add manual fix management commands

tubearchivist/config/management/commands/ta_fix_channels.py

@@ -0,0 +1,44 @@
+"""
+channel fix for update from v0.4.7 to v0.4.8
+reindex channels with 0 subscriber count
+python manage.py ta_fix_channels
+"""
+
+from django.core.management.base import BaseCommand
+from home.src.es.connect import IndexPaginate
+from home.tasks import check_reindex
+
+
+class Command(BaseCommand):
+    """fix comment link"""
+
+    def handle(self, *args, **options):
+        """run command"""
+        self.stdout.write("reindex failed channels")
+        channels = self._get_channels()
+        if not channels:
+            self.stdout.write("did not find any failed channels")
+            return
+
+        self.stdout.write(f"add {len(channels)} channels(s) to queue")
+        to_reindex = {"channel": [i["channel_id"] for i in channels]}
+        check_reindex.delay(data=to_reindex)
+        self.stdout.write(self.style.SUCCESS("    ✓ task queued\n"))
+
+    def _get_channels(self):
+        """get failed channels"""
+        self.stdout.write("search for failed channels")
+        es_query = {
+            "query": {
+                "bool": {
+                    "must": [
+                        {"term": {"channel_subs": {"value": 0}}},
+                        {"term": {"channel_active": {"value": True}}},
+                    ]
+                },
+            },
+            "_source": ["channel_id"],
+        }
+        channels = IndexPaginate("ta_channel", es_query).get_results()
+
+        return channels

tubearchivist/config/management/commands/ta_fix_comment_link.py

@@ -0,0 +1,76 @@
+"""
+comment link fix for update from v0.4.7 to v0.4.8
+scan your videos and comments to fix comment_count field
+python manage.py ta_fix_comment_link
+"""
+
+from django.core.management.base import BaseCommand, CommandError
+from home.src.es.connect import ElasticWrap, IndexPaginate
+
+
+class Command(BaseCommand):
+    """fix comment link"""
+
+    def handle(self, *args, **options):
+        """run command"""
+        self.stdout.write("run comment link fix")
+        expected_count = self._get_comment_indexed()
+        all_videos = self._get_videos()
+
+        self.stdout.write(f"checking {len(all_videos)} video(s)")
+        videos_updated = []
+        for video in all_videos:
+            video_id = video["youtube_id"]
+            comment_count = expected_count.get(video_id)
+            if not comment_count:
+                continue
+
+            data = {"doc": {"comment_count": comment_count}}
+            path = f"ta_video/_update/{video_id}"
+            response, status_code = ElasticWrap(path).post(data=data)
+
+            if status_code != 200:
+                message = (
+                    "failed to add comment count to video"
+                    + f"response code: {status_code}"
+                    + response
+                )
+                raise CommandError(message)
+
+            videos_updated.append(video_id)
+
+        self.stdout.write(f"fixed {len(videos_updated)} video(s)")
+        self.stdout.write(self.style.SUCCESS("    ✓ task completed\n"))
+
+    def _get_comment_indexed(self):
+        """get comment count by index"""
+        self.stdout.write("get comments")
+        src = "params['_source']['comment_comments'].length"
+        data = {
+            "script_fields": {
+                "comments_length": {
+                    "script": {"source": src, "lang": "painless"}
+                }
+            }
+        }
+        all_comments = IndexPaginate(
+            "ta_comment", data=data, keep_source=True
+        ).get_results()
+
+        expected_count = {
+            i["_id"]: i["fields"]["comments_length"][0] for i in all_comments
+        }
+
+        return expected_count
+
+    def _get_videos(self):
+        """get videos without comment_count"""
+        self.stdout.write("get videos")
+        data = {
+            "query": {
+                "bool": {"must_not": [{"exists": {"field": "comment_count"}}]}
+            }
+        }
+        all_videos = IndexPaginate("ta_video", data).get_results()
+
+        return all_videos