| Time |
S |
Nick |
Message |
| 01:31 |
|
|
talios joined #rest |
| 01:53 |
|
|
shrink0r_ joined #rest |
| 03:20 |
|
|
SlippinJimmy joined #rest |
| 06:55 |
|
|
lemur joined #rest |
| 08:53 |
|
|
TeddyMurray joined #rest |
| 09:02 |
|
|
TeddyMurray joined #rest |
| 09:37 |
|
|
rosstuck joined #rest |
| 10:19 |
|
|
Left_Turn joined #rest |
| 10:28 |
|
|
jcromartie joined #rest |
| 10:49 |
|
|
adaro joined #rest |
| 10:58 |
|
|
vanHoesel joined #rest |
| 11:45 |
|
|
shrink0r joined #rest |
| 12:47 |
|
|
mezod joined #rest |
| 14:33 |
|
|
TeddyMurray joined #rest |
| 14:34 |
|
|
TeddyMurray joined #rest |
| 14:43 |
|
|
Doc-Saintly joined #rest |
| 14:45 |
|
|
SlippinJimmy joined #rest |
| 14:46 |
|
|
jcromartie joined #rest |
| 15:32 |
|
|
shrink0r joined #rest |
| 16:18 |
|
|
pezra joined #rest |
| 16:31 |
|
dkm |
hello all. So I've designed an API to access real-time environmental data that is collected from right now one site, but eventually more than one site |
| 16:31 |
|
dkm |
now that I have the system more-or-less working I have some organizational and philosophical questions regarding best practices and endpoint design |
| 16:33 |
|
dkm |
1. data are collected for various metrics (e.g. water temperature, water pH, rain intensity, etc.), generally I have designed the endpoints to access data for one particular metric at a time |
| 16:34 |
|
dkm |
e.g. /sites/site1/metrics/rainfall/timeseries?since=2015-03-20 would return a json+hal response containing timeseries rainfall data for site 'site1' since the date specified |
| 16:36 |
|
dkm |
This seems to make sense and work fine for this case, but many times the end user will be interested in comparing data across multiple metrics (e.g. rainfall and water flow), for the case of just two metrics it probably is fine to just do two HTTP GETs to two different endpoints |
| 16:37 |
|
dkm |
but one feature my end-users need is a CSV download option to get a data set of say all the water quality parememters (7 or so) in one csv file. |
| 16:38 |
|
dkm |
actually, kind of 2 sub questions related to this: 1a what's the "proper" way to handle the return of CSV data when everything else is json+hal and 1b how to design endpoints to retreive this CSV data for a given list of metrics |
| 16:39 |
|
dkm |
right now I've done something like /sites/site1/archive.csv?metrics=2,3,4,5&since='2015-03-20', does this seem reasonable? are there cleaner ways to go about this? |
| 16:40 |
|
|
shrink0r joined #rest |
| 16:44 |
|
|
SlippinJimmy joined #rest |
| 17:54 |
|
|
lemur joined #rest |
| 18:34 |
|
|
Jefffrey joined #rest |
| 19:03 |
|
|
shrink0r joined #rest |
| 19:21 |
|
|
cigarshark joined #rest |
| 20:26 |
|
|
shrink0r joined #rest |
| 20:49 |
|
|
Jefffrey joined #rest |
| 21:12 |
|
|
TeddyMurray joined #rest |
| 21:37 |
|
|
SlippinJimmy joined #rest |
| 22:25 |
|
|
lemur joined #rest |
| 23:33 |
|
|
talios joined #rest |
| 23:58 |
|
|
Jefffrey joined #rest |
