[Rd] In-string variable/symbol substitution: What formats/syntax is out there?

Peter Meilstrup peter.meilstrup at gmail.com
Tue Dec 17 23:08:51 CET 2013


There's also knit_expand() in knitr, which uses the form "pi = {{pi}}"
with general expressions.

Peter

On Tue, Dec 17, 2013 at 1:44 PM, Henrik Bengtsson <hb at biostat.ucsf.edu> wrote:
> Hi,
>
> I'm try to collect a list of methods/packages available in R for doing
> in-string variable/symbol substitution, e.g. someFcn("pi=${pi}"),
> anotherFcn("pi=@pi@") and so on becomes "pi=3.141593".  I am aware of
> the following:
>
> ** gsubfn() in the 'gsubfn' package, e.g.
>> gsubfn( , , "pi = $pi, 2pi = `2*pi`")
> [1] "pi = 3.14159265358979, 2pi = 6.28318530717959"
>
>
> ** gstring() in the 'R.utils' package, e.g.
>> gstring("pi = ${pi}, 2pi = ${`2*pi`}")
> [1] "pi = 3.14159265358979, 2pi = 6.28318530717959"
>
>
> I'm sure there are other approaches - do you know of any in R?  They
> don't have to support in-line calculations such as in the first two
> examples, but if they do, it's a bonus.  I'm looking for simpler
> functions and not full blown literate programming methods (e.g.
> Sweave, noweb, knitr, brew, RSP, ...).  It should also be *in-string*
> substitution out of the box, so sub(), sprintf() and friends does not
> count.
>
> Thanks
>
> Henrik
>
> PS. The following is on the borderline because it does not do
> automatic variable look up, but since others may bring it up and/or
> know of a neater approach, I mention it too:
>
> ** copySubstitute() in the 'Biobase' package (with some efforts), e.g.
>> bbsubst <- function(fmt, ...) {
>   args <- lapply(list(...), FUN=as.character)
>   in <- textConnection(fmt)
>   out <- textConnection("res", open="w")
>   on.exit({ close(in); close(out) })
>   copySubstitute(in, out, symbolValues=args)
>   res
> }
>> bbsubst("pi = @pi@", pi=pi)
> [1] "pi = 3.14159265358979"
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel



More information about the R-devel mailing list