Merge pull request #4787 from kobotoolbox/edit-submission-with-xml-encoding-declaration

Fix error when editing submissions with XML encoding declaration

Author: jnm

kpi/deployment_backends/mock_backend.py

@@ -20,7 +20,7 @@
     from backports.zoneinfo import ZoneInfo
 
 from deepmerge import always_merger
-from dict2xml import dict2xml
+from dict2xml import dict2xml as dict2xml_real
 from django.conf import settings
 from django.urls import reverse
 from django.utils.translation import gettext as t
@@ -50,6 +50,11 @@
 from ..exceptions import KobocatBulkUpdateSubmissionsClientException
 
 
+def dict2xml(*args, **kwargs):
+    """ To facilitate mocking in unit tests """
+    return dict2xml_real(*args, **kwargs)
+
+
 class MockDeploymentBackend(BaseDeploymentBackend):
     """
     Only used for unit testing and interface testing.

kpi/tests/api/v2/test_api_submissions.py

@@ -2,6 +2,7 @@
 import copy
 import json
 import lxml
+import mock
 import random
 import string
 import time
@@ -14,6 +15,7 @@
 
 import pytest
 import responses
+from dict2xml import dict2xml
 from django.conf import settings
 from django.contrib.auth.models import User
 from django.urls import reverse
@@ -1247,6 +1249,47 @@ def test_edit_submission_with_different_root_name(self):
         )
         assert new_snapshot.pk != snapshot.pk
 
+    @responses.activate
+    def test_edit_submission_with_xml_encoding_declaration(self):
+        def dict2xml_with_encoding_declaration(*args, **kwargs):
+            return '<?xml version="1.0" encoding="utf-8"?>' + dict2xml(
+                *args, **kwargs
+            )
+
+        with mock.patch(
+            'kpi.deployment_backends.mock_backend.dict2xml'
+        ) as mock_dict2xml:
+            mock_dict2xml.side_effect = dict2xml_with_encoding_declaration
+            submission = self.submissions[-1]
+            submission_xml = self.asset.deployment.get_submissions(
+                user=self.asset.owner,
+                format_type=SUBMISSION_FORMAT_TYPE_XML,
+                submission_ids=[submission['_id']],
+            )[0]
+            assert submission_xml.startswith(
+                '<?xml version="1.0" encoding="utf-8"?>'
+            )
+
+            # Get edit endpoint
+            edit_url = reverse(
+                self._get_endpoint('submission-enketo-edit'),
+                kwargs={
+                    'parent_lookup_asset': self.asset.uid,
+                    'pk': submission['_id'],
+                },
+            )
+
+            # Set up a mock Enketo response and attempt the edit request
+            ee_url = f'{settings.ENKETO_URL}/{settings.ENKETO_EDIT_INSTANCE_ENDPOINT}'
+            responses.add_callback(
+                responses.POST,
+                ee_url,
+                callback=enketo_edit_instance_response_with_uuid_validation,
+                content_type='application/json',
+            )
+            response = self.client.get(edit_url, {'format': 'json'})
+            assert response.status_code == status.HTTP_200_OK
+
     @responses.activate
     def test_edit_submission_with_xml_missing_uuids(self):
         # Make a new submission without UUIDs

kpi/views/v2/data.py

@@ -655,6 +655,13 @@ def _get_enketo_link(
         submission_xml = deployment.get_submission(
             submission_id, user, SUBMISSION_FORMAT_TYPE_XML
         )
+        if isinstance(submission_xml, str):
+            # Workaround for "Unicode strings with encoding declaration are not
+            # supported. Please use bytes input or XML fragments without
+            # declaration."
+            # TODO: handle this in a unified way instead of haphazardly. See,
+            # e.g., `kpi.utils.xml.strip_nodes()`
+            submission_xml = submission_xml.encode()
         submission_xml_root = etree.fromstring(submission_xml)
         # The JSON version is needed to detect its version
         submission_json = deployment.get_submission(