A few scripts to ease the maintenance of Fallout's text files (1 and 2). Made for both practical (Fallout Fixt) and educational purposes (ugly code and lots of refactoring just for the sake of refactoring). Made for Python 3 in Windows (but it should work on UNIX systems too).
You'll need to install version 3.5 (or higher) of Python. Download here: http://www.python.org/download/
Although they are optional, you should also install both python-Levenshtein
and ujson. You can do this via cmd with the following lines:
pip install ujson python-Levenshtein
In case pip gives you an error (this usually means you don't have the
required compiler), you can use the binaries at:
http://www.lfd.uci.edu/~gohlke/pythonlibs/
For this, first upgrade pip:
python -m pip install --upgrade pip
And then use pip install followed by the .whl file path.
If the files don't have a correct syntax everything will most likely explode, so a good practice would be to run syntax_checker every time you make a change on the files.
Any module that uses lang.py (lazy_town, mismatch_finder and
similarity_finder) needs the files to be clear of line breaks (use
linebreak_remover with the parameter -e).
Every script now uses a command-line interface.
checks for duplicate index numbers, which usually equates to content been override and not used
removes line breaks, essential for the syntax_checker
spots ugly crash-inducing bugs
makes unmaintained localizations usable by "injecting" good lines (based on their similarity ratio -using Levenshtein distance-) into the Fixt files
looks for lines that should be same but aren't (translators fault)
ongoing project for creating a usable database of lines based on their similarity ratio. This would speed the localization process in the future and could work as a kind of translation memory.