[BioC] Rgraphviz, R 2.7.0 and Windows XP

Kort, Eric Eric.Kort at vai.org
Tue Jun 3 21:09:31 CEST 2008


I'm even more reluctant to chime in, but...

There are enough variables here that it seems a case could be made for making the graphviz dlls/shared libraries compatible with the BioC binary packages of Rgraphviz available on an applicable website to be copied into the lib directory of Rgraphviz.  Since graphviz is freely redistributable, this would simplify things for some users of the binary packages.

There are reasons for not doing this, of course.  But it seems likely that this is going to become/continue to be a recurrent theme without the shortcut.

-Eric

> -----Original Message-----
> From: bioconductor-bounces at stat.math.ethz.ch
> [mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of
> Tony Chiang
> Sent: Tuesday, June 03, 2008 2:58 PM
> To: Martin Morgan
> Cc: Jenny Drnevich; Myroslav Sypa; bioconductor at stat.math.ethz.ch
> Subject: Re: [BioC] Rgraphviz, R 2.7.0 and Windows XP
>
>
> Hi Martin et al,
>
> On Tue, Jun 3, 2008 at 7:36 PM, Martin Morgan
> <mtmorgan at fhcrc.org> wrote:
>
> > I'm a little reluctant to add my comments here, which seem
> to miss the
> > mark more often than not on this, but..
> >
> > Jenny Drnevich wrote:
> >
> >> Hi Tony,
> >>
> >> Actually, I have NOT been able to get Rgraphviz to work
> with R 2.7.0,
> >> and I've tried two different Windows machines. I'm not sure the
> >> problem completely lies with Rgraphviz, but maybe with Graphviz as
> >> well. Even though I want to use the Rgraphviz Windows
> binary instead
> >> of building Rgraphviz, I followed the README file found in the
> >> Rgraphviz_1.18.1.tar.gz source file:
> >>
> >> 1. I'm not sure how to "make sure your MSVCR80.DLL is from
> Microsoft
> >> Visual C++ 2005 SP1 Redistributable Package", but I downloaded and
> >> installed the package from the link provided, following all their
> >> instructions and warnings.
> >>
> >> 2. I downloaded and installed the
> graphviz-2.16.1.static.exe binary
> >> from Graphviz.org, which is the only 2.16 version available.
> >>
> >
> > Actually, you don't want this static version. Instead go
> looking here
> >
> > http://www.graphviz.org/pub/graphviz/ARCHIVE/
> >
> > Make sure to clean up as completely as possible your previous
> > installation.
> >
> > I *think*, contrary to previous posts of mine, that you can
> also use
> > the recent graphviz (2.18), so long as you're building Rgraphviz
> > yourself.
> >
>
> At least on my Macbook pro, this is true (at least it works
> for me without failing). Jenny, if you use biocLite, make
> sure that you set the type="source" to build it from source.
>
>
> >
> > Once built, your Rgraphviz will still have missing
> functionality, in
> > particular it will be unable to read / write graph objects using
> > agopen, agwrite, etc. This seems to be a complex issue in the
> > interaction between R and graphviz, tracing I think to the use of
> > different development tools for the two project. There might be
> > similar surprises in other functions; these are just the
> ones that the
> > Bioc build machine stumbles on when checking the package. You'll
> > encounter missing functionality in a most unpleasant way, abruptly
> > aborting your session.
> >
>
> I don't have access to a windows machine at the moment, but
> this also happened on my mac a few months back (always with
> the agopen and agwrite). After I grab the lastest X-code
> (with the latest compliers) and built both graphviz and
> Rgraphviz from source on the (also following Martin's advice
> to make sure that all previous incarnations of graphviz were
> completely removed before building graphviz again), this
> problem seems to have gone away.
>
>
> >
> > I think the rationale for not updating the instructions in the
> > package, or making some functionality 'optional' on windows, is the
> > hope for a more robust solution.
> >
> > Again I seem to have a poor track record on providing reliable
> > information about this, so please take with a grain of salt.
> >
>
> I should say ditto for me too now!
>
>
> >
> > Martin
> >
> >
> >  3. I created the new user variables as indicated, and made sure the
> >> C:\Program Files\graphviz-2.16\bin was added to my path
> (it was done
> >> automatically)
> >>
> >> 4. Then I used biocLite() to install Rgraphviz.
> >>
> >> When I tried to load Rgraphviz, I got both a Windows error and an
> >> error within R; The Windows error message is known to Graphviz
> >> because it is mentioned on the download page: "The application has
> >> failed to start because somefilename.dll was not found.
> Re-installing
> >> the application may fix this problem."  Graphviz says to install
> >> Microsoft Visual C++ 2005 Redistributable Package (x86) to fix it,
> >> but I've already done that!
> >>
> >> The R error message is:
> >>  > library(Rgraphviz)
> >> Loading required package: grid
> >> Error in inDL(x, as.logical(local), as.logical(now), ...)
> :  unable
> >> to load shared library
> >> 'C:/PROGRA~1/R/R-27~1.0/library/Rgraphviz/libs/Rgraphviz.dll':
> >>  LoadLibrary failure:  The specified module could not be found.
> >>
> >>
> >> Error : .onLoad failed in 'loadNamespace' for 'Rgraphviz'
> >> Error: package/namespace load failed for 'Rgraphviz'
> >>
> >>
> >> Finally, I finished the directions in the README file to
> safely load
> >> Rgraphviz but I got another error message:
> >>
> >>  > dyn.load("C:/Program Files/graphviz-2.16/bin/gvc.dll") Error in
> >> inDL(x, as.logical(local), as.logical(now), ...) :  unable to load
> >> shared library 'C:/Program
> >> Files/graphviz-2.16/bin/gvc.dll':
> >>  LoadLibrary failure:  The specified module could not be found.
> >>
> >>
> >> So is this gvc.dll file missing from Graphviz? Or would building
> >> Rgraphviz from source have created it? I'm still having
> the general
> >> Windows problem related to the Microsoft Visual C++ 2005
> >> Redistributable Package (x86), which I have no idea how to
> fix. I'd
> >> be very interested in hearing if anyone has gotten the Rgraphviz
> >> Windows _binary_ working with R 2.7.0 and how they did it.
> If someone
> >> knows what I'm doing wrong or have other suggestions to
> try, please
> >> let me now! Overall, I doubt that a generic/novice Windows R user
> >> could be able to get Rgraphviz to work...
> >>
> >> Thanks,
> >> Jenny
> >>
> >>  > sessionInfo()
> >> R version 2.7.0 (2008-04-22)
> >> i386-pc-mingw32
> >>
> >> locale:
> >> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> >> States.1252;LC_MONETARY=English_United
> >> States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
> >>
> >> attached base packages:
> >>  [1] grid      splines   tools     stats     graphics
> grDevices utils
> >>  [8] datasets  methods   base
> >>
> >> other attached packages:
> >>  [1] affyQCReport_1.18.0  geneplotter_1.18.0   lattice_0.17-8
> >>  [4] RColorBrewer_1.0-2   simpleaffy_2.16.0    made4_1.14.0
> >>  [7] ade4_1.4-7           affyPLM_1.16.0       affycoretools_1.12.0
> >> [10] annaffy_1.12.0       KEGG.db_2.2.0        gcrma_2.12.0
> >> [13] matchprobes_1.12.0   biomaRt_1.14.0       RCurl_0.8-1
> >> [16] GOstats_2.6.0        Category_2.6.0       genefilter_1.20.0
> >> [19] survival_2.34-1      RBGL_1.16.0          annotate_1.18.0
> >> [22] xtable_1.5-2         GO.db_2.2.0          AnnotationDbi_1.2.0
> >> [25] RSQLite_0.6-8        DBI_0.2-4            graph_1.18.1
> >> [28] limma_2.14.1         affy_1.18.0          preprocessCore_1.2.0
> >> [31] affyio_1.8.0         Biobase_2.0.1        RWinEdt_1.8-0
> >>
> >> loaded via a namespace (and not attached):
> >> [1] cluster_1.11.10    KernSmooth_2.22-22 XML_1.94-0.1
> >>  >
> >>
> >>
> >>
> >>
> >> At 10:39 AM 6/3/2008, Tony Chiang wrote:
> >>
> >>> I am sure that there are plenty of folks with Windows XP
> and R-2.7.0
> >>> who have gotten Rgraphviz to work. Unless you let us who
> what seems
> >>> to be the problem on your end, I would suggest reading the large
> >>> number of e-mails from the archives that deals with the different
> >>> issues that have come up...and maybe it might solve your problems.
> >>>
> >>> tony
> >>>
> >>> On Tue, Jun 3, 2008 at 4:09 PM, Myroslav Sypa
> <msypa at email.unc.edu>
> >>> wrote:
> >>>
> >>> > Hello all,
> >>> >
> >>> > I'm creating an application which use rJava to connect
> with R. But
> >>> > my question is about Rgraphviz. Could you tell me if Rgraphviz
> >>> > will (in
> >>> nearest
> >>> > future) work with R 2.7.0 as it has been working with R
> 2.6.*? I
> >>> > mean
> >>> that
> >>> > for installing Rgraphviz I need just run these two strings in R:
> >>> >
> >>> > /source("http://bioconductor.org/biocLite.R")
> >>> > biocLite("Rgraphviz")
> >>> >
> >>> > /?/
> >>> > /
> >>> >
> >>> > --
> >>> > Best regards, Myroslav
> >>> >
> >>> > _______________________________________________
> >>> > Bioconductor mailing list
> >>> > Bioconductor at stat.math.ethz.ch
> >>> > https://stat.ethz.ch/mailman/listinfo/bioconductor
> >>> > Search the archives:
> >>> >
> http://news.gmane.org/gmane.science.biology.informatics.conductor
> >>> >
> >>>
> >>>        [[alternative HTML version deleted]]
> >>>
> >>> _______________________________________________
> >>> Bioconductor mailing list
> >>> Bioconductor at stat.math.ethz.ch
> >>> https://stat.ethz.ch/mailman/listinfo/bioconductor
> >>> Search the archives:
> >>> http://news.gmane.org/gmane.science.biology.informatics.conductor
> >>>
> >>
> >> Jenny Drnevich, Ph.D.
> >>
> >> Functional Genomics Bioinformatics Specialist
> >> W.M. Keck Center for Comparative and Functional Genomics
> >> Roy J. Carver Biotechnology Center
> >> University of Illinois, Urbana-Champaign
> >>
> >> 330 ERML
> >> 1201 W. Gregory Dr.
> >> Urbana, IL 61801
> >> USA
> >>
> >> ph: 217-244-7355
> >> fax: 217-265-5066
> >> e-mail: drnevich at illinois.edu
> >>
> >> _______________________________________________
> >> Bioconductor mailing list
> >> Bioconductor at stat.math.ethz.ch
> >> https://stat.ethz.ch/mailman/listinfo/bioconductor
> >> Search the archives:
> >> http://news.gmane.org/gmane.science.biology.informatics.conductor
> >>
> >
> >
> > --
> > Martin Morgan
> > Computational Biology / Fred Hutchinson Cancer Research Center 1100
> > Fairview Ave. N. PO Box 19024 Seattle, WA 98109
> >
> > Location: Arnold Building M2 B169
> > Phone: (206) 667-2793
> >
>
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives:
> http://news.gmane.org/gmane.science.biology.informatics.conductor
>
This email message, including any attachments, is for th...{{dropped:6}}



More information about the Bioconductor mailing list