Re: [MoM]: r9659 - trunk/packages/fis-gtm/fis-gtm/trunk/debian
On Wed, Feb 15, 2012 at 11:49 AM, Karsten Hilbert
<Karsten.Hilbert@gmx.net> wrote:
> On Wed, Feb 15, 2012 at 04:29:16PM +0100, Andreas Tille wrote:
>
>> As a last resort for the *initial* packaging it might work out. Once we
>> are doing the real packaging we should not try such dirty tricks.
>
> Now that we can generate a mumps compiler by way of
> compiling only C code I wonder whether there's still a need
> to separate between "initial" and "real" packaging ? After
> all, no "bootstrapping" is needed anymore ?
>
> AFAIU we are currently creating a (standard) package for a
> "programming-language-with-database-called-M" from its
> FLOSS implementation "GT.M" by FiS ?
>
> Having done that we would be creating a package to install
> VistA as one (complex) application running in/on the
> installed-by-package M environment (incidentally being
> fis-gtm).
>
> While, yes, one requires the other both are "proper" or
> "real" packages ?
>
> Am I on the wrong track ?
>
> Karsten
> --
---------------------------------------
Karsten,
Yes, we have indeed dropped (or at least put on hold)
the bootstrapping approach of having two packages:
1) fis-gtm-initial
2) fis-gtm-server
and have worked mostly on the option of a pure-C
code base, for which we have put all the debian
configuration files in the directory:
fis-gtm/fis-gtm/trunk/debian
I believe Andreas was using "initial" as in "version 0.1"
of the packaging, e.g. the first quick and dirty way of
getting it done, and then revising it to follow better
practices and cleaner methods to become a "version 1.0".
... or at least that's how I'm reading it... :-)
Luis
Reply to: