Stream: bulk data
Topic: Healthcare Directories
Brian Postlethwaite (Feb 15 2019 at 03:23):
I've done a write-up on how we could do bulk data handling with Healthcare Directory content and looking for feedback from others here.
https://confluence.hl7.org/display/PA/Bulk+Data+and+Directories
Jason Walonoski (Feb 15 2019 at 14:26):
FWIW, I not sure about the proposed "deletions bundle." If you have a significant number of deleted resources (which seems reasonable when you're dealing with bulk data) that bundle could be enormous. Granted, lots of deletions might not be an issue with validated healthcare directories (?), but it seems what you do in this space might become a template for other areas. If deletions are really necessary, then I humbly suggest another nd-json file or set of files.
Brian Postlethwaite (Feb 15 2019 at 22:15):
Yes, set of bundles in the array. I might clarify that in the text.
Difference here is that the bundle itself can have multiple deletions in it. So what would the count be, the number of bundles or the number of deletions in the bundles.
Brian Postlethwaite (Feb 15 2019 at 22:16):
The bundle itself doesn't mean anything, other than the way to represent the resource deletion in the entry (the same way its represented elsewhere in fhir)
Last updated: Apr 12 2022 at 19:14 UTC
