Taking the last two labels of a hostname as the site made every .co.jp domain the same party. A $third-party rule needs the Public Suffix List
Background I run a small browser tool that checks ad-blocker filter rules (the Adblock Plus / uBlock Origin style syntax) and tells you whether a rule would match a given URL. https://hashitosystem.com/en/tools/filterlint/ You paste rules, enter a request URL and the page it loads on, and the tool shows which rules hit. The matcher is labelled "simplified" in the source, and I knew it cut…
The tool in question simplifies the process of determining if a website element is third-party by taking the last two labels of a hostname as the site. However, this method is flawed, as it fails to accurately distinguish between different entities on certain domains, particularly those ending in .co.jp. This oversight can lead to incorrect assessments of what is considered first or third-party content.
To rectify this issue, the correct approach involves identifying the longest public suffix plus an additional label from the Public Suffix List, which accurately determines the site's boundary. This method ensures that different companies and projects, such as tracker.co.jp and example.co.jp, are correctly identified as distinct entities, and GitHub Pages projects owned by different individuals are not mistakenly grouped together.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.