Re: [PATCH] Allow local branching to set up rebase by default.

	Hi, thanks for the response.

On May 7, 2008, at 9:22, Junio C Hamano wrote:

You got no reactions, neither
positive nor negative, partly due to this sentence, I suspect.

	I got one response from someone who said he had implemented the same  
thing in one of his branches.  That seemed to confirm the idea for  
me.  Some additional features were suggested to be built on top of  
this, but I didn't read it as necessary for proceeding.

Please sign-off your patches (see Documentation/SubmittingPatches).

	Oh thanks, I hadn't seen this.  I read it as well as the  
CodingGuidelines document.

quoted

+	When `never`, rebase is never automatically set to true.
+	When `local`, rebase is set to true for tracked branches of
+	other local branches.
+	When `remote`, rebase is set to true for tracked branches of
+	remote branches.
+	When `always`, rebase will be set to true for all tracking
+	branches.
+	This option defaults to never.

How does this interact with a similarly named configuration option,
autosetupmerge, whose settings can be false, true, and always?

	I can direct the user to the autosetupmerge config documentation  
since it also documents the flags that override its behavior.  Is this  
acceptable?

quoted

branch.<name>.remote::
	When in branch <name>, it tells `git fetch` which remote to fetch.
	If this option is not given, `git fetch` defaults to remote  
"origin".

This is not your fault, but can anybody guess what the exact command
sequence to cause the named branch to be advanced by rebasing, only by
reading this entry?

   branch.<name>.rebase::
           When true, rebase the branch <name> on top of the fetched  
branch,
           instead of merging the default branch from the default  
remote.

I can't.  Perhaps we would need "when "git pull" is run" or  
something at
the end.

	I'll submit this as a separate patch.

Is this correct?

	[...]

You might want to check how the setup_tracking() decides when to say
remote/local in its printf().

	Oh thanks, I clearly didn't understand this and just relied on my  
tests passing by coincidence.

	I'll resubmit my patch with your recommendations, thanks.

-- 
Dustin Sallings