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
{{ message }}
This repository was archived by the owner on Jan 28, 2025. It is now read-only.
I observed some inconsistencies between the live data in AHDS FHIR services and the data in parquet files exported by the corresponding FHIR-to-Synapse Sync Agent, specifically on the value of "coding" nested field. The fhir data has the proper JSON structure but in parquet files the same data is written in string formats.
Some examples I found across different FHIR resource types:
patient: identifier[].type.coding
fhir data: [{"system":"http://terminology.hl7.org/CodeSystem/v2-0203","code":"MR","display":"Medical Record Number"}]
parquet data: "[{\"system\":\"http://terminology.hl7.org/CodeSystem/v2-0203\",\"code\":\"MR\",\"display\":\"Medical Record Number\"}]"
Hi,
I observed some inconsistencies between the live data in AHDS FHIR services and the data in parquet files exported by the corresponding FHIR-to-Synapse Sync Agent, specifically on the value of "coding" nested field. The fhir data has the proper JSON structure but in parquet files the same data is written in string formats.
Some examples I found across different FHIR resource types:
[{"system":"http://terminology.hl7.org/CodeSystem/v2-0203","code":"MR","display":"Medical Record Number"}]"[{\"system\":\"http://terminology.hl7.org/CodeSystem/v2-0203\",\"code\":\"MR\",\"display\":\"Medical Record Number\"}]"[{"code":"age","display":"Age"}]"[{\"code\":\"age\",\"display\":\"Age\"}]"[{"code":"date","display":"Exact Date"}]"[{\"code\":\"date\",\"display\":\"Exact Date\"}]"There seems to be a pattern where all values of "coding" field are stored in string format instead of JSON structure.
Could anyone please check if you have seen the same issue?
Best regards,
Tony