ecological-state-toolkit 0.12.2

End-to-end ecological state construction, schemas, observation, and analysis conversion
Documentation
# Ecological State Toolkit for Python

This companion distribution converts completed Scientific Workflow JSONL
recordings into verified, C-contiguous NumPy arrays. Applications describe
streams and ecological encodings through the typed library API; the command
line interface accepts the same information in a conversion-request document.

The package contains no Dispatcher, GLV, Simulator, or study-specific logic.
It is analysis tooling, not a Workflow execution unit.

The public API is:

- `ArrayEncoding`: names the supported stable JSON-to-array decoders;
- `FieldSpec(name, encoding, output, category_count=None)`: selects one field
  and its output basename;
- `StreamSpec(name, fields)`: groups fields from one Workflow stream;
- `RecordingSpec(recording, identity, streams, metadata={})`: describes one
  completed recording using a `pathlib.Path`;
- `convert_recordings(recordings, output_directory, workers=4,
  progress=None)`: verifies and converts a batch concurrently;
- `ConvertedRecording`: returns source checksums, user/caller metadata, and
  typed `ArrayDescriptor` values.

The `ecological-state-convert` command accepts `--request`, `--output`, and
`--workers`; it publishes the same output through the library API.