google/global.txt: Can sometimes freak out on Google Images

[brian6932]

Eats every result, regardless of them being spam or not.

[chuahou]

This issue occurs for me as well on a variety of searches.

With the search query "memtest menu" in private browsing, every image result gets eaten for me.

I've tested that with every other filter disabled and only "uBlock-Origin-dev-filter – Google+DuckDuckGo – Global" enabled, the results get eaten; with "uBlock-Origin-dev-filter – Google+DuckDuckGo – Global" disabled, the results do not get eaten.

[chuahou]

It appears the problem is that the filter list selects the .MjjYud element. On the normal search results page, each .MjjYud corresponds to a single search result, which makes it work perfectly. On the image search results page, however, all the search results are wrapped in a .MjjYud, meaning every result gets eaten if at least one result matches one of the blacklisted domains.

[JoshuaZacek]

I've found a fix that seems to work for now, although it allows image results to surface from the site you want to blacklist.

I replaced :upwards(.MjjYud) in the filter with :upwards(.Ww4FFb).

So for example, to block Google results from example.com, you'd use google.com###rso .MjjYud a[href*="example.com"]:upward(.Ww4FFb).