How to obtain the root call log id?

[gianni.zoom (Gianni)]

@ddesimone

I can understand why the “new” /phone/call_history endpoint is Rate Limit Labeled HEAVY, due to the nature of potentially searching for a lot of data.

I don’t understand though why /phone/call_history/{callLogId} and /phone/call_history_detail/{callLogId} – note the use of the {callLogId} to fetch a single record – are labeled HEAVY and MEDIUM.

The previous (deprecated) /phone/call_logs/{callLogId} endpoint was LIGHT.

So now we have to build in more catch-retry or thread pause logic because we are experiencing the “Too Many Requests” error.

Lastly, these call log endpoint changes by Zoom have doubled the amount of API calls we make to get the data we need.

As an example, we have a process that fetches all prior day call history – which during a weekday could be 25k-30k+ calls.

Using /phone/call_history in batches of 300 is 80-100+ API calls. Not a big deal, but the JSON response schema for this endpoint gives a lot of detail but NOT ALL of the detail for the call.

So, now for each of the 30K+ calls we need to make 2 additional calls (instead of 1 before):

• /phone/call_history_detail/{callLogId} – this endpoint returns a few more pieces of information that the first didn’t – such as “hold_time” and “wait_time” (mislabeled in the documentation as “waiting_time”)
• /phone/call_history/{callLogId} – this endpoint returns us the call path data (ie: segments) that neither of the other endpoints contain. It used to be included in the deprecated “get call log details”.

Adding this up – 60,000+ API calls – happening every weekday.

Seems like incoming API traffic could be minimized by combining the 2 endpoints above. Move the call path data back into the call_history_detail endpoint and move the endpoint back to the LIGHT rate limit.

Good point @ddesimone – I’ve just shared with the PM and will discuss. Thank you and I’ll come back to the thread when I have more updates.