Jack Tanner ihok-PkbjNfxxIARBDgjK7y7TUQ@public.gmane.org writes:

This seems like it's going to be extremely useful when I get back to working on ggplot2. Thanks for your hard work. Unfortunately, that project just got punted down the priority list for me, so I can't really offer any constructive feedback right now.

OK, sounds like I should just get my undergarments unwadded, and wait for someone else to get a Round Tuit.

- Allen S. Rout

Tom Moertel tom-RKvYLT2W2Q9BDgjK7y7TUQ@public.gmane.org writes:

I am interested in seeing your script. It sounds like it embodies a lot of knowledge about building a bunch of R packages, and that knowledge is valuable. I'd hope we can make some use of it. :-)

http://nersp.osg.ufl.edu/~asr/media/R2spec-2.6.2-asr.src.rpm

It depends now on python-cheetah (available from EPEL).

Updated instructions:

----

for me to get the dependencies to build and install all (most) of ggplot2's broad depgraph, I must:

yum install openmpi openmpi-devel openmpi-libs unixODBC-devel pvm tcl tcl-devel perl-DateManip perl-Date-Calc perl-version

and then install from EPEL:

environment-modules-3.2.6-4.el5.x86_64.rpm mpich2-1.1.1-1.el5.x86_64.rpm perl-IO-stringy-2.110-5.el5.noarch.rpm perl-Jcode-2.06-6.el5.noarch.rpm perl-OLE-Storage_Lite-0.14-9.el5.noarch.rpm perl-Parse-RecDescent-1.96-1.el5.noarch.rpm perl-Spreadsheet-WriteExcel-2.18-1.el5.noarch.rpm perl-Unicode-Map-0.112-12.el5.x86_64.rpm

and then install from the graphviz people:

graphviz-2.24.0-1.el5.x86_64.rpm graphviz-devel-2.24.0-1.el5.x86_64.rpm

After installing the r2spec and the sysdeps, I made a temp directory, /var/tmp/Rbuild. I've got every script running happily from '.'; so there's no specialness about the tempdir any more.

There, I ran (as root): [...]

I heard the 'prefer not running as root' from earlier: since step N below is to install all these packages, is that really practical? In any case, the below can be run as a normal user, but the install steps below can't.

# clear; R2spec -n "Allen S. Rout" -e "asr-hnhdhkBXzx8@public.gmane.org" -c --force --suggests --verbose --package ggplot2

This downloaded the CRAN-like lists of packages, downloaded source tarballs, and built (and copied) the specfiles and source.

It also generated a few dot files, including the broad and narrow dependency graphs for the requested package.

It also generates a 'buildorderlist'; a path through the depgraph such that each package is attempted only after its dependencies are processed.

Then I stripped out the packages I knew would not complete their RPM build (mentioned below)

# mv buildorderlist buildorderlist.orig

# egrep -v "R-Rmpi|R-rsprng|R-multcomp|R-flexmix|R-RandomFields|R-MCMCpack|R-tcltk2" buildorderlist.orig > buildorderlist

Then I ran

/usr/lib/python2.4/site-packages/r2spec/buildfromlist.sh

Is it evil for me to have put the script in there? This is certainly not general-purpose; putting it in the default bin would be silly.

This will build and install all of the RPMs in the broad graph. ... If the individual builds work.

This script uses '.' or environment variable 'RBUILDLOG' as a tempdir extensively; it drops a build log and an install log for every package addressed. That way when things eventually bust you can see where they went wrong.

Then I ran

/usr/lib/python2.4/site-packages/r2spec/checkfromlist.sh

which runs 'R CMD check' on every package, again logging, and summarizing success or failure in 'checkresults'.

The check step took about 5.5 hours for me.

----

I still fail to build (and haven't investigated much yet):

tcltk2: Fails to install: need tclsh8.3, not 8.4 ?

multcomp: requires Survival >= 2.35.7 ? ...

flexmix: dep on multcomp

R-RandomFields, R-MCMCpack: uncaught debug files in the build tree.

MPI, sprng

----

- Allen S. Rout

"Tom "spot" Callaway" tcallawa-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org writes:

I think that there is no need to hand-craft every R package, but that each R spec file needs to be:

• Sanity checked, and adjusted as necessary
• Actively maintained

I don't want to simply run any tool, generate a bunch of spec files, get them into Fedora/EPEL, and then let them stagnate because one person owns all of CRAN. :)

Keen; can you refine those statements?

I might be able to address 'actively maintained' by analogy to the debian folks' efforts on this. They regularly run basically an update which detects package availability changes, and builds the new ones. So, a "reasonable" time after a new package appears on CRAN, it appears in the apt repo Dirk's maintaining. That's clearly one component of active maintenance; do you have more off the top of your head?

As for 'sanity checked', I don't know what that means. I'd like to pretend that any package for which R CMD check passes successfully has been sanity checked, but I know that to be a skewed persepctive. :) If you can give me criteria, I can start working to implement them.

- Allen S. Rout