Switch csv generation to QUOTE_MINIMAL

smart-on-fhir · Jun 24, 2024 · f7732b8 · f7732b8
1 parent b9c757a
commit f7732b8
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 5 deletions.
diff --git a/src/handlers/site_upload/powerset_merge.py b/src/handlers/site_upload/powerset_merge.py
@@ -228,8 +228,6 @@ def generate_csv_from_parquet(bucket_name: str, bucket_root: str, subbucket_path
     last_valid_df = last_valid_df.apply(
         lambda x: x.strip() if isinstance(x, str) else x
     ).replace('""', numpy.nan)
-    # Here we are removing internal commas from fields so we get a valid unquoted CSV
-    last_valid_df = last_valid_df.replace(to_replace=",", value="", regex=True)
     awswrangler.s3.to_csv(
         last_valid_df,
         (
@@ -238,7 +236,7 @@ def generate_csv_from_parquet(bucket_name: str, bucket_root: str, subbucket_path
             )
         ),
         index=False,
-        quoting=csv.QUOTE_NONE,
+        quoting=csv.QUOTE_MINIMAL,
     )
 
 

diff --git a/tests/site_upload/test_powerset_merge.py b/tests/site_upload/test_powerset_merge.py
@@ -377,7 +377,7 @@ def test_parquet_to_csv(mock_bucket):
     )
     assert list(df["race"].dropna().unique()) == [
         "White",
-        "Black or African American",
+        "Black, or African American",
         "Asian",
-        "American Indian or Alaska Native",
+        "American Indian, or Alaska Native",
     ]