Merge 9b8f948 into c99effd

datosgobar · Aug 1, 2018 · cfc5998 · cfc5998
2 parents c99effd + 9b8f948
commit cfc5998
Show file tree

Hide file tree

Showing 4 changed files with 16 additions and 4 deletions.
diff --git a/pydatajson/helpers.py b/pydatajson/helpers.py
@@ -233,9 +233,9 @@ def sheet_to_table(worksheet):
     return table
 
 
-def string_to_list(string, sep=","):
+def string_to_list(string, sep=",", filter_empty=False):
     """Transforma una string con elementos separados por `sep` en una lista."""
-    return [value.strip() for value in string.split(sep)]
+    return [value.strip() for value in string.split(sep) if (not filter_empty or value)]
 
 
 def add_dicts(one_dict, other_dict):

diff --git a/pydatajson/readers.py b/pydatajson/readers.py
@@ -456,14 +456,14 @@ def read_local_xlsx_catalog(xlsx_path, logger=None):
     # Transformo campos de texto separado por comas en listas
     if "catalog_language" in catalog:
         catalog["catalog_language"] = helpers.string_to_list(
-            catalog["catalog_language"])
+            catalog["catalog_language"], filter_empty=True)
 
     for dataset in catalog["catalog_dataset"]:
         array_fields = ["dataset_superTheme", "dataset_theme", "dataset_tags",
                         "dataset_keyword", "dataset_language"]
         for field in array_fields:
             if field in dataset:
-                dataset[field] = helpers.string_to_list(dataset[field])
+                dataset[field] = helpers.string_to_list(dataset[field], filter_empty=True)
 
     # Elimino los prefijos de los campos a nivel catálogo
     _remove_prefixes_and_unprefixed_fields(catalog, 'catalog_')

diff --git a/tests/samples/lists_extra_commas.xlsx b/tests/samples/lists_extra_commas.xlsx
diff --git a/tests/test_readers_and_writers.py b/tests/test_readers_and_writers.py
@@ -249,6 +249,18 @@ def test_read_write_both_formats_yields_the_same(self):
 
             self.assertEqual(catalog_json, catalog)
             self.assertEqual(catalog_xlsx, catalog)
+
+    def test_read_xlsx_lists_with_extra_commas(self):
+        # No hay valores vacíos a pesar que hay listas con comas extras
+        catalog = pydatajson.readers.read_catalog(self.get_sample("lists_extra_commas.xlsx"))
+        self.assertTrue(catalog['language'])
+        self.assertTrue(all(catalog['language']))
+        for dataset in catalog['dataset']:
+            for field in ['theme', 'superTheme', 'keyword']:
+                # Listas no vacias
+                self.assertTrue(dataset[field])
+                # Elementos no vacios
+                self.assertTrue(all(dataset[field]))
 
 
 if __name__ == '__main__':