[Phylobase-devl] New phylobase build approach using static libncl (Was: Rcpp and OS X compiliation)

[François Michonneau]

Hello,

  Thanks Mark for your spending the time to help to resolve this problem!

  I checked a few things and it seemed to work fine, so I committed
Mark's code. The error message that Mark was referring to is caused by
trees output by MrBayes which most likely will contain several nodes
with the same posterior probability. In our current implementation we
don't allow trees to have identical node labels. It's however possible
to export these posterior probabilities as data using the option
check.node.labels="asdata". Over spring break, I will spend some time
on phylobase, and in particular I will write unit tests for readNexus
(and hopefully finish to deal with allowing non unique labels).

  Cheers,
  -- François



On Tue, Mar 2, 2010 at 16:46, Mark Holder <mtholder at gmail.com> wrote:
> Hi again,
>
> Mark Holder and then Peter Cowan wrote:
>>>
>>> I should be able to fix that aspect of things tomorrow.  When I push
>>> those changes to NCL, I'll post to this list so that someone with write
>>> permissions to phylobase's svn repo can copy those changes into the
>>> phylobase.
>>
>> I'm happy to do this.
>
>
> At http://people.ku.edu/~mtholder/mth_diff_pkg.tar.gz you'll find a tar.gz
> archive.  If you unpack it, you'll see the contents of phylobase's pkg
> directory with the edits that I had to make to get things working. Several
> files were touched (the archive has all of the pkg if I run svn stat I see:
>
> M       pkg/src/NCLInterface.h
> M       pkg/src/ReadWithNCL.cpp
> M       pkg/src/nxspublicblocks.cpp
> M       pkg/src/nxscharactersblock.cpp
> M       pkg/src/ncl/nxsreader.h
> M       pkg/src/nxsreader.cpp
> M       pkg/src/NCLInterface.cpp
> M       pkg/inst/nexusfiles/treepluscharV01.nex
>
>
>
> Note that pkg/inst/nexusfiles/treepluscharV01.nex  is not a valid NEXUS file
> if you change line 74 from:
>
>        BEGIN CHARACTERS2;
> to
>
>        BEGIN CHARACTERS;
> then it should work.
>
> Also note that checkTree(object) in  checkdata.R is complaining about the
> data returned from pkg/inst/nexusfiles/co1.nex, but I don't know why.  It
> looks OK to me.  It is possible that I broke something, as I'm not too
> familiar with what was supposed to be returned by NCL (I tried to return the
> same syntax that was originally being returned).
>
>
>
>
>
>>>> Lastly, is there a way to control the output though Rcpp or otherwise, a
>>>> fair bit of what appears to be stderr gets printed in the R console, it'd be
>>>> nice to control this.
>>>
>>> The BASICCMDLINE implementation writes lots of status output to the
>>> standard error stream.  When I work with it tomorrow, I can make its
>>> chattiness controllable by an argument.
>>
>> That would be wonderful!
>
> I'm partially there. For the NCLInterface.cpp code used by phylobase I
> changed it so that it is quiet by default.  It is now be possible to send a
> numeric argument into BASICCMDLINE.Initialize();
>
> Currently (in ReadWithNCL.cpp) it says,
>        reader.Initialize(const_cast < char* > (filename.c_str()));
>
> If you change it to
>        reader.Initialize(const_cast < char* > (filename.c_str()), 0);
>
> you'll see more messages to stderr.  The numbers 0-7 generate less and less
> output.  8 or higher should be silent (which is the default).
>
>
> I did not add the hooks in ReadWithNCL.cpp to get a verbosity argument from
> R and pass that along to NCL.
>
>
>
>
>
>
>> If you have time, I'd like to have a more in depth conversation of what
>> NCL is capable and how it is organized.  Wrapping NCL is, in my opinion, one
>> of the key features of phylobase.  But, I also get the feeling that there is
>> more functionality we aren't taking advantage of.
>>
>> Perhaps a GSOC student project could come out of it.  Or at the very least
>> I can expand the developers guide section about NCL so we are better able to
>> maintain it in the future.
>
>
>
> The NCL docs are terribly out of date. Completely my fault. Paul Lewis is
> great about documenting his code, but I have not kept up with the
> documenting tasks.  I'm happy to chat.
>
> The main potential problems that I see with the ways that phylobase is using
> NCL now are:
>        1. in NCLInterface.cpp there are lots of call to
> RemoveUnderscoresAndSpaces to get rid of spaces and _ in names.  That makes
> names easier to deal with, but at some point will bite you (somebody will
> have dataset with a taxon labelled "AB" and another with "A B", after
> transformation there will be a name clash).
>
>        2. It was not clear to me how different character blocks should
> separated.  I just return the union of all character matrices.  It seems
> like, phylobase will need a richer interaction with NCL if phylobase wants
> to know about whether the data came in under different blocks.
>
>
>
> all the best,
> Mark
>
>
>
>
>
> Mark Holder
>
> mtholder at ku.edu
> http://phylo.bio.ku.edu/mark-holder
>
> ==============================================
> Department of Ecology and Evolutionary Biology
> University of Kansas
> 6031 Haworth Hall
> 1200 Sunnyside Avenue
> Lawrence, Kansas 66045
>
> lab phone:  785.864.5789
>
> fax (shared): 785.864.5860
> ==============================================
>
>
>
>
>
>
>
>
>