[PATCH 1 of 2] merge: drop underscore prefix from _checkunknown()

[Martin von Zweigbergk]

On Mon Nov 17 2014 at 2:34:29 PM Matt Mackall <mpm at selenic.com> wrote:

> On Fri, 2014-11-14 at 16:21 -0800, Martin von Zweigbergk wrote:
> > # HG changeset patch
> > # User Martin von Zweigbergk <martinvonz at google.com>
> > # Date 1415949135 28800
> > #      Thu Nov 13 23:12:15 2014 -0800
> > # Node ID c2491406f4623e148bee137b1862ff0b609e4235
> > # Parent  991098579940552536d0a99fa3602dd1661aa388
> > merge: drop underscore prefix from _checkunknown()
>
> > The method has been called from commands.py since 3eab42088be4
> > (update: just merge unknown file collisions, 2012-02-09), so drop the
> > underscore prefix that suggests that it's private.
>
> I look at the underscore in Python as advisory: you shouldn't be using
> this unless you have to. And when you see it used like this, you should
> think "ah, this is a hack of some sort".
>
> And in this case, it certainly is: we definitely don't want the commands
> layer to know about this obscure merge detail. So I think the underscore
> is correct as it stands and leaves an appropriate code smell here.
> Removing it is only going to reduce the probability that a future Martin
> will be bothered enough to properly refactor away the hack.
>

Oh, it didn't seems like an obscure merge detail to me at first, but that's
good to hear. It turns out the series of which this is the second patch
turned out deleting _checkunknown() altogether. I'm glad to hear it wasn't
a completely delusional idea.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://selenic.com/pipermail/mercurial-devel/attachments/20141118/9563b7e6/attachment.html>