[R] sas.get problem : the saga continues.

[John Kane]

--- "Nordlund, Dan (DSHS/RDA)" <NordlDJ at dshs.wa.gov>
wrote:

> > -----Original Message-----
> > From: r-help-bounces at stat.math.ethz.ch 
> > [mailto:r-help-bounces at stat.math.ethz.ch] On
> Behalf Of John Kane
> > Sent: Friday, April 13, 2007 1:37 PM
> > To: R R-help
> > Subject: Re: [R] sas.get problem : the saga
> continues.
> > 
> > 
> > #Continuing my sas.get saga.  Hmisc has been been
> > updated however I am still
> > getting the same problem. I seem to have aa SAS
> > problem  as well as
> >  an R problem.  There seems to be something dodgy
> in
> > the SAS data or
> > format files since SAS will not read the file
> unless I
> > use
> >  LIBNAME library = 'F:\testsas'
> >  Something like
> >  LIBNAME dd1 = 'F:\testsas'
> >  will not work and it should whereas Libname
> > dd1="C:\Temp\testsas"  does.
> >  
> > However I still cannot get sas.get to read a
> simple
> > file which does work with
> > LIBNAME dd1 = 'F:\testsas' .  Can anyone see if I
> am
> > misreading the Help or
> > making some kind of stupid typing mistake? I also
> have
> > tried running it on the
> > C drive with no luck. Thanks
> > 
> > Windows XP, R 2.4.1
> > 
> >  ###### EXAMPLE  #########
> >  #### SAS PROGRAM for simple file.  ###
> > libname  LIBRARY 'F:\testsas';
> > proc format library= LIBRARY;
> > value catfmt 1 = "Siam"
> >              2 = "Persian"
> >              3 = "Cougar";
> > value dogfmt 1 = "Lab"
> >              2 = "Collie"
> >              3 = "Coyote";
> > run;
> > data library.animals;
> >    input cat dog;
> >    datalines;
> > 1 3
> > 2 2
> > 3 1
> > ;
> > 
> > Data LIBRARY.doms;
> >  set LIBRARY.Animals ;
> >  format cat catfmt. dog  dogfmt.;
> >  proc freq;
> >  run;
> >  
> > #####
> > #   Apply sas.get
> > 
> > library(Hmisc)
> > mydata <- sas.get(library="F:/testsas",
> mem="doms",
> >              format.library="F:/testsas",
> >   sasprog = '"C:/Program Files/SAS/SAS
> 9.1/sas.exe"')
> > 
> > RESULTS
> > 
> > > library(Hmisc)
> > > mydata <- sas.get(library="C:/Temp/testsas",
> > mem="doms",
> > +              format.library="C:/Temp/testsas",
> > +   sasprog = '"C:/Program Files/SAS/SAS
> > 9.1/sas.exe"')
> > The filename, directory name, or volume label
> syntax
> > is incorrect.
> > Error in sas.get(library = "C:/Temp/testsas", mem
> =
> > "doms", format.library = "C:/Temp/testsas",  :
> >         SAS job failed with status 1
> > In addition: Warning message:
> > 'cmd' execution failed with error code 1 in:
> > shell(cmd, wait = TRUE, intern = output)
> 
> John,
> 
> You may have missed my earlier post about this.  The
> problem you are having with SAS is the value you are
> passing in the sasprog parameter.  You need to drop
> the single quotes, like the following 
> 
> sasprog = "C:/Program Files/SAS/SAS 9.1/sas.exe"

Ah, I'll give it a try. I added those single quotes
based on an early R-list posting  suggesting that the
space in "Program Files" was a problem. 

Actually by that time I was down to superstitious
learning but the idea seemed reasonable.
> 
> sas.get() calls a function, shQuote(), which wraps
> the program name in double quotes and then escapes
> the double quotes you are passing, which messes up
> the program name that Windows sees.

Oh I see what you mean.  
> 
> I don't understand exactly what problem you are
> having with the library names.  If I take your code
> and replace every instance of LIBRARY (in all caps)
> with dd1, your program runs just fine on my WinXP
> Pro system.  

I am not sure either but I suspect that SAS does not
tollerate such new-fangled things as USB sticks. Try
it on a USB and see what happens.  Things seem to work
just fine on a hard drive.  Currently I am blaming the
problem on SAS's outdated view of the world :) .

Most of my SAS (as opposed to R problems) seem to
disappear when I move the files to the C: drive.
Something I should have thought of sooner.


>If changing the sasprog parameter value
> as above doesn't work, write back to the R-help with
> the errors you get and someone (maybe even me :-)
> should be able to help.

Thanks, you have been very helpful already.  Stand by
for the difficult questions :)