Add support for :has from the CSS4 specification

[parthjoshi2007]

I was trying to determine the status of support for CSS4 selectors in scrapy but could not find any information. Is there any plan to add such support? I think the CSS4 working draft has some very useful extensions that other libraries like soupsieve, the CSS selector library used by Beautiful Soup, for example, already support.

In particular I would like to make the case for :has and possibly also :is. In the course of my experiments with scraping websites (for which scrapy is truly a lifesaver), I have come across several websites where I need to, just as an example, get the p tag after the p tag containing an image. See for example getting the job titles out of this webpage. Using :has this would be trivial with p:has(strong) + p but it is not possible currently with parsel.

And before people jump on me saying "Hey! Just use XPath selectors", I know that I could use XPath. But XPath selectors have a much steeper learning curve than CSS selectors and I can train people to use CSS much more easily than XPath.

[Gallaecio]

This is a duplicate of scrapy/cssselect#66 (parsel implements CSS support through cssselect).