[PATCH 2 of 2] debug: pass debug flag to remote ui object

Matt Mackall mpm at selenic.com
Fri Apr 17 10:08:17 CDT 2015


On Fri, 2015-04-17 at 04:31 -0400, Pierre-Yves David wrote:
> 
> On 04/16/2015 07:39 PM, Matt Mackall wrote:
> > On Wed, 2015-04-15 at 00:27 -0400, Pierre-Yves David wrote:
> >>
> >> On 04/14/2015 04:32 PM, Eric Sumner wrote:
> >>> # HG changeset patch
> >>> # User Eric Sumner <ericsumner at fb.com>
> >>> # Date 1429043033 14400
> >>> #      Tue Apr 14 16:23:53 2015 -0400
> >>> # Node ID b10599cba2adb8bf281ea9c913d6f235780618bf
> >>> # Parent  b30cb32ea6821e165aab65051767aa810fdb8464
> >>> debug: pass debug flag to remote ui object
> >>>
> >>> When we're talking to another repository, --debug shows more information than
> >>> setting ui.debug=True inside .hg/hgrc; this patch adds ui.debug to the config
> >>> options that get copied into the new UI object for remote repos
> >>
> >> Eric managed to convinces me that the "remote" is a small object that
> >> exit in the "local" world, it does not read the other repo
> >> configuration. So it make sense to reuse some of the local config (and
> >> actually already to it). However, this mean we should also copy other ui
> >> flag like 'verbose' or 'quiet'. I'm expecting a V2 with those, and some
> >> testing.
> >
> > This is odd. The way I'd expect this to work is:
> >
> > - user adds --debug
> > - "global" ui gets created with debug set
> > - "repo" ui gets created based on global ui, inherits debug
> > - "remote" ui gets created based on global ui, inherits debug
> >
> > If that's not working, then we should figure out why not.
> 
> Eric's case is where ui.debug=True is set in the local repo.

I see. I think this is more trouble than it's worth.

-- 
Mathematics is the supreme nostalgia of our time.




More information about the Mercurial-devel mailing list