[Phylobase-devl] Fwd: Re: CRAN packages phylobase and adephylo

Ben Bolker bbolker at gmail.com
Mon Nov 7 21:18:25 CET 2011

Hash: SHA1

  Does anyone out there have a Solaris machine?  I doubt it.

  I'm happy to global search-and-replace for the relevant functions and
prepend "std::" to them.  We could try this, submit to CRAN, and see if
it builds, but this seems like an inefficient way to debug.  Maybe I
will contact Brian Ripley (eek!) for advice.


On 11-11-06 12:58 PM, Mark Holder wrote:
> Hi, I don't have a Solaris machine to test on either.
> My guess:
> If the compiler on Solaris is putting all of the C-functions in C++'s
> "std"" namespace, then you would see something like this. Changing
> "sprintf" to "std::sprintf" might fix this (it looks like the std::
> namespace decorator will be needed on strcmp, strchr, and memmove at
> least, probably others). That should not break the builds on other
> platforms.
> Feel free to direct the Solaris user to me, and I'll try to help
> diagnose other issues with the NCL build.
> all the best, Mark
>  On Nov 4, 2011, at 7:35 PM, Ben Bolker wrote:
>> Hey guys,
>> Any opinions about this?
>> adephylo is not absolutely necessary to phylobase, but it bothers
>> me to spend effort removing it when we might be able to fix it
>> instead. Perhaps if Thibaut responds (the fix is fairly trivial)
>> and we then grovel to the R Gods/BDR ...
>> It may be necessary for political reasons to make these changes
>> and then revert them if/when we get adephylo fixed.
>> We have another issue -- the Solaris build (sigh) is failing, with
>> the following error:
>> "./ncl/nxsstring.h", line 467: Error: The function "sprintf" must
>> have a prototype.
>> Fixing this (by remote control) could be pretty challenging --
>> R-forge doesn't have a Solaris build, so it could involve more
>> grovelling to BDR for advice (another reason to temporarily do the
>> adephylo fix, unless we can get our act together quickly).
>> cheers Ben
>> -------- Original Message -------- Subject: Re: CRAN packages
>> phylobase and adephylo Date: Sun, 6 Nov 2011 16:50:41 +0000 (GMT) 
>> From: Prof Brian Ripley <ripley at stats.ox.ac.uk> To: Ben Bolker
>> <bbolker at gmail.com> CC: CRAN at r-project.org
>> On Fri, 4 Nov 2011, Ben Bolker wrote:
>>> On 11-11-04 04:31 AM, Prof Brian Ripley wrote:
>>>> Package adephylo has been archived: it was failing on R 2.14.0
>>>> and we have had no response to several messages to the
>>>> maintainer.
>>>> Package phylobase suggests adephylo, but still passes its
>>>> checks without it (and grepping the code suggests that it does
>>>> not actually use it). Nevertheless, the documentation needs
>>>> updating, so can we please have a revised version of
>>>> phylobase.
>>>> Brian Ripley
>>> I think I see the problem with adephylo -- I'm going to try to
>>> nag the maintainer (who I know personally) a bit more and see if
>>> we can get this resolved from that end.
>> It's far too late: adephylo has been archived.  His complete
>> failure to respond to messages means we may be reluctant to
>> re-admit it.
>> In any case, phylobase needs an update now.
>>> thanks Ben Bolker
>> -- 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 
>> <solaris_err.txt>_______________________________________________ 
>> 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
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/


More information about the Phylobase-devl mailing list