On Tue, Jul 3, 2012 at 12:20 PM, Lluís Batlle i Rossell <[email protected]>wrote:
> Maybe 'fossil stash' should handle better the renames situation? It looks > a bit > broken. > A stashed item is a snapshot of its state in the past, and that state is constant/final/immutable (as the past always is in fossil). A rename happens in the future, and a stash cannot be aware of that (nor even know if it should, for a given case, handle that - in the generic case this is not answerable). Perhaps stash is doing abstract the equivalent of "fossil revert" when it should instead be doing "fossil revert stashedfile1... stashedfileN"? i'm taking a look at it now, but the stash apply code is new to me and does manifest-level work i'm not familiar with :/. i'm not certain if i'll have this grokked in my last remaining hacking hour for the day :(. (i did, however, see an easy optimization opportunity to skip an unnecessary blob_compare() in one branch, and that will be checked in soon.) -- ----- stephan beal http://wanderinghorse.net/home/stephan/ http://gplus.to/sgbeal
_______________________________________________ fossil-users mailing list [email protected] http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
