Normalize sp_name and res_type columns from hfj_spidx_* tables - adde…

…d sql migration scripts

hapi-fhir-sql-migrate/src/main/resources/ca/uhn/fhir/jpa/migrate/taskdef/populate_spidx_identity/mssql.sql

@@ -0,0 +1,31 @@
+WITH cte_min_id AS (
+    SELECT
+        hash_identity,
+        MIN(sp_id) AS min_sp_id
+    FROM {0} -- replace {0} with hfj_spidx_* table
+    WHERE hash_identity NOT IN (
+        SELECT hash_identity
+        FROM hfj_spidx_identity
+    )
+    GROUP BY hash_identity
+)
+INSERT INTO HFJ_SPIDX_IDENTITY (
+    sp_identity_id,
+    hash_identity,
+    res_type,
+    sp_name
+)
+SELECT
+    NEXT VALUE FOR seq_spidx_identity OVER (ORDER BY src.res_type, src.sp_name) AS sp_identity_id,
+    src.hash_identity,
+    src.res_type,
+    src.sp_name
+FROM {0} AS src -- replace {0} with hfj_spidx_* table
+    JOIN cte_min_id AS cm
+ON src.hash_identity = cm.hash_identity
+    AND src.sp_id = cm.min_sp_id
+ORDER BY
+    src.res_type,
+    src.sp_name;
+
+

hapi-fhir-sql-migrate/src/main/resources/ca/uhn/fhir/jpa/migrate/taskdef/populate_spidx_identity/oracle.sql

@@ -0,0 +1,6 @@
+SELECT
+    hash_identity,
+    MAX(sp_name)   KEEP (DENSE_RANK FIRST ORDER BY sp_id) AS sp_name,
+    MAX(res_type)  KEEP (DENSE_RANK FIRST ORDER BY sp_id) AS res_type
+FROM hfj_spidx_token
+GROUP BY hash_identity;

hapi-fhir-sql-migrate/src/main/resources/ca/uhn/fhir/jpa/migrate/taskdef/populate_spidx_identity/postgres.sql

@@ -0,0 +1,31 @@
+INSERT INTO HFJ_SPIDX_IDENTITY (
+    sp_identity_id,
+    hash_identity,
+    res_type,
+    sp_name
+)
+WITH cte_min_id AS (
+    SELECT
+        hash_identity,
+        MIN(sp_id) AS min_sp_id
+    FROM {0} -- replace {0} with hfj_spidx_* table
+    WHERE hash_identity NOT IN (
+        SELECT hash_identity
+        FROM hfj_spidx_identity
+    )
+    GROUP BY hash_identity
+)
+SELECT
+    nextval(''SEQ_SPIDX_IDENTITY'') AS sp_identity_id,
+    src.hash_identity,
+    src.res_type,
+    src.sp_name
+FROM {0} AS src -- replace {0} with hfj_spidx_* table
+JOIN cte_min_id AS cm
+   ON src.hash_identity = cm.hash_identity
+  AND src.sp_id         = cm.min_sp_id
+ORDER BY src.res_type, src.sp_name;
+
+SELECT DISTINCT on (hash_identity)
+    hash_identity
+FROM hfj_spidx_token