| 1 | Revision history for Perl extension AnyEvent. |
1 | Revision history for Perl extension AnyEvent. |
|
|
2 | |
|
|
3 | TODO: document TCP_*** constants |
|
|
4 | TODO: add debug/trace logging to some modules |
|
|
5 | TODO: use the eval-with-cb-call trick also to speed up JSON? |
|
|
6 | TODO: FAQ: common pitfalls? invoke-before-return |
|
|
7 | TODO: FAQ: scope-ids? hosts file order? |
|
|
8 | TODO: AnyEvent::Handle with AnyEvent::IO? |
|
|
9 | |
|
|
10 | 7.02 Tue Aug 14 04:27:58 CEST 2012 |
|
|
11 | - AnyEvent::Util::run_cmd could block indefinitely (analyzed and test |
|
|
12 | program by Yorhel). |
|
|
13 | - verified that AnyEvent::Socket follows RFC5952. |
|
|
14 | - try to parse "ADDR#PORT" in addition to "ADDR PORT". |
|
|
15 | |
|
|
16 | 7.01 Sun May 13 01:03:17 CEST 2012 |
|
|
17 | - fail with EPROTO in AnyEvent::Handle wqhen TLS is requested but |
|
|
18 | not available, instead of throwing an exception. |
|
|
19 | - use File::Spec to get the tmpdir in t/*, to avoid needless |
|
|
20 | failures on (most, not mine :) windows boxes. |
|
|
21 | - new handle read types: tls_detect and tls_autostart. |
|
|
22 | |
|
|
23 | 7.0 Fri Apr 13 06:33:30 CEST 2012 |
|
|
24 | - child watchers are broken in POE 1.352 (also many earlier |
|
|
25 | versions) and there seems to be no way to work around it, as POE |
|
|
26 | itself is inherently racy. Document this regression and add a delay |
|
|
27 | in t/68_poe_03_child.t for the time being. |
|
|
28 | - new module AnyEvent::IO, that is a frontend to either a pure-perl |
|
|
29 | synchronous I/O implementation (AnyEvent::IO::Perl), or to the |
|
|
30 | asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO). |
|
|
31 | - load /etc/hosts only when DNS has no answer. |
|
|
32 | - stat /etc/hosts on every access and reload it if it changed. |
|
|
33 | - load /etc/hosts via AnyEvent::IO - potentially asynchronous. |
|
|
34 | - fix a buggy croak in the dh parameter reading in AnyEvent::TLS. |
|
|
35 | - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As |
|
|
36 | a side effect, they now use true appending as opposed to libc |
|
|
37 | appending, and the file might not have been opened when the function |
|
|
38 | returns. |
|
|
39 | - the default logging level was not properly documented in a variety of |
|
|
40 | places, this has hopefully been rectified. |
|
|
41 | - updated uts46data.pl for unicode 6.1.0. |
|
|
42 | - made log messages generated by AnyEvent submodules not |
|
|
43 | include the package name anymore, as it will be logged |
|
|
44 | by default already. |
|
|
45 | - upgrade to the trick used by common-sense 3.5 to work |
|
|
46 | around extra warning torture/breakage under perl 5.15.x. |
|
|
47 | - log messages by anyevent are now ucfirst, are usually full |
|
|
48 | sentences and do no longer include the package name. |
|
|
49 | - the storable read type would sometimes throw an exception instead |
|
|
50 | of causing EBADMSG (analyzed by Maxime Soulé). |
|
|
51 | - work around a bug in openssl 1.0.1 which enforces a minimum rsa |
|
|
52 | keysize (reported by Paul Howarth). |
|
|
53 | - documented Rocco fabricating statements about AnyEvent and me. |
|
|
54 | |
|
|
55 | 6.14 Tue Jan 31 20:00:24 CET 2012 |
|
|
56 | - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside, |
|
|
57 | probably an editing mistake (reported by Darin McBride). |
|
|
58 | |
|
|
59 | 6.13 Thu Jan 12 07:27:01 CET 2012 |
|
|
60 | - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way, |
|
|
61 | causing it to fail if POSIX wasn't loaded already (analysed by Rock Power). |
|
|
62 | - AnyEvent::Handle::push_read (line => did pass $1 directly, so |
|
|
63 | regex-matching inside the callback would change the parameter, despite |
|
|
64 | $1 being dynamically scoped per-block. this perl bug is now being |
|
|
65 | worked around (testcase by Cindy Wang). |
|
|
66 | |
|
|
67 | 6.12 Mon Dec 12 13:21:10 CET 2011 |
|
|
68 | - $! was clobbered by subsequent calls in tcp_connect, due to |
|
|
69 | the postpone not saving/restoring it (analyzed by Richard Garnish). |
|
|
70 | |
|
|
71 | 6.11 Tue Nov 22 10:36:05 CET 2011 |
|
|
72 | - Tk cannot create windows when tainted unless you set a title, so |
|
|
73 | set a dummy title for AnyEvent's dummy mainwindow. How dummy. |
|
|
74 | - escape any nonprintable/nonascii characters when stringifying |
|
|
75 | backtraces. |
|
|
76 | - log the reason loading an interface module fails to load at level debug. |
|
|
77 | |
|
|
78 | 6.1 Tue Oct 4 19:44:30 CEST 2011 |
|
|
79 | - INCOMPATIBLE CHANGE: the default log level is now "4" (error |
|
|
80 | and above), and some messages inside AnyEvent have been elevated |
|
|
81 | to higher log levels to print by default. |
|
|
82 | - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal. |
|
|
83 | - work around yet another signal race bug, newly introduced by POE. My |
|
|
84 | how that module sucks. |
|
|
85 | - add google public ipv6 dns as fallback servers, so AE::DNS can finally |
|
|
86 | support ipv6-only operation even with defaults (also use ::1 instead |
|
|
87 | of 127.0.0.1). |
|
|
88 | - give better guidelines on how to chose a logging level. |
|
|
89 | - AnyEvent::Log can now cap (limit) log message priority by context. |
|
|
90 | - AE::log will now save and restore $! and $@, so callers don't have to. |
|
|
91 | - AE::Strict now checks that registered fds are still valid at regular |
|
|
92 | intervals. |
|
|
93 | - skip frequently called 'require' calls if the module appears loaded. |
|
|
94 | - increase signal latency test timer considerably for poor little osx, |
|
|
95 | which can't be bothered with basic posix functionality such as signal |
|
|
96 | delivery. |
|
|
97 | - mention some debugging aids in the tutorial. |
|
|
98 | - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY. |
|
|
99 | - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name |
|
|
100 | of the context as title. |
|
|
101 | - AnyEvent::Strict would not restore read-write access to $_ when a callback |
|
|
102 | died - this is a user bug, but since AE::Strict exists to help dbeugging... |
|
|
103 | - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse |
|
|
104 | to work with it - the author is well aware of the issues but refuses |
|
|
105 | to fix them. This extreme step has been taken because the module actively |
|
|
106 | attacks the goals of AnyEvent and makes it harder for module authors to use |
|
|
107 | AnyEvent. |
|
|
108 | |
|
|
109 | 6.02 Fri Aug 26 20:08:31 CEST 2011 |
|
|
110 | - log_to_syslog now accepts facility strings. |
|
|
111 | - log_to_syslog would not allow facilities - "$facility|$level" |
|
|
112 | *literally* means that string. sheesh. |
|
|
113 | - the CPAN parser got confused and thought we provide Sys::Syslog, try |
|
|
114 | to work around. |
|
|
115 | - renamed AnyEvent::Impl::FLTK2 to FLTK. |
|
|
116 | |
|
|
117 | 6.01 Fri Aug 26 07:04:11 CEST 2011 |
|
|
118 | - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include |
|
|
119 | the TTL as fourth element - this affects ->resolve, ->request |
|
|
120 | and ->dns_unpack indirectly. Convenience functions such as |
|
|
121 | AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is |
|
|
122 | AnyEvent::Socket. |
|
|
123 | - INCOMPATIBLE CHANGE: backend authors now should not implement |
|
|
124 | one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll |
|
|
125 | methods. |
|
|
126 | - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see |
|
|
127 | AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details. |
|
|
128 | - DEPRECIATION: please change |
|
|
129 | "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and |
|
|
130 | "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run" |
|
|
131 | in your programs, for future compatibility. |
|
|
132 | - *blush* idle watcher emulation was borked. |
|
|
133 | - AnyEvent::Strict now checks the AE::xxx functions as well. |
|
|
134 | - querying too long domain names would barf deep inside AnyEvent::DNS - since |
|
|
135 | this is hard to avoid for applications, AnyEvent:.DNS::request now checks for |
|
|
136 | for too long domain names explicitly and passes an undef to the callback |
|
|
137 | instead (reported by James Bromberger). |
|
|
138 | - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield). |
|
|
139 | - new functions: AnyEvent::postpone, AnyEvent::log. |
|
|
140 | - new module: AnyEvent::Log, for simple logging needs. |
|
|
141 | - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config. |
|
|
142 | - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start |
|
|
143 | a debug shell. |
|
|
144 | - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable |
|
|
145 | watcher instrumentation/wrapping (see AnyEvent::Debug::wrap). |
|
|
146 | - new env variable, PERL_ANYEVENT_HOSTS, to override the place where |
|
|
147 | /etc/hosts can be found. |
|
|
148 | - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default |
|
|
149 | configuration. |
|
|
150 | - env variables can now be specified using AE_xyz instead of |
|
|
151 | PERL_ANYEVENT_xyz, subject to some rules. |
|
|
152 | - lots of new AnyEvent::Debug functionality. |
|
|
153 | - AnyEvent::Strict now checks for modifications of $_ by the callback |
|
|
154 | (by setting it to readonly while executing the callback, if possible). |
|
|
155 | - IO::Async sometimes cancels the wrong timer when you call cancel_timer. |
|
|
156 | Invest extra overhead to work around this peculiar behaviour. |
|
|
157 | - split the pure perl backend into a loop module and an impl |
|
|
158 | module (AnyEvent::Loop and AnyEvent::Impl::Perl). |
|
|
159 | - some read types (regex, netstring, json, storable) could |
|
|
160 | errornously keep an AE::Handle object alive on their own |
|
|
161 | (reported by Mohammad Toossi). |
|
|
162 | - AnyEvent::Socket::resolve_sockaddr and all functions using it now |
|
|
163 | supports /etc/hosts overrides. |
|
|
164 | - add more workarounds around child watcher bugs in IO::Async |
|
|
165 | and (newly introduced) in POE. |
|
|
166 | - use glib child watchers and try to work around its limitations, |
|
|
167 | if possible (based on a patch by Kevin Ryde). |
|
|
168 | - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2. |
|
|
169 | - many read types in AnyEvent::Handle would malfunction during |
|
|
170 | transitions to TLS because of a "delete $self->{rbuf}". |
|
|
171 | (reported by Gerald Galster). |
|
|
172 | - improve condvar blocking wait performance (probably only noticable |
|
|
173 | with faster backends), also make it easier for event loops that |
|
|
174 | do not support blocking waits. |
|
|
175 | - PERL_ANYEVENT_MODEL now also allows full module names, although |
|
|
176 | the usefulness of this feature is not yet clear. |
|
|
177 | - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS. |
|
|
178 | - add t/80_ssltext.t. |
|
|
179 | - no longer use AUTOLOAD for the initial loading (for no very good reason). |
|
|
180 | - tcp_server now tries to unlink unix domain sockets when it gets destroyed. |
|
|
181 | - do not run event loop detection multiple times when programs call their cached |
|
|
182 | copies of AnyEvent::detect. |
|
|
183 | - parse_hostport now accepts absolute paths als unix domain sockets. |
|
|
184 | - suppress the idiotic warning inside POE by patching the POE::Kernel |
|
|
185 | run flags instead of asking the user to jump through contortions. |
|
|
186 | - add Task::Weaken as dependency (as recommends only at this time). |
|
|
187 | - replace PF_UNSPEC by 0, for good old minix. |
|
|
188 | |
|
|
189 | 6.0 (not officially released, but was on CPAN for about 15 minutes :/) |
|
|
190 | |
|
|
191 | 5.34 Thu May 12 09:59:41 CEST 2011 |
|
|
192 | - AE::TLS didn't support both cert_file and key at the same time, |
|
|
193 | due to some copy&paste bug (reported and analyzed by vti). |
|
|
194 | - improve AE::Handle timeout documentation slightly. |
|
|
195 | |
|
|
196 | 5.33 Mon Feb 7 21:51:41 CET 2011 |
|
|
197 | - never appeared on CPAN, for unknown reasons. |
|
|
198 | - AnyEvent::Handle on_connect_error - do not try to destroy a handle |
|
|
199 | object when it's already gone (analyzed by Ryan Bullock). |
|
|
200 | - update idna algorithm(s) to UTS#46 version 6.0.0. |
|
|
201 | - preserve trailing dot in idn_to_ascii - testsuite says to preserve, |
|
|
202 | uts46 says it's an error, and testsuite makes more sense. |
|
|
203 | - remove leading dots in idn_to_ascii - testsuite says to remove, |
|
|
204 | uts46 says it's an error, and testsuite makes more sense. |
|
|
205 | - fixed a bug in idn ignored character processing. |
|
|
206 | - passes the uts46 6.0.0 testsuite except for three cases which |
|
|
207 | apparently are bugs in the testsuite itself. |
|
|
208 | - passes the uts46 6.0.1 testsuite except for eleven cases which |
|
|
209 | apparently are bugs in the testsuite itself. |
|
|
210 | |
|
|
211 | 5.31 Sun Jan 23 11:49:19 CET 2011 |
|
|
212 | - work around a POE bug where POE would first poll for new events and |
|
|
213 | invoke handlers BEFORE handling the message queue. |
|
|
214 | - fix parsing of "domain" statement in resolv.conf (Johannes Plunien). |
|
|
215 | - stop_read now does just that in TLS mode. see its |
|
|
216 | documentation for caveats. |
|
|
217 | - update to common::sense 3.4 warning set (some warnings in perl impose |
|
|
218 | an arbitrary 32kb limit on regex matches, which affects |
|
|
219 | AnyEvent::Handle). |
|
|
220 | - AE::Handle had a broken rbuf_max method and missing wbuf_max method. |
|
|
221 | - new experimental AE::Handle::resettls method. |
|
|
222 | |
|
|
223 | 5.3 Fri Dec 31 05:48:13 CET 2010 |
|
|
224 | - major IOAsync patch by Paul Evans - 0.33 is required, as the major |
|
|
225 | issues with IO::Async have been solved in that version, including |
|
|
226 | the availabilty of a default loop! |
|
|
227 | - implement new wbuf_max parameter for AnyEvent::Handle. |
|
|
228 | - added Cocoa::EventLoop backend by Daisuke Murase. |
|
|
229 | - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans). |
|
|
230 | - clarify handle on_prepare documentation. |
|
|
231 | - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd, |
|
|
232 | to avoid leaking them to any child processes. |
|
|
233 | - try to force files into the arch-dependend lib directory, for the |
|
|
234 | half-"turly obsessive compulsive person"(s) that depend on it. |
|
|
235 | |
|
|
236 | 5.29 Sun Dec 5 10:49:21 CET 2010 |
|
|
237 | - convert EV backend to EV 4.00 API (so better upgrade EV too). |
|
|
238 | - AE::Handle: implement dynamic read_size adjustment, add |
|
|
239 | max_read_size parameter, reduce default read_size to 2048. |
|
|
240 | - add some FAQ entries for servers and/or on_eof vs. on_error. |
|
|
241 | - work around OS bugs (cygwin again) when lingering in AE::Handle. |
|
|
242 | - reduce memory usage slightly when connect was used in AE::Handle. |
|
|
243 | - wrok around more segfaults in Tk. |
|
|
244 | - document tls_ctx => undef as valid. |
|
|
245 | - detect WNOHANG value at compiletime. |
|
|
246 | - include some probably linux-only support for building AnyEvent |
|
|
247 | as part of the perl core. |
|
|
248 | - improve compatibility of t/08_idna.t to perl 5.8. |
|
|
249 | - make AnyEvent::Strict truly optional (in case somebody builds a |
|
|
250 | minimal perl...). |
|
|
251 | |
|
|
252 | 5.28 Wed Oct 13 04:14:23 CEST 2010 |
|
|
253 | - due to a glitch, AnyEvent's internal getprotobyname wasn't |
|
|
254 | actually used in tcp_server. |
|
|
255 | - implement AnyEvent::CondVar->new method, which was documented but |
|
|
256 | missing. |
|
|
257 | - support autoloading of read types in unshift_read. |
|
|
258 | - AnyEvent::Strict: do the test for invalid fh arguments differently, |
|
|
259 | resulting in more informative error messages. |
|
|
260 | - new AnyEvent::FAQ manpage. |
|
|
261 | - clarify that you can actually call push_read etc. in on_read. |
|
|
262 | - put keys and certs in /tmp, not into the current directory, |
|
|
263 | and document this in the SECURITY CONSIDERATIONS sections |
|
|
264 | of AnyEvent::TLS. |
|
|
265 | - support ";" as resolv.conf comment character in AnyEvent::DNS, as |
|
|
266 | requested by Ron Isaacson. |
|
|
267 | - document $AnyEvent::DNS::RESOLVER variable. |
|
|
268 | - incorporated major typo patches by Abhijit Menon-Sen. |
|
|
269 | - AnyEvent::Handle now croaks when negative timeouts are passed. |
|
|
270 | - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util, |
|
|
271 | undocumented, but who knows when they might come in handy. |
|
|
272 | |
|
|
273 | 5.271 Tue Jun 8 12:05:46 CEST 2010 |
|
|
274 | - backport to perl 5.8.x. |
|
|
275 | |
|
|
276 | 5.27 Sun Jun 6 12:12:05 CEST 2010 |
|
|
277 | - postpone differently in AnyEvent::Socket now, as |
|
|
278 | when not, canceling the connection attempt might fail |
|
|
279 | (found by Felix Antonius Wilhelm Ostmann). |
|
|
280 | - explicitly check for non-stream sockets in AE::Handle, too many |
|
|
281 | clueless people fell into the trap of this somehow working. |
|
|
282 | - simplified and reworked the "OTHER MODULES" section. |
|
|
283 | - better/more condvar examples. |
|
|
284 | |
|
|
285 | 5.261 Wed Apr 28 16:13:36 CEST 2010 |
|
|
286 | - AF_INET6 was not properly used from Socket6 during configuration |
|
|
287 | time (found by Andy Grundman). |
|
|
288 | |
|
|
289 | 5.26 Mon Apr 12 04:49:35 CEST 2010 |
|
|
290 | - don't generate spurious readyness notifications when select |
|
|
291 | returns EINTR in the pure perl backend (analysed and testcase |
|
|
292 | by Bas Denissen). |
|
|
293 | - give same examples in SYNOPSIS sections of AnyEvent and AE. |
|
|
294 | - provider faster implementations for AE::time/now/now_update for |
|
|
295 | EV and Perl backends. |
|
|
296 | - provide faster AE::cv and AE::time implementations for all |
|
|
297 | backends. |
|
|
298 | - no longer support lower/mixed-case signal names (this was |
|
|
299 | never documented nor universally supported). |
|
|
300 | - some more memory µ-optimisations, and somewhat less messy |
|
|
301 | code to cope with both AnyEvent and AE APIs. |
| 2 | |
302 | |
| 3 | 5.251 Sat Mar 13 00:58:21 CET 2010 |
303 | 5.251 Sat Mar 13 00:58:21 CET 2010 |
| 4 | - make test failed if it was run before make install on perl <= |
304 | - make test failed if it was run before make install on perl <= |
| 5 | 5.8.8, and older versions of perl make our live very difficult, |
305 | 5.8.8, and older versions of perl make our live very difficult, |
| 6 | so write constants.pl during Makefile.PL time. |
306 | so write constants.pl during Makefile.PL time. |
| … | |
… | |
| 19 | - do not load Fcntl at runtime, saving memory and loading time. |
319 | - do not load Fcntl at runtime, saving memory and loading time. |
| 20 | - precompile a number of constants and use them instead of runtime |
320 | - precompile a number of constants and use them instead of runtime |
| 21 | detection and eval. |
321 | detection and eval. |
| 22 | - free detection code after detection and similar memory optimisations. |
322 | - free detection code after detection and similar memory optimisations. |
| 23 | - Perl backend timer interval best effort drift has been improved |
323 | - Perl backend timer interval best effort drift has been improved |
| 24 | (ame algorithm as EV). |
324 | (same algorithm as EV). |
| 25 | - update unicode idna mapping table. |
325 | - update unicode idna mapping table. |
| 26 | |
326 | |
| 27 | 5.24 Tue Jan 5 11:39:43 CET 2010 |
327 | 5.24 Tue Jan 5 11:39:43 CET 2010 |
| 28 | - cygwin never reports errors from failed connects. 1.5 just gives |
328 | - cygwin never reports errors from failed connects. 1.5 just gives |
| 29 | you continous readyness and EAGAIN, 1.7 is even more broken and |
329 | you continous readyness and EAGAIN, 1.7 is even more broken and |
| … | |
… | |
| 406 | 4.3 Fri Oct 3 09:18:43 CEST 2008 |
706 | 4.3 Fri Oct 3 09:18:43 CEST 2008 |
| 407 | - AnyEvent will now install a no-op signal handler for SIGPIPE, |
707 | - AnyEvent will now install a no-op signal handler for SIGPIPE, |
| 408 | unless one has been installed already. |
708 | unless one has been installed already. |
| 409 | - warn about not loaidng AnyEvent::Impl::POE early enough |
709 | - warn about not loaidng AnyEvent::Impl::POE early enough |
| 410 | (patch by Adam Rosenstein). |
710 | (patch by Adam Rosenstein). |
| 411 | - fixed a great number of bugs and corner cases in AnyEvent::Handle: |
711 | - fixed a great number of bugs and corner cases in AnyEvent::Handle. |
| 412 | - fix a bug in where in SSL connect mode, the |
712 | - fix a bug in where in SSL connect mode, the |
| 413 | client would first wait for some data by the server and |
713 | client would first wait for some data by the server and |
| 414 | otherwise hang. (reported and analysed in an absolutely |
714 | otherwise hang. (reported and analysed in an absolutely |
| 415 | exemplary manner by Adam Rosenstein). |
715 | exemplary manner by Adam Rosenstein). |
| 416 | - fix a bug in where SSL EOF would not be treated |
716 | - fix a bug in where SSL EOF would not be treated |
