Adapter documentation template
Copy this folder as a starting point for a new adapter track.
Overview
- What backend does this adapter target?
- Is it lazy, eager, or hybrid?
- What PlanFrame features are supported (joins, grouping, pivot, I/O, etc.)?
Installation
pip install <your-adapter-package>
Adapter author checklist (recommended)
- Run PlanFrame’s published conformance helper in your CI:
- Docs:
https://planframe.readthedocs.io/en/latest/planframe/guides/adapter-conformance/ - API:
planframe.adapter_conformance.run_minimal_adapter_conformance - Set conservative capability flags via
BaseAdapter.capabilities(see core docs andAdapterCapabilities).
Quickstart
Show a minimal example:
- define a schema
- construct a frame
- run 2–3 transforms
- collect / materialize
See the official adapters for concrete examples:
planframe-polarsplanframe-pandasplanframe-sparkless(Spark UI +sparklessengine)
Backend notes
- dtype mapping expectations
- null behavior
- join semantics differences (if any)
- performance notes