You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
The cqf-tooling produces a number of files as a result of an IG refresh that are quite large. In particular, the bundle directory and supporting directories can multiple GBs. As an example, see the Smart ANC IG. This makes checking out and using the repositories burdensome. Additionally, uploading large JSON files to a FHIR server can be problematic.
Describe the solution you'd like
I'd like the directories and files to be smaller. One way to accomplish this would be to add support for reading/writing GZipped files to the tooling directly so that JSON on disk is always compressed (this would apply to build output only - the source files still need to be human-readable).
In principle a FHIR server configured to accept GZipped content could use these files directly. We wouldn't need to decompress them prior to uploading to the FHIR server.
Describe alternatives you've considered
Another alternative would be to simply delete the duplicate files. This would reduce the on-disk size by around 50%, but we'd still have the issue of uploading large files to a FHIR server.
The text was updated successfully, but these errors were encountered:
JPercival
changed the title
Compress build/bundle files using GZip (comparable method)
Compress build/bundle files using GZip (or comparable method)
Dec 27, 2021
Is your feature request related to a problem? Please describe.
The cqf-tooling produces a number of files as a result of an IG refresh that are quite large. In particular, the
bundle
directory and supporting directories can multiple GBs. As an example, see the Smart ANC IG. This makes checking out and using the repositories burdensome. Additionally, uploading large JSON files to a FHIR server can be problematic.Describe the solution you'd like
I'd like the directories and files to be smaller. One way to accomplish this would be to add support for reading/writing GZipped files to the tooling directly so that JSON on disk is always compressed (this would apply to build output only - the source files still need to be human-readable).
In principle a FHIR server configured to accept GZipped content could use these files directly. We wouldn't need to decompress them prior to uploading to the FHIR server.
Describe alternatives you've considered
Another alternative would be to simply delete the duplicate files. This would reduce the on-disk size by around 50%, but we'd still have the issue of uploading large files to a FHIR server.
The text was updated successfully, but these errors were encountered: