def youtube_channel_search_gdocs(client, query, cfg):
search_query = f"site:youtube.com/channel \\\"{query}\\\""
search_results = gdoc_search(search_query, cfg)
channels = []
for result in search_results:
sanitized = "https://youtube.com/" + ('/'.join(result["link"].split('/')[3:5]))
if sanitized not in channels:
channels.append(sanitized)
if not channels:
return False
results = {"channels": [], "length": len(channels)}
channels = channels[:5]
for profil_url in channels:
req = client.get(profil_url)
source = req.text
data = json.loads(
source.split('window["ytInitialData"] = ')[1].split('window["ytInitialPlayerResponse"]')[0].split(';\n')[0])
avatar_link = data["metadata"]["channelMetadataRenderer"]["avatar"]["thumbnails"][0]["url"].split('=')[0]
req = client.get(avatar_link)
img = Image.open(BytesIO(req.content))
hash = image_hash(img)
title = data["metadata"]["channelMetadataRenderer"]["title"]
results["channels"].append({"profil_url": profil_url, "name": title, "hash": hash})
return results
def youtube_channel_search_gdocs(client, query, data_path, gdocs_public_doc):
search_query = f"site:youtube.com/channel \\\"{query}\\\""
search_results = gdoc_search(search_query, data_path, gdocs_public_doc)
channels = []
for result in search_results:
sanitized = "https://youtube.com/" + ('/'.join(
result["link"].split('/')[3:5]))
if sanitized not in channels:
channels.append(sanitized)
if not channels:
return False
results = {"channels": [], "length": len(channels)}
channels = channels[:5]
for profile_url in channels:
data = None
avatar_link = None
retries = 2
for retry in list(range(retries))[::-1]:
req = client.get(profile_url)
source = req.text
try:
data = json.loads(
source.split('window["ytInitialData"] = ')[1].split(
'window["ytInitialPlayerResponse"]')[0].split(';\n')
[0])
avatar_link = data["metadata"]["channelMetadataRenderer"][
"avatar"]["thumbnails"][0]["url"].split('=')[0]
except (KeyError, IndexError) as e:
if retry == 0:
return False
continue
else:
break
req = client.get(avatar_link)
img = Image.open(BytesIO(req.content))
hash = image_hash(img)
title = data["metadata"]["channelMetadataRenderer"]["title"]
results["channels"].append({
"profile_url": profile_url,
"name": title,
"hash": hash
})
return results
