Skip unquote allocation for non-quoted strings (#11813)

## Summary

Small optimization: no need to unquote if there aren't any quote
characters.

Cargo.lock

@@ -5531,6 +5531,7 @@ dependencies = [
  "indoc",
  "insta",
  "itertools 0.14.0",
+ "memchr",
  "regex",
  "reqwest",
  "reqwest-middleware",

crates/uv-requirements-txt/Cargo.toml

@@ -26,7 +26,7 @@ uv-pypi-types = { workspace = true }
 uv-warnings = { workspace = true }
 
 fs-err = { workspace = true }
-regex = { workspace = true }
+memchr = { workspace = true }
 reqwest = { workspace = true, optional = true }
 reqwest-middleware = { workspace = true, optional = true }
 thiserror = { workspace = true }
@@ -43,6 +43,7 @@ assert_fs = { version = "1.1.2" }
 indoc = { workspace = true }
 insta = { version = "1.40.0", features = ["filters"] }
 itertools = { version = "0.14.0" }
+regex = { workspace = true }
 tempfile = { workspace = true }
 test-case = { version = "3.3.1" }
 tokio = { version = "1.40.0" }

crates/uv-requirements-txt/src/lib.rs

@@ -523,7 +523,10 @@ fn parse_entry(
     let start = s.cursor();
     Ok(Some(if s.eat_if("-r") || s.eat_if("--requirement") {
         let filename = parse_value(content, s, |c: char| !is_terminal(c))?;
-        let filename = unquote(filename).unwrap_or_else(|_| filename.to_string());
+        let filename = unquote(filename)
+            .ok()
+            .flatten()
+            .unwrap_or_else(|| filename.to_string());
         let end = s.cursor();
         RequirementsTxtStatement::Requirements {
             filename,
@@ -532,7 +535,10 @@ fn parse_entry(
         }
     } else if s.eat_if("-c") || s.eat_if("--constraint") {
         let filename = parse_value(content, s, |c: char| !is_terminal(c))?;
-        let filename = unquote(filename).unwrap_or_else(|_| filename.to_string());
+        let filename = unquote(filename)
+            .ok()
+            .flatten()
+            .unwrap_or_else(|| filename.to_string());
         let end = s.cursor();
         RequirementsTxtStatement::Constraint {
             filename,
@@ -577,6 +583,8 @@ fn parse_entry(
     } else if s.eat_if("-i") || s.eat_if("--index-url") {
         let given = parse_value(content, s, |c: char| !is_terminal(c))?;
         let given = unquote(given)
+            .ok()
+            .flatten()
             .map(Cow::Owned)
             .unwrap_or(Cow::Borrowed(given));
         let expanded = expand_env_vars(given.as_ref());
@@ -606,6 +614,8 @@ fn parse_entry(
     } else if s.eat_if("--extra-index-url") {
         let given = parse_value(content, s, |c: char| !is_terminal(c))?;
         let given = unquote(given)
+            .ok()
+            .flatten()
             .map(Cow::Owned)
             .unwrap_or(Cow::Borrowed(given));
         let expanded = expand_env_vars(given.as_ref());
@@ -637,6 +647,8 @@ fn parse_entry(
     } else if s.eat_if("--find-links") || s.eat_if("-f") {
         let given = parse_value(content, s, |c: char| !is_terminal(c))?;
         let given = unquote(given)
+            .ok()
+            .flatten()
             .map(Cow::Owned)
             .unwrap_or(Cow::Borrowed(given));
         let expanded = expand_env_vars(given.as_ref());
@@ -666,6 +678,8 @@ fn parse_entry(
     } else if s.eat_if("--no-binary") {
         let given = parse_value(content, s, |c: char| !is_terminal(c))?;
         let given = unquote(given)
+            .ok()
+            .flatten()
             .map(Cow::Owned)
             .unwrap_or(Cow::Borrowed(given));
         let specifier = PackageNameSpecifier::from_str(given.as_ref()).map_err(|err| {
@@ -680,6 +694,8 @@ fn parse_entry(
     } else if s.eat_if("--only-binary") {
         let given = parse_value(content, s, |c: char| !is_terminal(c))?;
         let given = unquote(given)
+            .ok()
+            .flatten()
             .map(Cow::Owned)
             .unwrap_or(Cow::Borrowed(given));
         let specifier = PackageNameSpecifier::from_str(given.as_ref()).map_err(|err| {

crates/uv-requirements-txt/src/shquote.rs

@@ -142,12 +142,20 @@ fn unquote_open_escape(acc: &mut String, cursor: &mut std::iter::Enumerate<std::
 ///
 /// The result is canonical. There is only one valid unquoted result for a given input.
 ///
+/// If the string does not require any quoting or escaping, returns `Ok(None)`.
+///
 /// # Examples
 ///
 /// ```
 /// assert_eq!(r_shquote::unquote("foobar").unwrap(), "foobar");
 /// ```
-pub(crate) fn unquote(source: &str) -> Result<String, UnquoteError> {
+pub(crate) fn unquote(source: &str) -> Result<Option<String>, UnquoteError> {
+    // If the string does not contain any single-quotes, double-quotes, or escape sequences, it
+    // does not require any unquoting.
+    if memchr::memchr3(b'\'', b'"', b'\\', source.as_bytes()).is_none() {
+        return Ok(None);
+    }
+
     // An unquote-operation never results in a longer string. Furthermore, the common case is
     // most of the string is unquoted / unescaped. Hence, we simply allocate the same space
     // for the resulting string as the input.
@@ -182,7 +190,7 @@ pub(crate) fn unquote(source: &str) -> Result<String, UnquoteError> {
                 acc.push(next_ch);
             }
             None => {
-                break Ok(acc);
+                break Ok(Some(acc));
             }
         }
     }
@@ -194,10 +202,10 @@ mod tests {
 
     #[test]
     fn basic() {
-        assert_eq!(unquote("foobar").unwrap(), "foobar");
+        assert_eq!(unquote("foobar").unwrap(), None);
-        assert_eq!(unquote("foo'bar'").unwrap(), "foobar");
+        assert_eq!(unquote("foo'bar'").unwrap().unwrap(), "foobar");
-        assert_eq!(unquote("foo\"bar\"").unwrap(), "foobar");
+        assert_eq!(unquote("foo\"bar\"").unwrap().unwrap(), "foobar");
-        assert_eq!(unquote("\\foobar\\").unwrap(), "foobar");
+        assert_eq!(unquote("\\foobar\\").unwrap().unwrap(), "foobar");
-        assert_eq!(unquote("\\'foobar\\'").unwrap(), "'foobar'");
+        assert_eq!(unquote("\\'foobar\\'").unwrap().unwrap(), "'foobar'");
     }
 }