This plugin provides proof-of-concept level support for the ACT-R modelling language.
This originated out of procrastination of the tutorial/unit5/1hit-blackjack-model.lisp
assignment as part of the Cognitive Science BSc. Degree at TU Darmstadt in the winter semester 2019/2020.
I had no previous experience writing VS Code plugins, TextMate grammars or Oniguruma Regular Expressions, the latter two most likely being blatantly obvious when looking at the TextMate grammar.
In its current state the grammar does not parse the actual ACT-R grammar, but only parts of it and only if they are formatted in a certain form. If some variable or chunk name contains symbols besides [a-zA-Z0-9\-]
it has a good chance to break. So far the grammar is only guaranteed to cover the tutorial/unit5/1hit-blackjack-model.lisp
assignment code and any reasonable changes that could be made to it while completing the assignment.
- refactor grammar to a consistent style
- convert to
yaml
based definition because this supports comments - Implement a task runner to execute models directly from VS Code
- This is fairly simple and just requires calling
./apps/act-r-64
with various arguments to load initialization scripts (GUI optional) and finally the actually model file
- This is fairly simple and just requires calling
- Problem Matcher to extract warnings and errors from the task runner output
- Extracting the documentation from the reference manual to provide documentation for known functions/options/concepts
It is unlikely that I will invest much more if any time into this but will gladly support anyone who wants to attempt this. Open an issue or a PR for a feature you want to work on and I will be notified via GitHub.
- Initial release of prototype
- This mostly provides syntax highlighting support for the
tutorial/unit5/1hit-blackjack-model.lisp
assignment
- This mostly provides syntax highlighting support for the