Best headless browser for AI agents. Lite, Fast, High-Compatibility. Built in Rust
Rust ai-agentsai-toolsbrowserbrowser-automationcloud-browserheadlessheadless-browserkitesurfplaywrightpuppeteerrustservoweb-crawlerweb-crawlingweb-scraperweb-scraping
mod attribute;
mod combinator;
mod host_adapter;
mod pseudo;
mod query;
mod syntax;
use url::Url;
use crate::{
dom::{
NodeId,
native::{DomHost, NativeDom},
},
parser::HtmlParser,
};
use super::{QueryEngine, error::SelectorErrorKind};
pub(crate) fn parse_document(html: &str) -> NativeDom {
HtmlParser::SCRIPTING_ENABLED.parse(
Url::parse("http://example.test/").expect("valid test url"),
html.to_owned(),
)
}
pub(crate) fn find_by_id(document: &NativeDom, id: &str) -> NodeId {
document
.nodes()
.iter()
.find_map(|node| {
node.kind()
.as_element()
.and_then(|element| (element.attribute("id") == Some(id)).then_some(node.id()))
})
.expect("expected element by id")
}
pub(crate) fn host_from_html(html: &str) -> DomHost {
DomHost::from_dom(parse_document(html))
}
pub(crate) fn host_find_by_id(host: &DomHost, id: &str) -> NodeId {
host.element_handle_by_id(id)
.expect("expected host element by id")
}
// ---------------------------------------------------------------------------
// Helper: run a batch of (selector, expected_ids) assertions on one document.
// ---------------------------------------------------------------------------
pub(crate) fn assert_query_ids(
engine: &QueryEngine,
document: &NativeDom,
cases: &[(&str, Vec<&str>)],
) {
for (selector, expected) in cases {
let results = engine
.query_selector_all(document, selector)
.unwrap_or_else(|e| {
panic!("selector {selector:?} should succeed but got error: {e:?}")
});
let got: Vec<&str> = results
.iter()
.map(|nid| {
document
.node(*nid)
.and_then(|n| n.kind().as_element())
.and_then(|el| el.attribute("id"))
.unwrap_or("")
})
.collect();
assert_eq!(got, *expected, "selector {selector:?}");
}
}