⚡ Bolt: [performance improvement] Replace O(N^2) Array.find with O(N) Map lookup in OpenAPI scanner

.jules/bolt.md

@@ -8,3 +8,7 @@
 ## 2024-05-24 - Pre-compile RegExp in nested loops
 **Learning:** Instantiating `new RegExp()` inside nested array methods like `.filter` and `.some` creates a severe O(N*M) performance bottleneck, especially when matching two large lists (e.g., documented tests vs. actual test files).
 **Action:** Always pre-compile regular expressions and derived strings into an array of "matcher" objects outside of the loop before iterating, which shifts the instantiation cost from O(N*M) to O(N).
+
+## 2024-05-25 - O(N*M) Array.find bottlenecks
+**Learning:** Using `Array.find` inside nested loops to match entity properties (e.g., schemas and fields) causes an $O(N^2)$ algorithmic bottleneck, especially noticeable when parsing large documents like OpenAPI specifications.
+**Action:** Always precompute a `Map` of the target array for $O(1)$ lookups based on the identifier key before entering the nested loops.

cli/scanners/schemas.mjs

@@ -713,11 +713,16 @@ function scanRailsModels(dir) {
 
 function extractOpenAPIRelationships(schemas) {
   const relationships = [];
+
+  // PERFORMANCE OPTIMIZATION: Precompute an O(1) Map lookup of lowercased schema names
+  // to avoid O(N*M) array find lookups during the nested schema/field loops.
+  const schemaMap = new Map(schemas.map(s => [s.name.toLowerCase(), s]));
+
   for (const schema of schemas) {
     for (const field of schema.fields) {
       if (field.type !== 'string' && field.type !== 'number' && field.type !== 'boolean' && field.type !== 'integer') {
         // Likely a reference to another schema
-        const target = schemas.find(s => s.name.toLowerCase() === field.type.toLowerCase());
+        const target = schemaMap.get(field.type.toLowerCase());
         if (target) {
           relationships.push({
             from: schema.name,