roxmltree

Represent an XML 1.0 document as a read-only tree.

Why read-only?

Because in some cases all you need is to retrieve some data from the XML document. And for such cases, we can make a lot of optimizations.

As for roxmltree, it's fast not only because it's read-only, but also because it uses xmlparser, which is times faster then xml-rs. See Performance section for details.

Parsing behavior

Sadly, XML can be parsed in many different ways. The roxmltree is trying to mimic the Python's lxml behavior.

Unlike the lxml, roxmltree do support comments outside the root element.

Fo more details see docs/parsing.md.

Alternatives

* Rust besed for now

Legend:

• ✔ - supported
• ⚠ - parsing error
• ~ - partial
• nothing - not supported

Notes:

1. No default namespace propagation.
2. Nested/indirect entity references are not supported yet.
3. roxmltree keeps all nodes and attributes position in the original document, so you can easily retrieve it if you need it. See examples/print_pos.rs for details.
4. In the string_cache crate.
5. Binary size overhead according to the cargo-bloat.

Performance

test large_roxmltree     ... bench:   8,807,741 ns/iter (+/- 70,532)
test large_sdx_document  ... bench:   9,777,811 ns/iter (+/- 242,912)
test large_xmltree       ... bench:  31,041,407 ns/iter (+/- 27,171)
test large_treexml       ... bench:  32,048,129 ns/iter (+/- 29,860)
test large_elementtree   ... bench:  32,073,296 ns/iter (+/- 68,433)

test medium_roxmltree    ... bench:   1,735,369 ns/iter (+/- 3,218)
test medium_sdx_document ... bench:   3,569,814 ns/iter (+/- 10,518)
test medium_treexml      ... bench:  11,163,737 ns/iter (+/- 26,084)
test medium_xmltree      ... bench:  11,267,754 ns/iter (+/- 70,971)
test medium_elementtree  ... bench:  11,629,513 ns/iter (+/- 27,055)

roxmltree uses xmlparser internally, while sdx-document uses it's own one and xmltree, elementtree and treexml are using the xml-rs crate. Here is a comparison between xmlparser and xml-rs:

test large_xmlparser     ... bench:   2,019,245 ns/iter (+/- 693)
test large_xmlrs         ... bench:  29,086,480 ns/iter (+/- 22,741)

test medium_xmlparser    ... bench:     434,140 ns/iter (+/- 231)
test medium_xmlrs        ... bench:  10,391,411 ns/iter (+/- 24,738)

Note: tree crates may use different xml-rs crate versions.

You can try it yourself using cargo bench --features benchmark

Safety

• The library must not panic. Any panic considered as a critical bug and should be reported.
• The library forbids the unsafe code.

Non-goals

• Complete XML support.
• Tree modification and writing.
• XPath/XQuery.

API

The library uses Rust's idiomatic API based on iterators. In case you are more familiar with the browsers/JS DOM API - you can check out the tests/dom-api.rs to see how it can be converted into a Rust one.

Dependency

Rust >= 1.18

License

Licensed under either of

• Apache License v2.0
• MIT license

at your option.

Contribution

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.