Merge branch 'main' into no_std

domenukk · domenukk · commit 00efb8fc8249 · 2024-07-17T02:58:29.000+02:00
diff --git a/.github/dependabot.yml b/.github/dependabot.yml
@@ -0,0 +1,15 @@
+version: 2
+updates:
+  - package-ecosystem: "cargo"
+    directory: "/"
+    schedule:
+      interval: "monthly"
+    open-pull-requests-limit: 10
+  - package-ecosystem: "docker"
+    directory: "/"
+    schedule:
+      interval: "monthly"
+  - package-ecosystem: "github-actions"
+    directory: "/"
+    schedule:
+      interval: "monthly"
diff --git a/.github/workflows/codecov.yml b/.github/workflows/codecov.yml
@@ -2,23 +2,23 @@ name: Coverage
 
 on:
   push:
-    branches: ["master"]
+    branches: ["main"]
   pull_request:
 
 jobs:
   coverage:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: dtolnay/rust-toolchain@stable
       - uses: taiki-e/install-action@v2
         with:
           tool: cargo-tarpaulin
       - run: cargo tarpaulin --out Xml
-      - uses: codecov/codecov-action@v3
+      - uses: codecov/codecov-action@v4
         # A codecov token is not needed for public repos if the repo is linked
         # on codecov.io. See https://docs.codecov.com/docs/frequently-asked-questions#where-is-the-repository-upload-token-found
-      - uses: actions/upload-artifact@v2
+      - uses: actions/upload-artifact@v4
         with:
           name: code-coverage-report
           path: cobertura.xml
diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
@@ -2,7 +2,7 @@ name: CI
 
 on:
   push:
-    branches: ["master"]
+    branches: ["main"]
   pull_request:
   merge_group:
     types: [checks_requested]
@@ -31,7 +31,7 @@ jobs:
     runs-on: ${{ matrix.os }}
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: dtolnay/rust-toolchain@master
         with:
           toolchain: ${{ matrix.rust }}
@@ -70,16 +70,20 @@ jobs:
   WASM:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: dtolnay/rust-toolchain@stable
         with:
           targets: wasm32-unknown-unknown
+      - name: Install wasm-pack
+        run: curl https://rustwasm.github.io/wasm-pack/installer/init.sh -sSf | sh
       - run: cargo build --target wasm32-unknown-unknown
+      - run: cd url && wasm-pack test --headless --chrome
+      - run: cd url && wasm-pack test --headless --firefox
 
   Lint:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: dtolnay/rust-toolchain@stable
         with:
           components: rustfmt, clippy
@@ -90,7 +94,7 @@ jobs:
   Audit:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - uses: EmbarkStudios/cargo-deny-action@v1
       
   Result:
diff --git a/README.md b/README.md
@@ -9,6 +9,6 @@ rust-url
 
 URL library for Rust, based on the [URL Standard](https://url.spec.whatwg.org/).
 
-[Documentation](https://docs.rs/url/)
+[Documentation](https://docs.rs/url)
 
-Please see [UPGRADING.md](https://github.com/servo/rust-url/blob/master/UPGRADING.md) if you are upgrading from a previous version.
+Please see [UPGRADING.md](https://github.com/servo/rust-url/blob/main/UPGRADING.md) if you are upgrading from a previous version.
diff --git a/data-url/Cargo.toml b/data-url/Cargo.toml
@@ -17,6 +17,7 @@ alloc = []
 
 [dev-dependencies]
 tester = "0.9"
+unicode-width = "=0.1.12"
 serde = { version = "1.0", default-features = false, features = ["alloc", "derive"] }
 serde_json = "1.0"
 
diff --git a/data-url/src/lib.rs b/data-url/src/lib.rs
@@ -298,6 +298,7 @@ where
             // before this special byte
             if i > slice_start {
                 write_bytes(&bytes[slice_start..i])?;
+                slice_start = i;
             }
             // Then deal with the special byte.
             match byte {
diff --git a/data-url/tests/data-urls.json b/data-url/tests/data-urls.json
@@ -52,6 +52,24 @@
   ["data:text/plain;Charset=UTF-8,%C2%B1",
    "text/plain;charset=UTF-8",
    [194, 177]],
+  ["data:text/plain,%",
+   "text/plain",
+   [37]],
+  ["data:text/plain,X%",
+   "text/plain",
+   [88, 37]],
+  ["data:text/plain,X%%",
+   "text/plain",
+   [88, 37, 37]],
+  ["data:text/plain;Charset=UTF-8,X%X",
+   "text/plain;charset=UTF-8",
+   [88, 37, 88]],
+  ["data:text/plain;Charset=UTF-8,X%0",
+   "text/plain;charset=UTF-8",
+   [88, 37, 48]],
+  ["data:text/plain;Charset=UTF-8,X%0X",
+   "text/plain;charset=UTF-8",
+   [88, 37, 48, 88]],
   ["data:text/plain;charset=windows-1252,áñçə💩",
    "text/plain;charset=windows-1252",
    [195, 161, 195, 177, 195, 167, 201, 153, 240, 159, 146, 169]],
diff --git a/form_urlencoded/src/lib.rs b/form_urlencoded/src/lib.rs
@@ -396,6 +396,9 @@ pub(crate) fn encode<'a>(encoding_override: EncodingOverride<'_>, input: &'a str
     input.as_bytes().into()
 }
 
+// std::ptr::addr_eq was stabilized in rust 1.76. Once we upgrade
+// the MSRV we can remove this lint override.
+#[allow(ambiguous_wide_pointer_comparisons)]
 pub(crate) fn decode_utf8_lossy(input: Cow<'_, [u8]>) -> Cow<'_, str> {
     // Note: This function is duplicated in `percent_encoding/lib.rs`.
     match input {
diff --git a/idna/benches/all.rs b/idna/benches/all.rs
@@ -11,6 +11,12 @@ fn to_unicode_puny_label(bench: &mut Bencher) {
     bench.iter(|| config.to_unicode(black_box(encoded)));
 }
 
+fn to_ascii_already_puny_label(bench: &mut Bencher) {
+    let encoded = "abc.xn--mgbcm";
+    let config = Config::default();
+    bench.iter(|| config.to_ascii(black_box(encoded)));
+}
+
 fn to_unicode_ascii(bench: &mut Bencher) {
     let encoded = "example.com";
     let config = Config::default();
@@ -47,6 +53,7 @@ benchmark_group!(
     to_unicode_ascii,
     to_unicode_merged_label,
     to_ascii_puny_label,
+    to_ascii_already_puny_label,
     to_ascii_simple,
     to_ascii_merged,
 );
diff --git a/idna/src/punycode.rs b/idna/src/punycode.rs
@@ -15,7 +15,6 @@
 
 use alloc::{string::String, vec::Vec};
 use core::char;
-use core::u32;
 
 // Bootstring parameters for Punycode
 static BASE: u32 = 36;
@@ -41,10 +40,12 @@ fn adapt(mut delta: u32, num_points: u32, first_time: bool) -> u32 {
 
 /// Convert Punycode to an Unicode `String`.
 ///
-/// This is a convenience wrapper around `decode`.
+/// Return None on malformed input or overflow.
+/// Overflow can only happen on inputs that take more than
+/// 63 encoded bytes, the DNS limit on domain name labels.
 #[inline]
 pub fn decode_to_string(input: &str) -> Option<String> {
-    decode(input).map(|chars| chars.into_iter().collect())
+    Some(Decoder::default().decode(input).ok()?.collect())
 }
 
 /// Convert Punycode to Unicode.
@@ -110,7 +111,7 @@ impl Decoder {
                 if digit > (u32::MAX - i) / weight {
                     return Err(()); // Overflow
                 }
-                i += digit * weight;
+                i = i.checked_add(digit * weight).ok_or(())?;
                 let t = if k <= bias {
                     T_MIN
                 } else if k >= bias + T_MAX {
diff --git a/idna/src/uts46.rs b/idna/src/uts46.rs
@@ -475,7 +475,7 @@ impl Idna {
         errors
     }
 
-    /// http://www.unicode.org/reports/tr46/#ToASCII
+    /// <http://www.unicode.org/reports/tr46/#ToASCII>
     #[allow(clippy::wrong_self_convention)]
     pub fn to_ascii(&mut self, domain: &str, out: &mut String) -> Result<(), Errors> {
         let mut errors = self.to_ascii_inner(domain, out);
@@ -497,7 +497,7 @@ impl Idna {
         errors.into()
     }
 
-    /// http://www.unicode.org/reports/tr46/#ToUnicode
+    /// <http://www.unicode.org/reports/tr46/#ToUnicode>
     #[allow(clippy::wrong_self_convention)]
     pub fn to_unicode(&mut self, domain: &str, out: &mut String) -> Result<(), Errors> {
         if is_simple(domain) {
@@ -518,7 +518,7 @@ pub struct Config {
     use_idna_2008_rules: bool,
 }
 
-/// The defaults are that of https://url.spec.whatwg.org/#idna
+/// The defaults are that of <https://url.spec.whatwg.org/#idna>
 impl Default for Config {
     fn default() -> Self {
         Config {
@@ -566,14 +566,14 @@ impl Config {
         self
     }
 
-    /// http://www.unicode.org/reports/tr46/#ToASCII
+    /// <http://www.unicode.org/reports/tr46/#ToASCII>
     pub fn to_ascii(self, domain: &str) -> Result<String, Errors> {
         let mut result = String::with_capacity(domain.len());
         let mut codec = Idna::new(self);
         codec.to_ascii(domain, &mut result).map(|()| result)
     }
 
-    /// http://www.unicode.org/reports/tr46/#ToUnicode
+    /// <http://www.unicode.org/reports/tr46/#ToUnicode>
     pub fn to_unicode(self, domain: &str) -> (String, Result<(), Errors>) {
         let mut codec = Idna::new(self);
         let mut out = String::with_capacity(domain.len());
@@ -714,7 +714,7 @@ impl From<Errors> for Result<(), Errors> {
 #[cfg(feature = "std")]
 impl std::error::Error for Errors {}
 
-#[cfg(feature = "nightly")]
+#[cfg(feature = "unstable")]
 impl core::error::Error for Errors {}
 
 impl fmt::Display for Errors {
diff --git a/idna/tests/punycode.rs b/idna/tests/punycode.rs
@@ -7,7 +7,7 @@
 // except according to those terms.
 
 use crate::test::TestFn;
-use idna::punycode::{decode, encode_str};
+use idna::punycode::{decode, decode_to_string, encode_str};
 use serde_json::map::Map;
 use serde_json::Value;
 use std::panic::catch_unwind;
@@ -28,6 +28,17 @@ fn one_test(decoded: &str, encoded: &str) {
         }
     }
 
+    match decode_to_string(encoded) {
+        None => panic!("Decoding {} failed.", encoded),
+        Some(result) => assert!(
+            result == decoded,
+            "Incorrect decoding of \"{}\":\n   \"{}\"\n!= \"{}\"\n",
+            encoded,
+            result,
+            decoded
+        ),
+    }
+
     match encode_str(decoded) {
         None => panic!("Encoding {} failed.", decoded),
         Some(result) => assert!(
diff --git a/idna/tests/punycode_tests.json b/idna/tests/punycode_tests.json
@@ -1,6 +1,6 @@
 [
 {
-    "description": "These tests are copied from https://github.com/bestiejs/punycode.js/blob/master/tests/tests.js , used under the MIT license.",
+    "description": "These tests are copied from https://github.com/mathiasbynens/punycode.js/blob/main/tests/tests.js , used under the MIT license.",
     "decoded": "",
     "encoded": ""
 },
diff --git a/idna/tests/unit.rs b/idna/tests/unit.rs
@@ -39,6 +39,12 @@ fn test_punycode_prefix_without_length_check() {
     assert_eq!(config.to_ascii("xn--.example.org").unwrap(), ".example.org");
 }
 
+#[test]
+fn test_punycode_invalid_encoding() {
+    let config = idna::Config::default();
+    assert!(config.to_ascii("xn--55555577").is_err());
+}
+
 // http://www.unicode.org/reports/tr46/#Table_Example_Processing
 #[test]
 fn test_examples() {
diff --git a/idna/tests/uts46.rs b/idna/tests/uts46.rs
@@ -31,7 +31,7 @@ pub fn collect_tests<F: FnMut(String, TestFn)>(add_test: &mut F) {
         // ToUnicode
         let mut to_unicode = unescape(pieces.remove(0));
         if to_unicode.is_empty() {
-            to_unicode = source.clone();
+            to_unicode.clone_from(&source)
         }
         let to_unicode_status = status(pieces.remove(0));
 
diff --git a/percent_encoding/src/lib.rs b/percent_encoding/src/lib.rs
@@ -447,7 +447,10 @@ impl<'a> PercentDecode<'a> {
     }
 }
 
+// std::ptr::addr_eq was stabilized in rust 1.76. Once we upgrade
+// the MSRV we can remove this lint override.
 #[cfg(feature = "alloc")]
+#[allow(ambiguous_wide_pointer_comparisons)]
 fn decode_utf8_lossy(input: Cow<'_, [u8]>) -> Cow<'_, str> {
     // Note: This function is duplicated in `form_urlencoded/src/query_encoding.rs`.
     match input {
diff --git a/url/Cargo.toml b/url/Cargo.toml
@@ -2,7 +2,7 @@
 
 name = "url"
 # When updating version, also modify html_root_url in the lib.rs
-version = "2.5.0"
+version = "2.5.2"
 authors = ["The rust-url developers"]
 
 description = "URL library for Rust, based on the WHATWG URL Standard"
@@ -21,6 +21,9 @@ serde = { version = "1.0", features = ["derive"] }
 serde_json = "1.0"
 bencher = "0.1"
 
+[target.'cfg(all(target_arch = "wasm32", target_os = "unknown"))'.dev-dependencies]
+wasm-bindgen-test = "0.3"
+
 [dependencies]
 form_urlencoded = { version = "1.2.1", path = "../form_urlencoded", default-features = false, features = ["alloc"] }
 idna = { version = "0.5.0", path = "../idna", default-features = false, features = ["alloc"] }
diff --git a/url/src/host.rs b/url/src/host.rs
@@ -339,7 +339,7 @@ fn parse_ipv4addr(input: &str) -> ParseResult<Ipv4Addr> {
     }
     let mut ipv4 = numbers.pop().expect("a non-empty list of numbers");
     // Equivalent to: ipv4 >= 256 ** (4 − numbers.len())
-    if ipv4 > u32::max_value() >> (8 * numbers.len() as u32) {
+    if ipv4 > u32::MAX >> (8 * numbers.len() as u32) {
         return Err(ParseError::InvalidIpv4Address);
     }
     if numbers.iter().any(|x| *x > 255) {
diff --git a/url/src/lib.rs b/url/src/lib.rs
diff --git a/url/src/parser.rs b/url/src/parser.rs
diff --git a/url/src/quirks.rs b/url/src/quirks.rs
diff --git a/url/tests/expected_failures.txt b/url/tests/expected_failures.txt
diff --git a/url/tests/unit.rs b/url/tests/unit.rs
diff --git a/url/tests/wpt.rs b/url/tests/wpt.rs

Original file line number	Diff line number	Diff line change
`@@ -298,6 +298,7 @@ where`
`298`	`298`	`// before this special byte`
`299`	`299`	`if i > slice_start {`
`300`	`300`	`write_bytes(&bytes[slice_start..i])?;`
	`301`	`+ slice_start = i;`
`301`	`302`	`}`
`302`	`303`	`// Then deal with the special byte.`
`303`	`304`	`match byte {`
Original file line number	Diff line number	Diff line change
`@@ -396,6 +396,9 @@ pub(crate) fn encode<'a>(encoding_override: EncodingOverride<'_>, input: &'a str`
`396`	`396`	`input.as_bytes().into()`
`397`	`397`	`}`
`398`	`398`
	`399`	`+// std::ptr::addr_eq was stabilized in rust 1.76. Once we upgrade`
	`400`	`+// the MSRV we can remove this lint override.`
	`401`	`+#[allow(ambiguous_wide_pointer_comparisons)]`
`399`	`402`	`pub(crate) fn decode_utf8_lossy(input: Cow<'_, [u8]>) -> Cow<'_, str> {`
`400`	`403`	// Note: This function is duplicated in `percent_encoding/lib.rs`.
`401`	`404`	`match input {`