Source versus Patch (in pcmcia-cs)

[Pete Zaitcev]

On Fri, 11 Feb 2005 17:13:28 -0500, Dan Williams <dcbw at redhat.com> wrote:

> On Fri, 2005-02-11 at 14:07 -0800, Pete Zaitcev wrote:
> > Does anyone know why the startup script in pcmcia-cs is carried in the SRPM
> > as a Source, instead of Patch?
> > 
> > Source3: rc.pcmcia

> Well, rc.pcmcia isn't a patch, its the actual script.  So in that sense,
> it shouldn't be in a Patch: line in the specfile because it never gets
> applied anywhere, it just gets copied to where it needs to go during %
> install.  And it would be pretty silly to generate a patch with
> "diff /dev/null rc.pcmcia" so that it _would_ be a Patch:.

It's a copy of a downrev upstream script which we delete before installing
our own. We have to consider two points here:
 - The patch will include all these -"foo" +$"foo" for i18n
 - The script name is different: upstream uses rc.pcmcia and we use pcmcia.

My concern is that by forking the script wholesale we will slide behind
and lose some important modifications.

I am going to poke David Hinds about accepting our modifications.

Thanks,
-- Pete