⚡ Bolt: 성능 개선 (DBML 파싱 O(N^2) 해결 및 프론트엔드 GC 오버헤드 감소)

.jules/bolt.md

@@ -77,3 +77,6 @@ Optimized metric route processing to O(N) by creating a mapping of routes direct
 ## 2024-07-13 - [Optimize Export Dictionary FK lookups]
 **Learning:** Found O(N * C * E) performance bottleneck in ERD export dictionaries due to repeated array searching with `edges.some()` inside a nested loop over nodes and columns.
 **Action:** Replace repeated linear array scans for edges by precomputing O(1) Set lookups of foreign key column handles per node before looping.
+## 2024-07-15 - Optimize DBML import and Frontend GC overhead
+**Learning:** Using `sum(1 for ...)` inside a loop to calculate positions causes O(N^2) complexity, and `Array.from(string)` creates unnecessary array allocations leading to GC pressure.
+**Action:** Introduced O(1) dictionary counter in `dbml_import.py` and replaced `Array.from` with a `for...of` loop in `handleUtils.ts`.

CHANGELOG.md

@@ -11,3 +11,5 @@
 - [FE] `autoInfer.ts`에 대한 단위 테스트 및 UI 컴포넌트 단위 테스트를 추가하여 100% 테스트 커버리지를 유지합니다.
 - [FE] ⬇️ **DBML Export**: ERD 다이어그램을 DBML (Database Markup Language) 형식으로 내보낼 수 있는 기능을 추가했습니다. 상단의 DBML 버튼을 클릭하여 다운로드할 수 있습니다.
 - [FE] 📚 **Data Dictionary Export**: ERD 테이블/컬럼 메타데이터를 CSV 및 Markdown으로 내보내며, CSV formula injection과 Markdown 렌더링 escape를 적용했습니다.
+
+- [Performance] ⚡ **Bolt**: 백엔드 DBML 임포트 처리에서 O(N^2) 병목을 개선하고, 프론트엔드의 ERD 핸들 ID 생성 시 메모리 가비지 컬렉션(GC) 부하를 최적화했습니다.

backend/app/spec/dbml_import.py

@@ -131,6 +131,7 @@ def parse_dbml(text: str) -> dict[str, Any]:
     fk_specs: list[tuple[str, str, str, str, str, str]] = []  # child s/t/c, parent s/t/c
 
     oid_by_table: dict[tuple[str, str], int] = {}
+    col_count_by_oid: dict[int, int] = {}
     next_oid = 1
     current: tuple[str, str] | None = None
     in_ignored_block = 0
@@ -206,12 +207,16 @@ def parse_dbml(text: str) -> dict[str, Any]:
         col_name = (cm.group("qname") or cm.group("name")).strip('"')
         settings = (cm.group("settings") or "").lower()
         oid = oid_by_table[current]
+
+        col_position = col_count_by_oid.get(oid, 0) + 1
+        col_count_by_oid[oid] = col_position
+
         is_pk = bool(re.search(r"\bpk\b|primary\s+key", settings))
         columns.append(
             {
                 "relation_oid": oid,
                 "column_name": col_name,
-                "column_position": sum(1 for c in columns if c["relation_oid"] == oid) + 1,
+                "column_position": col_position,  # ⚡ Bolt: Use O(1) dictionary counter instead of O(N^2) inline generator expression for column positions.
                 "data_type": cm.group("type"),
                 "is_not_null": is_pk or "not null" in settings,
                 "has_default": "default:" in settings,

frontend/src/erd/handleUtils.ts

@@ -1,10 +1,12 @@
 export function sanitizeHandleId(columnName: string): string {
-  const encoded = Array.from(columnName, (char) => {
-    // Array.from only yields non-empty Unicode scalars, so codePointAt(0) is defined.
-    return char.codePointAt(0)!.toString(16).padStart(4, '0')
-  }).join('-')
+  const codes: string[] = [];
+  // ⚡ Bolt: Avoid Array.from(string) to prevent intermediate array allocations and reduce GC overhead.
+  for (const char of columnName) {
+    codes.push(char.codePointAt(0)!.toString(16).padStart(4, '0'));
+  }
+  const encoded = codes.join('-');
 
-  return `c-${encoded || 'empty'}`
+  return `c-${encoded || 'empty'}`;
 }
 
 export function sourceColumnHandleId(columnName: string): string {