Hi @Shida, when trying to download the contents of all documents of a Quick Dynalist user, the document rate limit becomes quickly a problem. I have a user with 500 documents, but the rate limit is at 50.
Would it be possible to increase the rate limit or create a new request type to query multiple documents at once?
So what we’re thinking of right now is that 500 documents sound like quite an abnormal use case.
We could look into increasing the limit to something more reasonable like 200, but I still feel like 500 documents is already at the point where it has degraded performance for various things like tag counter, global search, etc.
Does this strengthen the case for an Archive area for documents which, may have a degraded performance, but it’s intentional? The 500 documents use case could be abnormal but my guess is they are accessing only about 5% of them on a regular basis.
Personally, I would be in favour of this. Day to day performance outweighs occasional inconvenience for me. It would be good to hear the views of other users though.
I think we are discussing two different issues here though.
API
web browser frontend
Only because the web browser does not support such a large amount of documents, that that may not hold true for an application that accesses the API. E.g. What if I wanted to query the archived documents that you are suggesting?
I love this. I have a lot of old documents that I occasionally use for reference, but would prefer not to have them loaded and searchable most of the time.
@Shida querying all modified data since a given timestamp could also heavily reduce the load on your servers and the waiting time for quick dynalist users.
A version number would work equally well. So you would query all changes since a specific version number and would get the current version number plus changes back?
Apologies for the late reply! The rate increase is already in place, but hasn’t been deployed to the server yet, which will happen sometime this weekend.
As for the API changes, it’s still backlogged by a rare sync issue we’re currently investigating (disappearing documents, documents that can’t be deleted, etc), but I’ve been going around to clean things off the backlog from time to time so tickets don’t go stale for too long. Sorry about that!