A webrev containing fixes for: 368 hg recommit with no actual changes confuses the mortals 373 308 didn't do enough for pbchk and nits 377 cdm should not allow you to putback a null changeset
Is available at http://cr.opensolaris.org/~richlowe/scm_368 The 'fix' for #377 doesn't do the full amount necessary (partly because it needs to be hooked, and we don't have those bits yet). I decided to put it back as it stands, since John got into a situation where he may have putback such a cset to the xvm gate (or could have, at least), and was only really held back by the confusion caused by #368, which this wad fixes. It seemed better to introduce the warning while fixing the only thing that stopped it happening last time. It's also not correct in a multiple changesets per changegroup world (many other things are suboptimal in that situation, too). Do we still intend to encourage and support that, or do we intend to behave as we do with current policy and disallow it? (we discussed this before, long ago, and I'm generally in favour of allowing it, if we can enforce sane behaviour, but I'm starting to think that's harder than it first seemed.) Opinions welcome. -- Rich
