svn commit: r415078 - in head: . Mk

[Olivier Duchateau]

On Sat, 21 May 2016 20:36:43 +0200
Baptiste Daroussin <bapt at freebsd.org> wrote:

> On Sat, May 21, 2016 at 06:35:07PM +0000, Alexey Dokuchaev wrote:
> > On Sat, May 21, 2016 at 11:11:12AM -0600, Adam Weinberger wrote:
> > > > On 21 May, 2016, at 10:29, Alexey Dokuchaev <danfe at FreeBSD.org> wrote:
> > > > On Sat, May 21, 2016 at 02:25:22PM +0200, Baptiste Daroussin wrote:
> > > >> Implementing it as a Makefile variable would make it not automatically
> > > >> updated.
> > > > 
> > > > Even if we can agree on REP_TIMESTAMP variable, why `make makesum' cannot
> > > > sed -i the Makefile?
> > > 
> > > Oh god, please, no. Please do not mess with Makefiles.
> > > 
> > > Rather than having it come from 'makesum', can an svn hook write it to
> > > distinfo? May help it stay more current.
> > 
> > I understand your worries, but SVN already messes with Makefiles (keywords).
> > On the other hand, adding $Id$ to distinfo would both 1) solve the problem,
> > at least as I see it now, and 2) won't mix data with metadata which is I'm
> > debating against.
> > 
> > Baptiste, was adding $Id$ to distinfo considered/tested?  If yes, why was it
> > rejected?
> > 
> It won't work because keywords are not expanded in git and hg (and yes we do
> officially support our git mirror)

About Mercurial (hg), it supports keywords too. It's even an official extension.

> 
> Best regards,
> Bapt


-- 
Olivier Duchateau <olivierd at freebsd.org>