Odd values for "email" when asking for users

Description
It looks like starting sometime before May 23rd at 5AM, the value for email changed – normally, we received a lowercased version of the users’ email address, which is what would be added on user create since we use SSO integration and the SSO endpoint delivers the email in all lowercase.

However we’re now seeing values in all sorts of odd casing (which is fixable) as well as values like:

user@domain.edu,User@DOMAIN.edu

Which is also fixable but very perplexing, since no changes to your API have been recorded as of late. I’m worried that for the multiple email situations, I’ll end up with two wildly different emails.

Error
No error, just unexpected values.

Which App Type (OAuth / Chatbot / JWT / Webhook)?
JWT

Which Endpoint/s?
/v2/users

How To Reproduce (If applicable)
Steps to reproduce the behavior:

  1. Request:
    GET https://api.zoom.us/v2/users?page_number=1&page_size=100&status=active
    Authorization: Bearer [REDACTED]
    User-Agent: libwww-perl/6.08
    Cookie: cred=[REDACTED]
    Cookie2: $Version="1"
  2. No error, but I receive incorrect values for email.

Also observed this beginning 23 May 2021. It is causing user reconciliation problems as the observed values are inconsistent with expected values.

Hi @mterry and @JRBrownNKU,

Welcome both to the community!

I can confirm that this is expected and was recently changed on 23rd May.
The release notes for this update can be found below.

https://marketplace.zoom.us/docs/changelog#publications/api-updates-may-23-2021

I hope this helps!

Many thanks,
Alex

@alexmayo Okay, that explains the first problem I had, but what about the second, where there are multiple email addresses in the email field? Are they just caseless and cased versions of the same value?

@mterry not too sure about that!

I’ll let @tommy or @MaxM comment on it - it does look like a cased and lowercase version though as you say.

1 Like

I understand from my support case engineering was considering rolling this change back. This appears to have already occurred as the symptoms are no longer reproducible.

Hey @JRBrownNKU and @mterry,

I’m sorry that you encountered this issue when using our API! This was related to an issue that we identified with our latest update. As you mentioned, our engineering team has since rolled back this change and the issue should now be resolved.

Please reach out if you see any further instances of this.

Thanks,
Max