| Time |
S |
Nick |
Message |
| 01:00 |
|
|
pgicxplzs joined #rest |
| 02:37 |
|
|
tbsf joined #rest |
| 02:52 |
|
|
tbsf joined #rest |
| 05:07 |
|
|
tbsf joined #rest |
| 05:09 |
|
|
Coldblackice joined #rest |
| 06:08 |
|
|
tbsf joined #rest |
| 07:09 |
|
|
tbsf joined #rest |
| 07:16 |
|
|
searchbot joined #rest |
| 07:16 |
|
|
DrCode joined #rest |
| 07:16 |
|
|
blongden_ joined #rest |
| 07:16 |
|
|
StatelessCat joined #rest |
| 07:16 |
|
|
spaceone joined #rest |
| 07:29 |
|
|
nesselbosch joined #rest |
| 08:32 |
|
|
gigo1980 joined #rest |
| 09:09 |
|
|
interop_madness joined #rest |
| 09:45 |
|
|
tbsf joined #rest |
| 09:52 |
|
|
gigo1980 joined #rest |
| 10:05 |
|
|
timg___ joined #rest |
| 10:08 |
|
|
quimrstorres joined #rest |
| 10:31 |
|
|
moldy joined #rest |
| 10:31 |
|
moldy |
hi |
| 10:33 |
|
moldy |
i want to provide an interface for a client to verify that the attribute of a certain object that the client is about to POST is unique (e.g. that a username is not taken yet) |
| 10:33 |
|
moldy |
which method should i use here? i tend to say POST to avoid caching, but the request doesn't change the server state |
| 11:46 |
|
|
tbsf joined #rest |
| 12:12 |
|
|
timg___ joined #rest |
| 12:50 |
|
StatelessCat |
moldy: why not doing a GET and use the return code as the anwser ? |
| 12:50 |
|
moldy |
StatelessCat: you mean 404 to mean "free"? |
| 12:51 |
|
moldy |
StatelessCat: hmm. not sure. in my case, a number can be unavailable even if it doesn't publicly exist. |
| 12:52 |
|
StatelessCat |
404 means free, yes that's what i suggested |
| 12:52 |
|
moldy |
StatelessCat: i am using a "soft-delete" logic. if object with number 1 gets soft-deleted, i still treat that number as unavailable, but /objects/1 will 404. |
| 12:54 |
|
moldy |
so a dedicated validator endpoint seems to be more clean to me. here is my current server-side code (python/django): https://dpaste.de/9dYM |
| 12:54 |
|
StatelessCat |
moldy: in case the ressource exists but is not authorised for public access, you can return HTTP 401 http://httpstatusdogs.com/401-unauthorized |
| 12:55 |
|
moldy |
StatelessCat: hmm, i guess i could do that, but it would leak which instances are deleted. not sure. |
| 12:55 |
|
moldy |
so far, my approach was to hide the soft-deletion logic as much as possible |
| 12:55 |
|
moldy |
but maybe that approach was not very clever :) |
| 13:18 |
|
StatelessCat |
moldy: zith so nuch logic behind on the identifier, Maybe identifiers can be Ressources 1 |
| 13:19 |
|
StatelessCat |
and you will uses /identifiers/12344 as a Resource |
| 13:20 |
|
moldy |
StatelessCat: hmm, that may be an idea |
| 13:36 |
|
|
crystal77 joined #rest |
| 13:52 |
|
|
tbsf joined #rest |
| 14:35 |
|
|
eschmidbauer joined #rest |
| 14:50 |
|
|
tbsf joined #rest |
| 14:56 |
|
|
timg____ joined #rest |
| 14:59 |
|
|
timg_____ joined #rest |
| 15:58 |
|
|
tbsf joined #rest |
| 16:00 |
|
|
FritzLakritz joined #rest |
| 16:45 |
|
|
Macaveli joined #rest |
| 17:31 |
|
|
Macaveli joined #rest |
| 18:26 |
|
|
timg___ joined #rest |
| 18:31 |
|
|
timg___ joined #rest |
| 18:36 |
|
|
Devastator_ joined #rest |
| 18:40 |
|
|
Devastator_ joined #rest |
| 19:12 |
|
|
gigo1980 joined #rest |
| 21:42 |
|
|
Coldblackice joined #rest |
| 22:48 |
|
|
timg___ joined #rest |
| 22:53 |
|
|
timg___ joined #rest |
