Commit graph

22 commits

Author SHA1 Message Date
Joey Hess
f78ca9bc58 split out daemonstatus types 2012-10-30 14:11:14 -04:00
Joey Hess
87ba4f8677 split and lift Assistant.Pushes 2012-10-29 17:52:43 -04:00
Joey Hess
94ae5d14e5 more lifting 2012-10-29 16:28:45 -04:00
Joey Hess
76768ad977 converted 6 more threads 2012-10-29 11:40:22 -04:00
Joey Hess
4dbdc2b666 Assistant monad, stage 2.5
Converted several threads to run in the monad.

Added a lot of useful combinators for working with the monad.

Now the monad includes the name of the thread.

Some debugging messages are disabled pending converting other threads.
2012-10-29 02:21:04 -04:00
Joey Hess
64e1d7b579 send git-annex tag in initial presence
Will be used for finding other git-annex clients for pairing
2012-10-28 17:07:29 -04:00
Joey Hess
2dc40ecbd1 reconnect XMPP when NetWatcher notices a change 2012-10-27 00:42:14 -04:00
Joey Hess
9fc8257392 xmpp reconnection
If it managed to run for 5 minutes, reconnect immediately. Otherwise,
wait 5 minutes before reconnecting.
2012-10-27 00:06:17 -04:00
Joey Hess
0c0ef99341 gnutls segfault fixed
Adjust build deps to ensure that only a fixed version of the library will
be used.

Also, removed the bound thread stuff, which I now think was (probably)
a red herring.
2012-10-26 23:07:30 -04:00
Joey Hess
1d04dff4f0 split out xmpp utilities 2012-10-26 14:44:36 -04:00
Joey Hess
82b7e835e4 run xmpp test in bound thread 2012-10-26 14:39:22 -04:00
Joey Hess
07494cbb4b XMPP configuration form
Currently relies on SRV being set, or the JID's hostname being the server
hostname and the port being default. Future work: Allow manual
configuration of user name, hostname, and port.
2012-10-26 14:17:09 -04:00
Joey Hess
a11fb94c65 SRV record construction 2012-10-26 13:03:08 -04:00
Joey Hess
8916ff1a6c hook up SRV lookups for XMPP 2012-10-26 12:55:29 -04:00
Joey Hess
fe96b28b4d convert the assistant to use a bound thread for XMPP
This *may* solve the segfault I was seeing when the XMPP library called
startTLS. My hypothesis is as follows:

* TLS is documented
  (http://www.gnu.org/software/gnutls/manual/gnutls.html#Thread-safety)
  thread safe, but only when a single thread accesses it.
* forkIO threads are not bound to an OS thread, so it was possible for
  the threaded runtime to run part of the XMPP code on one thread, and
  then switch to another thread later.

So, forkOS, with its bound threads, should be used for the XMPP thread.
Since the crash doesn't happen reliably, I am not yet sure about this fix.

Note that I kept all the other threads in the assistant unbound, because
bound threads have significantly higher overhead.
2012-10-25 15:47:27 -04:00
Joey Hess
8c9c8e3110 only use one push attribute
When pushing to multiple UUIDs, combine them all into a single push
attribute.
2012-10-25 13:31:45 -04:00
Joey Hess
91c0c7b9ef switch from presence toggle hack to git-annex tag in presence extended content
Push notifications are actually working over XMPP now!
2012-10-25 13:04:43 -04:00
Joey Hess
52d380d7f0 add a separate field for the JID 2012-10-24 19:09:02 -04:00
Joey Hess
145202f21e flip availablility
Seems presence notifications are not sent to clients that have marked
themselves unavailable. (Testing with google talk.)

This is the death knell for the presence hack, because it has to stay
available, and even the toggle to unavailable and back could cause it to
miss a notification. Still, flipped it so it basically works, for now.
2012-10-24 17:23:21 -04:00
Joey Hess
422b426460 pull from one of the remotes in a push notification
Still need to do something about transfer queueing, however. This could be
a real can of worms.
2012-10-24 16:21:45 -04:00
Joey Hess
32497feb2a initial implementation of XMPP push notifier (untested)
Lacking error handling, reconnection, credentials configuration,
and doesn't actually do anything when it receives an incoming notification.

Other than that, it might work! :)
2012-10-24 15:42:02 -04:00
Joey Hess
ae8a3ab31e added push notifier thread, currently a no-op
Hooked up everything that needs to notify on pushes. Note that
syncNewRemote does not notify. This is probably ok, and I'd need to thread
more state through to make it do so.

This is only set up to support a single push notification method; I didn't
use a NotificationBroadcaster. Partly because I don't yet know what info
about pushes needs to be communicated, so my data types are only
preliminary.
2012-10-24 13:38:28 -04:00