scraper-rust 0.4.0__tar.gz → 0.4.2__tar.gz

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/.github/workflows/release.yml

@@ -10,12 +10,24 @@ permissions:
 
 jobs:
   build-wheels:
-    name: Build wheels on ${{ matrix.os }}
-    runs-on: ${{ matrix.os }}
+    name: Build wheels on ${{ matrix.name }}
+    runs-on: ${{ matrix.runner }}
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-latest, windows-latest, macos-latest]
+        include:
+          - name: linux-x86_64
+            runner: ubuntu-latest
+            platform: linux
+          - name: linux-arm64
+            runner: ubuntu-24.04-arm
+            platform: linux
+          - name: windows
+            runner: windows-latest
+            platform: windows
+          - name: macos
+            runner: macos-latest
+            platform: macos
 
     steps:
       - name: Checkout repository
@@ -27,7 +39,7 @@ jobs:
           python-version: '3.10'
 
       - name: Build wheels (Linux)
-        if: matrix.os == 'ubuntu-latest'
+        if: matrix.platform == 'linux'
         uses: PyO3/maturin-action@v1
         with:
           command: build
@@ -35,7 +47,7 @@ jobs:
           manylinux: auto
 
       - name: Build wheels (Windows/macOS)
-        if: matrix.os == 'windows-latest' || matrix.os == 'macos-latest'
+        if: matrix.platform == 'windows' || matrix.platform == 'macos'
         uses: PyO3/maturin-action@v1
         with:
           command: build
@@ -44,7 +56,7 @@ jobs:
       - name: Upload wheels as artifacts
         uses: actions/upload-artifact@v7
         with:
-          name: wheels-${{ matrix.os }}
+          name: wheels-${{ matrix.name }}
           path: dist/*.whl
 
       - name: Upload wheels to release
@@ -56,13 +68,25 @@ jobs:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
 
   build-free-threaded-wheels:
-    name: Build free-threaded wheels on ${{ matrix.os }} (Python ${{ matrix.python-version }})
-    runs-on: ${{ matrix.os }}
+    name: Build free-threaded wheels on ${{ matrix.name }} (Python ${{ matrix.python-version }})
+    runs-on: ${{ matrix.runner }}
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-latest, windows-latest, macos-latest]
         python-version: ['3.13t', '3.14t']
+        include:
+          - name: linux-x86_64
+            runner: ubuntu-latest
+            platform: linux
+          - name: linux-arm64
+            runner: ubuntu-24.04-arm
+            platform: linux
+          - name: windows
+            runner: windows-latest
+            platform: windows
+          - name: macos
+            runner: macos-latest
+            platform: macos
 
     steps:
       - name: Checkout repository
@@ -75,20 +99,20 @@ jobs:
           allow-prereleases: true
 
       - name: Install zig (for Linux manylinux builds)
-        if: matrix.os == 'ubuntu-latest'
+        if: matrix.platform == 'linux'
         uses: goto-bus-stop/setup-zig@v2
         with:
           version: 0.11.0
 
       - name: Build free-threaded wheels (Linux - using zig for manylinux)
-        if: matrix.os == 'ubuntu-latest'
+        if: matrix.platform == 'linux'
         uses: PyO3/maturin-action@v1
         with:
           command: build
           args: --release --strip --out dist --zig
 
       - name: Build free-threaded wheels (Windows/macOS)
-        if: matrix.os == 'windows-latest' || matrix.os == 'macos-latest'
+        if: matrix.platform == 'windows' || matrix.platform == 'macos'
         uses: PyO3/maturin-action@v1
         with:
           command: build
@@ -97,7 +121,7 @@ jobs:
       - name: Upload free-threaded wheels as artifacts
         uses: actions/upload-artifact@v7
         with:
-          name: wheels-free-threaded-${{ matrix.os }}-${{ matrix.python-version }}
+          name: wheels-free-threaded-${{ matrix.name }}-${{ matrix.python-version }}
           path: dist/*.whl
 
       - name: Upload free-threaded wheels to release

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/Cargo.lock

@@ -545,7 +545,17 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "1054432bae2f14e0061e33d23402fbaa67a921d319d56adc6bcf887ddad1cbc2"
 dependencies = [
  "log",
- "markup5ever",
+ "markup5ever 0.38.0",
+]
+
+[[package]]
+name = "html5ever"
+version = "0.39.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "46a1761807faccc9a19e86944bbf40610014066306f96edcdedc2fb714bcb7b8"
+dependencies = [
+ "log",
+ "markup5ever 0.39.0",
 ]
 
 [[package]]
@@ -1215,6 +1225,17 @@ dependencies = [
  "web_atoms",
 ]
 
+[[package]]
+name = "markup5ever"
+version = "0.39.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7122d987ec5f704ee56f6e5b41a7d93722e9aae27ae07cafa4036c4d3f9757de"
+dependencies = [
+ "log",
+ "tendril",
+ "web_atoms",
+]
+
 [[package]]
 name = "memchr"
 version = "2.8.0"
@@ -1820,7 +1841,7 @@ dependencies = [
  "cssparser",
  "ego-tree",
  "getopts",
- "html5ever",
+ "html5ever 0.38.0",
  "precomputed-hash",
  "selectors",
  "tendril",
@@ -1828,8 +1849,9 @@ dependencies = [
 
 [[package]]
 name = "scraper-rs"
-version = "0.4.0"
+version = "0.4.2"
 dependencies = [
+ "html5ever 0.39.0",
  "pyo3",
  "pyo3-async-runtimes",
  "scraper",

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "scraper-rs"
-version = "0.4.0"
+version = "0.4.2"
 edition = "2024"
 readme = "README.md"
 
@@ -20,6 +20,7 @@ tokio = { version = "1", features = ["rt", "macros"] }
 scraper = { version = "0.25.0", git = "https://github.com/rust-scraper/scraper", features = ["atomic"], rev = "d07444effb9ca192d9b3919fb920ab9c6c701c4f" }
 
 xee-xpath = "0.1.5"
+html5ever = "0.39.0"
 
 [profile.release]
 lto = "fat"

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: scraper-rust
-Version: 0.4.0
+Version: 0.4.2
 Classifier: Programming Language :: Python
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3 :: Only
@@ -166,6 +166,7 @@ If you have `just` installed, the repo includes helpers: `just build` (local whe
 ## Projects Using scraper-rs
 
 - [**silkworm**](https://github.com/BitingSnakes/silkworm) - Async web scraping framework on top of Rust
+- [**silkworm-mcp**](https://github.com/BitingSnakes/silkworm-mcp) - An MCP server for silkworm
 
 ## Development
 

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/README.md

@@ -139,6 +139,7 @@ If you have `just` installed, the repo includes helpers: `just build` (local whe
 ## Projects Using scraper-rs
 
 - [**silkworm**](https://github.com/BitingSnakes/silkworm) - Async web scraping framework on top of Rust
+- [**silkworm-mcp**](https://github.com/BitingSnakes/silkworm-mcp) - An MCP server for silkworm
 
 ## Development
 

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "scraper-rust"
-version = "0.4.0"
+version = "0.4.2"
 description = "Python bindings around rust-scraper/scraper with PyO3"
 authors = [{ name = "Yehor Smoliakov", email = "egorsmkv@gmail.com" }]
 requires-python = ">=3.10"

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/scraper_rs/__init__.py

@@ -9,6 +9,8 @@ __all__ = [
     "__version__",
     "first",
     "parse",
+    "parse_document",
+    "parse_fragment",
     "prettify",
     "select",
     "select_first",

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/scraper_rs.pyi

@@ -8,6 +8,21 @@ class ElementDict(TypedDict):
     html: str
     attrs: dict[str, str]
 
+class HtmlNodeDict(TypedDict, total=False):
+    node_type: str
+    tag: str
+    namespace: str
+    attrs: dict[str, str]
+    children: list[HtmlNodeDict]
+    text: str
+    quirks_mode: str
+    errors: list[str]
+    name: str
+    public_id: str
+    system_id: str
+    target: str
+    data: str
+
 class Element:
     tag: str
     text: str
@@ -62,6 +77,12 @@ class Document:
 def parse(
     html: str, *, max_size_bytes: int | None = ..., truncate_on_limit: bool = False
 ) -> Document: ...
+def parse_document(
+    html: str, *, max_size_bytes: int | None = ..., truncate_on_limit: bool = False
+) -> HtmlNodeDict: ...
+def parse_fragment(
+    html: str, *, max_size_bytes: int | None = ..., truncate_on_limit: bool = False
+) -> HtmlNodeDict: ...
 def prettify(
     html: str, *, max_size_bytes: int | None = ..., truncate_on_limit: bool = False
 ) -> str: ...

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/src/lib.rs

@@ -5,10 +5,17 @@ use std::future::Future;
 use std::rc::Rc;
 use std::sync::{Arc, Mutex, OnceLock};
 
+use html5ever::interface::{ElementFlags, NodeOrText, QuirksMode};
+use html5ever::tendril::{StrTendril, TendrilSink};
+use html5ever::tree_builder::TreeSink;
+use html5ever::{
+    Attribute, ParseOpts, QualName, local_name, ns, parse_document as html5ever_parse_document,
+    parse_fragment as html5ever_parse_fragment,
+};
 use pyo3::IntoPyObject;
 use pyo3::exceptions::PyValueError;
 use pyo3::prelude::*;
-use pyo3::types::PyDict;
+use pyo3::types::{PyDict, PyList};
 use pyo3::wrap_pyfunction;
 use scraper::{Html, Selector, element_ref::ElementRef};
 use xee_xpath::context::StaticContextBuilder;
@@ -110,6 +117,484 @@ fn ensure_within_size_limit(
     Ok(Cow::Borrowed(html))
 }
 
+#[derive(Clone)]
+enum Html5DictNodeData {
+    Document,
+    Doctype {
+        name: String,
+        public_id: String,
+        system_id: String,
+    },
+    Text {
+        text: String,
+    },
+    Comment {
+        text: String,
+    },
+    Element {
+        name: QualName,
+        attrs: Vec<Attribute>,
+        template_contents: Option<usize>,
+    },
+    ProcessingInstruction {
+        target: String,
+        data: String,
+    },
+}
+
+#[derive(Clone)]
+struct Html5DictNode {
+    parent: Option<usize>,
+    children: Vec<usize>,
+    data: Html5DictNodeData,
+}
+
+#[derive(Clone, Copy)]
+enum Html5ParseKind {
+    Document,
+    Fragment,
+}
+
+#[derive(Debug)]
+struct Html5ElemName {
+    ns: html5ever::Namespace,
+    local_name: html5ever::LocalName,
+}
+
+impl html5ever::interface::ElemName for Html5ElemName {
+    fn ns(&self) -> &html5ever::Namespace {
+        &self.ns
+    }
+
+    fn local_name(&self) -> &html5ever::LocalName {
+        &self.local_name
+    }
+}
+
+struct Html5DictTree {
+    nodes: Vec<Html5DictNode>,
+    quirks_mode: QuirksMode,
+    errors: Vec<String>,
+    parse_kind: Html5ParseKind,
+}
+
+struct Html5DictSink {
+    nodes: RefCell<Vec<Html5DictNode>>,
+    quirks_mode: std::cell::Cell<QuirksMode>,
+    errors: RefCell<Vec<String>>,
+    parse_kind: Html5ParseKind,
+}
+
+impl Html5DictSink {
+    fn new(parse_kind: Html5ParseKind) -> Self {
+        Self {
+            nodes: RefCell::new(vec![Html5DictNode {
+                parent: None,
+                children: Vec::new(),
+                data: Html5DictNodeData::Document,
+            }]),
+            quirks_mode: std::cell::Cell::new(QuirksMode::NoQuirks),
+            errors: RefCell::new(Vec::new()),
+            parse_kind,
+        }
+    }
+
+    fn create_node(&self, data: Html5DictNodeData) -> usize {
+        let mut nodes = self.nodes.borrow_mut();
+        let id = nodes.len();
+        nodes.push(Html5DictNode {
+            parent: None,
+            children: Vec::new(),
+            data,
+        });
+        id
+    }
+
+    fn remove_from_parent_impl(nodes: &mut [Html5DictNode], target: usize) {
+        let Some(parent) = nodes[target].parent.take() else {
+            return;
+        };
+        if let Some(position) = nodes[parent]
+            .children
+            .iter()
+            .position(|&child| child == target)
+        {
+            nodes[parent].children.remove(position);
+        }
+    }
+
+    fn append_text_node(&self, nodes: &mut Vec<Html5DictNode>, parent: usize, text: StrTendril) {
+        if let Some(&last_child) = nodes[parent].children.last()
+            && let Html5DictNodeData::Text { text: existing } = &mut nodes[last_child].data
+        {
+            existing.push_str(text.as_ref());
+            return;
+        }
+
+        let id = nodes.len();
+        nodes.push(Html5DictNode {
+            parent: Some(parent),
+            children: Vec::new(),
+            data: Html5DictNodeData::Text {
+                text: text.to_string(),
+            },
+        });
+        nodes[parent].children.push(id);
+    }
+
+    fn append_node(&self, nodes: &mut [Html5DictNode], parent: usize, child: usize) {
+        Self::remove_from_parent_impl(nodes, child);
+        nodes[child].parent = Some(parent);
+        nodes[parent].children.push(child);
+    }
+}
+
+impl TreeSink for Html5DictSink {
+    type Handle = usize;
+    type Output = Html5DictTree;
+    type ElemName<'a>
+        = Html5ElemName
+    where
+        Self: 'a;
+
+    fn finish(self) -> Self::Output {
+        Html5DictTree {
+            nodes: self.nodes.into_inner(),
+            quirks_mode: self.quirks_mode.get(),
+            errors: self.errors.into_inner(),
+            parse_kind: self.parse_kind,
+        }
+    }
+
+    fn parse_error(&self, msg: Cow<'static, str>) {
+        self.errors.borrow_mut().push(msg.into_owned());
+    }
+
+    fn get_document(&self) -> Self::Handle {
+        0
+    }
+
+    fn elem_name<'a>(&'a self, target: &'a Self::Handle) -> Self::ElemName<'a> {
+        let nodes = self.nodes.borrow();
+        match &nodes[*target].data {
+            Html5DictNodeData::Element { name, .. } => Html5ElemName {
+                ns: name.ns.clone(),
+                local_name: name.local.clone(),
+            },
+            _ => panic!("elem_name called on a non-element node"),
+        }
+    }
+
+    fn create_element(
+        &self,
+        name: QualName,
+        attrs: Vec<Attribute>,
+        flags: ElementFlags,
+    ) -> Self::Handle {
+        let template_contents = if flags.template {
+            Some(self.create_node(Html5DictNodeData::Document))
+        } else {
+            None
+        };
+
+        self.create_node(Html5DictNodeData::Element {
+            name,
+            attrs,
+            template_contents,
+        })
+    }
+
+    fn create_comment(&self, text: StrTendril) -> Self::Handle {
+        self.create_node(Html5DictNodeData::Comment {
+            text: text.to_string(),
+        })
+    }
+
+    fn create_pi(&self, target: StrTendril, data: StrTendril) -> Self::Handle {
+        self.create_node(Html5DictNodeData::ProcessingInstruction {
+            target: target.to_string(),
+            data: data.to_string(),
+        })
+    }
+
+    fn append(&self, parent: &Self::Handle, child: NodeOrText<Self::Handle>) {
+        let mut nodes = self.nodes.borrow_mut();
+        match child {
+            NodeOrText::AppendText(text) => self.append_text_node(&mut nodes, *parent, text),
+            NodeOrText::AppendNode(child) => self.append_node(&mut nodes, *parent, child),
+        }
+    }
+
+    fn append_based_on_parent_node(
+        &self,
+        element: &Self::Handle,
+        prev_element: &Self::Handle,
+        child: NodeOrText<Self::Handle>,
+    ) {
+        let has_parent = self.nodes.borrow()[*element].parent.is_some();
+        if has_parent {
+            self.append_before_sibling(element, child);
+        } else {
+            self.append(prev_element, child);
+        }
+    }
+
+    fn append_doctype_to_document(
+        &self,
+        name: StrTendril,
+        public_id: StrTendril,
+        system_id: StrTendril,
+    ) {
+        let id = self.create_node(Html5DictNodeData::Doctype {
+            name: name.to_string(),
+            public_id: public_id.to_string(),
+            system_id: system_id.to_string(),
+        });
+        self.append(&0, NodeOrText::AppendNode(id));
+    }
+
+    fn get_template_contents(&self, target: &Self::Handle) -> Self::Handle {
+        let nodes = self.nodes.borrow();
+        match &nodes[*target].data {
+            Html5DictNodeData::Element {
+                template_contents: Some(contents),
+                ..
+            } => *contents,
+            _ => panic!("get_template_contents called on a non-template element"),
+        }
+    }
+
+    fn same_node(&self, x: &Self::Handle, y: &Self::Handle) -> bool {
+        x == y
+    }
+
+    fn set_quirks_mode(&self, mode: QuirksMode) {
+        self.quirks_mode.set(mode);
+    }
+
+    fn append_before_sibling(&self, sibling: &Self::Handle, new_node: NodeOrText<Self::Handle>) {
+        let mut nodes = self.nodes.borrow_mut();
+        let Some(parent) = nodes[*sibling].parent else {
+            return;
+        };
+        let Some(position) = nodes[parent]
+            .children
+            .iter()
+            .position(|&child| child == *sibling)
+        else {
+            return;
+        };
+
+        match new_node {
+            NodeOrText::AppendText(text) => {
+                if position > 0 {
+                    let previous = nodes[parent].children[position - 1];
+                    if let Html5DictNodeData::Text { text: existing } = &mut nodes[previous].data {
+                        existing.push_str(text.as_ref());
+                        return;
+                    }
+                }
+
+                if let Html5DictNodeData::Text { text: existing } = &mut nodes[*sibling].data {
+                    existing.insert_str(0, text.as_ref());
+                    return;
+                }
+
+                let id = nodes.len();
+                nodes.push(Html5DictNode {
+                    parent: Some(parent),
+                    children: Vec::new(),
+                    data: Html5DictNodeData::Text {
+                        text: text.to_string(),
+                    },
+                });
+                nodes[parent].children.insert(position, id);
+            }
+            NodeOrText::AppendNode(child) => {
+                Self::remove_from_parent_impl(&mut nodes, child);
+                nodes[child].parent = Some(parent);
+                let Some(new_position) = nodes[parent]
+                    .children
+                    .iter()
+                    .position(|&node| node == *sibling)
+                else {
+                    return;
+                };
+                nodes[parent].children.insert(new_position, child);
+            }
+        }
+    }
+
+    fn add_attrs_if_missing(&self, target: &Self::Handle, attrs: Vec<Attribute>) {
+        let mut nodes = self.nodes.borrow_mut();
+        let Html5DictNodeData::Element {
+            attrs: existing_attrs,
+            ..
+        } = &mut nodes[*target].data
+        else {
+            panic!("add_attrs_if_missing called on a non-element node");
+        };
+
+        for attr in attrs {
+            if existing_attrs
+                .iter()
+                .all(|existing| existing.name != attr.name)
+            {
+                existing_attrs.push(attr);
+            }
+        }
+    }
+
+    fn remove_from_parent(&self, target: &Self::Handle) {
+        let mut nodes = self.nodes.borrow_mut();
+        Self::remove_from_parent_impl(&mut nodes, *target);
+    }
+
+    fn reparent_children(&self, node: &Self::Handle, new_parent: &Self::Handle) {
+        let children = {
+            let mut nodes = self.nodes.borrow_mut();
+            std::mem::take(&mut nodes[*node].children)
+        };
+
+        for child in children {
+            self.append(new_parent, NodeOrText::AppendNode(child));
+        }
+    }
+}
+
+fn quirks_mode_name(mode: QuirksMode) -> &'static str {
+    match mode {
+        QuirksMode::Quirks => "quirks",
+        QuirksMode::LimitedQuirks => "limited-quirks",
+        QuirksMode::NoQuirks => "no-quirks",
+    }
+}
+
+fn qual_name_to_string(name: &QualName) -> String {
+    name.local.to_string()
+}
+
+fn attrs_to_map(attrs: &[Attribute]) -> HashMap<String, String> {
+    attrs
+        .iter()
+        .map(|attr| (attr.name.local.to_string(), attr.value.to_string()))
+        .collect()
+}
+
+fn html5_dict_node_to_py(
+    py: Python<'_>,
+    tree: &Html5DictTree,
+    node_id: usize,
+    root_override: Option<&str>,
+) -> PyResult<Py<PyDict>> {
+    let node = &tree.nodes[node_id];
+    let dict = PyDict::new(py);
+
+    let node_type = root_override.unwrap_or(match &node.data {
+        Html5DictNodeData::Document => "document",
+        Html5DictNodeData::Doctype { .. } => "doctype",
+        Html5DictNodeData::Text { .. } => "text",
+        Html5DictNodeData::Comment { .. } => "comment",
+        Html5DictNodeData::Element { .. } => "element",
+        Html5DictNodeData::ProcessingInstruction { .. } => "processing_instruction",
+    });
+    dict.set_item("node_type", node_type)?;
+
+    match &node.data {
+        Html5DictNodeData::Document => {}
+        Html5DictNodeData::Doctype {
+            name,
+            public_id,
+            system_id,
+        } => {
+            dict.set_item("name", name)?;
+            dict.set_item("public_id", public_id)?;
+            dict.set_item("system_id", system_id)?;
+        }
+        Html5DictNodeData::Text { text } | Html5DictNodeData::Comment { text } => {
+            dict.set_item("text", text)?;
+        }
+        Html5DictNodeData::Element { name, attrs, .. } => {
+            dict.set_item("tag", qual_name_to_string(name))?;
+            dict.set_item("namespace", name.ns.to_string())?;
+            dict.set_item("attrs", attrs_to_map(attrs))?;
+        }
+        Html5DictNodeData::ProcessingInstruction { target, data } => {
+            dict.set_item("target", target)?;
+            dict.set_item("data", data)?;
+        }
+    }
+
+    let child_ids = if node_id == 0 && matches!(tree.parse_kind, Html5ParseKind::Fragment) {
+        let mut flattened = Vec::new();
+        for &child in &node.children {
+            if let Html5DictNodeData::Element { name, .. } = &tree.nodes[child].data
+                && name.local.as_ref() == "html"
+            {
+                flattened.extend(tree.nodes[child].children.iter().copied());
+            } else {
+                flattened.push(child);
+            }
+        }
+        flattened
+    } else {
+        node.children.clone()
+    };
+
+    let children = PyList::empty(py);
+    for child in child_ids {
+        children.append(html5_dict_node_to_py(py, tree, child, None)?)?;
+    }
+    dict.set_item("children", children)?;
+
+    if node_id == 0 {
+        dict.set_item("quirks_mode", quirks_mode_name(tree.quirks_mode))?;
+        dict.set_item("errors", &tree.errors)?;
+    }
+
+    Ok(dict.into())
+}
+
+fn parse_document_to_dict_tree(
+    html: &str,
+    max_size_bytes: Option<usize>,
+    truncate_on_limit: bool,
+) -> PyResult<Html5DictTree> {
+    let max_size_bytes = max_size_bytes.unwrap_or(DEFAULT_MAX_PARSE_BYTES);
+    let html_to_parse = ensure_within_size_limit(html, max_size_bytes, truncate_on_limit)?;
+    Ok(html5ever_parse_document(
+        Html5DictSink::new(Html5ParseKind::Document),
+        ParseOpts::default(),
+    )
+    .one(html_to_parse.as_ref()))
+}
+
+fn parse_fragment_to_dict_tree(
+    html: &str,
+    max_size_bytes: Option<usize>,
+    truncate_on_limit: bool,
+) -> PyResult<Html5DictTree> {
+    let max_size_bytes = max_size_bytes.unwrap_or(DEFAULT_MAX_PARSE_BYTES);
+    let html_to_parse = ensure_within_size_limit(html, max_size_bytes, truncate_on_limit)?;
+    let context_name = QualName::new(None, ns!(html), local_name!("div"));
+    Ok(html5ever_parse_fragment(
+        Html5DictSink::new(Html5ParseKind::Fragment),
+        ParseOpts::default(),
+        context_name,
+        Vec::new(),
+        false,
+    )
+    .one(html_to_parse.as_ref()))
+}
+
+fn html5_tree_to_py_dict(py: Python<'_>, tree: Html5DictTree) -> PyResult<Py<PyDict>> {
+    let root_override = match tree.parse_kind {
+        Html5ParseKind::Document => Some("document"),
+        Html5ParseKind::Fragment => Some("document_fragment"),
+    };
+    html5_dict_node_to_py(py, &tree, 0, root_override)
+}
+
 /// Tiny helper to truncate text in __repr__.
 fn truncate_for_repr(s: &str, max_chars: usize) -> String {
     let mut out = String::new();
@@ -1413,6 +1898,30 @@ fn parse(html: &str, max_size_bytes: Option<usize>, truncate_on_limit: bool) ->
     Document::from_html(html, max_size_bytes, truncate_on_limit)
 }
 
+#[pyfunction(name = "parse_document")]
+#[pyo3(signature = (html, *, max_size_bytes=None, truncate_on_limit=false))]
+fn parse_document_dict(
+    py: Python<'_>,
+    html: &str,
+    max_size_bytes: Option<usize>,
+    truncate_on_limit: bool,
+) -> PyResult<Py<PyDict>> {
+    let tree = parse_document_to_dict_tree(html, max_size_bytes, truncate_on_limit)?;
+    html5_tree_to_py_dict(py, tree)
+}
+
+#[pyfunction(name = "parse_fragment")]
+#[pyo3(signature = (html, *, max_size_bytes=None, truncate_on_limit=false))]
+fn parse_fragment_dict(
+    py: Python<'_>,
+    html: &str,
+    max_size_bytes: Option<usize>,
+    truncate_on_limit: bool,
+) -> PyResult<Py<PyDict>> {
+    let tree = parse_fragment_to_dict_tree(html, max_size_bytes, truncate_on_limit)?;
+    html5_tree_to_py_dict(py, tree)
+}
+
 #[pyfunction]
 #[pyo3(signature = (html, *, max_size_bytes=None, truncate_on_limit=false))]
 fn prettify(
@@ -1629,6 +2138,8 @@ fn scraper_rs(_py: Python<'_>, m: &Bound<'_, PyModule>) -> PyResult<()> {
 
     // Top-level functions
     m.add_function(wrap_pyfunction!(parse, m)?)?;
+    m.add_function(wrap_pyfunction!(parse_document_dict, m)?)?;
+    m.add_function(wrap_pyfunction!(parse_fragment_dict, m)?)?;
     m.add_function(wrap_pyfunction!(prettify, m)?)?;
     m.add_function(wrap_pyfunction!(select, m)?)?;
     m.add_function(wrap_pyfunction!(select_first, m)?)?;

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/tests/test_scraper.py

@@ -7,6 +7,8 @@ from scraper_rs import (
     __version__,
     first,
     parse,
+    parse_document,
+    parse_fragment,
     prettify,
     select,
     select_first,
@@ -204,6 +206,56 @@ def test_top_level_parse_and_select(sample_html: str) -> None:
     assert xpath_first(sample_html, "//div[@data-id='1']/a").text == "First"
 
 
+def test_parse_document_to_dict(sample_html: str) -> None:
+    parsed = parse_document(sample_html)
+
+    assert parsed["node_type"] == "document"
+    assert parsed["quirks_mode"] in {"quirks", "limited-quirks", "no-quirks"}
+    assert isinstance(parsed["errors"], list)
+
+    html_node = next(
+        child
+        for child in parsed["children"]
+        if child["node_type"] == "element" and child["tag"] == "html"
+    )
+    body_node = next(
+        child
+        for child in html_node["children"]
+        if child["node_type"] == "element" and child["tag"] == "body"
+    )
+    first_div = next(
+        child
+        for child in body_node["children"]
+        if child["node_type"] == "element" and child["tag"] == "div"
+    )
+
+    assert first_div["attrs"]["class"] == "item"
+    assert first_div["attrs"]["data-id"] == "1"
+
+
+def test_parse_fragment_to_dict() -> None:
+    parsed = parse_fragment(
+        '<div class="item">Hello <span>world</span><!-- note --></div>'
+    )
+
+    assert parsed["node_type"] == "document_fragment"
+    assert parsed["errors"] == []
+
+    div_node = next(
+        child
+        for child in parsed["children"]
+        if child["node_type"] == "element" and child["tag"] == "div"
+    )
+
+    assert div_node["attrs"]["class"] == "item"
+    assert div_node["children"][0]["node_type"] == "text"
+    assert div_node["children"][0]["text"] == "Hello "
+    assert div_node["children"][1]["tag"] == "span"
+    assert div_node["children"][1]["children"][0]["text"] == "world"
+    assert div_node["children"][2]["node_type"] == "comment"
+    assert div_node["children"][2]["text"] == " note "
+
+
 def test_css_alias_and_invalid_selector(sample_html: str) -> None:
     doc = Document(sample_html)
 

{scraper_rust-0.4.0 → scraper_rust-0.4.2}/uv.lock

@@ -43,26 +43,26 @@ wheels = [
 
 [[package]]
 name = "maturin"
-version = "1.12.4"
+version = "1.12.6"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "tomli", marker = "python_full_version < '3.11'" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/2f/a6/54e73f0ec0224488ae25196ce8b4df298cae613b099ad0c4f39dd7e3a8d2/maturin-1.12.4.tar.gz", hash = "sha256:06f6438be7e723aaf4b412fb34839854b540a1350f7614fadf5bd1db2b98d5f7", size = 262134, upload-time = "2026-02-21T10:24:25.64Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/0c/18/8b2eebd3ea086a5ec73d7081f95ec64918ceda1900075902fc296ea3ad55/maturin-1.12.6.tar.gz", hash = "sha256:d37be3a811a7f2ee28a0fa0964187efa50e90f21da0c6135c27787fa0b6a89db", size = 269165, upload-time = "2026-03-01T14:54:04.21Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/e1/cd/8285f37bf968b8485e3c7eb43349a5adbccfddfc487cd4327fb9104578cc/maturin-1.12.4-py3-none-linux_armv6l.whl", hash = "sha256:cf8a0eddef9ab8773bc823c77aed3de9a5c85fb760c86448048a79ef89794c81", size = 9758449, upload-time = "2026-02-21T10:24:35.382Z" },
-    { url = "https://files.pythonhosted.org/packages/d9/91/f51191db83735f77bc988c8034730bb63b750a4a1a04f9c8cba10f44ad45/maturin-1.12.4-py3-none-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:eba1bd1c1513d00fec75228da98622c68a9f50f9693aaa6fb7dacb244e7bbf26", size = 18938848, upload-time = "2026-02-21T10:24:10.701Z" },
-    { url = "https://files.pythonhosted.org/packages/65/47/03c422adeac93b903354b322bba632754fdb134b27ace71b5603feba5906/maturin-1.12.4-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:89749cfc0e6baf5517fa370729a98955552e42fefc406b95732d5c8e85bc90c0", size = 9791641, upload-time = "2026-02-21T10:24:21.72Z" },
-    { url = "https://files.pythonhosted.org/packages/5e/30/dd78acf6afc48d358512b5ed928fd24e2bc6b68db69b1f6bba3ffd7bcaed/maturin-1.12.4-py3-none-manylinux_2_12_i686.manylinux2010_i686.musllinux_1_1_i686.whl", hash = "sha256:4d68664e5b81f282144a3b717a7e8593ec94ac87d7ae563a4c464e93d6cde877", size = 9811625, upload-time = "2026-02-21T10:24:08.152Z" },
-    { url = "https://files.pythonhosted.org/packages/e3/9a/a6e358a18815ab090ef55187da0066df01a955c7c44a61fb83b127055f23/maturin-1.12.4-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.musllinux_1_1_x86_64.whl", hash = "sha256:88e09e6c386b08974fab0c7e4c07d7c7c50a0ba63095d31e930d80568488e1be", size = 10255812, upload-time = "2026-02-21T10:24:15.117Z" },
-    { url = "https://files.pythonhosted.org/packages/4a/c5/84dfcce1f3475237cba6e6201a1939980025afbb41c076aa5147b10ac202/maturin-1.12.4-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.musllinux_1_1_aarch64.whl", hash = "sha256:5cc56481b0f360571587c35a1d960ce6d0a0258d49aebb6af98fff9db837c337", size = 9645462, upload-time = "2026-02-21T10:24:28.814Z" },
-    { url = "https://files.pythonhosted.org/packages/de/82/0845fff86ea044028302db17bc611e9bfe1b7b2c992756162cbe71267df5/maturin-1.12.4-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.musllinux_1_1_armv7l.whl", hash = "sha256:8fd7eb0c9bb017e98d81aa86a1d440b912fe4f7f219571035dd6ab330c82071c", size = 9593649, upload-time = "2026-02-21T10:24:33.376Z" },
-    { url = "https://files.pythonhosted.org/packages/2b/14/6e8969cd48c7c8ea27d7638e572d46eeba9aa0cb370d3031eb6a3f10ff8d/maturin-1.12.4-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.musllinux_1_1_ppc64le.whl", hash = "sha256:5bb07c349dd066277a61e017a6d6e0860cd54b7b33f8ead10b9e5a4ffb740a0a", size = 12681515, upload-time = "2026-02-21T10:24:31.097Z" },
-    { url = "https://files.pythonhosted.org/packages/ac/8d/2ad86623dca3cfa394049f4220188dececa6e4cefd73ac1f1385fc79c876/maturin-1.12.4-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c21baaed066b5bec893db2d261bfe3b9da054d99c018326f0bdcf1dc4c3a1eb9", size = 10448453, upload-time = "2026-02-21T10:24:26.827Z" },
-    { url = "https://files.pythonhosted.org/packages/9c/eb/c66e2d3272e74dd590ae81bb51590bd98c3cd4e3f6629d4e4218bd6a5c28/maturin-1.12.4-py3-none-manylinux_2_31_riscv64.musllinux_1_1_riscv64.whl", hash = "sha256:939c4c57efa8ea982a991ee3ccb3992364622e9cbd1ede922b5cfb0f652bf517", size = 9970879, upload-time = "2026-02-21T10:24:12.881Z" },
-    { url = "https://files.pythonhosted.org/packages/38/a0/998f8063d67fa19639179af7e8ea46016ceaa12f85b9720a2e4846449f43/maturin-1.12.4-py3-none-win32.whl", hash = "sha256:d72f626616292cb3e283941f47835ffc608207ebd8f95f4c50523a6631ffcb2e", size = 8518146, upload-time = "2026-02-21T10:24:17.296Z" },
-    { url = "https://files.pythonhosted.org/packages/69/14/6ceea315db6e47093442ec70c2d01bb011d69f5243de5fc0e6a5fab97513/maturin-1.12.4-py3-none-win_amd64.whl", hash = "sha256:ab32c5ff7579a549421cae03e6297d3b03d7b81fa2934e3bdf24a102d99eb378", size = 9863686, upload-time = "2026-02-21T10:24:19.35Z" },
-    { url = "https://files.pythonhosted.org/packages/d4/28/73e14739c6f7605ff9b9d108726d3ff529d4f91a7838739b4dd0afd33ec1/maturin-1.12.4-py3-none-win_arm64.whl", hash = "sha256:b8c05d24209af50ed9ae9e5de473c84866b9676c637fcfad123ee57f4a9ed098", size = 8557843, upload-time = "2026-02-21T10:24:23.894Z" },
+    { url = "https://files.pythonhosted.org/packages/71/8b/9ddfde8a485489e3ebdc50ee3042ef1c854f00dfea776b951068f6ffe451/maturin-1.12.6-py3-none-linux_armv6l.whl", hash = "sha256:6892b4176992fcc143f9d1c1c874a816e9a041248eef46433db87b0f0aff4278", size = 9789847, upload-time = "2026-03-01T14:54:09.172Z" },
+    { url = "https://files.pythonhosted.org/packages/ef/e8/5f7fd3763f214a77ac0388dbcc71cc30aec5490016bd0c8e6bd729fc7b0a/maturin-1.12.6-py3-none-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:c0c742beeeef7fb93b6a81bd53e75507887e396fd1003c45117658d063812dad", size = 19023833, upload-time = "2026-03-01T14:53:46.743Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/7f/706ff3839c8b2046436d4c2bc97596c558728264d18abc298a1ad862a4be/maturin-1.12.6-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:2cb41139295eed6411d3cdafc7430738094c2721f34b7eeb44f33cac516115dc", size = 9821620, upload-time = "2026-03-01T14:54:12.04Z" },
+    { url = "https://files.pythonhosted.org/packages/0e/9c/70917fb123c8dd6b595e913616c9c72d730cbf4a2b6cac8077dc02a12586/maturin-1.12.6-py3-none-manylinux_2_12_i686.manylinux2010_i686.musllinux_1_1_i686.whl", hash = "sha256:351f3af1488a7cbdcff3b6d8482c17164273ac981378a13a4a9937a49aec7d71", size = 9849107, upload-time = "2026-03-01T14:53:48.971Z" },
+    { url = "https://files.pythonhosted.org/packages/59/ea/f1d6ad95c0a12fbe761a7c28a57540341f188564dbe8ad730a4d1788cd32/maturin-1.12.6-py3-none-manylinux_2_12_x86_64.manylinux2010_x86_64.musllinux_1_1_x86_64.whl", hash = "sha256:6dbddfe4dc7ddee60bbac854870bd7cfec660acb54d015d24597d59a1c828f61", size = 10242855, upload-time = "2026-03-01T14:53:44.605Z" },
+    { url = "https://files.pythonhosted.org/packages/93/1b/2419843a4f1d2fb4747f3dc3d9c4a2881cd97a3274dd94738fcdf0835e79/maturin-1.12.6-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.musllinux_1_1_aarch64.whl", hash = "sha256:8fdb0f63e77ee3df0f027a120e9af78dbc31edf0eb0f263d55783c250c33b728", size = 9674972, upload-time = "2026-03-01T14:53:52.763Z" },
+    { url = "https://files.pythonhosted.org/packages/71/46/b60ab2fc996d904b40e55bd475599dcdccd8f7ad3e649bf95e87970df466/maturin-1.12.6-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.musllinux_1_1_armv7l.whl", hash = "sha256:fa84b7493a2e80759cacc2e668fa5b444d55b9994e90707c42904f55d6322c1e", size = 9645755, upload-time = "2026-03-01T14:53:58.497Z" },
+    { url = "https://files.pythonhosted.org/packages/a4/96/03f2b55a8c226805115232fc23c4a4f33f0c9d39e11efab8166dc440f80d/maturin-1.12.6-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.musllinux_1_1_ppc64le.whl", hash = "sha256:e90dc12bc6a38e9495692a36c9e231c4d7e0c9bfde60719468ab7d8673db3c45", size = 12737612, upload-time = "2026-03-01T14:54:05.393Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/c2/648667022c5b53cdccefa67c245e8a984970f3045820f00c2e23bdb2aff4/maturin-1.12.6-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:06fc8d089f98623ce924c669b70911dfed30f9a29956c362945f727f9abc546b", size = 10455028, upload-time = "2026-03-01T14:54:07.349Z" },
+    { url = "https://files.pythonhosted.org/packages/63/d6/5b5efe3ca0c043357ed3f8d2b2d556169fdbf1ff75e50e8e597708a359d2/maturin-1.12.6-py3-none-manylinux_2_31_riscv64.musllinux_1_1_riscv64.whl", hash = "sha256:75133e56274d43b9227fd49dca9a86e32f1fd56a7b55544910c4ce978c2bb5aa", size = 10014531, upload-time = "2026-03-01T14:53:54.548Z" },
+    { url = "https://files.pythonhosted.org/packages/68/d5/39c594c27b1a8b32a0cb95fff9ad60b888c4352d1d1c389ac1bd20dc1e16/maturin-1.12.6-py3-none-win32.whl", hash = "sha256:3f32e0a3720b81423c9d35c14e728cb1f954678124749776dc72d533ea1115e8", size = 8553012, upload-time = "2026-03-01T14:53:50.706Z" },
+    { url = "https://files.pythonhosted.org/packages/94/66/b262832a91747e04051e21f986bd01a8af81fbffafacc7d66a11e79aab5f/maturin-1.12.6-py3-none-win_amd64.whl", hash = "sha256:977290159d252db946054a0555263c59b3d0c7957135c69e690f4b1558ee9983", size = 9890470, upload-time = "2026-03-01T14:53:56.659Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/47/76b8ca470ddc8d7d36aa8c15f5a6aed1841806bb93a0f4ead8ee61e9a088/maturin-1.12.6-py3-none-win_arm64.whl", hash = "sha256:bae91976cdc8148038e13c881e1e844e5c63e58e026e8b9945aa2d19b3b4ae89", size = 8606158, upload-time = "2026-03-01T14:54:02.423Z" },
 ]
 
 [[package]]
@@ -126,32 +126,32 @@ wheels = [
 
 [[package]]
 name = "ruff"
-version = "0.15.2"
+version = "0.15.7"
 source = { registry = "https://pypi.org/simple" }
-sdist = { url = "https://files.pythonhosted.org/packages/06/04/eab13a954e763b0606f460443fcbf6bb5a0faf06890ea3754ff16523dce5/ruff-0.15.2.tar.gz", hash = "sha256:14b965afee0969e68bb871eba625343b8673375f457af4abe98553e8bbb98342", size = 4558148, upload-time = "2026-02-19T22:32:20.271Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/a1/22/9e4f66ee588588dc6c9af6a994e12d26e19efbe874d1a909d09a6dac7a59/ruff-0.15.7.tar.gz", hash = "sha256:04f1ae61fc20fe0b148617c324d9d009b5f63412c0b16474f3d5f1a1a665f7ac", size = 4601277, upload-time = "2026-03-19T16:26:22.605Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/2f/70/3a4dc6d09b13cb3e695f28307e5d889b2e1a66b7af9c5e257e796695b0e6/ruff-0.15.2-py3-none-linux_armv6l.whl", hash = "sha256:120691a6fdae2f16d65435648160f5b81a9625288f75544dc40637436b5d3c0d", size = 10430565, upload-time = "2026-02-19T22:32:41.824Z" },
-    { url = "https://files.pythonhosted.org/packages/71/0b/bb8457b56185ece1305c666dc895832946d24055be90692381c31d57466d/ruff-0.15.2-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:a89056d831256099658b6bba4037ac6dd06f49d194199215befe2bb10457ea5e", size = 10820354, upload-time = "2026-02-19T22:32:07.366Z" },
-    { url = "https://files.pythonhosted.org/packages/2d/c1/e0532d7f9c9e0b14c46f61b14afd563298b8b83f337b6789ddd987e46121/ruff-0.15.2-py3-none-macosx_11_0_arm64.whl", hash = "sha256:e36dee3a64be0ebd23c86ffa3aa3fd3ac9a712ff295e192243f814a830b6bd87", size = 10170767, upload-time = "2026-02-19T22:32:13.188Z" },
-    { url = "https://files.pythonhosted.org/packages/47/e8/da1aa341d3af017a21c7a62fb5ec31d4e7ad0a93ab80e3a508316efbcb23/ruff-0.15.2-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:a9fb47b6d9764677f8c0a193c0943ce9a05d6763523f132325af8a858eadc2b9", size = 10529591, upload-time = "2026-02-19T22:32:02.547Z" },
-    { url = "https://files.pythonhosted.org/packages/93/74/184fbf38e9f3510231fbc5e437e808f0b48c42d1df9434b208821efcd8d6/ruff-0.15.2-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f376990f9d0d6442ea9014b19621d8f2aaf2b8e39fdbfc79220b7f0c596c9b80", size = 10260771, upload-time = "2026-02-19T22:32:36.938Z" },
-    { url = "https://files.pythonhosted.org/packages/05/ac/605c20b8e059a0bc4b42360414baa4892ff278cec1c91fff4be0dceedefd/ruff-0.15.2-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:2dcc987551952d73cbf5c88d9fdee815618d497e4df86cd4c4824cc59d5dd75f", size = 11045791, upload-time = "2026-02-19T22:32:31.642Z" },
-    { url = "https://files.pythonhosted.org/packages/fd/52/db6e419908f45a894924d410ac77d64bdd98ff86901d833364251bd08e22/ruff-0.15.2-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:42a47fd785cbe8c01b9ff45031af875d101b040ad8f4de7bbb716487c74c9a77", size = 11879271, upload-time = "2026-02-19T22:32:29.305Z" },
-    { url = "https://files.pythonhosted.org/packages/3e/d8/7992b18f2008bdc9231d0f10b16df7dda964dbf639e2b8b4c1b4e91b83af/ruff-0.15.2-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:cbe9f49354866e575b4c6943856989f966421870e85cd2ac94dccb0a9dcb2fea", size = 11303707, upload-time = "2026-02-19T22:32:22.492Z" },
-    { url = "https://files.pythonhosted.org/packages/d7/02/849b46184bcfdd4b64cde61752cc9a146c54759ed036edd11857e9b8443b/ruff-0.15.2-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b7a672c82b5f9887576087d97be5ce439f04bbaf548ee987b92d3a7dede41d3a", size = 11149151, upload-time = "2026-02-19T22:32:44.234Z" },
-    { url = "https://files.pythonhosted.org/packages/70/04/f5284e388bab60d1d3b99614a5a9aeb03e0f333847e2429bebd2aaa1feec/ruff-0.15.2-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:72ecc64f46f7019e2bcc3cdc05d4a7da958b629a5ab7033195e11a438403d956", size = 11091132, upload-time = "2026-02-19T22:32:24.691Z" },
-    { url = "https://files.pythonhosted.org/packages/fa/ae/88d844a21110e14d92cf73d57363fab59b727ebeabe78009b9ccb23500af/ruff-0.15.2-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:8dcf243b15b561c655c1ef2f2b0050e5d50db37fe90115507f6ff37d865dc8b4", size = 10504717, upload-time = "2026-02-19T22:32:26.75Z" },
-    { url = "https://files.pythonhosted.org/packages/64/27/867076a6ada7f2b9c8292884ab44d08fd2ba71bd2b5364d4136f3cd537e1/ruff-0.15.2-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:dab6941c862c05739774677c6273166d2510d254dac0695c0e3f5efa1b5585de", size = 10263122, upload-time = "2026-02-19T22:32:10.036Z" },
-    { url = "https://files.pythonhosted.org/packages/e7/ef/faf9321d550f8ebf0c6373696e70d1758e20ccdc3951ad7af00c0956be7c/ruff-0.15.2-py3-none-musllinux_1_2_i686.whl", hash = "sha256:1b9164f57fc36058e9a6806eb92af185b0697c9fe4c7c52caa431c6554521e5c", size = 10735295, upload-time = "2026-02-19T22:32:39.227Z" },
-    { url = "https://files.pythonhosted.org/packages/2f/55/e8089fec62e050ba84d71b70e7834b97709ca9b7aba10c1a0b196e493f97/ruff-0.15.2-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:80d24fcae24d42659db7e335b9e1531697a7102c19185b8dc4a028b952865fd8", size = 11241641, upload-time = "2026-02-19T22:32:34.617Z" },
-    { url = "https://files.pythonhosted.org/packages/23/01/1c30526460f4d23222d0fabd5888868262fd0e2b71a00570ca26483cd993/ruff-0.15.2-py3-none-win32.whl", hash = "sha256:fd5ff9e5f519a7e1bd99cbe8daa324010a74f5e2ebc97c6242c08f26f3714f6f", size = 10507885, upload-time = "2026-02-19T22:32:15.635Z" },
-    { url = "https://files.pythonhosted.org/packages/5c/10/3d18e3bbdf8fc50bbb4ac3cc45970aa5a9753c5cb51bf9ed9a3cd8b79fa3/ruff-0.15.2-py3-none-win_amd64.whl", hash = "sha256:d20014e3dfa400f3ff84830dfb5755ece2de45ab62ecea4af6b7262d0fb4f7c5", size = 11623725, upload-time = "2026-02-19T22:32:04.947Z" },
-    { url = "https://files.pythonhosted.org/packages/6d/78/097c0798b1dab9f8affe73da9642bb4500e098cb27fd8dc9724816ac747b/ruff-0.15.2-py3-none-win_arm64.whl", hash = "sha256:cabddc5822acdc8f7b5527b36ceac55cc51eec7b1946e60181de8fe83ca8876e", size = 10941649, upload-time = "2026-02-19T22:32:18.108Z" },
+    { url = "https://files.pythonhosted.org/packages/41/2f/0b08ced94412af091807b6119ca03755d651d3d93a242682bf020189db94/ruff-0.15.7-py3-none-linux_armv6l.whl", hash = "sha256:a81cc5b6910fb7dfc7c32d20652e50fa05963f6e13ead3c5915c41ac5d16668e", size = 10489037, upload-time = "2026-03-19T16:26:32.47Z" },
+    { url = "https://files.pythonhosted.org/packages/91/4a/82e0fa632e5c8b1eba5ee86ecd929e8ff327bbdbfb3c6ac5d81631bef605/ruff-0.15.7-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:722d165bd52403f3bdabc0ce9e41fc47070ac56d7a91b4e0d097b516a53a3477", size = 10955433, upload-time = "2026-03-19T16:27:00.205Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/10/12586735d0ff42526ad78c049bf51d7428618c8b5c467e72508c694119df/ruff-0.15.7-py3-none-macosx_11_0_arm64.whl", hash = "sha256:7fbc2448094262552146cbe1b9643a92f66559d3761f1ad0656d4991491af49e", size = 10269302, upload-time = "2026-03-19T16:26:26.183Z" },
+    { url = "https://files.pythonhosted.org/packages/eb/5d/32b5c44ccf149a26623671df49cbfbd0a0ae511ff3df9d9d2426966a8d57/ruff-0.15.7-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:6b39329b60eba44156d138275323cc726bbfbddcec3063da57caa8a8b1d50adf", size = 10607625, upload-time = "2026-03-19T16:27:03.263Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/f1/f0001cabe86173aaacb6eb9bb734aa0605f9a6aa6fa7d43cb49cbc4af9c9/ruff-0.15.7-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:87768c151808505f2bfc93ae44e5f9e7c8518943e5074f76ac21558ef5627c85", size = 10324743, upload-time = "2026-03-19T16:27:09.791Z" },
+    { url = "https://files.pythonhosted.org/packages/7a/87/b8a8f3d56b8d848008559e7c9d8bf367934d5367f6d932ba779456e2f73b/ruff-0.15.7-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:fb0511670002c6c529ec66c0e30641c976c8963de26a113f3a30456b702468b0", size = 11138536, upload-time = "2026-03-19T16:27:06.101Z" },
+    { url = "https://files.pythonhosted.org/packages/e4/f2/4fd0d05aab0c5934b2e1464784f85ba2eab9d54bffc53fb5430d1ed8b829/ruff-0.15.7-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e0d19644f801849229db8345180a71bee5407b429dd217f853ec515e968a6912", size = 11994292, upload-time = "2026-03-19T16:26:48.718Z" },
+    { url = "https://files.pythonhosted.org/packages/64/22/fc4483871e767e5e95d1622ad83dad5ebb830f762ed0420fde7dfa9d9b08/ruff-0.15.7-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:4806d8e09ef5e84eb19ba833d0442f7e300b23fe3f0981cae159a248a10f0036", size = 11398981, upload-time = "2026-03-19T16:26:54.513Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/99/66f0343176d5eab02c3f7fcd2de7a8e0dd7a41f0d982bee56cd1c24db62b/ruff-0.15.7-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:dce0896488562f09a27b9c91b1f58a097457143931f3c4d519690dea54e624c5", size = 11242422, upload-time = "2026-03-19T16:26:29.277Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/3a/a7060f145bfdcce4c987ea27788b30c60e2c81d6e9a65157ca8afe646328/ruff-0.15.7-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:1852ce241d2bc89e5dc823e03cff4ce73d816b5c6cdadd27dbfe7b03217d2a12", size = 11232158, upload-time = "2026-03-19T16:26:42.321Z" },
+    { url = "https://files.pythonhosted.org/packages/a7/53/90fbb9e08b29c048c403558d3cdd0adf2668b02ce9d50602452e187cd4af/ruff-0.15.7-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:5f3e4b221fb4bd293f79912fc5e93a9063ebd6d0dcbd528f91b89172a9b8436c", size = 10577861, upload-time = "2026-03-19T16:26:57.459Z" },
+    { url = "https://files.pythonhosted.org/packages/2f/aa/5f486226538fe4d0f0439e2da1716e1acf895e2a232b26f2459c55f8ddad/ruff-0.15.7-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:b15e48602c9c1d9bdc504b472e90b90c97dc7d46c7028011ae67f3861ceba7b4", size = 10327310, upload-time = "2026-03-19T16:26:35.909Z" },
+    { url = "https://files.pythonhosted.org/packages/99/9e/271afdffb81fe7bfc8c43ba079e9d96238f674380099457a74ccb3863857/ruff-0.15.7-py3-none-musllinux_1_2_i686.whl", hash = "sha256:1b4705e0e85cedc74b0a23cf6a179dbb3df184cb227761979cc76c0440b5ab0d", size = 10840752, upload-time = "2026-03-19T16:26:45.723Z" },
+    { url = "https://files.pythonhosted.org/packages/bf/29/a4ae78394f76c7759953c47884eb44de271b03a66634148d9f7d11e721bd/ruff-0.15.7-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:112c1fa316a558bb34319282c1200a8bf0495f1b735aeb78bfcb2991e6087580", size = 11336961, upload-time = "2026-03-19T16:26:39.076Z" },
+    { url = "https://files.pythonhosted.org/packages/26/6b/8786ba5736562220d588a2f6653e6c17e90c59ced34a2d7b512ef8956103/ruff-0.15.7-py3-none-win32.whl", hash = "sha256:6d39e2d3505b082323352f733599f28169d12e891f7dd407f2d4f54b4c2886de", size = 10582538, upload-time = "2026-03-19T16:26:15.992Z" },
+    { url = "https://files.pythonhosted.org/packages/2b/e9/346d4d3fffc6871125e877dae8d9a1966b254fbd92a50f8561078b88b099/ruff-0.15.7-py3-none-win_amd64.whl", hash = "sha256:4d53d712ddebcd7dace1bc395367aec12c057aacfe9adbb6d832302575f4d3a1", size = 11755839, upload-time = "2026-03-19T16:26:19.897Z" },
+    { url = "https://files.pythonhosted.org/packages/8f/e8/726643a3ea68c727da31570bde48c7a10f1aa60eddd628d94078fec586ff/ruff-0.15.7-py3-none-win_arm64.whl", hash = "sha256:18e8d73f1c3fdf27931497972250340f92e8c861722161a9caeb89a58ead6ed2", size = 11023304, upload-time = "2026-03-19T16:26:51.669Z" },
 ]
 
 [[package]]
 name = "scraper-rust"
-version = "0.4.0"
+version = "0.4.2"
 source = { editable = "." }
 
 [package.optional-dependencies]