Hello.

I have seen some academic papers about semantic search done with Neural Networks. These existed before LLMs were a thing. So I'm wondering. Do you know a tool which automatically indexes a codebase? So for the project developers, it'll be easier to find out where everything is. The tool should obviously know about the APIs and also it should recognize the programming language syntax.

Something like a graph of entities in the codebase. Like classes, methods, functions, constants and so on.

all 11 comments

sorted by: hot top controversial new old
[–] 6 points 3 weeks ago (8 children)

Do you mean like things like LSP and tags?

  • source
  • hideshow 8 child comments
  • [–] [S] 2 points 3 weeks ago (7 children)

    Yup. But in a web UI. So we can find other relevant entities.

  • source
  • parent
  • hideshow 7 child comments
  • [–] 8 points 3 weeks ago (4 children)
  • [–] [S] 2 points 3 weeks ago (2 children)

    oh yeah! that's it! But I wanted it for Lua. So the keyword is "documentation generator"?

  • source
  • parent
  • hideshow 2 child comments
  • [–] 4 points 3 weeks ago (1 child)

    https://github.com/colbymchenry/codegraph might approach what you're looking for. I've tried tokensave, a reimplementation in rust, and it seems to work ok when it comes to a semantic search over a code base (albeit it seems almost entirely genAI-coded so I'm reluctant to outright recommend it).

  • source
  • hideshow 1 child comment