Thijs Kinkhorst <[EMAIL PROTECTED]> writes: > On Thursday 10 May 2007 17:28, Russ Allbery wrote:
>> However, it's not correct that only the top-most changelog entries are >> relevant. > That's true in some cases, and indeed the .changes file might address > that. But I wonder whether there will be many more "false positives", > i.e. many-year-old changelog entries being flagged, than bugs that are > missed in those cases. At least I would see flagging a non-current > changelog entry that way as a false positive. > What course of action do you suggest? Fix the changelog retroactively? That's what I'd do, but you're right, there are a lot of hits that most people will consider a false positive, and which at the most should be info-level. But the problem in the current changelog entry should be a warning. I think the right course of action is to do what you suggest and change the test to only check the most recent entry, with follow-on work to check the *.changes file specifically. An argument could be made that changelogs should be fixed retroactively so that automated analysis of the changelog will correctly detect when past bugs were fixed, but I think that's a very minor benefit and most people aren't going to be interested in doing that. I looked at this again and I think I see how to fix it, with a bit more research. I'll try to get it changed in the next upload. -- Russ Allbery ([EMAIL PROTECTED]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
