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

Comparing AnyEvent/Changes (file contents):
Revision 1.376 by root, Mon Jan 11 17:41:59 2010 UTC vs.
Revision 1.539 by root, Fri Mar 30 21:31:52 2012 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: AnyEvent::IO
6tODO: maxim souffle handle storable
7TODO: document import tags aeio, also ae_ naming
8TODO: t/io
9TODO: use the eval-with-cb-call trick also to speed up JSON?
10lib/AnyEvent/DNS.pm: open my $fh, "<", $resolv_conf
11lib/AnyEvent/DNS.pm: if (open my $fh, "ipconfig /all |") {
12lib/AnyEvent/Log.pm: open my $fh, ">>", $path
13lib/AnyEvent/Log.pm: open my $fh, ">>", $path
14lib/AnyEvent/TLS.pm: or croak "$dh_file: failed to open DH parameter file: $!";
15TODO: disable logging to a context while it is being logged to?
16
176.2
18 - child watchers are broken in POE 1.352 (also many earlier
19 versions)and there seems to be no way to work around it, as POE
20 itself is inherently racy. Document this regression and sleep a
21 second in t/68_poe_03_child.t for the time being.
22 - new module AnyEvent::IO, that is a frontend to either a pure-perl
23 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
24 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
25 - load /etc/hosts only when dns fails has no answer.
26 - stat /etc/hosts on every access and reload it if it changed.
27 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
28 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
29 a side effect, they now use true appending as opposed to libc
30 appending, and the file might not have been opened when the function
31 returns.
32 - the default logging level was not properly documented in a variety of
33 places, this has hopefully been rectified.
34 - updated uts46data.pl for unicode 6.1.0.
35 - made log messages genearted by AnyEvent submodules not
36 include the package name anymore, as it will be logged
37 by default already.
38 - upgrade to the trick used by common-sense 3.5 to work
39 around extra warning torture/breakage under perl 5.15.x.
40 - log messages by anyevent are now ucfirst, are usually full
41 sentences and do no longer include the package name.
42 - the storable read type would sometimes throw an exception instead
43 of causing EBADMSG (analyzed by Maxime Soulé).
44 - documented Rocco fabricating statements about AnyEvent and me.
45
466.14 Tue Jan 31 20:00:24 CET 2012
47 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
48 probably an editing mistake (reported by Darin McBride).
49
506.13 Thu Jan 12 07:27:01 CET 2012
51 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
52 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
53 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
54 regex-matching inside the callback would change the parameter, despite
55 $1 being dynamically scoped per-block. this perl bug is now being
56 worked around (testcase by Cindy Wang).
57
586.12 Mon Dec 12 13:21:10 CET 2011
59 - $! was clobbered by subsequent calls in tcp_connect, due to
60 the postpone not saving/restoring it (analyzed by Richard Garnish).
61
626.11 Tue Nov 22 10:36:05 CET 2011
63 - Tk cannot create windows when tainted unless you set a title, so
64 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
65 - escape any nonprintable/nonascii characters when stringifying
66 backtraces.
67 - log the reason loading an interface module fails to load at level debug.
68
696.1 Tue Oct 4 19:44:30 CEST 2011
70 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
71 and above), and some messages inside AnyEvent have been elevated
72 to higher log levels to print by default.
73 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
74 - work around yet another signal race bug, newly introduced by POE. My
75 how that module sucks.
76 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
77 support ipv6-only operation even with defaults (also use ::1 instead
78 of 127.0.0.1).
79 - give better guidelines on how to chose a logging level.
80 - AnyEvent::Log can now cap (limit) log message priority by context.
81 - AE::log will now save and restore $! and $@, so callers don't have to.
82 - AE::Strict now checks that registered fds are still valid at regular
83 intervals.
84 - skip frequently called 'require' calls if the module appears loaded.
85 - increase signal latency test timer considerably for poor little osx,
86 which can't be bothered with basic posix functionality such as signal
87 delivery.
88 - mention some debugging aids in the tutorial.
89 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
90 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
91 of the context as title.
92 - AnyEvent::Strict would not restore read-write access to $_ when a callback
93 died - this is a user bug, but since AE::Strict exists to help dbeugging...
94 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
95 to work with it - the author is well aware of the issues but refuses
96 to fix them. This extreme step has been taken because the module actively
97 attacks the goals of AnyEvent and makes it harder for module authors to use
98 AnyEvent.
99
1006.02 Fri Aug 26 20:08:31 CEST 2011
101 - log_to_syslog now accepts facility strings.
102 - log_to_syslog would not allow facilities - "$facility|$level"
103 *literally* means that string. sheesh.
104 - the CPAN parser got confused and thought we provide Sys::Syslog, try
105 to work around.
106 - renamed AnyEvent::Impl::FLTK2 to FLTK.
107
1086.01 Fri Aug 26 07:04:11 CEST 2011
109 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
110 the TTL as fourth element - this affects ->resolve, ->request
111 and ->dns_unpack indirectly. Convenience functions such as
112 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
113 AnyEvent::Socket.
114 - INCOMPATIBLE CHANGE: backend authors now should not implement
115 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
116 methods.
117 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
118 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
119 - DEPRECIATION: please change
120 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
121 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
122 in your programs, for future compatibility.
123 - *blush* idle watcher emulation was borked.
124 - AnyEvent::Strict now checks the AE::xxx functions as well.
125 - querying too long domain names would barf deep inside AnyEvent::DNS - since
126 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
127 for too long domain names explicitly and passes an undef to the callback
128 instead (reported by James Bromberger).
129 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
130 - new functions: AnyEvent::postpone, AnyEvent::log.
131 - new module: AnyEvent::Log, for simple logging needs.
132 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
133 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
134 a debug shell.
135 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
136 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
137 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
138 /etc/hosts can be found.
139 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
140 configuration.
141 - env variables can now be specified using AE_xyz instead of
142 PERL_ANYEVENT_xyz, subject to some rules.
143 - lots of new AnyEvent::Debug functionality.
144 - AnyEvent::Strict now checks for modifications of $_ by the callback
145 (by setting it to readonly while executing the callback, if possible).
146 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
147 Invest extra overhead to work around this peculiar behaviour.
148 - split the pure perl backend into a loop module and an impl
149 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
150 - some read types (regex, netstring, json, storable) could
151 errornously keep an AE::Handle object alive on their own
152 (reported by Mohammad Toossi).
153 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
154 supports /etc/hosts overrides.
155 - add more workarounds around child watcher bugs in IO::Async
156 and (newly introduced) in POE.
157 - use glib child watchers and try to work around its limitations,
158 if possible (based on a patch by Kevin Ryde).
159 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
160 - many read types in AnyEvent::Handle would malfunction during
161 transitions to TLS because of a "delete $self->{rbuf}".
162 (reported by Gerald Galster).
163 - improve condvar blocking wait performance (probably only noticable
164 with faster backends), also make it easier for event loops that
165 do not support blocking waits.
166 - PERL_ANYEVENT_MODEL now also allows full module names, although
167 the usefulness of this feature is not yet clear.
168 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
169 - add t/80_ssltext.t.
170 - no longer use AUTOLOAD for the initial loading (for no very good reason).
171 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
172 - do not run event loop detection multiple times when programs call their cached
173 copies of AnyEvent::detect.
174 - parse_hostport now accepts absolute paths als unix domain sockets.
175 - suppress the idiotic warning inside POE by patching the POE::Kernel
176 run flags instead of asking the user to jump through contortions.
177 - add Task::Weaken as dependency (as recommends only at this time).
178 - replace PF_UNSPEC by 0, for good old minix.
179
1806.0 (not officially released, but was on CPAN for about 15 minutes :/)
181
1825.34 Thu May 12 09:59:41 CEST 2011
183 - AE::TLS didn't support both cert_file and key at the same time,
184 due to some copy&paste bug (reported and analyzed by vti).
185 - improve AE::Handle timeout documentation slightly.
186
1875.33 Mon Feb 7 21:51:41 CET 2011
188 - never appeared on CPAN, for unknown reasons.
189 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
190 object when it's already gone (analyzed by Ryan Bullock).
191 - update idna algorithm(s) to UTS#46 version 6.0.0.
192 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
193 uts46 says it's an error, and testsuite makes more sense.
194 - remove leading dots in idn_to_ascii - testsuite says to remove,
195 uts46 says it's an error, and testsuite makes more sense.
196 - fixed a bug in idn ignored character processing.
197 - passes the uts46 6.0.0 testsuite except for three cases which
198 apparently are bugs in the testsuite itself.
199 - passes the uts46 6.0.1 testsuite except for eleven cases which
200 apparently are bugs in the testsuite itself.
201
2025.31 Sun Jan 23 11:49:19 CET 2011
203 - work around a POE bug where POE would first poll for new events and
204 invoke handlers BEFORE handling the message queue.
205 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
206 - stop_read now does just that in TLS mode. see its
207 documentation for caveats.
208 - update to common::sense 3.4 warning set (some warnings in perl impose
209 an arbitrary 32kb limit on regex matches, which affects
210 AnyEvent::Handle).
211 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
212 - new experimental AE::Handle::resettls method.
213
2145.3 Fri Dec 31 05:48:13 CET 2010
215 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
216 issues with IO::Async have been solved in that version, including
217 the availabilty of a default loop!
218 - implement new wbuf_max parameter for AnyEvent::Handle.
219 - added Cocoa::EventLoop backend by Daisuke Murase.
220 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
221 - clarify handle on_prepare documentation.
222 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
223 to avoid leaking them to any child processes.
224 - try to force files into the arch-dependend lib directory, for the
225 half-"turly obsessive compulsive person"(s) that depend on it.
226
2275.29 Sun Dec 5 10:49:21 CET 2010
228 - convert EV backend to EV 4.00 API (so better upgrade EV too).
229 - AE::Handle: implement dynamic read_size adjustment, add
230 max_read_size parameter, reduce default read_size to 2048.
231 - add some FAQ entries for servers and/or on_eof vs. on_error.
232 - work around OS bugs (cygwin again) when lingering in AE::Handle.
233 - reduce memory usage slightly when connect was used in AE::Handle.
234 - wrok around more segfaults in Tk.
235 - document tls_ctx => undef as valid.
236 - detect WNOHANG value at compiletime.
237 - include some probably linux-only support for building AnyEvent
238 as part of the perl core.
239 - improve compatibility of t/08_idna.t to perl 5.8.
240 - make AnyEvent::Strict truly optional (in case somebody builds a
241 minimal perl...).
242
2435.28 Wed Oct 13 04:14:23 CEST 2010
244 - due to a glitch, AnyEvent's internal getprotobyname wasn't
245 actually used in tcp_server.
246 - implement AnyEvent::CondVar->new method, which was documented but
247 missing.
248 - support autoloading of read types in unshift_read.
249 - AnyEvent::Strict: do the test for invalid fh arguments differently,
250 resulting in more informative error messages.
251 - new AnyEvent::FAQ manpage.
252 - clarify that you can actually call push_read etc. in on_read.
253 - put keys and certs in /tmp, not into the current directory,
254 and document this in the SECURITY CONSIDERATIONS sections
255 of AnyEvent::TLS.
256 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
257 requested by Ron Isaacson.
258 - document $AnyEvent::DNS::RESOLVER variable.
259 - incorporated major typo patches by Abhijit Menon-Sen.
260 - AnyEvent::Handle now croaks when negative timeouts are passed.
261 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
262 undocumented, but who knows when they might come in handy.
263
2645.271 Tue Jun 8 12:05:46 CEST 2010
265 - backport to perl 5.8.x.
266
2675.27 Sun Jun 6 12:12:05 CEST 2010
268 - postpone differently in AnyEvent::Socket now, as
269 when not, canceling the connection attempt might fail
270 (found by Felix Antonius Wilhelm Ostmann).
271 - explicitly check for non-stream sockets in AE::Handle, too many
272 clueless people fell into the trap of this somehow working.
273 - simplified and reworked the "OTHER MODULES" section.
274 - better/more condvar examples.
275
2765.261 Wed Apr 28 16:13:36 CEST 2010
277 - AF_INET6 was not properly used from Socket6 during configuration
278 time (found by Andy Grundman).
279
2805.26 Mon Apr 12 04:49:35 CEST 2010
281 - don't generate spurious readyness notifications when select
282 returns EINTR in the pure perl backend (analysed and testcase
283 by Bas Denissen).
284 - give same examples in SYNOPSIS sections of AnyEvent and AE.
285 - provider faster implementations for AE::time/now/now_update for
286 EV and Perl backends.
287 - provide faster AE::cv and AE::time implementations for all
288 backends.
289 - no longer support lower/mixed-case signal names (this was
290 never documented nor universally supported).
291 - some more memory µ-optimisations, and somewhat less messy
292 code to cope with both AnyEvent and AE APIs.
293
2945.251 Sat Mar 13 00:58:21 CET 2010
295 - make test failed if it was run before make install on perl <=
296 5.8.8, and older versions of perl make our live very difficult,
297 so write constants.pl during Makefile.PL time.
298
2995.25 Sat Mar 13 00:23:14 CET 2010
300 - fix a race condition in AnyEvent::Handle that would cause
301 a "bio_read: unsupported method" error (in _another_
302 ssl connection) after user code threw an exception.
303 - added AnyEvent::Handle->destroyed method.
3 - speed up AnyEvent::Socket::format_address and ::format_ipv6. 304 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
305 - the AnyEvent::Util::fh_nonblocking function worked only
306 by lucky accident on win32.
307 - smaller and faster AnyEvent::Util::fh_nonblocking.
308 - when the (required!) Time::HiRes module is missing, AnyEvent
309 did not fall back to built-in time properly.
310 - do not load Fcntl at runtime, saving memory and loading time.
311 - precompile a number of constants and use them instead of runtime
312 detection and eval.
313 - free detection code after detection and similar memory optimisations.
314 - Perl backend timer interval best effort drift has been improved
315 (same algorithm as EV).
316 - update unicode idna mapping table.
4 317
55.24 Tue Jan 5 11:39:43 CET 2010 3185.24 Tue Jan 5 11:39:43 CET 2010
6 - cygwin never reports errors from failed connects. 1.5 just gives 319 - cygwin never reports errors from failed connects. 1.5 just gives
7 you continous readyness and EAGAIN, 1.7 is even more broken and 320 you continous readyness and EAGAIN, 1.7 is even more broken and
8 just hangs. work around both issues in a cygwin-specific hack. 321 just hangs. work around both issues in a cygwin-specific hack.
… …
3844.3 Fri Oct 3 09:18:43 CEST 2008 6974.3 Fri Oct 3 09:18:43 CEST 2008
385 - AnyEvent will now install a no-op signal handler for SIGPIPE, 698 - AnyEvent will now install a no-op signal handler for SIGPIPE,
386 unless one has been installed already. 699 unless one has been installed already.
387 - warn about not loaidng AnyEvent::Impl::POE early enough 700 - warn about not loaidng AnyEvent::Impl::POE early enough
388 (patch by Adam Rosenstein). 701 (patch by Adam Rosenstein).
389 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 702 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
390 - fix a bug in where in SSL connect mode, the 703 - fix a bug in where in SSL connect mode, the
391 client would first wait for some data by the server and 704 client would first wait for some data by the server and
392 otherwise hang. (reported and analysed in an absolutely 705 otherwise hang. (reported and analysed in an absolutely
393 exemplary manner by Adam Rosenstein). 706 exemplary manner by Adam Rosenstein).
394 - fix a bug in where SSL EOF would not be treated 707 - fix a bug in where SSL EOF would not be treated

Diff Legend

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