quick-xml 0.1.1

High performance xml pull reader


High performance xml pull reader for simple enough xmls.

Inspired by xml-rs.



extern crate quick_xml;


use quick_xml::{XmlReader, Event};

let xml = r#"<tag1 att1 = \"test\">
                <tag2><!--Test comment-->Test</b>
                    Test 2
let reader = XmlReader::from_str(xml).trim_text(true);
let mut count = 0;
let mut txt = Vec::new();
for r in reader {
    match r {
        Ok(Event::Start(ref e)) => {
            match e.as_bytes() {
                b"tag1" => println!("attributes values: {:?}", 
                                 e.attributes().map(|a| a.unwrap().1).collect::<Vec<_>>()),
                b"tag2" => count += 1,
                _ => (),
        Ok(Event::Text(e)) => txt.push(e.into_string()),
        Err(e) => println!("{:?}", e),
        _ => (),

Current state

quick-xml has been written to be fast.

On my first tests (200mb+ xmls) it performs much better (minimum 10x) than xml-rs.

While this is a still WIP and only basic xml specifications are implemented, you can already use it for simple enough xmls (no namespaces, no exotic characters etc ...).

This is particularly true when the xml is generated by a known source (e.g. OpenStreetMap).


There are many xml specifications not implemented yet:

  • namespaces
  • non-utf8
  • parse xml prologue
  • XQuries ?
  • more checks
  • ... and many other things I don't even know!


Any PR is welcomed!

I am not an expert in xml specifications, I simply have to work with big xmls. As a result, I may not implement some basic functionalities, simply because I don't know/need them.