Merge pull request #284 from datosgobar/283-explicit-error-on-missing…

…-xlsx-field 283 explicit error on missing xlsx field
datosgobar · Oct 4, 2019 · 879da46 · 879da46
2 parents c284204 + d14dba0
commit 879da46
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 8 deletions.
diff --git a/pydatajson/readers.py b/pydatajson/readers.py
@@ -91,8 +91,8 @@ def read_catalog(catalog, default_values=None, catalog_format=None,
                 catalog_dict = read_xlsx_catalog(catalog,
                                                  verify=verify,
                                                  timeout=timeout)
-            except openpyxl_exceptions + \
-                    (ValueError, AssertionError, IOError, BadZipfile) as e:
+            except openpyxl_exceptions + (KeyError, ValueError, AssertionError,
+                                          IOError, BadZipfile) as e:
                 raise ce.NonParseableCatalog(catalog, str(e))
         elif catalog_format == "json":
             try:
@@ -459,15 +459,15 @@ def read_local_xlsx_catalog(xlsx_path, logger=None):
                 '{}' (fila #{} de la hoja "Field"). Este campo no figurara en
                 el data.json de salida.""".format(
                     unidecode(field["dataset_title"]),
-                    unidecode(field["field_title"]), idx + 2))
+                    unidecode(field.get("field_title", "sin title")), idx + 2))
 
         elif distribution_index is None:
             logger.warning(
                 """No se encontro la distribucion '{}' especificada para el
                 campo'{}' (fila #{} de la hoja "Field"). Este campo no figurara
                 en el data.json de salida.""".format(
                     unidecode(field["distribution_title"]),
-                    unidecode(field["field_title"]), idx + 2))
+                    unidecode(field.get("field_title", )), idx + 2))
 
         else:
             dataset = catalog["catalog_dataset"][dataset_index]
@@ -539,10 +539,11 @@ def read_suffixless_catalog(catalog):
     try:
         catalog_dict = read_xlsx_catalog(catalog)
         return catalog_dict
-    except openpyxl_exceptions + (ValueError, AssertionError,
-                                  IOError, BadZipfile):
-        raise ce.NonParseableCatalog(
-            catalog, 'No es posible discernir el formato del catalogo')
+    except openpyxl_exceptions + (KeyError, ValueError, AssertionError,
+                                  IOError, BadZipfile) as e:
+        msg = 'No es posible discernir el formato del catalogo: {}'\
+            .format(str(e))
+        raise ce.NonParseableCatalog(catalog, msg)
 
 
 def read_table(path):

diff --git a/tests/samples/catalogo-justicia-missing-distribution-identifier.xlsx b/tests/samples/catalogo-justicia-missing-distribution-identifier.xlsx
diff --git a/tests/test_readers_and_writers.py b/tests/test_readers_and_writers.py
@@ -322,6 +322,12 @@ def test_xlsx_write_missing_optional_fields_and_themes(self):
         self.assertTrue('theme' not in written_dataset)
         self.assertTrue('field' not in written_distribution)
 
+    @nose.tools.raises(NonParseableCatalog)
+    def test_missing_mandatory_field_on_xlsx_catalog_raises(self):
+        sample = self.get_sample(
+            'catalogo-justicia-missing-distribution-identifier.xlsx')
+        pydatajson.readers.read_catalog(sample)
+
 
 if __name__ == '__main__':
     nose.run(defaultTest=__name__)