[pkg-bioc] CVS %p

[David Vernazobres]

Hi all, 
Hi Steffen,

I will try to answers commit mail after commit mail, without mistake!


On Sat, Apr 28, 2007 at 07:15:04PM +0000, CVS User smoe-guest wrote :
> Update of /cvsroot/pkg-bioc/tools
> In directory alioth:/tmp/cvs-serv9368
> 
> Modified Files:
> 	cran2deb.pl r_pkg_prepare.sh r_pkg_update.pl 
> Log Message:
[...] 
> * r_pkg_*
>   - moved from etch to lenny (hey, we got a release in the meantime :-) )
>     cdebootstrap fails when specifying 'sid', to spot that took me an hour,
>     someone with more insights please elucidate this one and send a bug
>     report/fix.
Hum, Yes, I was with a bunch of patch on my local repository, but I
somehow annonced it to the list 1 week ago (I should look for the date
in my history), saying that pbuilder was broken.
It's my fault, I have learn, I spent some time solving all my local
conflit !
Well, due to the release of Etch, cdebootstrap has been broken for some
time (the right version was in unstable). And in fact, the i386 is ok
now, but the amd64 pbuilder (at least Thurday) was still broken.
I was trying to figure out why, before repporting an "empty bug".

>   - added stable, testing and unstable to /etc/apt/sources.list, the r-base-dev
>     otherwise could not find the r-base-core 2.5

>   - some also allowing for --create-bioc in the argument check ... David, how could
>     you forget about it :o)
Well because my original version was --create-all ! 
And I think to remember that you add the diversion --create-cran. ;)

> What we now need I think is just a slight bit of an instruction. The
> Wiki page is about right there. Something noteworthy to add, learning
> from my erroneous expectations, is that r_pkg_prepare only prepares the
> chroot and that it is only the r_pkg_update and only when called as specified
> in the wiki that creates it .... however trivial it is.

Yep, wiki and documentation are old. It's on my plan to try to update
them, but somehow 1 day is just 24h.... But I know that code whitout
documentation is not code. 

> The error below so I presume results of our undecent build order, right?
> I am afraid that it was me who was once promising some improvements on
> this front.  Sigh. Maybe after this salad waiting downstairs for me.
> 
[...]

Best regards
david