Catch UnicodeDecodeError and retry (#138)

biothings · Jul 16, 2022 · 79e09de · 79e09de
1 parent cf3cb23
commit 79e09de
Showing 1 changed file with 5 additions and 2 deletions.
diff --git a/src/hub/dataload/sources/unii/unii_parser.py b/src/hub/dataload/sources/unii/unii_parser.py
@@ -3,8 +3,11 @@
 
 
 def load_data(input_file):
-
-    unii = pd.read_csv(input_file, sep='\t', low_memory=False, dtype=str)
+    try:
+        unii = pd.read_csv(input_file, sep='\t', low_memory=False, dtype=str)
+    except UnicodeDecodeError:
+        unii = pd.read_csv(input_file, sep='\t', low_memory=False, dtype=str, encoding='windows-1252')
+
     unii.rename(columns={'MF': 'molecular_formula',
                          'PT': 'preferred_term',
                          'RN': 'registry_number'}, inplace=True)