FHIR Chat · database schema

Stream: hapi

Topic: database schema

nicola (RIO/SS) (May 29 2019 at 20:19):

Hey HAPI guys - I'm working on a presentation on DevDays about FHIR storage implementations - is there some good top-level overview for HAPI - how data are stored in the database and how search works. As I understand from a codebase you have a resource table for all resources and tables for each search parameter type - is it most popular configuration?

James Agnew (May 31 2019 at 09:29):

I don't know that there are any good online explanations of it unfortunately- But I'd say your understanding is correct FWIW.

nicola (RIO/SS) (May 31 2019 at 10:27):

Thank you, @James Agnew !

Y.T.Dai (Dec 16 2021 at 00:37):

Hello, I'm now useing hapi server, I found that the Schema from hapi docunment is not yet comprehensive.
There is a sample data from https://www.hl7.org/fhir/observation-example-sample-data.json.html
"valueSampledData": {
        "origin": {
          "value": 2048
        },
        "period": 10,
        "factor": 1.612,
        "lowerLimit": -3300,
        "upperLimit": 3300,
        "dimensions": 1,
        "data": "2041 2043 2037 2047 2060 2062 2051 2023 2014 2027 2034 2033 2040 2047 2047 2053"
}
The question is, where can I found "valueSampledData" in Hapi PostgreSQL database? I search for several tabels in database, but still can not find the resoult. Thanks!

James Agnew (Dec 16 2021 at 14:09):

Assuming you haven't added any search parameter indexes on that field (which I wouldn't recommend for a SampledData), you'll find this in the table HFJ_RES_VER in the column RES_TEXT. Note that in default config, this column has gzipped contents so you'll need to un-gzip the contents.

Lin Zhang (Dec 16 2021 at 14:34):

For PG, the gzipped json is actually stored in the PG's system table something like "large_object..." that I don't remember its exact name. And hfj_res_ver.res_text just holds the OID for the gzipped data. Right?