Merge eaf63ca80cdd13b6ef801262ab1b47c82dc0fd4a into b1c79b50d4253e7acb839a93450311c1fca12ac8

Cargo.lock

@@ -545,6 +545,29 @@ version = "0.2.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "43da5946c66ffcc7745f48db692ffbb10a83bfe0afd96235c5c2a4fb23994929"
 
+[[package]]
+name = "cssparser"
+version = "0.34.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b7c66d1cd8ed61bf80b38432613a7a2f09401ab8d0501110655f8b341484a3e3"
+dependencies = [
+ "cssparser-macros",
+ "dtoa-short",
+ "itoa",
+ "phf",
+ "smallvec",
+]
+
+[[package]]
+name = "cssparser-macros"
+version = "0.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "13b588ba4ac1a99f7f2964d24b3d896ddc6bf847ee3855dbd4366f058cfcd331"
+dependencies = [
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "csv"
 version = "1.3.1"
@@ -592,6 +615,17 @@ dependencies = [
  "syn",
 ]
 
+[[package]]
+name = "derive_more"
+version = "0.99.20"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6edb4b64a43d977b8e99788fe3a04d483834fba1215a7e02caa415b626497f7f"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "dirs"
 version = "6.0.0"
@@ -630,6 +664,21 @@ version = "1.2.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "75b325c5dbd37f80359721ad39aca5a29fb04c89279657cffdda8736d0c0b9d2"
 
+[[package]]
+name = "dtoa"
+version = "1.0.10"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d6add3b8cff394282be81f3fc1a0605db594ed69890078ca6e2cab1c408bcf04"
+
+[[package]]
+name = "dtoa-short"
+version = "0.3.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cd1511a7b6a56299bd043a9c167a6d2bfb37bf84a6dfceaba651168adfb43c87"
+dependencies = [
+ "dtoa",
+]
+
 [[package]]
 name = "ecow"
 version = "0.2.3"
@@ -639,6 +688,12 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "ego-tree"
+version = "0.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b2972feb8dffe7bc8c5463b1dacda1b0dfbed3710e50f977d965429692d74cd8"
+
 [[package]]
 name = "either"
 version = "1.13.0"
@@ -861,6 +916,16 @@ version = "2.0.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "e6d5a32815ae3f33302d95fdcb2ce17862f8c65363dcfd29360480ba1001fc9c"
 
+[[package]]
+name = "futf"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "df420e2e84819663797d1ec6544b13c5be84629e7bb00dc960d6917db2987843"
+dependencies = [
+ "mac",
+ "new_debug_unreachable",
+]
+
 [[package]]
 name = "fxhash"
 version = "0.2.1"
@@ -970,6 +1035,18 @@ version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "2304e00983f87ffb38b55b444b5e3b60a884b5d30c0fca7d82fe33449bbe55ea"
 
+[[package]]
+name = "html5ever"
+version = "0.29.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3b7410cae13cbc75623c98ac4cbfd1f0bedddf3227afc24f370cf0f50a44a11c"
+dependencies = [
+ "log",
+ "mac",
+ "markup5ever",
+ "match_token",
+]
+
 [[package]]
 name = "httpdate"
 version = "1.0.3"
@@ -1540,6 +1617,37 @@ dependencies = [
  "pkg-config",
 ]
 
+[[package]]
+name = "mac"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c41e0c4fef86961ac6d6f8a82609f55f31b05e4fce149ac5710e439df7619ba4"
+
+[[package]]
+name = "markup5ever"
+version = "0.14.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c7a7213d12e1864c0f002f52c2923d4556935a43dec5e71355c2760e0f6e7a18"
+dependencies = [
+ "log",
+ "phf",
+ "phf_codegen",
+ "string_cache",
+ "string_cache_codegen",
+ "tendril",
+]
+
+[[package]]
+name = "match_token"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "88a9689d8d44bf9964484516275f5cd4c9b59457a6940c1d5d0ecbb94510a36b"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
 [[package]]
 name = "memchr"
 version = "2.7.4"
@@ -1612,6 +1720,12 @@ dependencies = [
  "tempfile",
 ]
 
+[[package]]
+name = "new_debug_unreachable"
+version = "1.0.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "650eef8c711430f1a879fdd01d4745a7deea475becfb90269c06775983bbf086"
+
 [[package]]
 name = "nom"
 version = "7.1.3"
@@ -1869,6 +1983,16 @@ dependencies = [
  "phf_shared",
 ]
 
+[[package]]
+name = "phf_codegen"
+version = "0.11.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aef8048c789fa5e851558d709946d6d79a8ff88c0440c587967f8e94bfb1216a"
+dependencies = [
+ "phf_generator",
+ "phf_shared",
+]
+
 [[package]]
 name = "phf_generator"
 version = "0.11.3"
@@ -1981,6 +2105,12 @@ dependencies = [
  "zerocopy",
 ]
 
+[[package]]
+name = "precomputed-hash"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "925383efa346730478fb4838dbe9137d2a47675ad789c546d150a6e1dd4ab31c"
+
 [[package]]
 name = "proc-macro2"
 version = "1.0.93"
@@ -2282,6 +2412,21 @@ version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 
+[[package]]
+name = "scraper"
+version = "0.23.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "527e65d9d888567588db4c12da1087598d0f6f8b346cc2c5abc91f05fc2dffe2"
+dependencies = [
+ "cssparser",
+ "ego-tree",
+ "getopts",
+ "html5ever",
+ "precomputed-hash",
+ "selectors",
+ "tendril",
+]
+
 [[package]]
 name = "security-framework"
 version = "2.11.1"
@@ -2305,6 +2450,25 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "selectors"
+version = "0.26.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fd568a4c9bb598e291a08244a5c1f5a8a6650bee243b5b0f8dbb3d9cc1d87fe8"
+dependencies = [
+ "bitflags 2.8.0",
+ "cssparser",
+ "derive_more",
+ "fxhash",
+ "log",
+ "new_debug_unreachable",
+ "phf",
+ "phf_codegen",
+ "precomputed-hash",
+ "servo_arc",
+ "smallvec",
+]
+
 [[package]]
 name = "self-replace"
 version = "1.5.0"
@@ -2388,6 +2552,15 @@ dependencies = [
  "unsafe-libyaml",
 ]
 
+[[package]]
+name = "servo_arc"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ae65c4249478a2647db249fb43e23cec56a2c8974a427e7bd8cb5a1d0964921a"
+dependencies = [
+ "stable_deref_trait",
+]
+
 [[package]]
 name = "shell-escape"
 version = "0.1.5"
@@ -2499,6 +2672,31 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "string_cache"
+version = "0.8.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bf776ba3fa74f83bf4b63c3dcbbf82173db2632ed8452cb2d891d33f459de70f"
+dependencies = [
+ "new_debug_unreachable",
+ "parking_lot",
+ "phf_shared",
+ "precomputed-hash",
+ "serde",
+]
+
+[[package]]
+name = "string_cache_codegen"
+version = "0.5.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c711928715f1fe0fe509c53b43e993a9a557babc2d0a3567d0a3006f1ac931a0"
+dependencies = [
+ "phf_generator",
+ "phf_shared",
+ "proc-macro2",
+ "quote",
+]
+
 [[package]]
 name = "strsim"
 version = "0.11.1"
@@ -2621,6 +2819,17 @@ dependencies = [
  "windows-sys 0.59.0",
 ]
 
+[[package]]
+name = "tendril"
+version = "0.4.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d24a120c5fc464a3458240ee02c299ebcb9d67b5249c8848b09d639dca8d7bb0"
+dependencies = [
+ "futf",
+ "mac",
+ "utf-8",
+]
+
 [[package]]
 name = "termcolor"
 version = "1.4.1"
@@ -3071,6 +3280,7 @@ dependencies = [
  "comemo",
  "csv",
  "ecow",
+ "ego-tree",
  "flate2",
  "fontdb",
  "glidesort",
@@ -3094,6 +3304,7 @@ dependencies = [
  "roxmltree",
  "rust_decimal",
  "rustybuzz",
+ "scraper",
  "serde",
  "serde_json",
  "serde_yaml 0.9.34+deprecated",
@@ -3426,6 +3637,12 @@ dependencies = [
  "xmlwriter",
 ]
 
+[[package]]
+name = "utf-8"
+version = "0.7.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "09cc8ee72d2a9becf2f2febe0205bbed8fc6615b7cb429ad062dc7b7ddd036a9"
+
 [[package]]
 name = "utf16_iter"
 version = "1.0.5"

Cargo.toml

@@ -54,6 +54,7 @@ csv = "1"
 ctrlc = "3.4.1"
 dirs = "6"
 ecow = { version = "0.2", features = ["serde"] }
+ego-tree = "0.10"
 env_proxy = "0.4"
 fastrand = "2.3"
 flate2 = "1"
@@ -104,6 +105,7 @@ roxmltree = "0.20"
 rust_decimal = { version = "1.36.0", default-features = false, features = ["maths"] }
 rustybuzz = "0.20"
 same-file = "1"
+scraper = "0.23.1"
 self-replace = "1.3.7"
 semver = "1"
 serde = { version = "1.0.184", features = ["derive"] }

crates/typst-library/Cargo.toml

@@ -27,6 +27,7 @@ codex = { workspace = true }
 comemo = { workspace = true }
 csv = { workspace = true }
 ecow = { workspace = true }
+ego-tree = { workspace = true }
 flate2 = { workspace = true }
 fontdb = { workspace = true }
 glidesort = { workspace = true }
@@ -50,6 +51,7 @@ regex-syntax = { workspace = true }
 roxmltree = { workspace = true }
 rust_decimal = { workspace = true }
 rustybuzz = { workspace = true }
+scraper = { workspace = true }
 serde = { workspace = true }
 serde_json = { workspace = true }
 serde_yaml = { workspace = true }

crates/typst-library/src/loading/html.rs

@@ -0,0 +1,98 @@
+use ecow::eco_format;
+use ego_tree::NodeRef;
+use scraper::Node;
+use typst_syntax::Spanned;
+
+use crate::diag::{At, FileError, SourceDiagnostic, SourceResult};
+use crate::engine::Engine;
+use crate::foundations::{dict, func, Array, Dict, IntoValue, Value};
+use crate::loading::{DataSource, Load};
+
+/// Reads structured data from an HTML file.
+///
+/// The HTML file is parsed into an array of dictionaries and strings. It is compatible with
+/// the XML format, parsed by the [`xml`]($xml) function.
+#[func(title = "HTML")]
+pub fn html_decode(
+    engine: &mut Engine,
+    /// A [path]($syntax/#paths) to an HTML file or raw HTML bytes.
+    source: Spanned<DataSource>,
+) -> SourceResult<Value> {
+    let data = source.load(engine.world)?;
+    let text = data.as_str().map_err(FileError::from).at(source.span)?;
+    let document = scraper::Html::parse_document(text);
+
+    if !document.errors.is_empty() {
+        let errors = document.errors.iter();
+        return Err(errors
+            .map(|msg| {
+                SourceDiagnostic::error(
+                    source.span,
+                    eco_format!("failed to parse HTML ({msg})"),
+                )
+            })
+            .collect());
+    }
+
+    Ok(convert_html(document.tree.root()))
+}
+
+/// Convert an HTML node to a Typst value.
+fn convert_html(node_ref: NodeRef<Node>) -> Value {
+    // `prefix` and `name` are part of the tag name. For example,
+    // in the following HTML, `html5` is the prefix and `div` is the name:
+    // ```
+    // <html5:div class="example" />
+    // ```
+    let (prefix, name, attrs) = match node_ref.value() {
+        Node::Text(text) => return (*text).into_value(),
+        Node::Document => return Value::Array(convert_html_children(node_ref)),
+        // todo: the namespace is ignored
+        Node::Element(element) => {
+            (element.name.prefix.as_ref(), &*element.name.local, Some(element.attrs()))
+        }
+        Node::Fragment => (None, "fragment", None),
+        // todo: doc type and processing instruction are ignored
+        // https://en.wikipedia.org/wiki/Processing_Instruction
+        Node::Doctype(..) | Node::ProcessingInstruction(..) => return Value::None,
+        Node::Comment(comment) => {
+            return Value::Dict(dict! {
+                "tag" => "",
+                "attrs" => dict! {},
+                "children" => [(*comment).into_value()].into_iter().collect::<Array>(),
+            });
+        }
+    };
+
+    let children = convert_html_children(node_ref);
+
+    let attrs: Dict = attrs
+        .into_iter()
+        .flatten()
+        .map(|(name, value)| (name.into(), value.into_value()))
+        .collect();
+
+    let mut converted = dict! {
+        "tag" => name.into_value(),
+        "attrs" => attrs,
+        "children" => children,
+    };
+
+    // In most cases, the prefix is not set, so we only add it if it exists.
+    if let Some(prefix) = prefix {
+        converted.insert("prefix".into(), (*prefix).into_value());
+    }
+
+    Value::Dict(converted)
+}
+
+/// Convert children an HTML node to a Typst value.
+fn convert_html_children(node_ref: NodeRef<Node>) -> Array {
+    node_ref
+        .children()
+        .filter(|v| {
+            !matches!(v.value(), Node::Doctype(..) | Node::ProcessingInstruction(..))
+        })
+        .map(convert_html)
+        .collect()
+}

crates/typst-library/src/loading/mod.rs

@@ -4,6 +4,8 @@
 mod cbor_;
 #[path = "csv.rs"]
 mod csv_;
+#[path = "html.rs"]
+mod html_;
 #[path = "json.rs"]
 mod json_;
 #[path = "read.rs"]
@@ -21,6 +23,7 @@ use typst_syntax::{FileId, Spanned};
 
 pub use self::cbor_::*;
 pub use self::csv_::*;
+pub use self::html_::*;
 pub use self::json_::*;
 pub use self::read_::*;
 pub use self::toml_::*;
@@ -37,6 +40,7 @@ pub(super) fn define(global: &mut Scope) {
     global.start_category(crate::Category::DataLoading);
     global.define_func::<read>();
     global.define_func::<csv>();
+    global.define_func::<html_decode>();
     global.define_func::<json>();
     global.define_func::<toml>();
     global.define_func::<yaml>();

tests/suite/loading/html.typ

@@ -0,0 +1,46 @@
+--- html ---
+// Test reading XML data.
+#let data = html-decode("/assets/text/example.html")
+#test(data, ((
+  tag: "html",
+  attrs: (:),
+  children: (
+    (
+      tag: "head",
+      attrs: (:),
+      children: (
+        "\n    ",
+        (
+          tag: "meta",
+          attrs: (charset: "UTF-8"),
+          children: (),
+        ),
+        "\n    ",
+        (
+          tag: "title",
+          attrs: (:),
+          children: ("Example document",),
+        ),
+        "\n  ",
+      ),
+    ),
+    "\n  ",
+    (
+      tag: "body",
+      attrs: (:),
+      children: (
+        "\n    ",
+        (
+          tag: "h1",
+          attrs: (:),
+          children: ("Hello, world!",),
+        ),
+        "\n  \n\n",
+      ),
+    ),
+  ),
+),))
+
+--- html-invalid ---
+// Error: 14-38 failed to parse HTML (Unexpected token)
+#html-decode("/assets/text/hello.txt")