Ok, DEFRA source (via scraping) should now be fixed. Did a quick look at the other two methods mentioned above as it'd be nice to use some other than scraping. A few quick observations.
There are coordinates in here, which is good (well they're a bounding box, but we can make a center point from that)
We'd have to do this for each of the ~160 locations, that seems way too heavy to do on each fetch
Each of the final xmls provides all data since the beginning of the year, again way to heavy to do every 10 minutes.
Unless I'm missing something, doesn't look like this would be a good alternative.
SOS API
This looks really promising, it should allows us to get the data via a RESTful API and consume JSON 💃 . The base url looks like https://uk-air.defra.gov.uk/sos-ukair/api/v1/. Did a quick check of stations and it reported back a whole bunch of things, which is great. There are coordinates. I'm unsure of how best to achieve our use case though. We don't want to have to query the system 500 times to get 3 pollutants from ~160 stations. Would be good to try and reach out to someone over there and see how to best handle that.
majesticio
commented
1 year ago
I'm seeking suggestions on a contact to confirm whether DEFRA's SOS API has a way to give latest measurements smoothly.
This convo started in a previous (now closed) issue when our scraper broke: https://github.com/openaq/openaq-fetch/issues/292
Below is from @jflasher, quoted from that thread:
Ok, DEFRA source (via scraping) should now be fixed. Did a quick look at the other two methods mentioned above as it'd be nice to use some other than scraping. A few quick observations.
Atom Feed
Unless I'm missing something, doesn't look like this would be a good alternative.
SOS API
This looks really promising, it should allows us to get the data via a RESTful API and consume JSON 💃 . The base url looks like
https://uk-air.defra.gov.uk/sos-ukair/api/v1/. Did a quick check of stations and it reported back a whole bunch of things, which is great. There are coordinates. I'm unsure of how best to achieve our use case though. We don't want to have to query the system 500 times to get 3 pollutants from ~160 stations. Would be good to try and reach out to someone over there and see how to best handle that.