惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

The GitHub Blog
The GitHub Blog
Jina AI
Jina AI
月光博客
月光博客
博客园 - Franky
小众软件
小众软件
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
V
Visual Studio Blog
有赞技术团队
有赞技术团队
V
V2EX
IT之家
IT之家
阮一峰的网络日志
阮一峰的网络日志
Stack Overflow Blog
Stack Overflow Blog
H
Help Net Security
Apple Machine Learning Research
Apple Machine Learning Research
腾讯CDC
D
DataBreaches.Net
Hugging Face - Blog
Hugging Face - Blog
Martin Fowler
Martin Fowler
罗磊的独立博客
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
WordPress大学
WordPress大学
C
Check Point Blog
Microsoft Azure Blog
Microsoft Azure Blog
Microsoft Security Blog
Microsoft Security Blog

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
⭐ One line of code that fixed my jumping sticky header
Giulio · 2026-06-03 · via DEV Community

You've built a sticky header. It works. You ship it.

Then someone opens your site on a slow connection, scrolls down, and the whole page jolts — content jumps up by a few pixels right as the header sticks. It looks broken. And the worst part? You can't reproduce it locally, because on your machine the fonts are already cached.

I hit this exact bug in a navigation plugin I maintain. Here's what's actually happening, and the fix that finally killed it.

The setup

A sticky header that's position: fixed is pulled out of the document flow. To stop the content below from jumping up the moment the header detaches, you insert a placeholder — a spacer — with the same height as the header:

var header  = document.querySelector('.site-header');
var spacer  = document.querySelector('.header-spacer');

spacer.style.height = header.offsetHeight + 'px';

Measure the header once, set the spacer, done. On every demo this works perfectly.

Why it breaks in the real world

The trap is when you measure.

If your header uses a custom font (Google Fonts, a self-hosted webfont, anything loaded asynchronously), the browser renders the page first with a fallback system font, then swaps in the real font once it downloads. This is the font-display: swap behavior — great for performance, brutal for layout measurements.

So the sequence on a cold load is:

  1. Page renders with Arial (fallback). Your header is, say, 64px tall.
  2. Your JS runs, measures 64px, sets the spacer to 64px.
  3. The webfont finishes downloading. Your header re-renders in Poppins, which has a slightly taller line-height. Now the header is 68px tall.
  4. The spacer is still 64px. There's now a 4px mismatch — and the moment the header goes sticky, the content jumps to close that gap.

On your dev machine steps 3 happens instantly (font is cached), so you never see it. On a real visitor's first load, the font arrives hundreds of milliseconds later. That's the jump.

The fix

The browser gives you a promise that resolves when all fonts have finished loading: document.fonts.ready. Re-measure when it fires.

function recalcSpacer() {
    spacer.style.height = header.offsetHeight + 'px';
}

// Measure now (fallback font)...
recalcSpacer();

// ...and again once the real fonts are in.
if (document.fonts && document.fonts.ready) {
    document.fonts.ready.then(recalcSpacer);
}

// Fonts aside, the header also changes height on resize.
window.addEventListener('resize', recalcSpacer);

That's the core of it. But there was one more wrinkle.

The wrinkle: measuring an element that's already fixed

If the header is already sticky when the font swaps (visitor scrolled fast), offsetHeight can return the wrong value, because a fixed element's box doesn't always reflect what the spacer needs. The fix is to measure it in its natural state: briefly remove the fixed class, read the height, put it back — all synchronously, so the browser never paints the intermediate state and the user sees nothing.

function recalcSpacerForce() {
    var wasFixed = header.classList.contains('is-sticky');
    if (wasFixed) header.classList.remove('is-sticky');

    spacer.style.height = header.offsetHeight + 'px';

    if (wasFixed) header.classList.add('is-sticky');
}

document.fonts.ready.then(recalcSpacerForce);

Because the class is removed and re-added within the same synchronous block — no await, no setTimeout — the layout change never reaches the screen. The browser batches it. You get the correct measurement with zero visible flicker.

The takeaway

Any layout measurement you take before fonts load is provisional. If you're setting heights, offsets, or scroll thresholds based on text-containing elements, hook into document.fonts.ready and measure again. It's one line, and it removes a class of "it works on my machine" bugs that are almost impossible to catch in dev.

I shipped this in a free WordPress menu plugin (Giuliomax Menu Builder) where the sticky header is user-configurable with any Google Font — which is exactly the scenario that surfaces this bug at scale. If you want to see it in a real codebase: https://github.com/Giulio001/menux-free-version

Has this one bitten you too? Curious how others handle layout measurements around async fonts.

wordpress: https://wordpress.org/plugins/giuliomax-menu-builder/