Not only is the Hive API reference cryptic and complicated to read, but it is incomplete. To get something done, I had to reverse engineer the api and beem.
https://developers.hive.io/apidefinitions/#bridge.get_ranked_posts
It clearly says there:
Query Parameters
{"sort": "", "tag": "", "observer": ""}
Those work, but after HOURS of tinkering and trial&error, I now know, that it also takes:
{"limit": 21, "start_author": "", "start_permlink": ""}
Where api.hive.blog accepts a limit between 1:100. If your expected result is any longer than 100, you need to call again, and reference:
start_author the author of the last post in the results (without @),
start_permlink the permlink of the last post of the results.
If that is some developer humor, I don't get it.
In my case, the actual result is a list of 193 items.
This is either autism or gatekeeping.
And we all know its the latter. It makes me sick. I fucking raged in voicechat over this.
You want Hive to grow?
You want to attract devs?
Fix that shit. ASAP.
The last month I extensively played around with OpenAI api.
Their API reference is a bit complicated, too, but I could find everything I needed.
If I sent a bad call, the API itself would reply with what it actually expects as params as error message not random code xyz, like hive api.
inb4: - anything @gtg has to say about this. No need for your elitist useless answer, anyways. - just use beem, lol
CUNTS!
For anyone wondering: building something so I can evaluate all entries to the maize challenge. Will publish a tool, which everybody can use. For normal humans.