fix transcripts parser (#192)

* fix linking to segment 1

it has version 0 and we were only checking 1+

CPCN-598

* avoid html conversion for transcripts

CPCN-607

* revert test change

* avoid any links in transcripts

server/cp/ingest/parser/cp_transcripts.py

@@ -2,13 +2,16 @@
 
 from superdesk import get_resource_service
 from superdesk.io.feed_parsers.ninjs import NINJSFeedParser
-from superdesk.text_utils import plain_text_to_html
 
 
-def get_previous_version(original_ingest_id: str, version_number: int) -> Optional[Dict[str, Any]]:
-    while version_number > 0:
+def get_previous_version(
+    original_ingest_id: str, version_number: int
+) -> Optional[Dict[str, Any]]:
+    while version_number >= 0:
         ingest_id = f"{original_ingest_id}.{version_number}"
-        prev_item = get_resource_service("archive").find_one(req=None, ingest_id=ingest_id)
+        prev_item = get_resource_service("archive").find_one(
+            req=None, ingest_id=ingest_id
+        )
 
         if prev_item is not None:
             return prev_item
@@ -27,12 +30,18 @@ def _transform_from_ninjs(self, ninjs: Dict[str, Any]):
         ninjs["guid"] = f"{original_guid}.{version}"
         item = super()._transform_from_ninjs(ninjs)
         item["version"] = version
-        item["body_html"] = plain_text_to_html(item["body_html"])
-        item.setdefault("extra", {}).update(dict(
-            publish_ingest_id_as_guid=True,
-            cp_version=version,
-            type="transcript",
-        ))
+        item["body_html"] = (
+            item["body_html"]
+            if item["body_html"].strip().startswith("<p>")
+            else "<p>{}</p>".format(item["body_html"])
+        )
+        item.setdefault("extra", {}).update(
+            dict(
+                publish_ingest_id_as_guid=True,
+                cp_version=version,
+                type="transcript",
+            )
+        )
 
         previous_item = get_previous_version(original_guid, version - 1)
         if previous_item is not None:

server/tests/ingest/parser/cp_transcripts.py

@@ -20,7 +20,7 @@ class CP_Transcripts_ParseTestCase(unittest.TestCase):
     def test_parse(self):
         with self.app.app_context(), patch.dict(superdesk.resources, resources):
             superdesk.resources["archive"].service.find_one.side_effect = [
-                {"ingest_id": "d3c8487a-1757-4dde-8bb5-22ca166c1e67.1", "version": 2, "extra": {"ap_version": 999}},
+                {"ingest_id": "d3c8487a-1757-4dde-8bb5-22ca166c1e67.0", "version": 0, "extra": {"ap_version": 999}},
             ]
             items = parser.parse(get_fixture_path("cp_transcripts.json", "cp_transcripts"), provider)
             superdesk.resources["archive"].service.find_one.side_effect = None
@@ -29,8 +29,8 @@ def test_parse(self):
         self.assertEqual("text", item["type"])
         self.assertEqual("transcript", item["extra"]["type"])
         self.assertEqual(True, item["extra"]["publish_ingest_id_as_guid"])
-        self.assertEqual(2, item["extra"]["cp_version"])
-        self.assertEqual("d3c8487a-1757-4dde-8bb5-22ca166c1e67.2", item["guid"])
-        self.assertEqual(2, item["version"])
-        self.assertEqual("d3c8487a-1757-4dde-8bb5-22ca166c1e67.1", item["rewrite_of"])
+        self.assertEqual(1, item["extra"]["cp_version"])
+        self.assertEqual("d3c8487a-1757-4dde-8bb5-22ca166c1e67.1", item["guid"])
+        self.assertEqual(1, item["version"])
+        self.assertEqual("d3c8487a-1757-4dde-8bb5-22ca166c1e67.0", item["rewrite_of"])
         self.assertTrue(item["body_html"].startswith("<p>laying around"))

server/tests/ingest/parser/fixtures/cp_transcripts/cp_transcripts.json

@@ -1,6 +1,6 @@
 {
     "guid": "d3c8487a-1757-4dde-8bb5-22ca166c1e67",
-    "version": "2",
+    "version": "1",
     "type": "text",
     "located": "Toronto, ON",
     "language": "en-CA",