Feature request: Allow matching URL patterns, not just domains

[raquo]

Not sure how hard this is to implement, but certainly would be nice... Some use cases:

• Highlight or filter certain subreddits or github projects
• Filter results from parts of websites that have robots.txt configured incorrectly. For example it's hard to find support EasyEda articles because Google search results are spammed with easyeda.com/search/* pages (example) (but the answer is likely on other, non-search easyeda.com URLs, so blocking the entire domain is not an option)

[seamusdemora]

@raquo 👍 Great idea, methinks. Imagine regex on search engine filtering!