Valid URLs containing more than one occurrence of string "http" are rejected

[tclapython]

Describe the bug
Commit 2ac828d "Support pasting multiple URLs in Add form" introduced extractURLs.js which is used to support multiple URL adds.

The link validation is const urlPattern = /https?://[^\s]+/gi;

this seems ok, but later

    .filter(url=>
        (url.match(/https?/g) || []).length == 1 &&
        isURL(url, { require_protocol: true, require_tld: true })

rejects valid URLs that contain the string http more than once. I have some of those in my bookmarks and noticed that new ones were getting rejected in the web UI.
Here are some examples of valid URLs that fail this check:

https://www.google.com/search?q=http
https://web.archive.org/web/20250225043633/https://www.nytimes.com/2025/02/24/science/asteroid-2024-yr4.html
https://www.tutorialspoint.com/http/index.htm

It's not a super crushing work stopper or anything, as I can paste half of the link to get it added, then later edit the link to add the rest and I suppose a different check is used because then it is accepted.

To Reproduce
Steps to reproduce the behavior:

1. Go to your web app raindrop.io
2. Click on '+ Add' to add a link
3. Paste one of the example links above, all of which are valid URLs, into the form
4. Click "Submit"
5. Receive error "Please enter a valid URL(s)"

Expected behavior
URLs of this form should be accepted as new link entries

Desktop:

• OS: Windows 11
• Browser Brave
• Version 1.75.181