[Bioc-devel] Harman not building on Windows due to mzR

[Hervé Pagès]

Hi Jason,

On 10/05/2016 05:08 AM, Jason.Ross at csiro.au wrote:
> Hi Steffen,
>
> I also develop on Linux, so I'm at a loss with compiling Bioconductor libs on Windows.
>
> As a guess, if there are no compiler errors but 'mzR.dll' cannot be found, I suppose the final linking of the compiled objects failed?

The error message is confusing: mzR.dll is actually here but it cannot
be loaded.

The man page for dyn.load (src/library/base/man/dynload.Rd) contains:

     Users have been confused by messages like
   \preformatted{
    error: unable to load shared object
    '.../library/rJava/libs/x64/rJava.dll':
    LoadLibrary failure:  The specified module could not be found.
   }
     The final line is a Windows (not \R) diagnostic: the \sQuote{module}
     that could not be found is not \file{rJava.dll} but something else
     Windows is looking for (here most likely Java DLLs): if you are lucky
     there will be a dialog box with more details.

Note that you don't see this when you open the man page on a
non-Windows system because this part of the man page is in a
#ifdef windows ... #endif directive.

Cheers,
H.

>
> I did a quick search and there seems to be a precedence for incompatibility between mzR and certain Rcpp versions: http://stackoverflow.com/questions/11181691/r-bioconductor-mzr-library-load-error
>
> Also on another tack, the line that seems to be doing the linking (prior to "installing to C:/Users/biocbuild/bbs-3.4-bioc/meat/mzR.buildbin-libdir/mzR/libs/i386") is fairly long. Has it has been clipped? Is it overflowing a shell variable? Windows has a very low command line limit of 8192 characters. By comparison, a typical Linux box has an arg limit of 2,097,152 chars (which can be found with `echo $(getconf ARG_MAX)`). https://mcuoneclipse.com/2015/03/29/solving-the-8192-character-command-line-limit-on-windows/
>
> Maybe overflowing is not the case? I expect it would throw an error and that line is only 3,845 chars in length.
>
> Anyway a couple of shots in the dark...
>
> Cheers,
> Jason
>
>
> ________________________________________
> From: Steffen Neumann <sneumann at ipb-halle.de>
> Sent: Wednesday, 5 October 2016 5:28 PM
> To: Ross, Jason (H&B, North Ryde); bioc-devel at r-project.org
> Subject: Re: [Bioc-devel] Harman not building on Windows due to mzR
>
> Hi, Jason,
>
> On Mi, 2016-10-05 at 03:40 +0000, Jason.Ross at csiro.au wrote:
>>
>> [...]
>> This is arising from the vignette and is due to the package msmsEDA
>> not being present. If I keep peeling the dependencies back, I get to
>> mzR, which has trouble loading the Windows dynamic link library. See:
>> http://bioconductor.org/checkResults/3.4/bioc-LATEST/mzR/tokay1-
>> install.html
>
> Together with the BioC team we're currently trying to find out
> what's going on there, we have the problem at least since the move
> to the new Tokay1 build host. Any help on the issue of "LoadLibrary
> failure:  The specified module could not be found." is highly welcome.
>
> On Linux I'd have ideas how to dig into the problem,
> and I'd be happy to try suggestions what to try next to fix
> the windows build.
>
> Sorry for the hassle,
> yours,
> Steffen
>
>
> --
> IPB Halle                    AG Massenspektrometrie & Bioinformatik
> Dr. Steffen Neumann          http://www.IPB-Halle.DE
> Weinberg 3                   Tel. +49 (0) 345 5582 - 1470
> 06120 Halle                       +49 (0) 345 5582 - 0
> sneumann(at)IPB-Halle.DE     Fax. +49 (0) 345 5582 - 1409
>
>
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>

-- 
Hervé Pagès

Program in Computational Biology
Division of Public Health Sciences
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N, M1-B514
P.O. Box 19024
Seattle, WA 98109-1024

E-mail: hpages at fredhutch.org
Phone:  (206) 667-5791
Fax:    (206) 667-1319