When I collaborate with other individuals, I generally receive text to be typeset in the form of a word document. However, it seems that there are almost always characters that, when I copy and paste - for lack of a better word, "bugger up" pdflatex.

Is there a way to sanitize these special characters and replace them with their plain text cognates?

with ', double-spaces with space, and so on.

Note: I could very well be referring to an issue associated with encoding, but I don't understand that - and neither will my collaborators.

PS: No idea what tag is relevant here, so please do so appropriately should you have the ability to do so.

  • 1
    Have you tried inputenc with the utf8 option set?
    – Canageek
    Mar 2, 2012 at 5:49
  • 2
    I recommend sed for tasks like this because I don't think a LaTeX package can/should solve these problems and because sed is made for things like this.
    – user10274
    Mar 2, 2012 at 6:10
  • 3
    @BrandonBertelsen Perhaps it's the other way around (windows isn't particularly sed friendly).
    – user10274
    Mar 2, 2012 at 6:28
  • 1
    sed should work fine in cygwin. I understand the overhead involved in that though. You may want to check out the regular expressions abilities of other languages. Python works extremely well with Windows for example.
    – qubyte
    Mar 2, 2012 at 7:20
  • 1
    You could try to open the doc-file in OpenOffice and then use the writer2latex plugin. Mar 2, 2012 at 8:43

1 Answer 1


A summary of the comments, in order of ease of use:

  • \usepackage[utf8]{inputenc}
  • Open in OpenOffice Writer, use writer2latex plugin
  • Use antiword and then pandoc to convert to tex.
  • Use sed to find replace encoded characters

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .