[PATCH] 1321 (partial): run Python pretxnchangegroup hooks before disk flush

[Matt Mackall]

On Fri, 2009-01-16 at 02:15 -0500, Doug Philips wrote:
> On or about Friday, January 16, 2009, at 01:42AM, Matt Mackall indited:
> >I'm convinced we need to do something here.
> ...
> >This would still need a new hook name, and we could provide a way for
> >in-process hooks to see the updated repo similar to what you've done.
> 
> I like the idea of having an environment variable that can keep hook-y code "in the know" and other code immune.
> 
> The question though, is why is a new hook needed?
> It seems from what I've read about the pretxnchangeset hook, the "outside visibility" is a bug, not a feature.
> As such, why not just fix the bug?

The 'outside visibility' is what lets a hook actually do useful work. As
such it cannot be fixed.

We can add a new hook that adds new magic, loudly mark the old hook as
deprecated, and then (in like a year) remove the old hook. But fixing
the old hook (better known as breaking it) is not allowed.

-- 
http://selenic.com : development and support for Mercurial and Linux