Question

Hi I’m working with a large data frame that I frequently need to subset in different combinations of variables. I’d like to be able to store the search in a string so I can just refer to the string when I want to see a subset.

x = read.table(textConnection("
                              cat1 cat2 value
                              A     Z    1
                              A     Y    2
                              A     X    3
                              B     N    2"),header=T,strip.white=T)
search_string="cat1== 'A' & cat2=='Z'"
with(x,subset(x,search))

doesn’t work. What I’d be looking for is the result of a search similar to the one below.

with(x,subset(x,cat1=='A' & cat2=='Z'))

I’d prefer not to just create multiple subsetted data frames at the start if another solution exists.

Is there a simple way to do what I’m trying?

Answer 1

You need to convert your string to an expression, and then evaluate this expression

 subset(x, eval(parse(text=search_string)))

  cat1 cat2 value
1    A    Z     1

Always remembering

fortune(106)

If the answer is parse() you should usually rethink the question.
   -- Thomas Lumley
      R-help (February 2005)

fortune(181)

Personally I have never regretted trying not to underestimate my own future stupidity.
   -- Greg Snow (explaining why eval(parse(...)) is often suboptimal, answering a question triggered by the
      infamous fortune(106))
      R-help (January 2007)

And noting in this case, it is far more typing to go the eval(parse(text==)))

You could also try using quote to save the call

search <- quote(cat1== 'A' & cat2=='Z')

and then just use

 subset(x, eval(search))

---

It is also important to remember that that subset has nonstandard evaluation so it might be safer to use `[`

x[eval(search,x),]

or

 x[eval(parse(text=search_string)), x),]