Hi all,
VS Code has a very nice optional spell checker (Code Spell aka cSpell) that can be customized in settings based on language. For the pretext-tools extension, I will set these default configurations automatically. If the user doesn't have cSpell installed, no harm. If they do, it will enable pretext as a language that should be checked, and set some regex patterns that should be ignored.
My question: what patterns should be ignored for spell checking? Right now, I'm thinking that anything between a pair of angle brackets should be ignored (this will include the tag name, attribute names, and attribute values), as would anything between math-mode tags (<m>, <me>, etc) and <latex-image> tags. Is this too much? Am I missing anything?
I suppose I could also do anything inside <program> or <pre>. Alternatively, I can load other language dictionaries, like python. But perhaps this would give some false negatives.
Another option is to add a dictionary for PreTeXt, which could ship in codespaces at least, or added by the CLI (in fact, I think it could be in the .ptx directory). But does maintaining a list of words seem like the right solution here?
I'd love to hear your thoughts.