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 was archived by the owner on Apr 19, 2026. It is now read-only.
I use Google Sign-In in my implementation of Polymer shop which I'm requesting a pull. Though I was suggested to use this google-signin component, I'm reluctant to do so for some reasons:
google-signin contains both view and logic (google-signin-aware). While I agree it's intuitive to use a single component to show a Google Sign-In button, there are use cases where only logic is needed.
google-signin-aware can handle multiple of itself. It's natural consequence if you assume to put multiple Google Sign-In buttons in a single page, but why is it needed? I only see it as an effort to resolve contradiction when putting multiple google-signin buttons.
There is an event for notifying successful authorization, but no events for authentication exists.
Though auth status changes can be propagated through events, there is no way to associate a user action to a successful sign-in with resulting id_token being passed to the invoker.
Here's my proposal:
Make AuthEngine as google-signin-aware itself and add iron-meta as a behavior so google-signin components can refer to a single google-signin-aware in a page. (This means implementer needs to put both google-signin-aware and google-signin in the same page.)
google-signin-aware-success event means "authorization" succeeded. Make it google-signin-aware-authz-success and add google-signin-aware-authn-success to indicate "authentication" success.
Return a promise on signIn() invocation. That way resolving function can receive result object and pass id_token or auth code to the server.
Hi,
I use Google Sign-In in my implementation of Polymer shop which I'm requesting a pull. Though I was suggested to use this
google-signincomponent, I'm reluctant to do so for some reasons:google-signincontains both view and logic (google-signin-aware). While I agree it's intuitive to use a single component to show a Google Sign-In button, there are use cases where only logic is needed.google-signin-awarecan handle multiple of itself. It's natural consequence if you assume to put multiple Google Sign-In buttons in a single page, but why is it needed? I only see it as an effort to resolve contradiction when putting multiplegoogle-signinbuttons.Here's my proposal:
google-signin-awareitself and addiron-metaas a behavior sogoogle-signincomponents can refer to a singlegoogle-signin-awarein a page. (This means implementer needs to put bothgoogle-signin-awareandgoogle-signinin the same page.)google-signin-aware-successevent means "authorization" succeeded. Make itgoogle-signin-aware-authz-successand addgoogle-signin-aware-authn-successto indicate "authentication" success.signIn()invocation. That way resolving function can receive result object and pass id_token or auth code to the server.Let me know what you think.