EV 4.00 API (so better upgrade EV too).
353 - AE::Handle: implement dynamic read_size adjustment, add
354 max_read_size parameter, reduce default read_size to 2048.
355 - add some FAQ entries for servers and/or on_eof vs. on_error.
356 - work around OS bugs (cygwin again) when lingering in AE::Handle.
357 - reduce memory usage slightly when connect was used in AE::Handle.
358 - wrok around more segfaults in Tk.
359 - document tls_ctx => undef as valid.
360 - detect WNOHANG value at compiletime.
361 - include some probably linux-only support for building AnyEvent
362 as part of the perl core.
363 - improve compatibility of t/08_idna.t to perl 5.8.
364 - make AnyEvent::Strict truly optional (in case somebody builds a
365 minimal perl...).
366
3675.28 Wed Oct 13 04:14:23 CEST 2010
368 - due to a glitch, AnyEvent's internal getprotobyname wasn't
369 actually used in tcp_server.
370 - implement AnyEvent::CondVar->new method, which was documented but
371 missing.
372 - support autoloading of read types in unshift_read.
373 - AnyEvent::Strict: do the test for invalid fh arguments differently,
374 resulting in more informative error messages.
375 - new AnyEvent::FAQ manpage.
376 - clarify that you can actually call push_read etc. in on_read.
377 - put keys and certs in /tmp, not into the current directory,
378 and document this in the SECURITY CONSIDERATIONS sections
379 of AnyEvent::TLS.
380 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
381 requested by Ron Isaacson.
382 - document $AnyEvent::DNS::RESOLVER variable.
383 - incorporated major typo patches by Abhijit Menon-Sen.
384 - AnyEvent::Handle now croaks when negative timeouts are passed.
385 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
386 undocumented, but who knows when they might come in handy.
387
3885.271 Tue Jun 8 12:05:46 CEST 2010
389 - backport to perl 5.8.x.
390
3915.27 Sun Jun 6 12:12:05 CEST 2010
392 - postpone differently in AnyEvent::Socket now, as
393 when not, canceling the connection attempt might fail
394 (found by Felix Antonius Wilhelm Ostmann).
395 - explicitly check for non-stream sockets in AE::Handle, too many
396 clueless people fell into the trap of this somehow working.
397 - simplified and reworked the "OTHER MODULES" section.
398 - better/more condvar examples.
399
4005.261 Wed Apr 28 16:13:36 CEST 2010
401 - AF_INET6 was not properly used from Socket6 during configuration
402 time (found by Andy Grundman).
403
4045.26 Mon Apr 12 04:49:35 CEST 2010
405 - don't generate spurious readyness notifications when select
406 returns EINTR in the pure perl backend (analysed and testcase
407 by Bas Denissen).
408 - give same examples in SYNOPSIS sections of AnyEvent and AE.
409 - provider faster implementations for AE::time/now/now_update for
410 EV and Perl backends.
411 - provide faster AE::cv and AE::time implementations for all
412 backends.
413 - no longer support lower/mixed-case signal names (this was
414 never documented nor universally supported).
415 - some more memory µ-optimisations, and somewhat less messy
416 code to cope with both AnyEvent and AE APIs.
417
4185.251 Sat Mar 13 00:58:21 CET 2010
419 - make test failed if it was run before make install on perl <=
420 5.8.8, and older versions of perl make our live very difficult,
421 so write constants.pl during Makefile.PL time.
422
4235.25 Sat Mar 13 00:23:14 CET 2010
424 - fix a race condition in AnyEvent::Handle that would cause
425 a "bio_read: unsupported method" error (in _another_
426 ssl connection) after user code threw an exception.
427 - added AnyEvent::Handle->destroyed method.
428 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
429 - the AnyEvent::Util::fh_nonblocking function worked only
430 by lucky accident on win32.
431 - smaller and faster AnyEvent::Util::fh_nonblocking.
432 - when the (required!) Time::HiRes module is missing, AnyEvent
433 did not fall back to built-in time properly.
434 - do not load Fcntl at runtime, saving memory and loading time.
435 - precompile a number of constants and use them instead of runtime
436 detection and eval.
437 - free detection code after detection and similar memory optimisations.
438 - Perl backend timer interval best effort drift has been improved
439 (same algorithm as EV).
440 - update unicode idna mapping table.
441
4425.24 Tue Jan 5 11:39:43 CET 2010
443 - cygwin never reports errors from failed connects. 1.5 just gives
444 you continous readyness and EAGAIN, 1.7 is even more broken and
445 just hangs. work around both issues in a cygwin-specific hack.
446 - improve idle watcher documentation slightly.
447
4485.23 Sun Dec 20 23:48:00 CET 2009
449 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
450 - implement punycode_encode/decode, idn_nameprep,
451 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
452 - provide $AE::VERSION.
453 - removed traces of "no strict 'refs'".
454
4555.22 Sat Dec 5 03:51:13 CET 2009
456 - downgrade-or-fail in AnyEvent::Handle::push_write, to
457 diagnose encoding failures earlier and more succinctly.
458 (this works around bugs in perl, throwing away encoding info
459 when passing scalar data to extensions).
460 - add more examples to AnyEvent::Socket manpage.
461 - upgrade internal warning set to the same as common::sense 2.03.
462 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
463 (John Beppu).
464 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
465 - add google public dns servers to fallback server set and make sure
466 we load-balance properly between the three sets. also add all
467 fallback dns servers, not just a random one, to each dns config.
468
4695.21 Thu Nov 19 02:48:47 CET 2009
470 - fix a problem where socket constants were called with parameters
471 (spotted by David Friedland).
472 - fork_call never use'd POSIX (reported by Daisuke Maki).
473 - improve perl 5.6 compatibility further (but it still won't work
474 unless you rip out everything but the core).
475 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
476 uses 10MB of RAM, but doesn't open a console window. *sigh*.
477
4785.202 Wed Oct 14 22:35:44 CEST 2009
479 - AnyEvent::DNS would unexpectedly clobber $_ under windows
480 (analysed by Matthias Waldorf).
481 - AnyEvent::Handle::run_cmd can now pass the PID of the
482 newly-created process, which is much less useful than it might
483 sound (based on patch by Yann Kerherve).
484
4855.201 Tue Sep 29 12:09:25 CEST 2009
486 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
487 (reported by Torsten Foertsch).
488 - common::sense 2.0 could cause tcp_server to throw an exception
489 (analysed by elmex).
490
4915.2 Mon Sep 14 07:04:49 CEST 2009
492 - INCOMPATIBLE CHANGE: do no longer support register_read_type
493 and register_write_type in AnyEvent::Handle, instead support
494 package names (the facility was mostly abused).
495 - implement "packagename-as-read/write type" support in
496 AnyEvent::Handle.
497 - AnyEvent::Handle: new options "keepalive" and "oobinline".
498 - oobinline set by default to avoid security issues.
499 - the pure-perl event loop backend wrongly detected times() fallback
500 support (spotted by Pavel Boldin).
501
5025.12 Tue Sep 1 20:26:50 CEST 2009
503 - be more lenient when parsing resolv.conf files, as some people
504 use hashmarks as comment indicator inside directives (reported
505 by Michael S. Fischer).
506 - use same set of warnings as common::sense 2.0.
507 - fix a potential 32 bit overflow issue due to perl having problems
508 with large hex constants in the Perl backend.
509 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
510 by default.
511 - fix some prototypes.
512 - (optionally) require Async::Interrupt 1.02.
513 - replace opendns fallbacks by verizon/level3 public dns servers.
514
5155.112 Fri Aug 21 13:59:04 CEST 2009
516 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
517 using starttls with an illegal/uncreatable tls context.
518 - port to broken win32 perls.
519
5205.111 Sun Aug 16 18:44:36 CEST 2009
521 - perl errornously requires sockaddr_un structures returned by the
522 kernel to have a specific length. work around this bug by padding
523 sockaddr_un structures (found to fail on OS X, as perl only uses
524 the correct code for GNU/Linux, but likely to fail on other
525 OSes as well).
526
5275.11 Wed Aug 12 17:49:37 CEST 2009
528 - fixed a bug in run_cmd with input from filename.
529 - create a json coder object when none is given in write direction
530 to reduce dependency on JSON.
531
5325.1 Tue Aug 11 03:17:32 CEST 2009
533 - tcp_connect now reports the correct error in $!, not an unrelated
534 one (was broken since 4.91).
535 - AE::Handle did not properly clear rtimeout/wtimeout watchers
536 when retrying a tcp_connect.
537 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
538 - fixed wrong documentation in AE::Socket.
539
5405.01 Mon Aug 10 03:16:32 CEST 2009
541 - last release broke EV child watchers :/.
542
5435.0 Sun Aug 9 17:32:09 CEST 2009
544 - officially introduce and document the AE API.
545 - lowering the timeout at runtime did not have immediate effect
546 in AnyEvent::Handle.
547 - AE::Handle now has separate and independent read and write timeout
548 settings.
549 - probe for Guard module when AnyEvent::Util is loaded, not at
550 runtime (Event doesn't like eval "use" from callbacks, and it avoids
551 exporting the probe function).
552 - Impl::IOAsync's io watchers did not use the filehandle emulation
553 code to work around it's watcher limits.
554 - work around IO::Async's problems with overloaded objects.
555 - try to disable wearnings inside IO::Async, it's just too buggy.
556 - do no longer use Event by default, only EV or Perl, Event is too buggy.
557 - hackishly provide fast AE:: interface when EV or Perl is the backend.
558 - inherit VERSION from AnyEvent::VERSION in most modules.
559
5604.91 Thu Aug 6 15:42:45 CEST 2009
561 - AE::Handle::starttls could get out of sync when the read buffer
562 already contains some TLS handshake.
563 - AE::Handle did not properly free the TLS session, causing bigger
564 memory leaks in Net::SSLeay than expected (reported by toaster).
565 - AE::Socket::tcp_connect will now artificially delay invoking
566 the callback to avoid returning after invoking the callback.
567 - convert many internal watcher uses to AE API.
568
5694.9 Sat Aug 1 11:07:01 CEST 2009
570 - Glib silently fails when registering a timeout with
571 a negative value, so avoid that.
572 - call condvar callback immediately when it is set
573 after the condition is already signalled.
574 - check rbuf_max condition only after trying to consume
575 data.
576 - normalise signal numbers to names when using the ae's
577 signal handling, but do not document this (yet).
578 - pure perl signal emulation did not properly set nonblocking
579 mode on the signal pipe on win32 (this is unlikely to have caused
580 any issues).
581 - new module: AnyEvent::Debug.
582 - AnyEvent::Strict now checks that a signal was specified
583 by name (not name or number).
584 - reduce memory footprint in typical cases by ~50kb by
585 only compiling signal code when necessary.
586 - add AnyEvent::Handle->rbuf_max.
587 - grab the AE namespace for future new API, implement stubs for
588 said future/alternative API.
589 - new function: AnyEvent::Socket::format_hostport.
590
5914.881 Tue Jul 28 12:51:53 CEST 2009
592 - work around a bug in local in pre-5.10 perls, causing
593 AnyEvent::Handle to recurse when it should not
594 (analysed by elmex).
595
5964.88 Tue Jul 28 04:04:37 CEST 2009
597 - re-bless the handle into a dummy package after calling
598 AnyEvent::Handle::destroy, so the user does not need to check
599 for errors after every push_write etc.
600 - do not attempt to run t/02_signals.t on obviously broken
601 platforms.
602
6034.87 Sun Jul 26 02:06:16 CEST 2009
604 - do not attempt to linger when there is no longer a valid fh.
605 - enforce tls mode to be either accept or connect, do not simply
606 segfault in Net::SSLeay.
607 - AnyEvent::Handle can now call tcp_connect itself (new parameters
608 connect, on_prepare, on_connect and on_connect_error). Updated
609 tutorial accordingly.
610 - add AnyEvent::Impl::Irssi backend.
611
6124.86 Mon Jul 20 23:52:29 CEST 2009
613 - since the verbose warning is not enough, explicitly document
614 that versions before 1.33 of Net::SSLeay are not secure.
615 - work around signal handling races in Event and (...) Event::Lib.
616 - try to align signal-race timer to full-second boundaries.
617 - work around Tk not liking negative timeouts.
618 - don't complain of different grades of environmental unfriendlyness
619 in IO::Async.
620
6214.85 Sat Jul 18 06:16:14 CEST 2009
622 - nail the signal race problem in perl once and for all
623 (see $AnyEvent::MAX_SIGNAL_LATENCY).
624 - take advantage of Async::Interrupt if it is available.
625 - load Time::HiRes and Guard modules on demand only.
626 - add optional/recommended modules section to AnyEvent
627 documentation.
628 - reduce memory usage considerably (and reduce startup penalty)
629 by not using "strict", "warnings" and "overload" modules.
630 - work around buggy windows/openbsd perls and provide EBADMSG
631 and EPROTO ourselves when missing.
632 - improve perl 5.6 compatibility of the core event loop.
633 - made Net::SSLeay version 1.33 a soft requirement.
634
6354.83 Fri Jul 17 16:56:26 CEST 2009
636 - implement AnyEvent::Socket::getprotobyname.
637 - AnyEvent::CondVar's will now detect recursive blocking
638 waits and will croak, as too many people fall into
639 this trap.
640 - AnyEvent::Handle will now call ->destroy on itself after
641 executing the on_error callback, instead of doing some
642 half-baked internal shutdown, for fatal errors.
643 - clarify on_eof behaviour w.r.t. the read queue and
644 on_read callbacks.
645 - ignore some possible spurious wake-ups in tcp_connect.
646
6474.82 Sat Jul 11 00:34:55 CEST 2009
648 - POE and Event backends didn't accept some callable objects as
649 callbacks.
650 - use Config module instead of POSIX module to detect signal names
651 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
652 module doesn't even have all POSIX signals :/.
653 - use more workarounds around the many refcnt/corruption bugs in
654 Event::Lib.
655 - work around a race condition in perl's select, causing t/03_child.t
656 to rarely fail.
657
6584.81 Thu Jul 9 10:30:30 CEST 2009
659 - AnyEvent::Handle didn't properly diagnose write errors
660 (it expected -1 from syswrite, how lame... :).
661 - support file descriptors in addition to file handles
662 in AnyEvent->io.
663 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
664 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
665 and PERL_ANYEVENT_CA_PATH.
666 - provide a sensible synopsis section for AnyEvent::TLS.
667 - add a "supported backends" section to the manpage.
668 - added simple io watcher test to testsuite, using a
669 portable_socketpair.
670 - tried to improve the stability of the Event::Lib backend,
671 YMMV.
672
6734.8 Mon Jul 6 23:45:16 CEST 2009
674 - AnyEvent::DNS did not properly follow CNAME records with
675 uppercase targets.
676 - AnyEvent::DNS would errornously return AAAA records
677 with v4 mapped addresses (a faulty record) as ipv4 addresses,
678 causing AnyEvent::Socket to throw an exception.
679 - added new module AnyEvent::TLS for easier SSL/TLS context
680 creation, with many options including hostname verification,
681 secure default configuration, lots of documentation and,
682 predefined diffie-hellman keys for perfect forward security
683 and much more. get it while it's still fresh!
684 - use AnyEvent::TLS in AnyEvent::Handle for context management.
685 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
686 so AnyEvent::Socket users have smaller memory footprint
687 in the common case.
688 - add AnyEvent::Handle->push_shutdown method.
689 - add an additional error message parameter to AnyEvent::Handle's
690 on_error callback (for TLS, $! is still available).
691 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
692 - make AnyEvent::Handle more robust against early connection
693 failures (during new), and return C<undef> in that case
694 from the constructor.
695 - AnyEvent::Handle will now try to load only JSON::XS first,
696 then fall back to JSON.
697 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
698 and because it was documented that way.
699
7004.451 Fri Jul 3 00:28:58 CEST 2009
701 - do not clear rbuf when shutting down an AnyEvent::Handle
702 object - doing so breaks AnyEvent::HTTP.
703
7044.45 Mon Jun 29 22:59:26 CEST 2009
705 - a write error could cause AnyEvent::Handle to create
706 an I/O watcher with an undefined $fh.
707 - special-case mapped ipv4 addresses in both
708 AnyEvent::Socket::format_address and parse_address,
709 to treat them just like ipv4 addresses.
710 - updated and overhauled the AnyEvent::Intro doc.
711 - implement AnyEvent::Socket::format_ipv4/ipv6.
712 - slightly speed up portable_pipe/socketpair.
713 - expand condvar begin/end documentation.
714
7154.42 Fri Jun 26 08:32:18 CEST 2009
716 - preliminary and neccesarily incomplete support for IO::Async.
717 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
718 case it was set to IGNORE, to ensure we can catch child statuses
719 even when the calling env acts stupidly.
720 - updated benchmarks with IO::Async, which performs very well.
721
7224.412 Wed Jun 24 01:35:57 CEST 2009
723 - support an "untaint" attribute for AnyEvent::DNS
724 and set it on the default resolver.
725 - implement and document AnyEvent::Impl::Perl::loop.
726 - remove all anyevent-env variables from %ENV
727 when running in tainted mode.
728 - mention and extend the IO::Lambda benchmark.
729
7304.411 Sun Jun 7 18:48:02 CEST 2009
731 - do not try to use F_SETFD on windows.
732
7334.41 Thu May 14 06:40:11 CEST 2009
734 - work around issues in older perls (5.8.0?) when
735 a signal handler is deleted from the %SIG hash.
736 - use POSIX::_exit in child test, to avoid running
737 destructors.
738 - speed up CHLD handling by relying on SIGCHLD being
739 synchronously delivered, even when we roll our own
740 implementation.
3 - AnyEvent::DNS: add the "dname" resource record name. 741 - AnyEvent::DNS: add the "dname" resource record name
742 for cosmetic reasons.
4 743
54.4 Sun Apr 26 20:12:33 CEST 2009 7444.4 Sun Apr 26 20:12:33 CEST 2009
6 - implemented idle watchers, where applicable. 745 - implemented idle watchers, where applicable.
7 - AnyEvent->time died when Event backend was in use. 746 - AnyEvent->time died when Event backend was in use.
8 - fix a memleak in the Tk backend. 747 - fix a memleak in the Tk backend.
9 - sped up Tk timer handling. 748 - sped up Tk timer handling.
10 - clip negative "after" values to 0 in AnyEvent::Impl::Event 749 - clip negative "after" values to 0 in AnyEvent::Impl::Event
11 to avoid spamming out warning messages. 750 to avoid spamming out warning messages.
12 - fix Qt timers without interval. 751 - fix Qt timers without interval.
13 - avoid Qt zero-timeout specialcase, in old libqt's. 752 - avoid Qt zero-timeout special case, in old libqt's.
14 753
154.352 Mon Apr 20 16:31:11 CEST 2009 7544.352 Mon Apr 20 16:31:11 CEST 2009
16 - fix AnyEvent::Strict error messages for child watchers. 755 - fix AnyEvent::Strict error messages for child watchers.
17 - fix/update Makefile.PL "recommends" versions. 756 - fix/update Makefile.PL "recommends" versions.
18 - add AnyEvent->now_update. 757 - add AnyEvent->now_update.
… …
824.3 Fri Oct 3 09:18:43 CEST 2008 8214.3 Fri Oct 3 09:18:43 CEST 2008
83 - AnyEvent will now install a no-op signal handler for SIGPIPE, 822 - AnyEvent will now install a no-op signal handler for SIGPIPE,
84 unless one has been installed already. 823 unless one has been installed already.
85 - warn about not loaidng AnyEvent::Impl::POE early enough 824 - warn about not loaidng AnyEvent::Impl::POE early enough
86 (patch by Adam Rosenstein). 825 (patch by Adam Rosenstein).
87 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 826 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
88 - fix a bug in where in SSL connect mode, the 827 - fix a bug in where in SSL connect mode, the
89 client would first wait for some data by the server and 828 client would first wait for some data by the server and
90 otherwise hang. (reported and analyzed in an absolutely 829 otherwise hang. (reported and analysed in an absolutely
91 exemplary manner by Adam Rosenstein). 830 exemplary manner by Adam Rosenstein).
92 - fix a bug in where SSL EOF would not be treated 831 - fix a bug in where SSL EOF would not be treated
93 as stream EOF, putting the connection into a hung state. 832 as stream EOF, putting the connection into a hung state.
94 (reported and analyzed in an absolutely exemplary manner by Adam 833 (reported and analysed in an absolutely exemplary manner by Adam
95 Rosenstein). 834 Rosenstein).
96 - fix a potential segfault when the TLS context 835 - fix a potential segfault when the TLS context
97 would go missing in a read callback (e.g. due to stoptls) - 836 would go missing in a read callback (e.g. due to stoptls) -
98 Net::SSLeay of course makes no type checking whatsoever. 837 Net::SSLeay of course makes no type checking whatsoever.
99 - AnyEvent::Handle will not stop the read watcher in TLS mode. 838 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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