# Apostrophe bug displaying text

When I typed, an apostrophe ' displays the text "âAZ". Does anyone have an idea of why this might be?

I'm using an atom package to compile the LaTeX file.

• what is atom ? I would guess you did not use the ' but the right quote ’ and that you entered the ' character in UTF-8 but declared the file encoding as latin 1 (or did not declare an encoding at all) but as you have shown no example, it is just a guess. – David Carlisle Mar 17 '17 at 21:20
• Welcome to TeX. SE. Please show a minimal working example, which will reproduce your probe, Help us to help you! – Zarko Mar 17 '17 at 21:28
• You have something like \usepackage[T1]{fontenc} in your file, but you don't have \usepackage[utf8]{inputenc} as well. Those three letters are what happens when the numeric value of the tokens created by \IeC{\'} get mapped to glyphs. If you include \usepackage[utf8]{inputenc} in your preamble you should get what you want. The atom editor is just running latexmk, so it shouldn't be the issue (I think). – Charles P. Schaum Mar 17 '17 at 22:03
• @CharlesP.Schaum most likely inputenc is the required solution although the standard T1 internal processing such as \IeC{\'} won't generate those three byte sequences, rather the input file must have UTF-8 characters which are not being combined (as ascii input is assumed) – David Carlisle Mar 17 '17 at 22:57

Presumably you do not get âAZ but âĂŹ

\documentclass{article}
\usepackage[T1]{fontenc}

\begin{document}

[’]

\end{document}


Produces

To see why, see that ’ is U+2019 that is, the character with value hex 2019.

In UTF-8 encoding, characters in that range take three bytes, in this case E2 80 99

Unless told to interpret such triples via the declaration

\usepackage[utf8]{inputenc}


pdftex will assume three bytes denote three input characters with codes E2, 80 and 99 respectively.

If you specify the 256-character single-byte encoding, T1, is being used for fonts with the declaration

\usepackage[T1]{fontenc}


then latex will typeset the characters in those slots in T1 encoding which are â, Ă, Ź

E2 (â) matches the encoding of â in Latin-1, but 80 and 99 are in the control character range in latin 1 and Unicode but the TeX-specific T1 encoding fills the full 256 range with characters as control character slots are not useful for typesetting.

\documentclass{article}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}
\begin{document}

[’]

\end{document}


Add \usepackage[utf8]{inputenc} to fix the issue.