georgesk
commented
9 years ago - fixed a lot of issues due to inconsistencies of the code initially
written for Python2, mainly confusions between byte strings
and unicode strings. The new code tries to use byte strings for
file-like streams and unicode strings for in-memory objects.
- added JavaDoc strings to document thoroughly functions and methods, mainly in odf/opendocument.py
- added a config.dox file to allow one to build a documentation in HTML and LaTeX, for library developpers and users.
- added many assert() clauses to ensure the types of parameters in functions and methods
- modified the utilities to make them usable with both Python2 and Python3
- implemented a feature announced in csv2ods manfiles but not previously active: -c / --encoding switch to take in account the encoding of the CSV source file.