Hi,
I have created an account and I tried to execute the sample commands in the GitHub homepage of the package:

> df <- cams_get_radition(username, lat=60, lon=15, 
                        date_begin="2016-01-01", date_end="2016-01-15")
Error in cams_get_radition(username, lat = 60, lon = 15, date_begin = "2016-01-01",  : 
  unused argument (lon = 15)

it seems that the argument is now lng instead of lon. I tried the right argument:

> df <- cams_get_radition(username, lat=60, lng =15, 
                        date_begin="2016-01-01", date_end="2016-01-15")
Error in match.arg(encode) : 
  'arg' should be one of “multipart”, “form”, “json”

the problem comes from httr::POST in cams_api, in line 95 you set as argument encode the value raw but my httr doc says:

POST(url = NULL, config = list(), ..., body = NULL,
  encode = c("multipart", "form", "json"), multipart = TRUE,
  handle = NULL)

the encode raw option is instead present in httr version 1.2.1 but not in the 1.1.0 I have on my Microsoft R 3.3.0. So, you should add a dependency to this particular version of httr.

UPDATE: I have installed the latest httr from Hadley's repository on GitHub and now I get:

> filename <- paste0(tempfile(), ".nc")
> 
> r <- cams_api(username, 60, 15, "2016-06-01", "2016-06-10", 
+               format = "application/x-netcdf", filename = filename)
No encoding supplied: defaulting to UTF-8.
xmlXPathEval: evaluation failed
xmlXPathEval: evaluation failed
xmlXPathEval: evaluation failed
Warning messages:
1: In node_find_all(x$node, x$doc, xpath = xpath, nsMap = ns) :
  Undefined namespace prefix [1219]
2: In node_find_all(x$node, x$doc, xpath = xpath, nsMap = ns) :
  Undefined namespace prefix [1219]
3: In node_find_all(x$node, x$doc, xpath = xpath, nsMap = ns) :
  Undefined namespace prefix [1219]

and r$ok is FALSE.

Hi,

I get the above error when trying to obtain solar data from within the camsRad package. Here is the code:

BanthamSolar <- cams_get_radiation(

lat=50.277

, lng=-3.879

, # Latitude & longitude coordinates

date_begin="2018-03-24", date_end="2018-11-11", # Date range

time_step="PT01H") # Use hourly time step

This code was working perfectly fine up until an update on the Soda-Pro website detailing a switch to a new version of CAMS radiation services. Here is a link to the release notes page on the Soda-pro website, please see note #1035 detailing possible errors relating to a new XML file format?

https://www.soda-pro.com/help/release-notes

It has links to an automatic-access page but this generates a manual data file. Any help on this would be appreciated!

👋 @lukas-rokka! we no longer transfer onboarded packages to ropenscilabs but instead directly to ropensci so we should move this repo. Are you ok with my doing it? If so when would you have the time to update all links? I wouldn't transfer before you have time to do that, and it's not urgent to transfer anyway.

If possible could you please add this badge to the README of rnaturalearth when updating CI links?
[![Project Status: Active – The project has reached a stable, usable state and is being actively developed.](http://www.repostatus.org/badges/latest/active.svg)](http://www.repostatus.org/#active)

Thanks for your patience! 🙏

👋 @lukas-rokka I'm using codemetar to create JSON-LD metadata for rOpenSci packages. I'd like some info about camsRad when you have the time, since you're the author / maintainer. In particular, I need your ORCID identifier as well as some keywords for the package that would help for search.

Thanks and let me know if you need me to elaborate on what I'm doing!

nc_open(r$response$content) in the example doesn't work on Mac. It works with ´nc_open(filename)´.

The CAMS Radiation Service relies on images from the Metosat satellite which is located at the at the prime meridian of 0° longitude (covering Europe, Africa and Middle East).

Are there any similar projects with open and free, high resolution (hourly, ~5 km) solar radiation data based on images from satellites above America and Asia?

I´m in first hand interested in services providing time series, as gridded data sources usually requires you to handle many TB of data.

Hi @lukas-rokka. In your rOpenSci post about camsRad, the link to https://rokka.shinyapps.io/smartep no longer works. Do you have a replacement?

If not, we could edit the post