What's Fixed

2 What's Fixed

The following Healthcare Data Repository (HDR) Fast Healthcare Interoperability Resources (FHIR) bugs/enhancements are fixed in this release:

Table 2-1 Fixed in 8.2.0.5

Bug Number	Description
Enhancement 38064577	The HDR FHIR server did not perform a soft delete when invoking the $expunge operation via a POST request at the instance level. As a result, permanently deleting a resource required a two-step process: 1. Soft delete using the DELETE operation. 2. Hard delete using the $expunge operation. Now, the HDR FHIR server supports performing both soft delete and hard delete in a single step using the $expunge POST operation at the instance level. This behavior can be controlled through the following option: Using an HTTP Request Header Set the following http header in the request: X-Force-Delete-Expunge: true If this header is present and set to true, the server will perform both soft delete and hard delete in a single step via the $expunge POST operation at instance level. If this header is not present, or if it is set to false or any value other than true, the server will default to the original two-step process.
Enhancement 38103866	The HDR FHIR server could not be configured to disable caching of FHIR search query strings and their associated results. As a result, for every FHIR search query invocation, entries were inserted into the HFJ_SEARCH, HFJ_SEARCH_INCLUDE and HFJ_SEARCH_RESULT tables. This led to enq: TX - index contention issues in the database due to frequent insert operations. Now, the HDR FHIR server can be configured to disable FHIR search queries and the associated result caching in the database using the following configuration property: search_query_result_cache_enabled: false If this property is set to false, the server will skip caching entries in the HFJ_SEARCH, HFJ_SEARCH_INCLUDE, and HFJ_SEARCH_RESULT tables during each FHIR search query execution. If this property is not present, or if it is set to true, the server will cache search queries and their results in the respective tables by default.

Bug Number

Description

Enhancement 38064577

The HDR FHIR server did not perform a soft delete when invoking the $expunge operation via a POST request at the instance level. As a result, permanently deleting a resource required a two-step process:

1. Soft delete using the DELETE operation.

2. Hard delete using the $expunge operation.

Now, the HDR FHIR server supports performing both soft delete and hard delete in a single step using the $expunge POST operation at the instance level. This behavior can be controlled through the following option:

Using an HTTP Request Header

Set the following http header in the request:

X-Force-Delete-Expunge: true

If this header is present and set to true, the server will perform both soft delete and hard delete in a single step via the $expunge POST operation at instance level.

If this header is not present, or if it is set to false or any value other than true, the server will default to the original two-step process.

Enhancement 38103866

The HDR FHIR server could not be configured to disable caching of FHIR search query strings and their associated results. As a result, for every FHIR search query invocation, entries were inserted into the HFJ_SEARCH, HFJ_SEARCH_INCLUDE and HFJ_SEARCH_RESULT tables. This led to enq: TX - index contention issues in the database due to frequent insert operations.

Now, the HDR FHIR server can be configured to disable FHIR search queries and the associated result caching in the database using the following configuration property:

search_query_result_cache_enabled: false

If this property is set to false, the server will skip caching entries in the HFJ_SEARCH, HFJ_SEARCH_INCLUDE, and HFJ_SEARCH_RESULT tables during each FHIR search query execution.
If this property is not present, or if it is set to true, the server will cache search queries and their results in the respective tables by default.