From c832b3a3e242e895929961f97d1251e9ddd5b431 Mon Sep 17 00:00:00 2001
From: DevSoftTech Server <info.devsofttech@gmail.com>
Date: Thu, 2 Jul 2026 19:09:36 +0530
Subject: [PATCH] Add remote SQL script for investigation category backfill

---
 ...-07-02_backfill_investigation_category.sql | 25 +++++++++++++++++
 .../README_investigation_category_backfill.md | 28 +++++++++++++++++++
 2 files changed, 53 insertions(+)
 create mode 100644 scripts/sql/2026-07-02_backfill_investigation_category.sql
 create mode 100644 scripts/sql/README_investigation_category_backfill.md

diff --git a/scripts/sql/2026-07-02_backfill_investigation_category.sql b/scripts/sql/2026-07-02_backfill_investigation_category.sql
new file mode 100644
index 0000000..fbfa02e
--- /dev/null
+++ b/scripts/sql/2026-07-02_backfill_investigation_category.sql
@@ -0,0 +1,25 @@
+-- Backfill blank investigation.category_name values using name-based rules.
+-- Safe to run multiple times: updates only rows where category_name is NULL/blank.
+
+START TRANSACTION;
+
+UPDATE investigation
+SET category_name = CASE
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(x[[:space:]]*[-]?[[:space:]]*ray|xray|radiograph|radiology)([^a-z]|$)' THEN 'X-Ray'
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(ultra[[:space:]]*sound|ultrasound|u\\.?s\\.?g\\.?|usg|sonography|doppler)([^a-z]|$)' THEN 'Ultra Sound'
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(ct[[:space:]]*[-]?[[:space:]]*scan|cect|hrct|ncct|computed tomography)([^a-z]|$)' THEN 'CT-Scan'
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(mri|mr[[:space:]]*i|magnetic resonance)([^a-z]|$)' THEN 'MRI'
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(echo|echocardiography|2d[[:space:]]*echo|doppler[[:space:]]*echo)([^a-z]|$)' THEN 'ECHO'
+    WHEN LOWER(Name) REGEXP '(^|[^a-z])(ecg|electrocardiogram|tmt|holter)([^a-z]|$)' THEN 'Cardiology'
+    ELSE 'Pathology'
+END
+WHERE category_name IS NULL OR TRIM(category_name) = '';
+
+SELECT ROW_COUNT() AS updated_rows;
+
+SELECT category_name, COUNT(*) AS total_rows
+FROM investigation
+GROUP BY category_name
+ORDER BY total_rows DESC, category_name ASC;
+
+COMMIT;
diff --git a/scripts/sql/README_investigation_category_backfill.md b/scripts/sql/README_investigation_category_backfill.md
new file mode 100644
index 0000000..df9ed94
--- /dev/null
+++ b/scripts/sql/README_investigation_category_backfill.md
@@ -0,0 +1,28 @@
+# Investigation Category Backfill (Remote)
+
+Run this SQL on remote DB:
+
+scripts/sql/2026-07-02_backfill_investigation_category.sql
+
+## Example (MySQL CLI)
+
+```bash
+mysql -u <db_user> -p <db_name> < scripts/sql/2026-07-02_backfill_investigation_category.sql
+```
+
+## What it does
+
+- Updates only blank category values in investigation.category_name
+- Uses name-based rules to map into:
+  - Pathology
+  - X-Ray
+  - Ultra Sound
+  - CT-Scan
+  - MRI
+  - ECHO
+  - Cardiology
+- Prints updated row count and final category distribution
+
+## Idempotency
+
+Safe to re-run. It only touches rows where category_name is blank/null.