On Wed, Jul 2, 2014 at 5:24 PM, David T. Lewis <lewis at mail.msen.com> wrote:

> On Wed, Jul 02, 2014 at 12:36:15PM -0700, Eliot Miranda wrote:
> > On Wed, Jul 2, 2014 at 12:04 PM, David T. Lewis <lewis at mail.msen.com>
> wrote:
> >
> > > The script is installed as /usr/local/bin/squeak.
> > >
> > > Yes that is the one that might get stepped on by the Cog install.
> > >
> > What cog install is this?  The Cog tarballs on my site don't install
> > anywhere specific.  They unpack to a directory called e.g. coglinuxht,
> > cogspurlinux, etc.  They don't stomp on anything.
>
> The squeakvm interpreter VM is typically installed in /usr/local/, so the
> start script is /usr/local/bin/squeak. This has been the case for the last
> 15 years or so.
>
> If a unix/linux installs Cog, they will typically want to install it in
> /usr/local/. In that case, the Cog start script stomps on the squeakvm
> start script.
>

No it doesn't.  Look at the tarballs.  They expand to a local directory
called coglinux, et al.  Further, their squeak scripts are cntained in that
directory hierarchy.  What you describe would only happen if one did a
build of a Cog VM and told the configure script to install in /usr/local.
 Note that my build scripts dont do this.  They "install" to a staging
directory in which the tarballs are made.


The same issue exists for installation in /usr/ as opposed to /usr/local/.
>
> Dave
>

I don't agree.  I don't see how one could do this inadvertently at all.
-- 
best,
Eliot
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/squeak-dev/attachments/20140702/4efac5b4/attachment.htm