Hey guys! 🙂 Is using the WatchServiceClient to store the ZedTokens in memory every time there is an mutation a good idea? Then use the stored ZedToken in the next check permission request

👋 hi, perhaps the first question would be what are you trying to achieve? You can technically do that, but it's not going to be great performance wise, you'd get a bad cache-hit rate because each time you do a check, there is nothing in the cache, because you always ask for the latest state of the system. It would be pretty much like doing

fully-consistency

all the time.

What I would like to achieve is using

at_least_as_fresh

with a zedtoken. However it didn't occur to me that the

changes_through

in the watch response has zedtoken but this token is not the product of performing a check request (where the response is cached), so that's why using the ZedToken from the watch response is probably going to like

fully-consistency

as you said?

Yeah, what you get via the

Watch

response is the product of some mutation in the database side, mainly changes to the relationships stored (there is also support for changes to the schema but its not yet exposed via the API)

Thanks a lot for your response. That was very useful 🙂