[PATCH] hgweb: support directory and subrepository navigation when descend mode is disabled

[Matt Mackall]

On Tue, 2012-02-14 at 00:36 +0100, Paul Boddie wrote:
> On Sunday 29 January 2012 20:56:39 Matt Mackall wrote:
> > On Sun, 2012-01-29 at 20:45 +0100, Paul Boddie wrote:
> > > On Sunday 29 January 2012 20:25:38 Matt Mackall wrote:
> > > > On Sun, 2012-01-29 at 20:03 +0100, Paul Boddie wrote:
> > > > > # HG changeset patch
> > > > > # User Paul Boddie <paul at boddie.org.uk>
> > > > > # Date 1327863292 -3600
> > > > > # Branch stable
> > > > > # Node ID 967d0f22ee70762029beb539e15daf97b36ace86
> > > > > # Parent  f8955a7f82e6a06a682cd95631fd1c5e55f46103
> > > > > hgweb: support directory and subrepository navigation when descend
> > > > > mode is disabled
> > > >
> > > > a) it's not clear that this won't be a security surprise to existing
> > > > users
> > >
> > > This shouldn't expose anything more than the directories and repositories
> > > already exposed by hgweb. It isn't generic filesystem navigation; it's
> > > just allowing people to group repositories inside directories and not
> > > have them all shown at once at the first opportunity.
> >
> > No, you've missed my point. If they aren't shown _today_, some people
> > will expect them to continue not being shown _tomorrow_. Thus changing
> > things is setting people up for a nasty surprise.
> >
> > But again: please hold this conversation until after Feb 1.
> 
> I understand the objection now, having looked at the descend behaviour and the 
> * and ** path behaviour as well. Would a separate value be appropriate for 
> descend? Something like "descend = Navigate"?

Probably not. I can't say I like not-quite-booleans. Perhaps an
independent variable.

-- 
Mathematics is the supreme nostalgia of our time.