From facfcf1b3c8a966b932706515185cb0fe4a4d5aa Mon Sep 17 00:00:00 2001
From: 211250101 <211250101@smail.nju.edu.cn>
Date: Sun, 17 Dec 2023 20:50:27 +0800
Subject: [PATCH] add to YouTube scrape

---
 .../youtube/Ubuntu/get_ubuntu.py              | 52 +++++++++++++++++++
 1 file changed, 52 insertions(+)
 create mode 100644 resouce_collection/youtube/Ubuntu/get_ubuntu.py

diff --git a/resouce_collection/youtube/Ubuntu/get_ubuntu.py b/resouce_collection/youtube/Ubuntu/get_ubuntu.py
new file mode 100644
index 0000000..3e2949c
--- /dev/null
+++ b/resouce_collection/youtube/Ubuntu/get_ubuntu.py
@@ -0,0 +1,52 @@
+import json
+import os
+from googleapiclient.discovery import build
+import socket
+socket.setdefaulttimeout(500)
+
+
+def search_youtube(api_key, query, max_results=50):
+  youtube = build('youtube', 'v3', developerKey=api_key)
+
+  search_response = youtube.search().list(
+      q=query,
+      part="id,snippet",
+      maxResults=max_results,
+      type="video"
+  ).execute()
+
+  videos = []
+
+  for search_result in search_response.get("items", []):
+      if search_result["id"]["kind"] == "youtube#video":
+          video_id = search_result["id"]["videoId"]
+          video_metadata = get_video_metadata(api_key, video_id)
+          videos.append(video_metadata)
+
+  return videos
+
+
+def get_video_metadata(api_key, video_id):
+    youtube = build('youtube', 'v3', developerKey=api_key)
+
+    request = youtube.videos().list(
+        part="snippet,contentDetails,statistics",
+        id=video_id
+    )
+    response = request.execute()
+    return response
+
+
+api_key = 'AIzaSyDI_BBExs-HypVZFxgnR5tj5S6-uKyU4vk'  # Replace with your actual API key
+
+# Search for videos related to "VLC player"
+vlc_related_videos = search_youtube(api_key, "Ubuntu Desktop Tutorial", max_results=10)
+
+# create data folder if not exist
+if not os.path.exists("data"):
+    os.makedirs("data")
+
+for video in vlc_related_videos:
+    # store the video metadata into a json file
+    with open(f"data/{video['etag']}.json", "w") as f:
+        json.dump(video, f, indent=4)
\ No newline at end of file