s-knibbs / sourcespell Goto Github PK

Character set detection is possible using the chardet library.

Currently the same character set is used for all files searched which will cause issues if the project contains files with different encodings.

Add tests for tox.ini and test python 3 compatibility. Both Pygments and PyEnchant support Python 3 so compatibility should not be a problem.

If a there is a misspelt word which would not be in the dictionary anyway, e.g. (Djangi -> Django), it would be nice to be able to manually enter the replacement, since the suggestion will not be present.

Pygments doesn't support markdown markup. The markup can probably just be passed straight to enchant without any problems though.

The token filtering from Pygments could be improved. Different file formats such as 'reStructuredText' and HTML need specific filtering rules applied.

For example, all HTML attribute values are currently being included but we probably only want the values of alt, title, value and any other attributes that contain text.

This can be implemented by adding separate filter classes and selecting based on the lexer name.