enable mypy ignore-without-code

mltb2/arangodb.py

@@ -107,13 +107,13 @@ def from_config_file(cls, config_file_name, aql_overwrite: Optional[str] = None)
         _check_config_keys(arango_config, expected_config_file_keys)
 
         return cls(
-            hosts=arango_config["hosts"],  # type: ignore
-            db_name=arango_config["db_name"],  # type: ignore
-            username=arango_config["username"],  # type: ignore
-            password=arango_config["password"],  # type: ignore
-            collection_name=arango_config["collection_name"],  # type: ignore
-            attribute_name=arango_config["attribute_name"],  # type: ignore
-            batch_size=int(arango_config["batch_size"]),  # type: ignore
+            hosts=arango_config["hosts"],  # type: ignore[arg-type]
+            db_name=arango_config["db_name"],  # type: ignore[arg-type]
+            username=arango_config["username"],  # type: ignore[arg-type]
+            password=arango_config["password"],  # type: ignore[arg-type]
+            collection_name=arango_config["collection_name"],  # type: ignore[arg-type]
+            attribute_name=arango_config["attribute_name"],  # type: ignore[arg-type]
+            batch_size=int(arango_config["batch_size"]),  # type: ignore[arg-type]
             aql_overwrite=aql_overwrite,
         )
 
@@ -150,12 +150,12 @@ def load_batch(self) -> Sequence:
                 aql = self.aql_overwrite
             cursor = connection.aql.execute(
                 aql,
-                bind_vars=bind_vars,  # type: ignore
+                bind_vars=bind_vars,  # type: ignore[arg-type]
                 batch_size=self.batch_size,
             )
-            with closing(cursor) as closing_cursor:  # type: ignore
-                batch = closing_cursor.batch()  # type: ignore
-        return batch  # type: ignore
+            with closing(cursor) as closing_cursor:  # type: ignore[type-var]
+                batch = closing_cursor.batch()  # type: ignore[union-attr]
+        return batch  # type: ignore[return-value]
 
     def save_batch(self, batch: Sequence) -> None:
         """Save a batch of data to the ArangoDB database.
@@ -195,24 +195,24 @@ def arango_collection_backup() -> None:
     output_file_name = f"./{args.col}_backup.jsonl.gz"
     print(f"Writing backup to '{output_file_name}'...")
 
-    with closing(ArangoClient(hosts=arango_config["hosts"])) as arango_client, gzip.open(  # type: ignore
+    with closing(ArangoClient(hosts=arango_config["hosts"])) as arango_client, gzip.open(  # type: ignore[arg-type]
         output_file_name, "w"
     ) as gzip_out:
         connection = arango_client.db(
-            arango_config["db_name"],  # type: ignore
-            arango_config["username"],  # type: ignore
-            arango_config["password"],  # type: ignore
+            arango_config["db_name"],  # type: ignore[arg-type]
+            arango_config["username"],  # type: ignore[arg-type]
+            arango_config["password"],  # type: ignore[arg-type]
         )
-        jsonlines_writer = jsonlines.Writer(gzip_out)  # type: ignore
+        jsonlines_writer = jsonlines.Writer(gzip_out)  # type: ignore[arg-type]
         try:
             cursor = connection.aql.execute(
                 "FOR doc IN @@coll RETURN doc",
                 bind_vars={"@coll": args.col},
                 batch_size=100,
-                max_runtime=60 * 60,  # type: ignore # 1 hour
+                max_runtime=60 * 60,  # type: ignore[arg-type] # 1 hour
                 stream=True,
             )
             for doc in tqdm(cursor):
                 jsonlines_writer.write(doc)
         finally:
-            cursor.close(ignore_missing=True)  # type: ignore
+            cursor.close(ignore_missing=True)  # type: ignore[union-attr]

mltb2/somajo_transformers.py

@@ -57,12 +57,12 @@ def __call__(self, text: str) -> List[str]:
         sentences = self.somajo_sentence_splitter(text)
         counts = self.transformers_token_counter(sentences)
 
-        assert len(sentences) == len(counts)  # type: ignore  # noqa: S101
+        assert len(sentences) == len(counts)  # type: ignore[arg-type]
 
         result_splits: List[str] = []
         current_sentences: List[str] = []
         current_count: int = 0
-        for sentence, count in zip(tqdm(sentences, disable=not self.show_progress_bar), counts):  # type: ignore
+        for sentence, count in zip(tqdm(sentences, disable=not self.show_progress_bar), counts):  # type: ignore[arg-type]
             if count > self.max_token:
                 if self.ignore_overly_long_sentences:
                     continue

mltb2/text.py

@@ -225,10 +225,10 @@ def fit(self, text: Union[str, Iterable[str]]) -> None:
             raise ValueError("fit must not be called after distance calculation!")
 
         if isinstance(text, str):
-            self._char_counter.update(text)  # type: ignore
+            self._char_counter.update(text)  # type: ignore[union-attr]
         else:
             for t in tqdm(text, disable=not self.show_progress_bar):
-                self._char_counter.update(t)  # type: ignore
+                self._char_counter.update(t)  # type: ignore[union-attr]
 
         self._fit_called = True
 
@@ -238,7 +238,7 @@ def _normalize_char_counter(self) -> None:
         This supports lazy postprocessing of the char counter.
         """
         if not self._distance_called:
-            self._normalized_char_counts = _normalize_counter_to_defaultdict(self._char_counter, self.max_dimensions)  # type: ignore
+            self._normalized_char_counts = _normalize_counter_to_defaultdict(self._char_counter, self.max_dimensions)  # type: ignore[arg-type]
             self._char_counter = None
             self._counted_char_set = set(self._normalized_char_counts)
             self._distance_called = True
@@ -260,9 +260,9 @@ def distance(self, text) -> float:
         text_vector = []
         text_count = Counter(text)
         text_count_defaultdict = _normalize_counter_to_defaultdict(text_count, self.max_dimensions)
-        for c in self._counted_char_set.union(text_count_defaultdict):  # type: ignore
+        for c in self._counted_char_set.union(text_count_defaultdict):  # type: ignore[union-attr]
             all_vector.append(
-                self._normalized_char_counts[c]  # type: ignore
+                self._normalized_char_counts[c]  # type: ignore[index]
             )  # if c is not in defaultdict, it will return 0
             text_vector.append(text_count_defaultdict[c])  # if c is not in defaultdict, it will return 0
         return cityblock(all_vector, text_vector)

pyproject.toml

@@ -168,3 +168,4 @@ notice-rgx = "(# Copyright \\(c\\) \\d{4}.*\\n)+# This software is distributed u
 
 [tool.mypy]
 ignore_missing_imports = true
+enable_error_code=["ignore-without-code"]