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

Comparing AnyEvent/Changes (file contents):
Revision 1.380 by root, Mon Feb 15 18:02:35 2010 UTC vs.
Revision 1.531 by root, Tue Mar 27 23:47:56 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: PERL_ANYEVENT_IO_MODEL docs
7tODO: maxim souffle handle storable
8TODO: t/io
9lib/AnyEvent/DNS.pm: open my $fh, "<", $resolv_conf
10lib/AnyEvent/DNS.pm: if (open my $fh, "ipconfig /all |") {
11lib/AnyEvent/Log.pm: open my $fh, ">>", $path
12lib/AnyEvent/Log.pm: open my $fh, ">>", $path
13lib/AnyEvent/Socket.pm: open my $fh, "<",
14lib/AnyEvent/TLS.pm: or croak "$dh_file: failed to open DH parameter file: $!";
15
16 - documented Rocco fabricating statements about AnyEvent and me.
17 - the default logging level was not properly documented in a variety of
18 places, this has hopefully been rectified.
19 - updated uts46data.pl for unicode 6.1.0.
20 - made log messages genearted by AnyEvent submodules not
21 include the package name anymore, as it will be logged
22 by default already.
23 - upgrade to the trick used by common-sense 3.5 to work
24 around extra warning torture/breakage under perl 5.15.x.
25 - log messages by anyevent are now ucfirst, are usually full
26 sentences and do no longer include the package name.
27
286.14 Tue Jan 31 20:00:24 CET 2012
29 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
30 probably an editing mistake (reported by Darin McBride).
31
326.13 Thu Jan 12 07:27:01 CET 2012
33 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
34 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
35 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
36 regex-matching inside the callback would change the parameter, despite
37 $1 being dynamically scoped per-block. this perl bug is now being
38 worked around (testcase by Cindy Wang).
39
406.12 Mon Dec 12 13:21:10 CET 2011
41 - $! was clobbered by subsequent calls in tcp_connect, due to
42 the postpone not saving/restoring it (analyzed by Richard Garnish).
43
446.11 Tue Nov 22 10:36:05 CET 2011
45 - Tk cannot create windows when tainted unless you set a title, so
46 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
47 - escape any nonprintable/nonascii characters when stringifying
48 backtraces.
49 - log the reason loading an interface module fails to load at level debug.
50
516.1 Tue Oct 4 19:44:30 CEST 2011
52 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
53 and above), and some messages inside AnyEvent have been elevated
54 to higher log levels to print by default.
55 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
56 - work around yet another signal race bug, newly introduced by POE. My
57 how that module sucks.
58 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
59 support ipv6-only operation even with defaults (also use ::1 instead
60 of 127.0.0.1).
61 - give better guidelines on how to chose a logging level.
62 - AnyEvent::Log can now cap (limit) log message priority by context.
63 - AE::log will now save and restore $! and $@, so callers don't have to.
64 - AE::Strict now checks that registered fds are still valid at regular
65 intervals.
66 - skip frequently called 'require' calls if the module appears loaded.
67 - increase signal latency test timer considerably for poor little osx,
68 which can't be bothered with basic posix functionality such as signal
69 delivery.
70 - mention some debugging aids in the tutorial.
71 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
72 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
73 of the context as title.
74 - AnyEvent::Strict would not restore read-write access to $_ when a callback
75 died - this is a user bug, but since AE::Strict exists to help dbeugging...
76 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
77 to work with it - the author is well aware of the issues but refuses
78 to fix them. This extreme step has been taken because the module actively
79 attacks the goals of AnyEvent and makes it harder for module authors to use
80 AnyEvent.
81
826.02 Fri Aug 26 20:08:31 CEST 2011
83 - log_to_syslog now accepts facility strings.
84 - log_to_syslog would not allow facilities - "$facility|$level"
85 *literally* means that string. sheesh.
86 - the CPAN parser got confused and thought we provide Sys::Syslog, try
87 to work around.
88 - renamed AnyEvent::Impl::FLTK2 to FLTK.
89
906.01 Fri Aug 26 07:04:11 CEST 2011
91 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
92 the TTL as fourth element - this affects ->resolve, ->request
93 and ->dns_unpack indirectly. Convenience functions such as
94 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
95 AnyEvent::Socket.
96 - INCOMPATIBLE CHANGE: backend authors now should not implement
97 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
98 methods.
99 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
100 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
101 - DEPRECIATION: please change
102 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
103 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
104 in your programs, for future compatibility.
105 - *blush* idle watcher emulation was borked.
106 - AnyEvent::Strict now checks the AE::xxx functions as well.
107 - querying too long domain names would barf deep inside AnyEvent::DNS - since
108 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
109 for too long domain names explicitly and passes an undef to the callback
110 instead (reported by James Bromberger).
111 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
112 - new functions: AnyEvent::postpone, AnyEvent::log.
113 - new module: AnyEvent::Log, for simple logging needs.
114 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
115 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
116 a debug shell.
117 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
118 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
119 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
120 /etc/hosts can be found.
121 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
122 configuration.
123 - env variables can now be specified using AE_xyz instead of
124 PERL_ANYEVENT_xyz, subject to some rules.
125 - lots of new AnyEvent::Debug functionality.
126 - AnyEvent::Strict now checks for modifications $_ by the callback
127 (by setting it to readonly while executing the callback, if possible).
128 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
129 Invest extra overhead to work around this peculiar behaviour.
130 - split the pure perl backend into a loop module and an impl
131 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
132 - some read types (regex, netstring, json, storable) could
133 errornously keep an AE::Handle object alive on their own
134 (reported by Mohammad Toossi).
135 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
136 supports /etc/hosts overrides.
137 - add more workarounds around child watcher bugs in IO::Async
138 and (newly introduced) in POE.
139 - use glib child watchers and try to work around its limitations,
140 if possible (based on a patch by Kevin Ryde).
141 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
142 - many read types in AnyEvent::Handle would malfunction during
143 transitions to TLS because of a "delete $self->{rbuf}".
144 (reported by Gerald Galster).
145 - improve condvar blocking wait performance (probably only noticable
146 with faster backends), also make it easier for event loops that
147 do not support blocking waits.
148 - PERL_ANYEVENT_MODEL now also allows full module names, although
149 the usefulness of this feature is not yet clear.
150 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
151 - add t/80_ssltext.t.
152 - no longer use AUTOLOAD for the initial loading (for no very good reason).
153 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
154 - do not run event loop detection multiple times when programs call their cached
155 copies of AnyEvent::detect.
156 - parse_hostport now accepts absolute paths als unix domain sockets.
157 - suppress the idiotic warning inside POE by patching the POE::Kernel
158 run flags instead of asking the user to jump through contortions.
159 - add Task::Weaken as dependency (as recommends only at this time).
160 - replace PF_UNSPEC by 0, for good old minix.
161
1626.0 (not officially released, but was on CPAN for about 15 minutes :/)
163
1645.34 Thu May 12 09:59:41 CEST 2011
165 - AE::TLS didn't support both cert_file and key at the same time,
166 due to some copy&paste bug (reported and analyzed by vti).
167 - improve AE::Handle timeout documentation slightly.
168
1695.33 Mon Feb 7 21:51:41 CET 2011
170 - never appeared on CPAN, for unknown reasons.
171 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
172 object when it's already gone (analyzed by Ryan Bullock).
173 - update idna algorithm(s) to UTS#46 version 6.0.0.
174 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
175 uts46 says it's an error, and testsuite makes more sense.
176 - remove leading dots in idn_to_ascii - testsuite says to remove,
177 uts46 says it's an error, and testsuite makes more sense.
178 - fixed a bug in idn ignored character processing.
179 - passes the uts46 6.0.0 testsuite except for three cases which
180 apparently are bugs in the testsuite itself.
181 - passes the uts46 6.0.1 testsuite except for eleven cases which
182 apparently are bugs in the testsuite itself.
183
1845.31 Sun Jan 23 11:49:19 CET 2011
185 - work around a POE bug where POE would first poll for new events and
186 invoke handlers BEFORE handling the message queue.
187 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
188 - stop_read now does just that in TLS mode. see its
189 documentation for caveats.
190 - update to common::sense 3.4 warning set (some warnings in perl impose
191 an arbitrary 32kb limit on regex matches, which affects
192 AnyEvent::Handle).
193 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
194 - new experimental AE::Handle::resettls method.
195
1965.3 Fri Dec 31 05:48:13 CET 2010
197 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
198 issues with IO::Async have been solved in that version, including
199 the availabilty of a default loop!
200 - implement new wbuf_max parameter for AnyEvent::Handle.
201 - added Cocoa::EventLoop backend by Daisuke Murase.
202 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
203 - clarify handle on_prepare documentation.
204 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
205 to avoid leaking them to any child processes.
206 - try to force files into the arch-dependend lib directory, for the
207 half-"turly obsessive compulsive person"(s) that depend on it.
208
2095.29 Sun Dec 5 10:49:21 CET 2010
210 - convert EV backend to EV 4.00 API (so better upgrade EV too).
211 - AE::Handle: implement dynamic read_size adjustment, add
212 max_read_size parameter, reduce default read_size to 2048.
213 - add some FAQ entries for servers and/or on_eof vs. on_error.
214 - work around OS bugs (cygwin again) when lingering in AE::Handle.
215 - reduce memory usage slightly when connect was used in AE::Handle.
216 - wrok around more segfaults in Tk.
217 - document tls_ctx => undef as valid.
218 - detect WNOHANG value at compiletime.
219 - include some probably linux-only support for building AnyEvent
220 as part of the perl core.
221 - improve compatibility of t/08_idna.t to perl 5.8.
222 - make AnyEvent::Strict truly optional (in case somebody builds a
223 minimal perl...).
224
2255.28 Wed Oct 13 04:14:23 CEST 2010
226 - due to a glitch, AnyEvent's internal getprotobyname wasn't
227 actually used in tcp_server.
228 - implement AnyEvent::CondVar->new method, which was documented but
229 missing.
230 - support autoloading of read types in unshift_read.
231 - AnyEvent::Strict: do the test for invalid fh arguments differently,
232 resulting in more informative error messages.
233 - new AnyEvent::FAQ manpage.
234 - clarify that you can actually call push_read etc. in on_read.
235 - put keys and certs in /tmp, not into the current directory,
236 and document this in the SECURITY CONSIDERATIONS sections
237 of AnyEvent::TLS.
238 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
239 requested by Ron Isaacson.
240 - document $AnyEvent::DNS::RESOLVER variable.
241 - incorporated major typo patches by Abhijit Menon-Sen.
242 - AnyEvent::Handle now croaks when negative timeouts are passed.
243 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
244 undocumented, but who knows when they might come in handy.
245
2465.271 Tue Jun 8 12:05:46 CEST 2010
247 - backport to perl 5.8.x.
248
2495.27 Sun Jun 6 12:12:05 CEST 2010
250 - postpone differently in AnyEvent::Socket now, as
251 when not, canceling the connection attempt might fail
252 (found by Felix Antonius Wilhelm Ostmann).
253 - explicitly check for non-stream sockets in AE::Handle, too many
254 clueless people fell into the trap of this somehow working.
255 - simplified and reworked the "OTHER MODULES" section.
256 - better/more condvar examples.
257
2585.261 Wed Apr 28 16:13:36 CEST 2010
259 - AF_INET6 was not properly used from Socket6 during configuration
260 time (found by Andy Grundman).
261
2625.26 Mon Apr 12 04:49:35 CEST 2010
263 - don't generate spurious readyness notifications when select
264 returns EINTR in the pure perl backend (analysed and testcase
265 by Bas Denissen).
266 - give same examples in SYNOPSIS sections of AnyEvent and AE.
267 - provider faster implementations for AE::time/now/now_update for
268 EV and Perl backends.
269 - provide faster AE::cv and AE::time implementations for all
270 backends.
271 - no longer support lower/mixed-case signal names (this was
272 never documented nor universally supported).
273 - some more memory ยต-optimisations, and somewhat less messy
274 code to cope with both AnyEvent and AE APIs.
275
2765.251 Sat Mar 13 00:58:21 CET 2010
277 - make test failed if it was run before make install on perl <=
278 5.8.8, and older versions of perl make our live very difficult,
279 so write constants.pl during Makefile.PL time.
280
2815.25 Sat Mar 13 00:23:14 CET 2010
282 - fix a race condition in AnyEvent::Handle that would cause
283 a "bio_read: unsupported method" error (in _another_
284 ssl connection) after user code threw an exception.
285 - added AnyEvent::Handle->destroyed method.
3 - speed up AnyEvent::Socket::format_address and ::format_ipv6. 286 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
4 - the AnyEvent::Util::fh_nonblocking function worked only 287 - the AnyEvent::Util::fh_nonblocking function worked only
5 by lucky accident on win32. 288 by lucky accident on win32.
6 - smaller and faster AnyEvent::Util::fh_nonblocking. 289 - smaller and faster AnyEvent::Util::fh_nonblocking.
7 - when the (required!) Time::HiRes module is missing, AnyEvent 290 - when the (required!) Time::HiRes module is missing, AnyEvent
8 did not fall back to built-in time properly. 291 did not fall back to built-in time properly.
292 - do not load Fcntl at runtime, saving memory and loading time.
293 - precompile a number of constants and use them instead of runtime
294 detection and eval.
9 - free detection code after detection and other memory optimisations. 295 - free detection code after detection and similar memory optimisations.
296 - Perl backend timer interval best effort drift has been improved
297 (same algorithm as EV).
298 - update unicode idna mapping table.
10 299
115.24 Tue Jan 5 11:39:43 CET 2010 3005.24 Tue Jan 5 11:39:43 CET 2010
12 - cygwin never reports errors from failed connects. 1.5 just gives 301 - cygwin never reports errors from failed connects. 1.5 just gives
13 you continous readyness and EAGAIN, 1.7 is even more broken and 302 you continous readyness and EAGAIN, 1.7 is even more broken and
14 just hangs. work around both issues in a cygwin-specific hack. 303 just hangs. work around both issues in a cygwin-specific hack.
… …
3904.3 Fri Oct 3 09:18:43 CEST 2008 6794.3 Fri Oct 3 09:18:43 CEST 2008
391 - AnyEvent will now install a no-op signal handler for SIGPIPE, 680 - AnyEvent will now install a no-op signal handler for SIGPIPE,
392 unless one has been installed already. 681 unless one has been installed already.
393 - warn about not loaidng AnyEvent::Impl::POE early enough 682 - warn about not loaidng AnyEvent::Impl::POE early enough
394 (patch by Adam Rosenstein). 683 (patch by Adam Rosenstein).
395 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 684 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
396 - fix a bug in where in SSL connect mode, the 685 - fix a bug in where in SSL connect mode, the
397 client would first wait for some data by the server and 686 client would first wait for some data by the server and
398 otherwise hang. (reported and analysed in an absolutely 687 otherwise hang. (reported and analysed in an absolutely
399 exemplary manner by Adam Rosenstein). 688 exemplary manner by Adam Rosenstein).
400 - fix a bug in where SSL EOF would not be treated 689 - fix a bug in where SSL EOF would not be treated

Diff Legend

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