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

Comparing AnyEvent/Changes (file contents):
Revision 1.250 by root, Sun Jun 28 11:43:51 2009 UTC vs.
Revision 1.464 by root, Sat Aug 13 16:11:33 2011 UTC

1Revision history for Perl extension AnyEvent. 1Revision history for Perl extension AnyEvent.
2 2
3TODO: parse ipv6 mapped addresses transparently 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: IO::Async sometimes cancels the wrong timer when you call cancel_timer -wait for IO::Async release.
9TODO: wrap
10TODO: parse_unix???
11TODO: logging?
12 - INCOMPATIBLE CHANGE: AnyEvent::DNS resource records now include
13 the TTL as fourth element - this affects ->resolve directly,
14 and ->request and ->dns_unpack indirectly. Convenience functions
15 such as AnyEvent::DNS::{a,txt,mx,srv} are not affected, neither
16 is AnyEvent::Socket.
17 - INCOMPATIBLE CHANGE: backend authors now should not implement
18 one_event or loop, but instead the AnyEvent::CondVar::_wait and _poll
19 methods.
20 - DEPRECIATION: please change
21 "use AnyEvent::Impl::Perl" to "use AnyEvent::Loop" and
22 "AnyEvent::Impl::Perl::loop" to "AnyEvent::Loop::run"
23 in your programs, for future compatibility.
24 - *blush* idle watcher emulation was borked.
25 - AnyEvent::Strict now checks the AE::xxx functions as well.
26 - querying too long domain names would barf deep inside AnyEvent::DNS - since
27 this is hard to avoid for applications, AnyEvent:.DNS::request now checks for
28 for too long domain names explicitly and passes an undef to the callback
29 instead (reported by James Bromberger).
30 - make AnyEvent::Util more robust against EINTR (analyzed by Andrew Suffield).
31 - new function: AnyEvent::postpone.
32 - new env variable, PERL_ANYEVENT_DEBUG_SHELL, to automatically start
33 a debug shell.
34 - new env variable, PERL_ANYEVENT_DEBUG_WRAP, to automatically enable
35 watcher instrumentation/wrapping (see AnyEvent::Debug::wrap).
36 - IO::Async sometimes cancels the wrong timer when you call cancel_timer.
37 Invest extra overhead to work around this peculiar behaviour.
38 - split the pure perl backend into a loop module and an impl
39 module (AnyEvent::Loop and AnyEvent::Impl::Perl).
40 - some read types (regex, netstring, json, storable) could
41 errornously keep an AE::Handle object alive on their own
42 (reported by Mohammad Toossi).
43 - add more workarounds around child watcher bugs in IO::Async
44 and (newly introduced) in POE.
45 - use glib child watchers and try to work around its limitations,
46 if possible (based on a patch by Kevin Ryde).
47 - add support for FLTK (fltk2), via AnyEvent::Impl::FLTK2.
48 - many read types in AnyEvent::Handle would malfunction during
49 transitions to TLS because of a "delete $self->{rbuf}".
50 (reported by Gerald Galster).
51 - resolve AnyEvent::Util::guard at first-use-time, not at load time,
52 for further speed and memory savings when not used.
53 - improve condvar blocking wait performance (probably only noticable
54 with faster backends), also make it easier for event loops that
55 do not support blocking waits.
56 - PERL_ANYEVENT_MODEL now also allows full module names, although
57 the usefulness of this feature is not yet clear.
58 - add tests for all supported event loops, conditional on PERL_ANYEVENT_LOOP_TESTS.
59 - add t/80_ssltext.t.
60 - no longer use AUTOLOAD for the initial loading (for no very good reason).
61
625.34 Thu May 12 09:59:41 CEST 2011
63 - AE::TLS didn't support both cert_file and key at the same time,
64 due to some copy&paste bug (reported and analyzed by vti).
65 - improve AE::Handle timeout documentation slightly.
66
675.33 Mon Feb 7 21:51:41 CET 2011
68 - never appeared on CPAN, for unknown reasons.
69 - AnyEvent::Handle on_connect_error - do not try to destroy a handle
70 object when it's already gone (analyzed by Ryan Bullock).
71 - update idna algorithm(s) to UTS#46 version 6.0.0.
72 - preserve trailing dot in idn_to_ascii - testsuite says to preserve,
73 uts46 says it's an error, and testsuite makes more sense.
74 - remove leading dots in idn_to_ascii - testsuite says to remove,
75 uts46 says it's an error, and testsuite makes more sense.
76 - fixed a bug in idn ignored character processing.
77 - passes the uts46 6.0.0 testsuite except for three cases which
78 apparently are bugs in the testsuite itself.
79 - passes the uts46 6.0.1 testsuite except for eleven cases which
80 apparently are bugs in the testsuite itself.
81
825.31 Sun Jan 23 11:49:19 CET 2011
83 - work around a POE bug where POE would first poll for new events and
84 invoke handlers BEFORE handling the message queue.
85 - fix parsing of "domain" statement in resolv.conf (Johannes Plunien).
86 - stop_read now does just that in TLS mode. see its
87 documentation for caveats.
88 - update to common::sense 3.4 warning set (some warnings in perl impose
89 an arbitrary 32kb limit on regex matches, which affects
90 AnyEvent::Handle).
91 - AE::Handle had a broken rbuf_max method and missing wbuf_max method.
92 - new experimental AE::Handle::resettls method.
93
945.3 Fri Dec 31 05:48:13 CET 2010
95 - major IOAsync patch by Paul Evans - 0.33 is required, as the major
96 issues with IO::Async have been solved in that version, including
97 the availabilty of a default loop!
98 - implement new wbuf_max parameter for AnyEvent::Handle.
99 - added Cocoa::EventLoop backend by Daisuke Murase.
100 - the default testsuite now honors PERL_ANYEVENT_MODEL (Paul Evans).
101 - clarify handle on_prepare documentation.
102 - set CLOEXEC flag on parent-side fd's in AnyEvent::Util::run_cmd,
103 to avoid leaking them to any child processes.
104 - try to force files into the arch-dependend lib directory, for the
105 half-"turly obsessive compulsive person"(s) that depend on it.
106
1075.29 Sun Dec 5 10:49:21 CET 2010
108 - convert EV backend to EV 4.00 API (so better upgrade EV too).
109 - AE::Handle: implement dynamic read_size adjustment, add
110 max_read_size parameter, reduce default read_size to 2048.
111 - add some FAQ entries for servers and/or on_eof vs. on_error.
112 - work around OS bugs (cygwin again) when lingering in AE::Handle.
113 - reduce memory usage slightly when connect was used in AE::Handle.
114 - wrok around more segfaults in Tk.
115 - document tls_ctx => undef as valid.
116 - detect WNOHANG value at compiletime.
117 - include some probably linux-only support for building AnyEvent
118 as part of the perl core.
119 - improve compatibility of t/08_idna.t to perl 5.8.
120 - make AnyEvent::Strict truly optional (in case somebody builds a
121 minimal perl...).
122
1235.28 Wed Oct 13 04:14:23 CEST 2010
124 - due to a glitch, AnyEvent's internal getprotobyname wasn't
125 actually used in tcp_server.
126 - implement AnyEvent::CondVar->new method, which was documented but
127 missing.
128 - support autoloading of read types in unshift_read.
129 - AnyEvent::Strict: do the test for invalid fh arguments differently,
130 resulting in more informative error messages.
131 - new AnyEvent::FAQ manpage.
132 - clarify that you can actually call push_read etc. in on_read.
133 - put keys and certs in /tmp, not into the current directory,
134 and document this in the SECURITY CONSIDERATIONS sections
135 of AnyEvent::TLS.
136 - support ";" as resolv.conf comment character in AnyEvent::DNS, as
137 requested by Ron Isaacson.
138 - document $AnyEvent::DNS::RESOLVER variable.
139 - incorporated major typo patches by Abhijit Menon-Sen.
140 - AnyEvent::Handle now croaks when negative timeouts are passed.
141 - add a shitload of TCP_xxx sockopt name constants to AnyEvent::Util,
142 undocumented, but who knows when they might come in handy.
143
1445.271 Tue Jun 8 12:05:46 CEST 2010
145 - backport to perl 5.8.x.
146
1475.27 Sun Jun 6 12:12:05 CEST 2010
148 - postpone differently in AnyEvent::Socket now, as
149 when not, canceling the connection attempt might fail
150 (found by Felix Antonius Wilhelm Ostmann).
151 - explicitly check for non-stream sockets in AE::Handle, too many
152 clueless people fell into the trap of this somehow working.
153 - simplified and reworked the "OTHER MODULES" section.
154 - better/more condvar examples.
155
1565.261 Wed Apr 28 16:13:36 CEST 2010
157 - AF_INET6 was not properly used from Socket6 during configuration
158 time (found by Andy Grundman).
159
1605.26 Mon Apr 12 04:49:35 CEST 2010
161 - don't generate spurious readyness notifications when select
162 returns EINTR in the pure perl backend (analysed and testcase
163 by Bas Denissen).
164 - give same examples in SYNOPSIS sections of AnyEvent and AE.
165 - provider faster implementations for AE::time/now/now_update for
166 EV and Perl backends.
167 - provide faster AE::cv and AE::time implementations for all
168 backends.
169 - no longer support lower/mixed-case signal names (this was
170 never documented nor universally supported).
171 - some more memory ยต-optimisations, and somewhat less messy
172 code to cope with both AnyEvent and AE APIs.
173
1745.251 Sat Mar 13 00:58:21 CET 2010
175 - make test failed if it was run before make install on perl <=
176 5.8.8, and older versions of perl make our live very difficult,
177 so write constants.pl during Makefile.PL time.
178
1795.25 Sat Mar 13 00:23:14 CET 2010
180 - fix a race condition in AnyEvent::Handle that would cause
181 a "bio_read: unsupported method" error (in _another_
182 ssl connection) after user code threw an exception.
183 - added AnyEvent::Handle->destroyed method.
184 - speed up AnyEvent::Socket::format_address and ::format_ipv6.
185 - the AnyEvent::Util::fh_nonblocking function worked only
186 by lucky accident on win32.
187 - smaller and faster AnyEvent::Util::fh_nonblocking.
188 - when the (required!) Time::HiRes module is missing, AnyEvent
189 did not fall back to built-in time properly.
190 - do not load Fcntl at runtime, saving memory and loading time.
191 - precompile a number of constants and use them instead of runtime
192 detection and eval.
193 - free detection code after detection and similar memory optimisations.
194 - Perl backend timer interval best effort drift has been improved
195 (same algorithm as EV).
196 - update unicode idna mapping table.
197
1985.24 Tue Jan 5 11:39:43 CET 2010
199 - cygwin never reports errors from failed connects. 1.5 just gives
200 you continous readyness and EAGAIN, 1.7 is even more broken and
201 just hangs. work around both issues in a cygwin-specific hack.
202 - improve idle watcher documentation slightly.
203
2045.23 Sun Dec 20 23:48:00 CET 2009
205 - support IDNs in resolve_sockaddr, and therefore in tcp_connect.
206 - implement punycode_encode/decode, idn_nameprep,
207 idn_to_ascii and idn_to_unicode operations in AnyEvent::Util.
208 - provide $AE::VERSION.
209 - removed traces of "no strict 'refs'".
210
2115.22 Sat Dec 5 03:51:13 CET 2009
212 - downgrade-or-fail in AnyEvent::Handle::push_write, to
213 diagnose encoding failures earlier and more succinctly.
214 (this works around bugs in perl, throwing away encoding info
215 when passing scalar data to extensions).
216 - add more examples to AnyEvent::Socket manpage.
217 - upgrade internal warning set to the same as common::sense 2.03.
218 - use pack "n/a*" for pre-5.8.9 perl compatibility in AnyEvent::DNS
219 (John Beppu).
220 - AnyEvent::Socket::inet_aton now properly supports ipv6, as documented.
221 - add google public dns servers to fallback server set and make sure
222 we load-balance properly between the three sets. also add all
223 fallback dns servers, not just a random one, to each dns config.
224
2255.21 Thu Nov 19 02:48:47 CET 2009
226 - fix a problem where socket constants were called with parameters
227 (spotted by David Friedland).
228 - fork_call never use'd POSIX (reported by Daisuke Maki).
229 - improve perl 5.6 compatibility further (but it still won't work
230 unless you rip out everything but the core).
231 - prefer Net::DNS::Resolver over ipconfig, if installed, on win32.
232 uses 10MB of RAM, but doesn't open a console window. *sigh*.
233
2345.202 Wed Oct 14 22:35:44 CEST 2009
235 - AnyEvent::DNS would unexpectedly clobber $_ under windows
236 (analysed by Matthias Waldorf).
237 - AnyEvent::Handle::run_cmd can now pass the PID of the
238 newly-created process, which is much less useful than it might
239 sound (based on patch by Yann Kerherve).
240
2415.201 Tue Sep 29 12:09:25 CEST 2009
242 - AnyEvent:Handle::on_starttls/on_stoptls methods were broken
243 (reported by Torsten Foertsch).
244 - common::sense 2.0 could cause tcp_server to throw an exception
245 (analysed by elmex).
246
2475.2 Mon Sep 14 07:04:49 CEST 2009
248 - INCOMPATIBLE CHANGE: do no longer support register_read_type
249 and register_write_type in AnyEvent::Handle, instead support
250 package names (the facility was mostly abused).
251 - implement "packagename-as-read/write type" support in
252 AnyEvent::Handle.
253 - AnyEvent::Handle: new options "keepalive" and "oobinline".
254 - oobinline set by default to avoid security issues.
255 - the pure-perl event loop backend wrongly detected times() fallback
256 support (spotted by Pavel Boldin).
257
2585.12 Tue Sep 1 20:26:50 CEST 2009
259 - be more lenient when parsing resolv.conf files, as some people
260 use hashmarks as comment indicator inside directives (reported
261 by Michael S. Fischer).
262 - use same set of warnings as common::sense 2.0.
263 - fix a potential 32 bit overflow issue due to perl having problems
264 with large hex constants in the Perl backend.
265 - do not use the slower 5.6 version fo the enc_name code in AnyEvent::DNS
266 by default.
267 - fix some prototypes.
268 - (optionally) require Async::Interrupt 1.02.
269 - replace opendns fallbacks by verizon/level3 public dns servers.
270
2715.112 Fri Aug 21 13:59:04 CEST 2009
272 - AnyEvent::Handle could cause hard crashes in Net::SSLeay when
273 using starttls with an illegal/uncreatable tls context.
274 - port to broken win32 perls.
275
2765.111 Sun Aug 16 18:44:36 CEST 2009
277 - perl errornously requires sockaddr_un structures returned by the
278 kernel to have a specific length. work around this bug by padding
279 sockaddr_un structures (found to fail on OS X, as perl only uses
280 the correct code for GNU/Linux, but likely to fail on other
281 OSes as well).
282
2835.11 Wed Aug 12 17:49:37 CEST 2009
284 - fixed a bug in run_cmd with input from filename.
285 - create a json coder object when none is given in write direction
286 to reduce dependency on JSON.
287
2885.1 Tue Aug 11 03:17:32 CEST 2009
289 - tcp_connect now reports the correct error in $!, not an unrelated
290 one (was broken since 4.91).
291 - AE::Handle did not properly clear rtimeout/wtimeout watchers
292 when retrying a tcp_connect.
293 - new functions: AnyEvent::Util::run_cmd and close_all_fds_except.
294 - fixed wrong documentation in AE::Socket.
295
2965.01 Mon Aug 10 03:16:32 CEST 2009
297 - last release broke EV child watchers :/.
298
2995.0 Sun Aug 9 17:32:09 CEST 2009
300 - officially introduce and document the AE API.
301 - lowering the timeout at runtime did not have immediate effect
302 in AnyEvent::Handle.
303 - AE::Handle now has separate and independent read and write timeout
304 settings.
305 - probe for Guard module when AnyEvent::Util is loaded, not at
306 runtime (Event doesn't like eval "use" from callbacks, and it avoids
307 exporting the probe function).
308 - Impl::IOAsync's io watchers did not use the filehandle emulation
309 code to work around it's watcher limits.
310 - work around IO::Async's problems with overloaded objects.
311 - try to disable wearnings inside IO::Async, it's just too buggy.
312 - do no longer use Event by default, only EV or Perl, Event is too buggy.
313 - hackishly provide fast AE:: interface when EV or Perl is the backend.
314 - inherit VERSION from AnyEvent::VERSION in most modules.
315
3164.91 Thu Aug 6 15:42:45 CEST 2009
317 - AE::Handle::starttls could get out of sync when the read buffer
318 already contains some TLS handshake.
319 - AE::Handle did not properly free the TLS session, causing bigger
320 memory leaks in Net::SSLeay than expected (reported by toaster).
321 - AE::Socket::tcp_connect will now artificially delay invoking
322 the callback to avoid returning after invoking the callback.
323 - convert many internal watcher uses to AE API.
324
3254.9 Sat Aug 1 11:07:01 CEST 2009
326 - Glib silently fails when registering a timeout with
327 a negative value, so avoid that.
328 - call condvar callback immediately when it is set
329 after the condition is already signalled.
330 - check rbuf_max condition only after trying to consume
331 data.
332 - normalise signal numbers to names when using the ae's
333 signal handling, but do not document this (yet).
334 - pure perl signal emulation did not properly set nonblocking
335 mode on the signal pipe on win32 (this is unlikely to have caused
336 any issues).
337 - new module: AnyEvent::Debug.
338 - AnyEvent::Strict now checks that a signal was specified
339 by name (not name or number).
340 - reduce memory footprint in typical cases by ~50kb by
341 only compiling signal code when necessary.
342 - add AnyEvent::Handle->rbuf_max.
343 - grab the AE namespace for future new API, implement stubs for
344 said future/alternative API.
345 - new function: AnyEvent::Socket::format_hostport.
346
3474.881 Tue Jul 28 12:51:53 CEST 2009
348 - work around a bug in local in pre-5.10 perls, causing
349 AnyEvent::Handle to recurse when it should not
350 (analysed by elmex).
351
3524.88 Tue Jul 28 04:04:37 CEST 2009
353 - re-bless the handle into a dummy package after calling
354 AnyEvent::Handle::destroy, so the user does not need to check
355 for errors after every push_write etc.
356 - do not attempt to run t/02_signals.t on obviously broken
357 platforms.
358
3594.87 Sun Jul 26 02:06:16 CEST 2009
360 - do not attempt to linger when there is no longer a valid fh.
361 - enforce tls mode to be either accept or connect, do not simply
362 segfault in Net::SSLeay.
363 - AnyEvent::Handle can now call tcp_connect itself (new parameters
364 connect, on_prepare, on_connect and on_connect_error). Updated
365 tutorial accordingly.
366 - add AnyEvent::Impl::Irssi backend.
367
3684.86 Mon Jul 20 23:52:29 CEST 2009
369 - since the verbose warning is not enough, explicitly document
370 that versions before 1.33 of Net::SSLeay are not secure.
371 - work around signal handling races in Event and (...) Event::Lib.
372 - try to align signal-race timer to full-second boundaries.
373 - work around Tk not liking negative timeouts.
374 - don't complain of different grades of environmental unfriendlyness
375 in IO::Async.
376
3774.85 Sat Jul 18 06:16:14 CEST 2009
378 - nail the signal race problem in perl once and for all
379 (see $AnyEvent::MAX_SIGNAL_LATENCY).
380 - take advantage of Async::Interrupt if it is available.
381 - load Time::HiRes and Guard modules on demand only.
382 - add optional/recommended modules section to AnyEvent
383 documentation.
384 - reduce memory usage considerably (and reduce startup penalty)
385 by not using "strict", "warnings" and "overload" modules.
386 - work around buggy windows/openbsd perls and provide EBADMSG
387 and EPROTO ourselves when missing.
388 - improve perl 5.6 compatibility of the core event loop.
389 - made Net::SSLeay version 1.33 a soft requirement.
390
3914.83 Fri Jul 17 16:56:26 CEST 2009
392 - implement AnyEvent::Socket::getprotobyname.
393 - AnyEvent::CondVar's will now detect recursive blocking
394 waits and will croak, as too many people fall into
395 this trap.
396 - AnyEvent::Handle will now call ->destroy on itself after
397 executing the on_error callback, instead of doing some
398 half-baked internal shutdown, for fatal errors.
399 - clarify on_eof behaviour w.r.t. the read queue and
400 on_read callbacks.
401 - ignore some possible spurious wake-ups in tcp_connect.
402
4034.82 Sat Jul 11 00:34:55 CEST 2009
404 - POE and Event backends didn't accept some callable objects as
405 callbacks.
406 - use Config module instead of POSIX module to detect signal names
407 in AnyEvent::Strict and AnyEvent::Impl::EventLib, as the POSIX
408 module doesn't even have all POSIX signals :/.
409 - use more workarounds around the many refcnt/corruption bugs in
410 Event::Lib.
411 - work around a race condition in perl's select, causing t/03_child.t
412 to rarely fail.
413
4144.81 Thu Jul 9 10:30:30 CEST 2009
415 - AnyEvent::Handle didn't properly diagnose write errors
416 (it expected -1 from syswrite, how lame... :).
417 - support file descriptors in addition to file handles
418 in AnyEvent->io.
419 - new env variables: PERL_ANYEVENT_RESOLV_CONF,
420 PERL_ANYEVENT_MAX_OUTSTANDING_DNS, PERL_ANYEVENT_CA_FILE
421 and PERL_ANYEVENT_CA_PATH.
422 - provide a sensible synopsis section for AnyEvent::TLS.
423 - add a "supported backends" section to the manpage.
424 - added simple io watcher test to testsuite, using a
425 portable_socketpair.
426 - tried to improve the stability of the Event::Lib backend,
427 YMMV.
428
4294.8 Mon Jul 6 23:45:16 CEST 2009
430 - AnyEvent::DNS did not properly follow CNAME records with
431 uppercase targets.
432 - AnyEvent::DNS would errornously return AAAA records
433 with v4 mapped addresses (a faulty record) as ipv4 addresses,
434 causing AnyEvent::Socket to throw an exception.
435 - added new module AnyEvent::TLS for easier SSL/TLS context
436 creation, with many options including hostname verification,
437 secure default configuration, lots of documentation and,
438 predefined diffie-hellman keys for perfect forward security
439 and much more. get it while it's still fresh!
440 - use AnyEvent::TLS in AnyEvent::Handle for context management.
441 - load AnyEvent::Handle only on demand in AnyEvent::DNS,
442 so AnyEvent::Socket users have smaller memory footprint
443 in the common case.
444 - add AnyEvent::Handle->push_shutdown method.
445 - add an additional error message parameter to AnyEvent::Handle's
446 on_error callback (for TLS, $! is still available).
447 - add AnyEvent::Handle on_starttls/on_stoptls callbacks.
448 - make AnyEvent::Handle more robust against early conenction
449 failures (during new), and return C<undef> in that case
450 from the constructor.
451 - AnyEvent::Handle will now try to load only JSON::XS first,
452 then fall back to JSON.
453 - format_ipv4/format_ipv6 are now exported by default, for symmetry,
454 and because it was documented that way.
455
4564.451 Fri Jul 3 00:28:58 CEST 2009
457 - do not clear rbuf when shutting down an AnyEvent::Handle
458 object - doing so breaks AnyEvent::HTTP.
459
4604.45 Mon Jun 29 22:59:26 CEST 2009
461 - a write error could cause AnyEvent::Handle to create
462 an I/O watcher with an undefined $fh.
463 - special-case mapped ipv4 addresses in both
464 AnyEvent::Socket::format_address and parse_address,
465 to treat them just like ipv4 addresses.
466 - updated and overhauled the AnyEvent::Intro doc.
467 - implement AnyEvent::Socket::format_ipv4/ipv6.
4 - slightly speed up portable_pipe/socketpair. 468 - slightly speed up portable_pipe/socketpair.
469 - expand condvar begin/end documentation.
5 470
64.42 Fri Jun 26 08:32:18 CEST 2009 4714.42 Fri Jun 26 08:32:18 CEST 2009
7 - preliminary and neccesarily incomplete support for IO::Async. 472 - preliminary and neccesarily incomplete support for IO::Async.
8 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in 473 - reset SIGCHLD to DEFAULT when AnyEvent is loaded, in
9 case it was set to IGNORE, to ensure we can catch child statuses 474 case it was set to IGNORE, to ensure we can catch child statuses
… …
1124.3 Fri Oct 3 09:18:43 CEST 2008 5774.3 Fri Oct 3 09:18:43 CEST 2008
113 - AnyEvent will now install a no-op signal handler for SIGPIPE, 578 - AnyEvent will now install a no-op signal handler for SIGPIPE,
114 unless one has been installed already. 579 unless one has been installed already.
115 - warn about not loaidng AnyEvent::Impl::POE early enough 580 - warn about not loaidng AnyEvent::Impl::POE early enough
116 (patch by Adam Rosenstein). 581 (patch by Adam Rosenstein).
117 - fixed a great number of bugs and corner cases in AnyEvent::Handle: 582 - fixed a great number of bugs and corner cases in AnyEvent::Handle.
118 - fix a bug in where in SSL connect mode, the 583 - fix a bug in where in SSL connect mode, the
119 client would first wait for some data by the server and 584 client would first wait for some data by the server and
120 otherwise hang. (reported and analyzed in an absolutely 585 otherwise hang. (reported and analysed in an absolutely
121 exemplary manner by Adam Rosenstein). 586 exemplary manner by Adam Rosenstein).
122 - fix a bug in where SSL EOF would not be treated 587 - fix a bug in where SSL EOF would not be treated
123 as stream EOF, putting the connection into a hung state. 588 as stream EOF, putting the connection into a hung state.
124 (reported and analyzed in an absolutely exemplary manner by Adam 589 (reported and analysed in an absolutely exemplary manner by Adam
125 Rosenstein). 590 Rosenstein).
126 - fix a potential segfault when the TLS context 591 - fix a potential segfault when the TLS context
127 would go missing in a read callback (e.g. due to stoptls) - 592 would go missing in a read callback (e.g. due to stoptls) -
128 Net::SSLeay of course makes no type checking whatsoever. 593 Net::SSLeay of course makes no type checking whatsoever.
129 - AnyEvent::Handle will not stop the read watcher in TLS mode. 594 - AnyEvent::Handle will not stop the read watcher in TLS mode.

Diff Legend

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