[gentoo-science] Errors installing Rcpp using R_Overlay

[Neil Shephard]

Hi,

Delighted to discover the R_Overlay that allows package installation from
CRAN under portage after a posting by maintainer Benda Xu a few weeks ago (
http://gentoo.2317880.n4.nabble.com/Re-new-eclass-to-install-R-modules-such-as-Rcpp-td331686.html
).

Found the time to sit down and start migrating/installing packages via the
overlay and decided to start by installing the tidyverse package as it
would pull in a host of dependencies and keep /var/lib/portage/world a bit
smaller (it pulls in 60 packages in total).

Unfortunately Rcpp fails to build and so the vast majority (n = 56) pulled
in when emerge tidyverse fail to install.

The build log is...

# cat /var/tmp/portage/sci-CRAN/Rcpp-0.12.15/temp/build.log
 * Package:    sci-CRAN/Rcpp-0.12.15
 * Repository: R_Overlay
 * USE:        abi_x86_64 amd64 elibc_glibc kernel_linux userland_GNU
 * FEATURES:   preserve-libs sandbox userpriv usersandbox
* installing *source* package ‘Rcpp’ ...
** package ‘Rcpp’ successfully unpacked and MD5 sums checked
** libs
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c Date.cpp -o Date.o
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c Module.cpp -o
Module.o
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c Rcpp_init.cpp -o
Rcpp_init.o
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c api.cpp -o api.o
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c attributes.cpp -o
attributes.o
x86_64-pc-linux-gnu-g++  -I/usr/lib64/R/include -DNDEBUG
-I../inst/include/     -fpic  -march=skylake -O2 -pipe -c barrier.cpp -o
barrier.o
x86_64-pc-linux-gnu-g++ -shared -L/usr/lib64/R/lib -Wl,-O1 -Wl,--as-needed
-o Rcpp.so Date.o Module.o Rcpp_init.o api.o attributes.o barrier.o
-L/usr/lib64/R/lib -lR
installing to /var/tmp/portage/sci-CRAN/Rcpp-0.12.15/work/Rcpp/libs
** R
** inst
** preparing package for lazy loading
** help
*** installing help indices
** building package indices
** installing vignettes
** testing if installed package can be loaded
* DONE (Rcpp)
Traceback (most recent call last):
  File "/usr/lib/portage/python3.5/doins.py", line 611, in <module>
    sys.exit(main(sys.argv[1:]))
  File "/usr/lib/portage/python3.5/doins.py", line 602, in main
    os.path.dirname(source)):
  File "/usr/lib/portage/python3.5/doins.py", line 450, in _doins
    return install_runner.install_file(source, os.path.dirname(dest))
  File "/usr/lib/portage/python3.5/doins.py", line 386, in install_file
    return self._ins_runner.run(source, dest_dir)
  File "/usr/lib/portage/python3.5/doins.py", line 195, in run
    sstat = os.stat(source)
FileNotFoundError: [Errno 2] No such file or directory:
b'/var/tmp/portage/sci-CRAN/Rcpp-0.12.15/work/Rcppy'
 * ERROR: sci-CRAN/Rcpp-0.12.15::R_Overlay failed (install phase):
 *   doins failed
 *
 * If you need support, post the output of `emerge --info
'=sci-CRAN/Rcpp-0.12.15::R_Overlay'`,
 * the complete build log and the output of `emerge -pqv
'=sci-CRAN/Rcpp-0.12.15::R_Overlay'`.
 * The complete build log is located at
'/var/tmp/portage/sci-CRAN/Rcpp-0.12.15/temp/build.log'.
 * The ebuild environment file is located at
'/var/tmp/portage/sci-CRAN/Rcpp-0.12.15/temp/environment'.
 * Working directory:
'/var/tmp/portage/sci-CRAN/Rcpp-0.12.15/work/Rcpp-0.12.15'
 * S: '/var/tmp/portage/sci-CRAN/Rcpp-0.12.15/work/Rcpp-0.12.15'


And I've placed the other information on pastebin...

#  emerge --info '=sci-CRAN/Rcpp-0.12.15::R_Overlay' | pastebinit
https://pastebin.com/mgrD6iJR

#  emerge -pqv '=sci-CRAN/Rcpp-0.12.15::R_Overlay' | pastebinit
https://pastebin.com/wUHkan1B


Any thoughts on how to solve this?  It looks like Rcpp builds ok within R
and the error arises when control is passed back to Portage.

I figured gentoo-science@l.g.o would be appropriate given
Benda's recent post but apologies if this is not the case.


Neil

--
Natural selection is a mechanism for generating an exceedingly high degree
of improbability. - R.A. Fisher


-----BEGIN PGP PUBLIC KEY BLOCK-----
Version: Mailvelope 0.11.0
Comment: Email security by Mailvelope - https://www.mailvelope.com
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=9T81
-----END PGP PUBLIC KEY BLOCK-----