[Debian-med-packaging] RFS: codonw

[Andreas Tille]

Uploaded, thanks for your work on this

    Andreas.

On Fri, Nov 27, 2015 at 10:46:47PM +0000, Sascha Steinbiss wrote:
> Hi Andreas,
> 
> as I have not heard back from upstream I have made the modifications as discussed by renaming the links to non-clashing variants prefixed with ‘coding-‘ and patching the source to behave as before when called using these links. These changes are pushed to git now.
> Since the main codonw binary can still be called on its own as well, I have put both that and the links into /usr/bin.
> 
> Cheers
> Sascha
> 
> > On 24 Nov 2015, at 20:44, Andreas Tille <andreas at an3as.eu> wrote:
> > 
> > On Tue, Nov 24, 2015 at 07:00:07PM +0000, Sascha Steinbiss wrote:
> >>>> 
> >>>> Well, these ‘tools’ are only symlinks to the main executable causing it to behave differently when called via the link. The functionality is also available using command line switches in the main codonw executable. Do you think it would be acceptable to drop or rename (e.g. codonw-cai, codonw-tidy, etc.) the linked tools? This would be my preference but could probably break existing third party scripts.
> >>> 
> >>> IMHO the most flexible idea is to move these to
> >>> 
> >>>  /usr/lib/codonw
> >> 
> >> What are ‘these’ in this case? The symlinks to the only binary produced (codonw)?
> > 
> > I'd suspect the binaries in /usr/lib/<packagename> but that might be
> > just me.  I'm not aware about any explicite policy statement about this.
> > 
> >>> *and* provide the /usr/bin/codonw-* files as links.  This might require
> >>> a certain amount of patching in the code but should be doable.
> >> 
> >> So you are thinking of links to links? Or should both the links in /usr/bin and /usr/lib/codonw link to the /usr/bin/codonw executable? Certainly doable but I am wondering about the reason for the extra layer of indirection if you meant the first case.
> > 
> > This might be a misunderstanding.  I do not want to make things more
> > complicated than needed.
> > 
> >>>> I can contact them and bring it up.
> >>> 
> >>> May be you could find a sensible solution together with upstream.
> >> 
> >> I have just contacted them but the original email address is invalid. Let me see if I can contact them another way.
> > 
> > I've read your mail(s) and think this is fine (and hope upstream will
> > respond).
> > 
> > Kind regards
> > 
> >      Andreas.
> > 
> > -- 
> > http://fam-tille.de
> > 
> > _______________________________________________
> > Debian-med-packaging mailing list
> > Debian-med-packaging at lists.alioth.debian.org
> > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging
> 
> 

-- 
http://fam-tille.de