I also added the requests.Session structure you suggested earlier. It seems quite convenient, but I think I ran into some rate limiting issues when I tried to download a lot of previous brawl data with it. Did you experience anything like that with it?
Quick search tells me that I can add an event hook to manually introduce a lag, but I'd rather not if there is any more standard solution available.
As far is a known the retry with backoff strategy is an solid one because the rate limit is determined by spl, which is necessary for them so the api traffic is not overloaded.
You can look if you can make the api call bigger so it includes more data at once (less calls). But i expect you need to do something for every brawl cycle.
These are the reason (for me to retrieve many transactions and balances) why there is sometimes a large wait to retrieve data from spl.
Not for your case but if you store the already retrieved data you do not have to retrieve it again in a later phase. Then you have to think on how to store it, this is where the application get more complicated imho.
For this I just store the raw returned data so its not that complicated, and rate limiting is just a small issue. I'm not even 100% that's what happened, but I found that I was not able to log into Splinterlands while my script was running 😅
you can look at event bases system that gives back the information asynchronous. that is more used in react framework, I'm not familiar with that in python
That's one I've interacted a lot with :)
I also added the requests.Session structure you suggested earlier. It seems quite convenient, but I think I ran into some rate limiting issues when I tried to download a lot of previous brawl data with it. Did you experience anything like that with it?
Quick search tells me that I can add an event hook to manually introduce a lag, but I'd rather not if there is any more standard solution available.
As far is a known the retry with backoff strategy is an solid one because the rate limit is determined by spl, which is necessary for them so the api traffic is not overloaded.
You can look if you can make the api call bigger so it includes more data at once (less calls). But i expect you need to do something for every brawl cycle.
These are the reason (for me to retrieve many transactions and balances) why there is sometimes a large wait to retrieve data from spl.
Not for your case but if you store the already retrieved data you do not have to retrieve it again in a later phase. Then you have to think on how to store it, this is where the application get more complicated imho.
For this I just store the raw returned data so its not that complicated, and rate limiting is just a small issue. I'm not even 100% that's what happened, but I found that I was not able to log into Splinterlands while my script was running 😅
you can look at event bases system that gives back the information asynchronous. that is more used in react framework, I'm not familiar with that in python