REFACTOR-#5332: define PQ_INDEX_REGEX as class variable (#5333)

Signed-off-by: Myachev <anatoly.myachev@intel.com>

modin/core/io/column_stores/parquet_dispatcher.py

@@ -14,8 +14,9 @@
 """Module houses `ParquetDispatcher` class, that is used for reading `.parquet` files."""
 
 import os
-
+import re
 import json
+
 import fsspec
 from fsspec.core import url_to_fs
 from fsspec.spec import AbstractBufferedFile
@@ -292,6 +293,8 @@ def _get_fastparquet_files(self):  # noqa: GL08
 class ParquetDispatcher(ColumnStoreDispatcher):
     """Class handles utils for reading `.parquet` files."""
 
+    index_regex = re.compile(r"__index_level_\d+__")
+
     @classmethod
     def get_dataset(cls, path, engine, storage_options):
         """
@@ -603,8 +606,6 @@ def _read(cls, path, engine, columns, **kwargs):
         ParquetFile API is used. Please refer to the documentation here
         https://arrow.apache.org/docs/python/parquet.html
         """
-        from modin.pandas.io import PQ_INDEX_REGEX
-
         if isinstance(path, str):
             if os.path.isdir(path):
                 path_generator = os.walk(path)
@@ -650,7 +651,7 @@ def _read(cls, path, engine, columns, **kwargs):
         columns = [
             c
             for c in column_names
-            if c not in index_columns and not PQ_INDEX_REGEX.match(c)
+            if c not in index_columns and not cls.index_regex.match(c)
         ]
 
         return cls.build_query_compiler(dataset, columns, index_columns, **kwargs)

modin/pandas/io.py

@@ -44,7 +44,6 @@
 )
 import pathlib
 import pickle
-import re
 from typing import (
     Union,
     IO,
@@ -69,8 +68,6 @@
 from modin.utils import _inherit_docstrings, Engine
 from . import _update_engine
 
-PQ_INDEX_REGEX = re.compile(r"__index_level_\d+__")
-
 
 def _read(**kwargs):
     """