============================
   Encoding Normalization
============================
:Author: Martin Blais <blais@furius.ca>
:Date: 2006-06-29
:Abstract:

    We are having problems with normalization of encodings.

.. contents::
..
    1  Input Cases
    2  Places where conversion occurs


Input Cases
===========

If we only consider two encodings different than ascii (latin-1 and UTF-8), we
have the four possible cases occurring with Nabu and what happens then:

1. File does not specify “coding”, is encoded in latin-1

2. File does not specify “coding”, is encoded in UTF-8

3. File specifies “coding” latin-1

4. File specifies “coding” UTF-8



Places where conversion occurs
==============================

- When you upload to docutils, docutils attempts to guess the input encoding.
  It does a pretty good job at it, i.e. looks for the coding directive, tries to
  convert from many different encodings.

