Merge pull request #206 from kornelski/html5up

Upgrade html5ever

.github/workflows/main.yml

@@ -39,7 +39,7 @@ jobs:
       - name: Set toolchain
         run: |
           rustup set profile minimal
-          rustup override set 1.61.0
+          rustup override set 1.65.0
       - name: Check
         run: cargo check --lib --all-features
         env:

CHANGELOG.md

@@ -1,6 +1,6 @@
 # Unreleased
 
-* Bump MSRV to 1.61.0
+* Bump MSRV to 1.65.0
 * Improve panic message when `clean_content_tags` conflicts with other options
 
 # 4.0.0

Cargo.toml

@@ -10,10 +10,10 @@ documentation = "https://docs.rs/ammonia/"
 repository = "https://github.com/rust-ammonia/ammonia"
 categories = [ "web-programming", "text-processing" ]
 edition = "2021"
-rust-version = "1.61"
+rust-version = "1.65"
 
 [dependencies]
-html5ever = "0.27"
+html5ever = "0.29"
 maplit = "1.0"
 tendril = "0.4"
 url = "2"

README.md

@@ -2,7 +2,7 @@ HTML Sanitization
 =================
 
 [![Crates.IO](https://img.shields.io/crates/v/ammonia.svg)](https://crates.io/crates/ammonia)
-![Requires rustc 1.61.0](https://img.shields.io/badge/rustc-1.61.0+-green.svg)
+![Requires rustc 1.65.0](https://img.shields.io/badge/rustc-1.65.0+-green.svg)
 
 Ammonia is a whitelist-based HTML sanitization library. It is designed to
 prevent cross-site scripting, layout breaking, and clickjacking caused

src/lib.rs

@@ -41,6 +41,7 @@ use maplit::{hashmap, hashset};
 use once_cell::sync::Lazy;
 use rcdom::{Handle, NodeData, RcDom, SerializableHandle};
 use std::borrow::{Borrow, Cow};
+use std::cell::Cell;
 use std::cmp::max;
 use std::collections::{HashMap, HashSet};
 use std::fmt::{self, Display};
@@ -179,37 +180,37 @@ pub fn is_html(input: &str) -> bool {
     let mut input = BufferQueue::default();
     input.push_back(chunk.try_reinterpret().unwrap());
 
-    let mut tok = Tokenizer::new(santok, Default::default());
+    let tok = Tokenizer::new(santok, Default::default());
     let _ = tok.feed(&mut input);
     tok.end();
-    tok.sink.was_sanitized
+    tok.sink.was_sanitized.get()
 }
 
-#[derive(Copy, Clone)]
+#[derive(Clone)]
 struct SanitizationTokenizer {
-    was_sanitized: bool,
+    was_sanitized: Cell<bool>,
 }
 
 impl SanitizationTokenizer {
     pub fn new() -> SanitizationTokenizer {
         SanitizationTokenizer {
-            was_sanitized: false,
+            was_sanitized: false.into(),
         }
     }
 }
 
 impl TokenSink for SanitizationTokenizer {
     type Handle = ();
-    fn process_token(&mut self, token: Token, _line_number: u64) -> TokenSinkResult<()> {
+    fn process_token(&self, token: Token, _line_number: u64) -> TokenSinkResult<()> {
         match token {
             Token::CharacterTokens(_) | Token::EOFToken | Token::ParseError(_) => {}
             _ => {
-                self.was_sanitized = true;
+                self.was_sanitized.set(true);
             }
         }
         TokenSinkResult::Continue
     }
-    fn end(&mut self) {}
+    fn end(&self) {}
 }
 
 /// An HTML sanitizer.

src/rcdom.rs

@@ -210,10 +210,10 @@ pub struct RcDom {
     pub document: Handle,
 
     /// Errors that occurred during parsing.
-    pub errors: Vec<Cow<'static, str>>,
+    pub errors: RefCell<Vec<Cow<'static, str>>>,
 
     /// The document's quirks mode.
-    pub quirks_mode: QuirksMode,
+    pub quirks_mode: Cell<QuirksMode>,
 }
 
 impl TreeSink for RcDom {
@@ -224,15 +224,17 @@ impl TreeSink for RcDom {
 
     type Handle = Handle;
 
-    fn parse_error(&mut self, msg: Cow<'static, str>) {
-        self.errors.push(msg);
+    type ElemName<'a> = ExpandedName<'a>;
+
+    fn parse_error(&self, msg: Cow<'static, str>) {
+        self.errors.borrow_mut().push(msg);
     }
 
-    fn get_document(&mut self) -> Handle {
+    fn get_document(&self) -> Handle {
         self.document.clone()
     }
 
-    fn get_template_contents(&mut self, target: &Handle) -> Handle {
+    fn get_template_contents(&self, target: &Handle) -> Handle {
         if let NodeData::Element {
             ref template_contents,
             ..
@@ -248,8 +250,8 @@ impl TreeSink for RcDom {
         }
     }
 
-    fn set_quirks_mode(&mut self, mode: QuirksMode) {
-        self.quirks_mode = mode;
+    fn set_quirks_mode(&self, mode: QuirksMode) {
+        self.quirks_mode.set(mode);
     }
 
     fn same_node(&self, x: &Handle, y: &Handle) -> bool {
@@ -264,7 +266,7 @@ impl TreeSink for RcDom {
     }
 
     fn create_element(
-        &mut self,
+        &self,
         name: QualName,
         attrs: Vec<Attribute>,
         flags: ElementFlags,
@@ -281,18 +283,18 @@ impl TreeSink for RcDom {
         })
     }
 
-    fn create_comment(&mut self, text: StrTendril) -> Handle {
+    fn create_comment(&self, text: StrTendril) -> Handle {
         Node::new(NodeData::Comment { contents: text })
     }
 
-    fn create_pi(&mut self, target: StrTendril, data: StrTendril) -> Handle {
+    fn create_pi(&self, target: StrTendril, data: StrTendril) -> Handle {
         Node::new(NodeData::ProcessingInstruction {
             target,
             contents: data,
         })
     }
 
-    fn append(&mut self, parent: &Handle, child: NodeOrText<Handle>) {
+    fn append(&self, parent: &Handle, child: NodeOrText<Handle>) {
         // Append to an existing Text node if we have one.
         if let NodeOrText::AppendText(ref text) = child {
             if let Some(h) = parent.children.borrow().last() {
@@ -313,7 +315,7 @@ impl TreeSink for RcDom {
         );
     }
 
-    fn append_before_sibling(&mut self, sibling: &Handle, child: NodeOrText<Handle>) {
+    fn append_before_sibling(&self, sibling: &Handle, child: NodeOrText<Handle>) {
         let (parent, i) = get_parent_and_index(sibling)
             .expect("append_before_sibling called on node without parent");
 
@@ -349,7 +351,7 @@ impl TreeSink for RcDom {
     }
 
     fn append_based_on_parent_node(
-        &mut self,
+        &self,
         element: &Self::Handle,
         prev_element: &Self::Handle,
         child: NodeOrText<Self::Handle>,
@@ -366,7 +368,7 @@ impl TreeSink for RcDom {
     }
 
     fn append_doctype_to_document(
-        &mut self,
+        &self,
         name: StrTendril,
         public_id: StrTendril,
         system_id: StrTendril,
@@ -381,7 +383,7 @@ impl TreeSink for RcDom {
         );
     }
 
-    fn add_attrs_if_missing(&mut self, target: &Handle, attrs: Vec<Attribute>) {
+    fn add_attrs_if_missing(&self, target: &Handle, attrs: Vec<Attribute>) {
         let mut existing = if let NodeData::Element { ref attrs, .. } = target.data {
             attrs.borrow_mut()
         } else {
@@ -399,11 +401,11 @@ impl TreeSink for RcDom {
         );
     }
 
-    fn remove_from_parent(&mut self, target: &Handle) {
+    fn remove_from_parent(&self, target: &Handle) {
         remove_from_parent(target);
     }
 
-    fn reparent_children(&mut self, node: &Handle, new_parent: &Handle) {
+    fn reparent_children(&self, node: &Handle, new_parent: &Handle) {
         let mut children = node.children.borrow_mut();
         let mut new_children = new_parent.children.borrow_mut();
         for child in children.iter() {
@@ -433,8 +435,8 @@ impl Default for RcDom {
     fn default() -> RcDom {
         RcDom {
             document: Node::new(NodeData::Document),
-            errors: vec![],
-            quirks_mode: tree_builder::NoQuirks,
+            errors: vec![].into(),
+            quirks_mode: tree_builder::NoQuirks.into(),
         }
     }
 }