On Tue, May 01, 2018 at 12:42:57PM +0200, Duy Nguyen wrote:
> On Mon, Apr 23, 2018 at 11:39:49PM +0000, brian m. carlson wrote:
> > Several of our shell scripts hard-code the object ID of the empty tree.
> > To avoid any problems when changing hashes, compute this value on
> > startup of the script.  For performance, store the value in a variable
> > and reuse it throughout the life of the script.
> > 
> > Signed-off-by: brian m. carlson <sandals@crustytoothpaste.net>
> > ---
> >  git-filter-branch.sh               | 4 +++-
> >  git-rebase--interactive.sh         | 4 +++-
> >  templates/hooks--pre-commit.sample | 2 +-
> >  3 files changed, 7 insertions(+), 3 deletions(-)
> > 
> > diff --git a/git-filter-branch.sh b/git-filter-branch.sh
> > index 64f21547c1..ccceaf19a7 100755
> > --- a/git-filter-branch.sh
> > +++ b/git-filter-branch.sh
> > @@ -11,6 +11,8 @@
> >  # The following functions will also be available in the commit filter:
> >  
> >  functions=$(cat << \EOF
> > +EMPTY_TREE=$(git hash-object -t tree /dev/null)
> 
> All scripts (except those example hooks) must source
> git-sh-setup. Should we define this in there instead?

I think at this point, I'm okay with special-casing these two uses, but
I would generally say that if we gain any more we should move it there.

There's a trade-off between the benefits of reuse here and the fact that
we're forking a process, which incurs a cost, especially on Windows.

I'm open to hearing other opinions, of course.
-- 
brian m. carlson: Houston, Texas, US
OpenPGP: https://keybase.io/bk2204