I would like to use the toggles from etoolbox to direct Sweave to evaluate or not evaluate chunks of code. Example:


 read.csv("My Big Fat Data file.csv") -> MyData

The idea would be to read in the data at the beginning of the work session and then \togglefalse so it would not need to be read again. I am using Eclipse and StatET, so object MyData would be resident in memory and accessible to R throughout the session.

However, the above code evaluates regardless of the toggle value.

I know I can achieve this result by changing eval=TRUE to eval=FALSE - but using toggles would be cleaner and give more flexibility.

  • Problem might be related to the @. Try adding \makeatletter before the \iftoggle and a \makeatother after the closing }. You also should have an additional {} at the end for the case where the \iftoggle evaluates to false. Apr 30 '12 at 17:41
  • Please add a minimal working example (MWE) that illustrates your problem. Also your \ifftoggle lacks the ‘false’-part.
    – Speravir
    Apr 30 '12 at 17:42

I encourage you to try the knitr package, which does this job more elegantly. See section 3.5 "Evaluation of Chunk Options" in the manual: https://github.com/downloads/yihui/knitr/knitr-manual.pdf

Your approach has two problems:

  1. Sweave will not recognize \iftoggle, because it does not analyze the TeX context; the tex document is processed purely as character strings -- whenever it sees <<>>=, it will treat the following lines as R code, regardless of if you have \iftoggle above the code chunk;
  2. you write more code by inserting \iftoggle{FirstTime}{} around your code chunks, and this is how you can do it with less efforts in knitr:

    # this variable controls if a chunk should be evaluated
    dothis <- TRUE
    <<chunk1, eval = dothis>>=
    print('say hello world')
    <<chunk2, eval = !dothis>>=
    print('silence is gold')

If you find the syntax strange, you may read how knitr differs with Sweave; basically knitr makes your document really programmable.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.