handling special characters e.g. accents in R

Question

I am doing some web scraping of names into a dataframe

For a name such as "Tomáš Rosický, I get a result "TomÃ¡Å¡ RosickÃ½"

I tried

Encoding("TomÃ¡Å¡ RosickÃ½") #  with latin1 response

but was not sure where to go from there to get the original name with accents back. Played around with iconv without success

I would be satisfied (and might even prefer) an output of "Tomas Rosicky"

How did you read the data.frame? Usually you can supply an encoding parameter such as `fileEncoding` to `read.table`. And as @Hong Ooi answered, UTF-8 seems to be the encoding you need. — Tommy, Mar 01 '12 at 06:48

score 13 · Accepted Answer · answered Mar 01 '12 at 06:28

13

You've read in a page encoded in UTF-8. if x is your column of names, use Encoding(x) <- "UTF-8".

answered Mar 01 '12 at 06:28

Hong Ooi

score 7 · Answer 2 · edited Jul 13 '15 at 17:51

7

You should use this:

df$colname <- iconv(df$colname, from="UTF-8", to="LATIN1")

edited Jul 13 '15 at 17:51

rink.attendant.6

answered Jul 13 '15 at 17:25

Roadkill

score 5 · Answer 3 · edited Apr 20 '16 at 15:24

5

To do a correct read of the file use the scan function:

namb <- scan(file='g:/testcodering.txt', fileEncoding='UTF-8',
what=character(), sep='\n', allowEscapes=T)
cat(namb)

This also works:

namc <- readLines(con <- file('g:/testcodering.txt', "r",
encoding='UTF-8')); close(con)
cat(namc)

This will read the file with the correct accents

edited Apr 20 '16 at 15:24

dpel

answered Mar 01 '12 at 11:08

Mischa Vreeburg

score 3 · Answer 4 · answered Nov 20 '13 at 15:40

3

A way to export accents correctly:

enc2utf8(as(dataframe$columnname, "character"))

answered Nov 20 '13 at 15:40

iulilia

4 Answers4