How do I speed up my 130 second youtube-api python example? [duplicate]

Question

This python snippet:

from datetime import datetime
from googleapiclient.discovery import build
import googleapiclient
import socket

with open("secret_youtube_api_key", "r") as api_key_file:
  api_key = api_key_file.read()

socket.setdefaulttimeout(30000)

youtube : googleapiclient.discovery.Resource = build('youtube', 'v3', developerKey=api_key)
print("youtube type " + str(type(youtube)))

request : googleapiclient.http.HttpRequest = youtube.channels().list(
  part='statistics',
  forUsername='PewDiePie'
)
print("request type " + str(type(request)))

print("\n" + str(datetime.now()) + " request.execute() start")
response = request.execute()
print("response type " + str(type(response)))
print(str(datetime.now()) + "\n request.execute() end")

print(response)

gives the following output:

youtube type <class 'googleapiclient.discovery.Resource'>
request type <class 'googleapiclient.http.HttpRequest'>

2022-04-20 03:22:35.541216 request.execute() start
response type <class 'dict'>
2022-04-20 03:24:46.061762
 request.execute() end
{'kind': 'youtube#channelListResponse', 'etag': '01QNwqKcy80m97clcBBjfozMTV0', 'pageInfo': {'totalResults': 1, 'resultsPerPage': 5}, 'items': [{'kind': 'youtube#channel', 'etag': 'O5elmeewmtTIKBg8LuPaid7kDFs', 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw', 'statistics': {'viewCount': '28262284610', 'subscriberCount': '111000000', 'hiddenSubscriberCount': False, 'videoCount': '4472'}}]}

The timestamp for the actual call execution are quiite regularly 130 seconds apart. My control command using wget gives a hint

wget --timeout=300000 "https://www.googleapis.com/youtube/v3/channels?key=$APP_KEY&forUsername=PewDiePie&part=statistics" 
--2022-04-20 03:24:44--  https://www.googleapis.com/youtube/v3/channels?key=REDACTED&forUsername=PewDiePie&part=statistics
Resolving www.googleapis.com (www.googleapis.com)... 2a00:1450:4001:810::200a, 2a00:1450:4001:811::200a, 2a00:1450:4001:80e::200a, ...
Connecting to www.googleapis.com (www.googleapis.com)|2a00:1450:4001:810::200a|:443... failed: Connection timed out.
Connecting to www.googleapis.com (www.googleapis.com)|2a00:1450:4001:811::200a|:443... failed: Connection timed out.
Connecting to www.googleapis.com (www.googleapis.com)|2a00:1450:4001:80e::200a|:443... failed: Connection timed out.
Connecting to www.googleapis.com (www.googleapis.com)|2a00:1450:4001:80f::200a|:443... failed: Connection timed out.
Connecting to www.googleapis.com (www.googleapis.com)|142.250.185.170|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: unspecified [application/json]
Saving to: ‘channels?key=REDACTED&forUsername=PewDiePie&part=statistics’

channels?key=REDACTED     [ <=>                                                                       ]     476  --.-KB/s    in 0s      

2022-04-20 03:33:28 (18,7 MB/s) - ‘channels?key=REDACTED&forUsername=PewDiePie&part=statistics’ saved [476]

This seems to be a common problem without a clear solution. There are related questions (link) that fall short in their answer.

It looks like IPv6 just generally fails. What could be wrong on my end. How would I fix IPV6 issues on my end? Given that my end may be unfixable - how would I need to modify my python code to exclude IPV6?

forcing ipv4 on wget returns instantly: wget -4 --timeout=300000 "https://www.googleapis.com/youtube/v3/channels?key=$APP_KEY&forUsername=PewDiePie&part=statistics"

Answer 1

This answer to a differently worded question fits here:

It answers how to force ipv4 - not how to fix ipv6.

from datetime import datetime
from googleapiclient.discovery import build
import googleapiclient
import socket

with open("secret_youtube_api_key", "r") as api_key_file:
  api_key = api_key_file.read()


#https://stackoverflow.com/a/50044152/536874
import socket
getaddrinfo_original = socket.getaddrinfo
def getaddrinfo_decorated(*args, **kwargs):
  # filter responses that are not ipv4
  return [response for response in getaddrinfo_original(*args, **kwargs) if response[0] == socket.AF_INET]
socket.getaddrinfo = getaddrinfo_decorated

youtube : googleapiclient.discovery.Resource = build('youtube', 'v3', developerKey=api_key)
print("youtube type " + str(type(youtube)))

request : googleapiclient.http.HttpRequest = youtube.channels().list(
  part='statistics',
  forUsername='PewDiePie'
)
print("request type " + str(type(request)))

print("\n" + str(datetime.now()) + " request.execute() start")
response = request.execute()
print("response type " + str(type(response)))
print(str(datetime.now()) + "\n request.execute() end")

print(response)

The updated code now outputs:

youtube type <class 'googleapiclient.discovery.Resource'>
request type <class 'googleapiclient.http.HttpRequest'>

2022-04-20 14:06:27.311410 request.execute() start
response type <class 'dict'>
2022-04-20 14:06:27.503453
 request.execute() end
{'kind': 'youtube#channelListResponse', 'etag': '_EY8e-LjjUX00BenLPIGIts8iS0', 'pageInfo': {'totalResults': 1, 'resultsPerPage': 5}, 'items': [{'kind': 'youtube#channel', 'etag': 'QmXjB1Hz-b5QX1ahMELzX1BTl6I', 'id': 'UC-lHJZR3Gqxm24_Vd_AJ5Yw', 'statistics': {'viewCount': '28263689871', 'subscriberCount': '111000000', 'hiddenSubscriberCount': False, 'videoCount': '4472'}}]}

Which is a 200ms call like it should be. A better answer would include how to find and fix whatever is wrong with ipv6 on my network.