Difference between revisions of "Linden Lab Official talk:Inventory API v3"
(Dissatisfied) |
(→Endpoint!!: new section) |
||
| (One intermediate revision by the same user not shown) | |||
| Line 3: | Line 3: | ||
[[User:Aleric Inglewood|Aleric Inglewood]] 12:41, 4 February 2014 (PST) | [[User:Aleric Inglewood|Aleric Inglewood]] 12:41, 4 February 2014 (PST) | ||
:I totally agree. One might wonder if this actually means that you can retrieve inventory data with off-world GET/POST requests, which would ''finally'' gives us a way to navigate through our directory in a ''sensible way''. | |||
:In other words: hello inventory-over-[https://en.wikipedia.org/wiki/Filesystem_in_Userspace FUSE]! | |||
:— [[User:Gwyneth Llewelyn|Gwyneth Llewelyn]] ([[User talk:Gwyneth Llewelyn|talk]]) 12:09, 15 May 2023 (PDT) | |||
== Endpoint!! == | |||
Okay, now that we have a fully documented API, described in minutiae, one can only wonder: where is the endpoint to call? How are the calls authenticated? | |||
The latter might be simpler to answer, i.e. just use LL's OAuth2/JWT provider. But allegedly that only works for internal domains and cannot be allowed outside anything which is ''not'' under the {{code|secondlife.com}} domain name... | |||
The endpoint is another story. It can be basically ''anything'', we just need to know ''what''. It would be reasonable to assume that it ''might'' be {{code|https://api.secondlife.com/...}} like some of the other, more recent APIs, but, nevertheless, you can't simply guess any missing path components from just that base URL... | |||
Or do we need to go via the capability service, and get an endpoint that way — as is done for the Registration API? | |||
Please elucidate us :) | |||
— [[User:Gwyneth Llewelyn|Gwyneth Llewelyn]] ([[User talk:Gwyneth Llewelyn|talk]]) 08:31, 26 February 2024 (PST) | |||
Latest revision as of 08:31, 26 February 2024
Lots of tables without explanation of what they mean, without examples, without basically any English text that explains. This page is not useful as-is.
Aleric Inglewood 12:41, 4 February 2014 (PST)
- I totally agree. One might wonder if this actually means that you can retrieve inventory data with off-world GET/POST requests, which would finally gives us a way to navigate through our directory in a sensible way.
- In other words: hello inventory-over-FUSE!
- — Gwyneth Llewelyn (talk) 12:09, 15 May 2023 (PDT)
Endpoint!!
Okay, now that we have a fully documented API, described in minutiae, one can only wonder: where is the endpoint to call? How are the calls authenticated?
The latter might be simpler to answer, i.e. just use LL's OAuth2/JWT provider. But allegedly that only works for internal domains and cannot be allowed outside anything which is not under the secondlife.com domain name...
The endpoint is another story. It can be basically anything, we just need to know what. It would be reasonable to assume that it might be https://api.secondlife.com/... like some of the other, more recent APIs, but, nevertheless, you can't simply guess any missing path components from just that base URL...
Or do we need to go via the capability service, and get an endpoint that way — as is done for the Registration API?
Please elucidate us :)
— Gwyneth Llewelyn (talk) 08:31, 26 February 2024 (PST)