feat(spider): add sitewide auditing

Cargo.toml

@@ -12,6 +12,6 @@ members = [
 smallvec = "1"
 selectors = "0.22.0"
 cssparser = "0.27.0"
-html5ever = "0.27"
+fast_html5ever = "0.26.1"
 ego-tree = "0.6.2"
 lazy_static = "1.4"

README.md

@@ -68,6 +68,7 @@ async fn main() {
 1. Shortest path CSS selectors for elements.
 1. i18n support for multiple languages.
 1. Re-creating layout tree to get element position coordinates.
+1. Crawling full websites lightning-fast using [spider](https://github.com/spider-rs/spider).
 
 ## [Benchmarks](./benches/)
 
@@ -78,6 +79,8 @@ audit-speed/core/audit: medium html (4k iterations)
 time: [824.07 µs 830.30 µs 839.37 µs]
 audit-speed/core/audit: large html (4k iterations)
 time: [1.1206 ms 1.1260 ms 1.1321 ms]
+audit-speed/core/audit: spider audit html (4k iterations)
+time: [263.33 ms 266.55 ms 269.93 ms]
 ```
 
 ## Examples
@@ -90,6 +93,7 @@ time: [1.1206 ms 1.1260 ms 1.1321 ms]
 1. [tokio](https://docs.rs/tokio/latest/tokio/): Enable tokio async runtime handling. Recommended for high freq server usage.
 1. [rayon](https://docs.rs/rayon/latest/rayon/): Parallelism with rayon. (Expensive test future handling)
 1. [rayon_wasm](https://lib.rs/crates/rayon-wasm): Enable the wasm runtime for rayon.
+1. [spider](https://docs.rs/spider-rs/latest/spider/): Crawl entire websites using spider. Full website audits of 100-1,000 pages within milliseconds.
 
 ### Contributing
 

accessibility-rs/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "accessibility-rs"
-version = "0.0.59"
+version = "0.0.60"
 authors = ["The A11yWatch Project Developers", "Jeff Mendez <[email protected]>"]
 edition = "2021"
 license = "MIT OR Apache-2.0"
@@ -12,8 +12,8 @@ include = ["/src", "../LICENSE_MIT", "../LICENSE_APACHE", "../README.md", "local
 
 [dependencies]
 lazy_static = { workspace = true }
-accessibility-scraper = { version = "0.0.11", features = ["main"], default-features = false, path = "../accessibility-scraper" }
-accessibility-tree = { version = "0.0.11", path = "../accessibility-tree/victor" }
+accessibility-scraper = { version = "0.0.12", features = ["main"], default-features = false, path = "../accessibility-scraper" }
+accessibility-tree = { version = "0.0.12", path = "../accessibility-tree/victor" }
 getrandom = { version = "0.2", features = ["js"] }
 taffy = { version = "0.4.0" }
 serde = { version = "1.0", features = ["derive"] }
@@ -31,12 +31,32 @@ rayon = { version = "1.10.0", optional = true }
 crossbeam-channel = { version = "0.5.12", optional = true }
 tokio = { version = "1.36.0", features = ["macros"], optional = true }
 tokio-stream = { version = "0.1.15", optional = true }
+spider = { version = "1.89.3", optional = true }
 
 [features]
 default = []
 rayon = ["dep:rayon", "dep:crossbeam-channel"]
 rayon_wasm = ["rayon/web_spin_lock"]
 tokio = ["dep:tokio", "dep:tokio-stream", "accessibility-scraper/tokio", "accessibility-tree/tokio"]
+spider = ["tokio", "dep:spider", "accessibility-scraper/spider"]
+spider_ua_generator = ["tokio", "spider/ua_generator"]
+spider_chrome = ["spider", "spider/chrome"]
+spider_chrome_cpu = ["spider", "spider/chrome_cpu"]
+spider_screenshot = ["spider", "spider/chrome_screenshot"]
+spider_smart = ["spider", "spider/smart"]
+spider_headed = ["spider", "spider/chrome_headed"]
+spider_store_page = ["spider", "spider/chrome_store_page"]
+spider_stealth = ["spider", "spider/chrome_stealth"]
+spider_intercept = ["spider", "spider/chrome_intercept"]
+spider_headless_new = ["spider", "spider/chrome_headless_new"]
+spider_real_browser = ["spider", "spider/real_browser"]
+spider_openai = ["spider", "spider/openai"]
+spider_openai_slim_fit = ["spider", "spider/openai_slim_fit"]
+spider_budget = ["spider", "spider/budget"]
+spider_cache = ["spider", "spider/cache"]
+spider_cache_mem = ["spider", "spider/cache_mem"]
+spider_sitemap = ["tokio", "spider/sitemap"]
+spider_control = ["tokio", "spider/control"]
 
 [dev-dependencies]
 maud = "0.25.0"

accessibility-rs/src/engine/issue.rs

@@ -1,7 +1,7 @@
 use serde::{Deserialize, Serialize};
 
 /// clip bounding box
-#[derive(Default, Debug, Serialize, Deserialize)]
+#[derive(Default, Debug, Clone, Serialize, Deserialize)]
 pub struct Clip {
     /// the x coords
     pub x: u32,
@@ -14,7 +14,7 @@ pub struct Clip {
 }
 
 /// Extra help information for the issue
-#[derive(Default, Debug, Serialize, Deserialize)]
+#[derive(Default, Debug, Clone, Serialize, Deserialize)]
 pub struct RunnerExtras {
     /// the url to get more information on the issue
     pub help_url: &'static str,
@@ -25,7 +25,7 @@ pub struct RunnerExtras {
 }
 
 /// Details of the problem
-#[derive(Default, Debug, Serialize, Deserialize)]
+#[derive(Default, Debug, Clone, Serialize, Deserialize)]
 pub struct Issue {
     /// the context of the issue or raw html
     pub context: String,

accessibility-rs/src/lib.rs

@@ -37,6 +37,15 @@
 //!     let audit = audit(config).await;
 //!     println!("{:?}", audit);
 //! }
+//!
+//! #[cfg(feature = "spider")]
+//! #[tokio::main]
+//! async fn main() {
+//!     let mut config = AuditConfig::default();
+//!     config.url = "https://example.com";
+//!     let audit = audit(config).await;
+//!     println!("{:?}", audit);
+//! }
 //! ```
 //!
 
@@ -45,6 +54,9 @@ extern crate lazy_static;
 #[macro_use]
 extern crate rust_i18n;
 
+#[cfg(feature = "spider")]
+pub use spider;
+
 /// the main engine for accessibility auditing.
 pub mod engine;
 /// locales for translations.
@@ -78,6 +90,9 @@ pub struct AuditConfig {
     pub locale: String,
     /// the guideline spec
     pub conformance: Conformance,
+    /// crawl and perform audits on the entire website
+    #[cfg(feature = "spider")]
+    pub url: String,
 }
 
 /// configs for the audit
@@ -94,6 +109,9 @@ pub struct AuditConfig<'a> {
     pub locale: &'a str,
     /// the guideline spec
     pub conformance: Conformance,
+    /// crawl and perform audits on the entire website
+    #[cfg(feature = "spider")]
+    pub url: &'a str,
 }
 
 #[cfg(not(feature = "tokio"))]
@@ -116,6 +134,29 @@ impl<'a> AuditConfig<'a> {
             ..Default::default()
         }
     }
+
+    /// a new audit configuration crawling a website. This does nothing without the 'spider' flag enabled.
+    #[cfg(feature = "spider")]
+    pub fn new_website(url: &'a str, css: &'a str, bounding_box: bool, locale: &'a str) -> Self {
+        AuditConfig {
+            url: url.into(),
+            css: css.into(),
+            bounding_box,
+            locale: locale.into(),
+            ..Default::default()
+        }
+    }
+
+    /// a new audit configuration crawling a website. This does nothing without the 'spider' flag enabled.
+    #[cfg(not(feature = "spider"))]
+    pub fn new_website(
+        _url: &'a str,
+        _css: &'a str,
+        _bounding_box: bool,
+        _locale: &'a str,
+    ) -> Self {
+        AuditConfig::default()
+    }
 }
 
 #[cfg(feature = "tokio")]
@@ -138,16 +179,78 @@ impl AuditConfig {
             ..Default::default()
         }
     }
+
+    /// a new audit configuration crawling a website. This does nothing without the 'spider' flag enabled.
+    #[cfg(feature = "spider")]
+    pub fn new_website(url: &str, css: &str, bounding_box: bool, locale: &str) -> Self {
+        AuditConfig {
+            url: url.into(),
+            css: css.into(),
+            bounding_box,
+            locale: locale.into(),
+            ..Default::default()
+        }
+    }
+    /// a new audit configuration crawling a website. This does nothing without the 'spider' flag enabled.
+    #[cfg(not(feature = "spider"))]
+    pub fn new_website(_url: &str, _css: &str, _bounding_box: bool, _locale: &str) -> Self {
+        AuditConfig::default()
+    }
 }
 
 /// audit a web page passing the html and css rules.
-#[cfg(feature = "tokio")]
+#[cfg(all(feature = "tokio", not(feature = "spider")))]
 pub async fn audit(config: AuditConfig) -> Vec<Issue> {
     let document = accessibility_scraper::Html::parse_document(&config.html).await;
     let auditor = Auditor::new(&document, &config.css, config.bounding_box, &config.locale);
     engine::audit::wcag::WCAGAAA::audit(auditor).await
 }
 
+#[cfg(feature = "spider")]
+#[derive(Debug, Clone)]
+/// The accessibility audit results either a single page or entire website.
+pub enum AuditResults {
+    /// Crawl results from multiple websites
+    Page(spider::hashbrown::HashMap<String, Vec<Issue>>),
+    /// Raw html markup results
+    Html(Vec<Issue>),
+}
+
+/// audit a web page passing the html and css rules.
+#[cfg(feature = "spider")]
+pub async fn audit(config: AuditConfig) -> AuditResults {
+    if !config.url.is_empty() {
+        use spider::website::Website;
+        let mut website: Website = Website::new(&config.url);
+        let mut rx2: tokio::sync::broadcast::Receiver<spider::page::Page> =
+            website.subscribe(16).unwrap();
+        let bounding_box = config.bounding_box;
+        let locale = config.locale;
+
+        let audits = tokio::spawn(async move {
+            let mut issues: spider::hashbrown::HashMap<String, Vec<Issue>> =
+                spider::hashbrown::HashMap::new();
+
+            while let Ok(res) = rx2.recv().await {
+                let document = accessibility_scraper::Html::parse_document(&res.get_html()).await;
+                let auditor = Auditor::new(&document, &"", bounding_box, &locale);
+                let issue = engine::audit::wcag::WCAGAAA::audit(auditor).await;
+                issues.insert(res.get_url().into(), issue);
+            }
+
+            issues
+        });
+
+        website.crawl().await;
+        website.unsubscribe();
+        AuditResults::Page(audits.await.unwrap_or_default())
+    } else {
+        let document = accessibility_scraper::Html::parse_document(&config.html).await;
+        let auditor = Auditor::new(&document, &config.css, config.bounding_box, &config.locale);
+        AuditResults::Html(engine::audit::wcag::WCAGAAA::audit(auditor).await)
+    }
+}
+
 /// audit a web page passing the html and css rules.
 #[cfg(not(feature = "tokio"))]
 pub fn audit(config: AuditConfig) -> Vec<Issue> {

accessibility-rs/tests/integration_test.rs

@@ -89,3 +89,12 @@ async fn _audit_xlarge() {
     .await;
     println!("{:?}", report)
 }
+
+#[cfg(feature = "spider")]
+#[spider::tokio::test]
+async fn _audit_website() {
+    let mut audit_config = AuditConfig::default();
+    audit_config.url = "https://choosealicense.com";
+    let report = accessibility_rs::audit(audit_config).await;
+    println!("{:?}", report)
+}

accessibility-rs/tests/mocks/mock.rs

@@ -1,3 +1,5 @@
+#![allow(dead_code)]
+
 /// mock website data from drake.com
 pub static MOCK_WEBSITE_HTML: &'static str = r###"
 <html class="no-js" lang="en"><!--<![endif]--><head><style>.hs-cta-wrapper p, .hs-cta-wrapper div { margin: 0; padding: 0; } a#cta_button_5479613_e9c3ff65-5303-42df-b466-cd64a84bfc67 {

accessibility-scraper/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "accessibility-scraper"
-version = "0.0.11"
+version = "0.0.12"
 edition = "2021"
 description = "HTML parsing and querying with CSS selectors with CSS binding styles to elements."
 keywords = ["html", "css", "selector", "scraping"]
@@ -14,7 +14,7 @@ readme = "README.md"
 [dependencies]
 cssparser = { workspace = true }
 ego-tree = { workspace = true }
-html5ever = { workspace = true }
+fast_html5ever = { workspace = true }
 selectors = { workspace = true }
 smallvec = { workspace = true }
 tendril = "0.4.3"
@@ -34,4 +34,5 @@ deterministic = ["indexmap"]
 main = ["getopts"]
 atomic = []
 errors = []
-tokio = ["dep:tokio", "dep:tokio-stream"]
+tokio = ["dep:tokio", "dep:tokio-stream"]
+spider = []

accessibility-scraper/README.md

@@ -132,7 +132,7 @@ assert_eq!(vec!["Hello, ", "world!"], text);
 ### Manipulating the DOM
 
 ```rust
-use html5ever::tree_builder::TreeSink;
+use fast_html5ever::tree_builder::TreeSink;
 use accessibility-scraper::{Html, Selector};
 
 let html = "<html><body>hello<p class=\"hello\">REMOVE ME</p></body></html>";

accessibility-scraper/src/element_ref/element.rs

@@ -1,6 +1,6 @@
 use super::ElementRef;
 use crate::selector::{CssLocalName, CssString, NonTSPseudoClass, PseudoElement, Simple};
-use html5ever::{LocalName, Namespace};
+use fast_html5ever::{LocalName, Namespace};
 use selectors::{
     attr::{AttrSelectorOperation, CaseSensitivity, NamespaceConstraint},
     matching, Element, OpaqueElement,

accessibility-scraper/src/element_ref/mod.rs

@@ -4,7 +4,7 @@ use std::ops::Deref;
 
 use ego_tree::iter::{Edge, Traverse};
 use ego_tree::NodeRef;
-use html5ever::serialize::{serialize, SerializeOpts, TraversalScope};
+use fast_html5ever::serialize::{serialize, SerializeOpts, TraversalScope};
 
 use crate::node::Element;
 use crate::{Node, Selector};

accessibility-scraper/src/element_ref/serializable.rs

@@ -1,6 +1,6 @@
 use std::io::Error;
 
-use html5ever::serialize::{Serialize, Serializer, TraversalScope};
+use fast_html5ever::serialize::{Serialize, Serializer, TraversalScope};
 
 use crate::ElementRef;