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

Comparing AnyEvent/Changes (file contents):
Revision 1.214 by root, Wed Jan 21 05:36:45 2009 UTC vs.
Revision 1.442 by root, Wed Feb 2 13:00:32 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
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
8 - update idna algorithm(s) to UTS#46 version 6.0.0.
9 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
10 uts46 says it's an error, and testsuite makes more sense.
11 - remove leading dots in idn_to_ascii - testsuite says to remove,
12 uts46 says it's an error, and testsuite makes more sense.
13 - fixed a bug in idn ignored character processing.
14 - passes the uts46 6.0.0 testsuite except for three cases which
15 apparently are bugs in the testsuite itself.
16 - passes the uts46 6.0.1 testsuite except for eleven cases which
17 apparently are bugs in the testsuite itself.
18
195.31 Sun Jan 23 11:49:19 CET 2011
20 - work around a POE bug where POE would first poll for new events and
21 invoke handlers BEFORE handling the message queue.
22 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
23 - stop_read now does just that in TLS mode. see its
24 documentation for caveats.
25 - update to common::sense 3.4 warning set (some warnings in perl impose
26 an arbitrary 32kb limit on regex matches, which affects
27 AnyEvent::Handle).
28 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
29 - new experimental AE::Handle::resettls method.
30
315.3 Fri Dec 31 05:48:13 CET 2010
32 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
33 issues with IO::Async have been solved in that version, including
34 the availabilty of a default loop!
35 - implement new wbuf_max parameter for AnyEvent::Handle.
36 - added Cocoa::EventLoop backend by Daisuke Murase.
37 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
38 - clarify handle on_prepare documentation.
39 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
40 to avoid leaking them to any child processes.
41 - try to force files into the arch-dependend lib directory, for the
42 half-"turly obsessive compulsive person"(s) that depend on it.
43
445.29 Sun Dec 5 10:49:21 CET 2010
45 - convert EV backend to EV 4.00 API (so better upgrade EV too).
46 - AE::Handle: implement dynamic read_size adjustment, add
47 max_read_size parameter, reduce default read_size to 2048.
48 - add some FAQ entries for servers and/or on_eof vs. on_error.
49 - work around OS bugs (cygwin again) when lingering in AE::Handle.
50 - reduce memory usage slightly when connect was used in AE::Handle.
51 - wrok around more segfaults in Tk.
52 - document tls_ctx => undef as valid.
53 - detect WNOHANG value at compiletime.
54 - include some probably linux-only support for building AnyEvent
55 as part of the perl core.
56 - improve compatibility of t/08_idna.t to perl 5.8.
57 - make AnyEvent::Strict truly optional (in case somebody builds a
58 minimal perl...).
59
605.28 Wed Oct 13 04:14:23 CEST 2010
61 - due to a glitch, AnyEvent's internal getprotobyname wasn't
62 actually used in tcp_server.
63 - implement AnyEvent::CondVar->new method, which was documented but
64 missing.
65 - support autoloading of read types in unshift_read.
66 - AnyEvent::Strict: do the test for invalid fh arguments differently,
67 resulting in more informative error messages.
68 - new AnyEvent::FAQ manpage.
69 - clarify that you can actually call push_read etc. in on_read.
70 - put keys and certs in /tmp, not into the current directory,
71 and document this in the SECURITY CONSIDERATIONS sections
72 of AnyEvent::TLS.
73 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
74 requested by Ron Isaacson.
75 - document $AnyEvent::DNS::RESOLVER variable.
76 - incorporated major typo patches by Abhijit Menon-Sen.
77 - AnyEvent::Handle now croaks when negative timeouts are passed.
78 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
79 undocumented, but who knows when they might come in handy.
80
815.271 Tue Jun 8 12:05:46 CEST 2010
82 - backport to perl 5.8.x.
83
845.27 Sun Jun 6 12:12:05 CEST 2010
85 - postpone differently in AnyEvent::Socket now, as
86 when not, canceling the connection attempt might fail
87 (found by Felix Antonius Wilhelm Ostmann).
88 - explicitly check for non-stream sockets in AE::Handle, too many
89 clueless people fell into the trap of this somehow working.
90 - simplified and reworked the "OTHER MODULES" section.
91 - better/more condvar examples.
92
935.261 Wed Apr 28 16:13:36 CEST 2010
94 - AF_INET6 was not properly used from Socket6 during configuration
95 time (found by Andy Grundman).
96
975.26 Mon Apr 12 04:49:35 CEST 2010
98 - don't generate spurious readyness notifications when select
99 returns EINTR in the pure perl backend (analysed and testcase
100 by Bas Denissen).
101 - give same examples in SYNOPSIS sections of AnyEvent and AE.
102 - provider faster implementations for AE::time/now/now_update for
103 EV and Perl backends.
104 - provide faster AE::cv and AE::time implementations for all
105 backends.
106 - no longer support lower/mixed-case signal names (this was
107 never documented nor universally supported).
108 - some more memory µ-optimisations, and somewhat less messy
109 code to cope with both AnyEvent and AE APIs.
110
1115.251 Sat Mar 13 00:58:21 CET 2010
112 - make test failed if it was run before make install on perl <=
113 5.8.8, and older versions of perl make our live very difficult,
114 so write constants.pl during Makefile.PL time.
115
1165.25 Sat Mar 13 00:23:14 CET 2010
117 - fix a race condition in AnyEvent::Handle that would cause
118 a "bio_read: unsupported method" error (in _another_
119 ssl connection) after user code threw an exception.
120 - added AnyEvent::Handle->destroyed method.
121 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
122 - the AnyEvent::Util::fh_nonblocking function worked only
123 by lucky accident on win32.
124 - smaller and faster AnyEvent::Util::fh_nonblocking.
125 - when the (required!) Time::HiRes module is missing, AnyEvent
126 did not fall back to built-in time properly.
127 - do not load Fcntl at runtime, saving memory and loading time.
128 - precompile a number of constants and use them instead of runtime
129 detection and eval.
130 - free detection code after detection and similar memory optimisations.
131 - Perl backend timer interval best effort drift has been improved
132 (same algorithm as EV).
133 - update unicode idna mapping table.
134
1355.24 Tue Jan 5 11:39:43 CET 2010
136 - cygwin never reports errors from failed connects. 1.5 just gives
137 you continous readyness and EAGAIN, 1.7 is even more broken and
138 just hangs. work around both issues in a cygwin-specific hack.
139 - improve idle watcher documentation slightly.
140
1415.23 Sun Dec 20 23:48:00 CET 2009
142 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
143 - implement punycode_encode/decode, idn_nameprep,
144 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
145 - provide $AE::VERSION.
146 - removed traces of "no strict 'refs'".
147
1485.22 Sat Dec 5 03:51:13 CET 2009
149 - downgrade-or-fail in AnyEvent::Handle::push_write, to
150 diagnose encoding failures earlier and more succinctly.
151 (this works around bugs in perl, throwing away encoding info
152 when passing scalar data to extensions).
153 - add more examples to AnyEvent::Socket manpage.
154 - upgrade internal warning set to the same as common::sense 2.03.
155 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
156 (John Beppu).
157 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
158 - add google public dns servers to fallback server set and make sure
159 we load-balance properly between the three sets. also add all
160 fallback dns servers, not just a random one, to each dns config.
161
1625.21 Thu Nov 19 02:48:47 CET 2009
163 - fix a problem where socket constants were called with parameters
164 (spotted by David Friedland).
165 - fork_call never use'd POSIX (reported by Daisuke Maki).
166 - improve perl 5.6 compatibility further (but it still won't work
167 unless you rip out everything but the core).
168 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
169 uses 10MB of RAM, but doesn't open a console window. *sigh*.
170
1715.202 Wed Oct 14 22:35:44 CEST 2009
172 - AnyEvent::DNS would unexpectedly clobber $_ under windows
173 (analysed by Matthias Waldorf).
174 - AnyEvent::Handle::run_cmd can now pass the PID of the
175 newly-created process, which is much less useful than it might
176 sound (based on patch by Yann Kerherve).
177
1785.201 Tue Sep 29 12:09:25 CEST 2009
179 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
180 (reported by Torsten Foertsch).
181 - common::sense 2.0 could cause tcp_server to throw an exception
182 (analysed by elmex).
183
1845.2 Mon Sep 14 07:04:49 CEST 2009
185 - INCOMPATIBLE CHANGE: do no longer support register_read_type
186 and register_write_type in AnyEvent::Handle, instead support
187 package names (the facility was mostly abused).
188 - implement "packagename-as-read/write type" support in
189 AnyEvent::Handle.
190 - AnyEvent::Handle: new options "keepalive" and "oobinline".
191 - oobinline set by default to avoid security issues.
192 - the pure-perl event loop backend wrongly detected times() fallback
193 support (spotted by Pavel Boldin).
194
1955.12 Tue Sep 1 20:26:50 CEST 2009
196 - be more lenient when parsing resolv.conf files, as some people
197 use hashmarks as comment indicator inside directives (reported
198 by Michael S. Fischer).
199 - use same set of warnings as common::sense 2.0.
200 - fix a potential 32 bit overflow issue due to perl having problems
201 with large hex constants in the Perl backend.
202 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
203 by default.
204 - fix some prototypes.
205 - (optionally) require Async::Interrupt 1.02.
206 - replace opendns fallbacks by verizon/level3 public dns servers.
207
2085.112 Fri Aug 21 13:59:04 CEST 2009
209 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
210 using starttls with an illegal/uncreatable tls context.
211 - port to broken win32 perls.
212
2135.111 Sun Aug 16 18:44:36 CEST 2009
214 - perl errornously requires sockaddr_un structures returned by the
215 kernel to have a specific length. work around this bug by padding
216 sockaddr_un structures (found to fail on OS X, as perl only uses
217 the correct code for GNU/Linux, but likely to fail on other
218 OSes as well).
219
2205.11 Wed Aug 12 17:49:37 CEST 2009
221 - fixed a bug in run_cmd with input from filename.
222 - create a json coder object when none is given in write direction
223 to reduce dependency on JSON.
224
2255.1 Tue Aug 11 03:17:32 CEST 2009
226 - tcp_connect now reports the correct error in $!, not an unrelated
227 one (was broken since 4.91).
228 - AE::Handle did not properly clear rtimeout/wtimeout watchers
229 when retrying a tcp_connect.
230 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
231 - fixed wrong documentation in AE::Socket.
232
2335.01 Mon Aug 10 03:16:32 CEST 2009
234 - last release broke EV child watchers :/.
235
2365.0 Sun Aug 9 17:32:09 CEST 2009
237 - officially introduce and document the AE API.
238 - lowering the timeout at runtime did not have immediate effect
239 in AnyEvent::Handle.
240 - AE::Handle now has separate and independent read and write timeout
241 settings.
242 - probe for Guard module when AnyEvent::Util is loaded, not at
243 runtime (Event doesn't like eval "use" from callbacks, and it avoids
244 exporting the probe function).
245 - Impl::IOAsync's io watchers did not use the filehandle emulation
246 code to work around it's watcher limits.
247 - work around IO::Async's problems with overloaded objects.
248 - try to disable wearnings inside IO::Async, it's just too buggy.
249 - do no longer use Event by default, only EV or Perl, Event is too buggy.
250 - hackishly provide fast AE:: interface when EV or Perl is the backend.
251 - inherit VERSION from AnyEvent::VERSION in most modules.
252
2534.91 Thu Aug 6 15:42:45 CEST 2009
254 - AE::Handle::starttls could get out of sync when the read buffer
255 already contains some TLS handshake.
256 - AE::Handle did not properly free the TLS session, causing bigger
257 memory leaks in Net::SSLeay than expected (reported by toaster).
258 - AE::Socket::tcp_connect will now artificially delay invoking
259 the callback to avoid returning after invoking the callback.
260 - convert many internal watcher uses to AE API.
261
2624.9 Sat Aug 1 11:07:01 CEST 2009
263 - Glib silently fails when registering a timeout with
264 a negative value, so avoid that.
265 - call condvar callback immediately when it is set
266 after the condition is already signalled.
267 - check rbuf_max condition only after trying to consume
268 data.
269 - normalise signal numbers to names when using the ae's
270 signal handling, but do not document this (yet).
271 - pure perl signal emulation did not properly set nonblocking
272 mode on the signal pipe on win32 (this is unlikely to have caused
273 any issues).
274 - new module: AnyEvent::Debug.
275 - AnyEvent::Strict now checks that a signal was specified
276 by name (not name or number).
277 - reduce memory footprint in typical cases by ~50kb by
278 only compiling signal code when necessary.
279 - add AnyEvent::Handle->rbuf_max.
280 - grab the AE namespace for future new API, implement stubs for
281 said future/alternative API.
282 - new function: AnyEvent::Socket::format_hostport.
283
2844.881 Tue Jul 28 12:51:53 CEST 2009
285 - work around a bug in local in pre-5.10 perls, causing
286 AnyEvent::Handle to recurse when it should not
287 (analysed by elmex).
288
2894.88 Tue Jul 28 04:04:37 CEST 2009
290 - re-bless the handle into a dummy package after calling
291 AnyEvent::Handle::destroy, so the user does not need to check
292 for errors after every push_write etc.
293 - do not attempt to run t/02_signals.t on obviously broken
294 platforms.
295
2964.87 Sun Jul 26 02:06:16 CEST 2009
297 - do not attempt to linger when there is no longer a valid fh.
298 - enforce tls mode to be either accept or connect, do not simply
299 segfault in Net::SSLeay.
300 - AnyEvent::Handle can now call tcp_connect itself (new parameters
301 connect, on_prepare, on_connect and on_connect_error). Updated
302 tutorial accordingly.
303 - add AnyEvent::Impl::Irssi backend.
304
3054.86 Mon Jul 20 23:52:29 CEST 2009
306 - since the verbose warning is not enough, explicitly document
307 that versions before 1.33 of Net::SSLeay are not secure.
308 - work around signal handling races in Event and (...) Event::Lib.
309 - try to align signal-race timer to full-second boundaries.
310 - work around Tk not liking negative timeouts.
311 - don't complain of different grades of environmental unfriendlyness
312 in IO::Async.
313
3144.85 Sat Jul 18 06:16:14 CEST 2009
315 - nail the signal race problem in perl once and for all
316 (see $AnyEvent::MAX_SIGNAL_LATENCY).
317 - take advantage of Async::Interrupt if it is available.
318 - load Time::HiRes and Guard modules on demand only.
319 - add optional/recommended modules section to AnyEvent
320 documentation.
321 - reduce memory usage considerably (and reduce startup penalty)
322 by not using "strict", "warnings" and "overload" modules.
323 - work around buggy windows/openbsd perls and provide EBADMSG
324 and EPROTO ourselves when missing.
325 - improve perl 5.6 compatibility of the core event loop.
326 - made Net::SSLeay version 1.33 a soft requirement.
327
3284.83 Fri Jul 17 16:56:26 CEST 2009
329 - implement AnyEvent::Socket::getprotobyname.
330 - AnyEvent::CondVar's will now detect recursive blocking
331 waits and will croak, as too many people fall into
332 this trap.
333 - AnyEvent::Handle will now call ->destroy on itself after
334 executing the on_error callback, instead of doing some
335 half-baked internal shutdown, for fatal errors.
336 - clarify on_eof behaviour w.r.t. the read queue and
337 on_read callbacks.
338 - ignore some possible spurious wake-ups in tcp_connect.
339
3404.82 Sat Jul 11 00:34:55 CEST 2009
341 - POE and Event backends didn't accept some callable objects as
342 callbacks.
343 - use Config module instead of POSIX module to detect signal names
344 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
345 module doesn't even have all POSIX signals :/.
346 - use more workarounds around the many refcnt/corruption bugs in
347 Event::Lib.
348 - work around a race condition in perl's select, causing t/03_child.t
349 to rarely fail.
350
3514.81 Thu Jul 9 10:30:30 CEST 2009
352 - AnyEvent::Handle didn't properly diagnose write errors
353 (it expected -1 from syswrite, how lame... :).
354 - support file descriptors in addition to file handles
355 in AnyEvent->io.
356 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
357 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
358 and PERL_ANYEVENT_CA_PATH.
359 - provide a sensible synopsis section for AnyEvent::TLS.
360 - add a "supported backends" section to the manpage.
361 - added simple io watcher test to testsuite, using a
362 portable_socketpair.
363 - tried to improve the stability of the Event::Lib backend,
364 YMMV.
365
3664.8 Mon Jul 6 23:45:16 CEST 2009
367 - AnyEvent::DNS did not properly follow CNAME records with
368 uppercase targets.
369 - AnyEvent::DNS would errornously return AAAA records
370 with v4 mapped addresses (a faulty record) as ipv4 addresses,
371 causing AnyEvent::Socket to throw an exception.
372 - added new module AnyEvent::TLS for easier SSL/TLS context
373 creation, with many options including hostname verification,
374 secure default configuration, lots of documentation and,
375 predefined diffie-hellman keys for perfect forward security
376 and much more. get it while it's still fresh!
377 - use AnyEvent::TLS in AnyEvent::Handle for context management.
378 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
379 so AnyEvent::Socket users have smaller memory footprint
380 in the common case.
381 - add AnyEvent::Handle->push_shutdown method.
382 - add an additional error message parameter to AnyEvent::Handle's
383 on_error callback (for TLS, $! is still available).
384 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
385 - make AnyEvent::Handle more robust against early conenction
386 failures (during new), and return C<undef> in that case
387 from the constructor.
388 - AnyEvent::Handle will now try to load only JSON::XS first,
389 then fall back to JSON.
390 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
391 and because it was documented that way.
392
3934.451 Fri Jul 3 00:28:58 CEST 2009
394 - do not clear rbuf when shutting down an AnyEvent::Handle
395 object - doing so breaks AnyEvent::HTTP.
396
3974.45 Mon Jun 29 22:59:26 CEST 2009
398 - a write error could cause AnyEvent::Handle to create
399 an I/O watcher with an undefined $fh.
400 - special-case mapped ipv4 addresses in both
401 AnyEvent::Socket::format_address and parse_address,
402 to treat them just like ipv4 addresses.
403 - updated and overhauled the AnyEvent::Intro doc.
404 - implement AnyEvent::Socket::format_ipv4/ipv6.
405 - slightly speed up portable_pipe/socketpair.
406 - expand condvar begin/end documentation.
407
4084.42 Fri Jun 26 08:32:18 CEST 2009
409 - preliminary and neccesarily incomplete support for IO::Async.
410 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
411 case it was set to IGNORE, to ensure we can catch child statuses
412 even when the calling env acts stupidly.
413 - updated benchmarks with IO::Async, which performs very well.
414
4154.412 Wed Jun 24 01:35:57 CEST 2009
416 - support an "untaint" attribute for AnyEvent::DNS
417 and set it on the default resolver.
418 - implement and document AnyEvent::Impl::Perl::loop.
419 - remove all anyevent-env variables from %ENV
420 when running in tainted mode.
421 - mention and extend the IO::Lambda benchmark.
422
4234.411 Sun Jun 7 18:48:02 CEST 2009
424 - do not try to use F_SETFD on windows.
425
4264.41 Thu May 14 06:40:11 CEST 2009
427 - work around issues in older perls (5.8.0?) when
428 a signal handler is deleted from the %SIG hash.
429 - use POSIX::_exit in child test, to avoid running
430 destructors.
431 - speed up CHLD handling by relying on SIGCHLD being
432 synchronously delivered, even when we roll our own
433 implementation.
434 - AnyEvent::DNS: add the "dname" resource record name
435 for cosmetic reasons.
436
4374.4 Sun Apr 26 20:12:33 CEST 2009
438 - implemented idle watchers, where applicable.
439 - AnyEvent->time died when Event backend was in use.
440 - fix a memleak in the Tk backend.
441 - sped up Tk timer handling.
442 - clip negative "after" values to 0 in AnyEvent::Impl::Event
443 to avoid spamming out warning messages.
444 - fix Qt timers without interval.
445 - avoid Qt zero-timeout special case, in old libqt's.
446
4474.352 Mon Apr 20 16:31:11 CEST 2009
448 - fix AnyEvent::Strict error messages for child watchers.
449 - fix/update Makefile.PL "recommends" versions.
450 - add AnyEvent->now_update.
451
4524.351 Sat Apr 11 07:56:14 CEST 2009
453 - actually make the signal pipe work under win32.
454 - localise $! in signal handler to avoid changing $!, although
455 perl itself does not seem to save/restore errno either.
456 - set the cloexec flag on the signal pipe (normally set by perl too).
457
4584.35 Fri Mar 27 11:48:20 CET 2009
459 - event models relying on AnyEvent's signal watcher emulation did
460 invoke the callback asynchronously, contradicting documentation
461 and causing signals to get lost (this includes AnyEvent's own
462 event loop). AnyEvent now uses the standard pipe trick to make
463 callback execution synchronous to the event loop.
464 - AnyEvent::Handle didn't free TLS context data on DESTROY
465 (patch by Pavel Shaydo).
466 - work around the ever-incompatibly-changing API of MakeMaker.
467 - document that changing global variables without restoring them
468 is a bad idea in Perl (noted by Adam Rosenstein).
469 - AnyEvent::Strict now barfs if ->io is passed a file.
470
4714.34 Thu Feb 12 18:32:45 CET 2009
472 - separately buffer TLS read data, as otherwise the read queue could
473 deadlock as receiving data is not expected while draining the
474 read queue (which cna only happen with TLS).
3 - raise EBADMSG error on JSON decoding errors. 475 - raise EBADMSG error on JSON decoding errors.
476 - fix some minor manpage bugs (reported by Maximilian Gaß).
477 - speed up select bitmask parsing quite a bit in the pure perl
478 backend.
479 - use CORE::select instead of just select in AnyEvent::Impl::Perl.
4 480
54.331 Tue Jan 6 21:07:25 CET 2009 4814.331 Tue Jan 6 21:07:25 CET 2009
6 - socketpair fails on many vista machines because vista has 482 - socketpair fails on many vista machines because vista has
7 completely broken accept/getpeername and getsockname functions, 483 completely broken accept/getpeername and getsockname functions,
8 so we provide our own socketpair emulation that kind of works 484 so we provide our own socketpair emulation that kind of works
… …
384.3 Fri Oct 3 09:18:43 CEST 2008 5144.3 Fri Oct 3 09:18:43 CEST 2008
39 - AnyEvent will now install a no-op signal handler for SIGPIPE, 515 - AnyEvent will now install a no-op signal handler for SIGPIPE,
40 unless one has been installed already. 516 unless one has been installed already.
41 - warn about not loaidng AnyEvent::Impl::POE early enough 517 - warn about not loaidng AnyEvent::Impl::POE early enough
42 (patch by Adam Rosenstein). 518 (patch by Adam Rosenstein).
43 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 519 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
44 - fix a bug in where in SSL connect mode, the 520 - fix a bug in where in SSL connect mode, the
45 client would first wait for some data by the server and 521 client would first wait for some data by the server and
46 otherwise hang. (reported and analyzed in an absolutely 522 otherwise hang. (reported and analysed in an absolutely
47 exemplary manner by Adam Rosenstein). 523 exemplary manner by Adam Rosenstein).
48 - fix a bug in where SSL EOF would not be treated 524 - fix a bug in where SSL EOF would not be treated
49 as stream EOF, putting the connection into a hung state. 525 as stream EOF, putting the connection into a hung state.
50 (reported and analyzed in an absolutely exemplary manner by Adam 526 (reported and analysed in an absolutely exemplary manner by Adam
51 Rosenstein). 527 Rosenstein).
52 - fix a potential segfault when the TLS context 528 - fix a potential segfault when the TLS context
53 would go missing in a read callback (e.g. due to stoptls) - 529 would go missing in a read callback (e.g. due to stoptls) -
54 Net::SSLeay of course makes no type checking whatsoever. 530 Net::SSLeay of course makes no type checking whatsoever.
55 - AnyEvent::Handle will not stop the read watcher in TLS mode. 531 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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