Minor tweaks and fixes

.rustfmt.toml

@@ -0,0 +1 @@
+# reset user settings to default

Cargo.toml

@@ -9,15 +9,7 @@ documentation = "https://docs.rs/lol-html"
 categories = ["parser-implementations", "web-programming"]
 keywords = ["html", "css-selectors", "parser", "rewriter", "streaming"]
 readme = "README.md"
-include = [
-    "/Cargo.toml",
-    "/LICENSE",
-    "/README.md",
-    "/src",
-    "/tests",
-    "/media",
-    "/benches"
-]
+include = ["/Cargo.toml", "/LICENSE", "/README.md", "/media", "/src"]
 autotests = false
 edition = "2021"
 

benches/cases/rewriting.rs

@@ -1,5 +1,5 @@
-use lol_html::html_content::*;
-use lol_html::*;
+use lol_html::html_content::ContentType;
+use lol_html::{element, Settings};
 
 define_group!(
     "Rewriting",

benches/cases/selector_matching.rs

@@ -1,4 +1,4 @@
-use lol_html::*;
+use lol_html::{element, Settings};
 
 define_group!(
     "Selector matching",

c-api/Cargo.toml

@@ -1,9 +1,9 @@
 [package]
 name = "lolhtml"
-version = "1.1.1"
+version = "1.1.2"
 authors = ["Ivan Nikulin <[email protected]>", "Joshua Nelson <[email protected]>"]
 edition = "2021"
-
+links = "lolhtml"
 publish = false
 
 [dependencies]

c-api/build.rs

@@ -0,0 +1,2 @@
+// Required for the links attribute
+fn main() {}

c-api/tests/build.rs

@@ -53,6 +53,8 @@ fn main() {
     }
 
     println!("cargo:rerun-if-changed=../include/lol_html.h");
+    println!("cargo:rerun-if-changed=src");
+    println!("cargo:rerun-if-changed=build.rs");
 
     // Collect all the C files from src/deps/picotest and src.
     let mut c_files = glob_c_files(PICOTEST_DIR);
@@ -62,15 +64,11 @@ fn main() {
     build
         .debug(true)
         .opt_level(0)
-        .flag_if_supported("-Wl,no-as-needed")
         .warnings(true)
         .extra_warnings(true)
         .warnings_into_errors(true)
         .include(INCLUDE_DIR)
         .include(PICOTEST_DIR)
         .files(c_files)
         .compile("lol_html_ctests");
-
-    // Link against the C API.
-    println!("cargo:rustc-link-lib=dylib=lolhtml");
 }

c-api/tests/src/main.rs

@@ -1,5 +1,8 @@
 //! The test runner for the C API tests.
 
+// ensure it's linked
+use lolhtml as _;
+
 extern "C" {
     fn run_tests() -> i32;
 }

js-api/src/lib.rs

@@ -15,6 +15,7 @@ struct Anchor<'r> {
 }
 
 impl<'r> Anchor<'r> {
+    #[inline]
     pub fn new(poisoned: Rc<Cell<bool>>) -> Self {
         Anchor {
             poisoned,

src/base/bytes.rs

@@ -63,6 +63,12 @@ impl<'b> Bytes<'b> {
         self.0[range.start..range.end].into()
     }
 
+    #[inline]
+    pub fn split_at(&self, pos: usize) -> (Bytes<'_>, Bytes<'_>) {
+        let (before, after) = self.0.split_at(pos);
+        (Bytes::from(before), Bytes::from(after))
+    }
+
     #[inline]
     pub fn opt_slice(&self, range: Option<Range>) -> Option<Bytes> {
         range.map(|range| self.slice(range))
@@ -165,6 +171,7 @@ impl<'b> From<&'b [u8]> for Bytes<'b> {
 }
 
 impl Debug for Bytes<'_> {
+    #[cold]
     fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
         write!(f, "`{}`", self.as_debug_string())
     }
@@ -173,6 +180,7 @@ impl Debug for Bytes<'_> {
 impl Deref for Bytes<'_> {
     type Target = [u8];
 
+    #[inline]
     fn deref(&self) -> &[u8] {
         &self.0
     }

src/base/debug_trace.rs

@@ -47,11 +47,11 @@ cfg_if! {
                 println!("{:#?}", $bookmark);
                 println!("Parser directive: `{:#?}`", $parser_directive);
 
-                let mut chunk_str = Bytes::from($chunk).as_debug_string();
+                // as_debug_string() is UTF-8, and the position for the input encoding is not guaranteed to match it
+                let chunk = Bytes::from($chunk);
+                let (before, after) = chunk.split_at($bookmark.pos);
 
-                chunk_str.insert_str($bookmark.pos, "|*|");
-
-                println!("Bookmark start: `{}`", chunk_str);
+                println!("Bookmark start: `{}|*|{}`", before.as_debug_string(), after.as_debug_string());
                 println!();
             };
 

src/base/encoding.rs

@@ -4,7 +4,7 @@ use std::sync::atomic::{AtomicUsize, Ordering};
 use std::sync::Arc;
 
 /// This serves as a map from integer to [`Encoding`], which allows more efficient
-/// sets/gets of the [SharedEncoding].
+/// sets/gets of the [`SharedEncoding`].
 static ALL_ENCODINGS: [&Encoding; 228] = [
     &encoding_rs::WINDOWS_1252_INIT,
     &encoding_rs::ISO_8859_2_INIT,
@@ -247,21 +247,23 @@ fn encoding_to_index(encoding: AsciiCompatibleEncoding) -> usize {
 
 /// A charset encoding that can be shared and modified.
 ///
-/// This is, for instance, used to adapt the charset dynamically in a [crate::HtmlRewriter] if it
+/// This is, for instance, used to adapt the charset dynamically in a [`crate::HtmlRewriter`] if it
 /// encounters a `meta` tag that specifies the charset (that behavior is dependent on
-/// [crate::Settings::adjust_charset_on_meta_tag]).
+/// [`crate::Settings::adjust_charset_on_meta_tag`]).
 #[derive(Clone)]
 pub struct SharedEncoding {
     encoding: Arc<AtomicUsize>,
 }
 
 impl SharedEncoding {
-    pub fn new(encoding: AsciiCompatibleEncoding) -> SharedEncoding {
-        SharedEncoding {
+    #[must_use]
+    pub fn new(encoding: AsciiCompatibleEncoding) -> Self {
+        Self {
             encoding: Arc::new(AtomicUsize::new(encoding_to_index(encoding))),
         }
     }
 
+    #[must_use]
     pub fn get(&self) -> &'static Encoding {
         let encoding = self.encoding.load(Ordering::Relaxed);
         ALL_ENCODINGS[encoding]

src/html/local_name.rs

@@ -30,12 +30,14 @@ pub struct LocalNameHash(Option<u64>);
 
 impl LocalNameHash {
     #[inline]
-    pub fn new() -> Self {
-        LocalNameHash(Some(0))
+    #[must_use]
+    pub const fn new() -> Self {
+        Self(Some(0))
     }
 
     #[inline]
-    pub fn is_empty(&self) -> bool {
+    #[must_use]
+    pub const fn is_empty(&self) -> bool {
         self.0.is_none()
     }
 
@@ -75,7 +77,7 @@ impl LocalNameHash {
 impl From<&str> for LocalNameHash {
     #[inline]
     fn from(string: &str) -> Self {
-        let mut hash = LocalNameHash::new();
+        let mut hash = Self::new();
 
         for ch in string.bytes() {
             hash.update(ch);
@@ -95,7 +97,7 @@ impl PartialEq<Tag> for LocalNameHash {
     }
 }
 
-/// LocalName is used for the comparison of tag names.
+/// `LocalName` is used for the comparison of tag names.
 /// In the majority of cases it will be represented as a hash, however for long
 /// non-standard tag names it fallsback to the Name representation.
 #[derive(Clone, Debug, Eq, Hash)]
@@ -106,6 +108,7 @@ pub enum LocalName<'i> {
 
 impl<'i> LocalName<'i> {
     #[inline]
+    #[must_use]
     pub fn new(input: &'i Bytes<'i>, range: Range, hash: LocalNameHash) -> Self {
         if hash.is_empty() {
             LocalName::Bytes(input.slice(range))
@@ -115,6 +118,7 @@ impl<'i> LocalName<'i> {
     }
 
     #[inline]
+    #[must_use]
     pub fn into_owned(self) -> LocalName<'static> {
         match self {
             LocalName::Bytes(b) => LocalName::Bytes(b.into_owned()),
@@ -142,15 +146,15 @@ impl PartialEq<Tag> for LocalName<'_> {
     fn eq(&self, tag: &Tag) -> bool {
         match self {
             LocalName::Hash(h) => h == tag,
-            _ => false,
+            LocalName::Bytes(_) => false,
         }
     }
 }
 
 impl PartialEq<LocalName<'_>> for LocalName<'_> {
     #[inline]
     fn eq(&self, other: &LocalName<'_>) -> bool {
-        use LocalName::*;
+        use LocalName::{Bytes, Hash};
 
         match (self, other) {
             (Hash(s), Hash(o)) => s == o,

src/html/namespace.rs

@@ -8,8 +8,8 @@ pub enum Namespace {
 
 impl Namespace {
     #[inline]
-    pub fn uri(self) -> &'static str {
-        use Namespace::*;
+    pub const fn uri(self) -> &'static str {
+        use Namespace::{Html, MathML, Svg};
 
         // NOTE: https://infra.spec.whatwg.org/#namespaces
         match self {

src/html/text_type.rs

@@ -41,8 +41,9 @@ impl TextType {
     ///
     /// [HTML entities]: https://developer.mozilla.org/en-US/docs/Glossary/Entity
     #[inline]
+    #[must_use]
     pub fn allows_html_entities(self) -> bool {
-        self == TextType::Data || self == TextType::RCData
+        self == Self::Data || self == Self::RCData
     }
 }
 

src/lib.rs

@@ -16,7 +16,8 @@
 //! [Cloudflare Workers]: https://www.cloudflare.com/en-gb/products/cloudflare-workers/
 //! [`HtmlRewriter`]: struct.HtmlRewriter.html
 //! [`rewrite_str`]: fn.rewrite_str.html
-
+#![allow(clippy::default_trait_access)]
+#![allow(clippy::module_name_repetitions)]
 #![cfg_attr(not(any(feature = "integration_test", test)), warn(missing_docs))]
 
 #[macro_use]
@@ -43,7 +44,7 @@ pub use self::rewriter::{
 pub use self::selectors_vm::Selector;
 pub use self::transform_stream::OutputSink;
 
-/// These module contains types to work with [`Send`]able [`HtmlRewriter`](crate::HtmlRewriter)s.
+/// These module contains types to work with [`Send`]able [`HtmlRewriter`]s.
 pub mod send {
     use crate::rewriter::{
         CommentHandlerSend, DoctypeHandlerSend, ElementHandlerSend, EndHandlerSend,
@@ -75,7 +76,7 @@ pub mod send {
     /// [`TextHandler`](crate::TextHandler) for [`Send`]able [`HtmlRewriter`](crate::HtmlRewriter)s.
     pub type TextHandler<'h> = TextHandlerSend<'h>;
 
-    /// [`Element`](crate::Element) for [`Send`]able [`HtmlRewriter`](crate::HtmlRewriter)s.
+    /// [`Element`](crate::rewritable_units::Element) for [`Send`]able [`HtmlRewriter`](crate::HtmlRewriter)s.
     pub type Element<'r, 't> = crate::rewritable_units::Element<'r, 't, SendHandlerTypes>;
 }
 
@@ -151,8 +152,9 @@ pub mod test_utils {
     }
 
     impl Output {
+        #[must_use]
         pub fn new(encoding: &'static Encoding) -> Self {
-            Output {
+            Self {
                 bytes: Vec::default(),
                 encoding,
                 finalizing_chunk_received: false,
@@ -174,7 +176,7 @@ pub mod test_utils {
     }
 
     impl From<Output> for String {
-        fn from(output: Output) -> String {
+        fn from(output: Output) -> Self {
             assert!(
                 output.finalizing_chunk_received,
                 "Finalizing chunk for the output hasn't been received."

src/memory/arena.rs

@@ -12,7 +12,7 @@ impl Arena {
     pub fn new(limiter: SharedMemoryLimiter, preallocated_size: usize) -> Self {
         limiter.preallocate(preallocated_size);
 
-        Arena {
+        Self {
             limiter,
             data: Vec::with_capacity(preallocated_size),
         }

src/memory/limited_vec.rs

@@ -13,8 +13,8 @@ pub struct LimitedVec<T> {
 }
 
 impl<T> LimitedVec<T> {
-    pub fn new(limiter: SharedMemoryLimiter) -> Self {
-        LimitedVec {
+    pub const fn new(limiter: SharedMemoryLimiter) -> Self {
+        Self {
             vec: vec![],
             limiter,
         }
@@ -122,7 +122,7 @@ mod tests {
     #[test]
     fn max_limit() {
         let limiter = SharedMemoryLimiter::new(2);
-        let mut vector: LimitedVec<u8> = LimitedVec::new(limiter.clone());
+        let mut vector: LimitedVec<u8> = LimitedVec::new(limiter);
 
         vector.push(1).unwrap();
         vector.push(2).unwrap();

src/memory/limiter.rs

@@ -17,8 +17,9 @@ pub struct SharedMemoryLimiter {
 }
 
 impl SharedMemoryLimiter {
-    pub fn new(max: usize) -> SharedMemoryLimiter {
-        SharedMemoryLimiter {
+    #[must_use]
+    pub fn new(max: usize) -> Self {
+        Self {
             current_usage: Arc::new(AtomicUsize::new(0)),
             max,
         }

src/parser/lexer/actions.rs

@@ -2,7 +2,7 @@ use super::*;
 use crate::parser::state_machine::StateMachineActions;
 
 use NonTagContentTokenOutline::*;
-use TagTokenOutline::*;
+use TagTokenOutline::{EndTag, StartTag};
 
 // NOTE: use macro instead of the function to make borrow
 // checker happy with range construction inside match arm
@@ -233,8 +233,8 @@ impl<S: LexemeSink> StateMachineActions for Lexer<S> {
     #[inline]
     fn finish_tag_name(&mut self, _context: &mut ParserContext<S>, _input: &[u8]) -> ActionResult {
         match self.current_tag_token {
-            Some(StartTag { ref mut name, .. }) | Some(EndTag { ref mut name, .. }) => {
-                *name = get_token_part_range!(self)
+            Some(StartTag { ref mut name, .. } | EndTag { ref mut name, .. }) => {
+                *name = get_token_part_range!(self);
             }
             _ => unreachable!("Tag should exist at this point"),
         }
@@ -246,12 +246,14 @@ impl<S: LexemeSink> StateMachineActions for Lexer<S> {
     fn update_tag_name_hash(&mut self, _context: &mut ParserContext<S>, input: &[u8]) {
         if let Some(ch) = input.get(self.pos()).copied() {
             match self.current_tag_token {
-                Some(StartTag {
-                    ref mut name_hash, ..
-                })
-                | Some(EndTag {
-                    ref mut name_hash, ..
-                }) => name_hash.update(ch),
+                Some(
+                    StartTag {
+                        ref mut name_hash, ..
+                    }
+                    | EndTag {
+                        ref mut name_hash, ..
+                    },
+                ) => name_hash.update(ch),
                 _ => unreachable!("Tag should exist at this point"),
             }
         }