How can I write special characters in RMarkdown latex documents?

Question

I am trying ot write a paper in RMarkdown pdf. But I dont know how to use special characters like "İ, ı, ğ, ü, ö". Those characters are present in Turkish. I can easily use them in Latex using e.g. \u{g} – ğ, \c{c} – ç, \”{u} – ü, {\i} – ı, etc.

Could you please tell how can I do the same in RMarkdown?

And where can I find some Rmarkdown paper or book templates?

Answer 1

HTML codes for Turkish characters will work (example from wikipedia):

---
title: "My doc"
output: pdf_document
---

HTML      char  detail                            anglicized
-------   --    ------------------------------    ---
Ğ   Ğ       Uppercase "G" with breve accent   gh1
İ     İ     Uppercase dotted "I"²               i (as in "tree")
Ş     Ş     Uppercase "S" with cedilla          sh
ğ     ğ     Lowercase "g" with breve accent   gh1
ı     ı     Lowercase dotless "i"³              ou (as in "in")
ş     ş     Lowercase "s" with cedilla          sh
-------   --    -------------------------------   --

Which renders the PDF like this:

For article templates, see the rticles package, or make your own

For book templates, have a look at pandoc ebook, gitbook and bookdown

---

This xelatex option I initially suggested will not work for Turkish characters:

The xelatex engine is recommended for this kind of thing. Then you can access your system fonts with the mainfont argument:

---
title: "My doc"
output:
  pdf_document:
    latex_engine: xelatex
mainfont: "name of your system font that has all those characters"
---

PDF output will be in the font you specify. 

Just type as normal with no special codes.

Answer 2

I guess this question per se is obsolete. As of now (rmarkdown V1.4 / knitr V1.15.1), knitr automatically follows the advice of this question and includes

\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}

in the preamble to LaTeX output (caveat: of course this means that your document editor will have to be using UTF-8, but e.g., RStudio does that by default).

Then this works as expected:

---
title: "Test"
output: pdf_document
---

Here are some special characters: İ ı Ş Ğ ş ğ ö ü

This is not a panacea, as many scripts still require more tinkering (I tried Korean, Chinese, and Japanese). See here for where to start on bolstering the preamble in case you need more multifarious fonts.

Answer 3

rmarkdown uses LaTeX to create the PDF, so you should be able to just use LaTeX markup. As an example:

---
title: "Test"
output: pdf_document
---

This is a special character: \u{g}.

Also, depending on the particular characters, you may be able to just type them in as-is (see, for example, here).

Answer 4

If you need special characters for other languages in addition to Turkish, looks like you can type in the code listed at this webpage:

http://www.starr.net/is/type/htmlcodes.html

Be sure to include the ;

Answer 5

This might be too late but I would like to add someothing in addition to the amazing and straightforward existing answers so far. This answer also works not only for pdf_documents but also for html_ and word_documents.

Solution: Use the

1. base R functions capture.output() and cat()
2. Unicode notation

Example: Consider writing ñ. A quick look at https://www.compart.com/en/unicode/U+00F1, gives us the U+00F1 code for ñ. In R, this is \u00f1. An MWE is as follows.

---
title: "Minimum Working Example"
author: "Mr. John Ate`r capture.output(cat('\u00f1'))`o"
output:
  pdf_document: default
  word_document: default
  html_document: default
---

## Some Example of Printing `r capture.output(cat('\u00f1'))`

This is an inline output of `r capture.output(cat('\u00f1'))`.

Simplification: It is quite cumbersome to use the capture.output(cat()) all the time so we can define a custom function say

diac <- function(x) {
  capture.output(cat(x))
}

and use it inline as follows: `r diac('\u00f1')`.