Adding comments that explain motivation behind row_id and person_seq_id.

OHDSI · Mar 17, 2023 · 456537f · 456537f
1 parent 457f0ed
commit 456537f
Show file tree

Hide file tree

Showing 2 changed files with 4 additions and 0 deletions.
diff --git a/inst/sql/sql_server/CreateCohorts.sql b/inst/sql/sql_server/CreateCohorts.sql
@@ -32,6 +32,8 @@ limitations under the License.
 
 DROP TABLE IF EXISTS #cohort_person;
 
+-- Generating a unique row_id that can be used instead of the compound key
+-- (subjectId, cohort_definition_id, cohort_start_date) for easier joining etc.
 SELECT ROW_NUMBER() OVER (ORDER BY person_id, cohort_start_date) AS row_id,
 	subject_id,
 	cohort_definition_id,

diff --git a/inst/sql/sql_server/GetCohorts.sql b/inst/sql/sql_server/GetCohorts.sql
@@ -1,6 +1,8 @@
 {DEFAULT @target_id = ''}
 {DEFAULT @sampled = FALSE}
 
+-- Storing person_id as a VARCHAR because R doesn't support 64-bit integers. Generating 
+-- a proper integer (person_seq_id) on the fly for easy joining, etc.
 SELECT row_id,
 	person_seq_id,
 	CAST(cohort.subject_id AS VARCHAR(30)) AS person_id,