try aarch64 simd jetscii with const new.

Author: kivikakk

Cargo.lock

@@ -43,7 +43,8 @@ arbitrary = { version = "1", optional = true, features = ["derive"] }
 bon = { version = "3", optional = true }
 caseless = "0.2.1"
 fmt2io = { version = "1.0.0", optional = true }
-jetscii = "0.5.3"
+# jetscii = "0.5.3"
+jetscii = { git = "https://github.com/kivikakk/jetscii", branch = "aarch64_simd_const" }
 
 [dev-dependencies]
 ntest = "0.9"

Cargo.toml

@@ -59,7 +59,7 @@ bench-all: binaries
 benches/samply-bench-input.md:
 	cat ${ROOT}/vendor/progit/*/*/*.markdown > $@
 
-SAMPLY_OPTIONS:=-r 10000 --iteration-count 20 --reuse-threads
+SAMPLY_OPTIONS:=-r 10000 --iteration-count 40 --reuse-threads
 SAMPLY_COMRAK_ARGS:=benches/samply-bench-input.md -o /dev/null
 
 samply-comrak-branch: benches/samply-bench-input.md build-comrak-branch

Makefile

@@ -1602,10 +1602,10 @@ fn tagfilter(literal: &str) -> bool {
 
 fn tagfilter_block(output: &mut dyn Write, buffer: &str) -> fmt::Result {
     let bytes = buffer.as_bytes();
-    let matcher = jetscii::bytes!(b'<');
+    const MATCHER: jetscii::BytesConst = jetscii::bytes!(b'<');
 
     let mut offset = 0;
-    while let Some(i) = matcher.find(&bytes[offset..]) {
+    while let Some(i) = MATCHER.find(&bytes[offset..]) {
         output.write_str(&buffer[offset..offset + i])?;
         if tagfilter(&buffer[offset + i..]) {
             output.write_str("&lt;")?;
@@ -1638,10 +1638,10 @@ pub fn dangerous_url(input: &str) -> bool {
 /// URLs in attributes.  See escape_href.
 pub fn escape(output: &mut dyn Write, buffer: &str) -> fmt::Result {
     let bytes = buffer.as_bytes();
-    let matcher = jetscii::bytes!(b'"', b'&', b'<', b'>');
+    const MATCHER: jetscii::BytesConst = jetscii::bytes!(b'"', b'&', b'<', b'>');
 
     let mut offset = 0;
-    while let Some(i) = matcher.find(&bytes[offset..]) {
+    while let Some(i) = MATCHER.find(&bytes[offset..]) {
         let esc: &str = match bytes[offset + i] {
             b'"' => "&quot;",
             b'&' => "&amp;",

src/html.rs

@@ -2485,10 +2485,10 @@ where
 
                     let matches_end_condition = match block_type {
                         1 => scanners::html_block_end_1(&line[self.first_nonspace..]),
-                        2 => scanners::html_block_end_2(&line[self.first_nonspace..]),
-                        3 => scanners::html_block_end_3(&line[self.first_nonspace..]),
-                        4 => scanners::html_block_end_4(&line[self.first_nonspace..]),
-                        5 => scanners::html_block_end_5(&line[self.first_nonspace..]),
+                        2 => Self::html_block_end_2(&line[self.first_nonspace..]),
+                        3 => Self::html_block_end_3(&line[self.first_nonspace..]),
+                        4 => Self::html_block_end_4(&line[self.first_nonspace..]),
+                        5 => Self::html_block_end_5(&line[self.first_nonspace..]),
                         _ => false,
                     };
 
@@ -2543,6 +2543,26 @@ where
         }
     }
 
+    fn html_block_end_2(buf: &str) -> bool {
+        const MATCHER: jetscii::Substring = jetscii::Substring::new("-->");
+        MATCHER.find(buf).is_some()
+    }
+
+    fn html_block_end_3(buf: &str) -> bool {
+        const MATCHER: jetscii::Substring = jetscii::Substring::new("?>");
+        MATCHER.find(buf).is_some()
+    }
+
+    fn html_block_end_4(buf: &str) -> bool {
+        const MATCHER: jetscii::BytesConst = jetscii::bytes!('>');
+        MATCHER.find(buf.as_bytes()).is_some()
+    }
+
+    fn html_block_end_5(buf: &str) -> bool {
+        const MATCHER: jetscii::Substring = jetscii::Substring::new("]]>");
+        MATCHER.find(buf).is_some()
+    }
+
     fn add_line(&mut self, node: Node<'a>, line: &str) {
         let mut ast = node.data.borrow_mut();
         assert!(ast.open);

src/parser/mod.rs

@@ -82,46 +82,6 @@ pub fn html_block_end_1(s: &str) -> bool {
 */
 }
 
-pub fn html_block_end_2(s: &str) -> bool {
-    let mut cursor = 0;
-    let mut marker = 0;
-    let len = s.len();
-/*!re2c
-    [^\n\x00]* '-->' { return true; }
-    * { return false; }
-*/
-}
-
-pub fn html_block_end_3(s: &str) -> bool {
-    let mut cursor = 0;
-    let mut marker = 0;
-    let len = s.len();
-/*!re2c
-    [^\n\x00]* '?>' { return true; }
-    * { return false; }
-*/
-}
-
-pub fn html_block_end_4(s: &str) -> bool {
-    let mut cursor = 0;
-    let mut marker = 0;
-    let len = s.len();
-/*!re2c
-    [^\n\x00]* '>' { return true; }
-    * { return false; }
-*/
-}
-
-pub fn html_block_end_5(s: &str) -> bool {
-    let mut cursor = 0;
-    let mut marker = 0;
-    let len = s.len();
-/*!re2c
-    [^\n\x00]* ']]>' { return true; }
-    * { return false; }
-*/
-}
-
 pub fn alert_start(s: &str) -> Option<AlertType> {
     let mut cursor = 0;
     let mut marker = 0;