fix for #68

datadotworld/models/query.py

@@ -102,6 +102,24 @@ def _iter_rows(self):
                             values.append(None)
 
                     table_row = schema_obj.cast_row(values)
+
+                    # when the column is a string value, the jsontableschema
+                    # library is incorrectly mapping the several literal
+                    # string values ('null', 'none', '-', etc.) to the python
+                    # `None` value - a deeper fix might be to reconsider using
+                    # that library, or maybe fixing this issue in that
+                    # library (since it's probably not a good idea to render
+                    # a number of strings un-representable) - this fixes the
+                    # problem for our result sets.  Essentially, this zips
+                    # over each result set and checks whether we mapped a
+                    # non-null value to `None` in a string field, and if
+                    # so it restores the non-null value before continuing
+                    table_row = map(lambda field, original, mapped:
+                                    original if (not mapped) and original
+                                       and field.type == 'string'
+                                    else mapped,
+                                    schema_obj.fields, values, table_row)
+
                     yield OrderedDict(zip(field_names, table_row))
             elif 'boolean' in self.raw_data:
                 # Results of an ASK query

tests/datadotworld/models/test_query.py

@@ -94,5 +94,15 @@ def test_dataframe(self, query_result_example, query_results):
                     equal_to((len(query_result_example['results']['bindings']),
                               len(metadata_names))))
 
+    def test_str_column(self, query_results):
+        '''
+        Test added for https://github.com/datadotworld/data.world-py/issues/68
+        added a value in the `st.name` column in sql_select.json of "NONE",
+        which was previously mapped to Python `None` - this verifies that we
+        have fixed that issue.
+        '''
+        for value in query_results.dataframe['st.name']:
+            assert_that(value)
+
     def test_str(self, query_results, query_result_example):
         assert_that(str(query_results), equal_to(str(query_result_example)))

tests/fixtures/queries/sql_select.json

@@ -589,7 +589,7 @@
         },
         "v_5": {
           "type": "literal",
-          "value": "UT West Mall @ Guadalupe"
+          "value": "NONE"
         },
         "v_6": {
           "type": "literal",