@kensanata hello 🙂

I want to understand something regarding mastodon-archive, that I can't find (or understand) in the documentation: when I use the --pace option, what archiving/deletion rate will be used ?
It fixed or dynamic ?

I suppose that without that option it does its request as much as it can, and stop when the limit is reached (blocked by the server ?).
But for instance for the deletion, the limit is 300requests/5min. What is the pace rate ?

Thanks 🙂

@Lapineige I think it works like you say it does. But actually, I have no idea. Check out the Mastodon.py docs: mastodonpy.readthedocs.io/en/s
If you use --pace for login, the Mastodon object will use ratelimit_method="pace". My hope would be that the library automatically determines the right pace based on rate limit. If not, I'd consider that a Mastodon.py bug and we should report it. github.com/halcy/Mastodon.py
@halcy

@kensanata @Lapineige yeah, that should basically be what it does.

Essentially, ratelimit modes: "wait" (default): Hit the server as fast as possible and when you get ratelimited, wait until you can go again

"pace": try to spread out the requests so that you stay barely below the rate limit. Possibly slightly slower overall, but more continuous progress and nicer to the server (just noticed that the docs are slightly wrong for what it does when it DOES get ratelimited: it actually waits in that case and does NOT throw an error)

"throw": go as hard as possible and when you get ratelimited throw an error.

@kensanata @Lapineige one potential problem with this is that this code is from when there was only ONE ratelimit, now there's different ones for different things I think? so mmmmaybe the code isn't 100% up to spec anymore. It should still mostly just work, though.

@halcy @kensanata in my case, without using --pace, it is limited quite soon indeed. My mastodon account becomes unusable.
With this option, it's *a lot* slower than the rate limit of 300/5min. It's more 300 for 1h… but my mastodon account is usable without any "throttled" warning

@Lapineige @kensanata that is... way too slow. essentially what SHOULD happen is (roughly)

1) it looks at how many requests are left per 5 minute slot

2) it then does one request every (num requests left / 5 min) / "pace factor", default 1.1

3) every time a request is made the numbers are updated

that SHOULD result in something like 270 requests per 5 minutes, less if you're concurrently doing other things, and is designed to allow you to do exactly what you said amongst other things, actually use the account while stuff like this runs.

I don't have time to look into it right now and probably not any time today, but please do file an issue on github with as many details as you have and I'll look at it when I have time

Sign in to participate in the conversation
La Quadrature du Net - Mastodon - Media Fédéré

Bienvenue dans le media fédéré de la Quadrature du Net association de défense des libertés. Les inscriptions sont ouvertes et libres.
Tout compte créé ici pourra a priori discuter avec l'ensemble des autres instances de Mastodon de la fédération, et sera visible sur les autres instances.
Nous maintiendrons cette instance sur le long terme.