Performance issues with large Robot Framework projects

[petrieh]

Hello all,

We’re running into performance and memory issues when using RobotCode on large Robot Framework projects.

I’ve uploaded a dummified version of our real project here:
👉 https://github.com/petrieh/dummified-robot

It’s about 460 MB of .robot and .resource files, trying to follow good design practices (no global resources, modular keywords, clear hierarchy). Even though it contains only simple keywords and calls, RobotCode takes ~30-40 minutes to analyze it and uses around 6 GB RAM, which isn’t freed afterward. The real project takes roughly 3 hours and uses up to 12 GB RAM.

This makes RobotCode hard to use in Dev Containers, since it seems to reanalyze everything each time the container starts, even when nothing changed. The extension data is mounted to the container and kept intact between restarts.

Component	Version
RobotCode	1.9.0
VS Code	1.105.0
Python	3.9.21
Robot Framework	7.2.2
OS	Rocky Linux 9.6

❓ Questions

• Any way to speed up or optimize analysis for large projects?
• Can analysis results be cached or reused between container sessions or lazy analysis used instead?
• How can we profile or debug the analyzer to locate the bottlenecks?

I’m happy to help investigate or fix the issue — I’d just need some guidance on where to start inside the RobotCode codebase.