On 2011-04-18, Dave Kuhlman wrote:
> (1) Does this patch have the effect,
> under Python 3.2, of generating an .odt file containing unicode
> (2) If so, wouldn't we actually want utf-8 as the encoding? Or does
> oowriter actually understand unicode?
Python 3 uses the `unicode` object for "generic" strings internally,
allowing encoding-independent representation of all Unicode characters.
Docutils uses `unicode` internally as well - but under Python 2 it needs
to explicitely decode things it reads and encode when writing (to disk or
So no, this patch does not change the behaviour that the encoding of the
output file is governed by the config setting
The text encoding for output.