* On 11 Sep 2014, Óscar Pereira wrote: > On Wed, Sep 10, 2014 at 12:32:04PM -0500, David Champion wrote: > >>>So I guess the question is... does someone still care about code > >>>indentation, or is it just the page [1] that is hopelessly outdated? > > > >I don't personally like the historical mutt coding style, but it's our > >coding style, and we need to stand by it until we formally change it. > > That's just it. The indent line I mentioned [1] comes from the page > on mutt's coding style [2]. Thus I would assume that mutt's source > code follows the style that one obtains from running the indent tool > with those options. But given the huge diff I obtained, such is > clearly not case. So I wanted to know the reason of such > discrepancy.
Ah, I see your point. I actually didn't get your first post - perhaps it was blocked for being so large? I don't know how the mail service is set up. I don't know specifically what you found deviant. When I run indent, some of what I see is because of ifdefed block ends (#ifdef __cplusplus; extern "C"; etc) causing spurious indentation changes. Some consists of genuine indentation changes, but from 2-space indents to 4. That's odd; we've used 2-space as far back as I recall, so perhaps the wiki is wrong, or perhaps indent is wrong. I confess I've never used its advice; I've just been assiduous in hand-formatting. I see some things that are obviously where people stopped caring about the style. Indent(1) is a bit of a problem in general because there are so many indents to choose from. None of them are fully mutually compatible, though each is fully unintelligible. I like uncrustify for this purpose. There's only one of it. I'll observe that whatever the prettifier tool, the configuration that results in the smallest diff is the one that most accurately describes the current code. At this stage of mutt's life, I'm more interested in that than in fitting a prescriptive recipe at some cost to the patches that reside in the wild. -- David Champion • d...@bikeshed.us
pgpjL2Ev8rY0g.pgp
Description: PGP signature