subrepo grand plan

[Matt Mackall]

On Thu, 2011-10-20 at 18:36 +0200, Dirkjan Ochtman wrote:
> On Thu, Oct 20, 2011 at 17:18, Matt Mackall <mpm at selenic.com> wrote:
> >> I'm with Patrick in that I prefer (b) even now. I find his argument
> >> about predictability and scalability of performance compelling, but
> >> mostly I find recursive commit (by default) plain scary.
> >
> > Would it still be scary if you had recursive diff and status by default?
> 
> Much less so, of course. Does rebase also recurse by default? log?
> parents? It still seems to me that the conceptual integrity would
> suffer, and it would take a whole lot of time and changesets to get
> everything integrated properly with the principle of least
> astonishment intact.
> 
> On Thu, Oct 20, 2011 at 17:32, Matt Mackall <mpm at selenic.com> wrote:
> > We'd need to add a -S switch to get the old behavior and a hint.
> >
> > I'm a little concerned that this will bring development on subrepo
> > integration to a stand-still.
> 
> Not sure what you mean? An -S switch for commit?

$ hg commit
abort: uncommitted subrepos
(use commit --subrepo to get the behavior we took away)

>  Why would this bring
> development to a stand-still? What kind of "integration"?

If we do (b) now, the pressing need for (a) will vanish.

-- 
Mathematics is the supreme nostalgia of our time.