[DOP-13846] - add bypass test and note for rootTag

onetl/file/format/xml.py

@@ -238,6 +238,11 @@ def parse_column(self, column: str | Column, schema: StructType) -> Column:
 
             This method assumes that the ``spark-xml`` package is installed and properly configured within your Spark environment.
 
+        .. note::
+
+            This method does not support XML strings with a root tag that is not specified as the ``rowTag``. If your XML data includes a root tag that encapsulates multiple row tags, ensure to preprocess the XML string to remove or ignore the root tag before parsing.
+
+
         Parameters
         ----------
         column : str | Column

tests/tests_integration/test_file_format_integration/test_xml_integration.py

@@ -175,6 +175,9 @@ def test_xml_parse_column(
     file_df_dataframe,
     file_df_schema,
 ):
+    from pyspark.sql.functions import col, expr
+    from pyspark.sql.types import StructType
+
     from onetl.file.format import XML
 
     spark_version = get_spark_version(spark)
@@ -188,8 +191,11 @@ def test_xml_parse_column(
         xml_data = file.read()
 
     df = spark.createDataFrame([(xml_data,)], ["xml_string"])
-    df.show(truncate=False)
-    xml = XML.parse({"rowTag": "item", "rootTag": "root"})
-    parsed_df = df.select(xml.parse_column("xml_string", schema=file_df_schema))
+    # remove the <root> tag from the XML string
+    df = df.withColumn("xml_string", expr("regexp_replace(xml_string, '^<root>|</root>$', '')"))
 
-    parsed_df.show(truncate=False)
+    xml = XML(row_tag="item")
+    parsed_df = df.select(xml.parse_column("xml_string", schema=file_df_schema))
+    assert isinstance(parsed_df.schema, StructType)
+    # check that there are no columns with only null values
+    assert parsed_df.filter(col("xml_string").isNotNull()).count() > 0