Seminar converter

Converts KSI tasks from before year 2014 into SQL statement that can be inserted into a backend database.

Requirements

Because the content of the tasks is inside hidden repository, it is necessary to first initialize submodules

git submodule init

which will clone all tasks inside the seminar-till-2014 directory.

All Python requirements are listed inside requirements.txt file and can be installed by running

pip install -r requirements.txt

The converter requires following commands to be available:

To generate HTML task files from seminar-till-2014 directory to output directory, execute

python main.py

To process the generated output directory into a single SQL statement, first adjust global constants inside generate_sql.py and then execute

python generate_sql.py

This will print the SQL import statement to the standard output.

The code is split into two folders:

tex contains functions that take care of locating all TeX tasks together with their solutions (locator.py) and for parsing task metadata (parser.py) - name, points and assets
html_writer converts TeX tasks into HTML strings that can be inserted into KSI database (converter.py)

Both sub-directories have a types.py file which contains custom-defined types.