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

Comparing AnyEvent/Changes (file contents):
Revision 1.343 by root, Tue Sep 1 17:36:34 2009 UTC vs.
Revision 1.489 by root, Thu Aug 25 01:41:07 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: pass values to anyevent condvar callback, make this count as recv. 3TODO: document TCP_*** constants
4
5TODO: ae::log examples
6TODO: trace-per-watcher AnyEvent::Debug
76.01
8 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
9 the TTL as fourth element - this affects ->resolve, ->request
10 and ->dns_unpack indirectly. Convenience functions such as
11 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
12 AnyEvent::Socket.
13 - INCOMPATIBLE CHANGE: backend authors now should not implement
14 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
15 methods.
16 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
17 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
18 - DEPRECIATION: please change
19 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
20 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
21 in your programs, for future compatibility.
22 - *blush* idle watcher emulation was borked.
23 - AnyEvent::Strict now checks the AE::xxx functions as well.
24 - querying too long domain names would barf deep inside AnyEvent::DNS - since
25 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
26 for too long domain names explicitly and passes an undef to the callback
27 instead (reported by James Bromberger).
28 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
29 - new functions: AnyEvent::postpone, AnyEvent::log.
30 - new module: AnyEvent::Log, for simple logging needs.
31 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
32 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
33 a debug shell.
34 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
35 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
36 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
37 F</etc/hosts> can be found.
38 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
39 configuration.
40 - env variables can now be specified using AE_xyz instead of
41 PERL_ANYEVENT_xyz, subject to some rules.
42 - lots of new AnyEvent::Debug functionality.
43 - AnyEvent::Strict now checks for modifications $_ by the callback
44 (by setting it to readonly while executing the callback, if possible).
45 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
46 Invest extra overhead to work around this peculiar behaviour.
47 - split the pure perl backend into a loop module and an impl
48 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
49 - some read types (regex, netstring, json, storable) could
50 errornously keep an AE::Handle object alive on their own
51 (reported by Mohammad Toossi).
52 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
53 supports /etc/hosts overrides.
54 - add more workarounds around child watcher bugs in IO::Async
55 and (newly introduced) in POE.
56 - use glib child watchers and try to work around its limitations,
57 if possible (based on a patch by Kevin Ryde).
58 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
59 - many read types in AnyEvent::Handle would malfunction during
60 transitions to TLS because of a "delete $self->{rbuf}".
61 (reported by Gerald Galster).
62 - improve condvar blocking wait performance (probably only noticable
63 with faster backends), also make it easier for event loops that
64 do not support blocking waits.
65 - PERL_ANYEVENT_MODEL now also allows full module names, although
66 the usefulness of this feature is not yet clear.
67 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
68 - add t/80_ssltext.t.
69 - no longer use AUTOLOAD for the initial loading (for no very good reason).
70 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
71 - do not run event loop detection multiple times when programs call their cached
72 copies of AnyEvent::detect.
73 - parse_hostport now accepts absolute paths als unix domain sockets.
74 - add Task::Weaken as dependency (as recommends only at this time).
75
766.0 (not officially released, but was on CPAN for about 15 minutes :/)
77
785.34 Thu May 12 09:59:41 CEST 2011
79 - AE::TLS didn't support both cert_file and key at the same time,
80 due to some copy&paste bug (reported and analyzed by vti).
81 - improve AE::Handle timeout documentation slightly.
82
835.33 Mon Feb 7 21:51:41 CET 2011
84 - never appeared on CPAN, for unknown reasons.
85 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
86 object when it's already gone (analyzed by Ryan Bullock).
87 - update idna algorithm(s) to UTS#46 version 6.0.0.
88 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
89 uts46 says it's an error, and testsuite makes more sense.
90 - remove leading dots in idn_to_ascii - testsuite says to remove,
91 uts46 says it's an error, and testsuite makes more sense.
92 - fixed a bug in idn ignored character processing.
93 - passes the uts46 6.0.0 testsuite except for three cases which
94 apparently are bugs in the testsuite itself.
95 - passes the uts46 6.0.1 testsuite except for eleven cases which
96 apparently are bugs in the testsuite itself.
97
985.31 Sun Jan 23 11:49:19 CET 2011
99 - work around a POE bug where POE would first poll for new events and
100 invoke handlers BEFORE handling the message queue.
101 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
102 - stop_read now does just that in TLS mode. see its
103 documentation for caveats.
104 - update to common::sense 3.4 warning set (some warnings in perl impose
105 an arbitrary 32kb limit on regex matches, which affects
106 AnyEvent::Handle).
107 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
108 - new experimental AE::Handle::resettls method.
109
1105.3 Fri Dec 31 05:48:13 CET 2010
111 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
112 issues with IO::Async have been solved in that version, including
113 the availabilty of a default loop!
114 - implement new wbuf_max parameter for AnyEvent::Handle.
115 - added Cocoa::EventLoop backend by Daisuke Murase.
116 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
117 - clarify handle on_prepare documentation.
118 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
119 to avoid leaking them to any child processes.
120 - try to force files into the arch-dependend lib directory, for the
121 half-"turly obsessive compulsive person"(s) that depend on it.
122
1235.29 Sun Dec 5 10:49:21 CET 2010
124 - convert EV backend to EV 4.00 API (so better upgrade EV too).
125 - AE::Handle: implement dynamic read_size adjustment, add
126 max_read_size parameter, reduce default read_size to 2048.
127 - add some FAQ entries for servers and/or on_eof vs. on_error.
128 - work around OS bugs (cygwin again) when lingering in AE::Handle.
129 - reduce memory usage slightly when connect was used in AE::Handle.
130 - wrok around more segfaults in Tk.
131 - document tls_ctx => undef as valid.
132 - detect WNOHANG value at compiletime.
133 - include some probably linux-only support for building AnyEvent
134 as part of the perl core.
135 - improve compatibility of t/08_idna.t to perl 5.8.
136 - make AnyEvent::Strict truly optional (in case somebody builds a
137 minimal perl...).
138
1395.28 Wed Oct 13 04:14:23 CEST 2010
140 - due to a glitch, AnyEvent's internal getprotobyname wasn't
141 actually used in tcp_server.
142 - implement AnyEvent::CondVar->new method, which was documented but
143 missing.
144 - support autoloading of read types in unshift_read.
145 - AnyEvent::Strict: do the test for invalid fh arguments differently,
146 resulting in more informative error messages.
147 - new AnyEvent::FAQ manpage.
148 - clarify that you can actually call push_read etc. in on_read.
149 - put keys and certs in /tmp, not into the current directory,
150 and document this in the SECURITY CONSIDERATIONS sections
151 of AnyEvent::TLS.
152 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
153 requested by Ron Isaacson.
154 - document $AnyEvent::DNS::RESOLVER variable.
155 - incorporated major typo patches by Abhijit Menon-Sen.
156 - AnyEvent::Handle now croaks when negative timeouts are passed.
157 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
158 undocumented, but who knows when they might come in handy.
159
1605.271 Tue Jun 8 12:05:46 CEST 2010
161 - backport to perl 5.8.x.
162
1635.27 Sun Jun 6 12:12:05 CEST 2010
164 - postpone differently in AnyEvent::Socket now, as
165 when not, canceling the connection attempt might fail
166 (found by Felix Antonius Wilhelm Ostmann).
167 - explicitly check for non-stream sockets in AE::Handle, too many
168 clueless people fell into the trap of this somehow working.
169 - simplified and reworked the "OTHER MODULES" section.
170 - better/more condvar examples.
171
1725.261 Wed Apr 28 16:13:36 CEST 2010
173 - AF_INET6 was not properly used from Socket6 during configuration
174 time (found by Andy Grundman).
175
1765.26 Mon Apr 12 04:49:35 CEST 2010
177 - don't generate spurious readyness notifications when select
178 returns EINTR in the pure perl backend (analysed and testcase
179 by Bas Denissen).
180 - give same examples in SYNOPSIS sections of AnyEvent and AE.
181 - provider faster implementations for AE::time/now/now_update for
182 EV and Perl backends.
183 - provide faster AE::cv and AE::time implementations for all
184 backends.
185 - no longer support lower/mixed-case signal names (this was
186 never documented nor universally supported).
187 - some more memory ยต-optimisations, and somewhat less messy
188 code to cope with both AnyEvent and AE APIs.
189
1905.251 Sat Mar 13 00:58:21 CET 2010
191 - make test failed if it was run before make install on perl <=
192 5.8.8, and older versions of perl make our live very difficult,
193 so write constants.pl during Makefile.PL time.
194
1955.25 Sat Mar 13 00:23:14 CET 2010
196 - fix a race condition in AnyEvent::Handle that would cause
197 a "bio_read: unsupported method" error (in _another_
198 ssl connection) after user code threw an exception.
199 - added AnyEvent::Handle->destroyed method.
200 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
201 - the AnyEvent::Util::fh_nonblocking function worked only
202 by lucky accident on win32.
203 - smaller and faster AnyEvent::Util::fh_nonblocking.
204 - when the (required!) Time::HiRes module is missing, AnyEvent
205 did not fall back to built-in time properly.
206 - do not load Fcntl at runtime, saving memory and loading time.
207 - precompile a number of constants and use them instead of runtime
208 detection and eval.
209 - free detection code after detection and similar memory optimisations.
210 - Perl backend timer interval best effort drift has been improved
211 (same algorithm as EV).
212 - update unicode idna mapping table.
213
2145.24 Tue Jan 5 11:39:43 CET 2010
215 - cygwin never reports errors from failed connects. 1.5 just gives
216 you continous readyness and EAGAIN, 1.7 is even more broken and
217 just hangs. work around both issues in a cygwin-specific hack.
218 - improve idle watcher documentation slightly.
219
2205.23 Sun Dec 20 23:48:00 CET 2009
221 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
222 - implement punycode_encode/decode, idn_nameprep,
223 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
224 - provide $AE::VERSION.
225 - removed traces of "no strict 'refs'".
226
2275.22 Sat Dec 5 03:51:13 CET 2009
228 - downgrade-or-fail in AnyEvent::Handle::push_write, to
229 diagnose encoding failures earlier and more succinctly.
230 (this works around bugs in perl, throwing away encoding info
231 when passing scalar data to extensions).
232 - add more examples to AnyEvent::Socket manpage.
233 - upgrade internal warning set to the same as common::sense 2.03.
234 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
235 (John Beppu).
236 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
237 - add google public dns servers to fallback server set and make sure
238 we load-balance properly between the three sets. also add all
239 fallback dns servers, not just a random one, to each dns config.
240
2415.21 Thu Nov 19 02:48:47 CET 2009
242 - fix a problem where socket constants were called with parameters
243 (spotted by David Friedland).
244 - fork_call never use'd POSIX (reported by Daisuke Maki).
245 - improve perl 5.6 compatibility further (but it still won't work
246 unless you rip out everything but the core).
247 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
248 uses 10MB of RAM, but doesn't open a console window. *sigh*.
249
2505.202 Wed Oct 14 22:35:44 CEST 2009
251 - AnyEvent::DNS would unexpectedly clobber $_ under windows
252 (analysed by Matthias Waldorf).
253 - AnyEvent::Handle::run_cmd can now pass the PID of the
254 newly-created process, which is much less useful than it might
255 sound (based on patch by Yann Kerherve).
256
2575.201 Tue Sep 29 12:09:25 CEST 2009
258 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
259 (reported by Torsten Foertsch).
260 - common::sense 2.0 could cause tcp_server to throw an exception
261 (analysed by elmex).
262
2635.2 Mon Sep 14 07:04:49 CEST 2009
264 - INCOMPATIBLE CHANGE: do no longer support register_read_type
265 and register_write_type in AnyEvent::Handle, instead support
266 package names (the facility was mostly abused).
267 - implement "packagename-as-read/write type" support in
268 AnyEvent::Handle.
269 - AnyEvent::Handle: new options "keepalive" and "oobinline".
270 - oobinline set by default to avoid security issues.
271 - the pure-perl event loop backend wrongly detected times() fallback
272 support (spotted by Pavel Boldin).
273
2745.12 Tue Sep 1 20:26:50 CEST 2009
4 - be more lenient when parsing resolv.conf files, as some people 275 - be more lenient when parsing resolv.conf files, as some people
5 use hashmarks as comment indicator inside directives (reported 276 use hashmarks as comment indicator inside directives (reported
6 by Michael S. Fischer). 277 by Michael S. Fischer).
7 - (optionally) require Async::Interrupt 1.02.
8 - use same set of warnings as common::sense 2.0. 278 - use same set of warnings as common::sense 2.0.
9 - fix a potential 32 bit overflow issue due to perl having problems 279 - fix a potential 32 bit overflow issue due to perl having problems
10 with large hex constants in the Perl backend. 280 with large hex constants in the Perl backend.
11 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS 281 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
12 by default. 282 by default.
13 - fix some prototypes. 283 - fix some prototypes.
284 - (optionally) require Async::Interrupt 1.02.
14 - replace opendns fallbacks by verizon/level3 public dns servers. 285 - replace opendns fallbacks by verizon/level3 public dns servers.
15 286
165.112 Fri Aug 21 13:59:04 CEST 2009 2875.112 Fri Aug 21 13:59:04 CEST 2009
17 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when 288 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
18 using starttls with an illegal/uncreatable tls context. 289 using starttls with an illegal/uncreatable tls context.
… …
90 - new function: AnyEvent::Socket::format_hostport. 361 - new function: AnyEvent::Socket::format_hostport.
91 362
924.881 Tue Jul 28 12:51:53 CEST 2009 3634.881 Tue Jul 28 12:51:53 CEST 2009
93 - work around a bug in local in pre-5.10 perls, causing 364 - work around a bug in local in pre-5.10 perls, causing
94 AnyEvent::Handle to recurse when it should not 365 AnyEvent::Handle to recurse when it should not
95 (analyzed by elmex). 366 (analysed by elmex).
96 367
974.88 Tue Jul 28 04:04:37 CEST 2009 3684.88 Tue Jul 28 04:04:37 CEST 2009
98 - re-bless the handle into a dummy package after calling 369 - re-bless the handle into a dummy package after calling
99 AnyEvent::Handle::destroy, so the user does not need to check 370 AnyEvent::Handle::destroy, so the user does not need to check
100 for errors after every push_write etc. 371 for errors after every push_write etc.
… …
3224.3 Fri Oct 3 09:18:43 CEST 2008 5934.3 Fri Oct 3 09:18:43 CEST 2008
323 - AnyEvent will now install a no-op signal handler for SIGPIPE, 594 - AnyEvent will now install a no-op signal handler for SIGPIPE,
324 unless one has been installed already. 595 unless one has been installed already.
325 - warn about not loaidng AnyEvent::Impl::POE early enough 596 - warn about not loaidng AnyEvent::Impl::POE early enough
326 (patch by Adam Rosenstein). 597 (patch by Adam Rosenstein).
327 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 598 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
328 - fix a bug in where in SSL connect mode, the 599 - fix a bug in where in SSL connect mode, the
329 client would first wait for some data by the server and 600 client would first wait for some data by the server and
330 otherwise hang. (reported and analyzed in an absolutely 601 otherwise hang. (reported and analysed in an absolutely
331 exemplary manner by Adam Rosenstein). 602 exemplary manner by Adam Rosenstein).
332 - fix a bug in where SSL EOF would not be treated 603 - fix a bug in where SSL EOF would not be treated
333 as stream EOF, putting the connection into a hung state. 604 as stream EOF, putting the connection into a hung state.
334 (reported and analyzed in an absolutely exemplary manner by Adam 605 (reported and analysed in an absolutely exemplary manner by Adam
335 Rosenstein). 606 Rosenstein).
336 - fix a potential segfault when the TLS context 607 - fix a potential segfault when the TLS context
337 would go missing in a read callback (e.g. due to stoptls) - 608 would go missing in a read callback (e.g. due to stoptls) -
338 Net::SSLeay of course makes no type checking whatsoever. 609 Net::SSLeay of course makes no type checking whatsoever.
339 - AnyEvent::Handle will not stop the read watcher in TLS mode. 610 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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