From 41090195a8a0ade2009f4b994601608e0ebc4f94 Mon Sep 17 00:00:00 2001 From: Chao Pang Date: Wed, 2 Oct 2024 10:14:57 -0400 Subject: [PATCH] added the unit column to patient events --- src/cehrbert_data/utils/spark_utils.py | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/src/cehrbert_data/utils/spark_utils.py b/src/cehrbert_data/utils/spark_utils.py index da715ac..e984b6b 100644 --- a/src/cehrbert_data/utils/spark_utils.py +++ b/src/cehrbert_data/utils/spark_utils.py @@ -150,11 +150,7 @@ def join_domain_tables(domain_tables): .withColumn("datetime", datetime_field_udf) ) - if domain_has_unit(sub_domain_table): - pass - else: - pass - + unit_udf = F.col("unit") if domain_has_unit(sub_domain_table) else F.lit(None).cast("string") sub_domain_table = sub_domain_table.select( sub_domain_table["person_id"], sub_domain_table[concept_id_field].alias("standard_concept_id"), @@ -163,6 +159,7 @@ def join_domain_tables(domain_tables): sub_domain_table["visit_occurrence_id"], F.lit(table_domain_field).alias("domain"), F.lit(-1).alias("concept_value"), + unit_udf.alias("unit"), ).distinct() # Remove "Patient Died" from condition_occurrence