Stream: bulk data
Topic: Public NDJSON Files
Vladimir Ignatov (Nov 15 2018 at 17:58):
Since there is growing interest in the bulk-data lately, people are already working on population-level health apps and experiments, hoping to see the spec implemented in production as soon as possible. We are working on adding bulk-data support to the sandbox servers but it will take a while.
At the same time, the bulk-data server at https://bulk-data.smarthealthit.org is great for testing clients but it simply does not output data that is good enough to feed into those pop-level apps.
That is why we've decided to accelerate the process by exporting the data from the sandbox servers as NDJSON. Instead of waiting for usable bulk-data servers to be released, you can use these NDJSON files as if they were downloaded from a bulk-data export endpoint.
The files are now publicly available at https://github.com/smart-on-fhir/flat-fhir-files
Vladimir Ignatov (Nov 15 2018 at 18:26):
Note that EVERY resource has its corresponding ndjson file. The file might be empty if there is no data. Please tell us if you would prefer to have the empty files removed from the repository. This would make it look more like a bulk data export and you won't have to filter them yourself, but it would also make it harder to determine what resources are empty.
Last updated: Apr 12 2022 at 19:14 UTC
