remove regexp_extract_all

quinn/__init__.py

@@ -24,7 +24,6 @@
     exists,
     forall,
     multi_equals,
-    regexp_extract_all,
     remove_all_whitespace,
     remove_non_word_characters,
     single_space,
@@ -75,7 +74,6 @@
     "week_start_date",
     "week_end_date",
     "approx_equal",
-    "regexp_extract_all",
     "business_days_between",
     "uuid5",
     "with_columns_renamed",

quinn/functions.py

@@ -230,20 +230,6 @@ def array_choice(col: Column, seed: int | None = None) -> Column:
     return col[index]
 
 
-@F.udf(returnType=ArrayType(StringType()))
-def regexp_extract_all(s: Column, regexp: Column) -> Column:
-    """Function uses the Python `re` library to extract regular expressions from a string (`s`) using a regex pattern (`regexp`).
-
-    It returns a list of all matches, or    `None` if `s` is `None`.
-
-    :param s: input string (`Column`)
-    :type s: str
-    :param regexp: string `re` pattern
-    :rtype: Column
-    """
-    return None if s is None else re.findall(regexp, s)
-
-
 def business_days_between(
     start_date: Column, end_date: Column, # noqa: ARG001
 ) -> Column:

tests/test_functions.py

@@ -329,20 +329,6 @@ def it_works_with_integer_values():
 #     chispa.assert_column_equality(actual_df, "random_letter", "expected")
 
 
-def test_regexp_extract_all():
-    df = quinn.create_df(
-        spark,
-        [("200 - 300 PA.", ["200", "300"]), ("400 PA.", ["400"]), (None, None)],
-        [
-            ("str", StringType(), True),
-            ("expected", ArrayType(StringType(), True), True),
-        ],
-    )
-    actual_df = df.withColumn(
-        "all_numbers", quinn.regexp_extract_all(F.col("str"), F.lit(r"(\d+)"))
-    )
-    chispa.assert_column_equality(actual_df, "all_numbers", "expected")
-
 
 def test_business_days_between():
     df = quinn.create_df(