Skip to content

substitute() bug? (PR#10525)

2 messages · rossibarra at gmail.com, Duncan Murdoch

rossibarra at gmail.com
# 
The first four lines of code below work as normal.  The fifth thorows  
an error: "Error in paste(theta[1], "=", 5) : object "theta" not found"


x=rnorm(1000);
bob=density(x);
topp=5;
plot(bob,xlab="", ylab="",  
main=substitute(paste(theta[1],"=",topp),list(topp=topp)), type="l");
plot(bob$y~bob$x,xlab="", ylab="",  
main=substitute(paste(theta[1],"=",topp),list(topp=topp)), type="l");


Note that theta is not defined anywhere:

ls()
  [1] "bob"      "dline"    "dplot"    "dpt"      "less"     "lesseq"
  [7] "more"     "moreeq"   "peak"     "peakline" "peakplot" "pop"
[13] "topp"     "x"


Version:
  platform = i386-apple-darwin8.10.1
  arch = i386
  os = darwin8.10.1
  system = i386, darwin8.10.1
  status =
  major = 2
  minor = 6.1
  year = 2007
  month = 11
  day = 26
  svn rev = 43537
  language = R
  version.string = R version 2.6.1 (2007-11-26)

Locale:
C

Search Path:
  .GlobalEnv, package:stats, package:graphics, package:grDevices,  
package:utils, package:datasets, package:methods,

____________

Jeffrey Ross-Ibarra
email: rossibarra at gmail.com
web: rossibarra.googlepages.com
Dept. Ecology and Evolutionary Biology
University of California
Irvine, Ca. 92697
Duncan Murdoch
#
rossibarra at gmail.com wrote:
I'm not sure this is a bug, but if it is, it's in 
graphics:::plot.formula, not in substitute.  The problem is that 
substitute returns an unevaluated call, and plot.formula tries to 
evaluate it.  You can avoid the bug by wrapping the substitute call in 
as.expression.

The docs for plotmath say that it requires an expression, but "in most 
cases" language objects will be coerced to one:  it appears this is one 
of the cases where that is not true.

Duncan Murdoch