linting

Author: florian-huber

ms2query/data_processing/fingerprint_computation.py

@@ -1,6 +1,6 @@
-import numpy as np
 import numba
-from numba import types, typed
+import numpy as np
+from numba import typed, types
 from rdkit import Chem
 from tqdm import tqdm
 

ms2query/database/ann_vector_index.py

@@ -297,7 +297,15 @@ def _build_where_clause(self, where_sql: Optional[str]) -> str:
             return f"WHERE {where_sql}"
 
     # ---------- querying ----------
-    def query(self, vector: np.ndarray, k: int = 10, ef: Optional[int] = None, assume_normalized: Optional[bool] = None):
+    def query(
+            self,
+            vector: np.ndarray,
+            k: int = 10,
+            ef: Optional[int] = None,
+            assume_normalized: Optional[bool] = None
+            ) -> List[Tuple[str, float]]:
+        """Query the index with a single vector.
+        """
         if self._index is None:
             raise RuntimeError("Index not built or loaded.")
         v = np.asarray(vector, dtype=np.float32).reshape(1, -1)
@@ -322,6 +330,8 @@ def query(self, vector: np.ndarray, k: int = 10, ef: Optional[int] = None, assum
 
     # ---------- persistence ----------
     def save_index(self, path_prefix: str) -> None:
+        """Save index to files with given prefix.
+        """
         if self._index is None or self._ids is None:
             raise RuntimeError("Index not built or loaded.")
         meta_path = f"{path_prefix}.meta.json"
@@ -335,6 +345,8 @@ def save_index(self, path_prefix: str) -> None:
             json.dump(meta, f)
 
     def load_index(self, path_prefix: str) -> None:
+        """Load index from files with given prefix.
+        """
         meta_path = f"{path_prefix}.meta.json"
         ids_path = f"{path_prefix}.ids.npy"
         hnsw_path = str(path_prefix)  #f"{path_prefix}.nmslib"

ms2query/database/spectral_database.py

@@ -318,7 +318,9 @@ def get_embeddings(
             cur.execute(f"SELECT spec_id, d, vec FROM {embeddings_table} ORDER BY spec_id ASC;")
         else:
             ph = ",".join("?" for _ in ids)
-            cur.execute(f"SELECT spec_id, d, vec FROM {embeddings_table} WHERE spec_id IN ({ph}) ORDER BY spec_id ASC;", ids)
+            cur.execute(
+                f"SELECT spec_id, d, vec FROM {embeddings_table} WHERE spec_id IN ({ph}) ORDER BY spec_id ASC;",
+                ids)
 
         sids: List[str] = []
         vecs: List[np.ndarray] = []

ms2query/library_io.py

@@ -8,7 +8,7 @@
 from tqdm import tqdm
 from ms2query import MS2QueryDatabase, MS2QueryLibrary
 from ms2query.data_processing.merging_utils import cluster_block, get_merged_spectra
-from ms2query.database import EmbeddingIndex, FingerprintSparseIndex
+from ms2query.database import EmbeddingIndex
 from ms2query.database.spectra_merging import _split_by_mode_charge
 
 

tests/test_ann_vector_index.py

@@ -66,9 +66,7 @@ def test_save_and_load_roundtrip_dense(tmp_path):
 def test_build_index_from_sqlite_streams_and_orders(batch_rows):
     conn = sqlite3.connect(":memory:")
     conn.execute("CREATE TABLE embeddings(spec_id TEXT, vec BLOB, d INTEGER)")
-    # Two 3D vectors; deliberately unsorted insertion order
-    v1 = np.array([1.0, 0.0, 0.0], np.float32)
-    v2 = np.array([1.0, 1.0, 0.0], np.float32)
+    # Add 3 vectors of dim 3
     conn.executemany( "INSERT INTO embeddings(spec_id, vec, d) VALUES (?,?,?)",
         [
             ("id_1", np.array([1.0, 0.0, 0.0], np.float32).tobytes(), 3),

tests/test_compound_database.py

@@ -2,8 +2,6 @@
 import sqlite3
 from pathlib import Path
 import numpy as np
-import pandas as pd
-import pytest
 from ms2query.data_processing import compute_morgan_fingerprints
 from ms2query.database.compound_database import (
     CompoundDatabase,