From 4ded8988c3e2e87296694062a476aaa7cb73f4cb Mon Sep 17 00:00:00 2001
From: Simon <simobilleter@gmail.com>
Date: Wed, 30 Aug 2023 18:42:03 +0700
Subject: [PATCH] add channel aggregation

---
 tubearchivist/api/src/aggs.py | 49 +++++++++++++++++++++++++++++++++++
 tubearchivist/api/urls.py     |  5 ++++
 tubearchivist/api/views.py    | 21 ++++++++++++++-
 3 files changed, 74 insertions(+), 1 deletion(-)

diff --git a/tubearchivist/api/src/aggs.py b/tubearchivist/api/src/aggs.py
index 08259146..a829469c 100644
--- a/tubearchivist/api/src/aggs.py
+++ b/tubearchivist/api/src/aggs.py
@@ -124,3 +124,52 @@ class DownloadHist(AggBase):
         buckets = aggregations[self.name]["buckets"]
 
         return {i.get("key_as_string"): i.get("doc_count") for i in buckets}
+
+
+class BiggestChannel(AggBase):
+    """get channel aggregations"""
+
+    name = "channel_stats"
+    path = "ta_video/_search"
+    data = {
+        "size": 0,
+        "aggs": {
+            name: {
+                "multi_terms": {
+                    "terms": [
+                        {"field": "channel.channel_name.keyword"},
+                        {"field": "channel.channel_id"},
+                    ],
+                    "order": {"doc_count": "desc"},
+                },
+                "aggs": {
+                    "doc_count": {"value_count": {"field": "_index"}},
+                    "duration": {"sum": {"field": "player.duration"}},
+                    "media_size": {"sum": {"field": "media_size"}},
+                },
+            },
+        },
+    }
+    order_choices = ["doc_count", "duration", "media_size"]
+
+    def process(self, order_by=False):
+        """process aggregation"""
+
+        if order_by and order_by in self.order_choices:
+            self.data["aggs"][self.name]["multi_terms"]["order"] = order_by
+
+        aggregations = self.get()
+        buckets = aggregations[self.name]["buckets"]
+
+        response = [
+            {
+                "id": i["key"][1],
+                "name": i["key"][0].title(),
+                "doc_count": i["doc_count"]["value"],
+                "duration": i["duration"]["value"],
+                "media_size": i["media_size"]["value"],
+            }
+            for i in buckets
+        ]
+
+        return response
diff --git a/tubearchivist/api/urls.py b/tubearchivist/api/urls.py
index 1e12ad78..6ad41638 100644
--- a/tubearchivist/api/urls.py
+++ b/tubearchivist/api/urls.py
@@ -151,4 +151,9 @@ urlpatterns = [
         views.StatDownloadHist.as_view(),
         name="api-stats-downloadhist",
     ),
+    path(
+        "stats/biggestchannels/",
+        views.StatBiggestChannel.as_view(),
+        name="api-stats-biggestchannels",
+    ),
 ]
diff --git a/tubearchivist/api/views.py b/tubearchivist/api/views.py
index 704e53a4..cf9521f4 100644
--- a/tubearchivist/api/views.py
+++ b/tubearchivist/api/views.py
@@ -1,6 +1,6 @@
 """all API views"""
 
-from api.src.aggs import DownloadHist, Primary, WatchProgress
+from api.src.aggs import BiggestChannel, DownloadHist, Primary, WatchProgress
 from api.src.search_processor import SearchProcess
 from home.src.download.queue import PendingInteract
 from home.src.download.subscriptions import (
@@ -1012,3 +1012,22 @@ class StatDownloadHist(ApiBaseView):
         # pylint: disable=unused-argument
 
         return Response(DownloadHist().process())
+
+
+class StatBiggestChannel(ApiBaseView):
+    """resolves to /api/stats/biggestchannels/
+    GET: return biggest channels
+    param: order
+    """
+
+    order_choices = ["doc_count", "duration", "media_size"]
+
+    def get(self, request):
+        """handle get request"""
+
+        order = request.GET.get("order")
+        if order and order not in self.order_choices:
+            message = {"message": f"invalid order parameter {order}"}
+            return Response(message, status=400)
+
+        return Response(BiggestChannel().process())