On 24 September 2005 at 14:30, Douglas Bates wrote:
| On 9/24/05, Dirk Eddelbuettel <[EMAIL PROTECTED]> wrote:
| >
| > On 24 September 2005 at 20:22, Kurt Roeckx wrote:
| > | Package: r-base
| > | Version: 2.2.0.alpha.20050908-1
| > | Severity: important
| > |
| > | Hi,
| > |
| > | Your package is failing to build on amd64 with the following
| > | error:
| > | # edd 05 Sep 2005  expand this to R_SHARE_DIR, R_INCLUDE_DIR, R_DOC_DIR
| > | (cd 
/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/bin/; \
| > |         ln -svf ../../../bin/R R;                               \
| > |         cd ../../../bin;                                        \
| > |         perl -p -i -e                                           \
| > |             's|^R_HOME_DIR=.*|R_HOME_DIR=/usr/lib/R|;\
| > |              s|^R_SHARE_DIR=.*|R_SHARE_DIR=/usr/lib/R/share|;\
| > |              s|^R_INCLUDE_DIR=.*|R_INCLUDE_DIR=/usr/lib/R/include|;\
| > |              s|^R_DOC_DIR=.*|R_DOC_DIR=/usr/lib/R/doc|'         R)
| > | /bin/sh: line 0: cd: 
/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/bin/: No 
such file or directory
| > | create symbolic link `R' to `../../../bin/R'
| > | Can't open R: No such file or directory.
| > | chmod a+x       
/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/echo.sh
         \
| > |                 
/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/help-links.sh
   \
| > |                 
/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/help-print.sh
| > | chmod: cannot access 
`/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/echo.sh':
 No such file or directory
| > | chmod: cannot access 
`/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/help-links.sh':
 No such file or directory
| > | chmod: cannot access 
`/build/buildd/r-base-2.2.0.beta.20050922/debian/r-base-core/usr/lib/R/share/sh/help-print.sh':
 No such file or directory
| > | make: *** [install-arch-stamp] Error 1
| > |
| > |
| > | I assume that this is not the actual reason why it's
| > | failing, and that it's something that happened before in
| > | the log, but I can't find it.
| >
| > Could you please look by hand? This very much looks like a local hickup in
| > the buildd.  If you look at the aggregation at
| >         http://buildd.debian.org/build.php?&pkg=r-base
| > you see that 2.2.0.beta.20050922 built everywhere (but ia64 where the gnome
| > conf script it calls fell over).
| >
| > Not R's bug, and I intend to downgrade this -- unless you can convince that
| > there is something wrong the package. I have no reason to believe that there
| > is.  The part failing above is also unchanged for many releases.
| 
| Is it because the files are going into a directory of the form
| .../usr/lib64/R/... not .../usr/lib/R/...?

Ah, nice catch. That's a possibility, would explain why it hits only amd64
and show it is our bug after all :)
 
| I know if I build R from the sources on an amd64 machine I end up with
| the R library in /usr/local/lib64/R

Fair point.

There are a ton of occurrences of "/lib" in debian/*.  Could be messy to make
all that conditional for amd64 esp. the various debian/*.files and
debian/*.links.  Can those even be per-arch ?  How is that handled otherwise
on amd64 ?

Doug, given that you actually _have_ an amd64, do you feel like prodding
around on this?  I can send you a fresh debian/* tarball ...

Dirk
 
| Doug
| 
| > Dirk
| >
| > | The logs are available at
| > | http://amd64.ftbfs.de/build.php?arch=&pkg=r-base
| > |
| > | PS: This is actually from a log of the 2.2.0.beta.20050922-1
| > | version, but the 2.2.0.alpha.20050908-1 version looks
| > | about the same.
| > |
| > |
| > | Kurt
| > |
| >
| > --
| > Statistics: The (futile) attempt to offer certainty about uncertainty.
| >          -- Roger Koenker, 'Dictionary of Received Ideas of Statistics'
| >
| >
| >

-- 
Statistics: The (futile) attempt to offer certainty about uncertainty.
         -- Roger Koenker, 'Dictionary of Received Ideas of Statistics'


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to