On Wed, 30 Jul 2003, Harold L Hunt II wrote:
Todd T. Fries wrote:
Penned by Daniel Stone on Wed, Jul 30, 2003 at 04:51:49PM +1000, we have: [..] | I did suggest this (mv hp,v hp.old,v).
If you guys care about history at all, aka the ability to checkout files in the past, you would not suggest nor impement this suggestion.
Does history matter when some people simply cannot checkout a tag set because of a mistake that was made? The balanced response here is that the history is not as important as making sure that all developers can move forward. Of course, we do this only in rare circumstances, but this is one of them.
I fetch older versions all the time - on xfree86, I did that, for instance to narrow down the range of dates for a bug introduced into the mouse driver. Having a known working version of something lets you cut down on the effort to isolate a bug. (Moving "forward" doesn't use the archives for anything except an odd backup format).
What's the problem here?
Go look at HEAD! HEAD checks out just fine and has already had the "hp" file moved out of the way. Your argument is moot. The only problem is that the decision that was made and applied to HEAD has not been applied to xf-4_3-branch.
Harold
_______________________________________________ Devel mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/devel
