Split atoms out into a separate crate (#599)

Signed-off-by: Nico Burns <nico@nicoburns.com>

Author: nicoburns

Cargo.toml

@@ -1,5 +1,6 @@
 [workspace]
 members = [
+    "web_atoms",
     "markup5ever",
     "html5ever",
     "rcdom",

html5ever/src/serialize/mod.rs

@@ -9,7 +9,7 @@
 
 use log::warn;
 pub use markup5ever::serialize::{AttrRef, Serialize, Serializer, TraversalScope};
-use markup5ever::{local_name, namespace_url, ns};
+use markup5ever::{local_name, ns};
 use std::io::{self, Write};
 
 use crate::{LocalName, QualName};

html5ever/src/tokenizer/mod.rs

@@ -25,7 +25,7 @@ use crate::util::str::lower_ascii_letter;
 
 use log::{debug, trace};
 use mac::format_if;
-use markup5ever::{namespace_url, ns, small_char_set, TokenizerResult};
+use markup5ever::{ns, small_char_set, TokenizerResult};
 use std::borrow::Cow::{self, Borrowed};
 use std::cell::{Cell, RefCell, RefMut};
 use std::collections::BTreeMap;

html5ever/src/tree_builder/mod.rs

@@ -33,7 +33,7 @@ use crate::tree_builder::tag_sets::*;
 use crate::util::str::to_escaped_string;
 use log::{debug, log_enabled, warn, Level};
 use mac::format_if;
-use markup5ever::{expanded_name, local_name, namespace_prefix, namespace_url, ns};
+use markup5ever::{expanded_name, local_name, namespace_prefix, ns};
 
 #[macro_use]
 mod tag_sets;

html5ever/src/tree_builder/rules.rs

@@ -19,7 +19,7 @@ use crate::tree_builder::{
     TreeSink,
 };
 use crate::QualName;
-use markup5ever::{expanded_name, local_name, namespace_url, ns};
+use markup5ever::{expanded_name, local_name, ns};
 use std::borrow::Cow::Borrowed;
 
 use crate::tendril::SliceExt;

html5ever/src/tree_builder/tag_sets.rs

@@ -10,7 +10,7 @@
 //! Various sets of HTML tag names, and macros for declaring them.
 
 use crate::ExpandedName;
-use markup5ever::{expanded_name, local_name, namespace_url, ns};
+use markup5ever::{expanded_name, local_name, ns};
 
 macro_rules! declare_tag_set_impl ( ($param:ident, $b:ident, $supr:ident, $($tag:tt)+) => (
     match $param {

markup5ever/Cargo.toml

@@ -6,7 +6,6 @@ license = "MIT OR Apache-2.0"
 repository = "https://github.com/servo/html5ever"
 description = "Common code for xml5ever and html5ever"
 documentation = "https://docs.rs/markup5ever"
-build = "build.rs"
 categories = [ "parser-implementations", "web-programming" ]
 edition = "2021"
 rust-version.workspace = true
@@ -15,11 +14,6 @@ rust-version.workspace = true
 path = "lib.rs"
 
 [dependencies]
-string_cache = "0.8"
-phf = "0.11"
+web_atoms = { version = "0.1", path = "../web_atoms" }
 tendril = "0.4"
-log = "0.4"
-
-[build-dependencies]
-string_cache_codegen = "0.5.4"
-phf_codegen = "0.11"
+log = "0.4"

markup5ever/interface/mod.rs

@@ -11,11 +11,11 @@
 use std::cell::Ref;
 use std::fmt;
 use tendril::StrTendril;
+use web_atoms::{LocalName, Namespace, Prefix};
 
 pub use self::tree_builder::{create_element, AppendNode, AppendText, ElementFlags, NodeOrText};
 pub use self::tree_builder::{ElemName, Tracer, TreeSink};
 pub use self::tree_builder::{LimitedQuirks, NoQuirks, Quirks, QuirksMode};
-use super::{LocalName, Namespace, Prefix};
 
 /// An [expanded name], containing the tag and the namespace.
 ///
@@ -165,8 +165,7 @@ pub mod tree_builder;
 ///  prefix (when resolved gives namespace_url `https://furniture.rs`)
 /// ```
 ///
-/// NOTE: `Prefix`, `LocalName` and `Prefix` are all derivative of
-/// `string_cache::atom::Atom` and `Atom` implements `Deref<str>`.
+/// NOTE: `Prefix`, `LocalName` and `Prefix` all implement `Deref<str>`.
 ///
 #[derive(PartialEq, Eq, PartialOrd, Ord, Hash, Debug, Clone)]
 #[cfg_attr(feature = "heap_size", derive(HeapSizeOf))]
@@ -401,7 +400,7 @@ pub struct Attribute {
 
 #[cfg(test)]
 mod tests {
-    use super::Namespace;
+    use web_atoms::{ns, Namespace};
 
     #[test]
     fn ns_macro() {

markup5ever/interface/tree_builder.rs

@@ -12,10 +12,10 @@
 //! It can be used by a parser to create the DOM graph structure in memory.
 
 use crate::interface::{Attribute, ExpandedName, QualName};
-use crate::{LocalName, Namespace};
 use std::borrow::Cow;
 use std::fmt::Debug;
 use tendril::StrTendril;
+use web_atoms::{LocalName, Namespace};
 
 pub use self::NodeOrText::{AppendNode, AppendText};
 pub use self::QuirksMode::{LimitedQuirks, NoQuirks, Quirks};

markup5ever/lib.rs

@@ -13,6 +13,9 @@
 
 pub use tendril;
 
+#[macro_use]
+extern crate web_atoms;
+
 /// Create a [`SmallCharSet`], with each space-separated number stored in the set.
 ///
 /// # Examples
@@ -34,9 +37,15 @@ macro_rules! small_char_set ( ($($e:expr)+) => (
     }
 ));
 
-include!(concat!(env!("OUT_DIR"), "/generated.rs"));
+pub use web_atoms::{
+    local_name, namespace_prefix, namespace_url, ns, LocalName, LocalNameStaticSet, Namespace,
+    NamespaceStaticSet, Prefix, PrefixStaticSet,
+};
 
-pub mod data;
+pub mod data {
+    pub use web_atoms::C1_REPLACEMENTS;
+    pub use web_atoms::NAMED_ENTITIES;
+}
 #[macro_use]
 pub mod interface;
 pub mod serialize;