is_chained + black

[xiangyan99]

No description provided.

[azure-sdk]

API change check

APIView has identified API level changes in this PR and created following API reviews.

azure-identity

[xiangyan99]

Azure/azure-sdk-for-java#35771

[pvaneck]

Looks like the keyword argument is_chained is now a part of the public API for these dev-time credentials. Was this is supposed to be an internal only param?

[xiangyan99]

Looks like the keyword argument is_chained is now a part of the public API for these dev-time credentials. Was this is supposed to be an internal only param?

Because those credentials did not accept kwargs, to accept extra parameter, there is no way to hide it from public API changes.

But they are not breaking changes.

[pvaneck]

Is within_credential_chain something we could use instead for this purpose?

Seems like we use it in quite a few places for determining the log-level depending on if the credential is within a chain or not (example).

[xiangyan99]

Is within_credential_chain something we could use instead for this purpose?

Seems like we use it in quite a few places for determining the log-level depending on if the credential is within a chain or not (example).

When I synced with Bill, we only wanted to swallow the error in DAC. But not in all chained credentials.

[pvaneck]

Just a comment about deduping error messages.

I'm generally okay with the added is_chained keyword argument. If we do get pushback for introducing this param to the API, then I think there are ways around it like setting credential._is_chained = True directly after instantiating it in DAC.