Re: [RFC/PATCH 0/3] fast-import: give importers access to the object store

[Jonathan Nieder <jrnieder@gmail.com>]

Ramkumar Ramachandra wrote:

> I finished writing tests for the svndiff0 applier
> and started modifying svn-fe to use it. I've applied your patches and
> rebased my work on top of it.

Nice.

> Ideally, we'd like to be able to supply a commit and a pathname.

To expand on this a little, avoiding dwim-ery of this kind was part
of the spirit I was going for: minimal as possible.  Either the
frontend or the backend is going to have to read the first line of
the commit:

 "tree ...tree name...
  parent [etc]"

and I thought, why not let the frontend do that for now?

Hopefully after using something just barely sufficient like this, we
will know what primitives are needed to make code clear and
efficient (ls?  cat --batch-check?  cat byte range?).  And such
primitives could be at a higher level of abstraction, which might make
the protocol guardians happier.

> On a related
> note, we should probably extend the git-remote-testgit helper to use
> this feature in future to facilitate writing unittests.

Hopefully t/t9011-svn-helper.sh will exercise it in the future.

Thanks for the comments,
Jonathan