Date: 2026-09-06 13:11 Summary: convert-html-to-pdf no longer auto-fetches or auto-converts on page entry; Load URL fills preview only, Convert runs on click. Visibility: people
[question] ”Could not load that URL. Check the address, or the site may block fetches“ 这个不再做自动load，点击才load convert
Tool links (English)
- https://onlinefreetools.org/tools/convert-html-to-pdf Chinese: https://onlinefreetools.org/zh/tools/convert-html-to-pdf
[try to solve]
Context
Opening the tool showed the URL-fetch error even though the visitor had not asked to load a page. The URL field defaults to https://example.com. Page-entry loadSample() plus Load URL calling runConvert() made fetch/convert happen without an explicit Convert click. Sites that block the Worker fetch then painted Could not load that URL… immediately.
Process
- Removed the init
loadSample()call. TheloadSamplefunction stays for the Sample button (lint still requires it). - Load URL now writes the fetched HTML into the textarea and refreshes the HTML preview only. It does not call
runConvert(). - Convert still runs only on the Convert button (and Sample / Paste, which are explicit clicks).
- Local check: after 2.5s idle, error empty, textarea empty, no
/api/tools/convert-html-to-pdfrequests, download disabled. Sample click with wait 0 still reaches 「完成」.
Root cause / analysis
Auto-run on entry treated a default URL / sample as a user request. A blocked fetch then looked like a broken tool. Fetch and PDF export should be separate, click-driven steps.
Solution
Idle page is empty aside from the default URL value. Click Load URL to fetch; click Convert to export PDF. PAGES_CACHE_VERSION 4.35.
Notes / boundaries
- Enter in the URL field still triggers Load URL (same as the button), not Convert.
- Sample still converts immediately because that button means “run a demo.”
[actions]
src/pages/convertHtmlToPdfPage.ts: drop init auto-run; Load URL no longer convertswrangler.jsonc: cache 4.35src/site/tool-catalog.d/convert-html-to-pdf.json:updatedAt