You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jan 31, 2021. It is now read-only.
If an error is encountered at some point, save the partial data to the file with "nextPageToken" being set.
If the user runs the same command again, check if the "nextPageToken" is not empty, and if so resume the operation using it.
And some notes about this approach:
"nextPageToken": null means everything is successfully downloaded.
On step 2 above, when there is an error, the script can actually make a second try before failing and saving partial data.
To decrease the chance of error, there may be delay between requests (in fact I looked at the source code to see if this is already done but failed; because JS code, to me, is very hard to follow)
The text was updated successfully, but these errors were encountered:
@maliayas can you download youtube data using this code?. can you help me to get the data from youtube?. because I gave an error Error API response does not contain a "content_html" field #47. I want to get some data for my final year project. previously I have got data using this program. please help me.
This is just a proposal; there is no code yet. Unfortunately I'm not a JS dev, but I guess @philbot9 or someone else can easily implement this and then yes, it will fix #47.
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Youtube API sometimes may return error pages (such as #47) or simply a network error may break the whole fetching process.
To overcome this, I'll kindly propose the below approach:
Turn the scheme into something like this:
If an error is encountered at some point, save the partial data to the file with "nextPageToken" being set.
If the user runs the same command again, check if the "nextPageToken" is not empty, and if so resume the operation using it.
And some notes about this approach:
"nextPageToken": null
means everything is successfully downloaded.The text was updated successfully, but these errors were encountered: