Duy Nguyen wrote: > On Mon, Mar 25, 2013 at 1:17 AM, Jonathan Nieder <jrnieder@xxxxxxxxx> wrote: >> Hm, should this be the default? >> >> In principle, I would expect >> >> git checkout -- . >> >> to make the worktree match the index, respecting the sparse checkout. >> And something like >> >> git checkout --widen -- . >> >> to change the sparse checkout pattern. [...] > Changing the default may involve a painful transition phase (e.g. "add > -u"). I don't think it needs to. There aren't many people using sparse checkout even these days, and I think they'd generally be happy about the change. But if we want to be conservative until some later point (like 2.1), perhaps --sparse should be named something like --no-widen? That way, I can do git checkout --no-widen -- . to make the worktree match the index, respecting the sparse checkout. And I can do git checkout --widen -- . to change the sparse checkout pattern. Meanwhile the confusing command git checkout -- . would be ill-defined for sparse checkouts --- in past git versions, if I understand you correctly it acted like --widen, while in some unspecified future version it may change to mean --no-widen. No need for warnings because I doubt anyone is relying on either behavior. Would that work? Jonathan -- To unsubscribe from this list: send the line "unsubscribe git" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html