Skip to content

long formulas in terms() (PR#873)

3 messages · Jerome Asselin, Thomas Lumley, Brian Ripley

Original
Jerome Asselin
# 
Full_Name: Jerome Asselin
Version: 1.2.2
OS: Windows 95
Submission from: (NULL) (142.52.81.1)



I need to use the "terms()" function to decompose a formula.
The function seems to work for small formulas, but not for long formulas.

##############################
#The following works!
form <- cbind(log(inflowd1),log(inflowd2),log(inflowd3),
    log(inflowd4)) ~ precip*I(Tmax^2)
terms(form)

#But enlarging the left-hand side of the formula causes an error!
form <- cbind(log(inflowd1),log(inflowd2),log(inflowd3),
    log(inflowd4),log(inflowd5),log(inflowd6)) ~ precip*I(Tmax^2)
terms(form) # ERROR
##############################

I noticed that when the left-hand side term of the formula is
too large to be written on one line, an error occurs in terms().
However, increasing the line width by "options(width=200)" did
not solve the problem.

platform i386-pc-mingw32
arch     x86            
os       Win32          
system   x86, Win32     
status                  
major    1              
minor    2.2            
year     2001           
month    02             
day      26             
language R              


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Thomas Lumley
#
On Thu, 8 Mar 2001 jerome@stat.ubc.ca wrote:

            

      
      
      
    

        
<snip>

            

      
      
      
    

        
The problem is in fixFormulaObject in terms.formula(). It uses deparse()
on the lhs of the formula.

A work-around is to change the line near the top of fixFormulaObject in
terms.formula from
	else deparse(form[[2]])
 to
	else deparse(form[[2]],width.cutoff=200)

but this may not be the final solution -- we should have something without
a width limit

	-thomas

Thomas Lumley			Asst. Professor, Biostatistics
tlumley@u.washington.edu	University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Thu, 8 Mar 2001 jerome@stat.ubc.ca wrote:

        

            

      
      
      
    

        
You can edit fixFormulaObject in terms.formula and set

deparse(form[[2]], width.cutoff = 500)

This still has a limit, but a much higher one.  Or write a shorter lhs?

deparse has been changed since that function was written, I believe,
but both a long time ago.