dargs 0.5.0.post0

Process arguments for the deep modeling project.

Argument checking for python programs

This is a minimum version for checking the input argument dict. It would examine argument's type, as well as keys and types of its sub-arguments.

A special case called variant is also handled, where you can determine the items of a dict based the value of on one of its flag_name key.

There are three main methods of Argument class:

• check method that takes a dict and see if its type follows the definition in the class
• normalize method that takes a dict and convert alias and add default value into it
• gendoc method that outputs the defined argument structure and corresponding docs

There are also check_value and normalize_value that ignore the leading key comparing to the base version.

When targeting to html rendering, additional anchor can be made for cross reference. Set make_anchor=True when calling gendoc function and use standard ref syntax in rst. The id is the same as the argument path. Variant types would be in square brackets.

Please refer to test files for detailed usage.

Additional features

• PEP 484 type annotations
• Native integration with Sphinx, DP-GUI, and Jupyter Notebook
• JSON encoder for Argument and Variant classes
• Generate JSON schema from an Argument, which can be further integrated with JSON editors such as Visual Studio Code
• Load dict values from external JSON/YAML files via the $ref key