Merge pull request #2696 from paolobarbolini/once-cell-to-std

Replace `once_cell::sync::Lazy` with `std::sync::LazyLock`

Author: ehuss

Cargo.lock

@@ -30,7 +30,6 @@ anyhow = "1.0.71"
 chrono = { version = "0.4.24", default-features = false, features = ["clock"] }
 clap = { version = "4.3.12", features = ["cargo", "wrap_help"] }
 clap_complete = "4.3.2"
-once_cell = "1.17.1"
 env_logger = "0.11.1"
 handlebars = "6.0"
 hex = "0.4.3"

Cargo.toml

@@ -1,11 +1,10 @@
 use regex::Regex;
-use std::path::Path;
+use std::{path::Path, sync::LazyLock};
 
 use super::{Preprocessor, PreprocessorContext};
 use crate::book::{Book, BookItem};
 use crate::errors::*;
 use log::warn;
-use once_cell::sync::Lazy;
 
 /// A preprocessor for converting file name `README.md` to `index.md` since
 /// `README.md` is the de facto index file in markdown-based documentation.
@@ -68,7 +67,7 @@ fn warn_readme_name_conflict<P: AsRef<Path>>(readme_path: P, index_path: P) {
 }
 
 fn is_readme_file<P: AsRef<Path>>(path: P) -> bool {
-    static RE: Lazy<Regex> = Lazy::new(|| Regex::new(r"(?i)^readme$").unwrap());
+    static RE: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"(?i)^readme$").unwrap());
 
     RE.is_match(
         path.as_ref()

src/preprocess/index.rs

@@ -7,11 +7,11 @@ use regex::{CaptureMatches, Captures, Regex};
 use std::fs;
 use std::ops::{Bound, Range, RangeBounds, RangeFrom, RangeFull, RangeTo};
 use std::path::{Path, PathBuf};
+use std::sync::LazyLock;
 
 use super::{Preprocessor, PreprocessorContext};
 use crate::book::{Book, BookItem};
 use log::{error, warn};
-use once_cell::sync::Lazy;
 
 const ESCAPE_CHAR: char = '\\';
 const MAX_LINK_NESTED_DEPTH: usize = 10;
@@ -409,7 +409,7 @@ impl<'a> Iterator for LinkIter<'a> {
 fn find_links(contents: &str) -> LinkIter<'_> {
     // lazily compute following regex
     // r"\\\{\{#.*\}\}|\{\{#([a-zA-Z0-9]+)\s*([^}]+)\}\}")?;
-    static RE: Lazy<Regex> = Lazy::new(|| {
+    static RE: LazyLock<Regex> = LazyLock::new(|| {
         Regex::new(
             r"(?x)              # insignificant whitespace mode
         \\\{\{\#.*\}\}      # match escaped link

src/preprocess/links.rs

@@ -12,11 +12,11 @@ use std::collections::BTreeMap;
 use std::collections::HashMap;
 use std::fs::{self, File};
 use std::path::{Path, PathBuf};
+use std::sync::LazyLock;
 
 use crate::utils::fs::get_404_output_file;
 use handlebars::Handlebars;
 use log::{debug, trace, warn};
-use once_cell::sync::Lazy;
 use regex::{Captures, Regex};
 use serde_json::json;
 
@@ -664,7 +664,7 @@ fn make_data(
 /// Goes through the rendered HTML, making sure all header tags have
 /// an anchor respectively so people can link to sections directly.
 fn build_header_links(html: &str) -> String {
-    static BUILD_HEADER_LINKS: Lazy<Regex> = Lazy::new(|| {
+    static BUILD_HEADER_LINKS: LazyLock<Regex> = LazyLock::new(|| {
         Regex::new(r#"<h(\d)(?: id="([^"]+)")?(?: class="([^"]+)")?>(.*?)</h\d>"#).unwrap()
     });
     static IGNORE_CLASS: &[&str] = &["menu-title"];
@@ -725,8 +725,8 @@ fn insert_link_into_header(
 // ```
 // This function replaces all commas by spaces in the code block classes
 fn fix_code_blocks(html: &str) -> String {
-    static FIX_CODE_BLOCKS: Lazy<Regex> =
-        Lazy::new(|| Regex::new(r##"<code([^>]+)class="([^"]+)"([^>]*)>"##).unwrap());
+    static FIX_CODE_BLOCKS: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r##"<code([^>]+)class="([^"]+)"([^>]*)>"##).unwrap());
 
     FIX_CODE_BLOCKS
         .replace_all(html, |caps: &Captures<'_>| {
@@ -739,8 +739,8 @@ fn fix_code_blocks(html: &str) -> String {
         .into_owned()
 }
 
-static CODE_BLOCK_RE: Lazy<Regex> =
-    Lazy::new(|| Regex::new(r##"((?s)<code[^>]?class="([^"]+)".*?>(.*?)</code>)"##).unwrap());
+static CODE_BLOCK_RE: LazyLock<Regex> =
+    LazyLock::new(|| Regex::new(r##"((?s)<code[^>]?class="([^"]+)".*?>(.*?)</code>)"##).unwrap());
 
 fn add_playground_pre(
     html: &str,
@@ -808,8 +808,10 @@ fn add_playground_pre(
 /// Modifies all `<code>` blocks to convert "hidden" lines and to wrap them in
 /// a `<span class="boring">`.
 fn hide_lines(html: &str, code_config: &Code) -> String {
-    static LANGUAGE_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new(r"\blanguage-(\w+)\b").unwrap());
-    static HIDELINES_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new(r"\bhidelines=(\S+)").unwrap());
+    static LANGUAGE_REGEX: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"\blanguage-(\w+)\b").unwrap());
+    static HIDELINES_REGEX: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"\bhidelines=(\S+)").unwrap());
 
     CODE_BLOCK_RE
         .replace_all(html, |caps: &Captures<'_>| {
@@ -850,7 +852,8 @@ fn hide_lines(html: &str, code_config: &Code) -> String {
 }
 
 fn hide_lines_rust(content: &str) -> String {
-    static BORING_LINES_REGEX: Lazy<Regex> = Lazy::new(|| Regex::new(r"^(\s*)#(.?)(.*)$").unwrap());
+    static BORING_LINES_REGEX: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"^(\s*)#(.?)(.*)$").unwrap());
 
     let mut result = String::with_capacity(content.len());
     let mut lines = content.lines().peekable();

src/renderer/html_handlebars/hbs_renderer.rs

@@ -1,9 +1,9 @@
 use std::borrow::Cow;
 use std::collections::{HashMap, HashSet};
 use std::path::{Path, PathBuf};
+use std::sync::LazyLock;
 
 use elasticlunr::{Index, IndexBuilder};
-use once_cell::sync::Lazy;
 use pulldown_cmark::*;
 
 use crate::book::{Book, BookItem, Chapter};
@@ -314,7 +314,7 @@ fn write_to_json(index: Index, search_config: &Search, doc_urls: Vec<String>) ->
 }
 
 fn clean_html(html: &str) -> String {
-    static AMMONIA: Lazy<ammonia::Builder<'static>> = Lazy::new(|| {
+    static AMMONIA: LazyLock<ammonia::Builder<'static>> = LazyLock::new(|| {
         let mut clean_content = HashSet::new();
         clean_content.insert("script");
         clean_content.insert("style");

src/renderer/html_handlebars/search.rs

@@ -1,7 +1,6 @@
 //! Support for writing static files.
 
 use log::{debug, warn};
-use once_cell::sync::Lazy;
 
 use crate::config::HtmlConfig;
 use crate::errors::*;
@@ -13,6 +12,7 @@ use std::borrow::Cow;
 use std::collections::HashMap;
 use std::fs::{self, File};
 use std::path::{Path, PathBuf};
+use std::sync::LazyLock;
 
 /// Map static files to their final names and contents.
 ///
@@ -231,8 +231,8 @@ impl StaticFiles {
         use regex::bytes::{Captures, Regex};
         // The `{{ resource "name" }}` directive in static resources look like
         // handlebars syntax, even if they technically aren't.
-        static RESOURCE: Lazy<Regex> =
-            Lazy::new(|| Regex::new(r#"\{\{ resource "([^"]+)" \}\}"#).unwrap());
+        static RESOURCE: LazyLock<Regex> =
+            LazyLock::new(|| Regex::new(r#"\{\{ resource "([^"]+)" \}\}"#).unwrap());
         fn replace_all<'a>(
             hash_map: &HashMap<String, String>,
             data: &'a [u8],

src/renderer/html_handlebars/static_files.rs

@@ -5,14 +5,14 @@ mod string;
 pub(crate) mod toml_ext;
 use crate::errors::Error;
 use log::error;
-use once_cell::sync::Lazy;
 use pulldown_cmark::{html, CodeBlockKind, CowStr, Event, Options, Parser, Tag, TagEnd};
 use regex::Regex;
 
 use std::borrow::Cow;
 use std::collections::HashMap;
 use std::fmt::Write;
 use std::path::Path;
+use std::sync::LazyLock;
 
 pub use self::string::{
     take_anchored_lines, take_lines, take_rustdoc_include_anchored_lines,
@@ -21,7 +21,7 @@ pub use self::string::{
 
 /// Replaces multiple consecutive whitespace characters with a single space character.
 pub fn collapse_whitespace(text: &str) -> Cow<'_, str> {
-    static RE: Lazy<Regex> = Lazy::new(|| Regex::new(r"\s\s+").unwrap());
+    static RE: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"\s\s+").unwrap());
     RE.replace_all(text, " ")
 }
 
@@ -50,7 +50,7 @@ pub fn id_from_content(content: &str) -> String {
     let mut content = content.to_string();
 
     // Skip any tags or html-encoded stuff
-    static HTML: Lazy<Regex> = Lazy::new(|| Regex::new(r"(<.*?>)").unwrap());
+    static HTML: LazyLock<Regex> = LazyLock::new(|| Regex::new(r"(<.*?>)").unwrap());
     content = HTML.replace_all(&content, "").into();
     const REPL_SUB: &[&str] = &["&lt;", "&gt;", "&amp;", "&#39;", "&quot;"];
     for sub in REPL_SUB {
@@ -93,9 +93,10 @@ pub fn unique_id_from_content(content: &str, id_counter: &mut HashMap<String, us
 /// None. Ideally, print page links would link to anchors on the print page,
 /// but that is very difficult.
 fn adjust_links<'a>(event: Event<'a>, path: Option<&Path>) -> Event<'a> {
-    static SCHEME_LINK: Lazy<Regex> = Lazy::new(|| Regex::new(r"^[a-z][a-z0-9+.-]*:").unwrap());
-    static MD_LINK: Lazy<Regex> =
-        Lazy::new(|| Regex::new(r"(?P<link>.*)\.md(?P<anchor>#.*)?").unwrap());
+    static SCHEME_LINK: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"^[a-z][a-z0-9+.-]*:").unwrap());
+    static MD_LINK: LazyLock<Regex> =
+        LazyLock::new(|| Regex::new(r"(?P<link>.*)\.md(?P<anchor>#.*)?").unwrap());
 
     fn fix<'a>(dest: CowStr<'a>, path: Option<&Path>) -> CowStr<'a> {
         if dest.starts_with('#') {
@@ -148,8 +149,8 @@ fn adjust_links<'a>(event: Event<'a>, path: Option<&Path>) -> Event<'a> {
         // There are dozens of HTML tags/attributes that contain paths, so
         // feel free to add more tags if desired; these are the only ones I
         // care about right now.
-        static HTML_LINK: Lazy<Regex> =
-            Lazy::new(|| Regex::new(r#"(<(?:a|img) [^>]*?(?:src|href)=")([^"]+?)""#).unwrap());
+        static HTML_LINK: LazyLock<Regex> =
+            LazyLock::new(|| Regex::new(r#"(<(?:a|img) [^>]*?(?:src|href)=")([^"]+?)""#).unwrap());
 
         HTML_LINK
             .replace_all(&html, |caps: &regex::Captures<'_>| {

src/utils/mod.rs

@@ -1,7 +1,7 @@
-use once_cell::sync::Lazy;
 use regex::Regex;
 use std::ops::Bound::{Excluded, Included, Unbounded};
 use std::ops::RangeBounds;
+use std::sync::LazyLock;
 
 /// Take a range of lines from a string.
 pub fn take_lines<R: RangeBounds<usize>>(s: &str, range: R) -> String {
@@ -24,10 +24,10 @@ pub fn take_lines<R: RangeBounds<usize>>(s: &str, range: R) -> String {
     }
 }
 
-static ANCHOR_START: Lazy<Regex> =
-    Lazy::new(|| Regex::new(r"ANCHOR:\s*(?P<anchor_name>[\w_-]+)").unwrap());
-static ANCHOR_END: Lazy<Regex> =
-    Lazy::new(|| Regex::new(r"ANCHOR_END:\s*(?P<anchor_name>[\w_-]+)").unwrap());
+static ANCHOR_START: LazyLock<Regex> =
+    LazyLock::new(|| Regex::new(r"ANCHOR:\s*(?P<anchor_name>[\w_-]+)").unwrap());
+static ANCHOR_END: LazyLock<Regex> =
+    LazyLock::new(|| Regex::new(r"ANCHOR_END:\s*(?P<anchor_name>[\w_-]+)").unwrap());
 
 /// Take anchored lines from a string.
 /// Lines containing anchor are ignored.