This repository provides standardized metadata schemas to describe electrochemical systems and
the collected data, and the mdstools Python package to generate, validate and work with them.
It is part of the echemdb community project.
Documentation: https://echemdb.github.io/metadata-schema/
The schemas cover the electrochemical system (electrodes, electrolyte, cell), the experimental
setup, the source of the data and its curation. They are defined in LinkML
(linkml/), the single source of truth, from which we generate
- JSON Schemas (
schemas/), to validate metadata with any tool that supports JSON Schema, and - Pydantic models (
mdstools/models/).
The schema reference, example files and JSON Schema downloads are on the documentation website.
The mdstools Python package works with metadata files: it flattens them to Excel or CSV,
enriches them with descriptions and examples from the schemas, validates them, and migrates
them to newer schema versions.
Install pixi, then
git clone https://github.com/echemdb/metadata-schema.git
cd metadata-schema
pixi install# Flatten a YAML file to enriched Excel and CSV files in generated/
mdstools flatten tests/example_metadata.yaml
# Convert an Excel or CSV file back to YAML
mdstools unflatten generated/example_metadata.xlsx --schema-file schemas/minimum_echemdb.json
# Migrate a metadata file to the current schema version (dry run without --in-place)
mdstools update metadata.yaml --in-placeRun mdstools <command> --help for all options, or prefix the commands with pixi run to use
the pixi environment. The CLI, the Python API and the schema generation are described in the
documentation.
Contributions are welcome. See CONTRIBUTING.md for the setup, the workflow for schema changes and migrations, and the checks:
pixi install -e dev
pixi run test
pixi run validate
pixi run lint
pixi run check-news
pixi run doc-fast # build the documentation in doc/generated/html