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

Comparing AnyEvent/Changes (file contents):
Revision 1.260 by root, Fri Jul 3 21:44:14 2009 UTC vs.
Revision 1.417 by root, Sun Nov 14 02:24:51 2010 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3ZERO_RETURN - stoptls? 3TODO: document TCP_*** constants
4TODO: catch AE::xxx for anyevent struct and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict etc. for AE::
7TODO: anyevent strict checking for $_ changes?
8
9 - convert EV backend to EV 4.00 API.
10 - AE::Handle: implement dynamic read_size adjustment, add
11 max_read_size parameter, reduce default read_size to 2048.
12 - work around OS bugs (cygwin again) when lingering in AE::Handle.
13
145.28 Wed Oct 13 04:14:23 CEST 2010
15 - due to a glitch, AnyEvent's internal getprotobyname wasn't
16 actually used in tcp_server.
17 - implement AnyEvent::CondVar->new method, which was documented but
18 missing.
19 - support autoloading of read types in unshift_read.
20 - AnyEvent::Strict: do the test for invalid fh arguments differently,
21 resulting in more informative error messages.
22 - new AnyEvent::FAQ manpage.
23 - clarify that you can actually call push_read etc. in on_read.
24 - put keys and certs in /tmp, not into the current directory,
25 and document this in the SECURITY CONSIDERATIONS sections
26 of AnyEvent::TLS.
27 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
28 requested by Ron Isaacson.
29 - document $AnyEvent::DNS::RESOLVER variable.
30 - incorporated major typo patches by Abhijit Menon-Sen.
31 - AnyEvent::Handle now croaks when negative timeouts are passed.
32 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
33 undocumented, but who knows when they might come in handy.
34
355.271 Tue Jun 8 12:05:46 CEST 2010
36 - backport to perl 5.8.x.
37
385.27 Sun Jun 6 12:12:05 CEST 2010
39 - postpone differently in AnyEvent::Socket now, as
40 when not, canceling the connection attempt might fail
41 (found by Felix Antonius Wilhelm Ostmann).
42 - explicitly check for non-stream sockets in AE::Handle, too many
43 clueless people fell into the trap of this somehow working.
44 - simplified and reworked the "OTHER MODULES" section.
45 - better/more condvar examples.
46
475.261 Wed Apr 28 16:13:36 CEST 2010
48 - AF_INET6 was not properly used from Socket6 during configuration
49 time (found by Andy Grundman).
50
515.26 Mon Apr 12 04:49:35 CEST 2010
52 - don't generate spurious readyness notifications when select
53 returns EINTR in the pure perl backend (analysed and testcase
54 by Bas Denissen).
55 - give same examples in SYNOPSIS sections of AnyEvent and AE.
56 - provider faster implementations for AE::time/now/now_update for
57 EV and Perl backends.
58 - provide faster AE::cv and AE::time implementations for all
59 backends.
60 - no longer support lower/mixed-case signal names (this was
61 never documented nor universally supported).
62 - some more memory ยต-optimisations, and somewhat less messy
63 code to cope with both AnyEvent and AE APIs.
64
655.251 Sat Mar 13 00:58:21 CET 2010
66 - make test failed if it was run before make install on perl <=
67 5.8.8, and older versions of perl make our live very difficult,
68 so write constants.pl during Makefile.PL time.
69
705.25 Sat Mar 13 00:23:14 CET 2010
71 - fix a race condition in AnyEvent::Handle that would cause
72 a "bio_read: unsupported method" error (in _another_
73 ssl connection) after user code threw an exception.
74 - added AnyEvent::Handle->destroyed method.
75 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
76 - the AnyEvent::Util::fh_nonblocking function worked only
77 by lucky accident on win32.
78 - smaller and faster AnyEvent::Util::fh_nonblocking.
79 - when the (required!) Time::HiRes module is missing, AnyEvent
80 did not fall back to built-in time properly.
81 - do not load Fcntl at runtime, saving memory and loading time.
82 - precompile a number of constants and use them instead of runtime
83 detection and eval.
84 - free detection code after detection and similar memory optimisations.
85 - Perl backend timer interval best effort drift has been improved
86 (same algorithm as EV).
87 - update unicode idna mapping table.
88
895.24 Tue Jan 5 11:39:43 CET 2010
90 - cygwin never reports errors from failed connects. 1.5 just gives
91 you continous readyness and EAGAIN, 1.7 is even more broken and
92 just hangs. work around both issues in a cygwin-specific hack.
93 - improve idle watcher documentation slightly.
94
955.23 Sun Dec 20 23:48:00 CET 2009
96 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
97 - implement punycode_encode/decode, idn_nameprep,
98 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
99 - provide $AE::VERSION.
100 - removed traces of "no strict 'refs'".
101
1025.22 Sat Dec 5 03:51:13 CET 2009
103 - downgrade-or-fail in AnyEvent::Handle::push_write, to
104 diagnose encoding failures earlier and more succinctly.
105 (this works around bugs in perl, throwing away encoding info
106 when passing scalar data to extensions).
107 - add more examples to AnyEvent::Socket manpage.
108 - upgrade internal warning set to the same as common::sense 2.03.
109 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
110 (John Beppu).
111 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
112 - add google public dns servers to fallback server set and make sure
113 we load-balance properly between the three sets. also add all
114 fallback dns servers, not just a random one, to each dns config.
115
1165.21 Thu Nov 19 02:48:47 CET 2009
117 - fix a problem where socket constants were called with parameters
118 (spotted by David Friedland).
119 - fork_call never use'd POSIX (reported by Daisuke Maki).
120 - improve perl 5.6 compatibility further (but it still won't work
121 unless you rip out everything but the core).
122 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
123 uses 10MB of RAM, but doesn't open a console window. *sigh*.
124
1255.202 Wed Oct 14 22:35:44 CEST 2009
126 - AnyEvent::DNS would unexpectedly clobber $_ under windows
127 (analysed by Matthias Waldorf).
128 - AnyEvent::Handle::run_cmd can now pass the PID of the
129 newly-created process, which is much less useful than it might
130 sound (based on patch by Yann Kerherve).
131
1325.201 Tue Sep 29 12:09:25 CEST 2009
133 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
134 (reported by Torsten Foertsch).
135 - common::sense 2.0 could cause tcp_server to throw an exception
136 (analysed by elmex).
137
1385.2 Mon Sep 14 07:04:49 CEST 2009
139 - INCOMPATIBLE CHANGE: do no longer support register_read_type
140 and register_write_type in AnyEvent::Handle, instead support
141 package names (the facility was mostly abused).
142 - implement "packagename-as-read/write type" support in
143 AnyEvent::Handle.
144 - AnyEvent::Handle: new options "keepalive" and "oobinline".
145 - oobinline set by default to avoid security issues.
146 - the pure-perl event loop backend wrongly detected times() fallback
147 support (spotted by Pavel Boldin).
148
1495.12 Tue Sep 1 20:26:50 CEST 2009
150 - be more lenient when parsing resolv.conf files, as some people
151 use hashmarks as comment indicator inside directives (reported
152 by Michael S. Fischer).
153 - use same set of warnings as common::sense 2.0.
154 - fix a potential 32 bit overflow issue due to perl having problems
155 with large hex constants in the Perl backend.
156 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
157 by default.
158 - fix some prototypes.
159 - (optionally) require Async::Interrupt 1.02.
160 - replace opendns fallbacks by verizon/level3 public dns servers.
161
1625.112 Fri Aug 21 13:59:04 CEST 2009
163 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
164 using starttls with an illegal/uncreatable tls context.
165 - port to broken win32 perls.
166
1675.111 Sun Aug 16 18:44:36 CEST 2009
168 - perl errornously requires sockaddr_un structures returned by the
169 kernel to have a specific length. work around this bug by padding
170 sockaddr_un structures (found to fail on OS X, as perl only uses
171 the correct code for GNU/Linux, but likely to fail on other
172 OSes as well).
173
1745.11 Wed Aug 12 17:49:37 CEST 2009
175 - fixed a bug in run_cmd with input from filename.
176 - create a json coder object when none is given in write direction
177 to reduce dependency on JSON.
178
1795.1 Tue Aug 11 03:17:32 CEST 2009
180 - tcp_connect now reports the correct error in $!, not an unrelated
181 one (was broken since 4.91).
182 - AE::Handle did not properly clear rtimeout/wtimeout watchers
183 when retrying a tcp_connect.
184 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
185 - fixed wrong documentation in AE::Socket.
186
1875.01 Mon Aug 10 03:16:32 CEST 2009
188 - last release broke EV child watchers :/.
189
1905.0 Sun Aug 9 17:32:09 CEST 2009
191 - officially introduce and document the AE API.
192 - lowering the timeout at runtime did not have immediate effect
193 in AnyEvent::Handle.
194 - AE::Handle now has separate and independent read and write timeout
195 settings.
196 - probe for Guard module when AnyEvent::Util is loaded, not at
197 runtime (Event doesn't like eval "use" from callbacks, and it avoids
198 exporting the probe function).
199 - Impl::IOAsync's io watchers did not use the filehandle emulation
200 code to work around it's watcher limits.
201 - work around IO::Async's problems with overloaded objects.
202 - try to disable wearnings inside IO::Async, it's just too buggy.
203 - do no longer use Event by default, only EV or Perl, Event is too buggy.
204 - hackishly provide fast AE:: interface when EV or Perl is the backend.
205 - inherit VERSION from AnyEvent::VERSION in most modules.
206
2074.91 Thu Aug 6 15:42:45 CEST 2009
208 - AE::Handle::starttls could get out of sync when the read buffer
209 already contains some TLS handshake.
210 - AE::Handle did not properly free the TLS session, causing bigger
211 memory leaks in Net::SSLeay than expected (reported by toaster).
212 - AE::Socket::tcp_connect will now artificially delay invoking
213 the callback to avoid returning after invoking the callback.
214 - convert many internal watcher uses to AE API.
215
2164.9 Sat Aug 1 11:07:01 CEST 2009
217 - Glib silently fails when registering a timeout with
218 a negative value, so avoid that.
219 - call condvar callback immediately when it is set
220 after the condition is already signalled.
221 - check rbuf_max condition only after trying to consume
222 data.
223 - normalise signal numbers to names when using the ae's
224 signal handling, but do not document this (yet).
225 - pure perl signal emulation did not properly set nonblocking
226 mode on the signal pipe on win32 (this is unlikely to have caused
227 any issues).
228 - new module: AnyEvent::Debug.
229 - AnyEvent::Strict now checks that a signal was specified
230 by name (not name or number).
231 - reduce memory footprint in typical cases by ~50kb by
232 only compiling signal code when necessary.
233 - add AnyEvent::Handle->rbuf_max.
234 - grab the AE namespace for future new API, implement stubs for
235 said future/alternative API.
236 - new function: AnyEvent::Socket::format_hostport.
237
2384.881 Tue Jul 28 12:51:53 CEST 2009
239 - work around a bug in local in pre-5.10 perls, causing
240 AnyEvent::Handle to recurse when it should not
241 (analysed by elmex).
242
2434.88 Tue Jul 28 04:04:37 CEST 2009
244 - re-bless the handle into a dummy package after calling
245 AnyEvent::Handle::destroy, so the user does not need to check
246 for errors after every push_write etc.
247 - do not attempt to run t/02_signals.t on obviously broken
248 platforms.
249
2504.87 Sun Jul 26 02:06:16 CEST 2009
251 - do not attempt to linger when there is no longer a valid fh.
252 - enforce tls mode to be either accept or connect, do not simply
253 segfault in Net::SSLeay.
254 - AnyEvent::Handle can now call tcp_connect itself (new parameters
255 connect, on_prepare, on_connect and on_connect_error). Updated
256 tutorial accordingly.
257 - add AnyEvent::Impl::Irssi backend.
258
2594.86 Mon Jul 20 23:52:29 CEST 2009
260 - since the verbose warning is not enough, explicitly document
261 that versions before 1.33 of Net::SSLeay are not secure.
262 - work around signal handling races in Event and (...) Event::Lib.
263 - try to align signal-race timer to full-second boundaries.
264 - work around Tk not liking negative timeouts.
265 - don't complain of different grades of environmental unfriendlyness
266 in IO::Async.
267
2684.85 Sat Jul 18 06:16:14 CEST 2009
269 - nail the signal race problem in perl once and for all
270 (see $AnyEvent::MAX_SIGNAL_LATENCY).
271 - take advantage of Async::Interrupt if it is available.
272 - load Time::HiRes and Guard modules on demand only.
273 - add optional/recommended modules section to AnyEvent
274 documentation.
275 - reduce memory usage considerably (and reduce startup penalty)
276 by not using "strict", "warnings" and "overload" modules.
277 - work around buggy windows/openbsd perls and provide EBADMSG
278 and EPROTO ourselves when missing.
279 - improve perl 5.6 compatibility of the core event loop.
280 - made Net::SSLeay version 1.33 a soft requirement.
281
2824.83 Fri Jul 17 16:56:26 CEST 2009
283 - implement AnyEvent::Socket::getprotobyname.
284 - AnyEvent::CondVar's will now detect recursive blocking
285 waits and will croak, as too many people fall into
286 this trap.
287 - AnyEvent::Handle will now call ->destroy on itself after
288 executing the on_error callback, instead of doing some
289 half-baked internal shutdown, for fatal errors.
290 - clarify on_eof behaviour w.r.t. the read queue and
291 on_read callbacks.
292 - ignore some possible spurious wake-ups in tcp_connect.
293
2944.82 Sat Jul 11 00:34:55 CEST 2009
295 - POE and Event backends didn't accept some callable objects as
296 callbacks.
297 - use Config module instead of POSIX module to detect signal names
298 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
299 module doesn't even have all POSIX signals :/.
300 - use more workarounds around the many refcnt/corruption bugs in
301 Event::Lib.
302 - work around a race condition in perl's select, causing t/03_child.t
303 to rarely fail.
304
3054.81 Thu Jul 9 10:30:30 CEST 2009
306 - AnyEvent::Handle didn't properly diagnose write errors
307 (it expected -1 from syswrite, how lame... :).
308 - support file descriptors in addition to file handles
309 in AnyEvent->io.
310 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
311 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
312 and PERL_ANYEVENT_CA_PATH.
313 - provide a sensible synopsis section for AnyEvent::TLS.
314 - add a "supported backends" section to the manpage.
315 - added simple io watcher test to testsuite, using a
316 portable_socketpair.
317 - tried to improve the stability of the Event::Lib backend,
318 YMMV.
319
3204.8 Mon Jul 6 23:45:16 CEST 2009
321 - AnyEvent::DNS did not properly follow CNAME records with
322 uppercase targets.
323 - AnyEvent::DNS would errornously return AAAA records
324 with v4 mapped addresses (a faulty record) as ipv4 addresses,
325 causing AnyEvent::Socket to throw an exception.
4 - added new module AnyEvent::TLS for easier SSL/TLS context 326 - added new module AnyEvent::TLS for easier SSL/TLS context
5 creation, with many options including hostname verification, 327 creation, with many options including hostname verification,
6 secure default configuration, lots of documentation and, 328 secure default configuration, lots of documentation and,
7 predefined diffie-hellman keys for perfect forward security 329 predefined diffie-hellman keys for perfect forward security
8 and much more. get it while it's still fresh! 330 and much more. get it while it's still fresh!
… …
11 so AnyEvent::Socket users have smaller memory footprint 333 so AnyEvent::Socket users have smaller memory footprint
12 in the common case. 334 in the common case.
13 - add AnyEvent::Handle->push_shutdown method. 335 - add AnyEvent::Handle->push_shutdown method.
14 - add an additional error message parameter to AnyEvent::Handle's 336 - add an additional error message parameter to AnyEvent::Handle's
15 on_error callback (for TLS, $! is still available). 337 on_error callback (for TLS, $! is still available).
338 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
16 - make AnyEvent::Handle more robust against early conenction 339 - make AnyEvent::Handle more robust against early conenction
17 failures (during new), and return C<undef> in that case 340 failures (during new), and return C<undef> in that case
18 from the constructor. 341 from the constructor.
19 - AnyEvent::Handle will now try to load only JSON::XS first, 342 - AnyEvent::Handle will now try to load only JSON::XS first,
20 then fall back to JSON. 343 then fall back to JSON.
344 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
345 and because it was documented that way.
21 346
224.451 Fri Jul 3 00:28:58 CEST 2009 3474.451 Fri Jul 3 00:28:58 CEST 2009
23 - do not clear rbuf when shutting down an AnyEvent::Handle 348 - do not clear rbuf when shutting down an AnyEvent::Handle
24 object - doing so breaks AnyEvent::HTTP. 349 object - doing so breaks AnyEvent::HTTP.
25 350
… …
1434.3 Fri Oct 3 09:18:43 CEST 2008 4684.3 Fri Oct 3 09:18:43 CEST 2008
144 - AnyEvent will now install a no-op signal handler for SIGPIPE, 469 - AnyEvent will now install a no-op signal handler for SIGPIPE,
145 unless one has been installed already. 470 unless one has been installed already.
146 - warn about not loaidng AnyEvent::Impl::POE early enough 471 - warn about not loaidng AnyEvent::Impl::POE early enough
147 (patch by Adam Rosenstein). 472 (patch by Adam Rosenstein).
148 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 473 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
149 - fix a bug in where in SSL connect mode, the 474 - fix a bug in where in SSL connect mode, the
150 client would first wait for some data by the server and 475 client would first wait for some data by the server and
151 otherwise hang. (reported and analyzed in an absolutely 476 otherwise hang. (reported and analysed in an absolutely
152 exemplary manner by Adam Rosenstein). 477 exemplary manner by Adam Rosenstein).
153 - fix a bug in where SSL EOF would not be treated 478 - fix a bug in where SSL EOF would not be treated
154 as stream EOF, putting the connection into a hung state. 479 as stream EOF, putting the connection into a hung state.
155 (reported and analyzed in an absolutely exemplary manner by Adam 480 (reported and analysed in an absolutely exemplary manner by Adam
156 Rosenstein). 481 Rosenstein).
157 - fix a potential segfault when the TLS context 482 - fix a potential segfault when the TLS context
158 would go missing in a read callback (e.g. due to stoptls) - 483 would go missing in a read callback (e.g. due to stoptls) -
159 Net::SSLeay of course makes no type checking whatsoever. 484 Net::SSLeay of course makes no type checking whatsoever.
160 - AnyEvent::Handle will not stop the read watcher in TLS mode. 485 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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