Fix importlib DeprecationWarning (#1584)

* Update importlib API

* resolve broken test with Python3.11 error message

* fix some FutureWarnings

docs/source/release_notes.rst

@@ -3,15 +3,17 @@
 Release Notes
 -------------
 
-.. Future Release
-  ==============
+Future Release
+==============
     * Enhancements
     * Fixes
+        * Fix importlib DeprecationWarning in ``inference_functions.py`` (:pr:`1584`)
     * Changes
     * Documentation Changes
     * Testing Changes
 
-.. Thanks to the following people for contributing to this release:
+    Thanks to the following people for contributing to this release:
+    :user:`sbadithe`
 
 v0.21.0 December 1, 2022
 ========================

pyproject.toml

@@ -40,6 +40,7 @@ dependencies = [
     "scikit-learn >= 0.22",
     "python-dateutil >= 2.8.1",
     "scipy >= 1.4.0",
+    "importlib-resources >= 5.10.0"
 ]
 
 [project.urls]

woodwork/tests/accessor/test_statistics.py

@@ -58,6 +58,7 @@
 from woodwork.tests.testing_utils import (
     _check_close,
     check_empty_box_plot_dict,
+    concat_dataframe_or_series,
     dep_between_cols,
     to_pandas,
 )
@@ -2087,7 +2088,10 @@ def test_get_low_high_bound_warnings():
 
 def test_get_medcouple(outliers_df_pandas, skewed_outliers_df_pandas):
     has_outliers_series = outliers_df_pandas["has_outliers"]
-    has_outliers_series = has_outliers_series.append(pd.Series([39]), ignore_index=True)
+    has_outliers_series = pd.concat(
+        [has_outliers_series, pd.Series([39], dtype="int64")],
+        ignore_index=True,
+    )
     has_outliers_series.ww.init()
     mc = _get_medcouple_statistic(has_outliers_series)
     assert mc == 0.122
@@ -2122,7 +2126,8 @@ def test_determine_best_outlier_method_sampling_outcome(skewed_outliers_df_panda
 
 
 def test_determine_best_outlier_method_equivalent_outcome(
-    outliers_df_pandas, skewed_outliers_df_pandas
+    outliers_df_pandas,
+    skewed_outliers_df_pandas,
 ):
     contains_nans_series_skewed = skewed_outliers_df_pandas["right_skewed_outliers"]
     contains_nans_series_skewed.ww.init()
@@ -2313,7 +2318,7 @@ def test_spearman_ordinal(df_mi, use_ordinal):
     else:
         df_mi.ww.init()
     sp = df_mi.ww.dependence(measures=["spearman"])
-    valid_sp_columns = (sp.column_1.append(sp.column_2)).unique()
+    valid_sp_columns = concat_dataframe_or_series(sp.column_1, sp.column_2).unique()
     assert "strs" not in valid_sp_columns
     if use_ordinal:
         assert "strs2" in valid_sp_columns

woodwork/tests/accessor/test_table_accessor.py

@@ -2456,8 +2456,7 @@ def test_accessor_schema_properties(sample_df):
         assert prop_from_accessor == prop_from_schema
 
         # Assumes we don't have setters for any of these attributes
-        error = "can't set attribute"
-        with pytest.raises(AttributeError, match=error):
+        with pytest.raises(AttributeError):
             setattr(sample_df.ww, schema_property, "new_value")
 
 

woodwork/tests/fixtures/datetime_freq.py

@@ -50,8 +50,8 @@
 def pad_datetime_series(dates, freq, pad_start=0, pad_end=100):
     dates = [pd.Timestamp(d) for d in dates]
 
-    head = pd.Series([])
-    tail = pd.Series([])
+    head = pd.Series([], dtype="datetime64[ns]")
+    tail = pd.Series([], dtype="datetime64[ns]")
 
     if pad_start > 0:
         head = (

woodwork/tests/requirement_files/minimum_core_requirements.txt

@@ -2,3 +2,4 @@ pandas==1.4.0
 python-dateutil==2.8.1
 scikit-learn==0.22
 scipy==1.4.0
+importlib-resources==5.10.0

woodwork/tests/requirement_files/minimum_dask_requirements.txt

@@ -3,3 +3,4 @@ pandas==1.4.0
 python-dateutil==2.8.1
 scikit-learn==0.22
 scipy==1.4.0
+importlib-resources==5.10.0

woodwork/tests/requirement_files/minimum_spark_requirements.txt

@@ -3,3 +3,4 @@ pyspark==3.2.0
 python-dateutil==2.8.1
 scikit-learn==0.22
 scipy==1.4.0
+importlib-resources==5.10.0

woodwork/type_sys/inference_functions.py

@@ -1,13 +1,12 @@
 import sys
-from importlib import resources as pkg_resources
 from typing import Any, Callable, Iterable, Union
 
 import numpy as np
 import pandas as pd
+from importlib_resources import files
 from pandas.api import types as pdtypes
 
 import woodwork as ww
-from woodwork import data
 from woodwork.config import config
 from woodwork.type_sys.utils import _is_categorical_series, col_is_datetime
 
@@ -16,7 +15,9 @@
 Tokens = Iterable[str]
 
 COMMON_WORDS_SET = set(
-    line.strip().lower() for line in pkg_resources.open_text(data, "1-1000.txt")
+    word.strip().lower()
+    for word in files("woodwork.data").joinpath("1-1000.txt").read_text().split("\n")
+    if len(word) > 0
 )
 
 NL_delimiters = r"[- \[\].,!\?;\n]"