[Phylobase-devl] Fwd: CRAN package phylobase

[François Michonneau]

Hello all,

  Over the last few weeks, I have committed some changes to phylobase. Most
have been brewing on my computer for quite some time and were almost ready
for the outside world. The main change is that most of the testing about
the validity of phylo4(d) objects is now done in C++. I was hoping it would
make things faster. It does, but I'm not sure by how much at this stage (on
some tests it was 10x faster on larger trees).

  I have also started to convert the unit tests to the test_that package,
as ultimately I'd like to make phylobase compatible with the devtools
package.

  I don't understand or know how to emulate a sparc environment on my
computer to reproduce the bug mentioned and see what can be done about it.
The test that causes the bug is among the ones that I converted to
test_that. If we are lucky that might be all that's needed to make it pass
the test (i.e., it was a bug with RUnit). If not, I'll try harder to
reproduce the bug that is causing the check to fail on the sparc
architecture. In the worst case, we can always omit this test from the
suite.

  If you have some time, please report if this updated version (0.6.7,
r862) passes the R CMD check on your system.

  Thanks,
  -- François



On Sun, Mar 9, 2014 at 9:38 AM, Ben Bolker <bbolker at gmail.com> wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
>   Here we go ...
>
> - -------- Original Message --------
> Subject: CRAN package phylobase
> Resent-Date: Sun, 09 Mar 2014 08:46:51 -0400
> Date: Sun, 09 Mar 2014 12:45:47 +0000
> From: Prof Brian Ripley <ripley at stats.ox.ac.uk>
> To: Ben Bolker <bolker at mcmaster.ca>, CRAN <CRAN at r-project.org>
>
> This applies also to phylobase.
>
>   Running ‘doRUnit.R’
>  ERROR
> Running the tests in ‘tests/doRUnit.R’ failed.
> Last 13 lines of output:
>   Number of test functions: 99
>   Number of errors: 0
>   Number of failures: 1
>
>
>   1 Test Suite :
>   phylobase unit testing - 99 test functions, 0 errors, 1 failure
>   FAILURE in test.nodeDepth.phylo4: Error in
> checkIdentical(nodeDepth(phy.alt), allDepths) : FALSE
>   Error:
>
>   unit testing failed (#test failures: 1, #R errors: 0)
>
>   Execution halted
>
>
> - -------- Original Message --------
> Subject: CRAN packages EBS MPSEM chgnpt emplik
> Date: Wed, 05 Mar 2014 09:45:11 +0000
> From: Prof Brian Ripley <ripley at stats.ox.ac.uk>
> To: Alice Cleynen <alice.cleynen at agroparistech.fr>,        Guillaume
> Guenard <guillaume.guenard at gmail.com>,        Youyi Fong
> <yfong at fhcrc.org>, Mai Zhou <mai at ms.uky.edu>
> CC: CRAN <CRAN at r-project.org>
>
> These have long failed their tests on Sparc Solaris.  There are two main
> differences from the common R platforms:
>
> 1) It is big-endian.
>
> 2) It does not have extended-precision hardware.  You can emulate that
> on other platforms: see
>
> http://cran.r-project.org/doc/manuals/r-devel/R-exts.html#Writing-portable-packages
>
>
> .
>
> There are other platforms with each of these characteristics, but not in
> the check farm.
>
> I was able to reproduce the errors by emulation on x86_64 Linux.
>
> Please submit corrected updates before Mar 21 if you want to retain the
> packages on CRAN.
>
> - --
> Brian D. Ripley,                  ripley at stats.ox.ac.uk
> Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
> University of Oxford,             Tel:  +44 1865 272861 (self)
> 1 South Parks Road,                     +44 1865 272866 (PA)
> Oxford OX1 3TG, UK                Fax:  +44 1865 272595
>
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.11 (GNU/Linux)
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
>
> iQEcBAEBAgAGBQJTHG7KAAoJEOCV5YRblxUHWUQH/jVEpE6L/gEBxSFwByJdaJU+
> DzgxEy3NJKIkTYRQmj5KrhWSsO8V8VnK5X32+6uyMUeB+55JCWuDMa8+iSXOP2++
> tDsJBFwufpvJ3POOQGUjRJV4jSOmpuqBCjnnmijzKv8aKl4lUq+Y7/281MqeLyrh
> t7/X3fwLElKjWNhAqn6PhSrer3GxzRP7YDRFx1fl+aM1W0MFnWBnqsxNwMnEQ0RU
> 6S1VwrlK+tYhscVPcC+xjbqqXRrDOtnXvUUs2oFrUuXHLs+0iMnfiRjUjt5MVF/R
> f+0JuVBovVWqQmyNDyiiE6FFPMTtyT5ZYwWItN/ikxKiEGg9kD5hPo7TD89/n8A=
> =XCil
> -----END PGP SIGNATURE-----
> _______________________________________________
> Phylobase-devl mailing list
> Phylobase-devl at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/phylobase-devl
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/phylobase-devl/attachments/20140310/8b2eb89f/attachment.html>