Can I use it with html5ever?

[LeMoussel]

Can I use sxd-xpath with html5ever to find node in HTML document?

For example:

let html_file =  Path::new("cocon1.html").to_path_buf();
let dom = parse_document(RcDom::default(), Default::default())
                .from_utf8()
                .from_file(&html_file)
                .unwrap();
let html_content = document.to_string();

[shepmaster]

No, but this could be interesting to investigate. I don't think it would be very easy though, as I believe (but have not verified) that the underlying access to the DOMs is not the most similar.

[mildred]

Does sxd-xpath depends directly on a sxd-document object or does it depends on a trait which is implemented using sxd-document but can be re-implemented for any parser with similar access? If not, is that complicated to do so?

[shepmaster]

Does sxd-xpath depends directly on a sxd-document object

Yes, basically everywhere uses a concrete sxd_xpath::nodeset::Node which itself refers to 7 concrete types from sxd-document. The code then matches on that in various places and uses methods on those types. The Node is also a public API.

I don't have an answer for "is it complicated". My gut feeling is that it would be, but sometimes the only way to find out is to just try.

[carlolefou]

@shepmaster I needed to evaluate xpaths against html for scraping so I wrote this helper library to bridge html5ever with sxd-document. This is pretty much my first real rust code so, if you have some time, please take a look and send any feedback my way. thanks! https://github.com/carlolefou/sxd-html