| Time |
S |
Nick |
Message |
| 01:15 |
|
|
fuzzyhorns joined #rest |
| 02:25 |
|
|
baweaver joined #rest |
| 02:34 |
|
|
fuzzyhorns joined #rest |
| 02:50 |
|
|
fuzzyhorns joined #rest |
| 02:52 |
|
|
fuzzyhorns joined #rest |
| 03:13 |
|
|
wsiqueir joined #rest |
| 05:37 |
|
|
_ollie joined #rest |
| 06:37 |
|
|
vanHoese_ joined #rest |
| 08:15 |
|
|
graste joined #rest |
| 08:31 |
|
|
chthon joined #rest |
| 08:53 |
|
|
darkangel joined #rest |
| 08:55 |
|
darkangel |
Is there any neat/efficient way to delete a selection of resources? I don't really like the 2-step option. |
| 08:56 |
|
trygvis |
not really |
| 08:57 |
|
trygvis |
you can make a serivce that can do deletes for you, but it becomes very specific |
| 08:57 |
|
darkangel |
Is "DELETE /articles?ids=1,2,3" very very bad? |
| 08:58 |
|
trygvis |
not if the hypermedia describes how to do create the request |
| 08:58 |
|
darkangel |
We're not really using hypermedia yet. |
| 08:58 |
|
trygvis |
but that is a kind of service that I mentioned. I would probably POST a list of URIs for the service to delete |
| 08:58 |
|
trygvis |
well, if you're all about RPC you're in the wrong channel :) |
| 08:59 |
|
darkangel |
Trying to keep it ReST-like at least. |
| 08:59 |
|
trygvis |
at least use the URLs then instead of IDs |
| 09:01 |
|
darkangel |
Seems redundant. |
| 09:05 |
|
darkangel |
trygvis: What would you name that resource (where you post the list of URIs)? |
| 09:08 |
|
trygvis |
"asdlkasjdaklsdjfioehfgvdfbgj" is a good example |
| 09:08 |
|
darkangel |
lol |
| 09:09 |
|
trygvis |
it's rest, the URLs are not important |
| 09:10 |
|
trygvis |
(their text representation isn't) |
| 09:10 |
|
trygvis |
using and discovering them is important |
| 09:11 |
|
darkangel |
Well I haven't seen any public APIs with names like that. :) |
| 09:12 |
|
graste |
how about DELETE /articles with payload or article urls and then that returns LOCATION "articles/434754df-67d3-342c-8a53-8555f01c51a6" which you can GET the status on the deletion of the POSTed article urls :-) |
| 09:12 |
|
darkangel |
DELETE is not really supposed to have a payload. |
| 09:13 |
|
graste |
not really as in must not or not really as in unusual |
| 09:13 |
|
darkangel |
As in some proxies and other tools strip the body out. |
| 09:14 |
|
graste |
there is always the option of RPC style POST abusal ^^ |
| 09:15 |
|
graste |
POST the urls of resources to be deleted to some url |
| 09:15 |
|
darkangel |
Ya like verbs in the URI ... I'm sure it's gonna happen at some point. |
| 09:15 |
|
graste |
-.- |
| 09:15 |
|
graste |
you're late to the party as in rails? you can add _method=delete ^^ |
| 09:16 |
|
trygvis |
sending DELETE with a payload is much worse than using POST to delete data |
| 09:17 |
|
graste |
so we've got a winner via POST there? |
| 09:18 |
|
darkangel |
POST /articles/for-deletion [1, 2] |
| 09:18 |
|
darkangel |
lol |
| 09:18 |
|
darkangel |
I hate this sh*t. |
| 09:19 |
|
graste |
depending on usecase you might foreach $article DELETE $url in your client code instead of trying to create this bulk deletion url |
| 09:19 |
|
graste |
as in that case you will need to think about partial success etc as well |
| 09:19 |
|
|
interop_madness joined #rest |
| 09:21 |
|
darkangel |
And transactions, etc. But anyway, that might be okay for a few articles, but if you're deleting 1,000 then it's far from ideal (esp. without HTTP/2). |
| 09:27 |
|
graste |
yes |
| 09:32 |
|
|
mezod joined #rest |
| 10:24 |
|
|
vanHoesel joined #rest |
| 11:12 |
|
|
vanHoesel joined #rest |
| 12:39 |
|
|
mezod joined #rest |
| 12:54 |
|
|
jangid joined #rest |
| 13:40 |
|
|
mezod joined #rest |
| 13:47 |
|
|
wavded joined #rest |
| 14:18 |
|
|
`0660 joined #rest |
| 15:01 |
|
|
timg___ joined #rest |
| 15:36 |
|
|
jangid joined #rest |
| 15:37 |
|
|
wsiqueir joined #rest |
| 15:39 |
|
|
vanHoesel joined #rest |
| 15:43 |
|
|
jangid joined #rest |
| 15:53 |
|
|
vanHoesel joined #rest |
| 15:55 |
|
|
ralphschindler joined #rest |
| 16:31 |
|
|
jangid joined #rest |
| 16:33 |
|
|
nkoza joined #rest |
| 16:47 |
|
|
ralphschindler joined #rest |
| 16:50 |
|
|
jangid joined #rest |
| 16:51 |
|
jangid |
what is peak time for this channel? No one online since very long. |
| 16:53 |
|
|
mezod joined #rest |
| 16:54 |
|
sfisque |
patience. just post a question and someone at some point may see it and respond to you |
| 17:08 |
|
jangid |
:-) Thanks |
| 17:24 |
|
|
timg___ joined #rest |
| 18:00 |
|
|
jangid joined #rest |
| 18:05 |
|
|
vanHoesel joined #rest |
| 19:08 |
|
|
vanHoesel joined #rest |
| 19:18 |
|
|
jangid joined #rest |
| 19:24 |
|
|
jangid joined #rest |
| 19:58 |
|
|
foist joined #rest |
| 20:18 |
|
pdurbin |
I don't know that anyone keeps metrics on the peak time for this channel. |
| 20:23 |
|
fumanchu_ |
3am Pacific: 0. 4am: 0. 5am: 0. 6am: 0. 7am: 0. 8am: 0. 9am: 2 messages. 10am: 1. 11am: 0. 12noon: 0. 1pm: 4 |
| 20:23 |
|
fumanchu_ |
I hereby conclude: 1pm Pacific |
| 20:24 |
|
pdurbin |
:) |
| 20:47 |
|
|
Coldblackice joined #rest |
| 20:52 |
|
|
mezod joined #rest |
| 20:54 |
|
|
vanHoesel joined #rest |
| 21:22 |
|
|
ralphschindler joined #rest |
| 21:56 |
|
|
vanHoesel joined #rest |
| 21:56 |
|
|
ralphschindler joined #rest |
| 22:04 |
|
|
talios joined #rest |
| 23:43 |
|
|
vanHoesel joined #rest |
