⚡ Bolt: Optimize $O(N^2)$ list inclusion checks in case extraction

.jules/bolt.md

@@ -0,0 +1,3 @@
+## 2024-05-02 - O(N^2) Anti-Pattern in Unique Value Extraction
+**Learning:** O(N) inclusion checks (`not in seen` where `seen` is a list) inside an O(N) loop when filtering uniqueness out of cases records in `blocking_commitment_ids` and `blocked_peer_host_ids` scales quadratically O(N^2) and is noticeably slower even on small N due to repeated list scans. `cases.py` functions use this pattern, limiting performance as the number of cases grows.
+**Action:** Always use sets (`seen = set()`) for O(1) inclusion checks and uniqueness constraints when accumulating IDs or references from loops, then convert the set back to a sorted list (`sorted(list(seen))` or `sorted(seen)`) at the end. I will apply this to `kernel/kernel/cases.py` and `intelligence/company/meridian_platform/cases.py` for `blocking_commitment_ids` and `blocked_peer_host_ids`.

intelligence/company/meridian_platform/cases.py

@@ -230,21 +230,21 @@ def peer_can_be_thawed(peer_host_id, org_id=None, *, claim_types=None):
 
 
 def blocking_commitment_ids(org_id=None):
-    seen = []
+    seen = set()
     for row in blocking_cases(org_id):
         commitment_id = (row.get('linked_commitment_id') or '').strip()
-        if commitment_id and commitment_id not in seen:
-            seen.append(commitment_id)
-    return sorted(seen)
+        if commitment_id:
+            seen.add(commitment_id)
+    return sorted(list(seen))
 
 
 def blocked_peer_host_ids(org_id=None):
-    seen = []
+    seen = set()
     for row in blocking_cases(org_id):
         peer_host_id = (row.get('target_host_id') or '').strip()
-        if peer_host_id and case_requires_peer_block(row) and peer_host_id not in seen:
-            seen.append(peer_host_id)
-    return sorted(seen)
+        if peer_host_id and case_requires_peer_block(row):
+            seen.add(peer_host_id)
+    return sorted(list(seen))
 
 
 def _commitment_counterparty_binding(commitment_record):

kernel/kernel/cases.py

@@ -309,21 +309,21 @@ def blocking_peer_case(peer_host_id, org_id=None, *, claim_types=None):
 
 
 def blocking_commitment_ids(org_id=None):
-    seen = []
+    seen = set()
     for row in blocking_cases(org_id):
         commitment_id = (row.get('linked_commitment_id') or '').strip()
-        if commitment_id and commitment_id not in seen:
-            seen.append(commitment_id)
-    return sorted(seen)
+        if commitment_id:
+            seen.add(commitment_id)
+    return sorted(list(seen))
 
 
 def blocked_peer_host_ids(org_id=None):
-    seen = []
+    seen = set()
     for row in blocking_cases(org_id):
         peer_host_id = (row.get('target_host_id') or '').strip()
-        if peer_host_id and case_requires_peer_block(row) and peer_host_id not in seen:
-            seen.append(peer_host_id)
-    return sorted(seen)
+        if peer_host_id and case_requires_peer_block(row):
+            seen.add(peer_host_id)
+    return sorted(list(seen))
 
 
 def _commitment_counterparty_binding(commitment_record):