On Thu, Aug 29, 2019 at 08:22:49PM +0900, Mike Hommey wrote: > It is possible to end up in situations where a replace ref points to > itself. In that case, it would seem better to stop the lookup rather > than try to follow the link infinitely and fail with "replace depth too > high". I don't think this is worth doing. It's just a special case (a 2-node cycle in the replace graph) of a more general one (an n-node cycle). We catch the general case with the depth counter (though of course there are other techniques, which we could debate). Is it worth adding extra code to cover this special one? > I'm not entirely sure whether it's actually worth fixing. Arguably, `git > replace` should also avoid the footgun in the first place (although in > my case, it wasn't due to `git replace` itself). Yes, if "git replace $OID $OID" does not complain, it probably should. Perhaps it should even confirm that the replacement can be resolved, and does not point to the original object. That covers the n-node case, as well. -Peff
