change the NINJS ingest parser to use the translations of the respect…

…ive language from the respective CV [SDESK-7169] (#2525)

* change the NINJS ingest parser to use the translations of the respective language from the respective CV [SDESK-7169]

* update types

* update logic for anpa_category and update tests

* fix mypy

* update types

* address comment

* update func to retrive translations attribute

* optimize code

* remove unused imports

superdesk/io/feed_parsers/ninjs.py

@@ -17,6 +17,8 @@
 from superdesk.io.feed_parsers import FeedParser
 from superdesk.utc import utc
 from superdesk.metadata.utils import generate_tag_from_url
+from typing import Optional, Dict, List, Any
+from superdesk import get_resource_service
 
 logger = logging.getLogger(__name__)
 
@@ -97,7 +99,7 @@ def _transform_from_ninjs(self, ninjs):
             item["genre"] = self._format_qcodes(ninjs["genre"])
 
         if ninjs.get("service"):
-            item["anpa_category"] = self._format_qcodes(ninjs["service"])
+            item["anpa_category"] = self._format_qcodes(ninjs["service"], "categories")
 
         if ninjs.get("subject"):
             item["subject"] = self._format_qcodes(ninjs["subject"])
@@ -182,12 +184,23 @@ def parse_renditions(self, renditions):
                 rend[rendition_name] = parsed_rendition
         return rend
 
-    def _format_qcodes(self, items):
+    def _format_qcodes(self, items: List[Dict[str, Any]], cv_name: Optional[str] = None) -> List[Dict[str, Any]]:
         subjects = []
+        cv = get_resource_service("vocabularies").find_one(req=None, _id=cv_name) or {}
+        cv_items = {item["qcode"]: item for item in cv.get("items") or []}
+
         for item in items:
-            subject = {"name": item.get("name"), "qcode": item.get("code")}
-            if item.get("scheme"):
-                subject["scheme"] = item.get("scheme")
+            if cv_items.get(item.get("code")):
+                subject = cv_items[item["code"]]
+            else:
+                subject = {
+                    "name": item.get("name"),
+                    "qcode": item.get("code"),
+                }
+            if not subject.get("translations") and item.get("translations"):
+                subject["translations"] = item["translations"]
+            if not subject.get("scheme") and item.get("scheme"):
+                subject["scheme"] = item["scheme"]
             subjects.append(subject)
 
         return subjects

tests/io/feed_parsers/ninjs_test.py

@@ -15,7 +15,19 @@
 
 
 class NINJSTestCase(TestCase):
-    vocab = [{"_id": "genre", "items": [{"name": "Current"}]}]
+    vocab = [
+        {"_id": "genre", "items": [{"name": "Current"}]},
+        {
+            "_id": "categories",
+            "items": [
+                {
+                    "name": "Advisory",
+                    "qcode": "m",
+                    "translations": {"name": {"en": "Advisory", "fr": "Avis"}},
+                },
+            ],
+        },
+    ]
 
     def setUp(self):
         with self.app.app_context():
@@ -46,6 +58,13 @@ def test_headline(self):
         self.assertEqual(self.items[0]["original_source"], "AAP")
         self.assertEqual("2017-08-24T04:38:34+00:00", self.items[0]["versioncreated"].isoformat())
 
+    def test_translated_value(self):
+        self.assertEqual(self.items[0].get("headline"), "headline")
+        self.assertEqual(
+            self.items[0].get("anpa_category"),
+            [{"name": "Advisory", "qcode": "m", "translations": {"name": {"en": "Advisory", "fr": "Avis"}}}],
+        )
+
 
 class AssociatedTestCase(NINJSTestCase):
     filename = "ninjs2.json"

tests/io/fixtures/ninjs1.json

@@ -28,10 +28,11 @@
     "firstcreated": "2017-08-24T04:37:19+0000",
     "description_html": "<p>abstract</p>",
     "slugline": "slugline",
+    "language": "fr",
     "service": [
         {
-            "code": "i",
-            "name": "International News"
+            "name": "Advisory",
+            "code": "m"
         }
     ],
     "usageterms": "Usage",