Heya,
On Sun, Aug 29, 2010 at 21:02, Jonathan Nieder [off-list ref] wrote:
I was thinking specifically of the case where one is unlucky enough
for the stream to end at a valid, early spot.
I think it makes sense to say that if you issue a 'feature done', we
change the code that checks for EOF to error instead of quit.
The way all importers seem to end up is to call "git fast-import" as a
child process (rather than advertising an interface like
svnrdump dump <URI> | svn-fe | git fast-import
) so maybe this is not such a big deal.
Does it matter much which way the importer is called? If it ends early
at a valid point nobody will know regardless of how it is called, no?
--
Cheers,
Sverre Rabbelier