fix function names

ikuyamada · ikuyamada · commit bf9be0e19dbd · 2024-01-02T17:15:33.000+09:00
diff --git a/wikipedia2vec/dictionary.py b/wikipedia2vec/dictionary.py
@@ -193,7 +193,7 @@ def build(
         entity_counter = Counter()
         entity_doc_counter = Counter()
 
-        with closing(Pool(pool_size, initializer=init_worker, initargs=(dump_db, tokenizer))) as pool:
+        with closing(Pool(pool_size, initializer=_init_worker, initargs=(dump_db, tokenizer))) as pool:
             with tqdm(total=dump_db.page_size(), mininterval=0.5, disable=not progressbar) as bar:
                 f = partial(_process_page, lowercase=lowercase, min_paragraph_len=min_paragraph_len)
                 for word_cnt, entity_cnt in pool.imap_unordered(f, dump_db.titles(), chunksize=chunk_size):
@@ -346,7 +346,7 @@ def load(target: Union[str, dict], mmap: bool = True) -> "Dictionary":
 _tokenizer: Optional[BaseTokenizer] = None
 
 
-def init_worker(dump_db: DumpDB, tokenizer: BaseTokenizer):
+def _init_worker(dump_db: DumpDB, tokenizer: BaseTokenizer):
     global _dump_db, _tokenizer
 
     _dump_db = dump_db
diff --git a/wikipedia2vec/link_graph.py b/wikipedia2vec/link_graph.py
@@ -82,7 +82,7 @@ def build(
         logger.info("Step 1/2: Processing Wikipedia pages...")
 
         with closing(
-            Pool(pool_size, initializer=init_worker, initargs=(dump_db, dictionary.serialize(shared_array=True)))
+            Pool(pool_size, initializer=_init_worker, initargs=(dump_db, dictionary.serialize(shared_array=True)))
         ) as pool:
             rows = []
             cols = []
@@ -131,7 +131,7 @@ def build(
 _dictionary: Optional[Dictionary] = None
 
 
-def init_worker(dump_db: DumpDB, dictionary_obj: dict):
+def _init_worker(dump_db: DumpDB, dictionary_obj: dict):
     global _dump_db, _dictionary
 
     _dump_db = dump_db
diff --git a/wikipedia2vec/mention_db.py b/wikipedia2vec/mention_db.py
@@ -169,7 +169,7 @@ def build(
         name_dict = defaultdict(lambda: Counter())
         init_args = [dump_db, dictionary.serialize(shared_array=True), tokenizer, None]
 
-        with closing(Pool(pool_size, initializer=init_worker, initargs=init_args)) as pool:
+        with closing(Pool(pool_size, initializer=_init_worker, initargs=init_args)) as pool:
             with tqdm(total=dump_db.page_size(), mininterval=0.5, disable=not progressbar) as bar:
                 f = partial(_extract_links, max_mention_len=max_mention_len, case_sensitive=case_sensitive)
                 for ret in pool.imap_unordered(f, dump_db.titles(), chunksize=chunk_size):
@@ -184,7 +184,7 @@ def build(
         name_counter = Counter()
         init_args[3] = name_trie
 
-        with closing(Pool(pool_size, initializer=init_worker, initargs=init_args)) as pool:
+        with closing(Pool(pool_size, initializer=_init_worker, initargs=init_args)) as pool:
             with tqdm(total=dump_db.page_size(), mininterval=0.5, disable=not progressbar) as bar:
                 f = partial(_count_occurrences, max_mention_len=max_mention_len, case_sensitive=case_sensitive)
                 for names in pool.imap_unordered(f, dump_db.titles(), chunksize=chunk_size):
@@ -264,7 +264,7 @@ def load(target: Union[str, dict], dictionary: Dictionary) -> "MentionDB":
 _name_trie: Optional[Trie] = None
 
 
-def init_worker(dump_db: DumpDB, dictionary_obj: dict, tokenizer: BaseTokenizer, name_trie: Trie = None):
+def _init_worker(dump_db: DumpDB, dictionary_obj: dict, tokenizer: BaseTokenizer, name_trie: Trie = None):
     global _dump_db, _dictionary, _tokenizer, _name_trie
 
     _dump_db = dump_db
