ruma_html/lib.rs
1#![doc(html_favicon_url = "https://ruma.dev/favicon.ico")]
2#![doc(html_logo_url = "https://ruma.dev/images/logo.png")]
3//! Opinionated HTML parsing and manipulating library.
4//!
5//! Like the rest of the Ruma crates, this crate is primarily meant to be used for
6//! the Matrix protocol. It should be able to be used to interact with any HTML
7//! document but will offer APIs focused on specificities of HTML in the Matrix
8//! specification..
9//!
10//! # Features
11//!
12//! * `matrix` - Allow to convert HTML elements data into enums with variants for elements and
13//! attributes [suggested by the Matrix Specification][spec].
14//!
15//! [spec]: https://spec.matrix.org/latest/client-server-api/#mroommessage-msgtypes
16
17#![warn(missing_docs)]
18#![cfg_attr(docsrs, feature(doc_auto_cfg))]
19
20pub use html5ever::{tendril::StrTendril, Attribute, LocalName, Namespace, Prefix, QualName};
21
22mod helpers;
23mod html;
24mod sanitizer_config;
25
26pub use self::{helpers::*, html::*, sanitizer_config::*};
27
28/// What [HTML elements and attributes] should be kept by the sanitizer.
29///
30/// [HTML elements and attributes]: https://spec.matrix.org/latest/client-server-api/#mroommessage-msgtypes
31#[derive(Clone, Copy, Debug, PartialEq, Eq)]
32#[allow(clippy::exhaustive_enums)]
33pub enum HtmlSanitizerMode {
34 /// Keep only the elements and attributes suggested in the Matrix specification.
35 ///
36 /// In addition to filtering elements and attributes listed in the Matrix specification, it
37 /// also removes elements that are nested more than 100 levels deep.
38 ///
39 /// Deprecated elements and attributes are also replaced when applicable.
40 Strict,
41
42 /// Like `Strict` mode, with additional elements and attributes that are not yet included in
43 /// the spec, but are reasonable to keep.
44 ///
45 /// Differences with `Strict` mode:
46 ///
47 /// * The `matrix` scheme is allowed in links.
48 Compat,
49}