Re: base64 patch fix

From: H. Peter Anvin <hpa@zytor.com>
Date: 2005-04-13 18:03:21
Ingo Molnar wrote:
> 
> but applying it would break the self-hosting property of this project.  
> Which matters, even if it's just 1 week old :-)
> 
> A solution to the migration problem would be to put conversion code into 
> fsck (if it encounters an old hash, it renames it to the new encoding), 
> and to commit that code first, in the old repository format. Then Linus 
> would keep the old hash-files in his repository for a migration period, 
> so that people having the pre-conversion tree can still sync up to the 
> point where the fsck conversion feature was integrated. Then the 
> migration would be painless - people will have to fsck once and be done.  
> (if they dont fsck then they'll just stay at the old branch)
> 

If the decision is to go with base64, *and* that having a flag day even 
this early on in unacceptable, I'll write a repo converter -- something 
that just reads an old repo and creates a new one.

However, I would like to get a thumbs up/thumbs down first.

	-hpa
-
To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Received on Wed Apr 13 01:07:09 2005

This archive was generated by hypermail 2.1.8 : 2005-04-15 12:56:42 EST