On 07/11/2017 02:58 AM, Kashyap Chamarthy wrote: > On Mon, Jul 10, 2017 at 09:36:50PM -0500, Eric Blake wrote: >> On 07/10/2017 03:15 AM, Kashyap Chamarthy wrote: > > [...] > >>> Signed-off-by: Kashyap Chamarthy <kcham...@redhat.com> >>> Reviewed-by: John Snow <js...@redhat.com> >>> --- >> >>> --- >>> docs/devel/bitmaps.md | 505 ------------------------------------------ >>> docs/interop/bitmaps.rst | 555 >>> +++++++++++++++++++++++++++++++++++++++++++++++ >> >> A shame that git rename detection doesn't see these as the same rough >> contents, but not too bad. > > _Should_ it detect?
It depends on the similarities between the file, and on whether you have 'git config diff.renames true' set (there are some other parameters you can set to fine-tune how much effort git expends on detecting file renames, but the defaults are usually good enough). > >> I'll just review the new text; if I point out >> something that was pre-existing in the old text, it may be nicer to >> split the cleanups into a separate followup patch, but I'm also okay if >> they go in as part of this patch. > > Indeed -- the things you point out further below were already > pre-existing. > > Asking out of curiosity: You say it is nicer split because we'll retain > the `git-bisect`-ability? Or just to keep text motion and actual > changes separate? (I think both.) bisect-ability is important on code, but less so on docs. If you were doing code motion, I'd absolutely insist that cleanups be separate from the motion; but here I don't have a strong preference (separating the cleanups DOES make it easier to perform a backport of the cleanup without having to backport the rename, but backporting docs tends to be less of a priority than backporting code functionality). -- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
signature.asc
Description: OpenPGP digital signature
