fix(harvard_merge): adjust havard_id column type, index

- Use better indexing approach
- Use IntegerField
- Make nullable

cl/search/migrations/0037_add_harvard_id_to_opinioncluster.py

@@ -1,4 +1,4 @@
-# Generated by Django 5.1.2 on 2024-10-25 18:54
+# Generated by Django 5.1.2 on 2024-10-25 22:37
 
 import pgtrigger.compiler
 import pgtrigger.migrations
@@ -8,6 +8,7 @@
 class Migration(migrations.Migration):
 
     dependencies = [
+        ("people_db", "0017_pghistory_v3_4_0_trigger_update"),
         ("search", "0036_add_searchquery"),
     ]
 
@@ -23,18 +24,23 @@ class Migration(migrations.Migration):
         migrations.AddField(
             model_name="opinioncluster",
             name="harvard_id",
-            field=models.CharField(
-                db_index=True,
-                default=0,
+            field=models.IntegerField(
                 help_text="The ID of the item in the Caselaw Access Project (Harvard)",
+                null=True,
             ),
         ),
         migrations.AddField(
             model_name="opinionclusterevent",
             name="harvard_id",
-            field=models.CharField(
-                default=0,
+            field=models.IntegerField(
                 help_text="The ID of the item in the Caselaw Access Project (Harvard)",
+                null=True,
+            ),
+        ),
+        migrations.AddIndex(
+            model_name="opinioncluster",
+            index=models.Index(
+                fields=["harvard_id"], name="search_opin_harvard_9f1d7e_idx"
             ),
         ),
         pgtrigger.migrations.AddTrigger(

cl/search/migrations/0037_add_harvard_id_to_opinioncluster.sql

@@ -1,8 +1,5 @@
 BEGIN;
-ALTER TABLE "search_opinioncluster" ADD COLUMN "harvard_id" varchar DEFAULT '0' NOT NULL;
-ALTER TABLE "search_opinioncluster" ALTER COLUMN "harvard_id" DROP DEFAULT;
-ALTER TABLE "search_opinionclusterevent" ADD COLUMN "harvard_id" varchar DEFAULT '0' NOT NULL;
-ALTER TABLE "search_opinionclusterevent" ALTER COLUMN "harvard_id" DROP DEFAULT;
-CREATE INDEX "search_opinioncluster_harvard_id_b7c3eb52" ON "search_opinioncluster" ("harvard_id");
-CREATE INDEX "search_opinioncluster_harvard_id_b7c3eb52_like" ON "search_opinioncluster" ("harvard_id" varchar_pattern_ops);
+ALTER TABLE "search_opinioncluster" ADD COLUMN "harvard_id" integer NULL;
+ALTER TABLE "search_opinionclusterevent" ADD COLUMN "harvard_id" integer NULL;
+CREATE INDEX "search_opin_harvard_9f1d7e_idx" ON "search_opinioncluster" ("harvard_id");
 COMMIT;

cl/search/migrations/0037_add_harvard_id_to_opinioncluster_customers.sql

@@ -1,6 +1,4 @@
 BEGIN;
-ALTER TABLE "search_opinioncluster" ADD COLUMN "harvard_id" varchar DEFAULT '0' NOT NULL;
-ALTER TABLE "search_opinioncluster" ALTER COLUMN "harvard_id" DROP DEFAULT;
-CREATE INDEX "search_opinioncluster_harvard_id_b7c3eb52" ON "search_opinioncluster" ("harvard_id");
-CREATE INDEX "search_opinioncluster_harvard_id_b7c3eb52_like" ON "search_opinioncluster" ("harvard_id" varchar_pattern_ops);
+ALTER TABLE "search_opinioncluster" ADD COLUMN "harvard_id" integer NULL;
+CREATE INDEX "search_opin_harvard_9f1d7e_idx" ON "search_opinioncluster" ("harvard_id");
 COMMIT;

cl/search/models.py

@@ -2731,10 +2731,9 @@ class OpinionCluster(AbstractDateTimeModel):
         storage=IncrementingAWSMediaStorage(),
         blank=True,
     )
-    harvard_id = models.CharField(
+    harvard_id = models.IntegerField(
         help_text="The ID of the item in the Caselaw Access Project (Harvard)",
-        default=0,
-        db_index=True,
+        null=True,
     )
     arguments = models.TextField(
         help_text="The attorney(s) and legal arguments presented as HTML text. "
@@ -2784,6 +2783,11 @@ class OpinionCluster(AbstractDateTimeModel):
         ]
     )
 
+    class Meta:
+        indexes = [
+            models.Index(fields=["harvard_id"]),
+        ]
+
     async def acaption(self):
         """Make a proper caption
 

cl/search/tests/test_import_harvard_pdfs.py

@@ -183,4 +183,4 @@ def test_assign_harvard_id(
 
         # Verify that the cluster's harvard_id field was updated
         self.cluster.refresh_from_db()
-        self.assertEqual(self.cluster.harvard_id, "1")
+        self.assertEqual(self.cluster.harvard_id, 1)