Using `process()`

process() is an R function which can be used to create a LaTeX document (usually but not necessarily in PDF format) from a simple text file. Any R commands are executed in the background and any numerical or graphical output is incorporated into the final document. It uses the R Sweave() utility.

It is assumed here that you have working versions of R and LaTeX, have a special dedicated process() folder and have successfully loaded process() into R. See the detailed instructions for setting up and loading process() if necessary.

Hello World

The simplest example is as follows: consider a text file test consisting of just two lines:

Heading
Hello World

If this is in the same directory that R was started in and all is working properly, then if you type

process(test)

at the R prompt then a file test.pdf is opened in a PDF viewer. Note that the first line is a centred, large bold heading, the username is inserted as well as the date, and the second line is typeset as normal text.

R code and graphics

Suppose the file test2 contains the following lines:

Heading
Here is some code:
\code
z <- rnorm(100)
\end
Here is a graph:
\graph
hist(z)
\end

Here is a final remark.

Then typing

process(test2)

opens a file test2.pdf in a PDF viewer; note that the blank line before the final remark ensures that the text after the graph starts on a new line.

General Sweave

An Sweave code chunk is generally of the form

<<(chunk parameters go here)>>=
(R code goes here)
@

The utilities \code...\end and \graph...\end are equivalent to

<<>>=
...
@

and

<<fig=TRUE>>=
...
@

respectively, as the following example shows:

Heading
Here is some code:
<<>>=
z <- rnorm(100)
@
Here is a graph:
<<fig=TRUE>>=
hist(z)
@

Here is a final remark.

This gives identical output to the previous example. More sophisticated Sweave operations can be performed in this way, for example the scaling of graphics: supposing test3 is

Heading
Here is some code:
<<>>=
z <- rnorm(100)
@
Here is a graph:
<<fig=TRUE,width=4,height=7>>=
hist(z)
@

Here is a final remark.

then process(test3) gives test3.pdf with a bigger histogram.

General LaTeX

Experienced users of LaTeX can insert command sequences in their text file as they would in a standard LaTeX file. Suppose test4 is

Heading
Here is some code: we are going to generate some pseudo-random numbers
from the $N(\mu,\sigma^2)$ distribution with mean $\mu=0$
and variance $\sigma^2=1$.
\code
z <- rnorm(100)
\end
Here is a graph:
\graph
hist(z)
\end

Here is a displayed equation
$$\int_{-\infty}^\infty e^{-\frac{x^2}2}\,dx=\sqrt{2\pi}\,.$$

Then process(test4) gives test4.pdf.

Michael Stewart

Using process()