When testing with my Google account, I found a couple of very minor differences in the bodies of the responses from these two endpoints:
The current endpoint returns a picture URL with a sz=50 query string, which links to a small photo. The OpenID endpoint returns the same URL but with no query string, which links to a large version of the same photo.
The current endpoint contains "kind" => "plus#personOpenIdConnect",, and the new OpenID endpoint omits this field.
See discussion in #9. This new endpoint for obtaining user information returns almost the same information as the old endpoint, but does not require turning on the Google+ API. This follows the OpenID Connect directions at https://developers.google.com/identity/protocols/OpenIDConnect#obtaininguserprofileinformation instead of the Google Sign-In directions.
When testing with my Google account, I found a couple of very minor differences in the bodies of the responses from these two endpoints:
picture
URL with asz=50
query string, which links to a small photo. The OpenID endpoint returns the same URL but with no query string, which links to a large version of the same photo."kind" => "plus#personOpenIdConnect",
, and the new OpenID endpoint omits this field.I also renamed
@token_url
to avoid confusion with https://github.com/ueberauth/ueberauth_google/blob/master/lib/ueberauth/strategy/google/oauth.ex#L17.@token_url
was actually not a URL for exchanging an authorization code for an access token, as the variable name implied. Instead, it was just an endpoint to retrieve user info once already authenticated with Google.