tests/test_dataset_schema.py

# from datetime import date
# from typing import Any
from pathlib import Path

# from pydantic import BaseModel

# # {
# #     "text": "SAMLEDE VÆRKER\n\nJEPPE AAKJÆR GYLDENDALSKE BOGHANDEL...",
# #     "source": "adl",
# #     "id": "adl_aakjaer06val",
# #     "added": "2020-09-14",
# #     "created": "1700-01-01, 2022-01-01",
# #     "license": "Creative Commons Legal Code\n\nCC0 1.0 Universal",
# #     "domain": "Wiki & Books",
# #     "metadata": {"source-pretty": "Archive for Danish Literature"},
# # }


# class SampleSchema(BaseModel):
#     text: str
#     source: str
#     id: str
#     added: date  # date.fromisoformat
#     created: tuple[date, date]
#     license: str  # TODO: should probably be a literal
#     domain: str  # TODO: convert to literal
#     metadata: dict[str, Any]

#     def metadata_validator(cls):
#         assert "source-pretty" in cls.metadata

# def test_sample_schema():
#     """Ensure that the dataset samples follow the correct schema"""
#     pass

# def test_dataset_schema():
#     """tests that the dataset frontmatter and markdown follows the correct format."""
#     # test correct yaml
#     # test it has place for descriptive stats
#     # test license is specified in text in "other" is used.
#     pass


def test_dataset_folder_structure(repo_path: Path):
    """tests that the dataset folder structure is as follows.

    dataset_name
    |- dataset_name.md
    |- dataset_name.parquet

    If there is a python file, there should at least be one called `create.py`, but there can be additional.
    """
    datasets = (repo_path / "data").glob("*")
    for path in datasets:
        assert (path / f"{path.name}.parquet").exists()
        assert (path / f"{path.name}.md").exists()

        if any(p.name.endswith(".py") for p in path.glob("*")):
            assert (path / "create.py").exists()