Raphael Geissert <[email protected]> writes: > On Wednesday 16 September 2009 13:27:55 Russ Allbery wrote:
>> Also, I think that thread identified some other problems with using a >> date from the .TH line to do this out-of-date check. I'm not sure that >> the check is going to be viable, unfortunately, although I suppose we >> could try it as an experimental check and see how bad the false >> positive situation is. But for me personally it dropped in priority >> due to the difficulties it looks like we'll have with doing it >> properly. > I think it might also be implemented by comparing the mtime of the man > pages; still as an experimental test, but it might lead to better > results, don't you think?. Maybe, particularly if we apply some sort of a fudge to allow for stuff committed in odd orders or merged from other branches. Don't warn if the man pages are dated within a day of each other, for instance. -- Russ Allbery ([email protected]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected]
