Once the dataset is public, we'll want a function that can pull the data directly from the Austraits page on Zenodo: https://doi.org/10.5281/zenodo.3568417
On Zenodo each release has a doi. There's also an overall "concept doi", which is pointer to the entire project (and by default redirects to the latest release).
For any given version, we can access a json of the metadata via a link like https://zenodo.org/api/records/4316550, where the number 4316550 comes form the doi: 10.5281/zenodo.4316550
Using this we can load the metadata straight into R
x <-
jsonlite::fromJSON("https://zenodo.org/api/records/4316550")
# or
# httr::GET("https://zenodo.org/api/records/4316550") |> httr::content(as = "text") |> fromJSON()This information includes links to the files, which we can use to download the file
url <- x$files$links$download[1]
filename <- basename( x$files$filename[1] )
download.file(url, filename)
So that will work for retrieving
- the latest version, whatever it is
- any version when you know the DOI
A remaining challenge is to discover all the versions associated with the conceptdoi. So far I can't see how to do that from zenodo. But it is possible via doi resolution service.
Once the dataset is public, we'll want a function that can pull the data directly from the Austraits page on Zenodo: https://doi.org/10.5281/zenodo.3568417
On Zenodo each release has a doi. There's also an overall "concept doi", which is pointer to the entire project (and by default redirects to the latest release).
For any given version, we can access a json of the metadata via a link like https://zenodo.org/api/records/4316550, where the number 4316550 comes form the doi: 10.5281/zenodo.4316550
Using this we can load the metadata straight into R
This information includes links to the files, which we can use to download the file
So that will work for retrieving
A remaining challenge is to discover all the versions associated with the conceptdoi. So far I can't see how to do that from zenodo. But it is possible via doi resolution service.