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

Comparing AnyEvent/Changes (file contents):
Revision 1.307 by root, Wed Jul 29 12:39:21 2009 UTC vs.
Revision 1.542 by root, Thu Apr 5 04:09:49 2012 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: release a:i 1.0 3TODO: document TCP_*** constants
4TODO: ae::strict should chekc for signal names (enforced) 4TODO: add debug/trace logging to some modules
5TODO: use the eval-with-cb-call trick also to speed up JSON?
6lib/AnyEvent/DNS.pm: open my $fh, "<", $resolv_conf
7lib/AnyEvent/DNS.pm: if (open my $fh, "ipconfig /all |") {
8lib/AnyEvent/TLS.pm: or croak "$dh_file: failed to open DH parameter file: $!";
9
106.2
11 - child watchers are broken in POE 1.352 (also many earlier
12 versions)and there seems to be no way to work around it, as POE
13 itself is inherently racy. Document this regression and sleep a
14 second in t/68_poe_03_child.t for the time being.
15 - new module AnyEvent::IO, that is a frontend to either a pure-perl
16 synchronous I/O implementation (AnyEvent::IO::Perl), or to the
17 asynchronous I/O extension IO::AIO (AnyEvent::IO::IOAIO).
18 - load /etc/hosts only when dns fails has no answer.
19 - stat /etc/hosts on every access and reload it if it changed.
20 - load /etc/hosts via AnyEvent::IO - potentially asynchronous.
21 - AnyEvent::Log log_to_file and log_to_path now use AnyEvent::IO. As
22 a side effect, they now use true appending as opposed to libc
23 appending, and the file might not have been opened when the function
24 returns.
25 - the default logging level was not properly documented in a variety of
26 places, this has hopefully been rectified.
27 - updated uts46data.pl for unicode 6.1.0.
28 - made log messages genearted by AnyEvent submodules not
29 include the package name anymore, as it will be logged
30 by default already.
31 - upgrade to the trick used by common-sense 3.5 to work
32 around extra warning torture/breakage under perl 5.15.x.
33 - log messages by anyevent are now ucfirst, are usually full
34 sentences and do no longer include the package name.
35 - the storable read type would sometimes throw an exception instead
36 of causing EBADMSG (analyzed by Maxime Soulé).
37 - documented Rocco fabricating statements about AnyEvent and me.
38
396.14 Tue Jan 31 20:00:24 CET 2012
40 - AnyEvent::Impl::Tk was broken due to a mysterious "die" inside,
41 probably an editing mistake (reported by Darin McBride).
42
436.13 Thu Jan 12 07:27:01 CET 2012
44 - AnyEvent::Util::fork_call checked for POSIX availability in the wrong way,
45 causing it to fail if POSIX wasn't loaded already (analysed by Rock Power).
46 - AnyEvent::Handle::push_read (line => did pass $1 directly, so
47 regex-matching inside the callback would change the parameter, despite
48 $1 being dynamically scoped per-block. this perl bug is now being
49 worked around (testcase by Cindy Wang).
50
516.12 Mon Dec 12 13:21:10 CET 2011
52 - $! was clobbered by subsequent calls in tcp_connect, due to
53 the postpone not saving/restoring it (analyzed by Richard Garnish).
54
556.11 Tue Nov 22 10:36:05 CET 2011
56 - Tk cannot create windows when tainted unless you set a title, so
57 set a dummy title for AnyEvent's dummy mainwindow. How dummy.
58 - escape any nonprintable/nonascii characters when stringifying
59 backtraces.
60 - log the reason loading an interface module fails to load at level debug.
61
626.1 Tue Oct 4 19:44:30 CEST 2011
63 - INCOMPATIBLE CHANGE: the default log level is now "4" (error
64 and above), and some messages inside AnyEvent have been elevated
65 to higher log levels to print by default.
66 - AnyEvent::log, unlike AnyEvent::Log::log, did not always exit on fatal.
67 - work around yet another signal race bug, newly introduced by POE. My
68 how that module sucks.
69 - add google public ipv6 dns as fallback servers, so AE::DNS can finally
70 support ipv6-only operation even with defaults (also use ::1 instead
71 of 127.0.0.1).
72 - give better guidelines on how to chose a logging level.
73 - AnyEvent::Log can now cap (limit) log message priority by context.
74 - AE::log will now save and restore $! and $@, so callers don't have to.
75 - AE::Strict now checks that registered fds are still valid at regular
76 intervals.
77 - skip frequently called 'require' calls if the module appears loaded.
78 - increase signal latency test timer considerably for poor little osx,
79 which can't be bothered with basic posix functionality such as signal
80 delivery.
81 - mention some debugging aids in the tutorial.
82 - new env variable PERL_ANYEVENT_MAX_SIGNAL_LATENCY.
83 - private (%namespace) log contexts in PERL_ANYEVENT_LOG now have the name
84 of the context as title.
85 - AnyEvent::Strict would not restore read-write access to $_ when a callback
86 died - this is a user bug, but since AE::Strict exists to help dbeugging...
87 - detect the extremely evil IO::Async::Loop::AnyEvent module and refuse
88 to work with it - the author is well aware of the issues but refuses
89 to fix them. This extreme step has been taken because the module actively
90 attacks the goals of AnyEvent and makes it harder for module authors to use
91 AnyEvent.
92
936.02 Fri Aug 26 20:08:31 CEST 2011
94 - log_to_syslog now accepts facility strings.
95 - log_to_syslog would not allow facilities - "$facility|$level"
96 *literally* means that string. sheesh.
97 - the CPAN parser got confused and thought we provide Sys::Syslog, try
98 to work around.
99 - renamed AnyEvent::Impl::FLTK2 to FLTK.
100
1016.01 Fri Aug 26 07:04:11 CEST 2011
102 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
103 the TTL as fourth element - this affects ->resolve, ->request
104 and ->dns_unpack indirectly. Convenience functions such as
105 AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither is
106 AnyEvent::Socket.
107 - INCOMPATIBLE CHANGE: backend authors now should not implement
108 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
109 methods.
110 - INCOMPATIBLE CHANGE: verbosity levels have been redefined, see
111 AnyEvent (PERL_ANYEVENT_VERBOSE) and AnyEvent::Log for details.
112 - DEPRECIATION: please change
113 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
114 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
115 in your programs, for future compatibility.
116 - *blush* idle watcher emulation was borked.
117 - AnyEvent::Strict now checks the AE::xxx functions as well.
118 - querying too long domain names would barf deep inside AnyEvent::DNS - since
119 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
120 for too long domain names explicitly and passes an undef to the callback
121 instead (reported by James Bromberger).
122 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
123 - new functions: AnyEvent::postpone, AnyEvent::log.
124 - new module: AnyEvent::Log, for simple logging needs.
125 - new env variable, PERL_ANYEVENT_LOG, for fine-grained logging config.
126 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
127 a debug shell.
128 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
129 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
130 - new env variable, PERL_ANYEVENT_HOSTS, to override the place where
131 /etc/hosts can be found.
132 - an empty PERL_ANYEVENT_RESOLV_CONF now selects the default
133 configuration.
134 - env variables can now be specified using AE_xyz instead of
135 PERL_ANYEVENT_xyz, subject to some rules.
136 - lots of new AnyEvent::Debug functionality.
137 - AnyEvent::Strict now checks for modifications of $_ by the callback
138 (by setting it to readonly while executing the callback, if possible).
139 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
140 Invest extra overhead to work around this peculiar behaviour.
141 - split the pure perl backend into a loop module and an impl
142 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
143 - some read types (regex, netstring, json, storable) could
144 errornously keep an AE::Handle object alive on their own
145 (reported by Mohammad Toossi).
146 - AnyEvent::Socket::resolve_sockaddr and all functions using it now
147 supports /etc/hosts overrides.
148 - add more workarounds around child watcher bugs in IO::Async
149 and (newly introduced) in POE.
150 - use glib child watchers and try to work around its limitations,
151 if possible (based on a patch by Kevin Ryde).
152 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
153 - many read types in AnyEvent::Handle would malfunction during
154 transitions to TLS because of a "delete $self->{rbuf}".
155 (reported by Gerald Galster).
156 - improve condvar blocking wait performance (probably only noticable
157 with faster backends), also make it easier for event loops that
158 do not support blocking waits.
159 - PERL_ANYEVENT_MODEL now also allows full module names, although
160 the usefulness of this feature is not yet clear.
161 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
162 - add t/80_ssltext.t.
163 - no longer use AUTOLOAD for the initial loading (for no very good reason).
164 - tcp_server now tries to unlink unix domain sockets when it gets destroyed.
165 - do not run event loop detection multiple times when programs call their cached
166 copies of AnyEvent::detect.
167 - parse_hostport now accepts absolute paths als unix domain sockets.
168 - suppress the idiotic warning inside POE by patching the POE::Kernel
169 run flags instead of asking the user to jump through contortions.
170 - add Task::Weaken as dependency (as recommends only at this time).
171 - replace PF_UNSPEC by 0, for good old minix.
172
1736.0 (not officially released, but was on CPAN for about 15 minutes :/)
174
1755.34 Thu May 12 09:59:41 CEST 2011
176 - AE::TLS didn't support both cert_file and key at the same time,
177 due to some copy&paste bug (reported and analyzed by vti).
178 - improve AE::Handle timeout documentation slightly.
179
1805.33 Mon Feb 7 21:51:41 CET 2011
181 - never appeared on CPAN, for unknown reasons.
182 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
183 object when it's already gone (analyzed by Ryan Bullock).
184 - update idna algorithm(s) to UTS#46 version 6.0.0.
185 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
186 uts46 says it's an error, and testsuite makes more sense.
187 - remove leading dots in idn_to_ascii - testsuite says to remove,
188 uts46 says it's an error, and testsuite makes more sense.
189 - fixed a bug in idn ignored character processing.
190 - passes the uts46 6.0.0 testsuite except for three cases which
191 apparently are bugs in the testsuite itself.
192 - passes the uts46 6.0.1 testsuite except for eleven cases which
193 apparently are bugs in the testsuite itself.
194
1955.31 Sun Jan 23 11:49:19 CET 2011
196 - work around a POE bug where POE would first poll for new events and
197 invoke handlers BEFORE handling the message queue.
198 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
199 - stop_read now does just that in TLS mode. see its
200 documentation for caveats.
201 - update to common::sense 3.4 warning set (some warnings in perl impose
202 an arbitrary 32kb limit on regex matches, which affects
203 AnyEvent::Handle).
204 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
205 - new experimental AE::Handle::resettls method.
206
2075.3 Fri Dec 31 05:48:13 CET 2010
208 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
209 issues with IO::Async have been solved in that version, including
210 the availabilty of a default loop!
211 - implement new wbuf_max parameter for AnyEvent::Handle.
212 - added Cocoa::EventLoop backend by Daisuke Murase.
213 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
214 - clarify handle on_prepare documentation.
215 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
216 to avoid leaking them to any child processes.
217 - try to force files into the arch-dependend lib directory, for the
218 half-"turly obsessive compulsive person"(s) that depend on it.
219
2205.29 Sun Dec 5 10:49:21 CET 2010
221 - convert EV backend to EV 4.00 API (so better upgrade EV too).
222 - AE::Handle: implement dynamic read_size adjustment, add
223 max_read_size parameter, reduce default read_size to 2048.
224 - add some FAQ entries for servers and/or on_eof vs. on_error.
225 - work around OS bugs (cygwin again) when lingering in AE::Handle.
226 - reduce memory usage slightly when connect was used in AE::Handle.
227 - wrok around more segfaults in Tk.
228 - document tls_ctx => undef as valid.
229 - detect WNOHANG value at compiletime.
230 - include some probably linux-only support for building AnyEvent
231 as part of the perl core.
232 - improve compatibility of t/08_idna.t to perl 5.8.
233 - make AnyEvent::Strict truly optional (in case somebody builds a
234 minimal perl...).
235
2365.28 Wed Oct 13 04:14:23 CEST 2010
237 - due to a glitch, AnyEvent's internal getprotobyname wasn't
238 actually used in tcp_server.
239 - implement AnyEvent::CondVar->new method, which was documented but
240 missing.
241 - support autoloading of read types in unshift_read.
242 - AnyEvent::Strict: do the test for invalid fh arguments differently,
243 resulting in more informative error messages.
244 - new AnyEvent::FAQ manpage.
245 - clarify that you can actually call push_read etc. in on_read.
246 - put keys and certs in /tmp, not into the current directory,
247 and document this in the SECURITY CONSIDERATIONS sections
248 of AnyEvent::TLS.
249 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
250 requested by Ron Isaacson.
251 - document $AnyEvent::DNS::RESOLVER variable.
252 - incorporated major typo patches by Abhijit Menon-Sen.
253 - AnyEvent::Handle now croaks when negative timeouts are passed.
254 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
255 undocumented, but who knows when they might come in handy.
256
2575.271 Tue Jun 8 12:05:46 CEST 2010
258 - backport to perl 5.8.x.
259
2605.27 Sun Jun 6 12:12:05 CEST 2010
261 - postpone differently in AnyEvent::Socket now, as
262 when not, canceling the connection attempt might fail
263 (found by Felix Antonius Wilhelm Ostmann).
264 - explicitly check for non-stream sockets in AE::Handle, too many
265 clueless people fell into the trap of this somehow working.
266 - simplified and reworked the "OTHER MODULES" section.
267 - better/more condvar examples.
268
2695.261 Wed Apr 28 16:13:36 CEST 2010
270 - AF_INET6 was not properly used from Socket6 during configuration
271 time (found by Andy Grundman).
272
2735.26 Mon Apr 12 04:49:35 CEST 2010
274 - don't generate spurious readyness notifications when select
275 returns EINTR in the pure perl backend (analysed and testcase
276 by Bas Denissen).
277 - give same examples in SYNOPSIS sections of AnyEvent and AE.
278 - provider faster implementations for AE::time/now/now_update for
279 EV and Perl backends.
280 - provide faster AE::cv and AE::time implementations for all
281 backends.
282 - no longer support lower/mixed-case signal names (this was
283 never documented nor universally supported).
284 - some more memory µ-optimisations, and somewhat less messy
285 code to cope with both AnyEvent and AE APIs.
286
2875.251 Sat Mar 13 00:58:21 CET 2010
288 - make test failed if it was run before make install on perl <=
289 5.8.8, and older versions of perl make our live very difficult,
290 so write constants.pl during Makefile.PL time.
291
2925.25 Sat Mar 13 00:23:14 CET 2010
293 - fix a race condition in AnyEvent::Handle that would cause
294 a "bio_read: unsupported method" error (in _another_
295 ssl connection) after user code threw an exception.
296 - added AnyEvent::Handle->destroyed method.
297 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
298 - the AnyEvent::Util::fh_nonblocking function worked only
299 by lucky accident on win32.
300 - smaller and faster AnyEvent::Util::fh_nonblocking.
301 - when the (required!) Time::HiRes module is missing, AnyEvent
302 did not fall back to built-in time properly.
303 - do not load Fcntl at runtime, saving memory and loading time.
304 - precompile a number of constants and use them instead of runtime
305 detection and eval.
306 - free detection code after detection and similar memory optimisations.
307 - Perl backend timer interval best effort drift has been improved
308 (same algorithm as EV).
309 - update unicode idna mapping table.
310
3115.24 Tue Jan 5 11:39:43 CET 2010
312 - cygwin never reports errors from failed connects. 1.5 just gives
313 you continous readyness and EAGAIN, 1.7 is even more broken and
314 just hangs. work around both issues in a cygwin-specific hack.
315 - improve idle watcher documentation slightly.
316
3175.23 Sun Dec 20 23:48:00 CET 2009
318 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
319 - implement punycode_encode/decode, idn_nameprep,
320 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
321 - provide $AE::VERSION.
322 - removed traces of "no strict 'refs'".
323
3245.22 Sat Dec 5 03:51:13 CET 2009
325 - downgrade-or-fail in AnyEvent::Handle::push_write, to
326 diagnose encoding failures earlier and more succinctly.
327 (this works around bugs in perl, throwing away encoding info
328 when passing scalar data to extensions).
329 - add more examples to AnyEvent::Socket manpage.
330 - upgrade internal warning set to the same as common::sense 2.03.
331 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
332 (John Beppu).
333 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
334 - add google public dns servers to fallback server set and make sure
335 we load-balance properly between the three sets. also add all
336 fallback dns servers, not just a random one, to each dns config.
337
3385.21 Thu Nov 19 02:48:47 CET 2009
339 - fix a problem where socket constants were called with parameters
340 (spotted by David Friedland).
341 - fork_call never use'd POSIX (reported by Daisuke Maki).
342 - improve perl 5.6 compatibility further (but it still won't work
343 unless you rip out everything but the core).
344 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
345 uses 10MB of RAM, but doesn't open a console window. *sigh*.
346
3475.202 Wed Oct 14 22:35:44 CEST 2009
348 - AnyEvent::DNS would unexpectedly clobber $_ under windows
349 (analysed by Matthias Waldorf).
350 - AnyEvent::Handle::run_cmd can now pass the PID of the
351 newly-created process, which is much less useful than it might
352 sound (based on patch by Yann Kerherve).
353
3545.201 Tue Sep 29 12:09:25 CEST 2009
355 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
356 (reported by Torsten Foertsch).
357 - common::sense 2.0 could cause tcp_server to throw an exception
358 (analysed by elmex).
359
3605.2 Mon Sep 14 07:04:49 CEST 2009
361 - INCOMPATIBLE CHANGE: do no longer support register_read_type
362 and register_write_type in AnyEvent::Handle, instead support
363 package names (the facility was mostly abused).
364 - implement "packagename-as-read/write type" support in
365 AnyEvent::Handle.
366 - AnyEvent::Handle: new options "keepalive" and "oobinline".
367 - oobinline set by default to avoid security issues.
368 - the pure-perl event loop backend wrongly detected times() fallback
369 support (spotted by Pavel Boldin).
370
3715.12 Tue Sep 1 20:26:50 CEST 2009
372 - be more lenient when parsing resolv.conf files, as some people
373 use hashmarks as comment indicator inside directives (reported
374 by Michael S. Fischer).
375 - use same set of warnings as common::sense 2.0.
376 - fix a potential 32 bit overflow issue due to perl having problems
377 with large hex constants in the Perl backend.
378 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
379 by default.
380 - fix some prototypes.
381 - (optionally) require Async::Interrupt 1.02.
382 - replace opendns fallbacks by verizon/level3 public dns servers.
383
3845.112 Fri Aug 21 13:59:04 CEST 2009
385 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
386 using starttls with an illegal/uncreatable tls context.
387 - port to broken win32 perls.
388
3895.111 Sun Aug 16 18:44:36 CEST 2009
390 - perl errornously requires sockaddr_un structures returned by the
391 kernel to have a specific length. work around this bug by padding
392 sockaddr_un structures (found to fail on OS X, as perl only uses
393 the correct code for GNU/Linux, but likely to fail on other
394 OSes as well).
395
3965.11 Wed Aug 12 17:49:37 CEST 2009
397 - fixed a bug in run_cmd with input from filename.
398 - create a json coder object when none is given in write direction
399 to reduce dependency on JSON.
400
4015.1 Tue Aug 11 03:17:32 CEST 2009
402 - tcp_connect now reports the correct error in $!, not an unrelated
403 one (was broken since 4.91).
404 - AE::Handle did not properly clear rtimeout/wtimeout watchers
405 when retrying a tcp_connect.
406 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
407 - fixed wrong documentation in AE::Socket.
408
4095.01 Mon Aug 10 03:16:32 CEST 2009
410 - last release broke EV child watchers :/.
411
4125.0 Sun Aug 9 17:32:09 CEST 2009
413 - officially introduce and document the AE API.
414 - lowering the timeout at runtime did not have immediate effect
415 in AnyEvent::Handle.
416 - AE::Handle now has separate and independent read and write timeout
417 settings.
418 - probe for Guard module when AnyEvent::Util is loaded, not at
419 runtime (Event doesn't like eval "use" from callbacks, and it avoids
420 exporting the probe function).
421 - Impl::IOAsync's io watchers did not use the filehandle emulation
422 code to work around it's watcher limits.
423 - work around IO::Async's problems with overloaded objects.
424 - try to disable wearnings inside IO::Async, it's just too buggy.
425 - do no longer use Event by default, only EV or Perl, Event is too buggy.
426 - hackishly provide fast AE:: interface when EV or Perl is the backend.
427 - inherit VERSION from AnyEvent::VERSION in most modules.
428
4294.91 Thu Aug 6 15:42:45 CEST 2009
430 - AE::Handle::starttls could get out of sync when the read buffer
431 already contains some TLS handshake.
432 - AE::Handle did not properly free the TLS session, causing bigger
433 memory leaks in Net::SSLeay than expected (reported by toaster).
434 - AE::Socket::tcp_connect will now artificially delay invoking
435 the callback to avoid returning after invoking the callback.
436 - convert many internal watcher uses to AE API.
437
4384.9 Sat Aug 1 11:07:01 CEST 2009
439 - Glib silently fails when registering a timeout with
440 a negative value, so avoid that.
441 - call condvar callback immediately when it is set
442 after the condition is already signalled.
443 - check rbuf_max condition only after trying to consume
444 data.
5 - normalise signal numbers to names when using the ae's 445 - normalise signal numbers to names when using the ae's
6 signal handling, but do not document this (yet). 446 signal handling, but do not document this (yet).
447 - pure perl signal emulation did not properly set nonblocking
448 mode on the signal pipe on win32 (this is unlikely to have caused
449 any issues).
450 - new module: AnyEvent::Debug.
451 - AnyEvent::Strict now checks that a signal was specified
452 by name (not name or number).
453 - reduce memory footprint in typical cases by ~50kb by
454 only compiling signal code when necessary.
455 - add AnyEvent::Handle->rbuf_max.
456 - grab the AE namespace for future new API, implement stubs for
457 said future/alternative API.
458 - new function: AnyEvent::Socket::format_hostport.
7 459
84.881 Tue Jul 28 12:51:53 CEST 2009 4604.881 Tue Jul 28 12:51:53 CEST 2009
9 - work around a bug in local in pre-5.10 perls, causing 461 - work around a bug in local in pre-5.10 perls, causing
10 AnyEvent::Handle to recurse when it should not 462 AnyEvent::Handle to recurse when it should not
11 (analyzed by elmex). 463 (analysed by elmex).
12 464
134.88 Tue Jul 28 04:04:37 CEST 2009 4654.88 Tue Jul 28 04:04:37 CEST 2009
14 - re-bless the handle into a dummy package after calling 466 - re-bless the handle into a dummy package after calling
15 AnyEvent::Handle::destroy, so the user does not need to check 467 AnyEvent::Handle::destroy, so the user does not need to check
16 for errors after every push_write etc. 468 for errors after every push_write etc.
… …
2384.3 Fri Oct 3 09:18:43 CEST 2008 6904.3 Fri Oct 3 09:18:43 CEST 2008
239 - AnyEvent will now install a no-op signal handler for SIGPIPE, 691 - AnyEvent will now install a no-op signal handler for SIGPIPE,
240 unless one has been installed already. 692 unless one has been installed already.
241 - warn about not loaidng AnyEvent::Impl::POE early enough 693 - warn about not loaidng AnyEvent::Impl::POE early enough
242 (patch by Adam Rosenstein). 694 (patch by Adam Rosenstein).
243 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 695 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
244 - fix a bug in where in SSL connect mode, the 696 - fix a bug in where in SSL connect mode, the
245 client would first wait for some data by the server and 697 client would first wait for some data by the server and
246 otherwise hang. (reported and analyzed in an absolutely 698 otherwise hang. (reported and analysed in an absolutely
247 exemplary manner by Adam Rosenstein). 699 exemplary manner by Adam Rosenstein).
248 - fix a bug in where SSL EOF would not be treated 700 - fix a bug in where SSL EOF would not be treated
249 as stream EOF, putting the connection into a hung state. 701 as stream EOF, putting the connection into a hung state.
250 (reported and analyzed in an absolutely exemplary manner by Adam 702 (reported and analysed in an absolutely exemplary manner by Adam
251 Rosenstein). 703 Rosenstein).
252 - fix a potential segfault when the TLS context 704 - fix a potential segfault when the TLS context
253 would go missing in a read callback (e.g. due to stoptls) - 705 would go missing in a read callback (e.g. due to stoptls) -
254 Net::SSLeay of course makes no type checking whatsoever. 706 Net::SSLeay of course makes no type checking whatsoever.
255 - AnyEvent::Handle will not stop the read watcher in TLS mode. 707 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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