[PATCH 1 of 6 RFC] ui.config: fix bug in config alternatives from cc669e4fec95

Matt Mackall mpm at selenic.com
Wed Aug 7 17:13:53 CDT 2013


On Fri, 2013-08-02 at 10:15 -0400, Augie Fackler wrote:
> # HG changeset patch
> # User Augie Fackler <durin42 at gmail.com>
> # Date 1374707619 14400
> #      Wed Jul 24 19:13:39 2013 -0400
> # Node ID 123bd5dda28b5452afd8881488d14056f53ae421
> # Parent  a58251c0568fc5e86089a803ca75f75cc8c01678
> ui.config: fix bug in config alternatives from cc669e4fec95

Queued the first one.

I think we need another think about the namespace issue we discussed on
IRC:

[foo]
barbaz = False
bar_baz = True # this one should take precedence if we accept both forms

How do we handle this intelligently?

Thought: we give config a set of characters to fold on (ie [-_]) and it
builds a parallel tree of folded identifiers. Iff we do a lookup with
alternatives, we consult that tree, otherwise we consult the normal
tree.

-- 
Mathematics is the supreme nostalgia of our time.




More information about the Mercurial-devel mailing list