Non-objective LaTeX behaviour

Question

Using: Win 10, MikTex 2.9.6210, TeXnicCenter 2.02. Compiling with TeXnicCenter's default profile LaTeX => PDF, which uses pdflatex.exe

I seem to have ran into a pickle. I have two files:

• _Main.tex -> created by editing a template file I downloaded from my faculty's website.
• TestingCRLF.tex -> created with RightClick > New > Text Document and then edited.

They both contain the code:

\documentclass[11pt,openright,twoside]{report}
\usepackage[utf8]{inputenc}
\usepackage[portuguese,english]{babel}

\title{Testing special characters.}
\author{The very Me}

\begin{document}
    \maketitle
    Testânto caractéres epecíais.

\end{document}

_Main.tex compiles withouth error but TestingCRLF.tex has 3 errors. One for each of the special characters within document. The errors I get are:

! Package inputenc Error: Unicode char ânt (U+E4)
(inputenc)                not set up for use with LaTeX.
See the inputenc package documentation for explanation.
Type H <return> for immediate help.
...

l.10 Testânt
           o caractéres especíais.

The other two errors mention characters (U+E0) and (U+D7).

In the past I've faced the issue of porting a tex file from linux, which uses only Line Feed ('\n') character for ending a line, while Windows used Carriage Return and Line Feed ("\r\n") to end a line. This caused this type of errors when compiling.

To check for that issue I used NotePad++ which allows you to see special hidden characters. I tell you, both files have "\r\n" at the end of each line. So I can only conclude both files are IDENTICAL. But one compiles and the other does not.

Answer 1

As Ulrike Fischer said in the question's comment, the issue was the Encoding of the file. This means the issue is not related with LaTeX.

As far as I've understood, when editing a file, the character encoding determines what bytes sequences represent which characters. TestingCRLF.tex had ANSI encoding, so when I entered the special characters they were not in UTF-8 encoding.

To fix this, I opened the file in Notepad++, which allows me to change the encoding used to interpret the file. It was ANSI and I changed it to UTF-8. All special characters showed up unencoded (as hex characters). Now in UTF-8 mode, I replaced them all with correct versions.

That's it. So, in the end guess what.. The files were different after all! haha