James Carlson <james.d.carlson at sun.com> writes: > Richard Lowe writes: >> 417 cdm should sort hg nits output by filename >> >> http://cr.opensolaris.org/~richlowe/scm_417 > > The changes themselves look good and complete. > > There seem to be other places (such as cdm_renamed and cdm_turdchk) > where the user can get an unsorted view of the active list. I get why > we're doing this with 'nits', but are there other cases that should be > fixed?
renamed could sort and perhaps should, yes, I'll do that. turdchk walks changesets, which should be sorted into increasing numerical order by revision anyway. I can make that more explicit at the expense of a needless sort, if you'd prefer? > A somewhat off-topic comment: should we really call merge errors > 'turds'? I know that's what the gatekeepers call the problem, but our > developer tools haven't previously used that word, as far as I know. > At least in some places, I think that scatological term might be > somewhat (and needlessly) offensive. I asked this at the time, and while one person did say something much like the above, everyone else's view was that it was the common term in use. I'd say if you feel strongly enough that it should be changed, come up with alternate but equally understandable/common wording and file a bug. > (As for the filter itself, a nit: using a double-negative -- as in > 'unmodified=False' -- seems as though it makes the code a bit harder > to read. But making a readability comment in Python code is probably > raising a moot point anyway. :->) Yeah, I'm not keen on it at this point, either. I expect to remove it with the fix for #416. -- Rich
