from pathlib import Path
import sys

import pytest


SCRIPTS = Path(__file__).resolve().parents[1] / "scripts"
sys.path.insert(0, str(SCRIPTS))

from fetch_utils import fetch_with_cache  # noqa: E402
from source_registry import (  # noqa: E402
    SourceSpec,
    _REGISTRY,
    get_source,
    register,
    register_defaults,
)


def setup_function():
    _REGISTRY.clear()


def test_source_spec_requires_url_and_license_class():
    spec = SourceSpec(
        source_id="nhdplus_flow",
        url="https://example.invalid/nhd",
        organization="US EPA",
        license_class="public",
        expected_format="geojson",
    )
    assert spec.source_id == "nhdplus_flow"


def test_source_spec_rejects_missing_url():
    with pytest.raises(ValueError, match="url"):
        SourceSpec(
            source_id="missing_url",
            url="",
            organization="US EPA",
            license_class="public",
            expected_format="geojson",
        )


def test_source_spec_rejects_unknown_license_class():
    with pytest.raises(ValueError, match="license_class"):
        SourceSpec(
            source_id="bad_license",
            url="https://example.invalid/data",
            organization="US EPA",
            license_class="unclear",
            expected_format="geojson",
        )


def test_unknown_source_raises():
    register_defaults()
    with pytest.raises(KeyError):
        get_source("not_a_real_source")


def test_register_defaults_includes_nhdplus_or_wbd():
    register_defaults()
    spec = get_source("wbd_huc8")
    assert "http" in spec.url


def test_register_adds_custom_source():
    spec = SourceSpec(
        source_id="custom_source",
        url="https://example.invalid/custom.csv",
        organization="Example",
        license_class="open",
        expected_format="csv",
    )
    register(spec)

    assert get_source("custom_source") == spec


def test_fetch_with_cache_writes_data_digest_and_retrieval_time(tmp_path):
    spec = SourceSpec(
        source_id="cached_source",
        url="https://example.invalid/cached.csv",
        organization="Example",
        license_class="open",
        expected_format="csv",
    )
    destination = tmp_path / "cached.csv"

    fetch_with_cache(spec, destination, content=b"sample,data\n1,2\n")

    assert destination.read_bytes() == b"sample,data\n1,2\n"
    assert (tmp_path / "cached.csv.sha256").exists()
    assert (tmp_path / "cached.csv.retrieved_at.txt").exists()