Assure Fortran documentation consistence.
As a project increases in size, it is harder to track the quality of the present
code. In the case of functionality unit testing and coverage metrics can be
enough to ensure there are no problems. But keeping track of correct
documentation needs even more meticulous work, external tools can be used to
help this process (like )
in Python.
fordoctest aims to be an equivalent tool for Fortran documentation.
fordoctest <ford_project_file> [ford_flags]Right now fordoctest is in a pre-alpha state, it will check on all the source
files included in a FORD file and send a warning if any of the following cases
occur:
- More than one module per-file
- Undocumented:
- Isolated procedures
- Modules
- Modules procedures
- Procedures arguments
The output of `fordoctest will be something like:"
fordoctest aims to be a complete tool used in Fortran projects CI systems that
can assure that all the code is correctly documented, and maybe also include
extra checks like file headers, file naming checks, usage with other tools
beside FORD, like Doxygen and Sphinx. We'll consider fordoctest to be in
beta-stage once:
- It becames FORD agnostic (right now it fully depends on the FORD cli parser so it's hard to easy add up methods)
- Includes documentation checks of all the different kinds entities (like not included yet, derived types)
- Include documentation restrictions (like forcing imperative verbs in a procedure documentation)
- Fully tested with coverage of at least 90%
Since this is still an alpha release, we haven't yet released an installable
on PyPI, but the development version can be installed with pip with:
pip install git+ssh://git@github.com/fedebenelli/fordoctest.git