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 1.433 by root, Fri Jan 14 05:36:37 2011 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 strict and anyevent debug
5TODO: anyevent::trace? or debug tracing all watchers?
6TODO: anyevent strict checking for $_ changes?
7
8TODO: resetttls?
95.3 Fri Dec 31 05:48:13 CET 2010
10 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
11 issues with IO::Async have been solved in that version, including
12 the availabilty of a default loop!
13 - implement new wbuf_max parameter for AnyEvent::Handle.
14 - added Cocoa::EventLoop backend by Daisuke Murase.
15 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
16 - clarify handle on_prepare documentation.
17 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
18 to avoid leaking them to any child processes.
19 - try to force files into the arch-dependend lib directory, for the
20 half-"turly obsessive compulsive person"(s) that depend on it.
21
225.29 Sun Dec 5 10:49:21 CET 2010
23 - convert EV backend to EV 4.00 API (so better upgrade EV too).
24 - AE::Handle: implement dynamic read_size adjustment, add
25 max_read_size parameter, reduce default read_size to 2048.
26 - add some FAQ entries for servers and/or on_eof vs. on_error.
27 - work around OS bugs (cygwin again) when lingering in AE::Handle.
28 - reduce memory usage slightly when connect was used in AE::Handle.
29 - wrok around more segfaults in Tk.
30 - document tls_ctx => undef as valid.
31 - detect WNOHANG value at compiletime.
32 - include some probably linux-only support for building AnyEvent
33 as part of the perl core.
34 - improve compatibility of t/08_idna.t to perl 5.8.
35 - make AnyEvent::Strict truly optional (in case somebody builds a
36 minimal perl...).
37
385.28 Wed Oct 13 04:14:23 CEST 2010
39 - due to a glitch, AnyEvent's internal getprotobyname wasn't
40 actually used in tcp_server.
41 - implement AnyEvent::CondVar->new method, which was documented but
42 missing.
43 - support autoloading of read types in unshift_read.
44 - AnyEvent::Strict: do the test for invalid fh arguments differently,
45 resulting in more informative error messages.
46 - new AnyEvent::FAQ manpage.
47 - clarify that you can actually call push_read etc. in on_read.
48 - put keys and certs in /tmp, not into the current directory,
49 and document this in the SECURITY CONSIDERATIONS sections
50 of AnyEvent::TLS.
51 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
52 requested by Ron Isaacson.
53 - document $AnyEvent::DNS::RESOLVER variable.
54 - incorporated major typo patches by Abhijit Menon-Sen.
55 - AnyEvent::Handle now croaks when negative timeouts are passed.
56 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
57 undocumented, but who knows when they might come in handy.
58
595.271 Tue Jun 8 12:05:46 CEST 2010
60 - backport to perl 5.8.x.
61
625.27 Sun Jun 6 12:12:05 CEST 2010
63 - postpone differently in AnyEvent::Socket now, as
64 when not, canceling the connection attempt might fail
65 (found by Felix Antonius Wilhelm Ostmann).
66 - explicitly check for non-stream sockets in AE::Handle, too many
67 clueless people fell into the trap of this somehow working.
68 - simplified and reworked the "OTHER MODULES" section.
69 - better/more condvar examples.
70
715.261 Wed Apr 28 16:13:36 CEST 2010
72 - AF_INET6 was not properly used from Socket6 during configuration
73 time (found by Andy Grundman).
74
755.26 Mon Apr 12 04:49:35 CEST 2010
76 - don't generate spurious readyness notifications when select
77 returns EINTR in the pure perl backend (analysed and testcase
78 by Bas Denissen).
79 - give same examples in SYNOPSIS sections of AnyEvent and AE.
80 - provider faster implementations for AE::time/now/now_update for
81 EV and Perl backends.
82 - provide faster AE::cv and AE::time implementations for all
83 backends.
84 - no longer support lower/mixed-case signal names (this was
85 never documented nor universally supported).
86 - some more memory ยต-optimisations, and somewhat less messy
87 code to cope with both AnyEvent and AE APIs.
88
895.251 Sat Mar 13 00:58:21 CET 2010
90 - make test failed if it was run before make install on perl <=
91 5.8.8, and older versions of perl make our live very difficult,
92 so write constants.pl during Makefile.PL time.
93
945.25 Sat Mar 13 00:23:14 CET 2010
95 - fix a race condition in AnyEvent::Handle that would cause
96 a "bio_read: unsupported method" error (in _another_
97 ssl connection) after user code threw an exception.
98 - added AnyEvent::Handle->destroyed method.
99 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
100 - the AnyEvent::Util::fh_nonblocking function worked only
101 by lucky accident on win32.
102 - smaller and faster AnyEvent::Util::fh_nonblocking.
103 - when the (required!) Time::HiRes module is missing, AnyEvent
104 did not fall back to built-in time properly.
105 - do not load Fcntl at runtime, saving memory and loading time.
106 - precompile a number of constants and use them instead of runtime
107 detection and eval.
108 - free detection code after detection and similar memory optimisations.
109 - Perl backend timer interval best effort drift has been improved
110 (same algorithm as EV).
111 - update unicode idna mapping table.
112
1135.24 Tue Jan 5 11:39:43 CET 2010
114 - cygwin never reports errors from failed connects. 1.5 just gives
115 you continous readyness and EAGAIN, 1.7 is even more broken and
116 just hangs. work around both issues in a cygwin-specific hack.
117 - improve idle watcher documentation slightly.
118
1195.23 Sun Dec 20 23:48:00 CET 2009
120 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
121 - implement punycode_encode/decode, idn_nameprep,
122 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
123 - provide $AE::VERSION.
124 - removed traces of "no strict 'refs'".
125
1265.22 Sat Dec 5 03:51:13 CET 2009
127 - downgrade-or-fail in AnyEvent::Handle::push_write, to
128 diagnose encoding failures earlier and more succinctly.
129 (this works around bugs in perl, throwing away encoding info
130 when passing scalar data to extensions).
131 - add more examples to AnyEvent::Socket manpage.
132 - upgrade internal warning set to the same as common::sense 2.03.
133 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
134 (John Beppu).
135 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
136 - add google public dns servers to fallback server set and make sure
137 we load-balance properly between the three sets. also add all
138 fallback dns servers, not just a random one, to each dns config.
139
1405.21 Thu Nov 19 02:48:47 CET 2009
141 - fix a problem where socket constants were called with parameters
142 (spotted by David Friedland).
143 - fork_call never use'd POSIX (reported by Daisuke Maki).
144 - improve perl 5.6 compatibility further (but it still won't work
145 unless you rip out everything but the core).
146 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
147 uses 10MB of RAM, but doesn't open a console window. *sigh*.
148
1495.202 Wed Oct 14 22:35:44 CEST 2009
150 - AnyEvent::DNS would unexpectedly clobber $_ under windows
151 (analysed by Matthias Waldorf).
152 - AnyEvent::Handle::run_cmd can now pass the PID of the
153 newly-created process, which is much less useful than it might
154 sound (based on patch by Yann Kerherve).
155
1565.201 Tue Sep 29 12:09:25 CEST 2009
157 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
158 (reported by Torsten Foertsch).
159 - common::sense 2.0 could cause tcp_server to throw an exception
160 (analysed by elmex).
161
1625.2 Mon Sep 14 07:04:49 CEST 2009
163 - INCOMPATIBLE CHANGE: do no longer support register_read_type
164 and register_write_type in AnyEvent::Handle, instead support
165 package names (the facility was mostly abused).
166 - implement "packagename-as-read/write type" support in
167 AnyEvent::Handle.
168 - AnyEvent::Handle: new options "keepalive" and "oobinline".
169 - oobinline set by default to avoid security issues.
170 - the pure-perl event loop backend wrongly detected times() fallback
171 support (spotted by Pavel Boldin).
172
1735.12 Tue Sep 1 20:26:50 CEST 2009
174 - be more lenient when parsing resolv.conf files, as some people
175 use hashmarks as comment indicator inside directives (reported
176 by Michael S. Fischer).
177 - use same set of warnings as common::sense 2.0.
178 - fix a potential 32 bit overflow issue due to perl having problems
179 with large hex constants in the Perl backend.
180 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
181 by default.
182 - fix some prototypes.
183 - (optionally) require Async::Interrupt 1.02.
184 - replace opendns fallbacks by verizon/level3 public dns servers.
185
1865.112 Fri Aug 21 13:59:04 CEST 2009
187 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
188 using starttls with an illegal/uncreatable tls context.
189 - port to broken win32 perls.
190
1915.111 Sun Aug 16 18:44:36 CEST 2009
192 - perl errornously requires sockaddr_un structures returned by the
193 kernel to have a specific length. work around this bug by padding
194 sockaddr_un structures (found to fail on OS X, as perl only uses
195 the correct code for GNU/Linux, but likely to fail on other
196 OSes as well).
197
1985.11 Wed Aug 12 17:49:37 CEST 2009
199 - fixed a bug in run_cmd with input from filename.
200 - create a json coder object when none is given in write direction
201 to reduce dependency on JSON.
202
2035.1 Tue Aug 11 03:17:32 CEST 2009
204 - tcp_connect now reports the correct error in $!, not an unrelated
205 one (was broken since 4.91).
206 - AE::Handle did not properly clear rtimeout/wtimeout watchers
207 when retrying a tcp_connect.
208 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
209 - fixed wrong documentation in AE::Socket.
210
2115.01 Mon Aug 10 03:16:32 CEST 2009
212 - last release broke EV child watchers :/.
213
2145.0 Sun Aug 9 17:32:09 CEST 2009
215 - officially introduce and document the AE API.
216 - lowering the timeout at runtime did not have immediate effect
217 in AnyEvent::Handle.
218 - AE::Handle now has separate and independent read and write timeout
219 settings.
220 - probe for Guard module when AnyEvent::Util is loaded, not at
221 runtime (Event doesn't like eval "use" from callbacks, and it avoids
222 exporting the probe function).
223 - Impl::IOAsync's io watchers did not use the filehandle emulation
224 code to work around it's watcher limits.
225 - work around IO::Async's problems with overloaded objects.
226 - try to disable wearnings inside IO::Async, it's just too buggy.
227 - do no longer use Event by default, only EV or Perl, Event is too buggy.
228 - hackishly provide fast AE:: interface when EV or Perl is the backend.
229 - inherit VERSION from AnyEvent::VERSION in most modules.
230
2314.91 Thu Aug 6 15:42:45 CEST 2009
232 - AE::Handle::starttls could get out of sync when the read buffer
233 already contains some TLS handshake.
234 - AE::Handle did not properly free the TLS session, causing bigger
235 memory leaks in Net::SSLeay than expected (reported by toaster).
236 - AE::Socket::tcp_connect will now artificially delay invoking
237 the callback to avoid returning after invoking the callback.
238 - convert many internal watcher uses to AE API.
239
2404.9 Sat Aug 1 11:07:01 CEST 2009
241 - Glib silently fails when registering a timeout with
242 a negative value, so avoid that.
243 - call condvar callback immediately when it is set
244 after the condition is already signalled.
245 - check rbuf_max condition only after trying to consume
246 data.
247 - normalise signal numbers to names when using the ae's
248 signal handling, but do not document this (yet).
249 - pure perl signal emulation did not properly set nonblocking
250 mode on the signal pipe on win32 (this is unlikely to have caused
251 any issues).
252 - new module: AnyEvent::Debug.
253 - AnyEvent::Strict now checks that a signal was specified
254 by name (not name or number).
255 - reduce memory footprint in typical cases by ~50kb by
256 only compiling signal code when necessary.
257 - add AnyEvent::Handle->rbuf_max.
258 - grab the AE namespace for future new API, implement stubs for
259 said future/alternative API.
260 - new function: AnyEvent::Socket::format_hostport.
261
2624.881 Tue Jul 28 12:51:53 CEST 2009
263 - work around a bug in local in pre-5.10 perls, causing
264 AnyEvent::Handle to recurse when it should not
265 (analysed by elmex).
266
2674.88 Tue Jul 28 04:04:37 CEST 2009
268 - re-bless the handle into a dummy package after calling
269 AnyEvent::Handle::destroy, so the user does not need to check
270 for errors after every push_write etc.
271 - do not attempt to run t/02_signals.t on obviously broken
272 platforms.
273
2744.87 Sun Jul 26 02:06:16 CEST 2009
275 - do not attempt to linger when there is no longer a valid fh.
276 - enforce tls mode to be either accept or connect, do not simply
277 segfault in Net::SSLeay.
278 - AnyEvent::Handle can now call tcp_connect itself (new parameters
279 connect, on_prepare, on_connect and on_connect_error). Updated
280 tutorial accordingly.
281 - add AnyEvent::Impl::Irssi backend.
282
2834.86 Mon Jul 20 23:52:29 CEST 2009
284 - since the verbose warning is not enough, explicitly document
285 that versions before 1.33 of Net::SSLeay are not secure.
286 - work around signal handling races in Event and (...) Event::Lib.
287 - try to align signal-race timer to full-second boundaries.
288 - work around Tk not liking negative timeouts.
289 - don't complain of different grades of environmental unfriendlyness
290 in IO::Async.
291
2924.85 Sat Jul 18 06:16:14 CEST 2009
293 - nail the signal race problem in perl once and for all
294 (see $AnyEvent::MAX_SIGNAL_LATENCY).
295 - take advantage of Async::Interrupt if it is available.
296 - load Time::HiRes and Guard modules on demand only.
297 - add optional/recommended modules section to AnyEvent
298 documentation.
299 - reduce memory usage considerably (and reduce startup penalty)
300 by not using "strict", "warnings" and "overload" modules.
301 - work around buggy windows/openbsd perls and provide EBADMSG
302 and EPROTO ourselves when missing.
303 - improve perl 5.6 compatibility of the core event loop.
304 - made Net::SSLeay version 1.33 a soft requirement.
305
3064.83 Fri Jul 17 16:56:26 CEST 2009
307 - implement AnyEvent::Socket::getprotobyname.
308 - AnyEvent::CondVar's will now detect recursive blocking
309 waits and will croak, as too many people fall into
310 this trap.
311 - AnyEvent::Handle will now call ->destroy on itself after
312 executing the on_error callback, instead of doing some
313 half-baked internal shutdown, for fatal errors.
314 - clarify on_eof behaviour w.r.t. the read queue and
315 on_read callbacks.
316 - ignore some possible spurious wake-ups in tcp_connect.
317
3184.82 Sat Jul 11 00:34:55 CEST 2009
319 - POE and Event backends didn't accept some callable objects as
320 callbacks.
321 - use Config module instead of POSIX module to detect signal names
322 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
323 module doesn't even have all POSIX signals :/.
324 - use more workarounds around the many refcnt/corruption bugs in
325 Event::Lib.
326 - work around a race condition in perl's select, causing t/03_child.t
327 to rarely fail.
328
3294.81 Thu Jul 9 10:30:30 CEST 2009
330 - AnyEvent::Handle didn't properly diagnose write errors
331 (it expected -1 from syswrite, how lame... :).
332 - support file descriptors in addition to file handles
333 in AnyEvent->io.
334 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
335 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
336 and PERL_ANYEVENT_CA_PATH.
337 - provide a sensible synopsis section for AnyEvent::TLS.
338 - add a "supported backends" section to the manpage.
339 - added simple io watcher test to testsuite, using a
340 portable_socketpair.
341 - tried to improve the stability of the Event::Lib backend,
342 YMMV.
343
3444.8 Mon Jul 6 23:45:16 CEST 2009
345 - AnyEvent::DNS did not properly follow CNAME records with
346 uppercase targets.
347 - AnyEvent::DNS would errornously return AAAA records
348 with v4 mapped addresses (a faulty record) as ipv4 addresses,
349 causing AnyEvent::Socket to throw an exception.
4 - added new module AnyEvent::TLS for easier SSL/TLS context 350 - added new module AnyEvent::TLS for easier SSL/TLS context
5 creation, with many options including hostname verification, 351 creation, with many options including hostname verification,
6 secure default configuration, lots of documentation and, 352 secure default configuration, lots of documentation and,
7 predefined diffie-hellman keys for perfect forward security 353 predefined diffie-hellman keys for perfect forward security
8 and much more. get it while it's still fresh! 354 and much more. get it while it's still fresh!
… …
11 so AnyEvent::Socket users have smaller memory footprint 357 so AnyEvent::Socket users have smaller memory footprint
12 in the common case. 358 in the common case.
13 - add AnyEvent::Handle->push_shutdown method. 359 - add AnyEvent::Handle->push_shutdown method.
14 - add an additional error message parameter to AnyEvent::Handle's 360 - add an additional error message parameter to AnyEvent::Handle's
15 on_error callback (for TLS, $! is still available). 361 on_error callback (for TLS, $! is still available).
362 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
16 - make AnyEvent::Handle more robust against early conenction 363 - make AnyEvent::Handle more robust against early conenction
17 failures (during new), and return C<undef> in that case 364 failures (during new), and return C<undef> in that case
18 from the constructor. 365 from the constructor.
19 - AnyEvent::Handle will now try to load only JSON::XS first, 366 - AnyEvent::Handle will now try to load only JSON::XS first,
20 then fall back to JSON. 367 then fall back to JSON.
368 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
369 and because it was documented that way.
21 370
224.451 Fri Jul 3 00:28:58 CEST 2009 3714.451 Fri Jul 3 00:28:58 CEST 2009
23 - do not clear rbuf when shutting down an AnyEvent::Handle 372 - do not clear rbuf when shutting down an AnyEvent::Handle
24 object - doing so breaks AnyEvent::HTTP. 373 object - doing so breaks AnyEvent::HTTP.
25 374
… …
1434.3 Fri Oct 3 09:18:43 CEST 2008 4924.3 Fri Oct 3 09:18:43 CEST 2008
144 - AnyEvent will now install a no-op signal handler for SIGPIPE, 493 - AnyEvent will now install a no-op signal handler for SIGPIPE,
145 unless one has been installed already. 494 unless one has been installed already.
146 - warn about not loaidng AnyEvent::Impl::POE early enough 495 - warn about not loaidng AnyEvent::Impl::POE early enough
147 (patch by Adam Rosenstein). 496 (patch by Adam Rosenstein).
148 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 497 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
149 - fix a bug in where in SSL connect mode, the 498 - fix a bug in where in SSL connect mode, the
150 client would first wait for some data by the server and 499 client would first wait for some data by the server and
151 otherwise hang. (reported and analyzed in an absolutely 500 otherwise hang. (reported and analysed in an absolutely
152 exemplary manner by Adam Rosenstein). 501 exemplary manner by Adam Rosenstein).
153 - fix a bug in where SSL EOF would not be treated 502 - fix a bug in where SSL EOF would not be treated
154 as stream EOF, putting the connection into a hung state. 503 as stream EOF, putting the connection into a hung state.
155 (reported and analyzed in an absolutely exemplary manner by Adam 504 (reported and analysed in an absolutely exemplary manner by Adam
156 Rosenstein). 505 Rosenstein).
157 - fix a potential segfault when the TLS context 506 - fix a potential segfault when the TLS context
158 would go missing in a read callback (e.g. due to stoptls) - 507 would go missing in a read callback (e.g. due to stoptls) -
159 Net::SSLeay of course makes no type checking whatsoever. 508 Net::SSLeay of course makes no type checking whatsoever.
160 - AnyEvent::Handle will not stop the read watcher in TLS mode. 509 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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