chore(deps): bump ordered-float from 4.6.0 to 5.3.0#25307
chore(deps): bump ordered-float from 4.6.0 to 5.3.0#25307dependabot[bot] wants to merge 1 commit intomasterfrom
Conversation
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: fcfbdfb0f3
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| metrics-util = { version = "0.18.0", default-features = false, features = ["registry"] } | ||
| nom = { version = "8.0.0", default-features = false } | ||
| ordered-float = { version = "4.6.0", default-features = false } | ||
| ordered-float = { version = "5.3.0", default-features = false } |
There was a problem hiding this comment.
Pin ordered-float below 5 until NotNan ops are updated
Upgrading ordered-float to 5.3.0 introduces a breaking API change where NotNan<T> + T now returns T (not NotNan<T>), so existing code like AddNumbersMerger::add in src/transforms/reduce/merge_strategy.rs (NumberMergerValue::Float(f + j as f64)) no longer type-checks because that enum variant requires NotNan<f64>. With this dependency bump alone, the project stops compiling unless those call sites are rewritten (or the dependency stays on 4.x).
Useful? React with 👍 / 👎.
dddc6ee to
3df2ae6
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 3df2ae68dd
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| metrics-util = { version = "0.18.0", default-features = false, features = ["registry"] } | ||
| nom = { version = "8.0.0", default-features = false } | ||
| ordered-float = { version = "4.6.0", default-features = false } | ||
| ordered-float = { version = "5.3.0", default-features = false } |
There was a problem hiding this comment.
Pin ordered-float to 4.x until NotNan ops are migrated
Bumping ordered-float to 5.3.0 breaks existing call sites that rely on NotNan<f64> + f64 returning NotNan<f64>; in v5 those mixed operators return plain f64 (per the v5 breaking change), so assignments like NumberMergerValue::Float(f + j as f64) in src/transforms/reduce/merge_strategy.rs no longer satisfy the enum’s NotNan<f64> variant type. This makes the workspace fail to compile until those arithmetic sites are rewritten to re-wrap/check with NotNan::new(...) or the dependency is kept on 4.x.
Useful? React with 👍 / 👎.
3df2ae6 to
9f8187b
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 9f8187b4c0
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| metrics-util = { version = "0.18.0", default-features = false, features = ["registry"] } | ||
| nom = { version = "8.0.0", default-features = false } | ||
| ordered-float = { version = "4.6.0", default-features = false } | ||
| ordered-float = { version = "5.3.0", default-features = false } |
There was a problem hiding this comment.
Keep ordered-float on 4.x until NotNan ops are migrated
Bumping ordered-float to 5.3.0 introduces a breaking API change for mixed arithmetic (NotNan<T> op T now yields T), but current code still relies on the old return type; for example, AddNumbersMerger::add assigns NumberMergerValue::Float(f + j as f64) in src/transforms/reduce/merge_strategy.rs:419, while NumberMergerValue::Float requires NotNan<f64> (merge_strategy.rs:381). With this dependency change alone, those call sites no longer type-check and the workspace fails to compile until they are rewritten or the crate is pinned to 4.x.
Useful? React with 👍 / 👎.
Bumps [ordered-float](https://github.com/reem/rust-ordered-float) from 4.6.0 to 5.3.0. - [Release notes](https://github.com/reem/rust-ordered-float/releases) - [Commits](reem/rust-ordered-float@v4.6.0...v5.3.0) --- updated-dependencies: - dependency-name: ordered-float dependency-version: 5.3.0 dependency-type: direct:production update-type: version-update:semver-major ... Signed-off-by: dependabot[bot] <support@github.com>
9f8187b to
656abe4
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 656abe4f4c
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
| metrics-util = { version = "0.18.0", default-features = false, features = ["registry"] } | ||
| nom = { version = "8.0.0", default-features = false } | ||
| ordered-float = { version = "4.6.0", default-features = false } | ||
| ordered-float = { version = "5.3.0", default-features = false } |
There was a problem hiding this comment.
Rework NotNan mixed arithmetic before version bump
Bumping ordered-float to 5.3.0 introduces a breaking operator change: NotNan<T> op T now yields plain T, so existing mixed arithmetic no longer type-checks where a NotNan<f64> is required. In this repo, src/transforms/reduce/merge_strategy.rs:419 assigns NumberMergerValue::Float(f + j as f64) while NumberMergerValue::Float is defined as NotNan<f64> (merge_strategy.rs:381), so this dependency update makes the workspace fail to compile until those call sites are rewritten (or the dependency remains on 4.x). Fresh evidence: the 5.3.0 NotNan docs define impl Add<T> for NotNan<T> { type Output = T; }.
Useful? React with 👍 / 👎.
Bumps ordered-float from 4.6.0 to 5.3.0.
Release notes
Sourced from ordered-float's releases.
Commits
6cca9b8v5.3.0a980aa3Rkyv v0.8 compatibility (#182)09998b4Fix broken link in docs35507c2Version 5.2.0f6b9d9bGeneralise lifetimes for binop implementations (#181)25da208Remove old CI configuration2bb1184Install clippy and rustfmtfea67dcVersion 5.1.01841f05Merge pull request #176 from DouisLavid/master1ae6383add a missing feature check for use statement