git-annex-gpl - git-annex without the AGPL

	Commit message (Collapse)	Author	Age
*	assistant: Fix syncing to just created ssh remotes.	Joey Hess	2012-12-06
\| \| \| \| \|	This got broken when I optimised reconnecting with remotes, to not do a full scan if the remote was not diverged.
*	webapp: Allow user to specify the ssh port when setting up a remote.	Joey Hess	2012-12-06
\|
*	--auto fixes	Joey Hess	2012-12-06
\| \| \| \| \| \| \|	* get/copy --auto: Transfer data even if it would exceed numcopies, when preferred content settings want it. * drop --auto: Fix dropping content when there are no preferred content settings.
*	assistant: Allow periods in ssh key comments.	Joey Hess	2012-12-06
\|
*	assistant: Avoid trying to drop content from remotes that don't have it.	Joey Hess	2012-12-05
\|
*	typo	Joey Hess	2012-12-05
\|
*	avoid crashing threads that drop content if the drop crashes	Joey Hess	2012-12-05
\|
*	remove dead code	Joey Hess	2012-12-01
\|
*	better function name	Joey Hess	2012-12-01
\|
*	remove hard link when sanity check failed	Joey Hess	2012-11-29
\| \| \| \|	See http://git-annex.branchable.com/forum/dot_git_slash_annex_slash_tmp/
*	assistant: Retrival from glacier now handled.	Joey Hess	2012-11-29
\|
*	typo	Joey Hess	2012-11-28
\|
*	Fix broken .config/git-annex/program installed by standalone tarball.	Joey Hess	2012-11-28
\| \| \| \| \| \| \| \|	It was doubly broken; both missing a slash, and containing "runshell git-annex", while some parts of the code expected it to be a simple path to a program. This appears to include the transfer queue runner, and the code that starts a new assistant process when switching to another repository in the webapp.
*	avoid using Data.AssocList, which is cunningly part of hxt	Joey Hess	2012-11-27
\|
*	The standalone builds now unset their special path and library path ↵	Joey Hess	2012-11-27
\| \| \| \| \| \|	variables before running the system web browser. Should fix a crash reported on OSX.
*	allow building webapp w/o webdav	Joey Hess	2012-11-25
\|
*	webapp and assistant glacier support	Joey Hess	2012-11-24
\|
*	queue uploads when a new or renamed symlink is handled	Joey Hess	2012-11-24
\|
*	webapp: support box.com	Joey Hess	2012-11-17
\|
*	update to dav 0.1, and basic uploading is working!	Joey Hess	2012-11-15
\|
*	factor out Creds	Joey Hess	2012-11-14
\|
*	use System.Timeout	Joey Hess	2012-11-14
\| \| \| \|	I'd forgotten this existed!
*	automatic repolist updating	Joey Hess	2012-11-13
\|
*	Fix build of assistant without yesod.	Joey Hess	2012-11-13
\|
*	upper-case first char of buddy name	Joey Hess	2012-11-12
\|
*	add a yes/no confirm for xmpp pairing requests	Joey Hess	2012-11-11
\|
*	auto-accept pair requests from JIDs already paired with	Joey Hess	2012-11-11
\|
*	don't try to transfer data to/from XMPP remotes	Joey Hess	2012-11-11
\| \| \| \| \|	Partition syncRemotes into ones needing git sync (ie, non-special remotes), and ones needing data sync (ie, non-XMPP remotes).
*	allow both one push and one receive-pack to run at the same time	Joey Hess	2012-11-11
\| \| \| \| \| \|	Noticed that when pairing, sometimes both sides start to push, and the other side sends a PushRequest, and the two deadlock, neither doing anything. (Timeout eventually breaks this.) So, let both run at the same time.
*	add 120 second timeout when waiting for XMPP packets	Joey Hess	2012-11-11
\|
*	use trustExclude	Joey Hess	2012-11-11
\|
*	reword alert	Joey Hess	2012-11-10
\|
*	fix creation of creds directory for xmpp	Joey Hess	2012-11-10
\|
*	show when a buddy is already paired	Joey Hess	2012-11-10
\|
*	set negative XMPP presence priority	Joey Hess	2012-11-10
\| \| \| \| \| \| \| \|	This should help prevent git-annex clients receiving messages that were intended for normal clients they're sharing the account with. Changed XMPP protocol use to always send chat messages directed at the specific client, as the negative priority blocks less directed messages.
*	avoid sending xmpp push notifications when not pushing to any remotes	Joey Hess	2012-11-10
\|
*	full-on git-annex assistant syncing now works over XMPP!	Joey Hess	2012-11-10
\| \| \| \| \| \| \| \| \| \|	I decided to use the fallback push mode from the beginning for XMPP, since while it uses some ugly branches, it avoids the possibility of a normal push failing, and needing to pull and re-push. Due to the overhead of XMPP, and the difficulty of building such a chain of actions due to the async implementation, this seemed reasonable. It seems to work great!
*	more refactoring	Joey Hess	2012-11-10
\|
*	more nice refactoring	Joey Hess	2012-11-10
\|
*	better data types allowed marvelous refactoring	Joey Hess	2012-11-10
\|
*	separate data type for push stages	Joey Hess	2012-11-10
\| \| \| \|	This improves type safety.
*	refactor	Joey Hess	2012-11-10
\|
*	use a lookup table for speed	Joey Hess	2012-11-10
\|
*	on second thought, do allow deferring StartingPush	Joey Hess	2012-11-10
\| \| \| \| \| \|	My reasoning is that StartingPush could be received after another push starts being received, and it would be better to respond to it afterwards than not.
*	refactor	Joey Hess	2012-11-10
\|
*	fix deferring of CanPush, and stop deferring StartingPush	Joey Hess	2012-11-10
\|
*	refactor	Joey Hess	2012-11-10
\|
*	refactor	Joey Hess	2012-11-10
\|
*	refactor	Joey Hess	2012-11-10
\|
*	fix exit code propigation	Joey Hess	2012-11-10
\|