On Sun, May 22, 2005 at 10:40:45PM -0500, Dirk Eddelbuettel wrote: > On 22 May 2005 at 19:51, Steve Langasek wrote: > | On Sun, May 22, 2005 at 02:28:21PM -0500, Dirk Eddelbuettel wrote:
> | > | It looks like rpy 0.4.1-4 fixes this bug by adding a full set of R-2.1.0 > | > I recalled that we had fixed it; I guess I confused 2.0.1 with 2.1.0 here. > | > | headers inside the tarball. Is this really the appropriate fix? If > so, I > | > | can push 0.4.1-4 into sarge; but it looks like these headers are > duplicates > | > | of the ones already present in r-base-core, and that this is actually a > bug > | > | with the upstream build-scripts? > | > You need to talk to Greg (== upstream) about that. He calls this > "batteries > | > ^H generators included". For other less stringently organised upstream > | > systems, shipping the headers is appropriate. For us, it is overkill, but > | > then this ain't a Debian-native package so ... > | Eh, these headers were all added to the package in the Debian diff between > | 0.4.1-2 and 0.4.1-4; it doesn't look to me like upstream's to blame for > | their presence. > Right, but upstream is to blame for the decision to include all headers and > kitchensinks : > [EMAIL PROTECTED]:~/src/debian/Python/rpy-0.4.1> du -csk R-[12]* > 496 R-1.8.0 > 496 R-1.8.1 > 504 R-1.9.0 > 504 R-1.9.1 > 592 R-2.0.0 > 592 R-2.0.1 > 520 R-2.1.0 > 3704 total > I just checked the changelog. 0.4.1 (upstream) added headers for R 2.0.1. > Debian 0.4.1-3 added R-2.1.0 building on the existing "batteries err > generators included" framework of yielding to all known recent R versions. > Which is, as I agree, a defensible strategy. > | > What do I have to do to get 0.4.1-4 into sarge? Looking from the > different > | > 'build', 'excuses' and 'more' links off my qa summary page > | > (http://people.debian.org/~igloo/status.php?packages=rpy), it looks like > this > | > has been built everywhere. Can you push it into sarge? > | Only if you can really explain why the package grew all of these headers in > | a Debian revision, and why rpy can't be fixed to use the r-base headers > | installed on the system (as a result of the build-dep) instead. :) > See above for part one, and see Greg for part two as I alluded to earlier :) Ok. Thanks to Don's analysis that pointing at the r-base packaged headers is non-trivial, I've gone ahead and hinted 0.4.1-4 in. Thanks, -- Steve Langasek postmodern programmer
signature.asc
Description: Digital signature
