On Thu, Jul 03, 2008 at 12:41:43PM -0700, Junio C Hamano wrote: > Once reflog entries begin expiring, we won't have the null_sha1 on the > previous side of them, and after that we will keep returning the > oldest-known one, so nobody should be expecting to find out when a ref was > first created in this repository by checking with what timestamp that > syntax breaks in the existing interface. Right, that was my thought: though perhaps not technically as precise, it is much more DWIM. My other concern, which I failed to voice in the original mail was: should we do this magic for other reflog entries besides the first one? It is possible to end up with a null sha1 in any other entry? Branch creation shouldn't do it, because we will have deleted any previous reflog when killing off the branch before, so we always end up putting in the "oldest" reflog entry. Cloning shouldn't do it for obvious reasons. > Sorry for not being Shawn ;-) Heh. -Peff -- 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
