d8a80f6272
이름만 다른(표기 변형) [[위키링크]]를 대상 문서의 canonical 제목으로 치환해 끊겼던 1,200개 링크를 연결. 제목/파일명 정규화 일치만 적용하고 별칭 매칭은 과병합 위험으로 제외(애매성 가드). 원본은 _link_reconcile_backup/ 에 백업. 도구: Datacollect/scripts/link_reconcile_apply.mjs Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
6.2 KiB
6.2 KiB
id, title, category, status, canonical_id, aliases, duplicate_of, source_trust_level, confidence_score, verification_status, tags, raw_sources, last_reinforced, github_commit, tech_stack
| id | title | category | status | canonical_id | aliases | duplicate_of | source_trust_level | confidence_score | verification_status | tags | raw_sources | last_reinforced | github_commit | tech_stack | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| wiki-2026-0508-long-animation-frames-api | Long Animation Frames API | 10_Wiki/Topics | verified | self |
|
none | A | 0.9 | applied |
|
2026-05-10 | pending |
|
Long Animation Frames API
매 한 줄
"매 Long Tasks 의 후계자". 50ms+ frame 을 잡고 어떤 script 가 원인인지 attribution 까지 — INP 디버깅의 핵심.
매 핵심
매 vs Long Tasks
| Long Tasks | Long Animation Frames (LoAF) | |
|---|---|---|
| 단위 | 단일 task | rendering frame 전체 (start ~ render) |
| Threshold | 50ms+ | 50ms+ frame duration |
| Attribution | 거의 없음 | scripting source URL, invoker, function |
| Render 정보 | 없음 | renderStart, styleAndLayoutStart, paint |
매 timing 분해
startTime→ frame 시작renderStart→ render phase 시작styleAndLayoutStart→ style/layout 시작duration→ 총 frameblockingDuration→ blocking 시간scripts[]→ 어떤 스크립트가 얼마나 점유
매 응용
- INP regression 디버깅
- 3rd-party script 영향 정량화
- Hydration cost 측정 (Next/Nuxt)
- RUM 에 attribution 보내기
- Frame budget violation 알림
💻 패턴
Pattern 1: Basic observer
const obs = new PerformanceObserver((list) => {
for (const entry of list.getEntries()) {
if (entry.duration < 50) continue;
console.log({
duration: entry.duration,
blocking: entry.blockingDuration,
renderStart: entry.renderStart,
scripts: entry.scripts.map(s => ({
name: s.name,
invoker: s.invoker,
source: s.sourceURL,
duration: s.duration,
})),
});
}
});
obs.observe({ type: "long-animation-frame", buffered: true });
Pattern 2: Send to RUM with worst script
new PerformanceObserver((list) => {
for (const e of list.getEntries()) {
const worst = e.scripts.reduce((a, b) => b.duration > a.duration ? b : a, { duration: 0 });
sendBeacon("/rum/loaf", JSON.stringify({
url: location.href,
duration: e.duration,
blocking: e.blockingDuration,
worstScript: worst.sourceURL,
worstFn: worst.invoker,
}));
}
}).observe({ type: "long-animation-frame", buffered: true });
Pattern 3: Budget violation alarm
const FRAME_BUDGET_MS = 100;
new PerformanceObserver((list) => {
for (const e of list.getEntries()) {
if (e.duration > FRAME_BUDGET_MS) {
reportBudgetMiss({
duration: e.duration,
scripts: e.scripts.map(s => s.sourceURL),
});
}
}
}).observe({ type: "long-animation-frame", buffered: true });
Pattern 4: 3rd-party attribution
function attributeScript(s) {
const u = new URL(s.sourceURL || location.href);
if (u.hostname === location.hostname) return "first-party";
if (/google|facebook|hotjar|segment/.test(u.hostname)) return "analytics";
return u.hostname;
}
new PerformanceObserver((list) => {
const buckets = {};
for (const e of list.getEntries())
for (const s of e.scripts) {
const k = attributeScript(s);
buckets[k] = (buckets[k] || 0) + s.duration;
}
console.table(buckets);
}).observe({ type: "long-animation-frame", buffered: true });
Pattern 5: Hydration measurement
let hydrationFrames = [];
new PerformanceObserver((list) => {
if (performance.now() < 5000) { // 첫 5s = hydration phase 가정
hydrationFrames.push(...list.getEntries());
}
}).observe({ type: "long-animation-frame", buffered: true });
window.addEventListener("load", () => {
console.log("Hydration LoAFs:", hydrationFrames.length,
"total blocking:", hydrationFrames.reduce((s, e) => s + e.blockingDuration, 0));
});
Pattern 6: Tie LoAF to INP event
let recentLoafs = [];
new PerformanceObserver(list => {
recentLoafs.push(...list.getEntries());
recentLoafs = recentLoafs.slice(-20);
}).observe({ type: "long-animation-frame", buffered: true });
new PerformanceObserver(list => {
for (const ev of list.getEntries()) {
const overlapping = recentLoafs.filter(l =>
l.startTime <= ev.startTime + ev.duration && l.startTime + l.duration >= ev.startTime
);
console.log("INP", ev.duration, "overlapping LoAFs:", overlapping);
}
}).observe({ type: "event", durationThreshold: 40 });
Pattern 7: Feature detection
const supportsLoAF = PerformanceObserver.supportedEntryTypes?.includes("long-animation-frame");
if (supportsLoAF) { /* observe */ }
else { /* fallback to long-task */ }
매 결정 기준
| 상황 | API |
|---|---|
| INP debugging | LoAF (attribution 필수) |
| 단순 long task 개수 | longtask 도 충분 |
| Production RUM | LoAF (worst script만 전송) |
| 3rd-party 영향 분석 | LoAF |
| 호환성 필요 | LoAF + longtask fallback |
기본값: LoAF observer + RUM beacon + budget alarm.
🔗 Graph
- 부모: Web_Performance
- 변형: Long Tasks (구세대)
- 응용: INP, Core Web Vitals Optimization (INP, LCP, CLS)
- Adjacent: Lighthouse, RUM
🤖 LLM 활용
언제: LoAF dump → "어떤 script 가 가장 비용 비싼가" 분석, 패턴 인식. 언제 X: 실시간 sampling 결정 (deterministic threshold), 보안 critical attribution (cross-origin source 제한).
❌ 안티패턴
- 모든 LoAF 를 서버로 전송 → bandwidth 폭발 (worst만 전송)
- Long Task 만 보고 INP 디버깅 → 원인 못 찾음
buffered: true안 씀 → 초기 frame 놓침- Feature detection 없이 사용 → 구 브라우저 throw
- Cross-origin script
sourceURL가려짐 무시 → "(unknown)" 비율 보고
🧪 검증 / 중복
- Verified (W3C LongAnimationFrameTiming spec, web.dev/articles/long-animation-frames). 신뢰도 A.
- Chrome 123+ 안정 지원, Safari/Firefox 진행 중 (2026-05 기준).
🕓 Changelog
| 날짜 | 변경 |
|---|---|
| 2026-05-08 | Phase 1 |
| 2026-05-10 | Manual cleanup — vs LongTasks, attribution + INP correlation patterns |