On Wed, Feb 05, 2014 at 08:57:12PM +0400, Kirill Smelkov wrote:

> Since diff_tree_sha1() can now accept empty trees via NULL sha1, we
> could just call it without manually reading trees into tree_desc and
> duplicating code.
> Besides, that
>       if (!tree)
>               return 0;
> looked suspect - we were saying an invalid tree != empty tree, but maybe it is
> better to just say the tree is invalid here, which is what diff_tree_sha1()
> does for such case.

I think that is sensible. The assertion that "invalid != empty" is
probably sane, because we handle the empty tree as internal magic. But I
do not see any reason we should be hitting this code path regularly with
an invalid tree, short of repository corruption, so in practice I don't
think it matters.

This does introduce a die() where there was not one previously, and that
can make things harder to diagnose/debug in a corrupted repository. But
it looks like this is limited to the history-simplification code, and I
suspect that it is not commonly used in the case of corruption.

So I think the patch looks fine.

To unsubscribe from this list: send the line "unsubscribe git" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to