spkg-configure for R

[dimpase]

a minimum working on standard distros setup.

Note that R does not have a working uninstall, so one needs to manually uninstall by

rm -rf $SAGE_LOCAL/lib/R
rm $SAGE_LOCAL/bin/R*

followed by

./bootstrap
./configure
./sage -f rpy2
make build

and then test. (Or, of course start from make distclean)

Depends on #27870

CC: @sagetrac-tmonteil @kiwifb @isuruf @EmmanuelCharpentier @embray

Component: build: configure

Author: Dima Pasechnik

Branch/Commit: 1f4a7e1

Reviewer: Emmanuel Charpentier

Issue created by migration from https://trac.sagemath.org/ticket/28884

[dimpase]

comment:1

TODO: check that versions of R and libR match

[kiwifb]

comment:6

I guess we could check for matching versions with that kind of stuff

fbissey@moonloop ~ $ pkg-config --modversion libR
3.4.1
fbissey@moonloop ~ $ R --version
R version 3.4.1 (2017-06-30) -- "Single Candle"
Copyright (C) 2017 The R Foundation for Statistical Computing
Platform: x86_64-pc-linux-gnu (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under the terms of the
GNU General Public License versions 2 or 3.
For more information about these matters see
http://www.gnu.org/licenses/.

Nice that libR comes with a .pc file.

[dimpase]

comment:7

I'll do the TODO as soon as the rest looks OK. I've done similar checks for pari...

[EmmanuelCharpentier]

comment:8

Does this install on top of #27870, includes (i. e. subsumes) #27870, or what ?

[dimpase]

comment:10

#27870 should have been listed as a dependency, sorry. Anyway, the branch here includes the branch of #27870, so you just need the branch u/dimpase/packages/rconf

[dimpase]

Dependencies: #27870

[EmmanuelCharpentier]

comment:11

Replying to @dimpase:

#27870 should have been listed as a dependency, sorry. Anyway, the branch here includes the branch of #27870, so you just need the branch u/dimpase/packages/rconf

Thanks ! Looking into it...

[EmmanuelCharpentier]

comment:12

LGTM:

• Sage -R works as expected. Superficial testing is OK.

charpent@zen-book-flip:/usr/local/sage-9$ sage -t --long src/sage/interfaces/r.py 
Running doctests with ID 2019-12-15-10-50-43-b4fc9d4c.
Git branch: t/28884/packages/rconf
Using --optional=build,dochtml,dot2tex,fricas,gap_packages,giacpy_sage,libsemigroups,memlimit,python2,sage,sagenb
Doctesting 1 file.
sage -t --long --warn-long 184.3 src/sage/interfaces/r.py
    [257 tests, 5.48 s]
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 5.7 seconds
    cpu time: 5.3 seconds
    cumulative wall time: 5.5 seconds

charpent@zen-book-flip:/usr/local/sage-9$ sage -t --long src/sage/stats/r.py 
Running doctests with ID 2019-12-15-10-51-20-adfb7e0b.
Git branch: t/28884/packages/rconf
Using --optional=build,dochtml,dot2tex,fricas,gap_packages,giacpy_sage,libsemigroups,memlimit,python2,sage,sagenb
Doctesting 1 file.
sage -t --long --warn-long 184.3 src/sage/stats/r.py
    [2 tests, 0.54 s]
----------------------------------------------------------------------
All tests passed!
----------------------------------------------------------------------
Total time for all tests: 0.7 seconds
    cpu time: 0.5 seconds
    cumulative wall time: 0.5 seconds

make ptestalllong underway...

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. This was a forced push. New commits:

e5c91db	update pkgconfig to v1.5.1. See trac #28883
1720b38	a bare minumim openblas config
925266b	use openblas.pc for blas and lapack, don't require atlas is to be used
b89df80	move blas handling to build/pkgs/
6341576	create cblas.pc link, too, not only blas and lapack
f66ea24	spkg-configure for R

[sagetrac-git]

Changed commit from bbea194 to f66ea24

[dimpase]

comment:14

added forgotten commit sagemath/sagetrac-mirror@6341576 from #27870 to this branch (should not matter, review-wise, I guess)

[EmmanuelCharpentier]

comment:16

Replying to @EmmanuelCharpentier:

[ Snip... ]

make ptestalllong underway...

ptestalllong gets 6 transient and 3 permanent failures :

File	Result	P/T
src/sage/plot/animate.py	7 doctests failed	T
src/sage/modular/ssmod/ssmod.py	2 doctests failed	T
src/sage/misc/latex.py	1 doctest failed	T
src/sage/numerical/backends/glpk_backend.pyx	1 doctest failed	P
src/sage/repl/load.py	1 doctest failed	T
src/sage/tests/gap_packages.py	1 doctest failed	P
src/sage/libs/glpk/error.pyx	1 doctest failed	P
src/sage/combinat/designs/ext_rep.py	1 doctest failed	T
src/sage/databases/findstat.py	1 doctest failed	T

All of these have already been reported for previous release(s) of Python 3-based Sage 9 betas, and therefore considered not to be related to this ticket.

==> positive_review

Given the importance of this ticket (a large headache off the shoulders of active Sage-R users...), I recommend tests on other platforms (Mac OS X and its shenanigans, Windows and the coexistence of Windows and Cygwin libraries...).

[EmmanuelCharpentier]

Reviewer: Emmanuel Charpentier

[EmmanuelCharpentier]

comment:17

Replying to @dimpase:

added forgotten commit sagemath/sagetrac-mirror@6341576 from #27870 to this branch (should not matter, review-wise, I guess)

Aaaarghhhh...

Another check may be worthy. But maybe after tests on other platforms... Please ask as neeeded.

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1 and set ticket back to needs_review. New commits:

3812f1c

list more relevant system packages to install on Debian/Fedora

[sagetrac-git]

Changed commit from f66ea24 to 3812f1c

[EmmanuelCharpentier]

comment:19

Replying to @sagetrac-git:

Branch pushed to git repo; I updated commit sha1 and set ticket back to needs_review. New commits:

3812f1c

list more relevant system packages to install on Debian/Fedora

Similar results on this second pass. positive review. Same recommendation as before.

[dimpase]

comment:23

It works on MacOS 10.13.6 with Homebrew's R and openblas packages.

It doesn't work with R installed with standalone isntaller from R Project,
probably as there is some kind of library/compiler incompatibility - they don't build R with Xcode compiler, they use their "own"; more precisely, OpenMP-enabled
clang 8, see https://cran.r-project.org/bin/macosx/tools.
Somehow I see an error message while building rpy2, attempting to use -fopenmp switch with Xcode's clang.

[vbraun]

Changed branch from u/dimpase/packages/rconf to 6aba5b4

[vbraun]

Changed commit from 6aba5b4 to none

[vbraun]

comment:25

More stuff that has #28883 merged in

[vbraun]

Commit: 6aba5b4

[vbraun]

New commits:

e5c91db	update pkgconfig to v1.5.1. See trac #28883
1720b38	a bare minumim openblas config
925266b	use openblas.pc for blas and lapack, don't require atlas is to be used
b89df80	move blas handling to build/pkgs/
6341576	create cblas.pc link, too, not only blas and lapack
f66ea24	spkg-configure for R
3812f1c	list more relevant system packages to install on Debian/Fedora
6aba5b4	lower versions of R and openblas

[vbraun]

Changed branch from 6aba5b4 to u/dimpase/packages/rconf

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1 and set ticket back to needs_review. This was a forced push. New commits:

f74f554	update pkgconfig to v1.5.1. See trac #28883
fb8f4e5	allow absent zlib.pc
b3a8891	a bare minumim openblas config
c031c9e	use openblas.pc for blas and lapack, don't require atlas is to be used
b49f79e	move blas handling to build/pkgs/
81bf522	create cblas.pc link, too, not only blas and lapack
6df5985	spkg-configure for R
0bad8b4	list more relevant system packages to install on Debian/Fedora
1f4a7e1	lower versions of R and openblas

[sagetrac-git]

Changed commit from 6aba5b4 to 1f4a7e1

[dimpase]

comment:30

routine rebase over updated #27870

[vbraun]

Changed branch from u/dimpase/packages/rconf to 1f4a7e1