str checks use plain string instead of re.Pattern (unionai-oss#1729)

Signed-off-by: cosmicBboy <niels.bantilan@gmail.com>

pandera/api/checks.py

@@ -436,7 +436,7 @@ def str_matches(cls, pattern: Union[str, re.Pattern], **kwargs) -> "Check":
         :param kwargs: key-word arguments passed into the `Check` initializer.
         """
         try:
-            pattern_mod = re.compile(pattern)
+            re.compile(pattern)
         except TypeError as exc:
             raise ValueError(
                 f'pattern="{pattern}" cannot be compiled as regular expression'
@@ -446,7 +446,7 @@ def str_matches(cls, pattern: Union[str, re.Pattern], **kwargs) -> "Check":
             kwargs,
             error=f"str_matches('{pattern}')",
             statistics={"pattern": pattern},
-            pattern=pattern_mod,
+            pattern=pattern,
         )
 
     @classmethod
@@ -459,7 +459,7 @@ def str_contains(
         :param kwargs: key-word arguments passed into the `Check` initializer.
         """
         try:
-            pattern_mod = re.compile(pattern)
+            re.compile(pattern)
         except TypeError as exc:
             raise ValueError(
                 f'pattern="{pattern}" cannot be compiled as regular expression'
@@ -469,7 +469,7 @@ def str_contains(
             kwargs,
             error=f"str_contains('{pattern}')",
             statistics={"pattern": pattern},
-            pattern=pattern_mod,
+            pattern=pattern,
         )
 
     @classmethod

pandera/backends/polars/builtin_checks.py

@@ -204,7 +204,7 @@ def str_matches(
 )
 def str_contains(
     data: PolarsData,
-    pattern: re.Pattern,
+    pattern: Union[str, re.Pattern],
 ) -> pl.LazyFrame:
     """Ensure that a pattern can be found in the string.
 

pandera/backends/pyspark/builtin_checks.py

@@ -1,6 +1,5 @@
 """PySpark implementation of built-in checks"""
 
-import re
 from typing import Any, Iterable, TypeVar
 
 import pyspark.sql.types as pst
@@ -280,9 +279,7 @@ def notin(
     error="str_contains('{pattern}')",
 )
 @register_input_datatypes(acceptable_datatypes=convert_to_list(STRING_TYPE))
-def str_contains(
-    data: PysparkDataframeColumnObject, pattern: re.Pattern
-) -> bool:
+def str_contains(data: PysparkDataframeColumnObject, pattern: str) -> bool:
     """Ensure that a pattern can be found within each row.
 
     Remember it can be a compute intensive check on large dataset. So, use it with caution.
@@ -291,9 +288,8 @@ def str_contains(
                 to access the dataframe is "dataframe" and column name using "column_name".
     :param pattern: Regular expression pattern to use for searching
     """
-
     return (
-        data.dataframe.filter(~col(data.column_name).rlike(pattern.pattern))
+        data.dataframe.filter(~col(data.column_name).rlike(pattern))
         .limit(1)
         .count()
         == 0