Hello everyone,
I’m developing a standalone utility that reads Maya ASCII (.ma) files and produces a comprehensive render-dependency list of files.
Intended output
• Textures & file paths
• Arnold assets: shaders, procedurals, stand-ins (.ass), .tx, OSL, etc.
• Scene references: .ma / .mb
• Volumes: VDB
• Light profiles: IES
• Environment maps: HDRI/EXR
• Layer-wise organization of all of the above
Motivation: faster audits, missing-asset tracing, and clearer visibility into scene dependencies.
Questions
1. Prior art / similar tools: Are there existing projects or approaches I should review?
2. Arnold parsing strategy: Recommended methods for identifying ai* nodes and resolving their file-bearing attributes directly from ASCII (e.g., .tx/.ass/.osl/.vdb/IES/HDRI)? Any reliable attribute patterns or pitfalls?
3. Render layers: Known issues when parsing legacy Render Layers vs Render Setup from .ma—especially regarding per-layer overrides, namespaces, or reference edits that may mislead a naïve parser?
4. Language & libraries: I’m leaning toward Python for ecosystem alignment and prototyping speed. Are there strong reasons to prefer another language, or specific Python libraries you recommend for robust tokenization/parsing?
Any guidance, examples, or cautions from experience would be greatly appreciated. Thank you!
