uv: migrate to rkyv 0.8

Recently, rkyv 0.8 was released. Its API is a fair bit simpler now for higher level uses (like for us in `uv`) and results in us being able to delete a fair bit of code. This also removes our last dependency on `syn 1.0`, and thus drops that dependency. Performance (via testing on the `transformers` example) seems to remain about the same, which is what was expected: ``` $ hyperfine -w5 -r100 'uv lock' 'uv-ag-rkyv-update lock' Benchmark 1: uv lock Time (mean ± σ): 55.6 ms ± 6.4 ms [User: 30.4 ms, System: 35.1 ms] Range (min … max): 43.0 ms … 73.1 ms 100 runs Benchmark 2: uv-ag-rkyv-update lock Time (mean ± σ): 56.5 ms ± 7.2 ms [User: 30.5 ms, System: 36.3 ms] Range (min … max): 39.1 ms … 71.5 ms 100 runs Summary uv lock ran 1.02 ± 0.18 times faster than uv-ag-rkyv-update lock ``` Closes #7415
astral-sh · Sep 18, 2024 · 1379b53 · 1379b53
1 parent 91a574c
commit 1379b53
Show file tree

Hide file tree

Showing 22 changed files with 203 additions and 437 deletions.
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/Cargo.toml b/Cargo.toml
@@ -70,6 +70,7 @@ axoupdater = { version = "0.7.2", default-features = false }
 backoff = { version = "0.4.0" }
 base64 = { version = "0.22.1" }
 boxcar = { version = "0.2.5" }
+bytecheck = { version = "0.8.0" }
 cachedir = { version = "0.3.1" }
 cargo-util = { version = "0.2.14" }
 clap = { version = "4.5.17" }
@@ -125,7 +126,7 @@ regex = { version = "1.10.6" }
 reqwest = { version = "0.12.7", default-features = false, features = ["json", "gzip", "stream", "rustls-tls", "rustls-tls-native-roots", "socks"] }
 reqwest-middleware = { git = "https://github.com/astral-sh/reqwest-middleware", rev = "5e3eaf254b5bd481c75d2710eed055f95b756913" }
 reqwest-retry = { git = "https://github.com/astral-sh/reqwest-middleware", rev = "5e3eaf254b5bd481c75d2710eed055f95b756913" }
-rkyv = { version = "0.7.45", features = ["strict", "validation"] }
+rkyv = { version = "0.8.8", features = ["bytecheck"] }
 rmp-serde = { version = "1.3.0" }
 rust-netrc = { version = "0.1.1" }
 rustc-hash = { version = "2.0.0" }

diff --git a/crates/distribution-filename/src/build_tag.rs b/crates/distribution-filename/src/build_tag.rs
@@ -32,8 +32,7 @@ pub enum BuildTagError {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct BuildTag(u64, Option<Arc<str>>);
 
 impl FromStr for BuildTag {

diff --git a/crates/distribution-filename/src/extension.rs b/crates/distribution-filename/src/extension.rs
@@ -25,8 +25,7 @@ pub enum DistExtension {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub enum SourceDistExtension {
     Zip,
     TarGz,

diff --git a/crates/distribution-filename/src/source_dist.rs b/crates/distribution-filename/src/source_dist.rs
@@ -20,8 +20,7 @@ use uv_normalize::{InvalidNameError, PackageName};
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct SourceDistFilename {
     pub name: PackageName,
     pub version: Version,

diff --git a/crates/distribution-filename/src/wheel.rs b/crates/distribution-filename/src/wheel.rs
@@ -12,8 +12,7 @@ use uv_normalize::{InvalidNameError, PackageName};
 use crate::{BuildTag, BuildTagError};
 
 #[derive(Debug, Clone, Eq, PartialEq, Hash, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct WheelFilename {
     pub name: PackageName,
     pub version: Version,

diff --git a/crates/distribution-types/src/file.rs b/crates/distribution-types/src/file.rs
@@ -24,8 +24,7 @@ pub enum FileConversionError {
 #[derive(
     Debug, Clone, Hash, Serialize, Deserialize, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct File {
     pub dist_info_metadata: bool,
     pub filename: String,
@@ -72,8 +71,7 @@ impl File {
 #[derive(
     Debug, Clone, Hash, Serialize, Deserialize, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub enum FileLocation {
     /// URL relative to the base URL.
     RelativeUrl(String, String),
@@ -150,8 +148,7 @@ impl Display for FileLocation {
     rkyv::Serialize,
 )]
 #[serde(transparent)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct UrlString(String);
 
 impl UrlString {

diff --git a/crates/pep440-rs/src/version.rs b/crates/pep440-rs/src/version.rs
@@ -27,8 +27,7 @@ use std::{
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 #[cfg_attr(feature = "pyo3", pyclass)]
 pub enum Operator {
     /// `== 1.2.3`
@@ -288,15 +287,13 @@ impl std::fmt::Display for OperatorParseError {
 /// let version = Version::from_str("1.19").unwrap();
 /// ```
 #[derive(Clone, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 pub struct Version {
     inner: Arc<VersionInner>,
 }
 
 #[derive(Clone, Debug, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 enum VersionInner {
     Small { small: VersionSmall },
     Full { full: VersionFull },
@@ -885,8 +882,7 @@ impl FromStr for Version {
 /// Thankfully, such versions are incredibly rare. Virtually all versions have
 /// zero or one pre, dev or post release components.
 #[derive(Clone, Debug, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 struct VersionSmall {
     /// The representation discussed above.
     repr: u64,
@@ -1227,8 +1223,7 @@ impl VersionSmall {
 /// In general, the "full" representation is rarely used in practice since most
 /// versions will fit into the "small" representation.
 #[derive(Clone, Debug, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 struct VersionFull {
     /// The [versioning
     /// epoch](https://peps.python.org/pep-0440/#version-epochs). Normally
@@ -1361,8 +1356,7 @@ impl FromStr for VersionPattern {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 #[cfg_attr(feature = "pyo3", pyclass)]
 pub struct Prerelease {
     /// The kind of pre-release.
@@ -1387,8 +1381,7 @@ pub struct Prerelease {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 #[cfg_attr(feature = "pyo3", pyclass)]
 pub enum PrereleaseKind {
     /// alpha pre-release
@@ -1431,8 +1424,7 @@ impl std::fmt::Display for Prerelease {
 ///
 /// Luckily the default `Ord` implementation for `Vec<LocalSegment>` matches the PEP 440 rules.
 #[derive(Eq, PartialEq, Debug, Clone, Hash, rkyv::Archive, rkyv::Deserialize, rkyv::Serialize)]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
+#[rkyv(derive(Debug, Eq, PartialEq, PartialOrd, Ord))]
 pub enum LocalSegment {
     /// Not-parseable as integer segment of local version
     String(String),

diff --git a/crates/pep440-rs/src/version_specifier.rs b/crates/pep440-rs/src/version_specifier.rs
@@ -47,8 +47,7 @@ use crate::{
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 #[cfg_attr(feature = "pyo3", pyclass(sequence))]
 pub struct VersionSpecifiers(Vec<VersionSpecifier>);
 
@@ -293,8 +292,7 @@ impl std::error::Error for VersionSpecifiersParseError {}
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 #[cfg_attr(feature = "pyo3", pyclass(get_all))]
 pub struct VersionSpecifier {
     /// ~=|==|!=|<=|>=|<|>|===, plus whether the version ended with a star

diff --git a/crates/pypi-types/src/simple_json.rs b/crates/pypi-types/src/simple_json.rs
@@ -99,8 +99,7 @@ impl CoreMetadata {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 #[serde(untagged)]
 pub enum Yanked {
     Bool(bool),
@@ -303,8 +302,7 @@ impl FromStr for Hashes {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub enum HashAlgorithm {
     Md5,
     Sha256,
@@ -352,8 +350,7 @@ impl std::fmt::Display for HashAlgorithm {
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 pub struct HashDigest {
     pub algorithm: HashAlgorithm,
     pub digest: Box<str>,

diff --git a/crates/uv-cache/src/lib.rs b/crates/uv-cache/src/lib.rs
@@ -766,7 +766,7 @@ impl CacheBucket {
             Self::Interpreter => "interpreter-v2",
             // Note that when bumping this, you'll also need to bump it
             // in crates/uv/tests/cache_clean.rs.
-            Self::Simple => "simple-v12",
+            Self::Simple => "simple-v13",
             Self::Wheels => "wheels-v1",
             Self::Archive => "archive-v0",
             Self::Builds => "builds-v0",

diff --git a/crates/uv-client/Cargo.toml b/crates/uv-client/Cargo.toml
@@ -27,6 +27,7 @@ anyhow = { workspace = true }
 async-trait = { workspace = true }
 async_http_range_reader = { workspace = true }
 async_zip = { workspace = true }
+bytecheck = { workspace = true }
 fs-err = { workspace = true, features = ["tokio"] }
 futures = { workspace = true }
 html-escape = { workspace = true }

diff --git a/crates/uv-client/src/cached_client.rs b/crates/uv-client/src/cached_client.rs
@@ -75,9 +75,13 @@ impl<T: Serialize + DeserializeOwned> Cacheable for SerdeCacheable<T> {
 /// All `OwnedArchive` values are cacheable.
 impl<A> Cacheable for OwnedArchive<A>
 where
-    A: rkyv::Archive + rkyv::Serialize<crate::rkyvutil::Serializer<4096>>,
-    A::Archived: for<'a> rkyv::CheckBytes<rkyv::validation::validators::DefaultValidator<'a>>
-        + rkyv::Deserialize<A, rkyv::de::deserializers::SharedDeserializeMap>,
+    // A: rkyv::Archive + rkyv::Serialize<crate::rkyvutil::Serializer<4096>>,
+    // A::Archived: for<'a> rkyv::bytecheck::CheckBytes<rkyv::validation::validators::DefaultValidator<'a>>
+    // + rkyv::Deserialize<A, rkyv::de::deserializers::SharedDeserializeMap>,
+    A: rkyv::Archive + for<'a> rkyv::Serialize<crate::rkyvutil::Serializer<'a>>,
+    A::Archived: rkyv::Portable
+        + rkyv::Deserialize<A, crate::rkyvutil::Deserializer>
+        + for<'a> rkyv::bytecheck::CheckBytes<crate::rkyvutil::Validator<'a>>,
 {
     type Target = Self;
 

diff --git a/crates/uv-client/src/error.rs b/crates/uv-client/src/error.rs
@@ -227,7 +227,7 @@ pub enum ErrorKind {
     ArchiveRead(String),
 
     #[error("Writing to cache archive failed: {0}")]
-    ArchiveWrite(#[source] crate::rkyvutil::SerializerError),
+    ArchiveWrite(String),
 
     #[error("Network connectivity is disabled, but the requested data wasn't found in the cache for: `{0}`")]
     Offline(String),

diff --git a/crates/uv-client/src/httpcache/control.rs b/crates/uv-client/src/httpcache/control.rs
@@ -20,8 +20,7 @@ use crate::rkyvutil::OwnedArchive;
     rkyv::Deserialize,
     rkyv::Serialize,
 )]
-#[archive(check_bytes)]
-#[archive_attr(derive(Debug))]
+#[rkyv(derive(Debug))]
 #[allow(clippy::struct_excessive_bools)]
 pub struct CacheControl {
     // directives for requests and responses