Fix coordinator worker integration

This commit is contained in:
Emil
2026-07-23 21:33:37 +03:00
parent b4a89dd7c2
commit 983c5843ec
25 changed files with 748 additions and 158 deletions
+17
View File
@@ -2,6 +2,7 @@ from __future__ import annotations
from pathlib import Path
import pytest
from rdkit import DataStructs
from scimesh.chemistry.dataset import DatasetStats, iter_valid_molecules
@@ -61,3 +62,19 @@ def test_graph_supports_less_than_threshold_direction(small_dataset: Path) -> No
)
assert all(edge.similarity <= 0.15 for edge in result.edges)
def test_graph_rejects_duplicate_identifiers(tmp_path: Path) -> None:
dataset = tmp_path / "duplicate_ids.tsv"
dataset.write_text(
"chembl_id\tcanonical_smiles\nDUP\tCCO\nDUP\tCCC\n", encoding="utf-8"
)
with pytest.raises(ValueError, match="duplicate chembl_id"):
build_similarity_graph(dataset, threshold=0.1, block_size=1)
def test_graph_writer_creates_missing_output_directory(tmp_path: Path) -> None:
output = tmp_path / "nested" / "edges.csv"
write_graph_edges(output, [])
assert output.read_text(encoding="utf-8").startswith("source_id,target_id")