You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Keep the result of nearest rational-to-float rounding instead of converting every inexact maximum-finite result into infinity. A quotient just beyond the largest finite float should still round to it until the halfway overflow threshold.
Apply the same boundary behavior to float(int): retain the existing fast conversion and retry only when malachite-bigint reports the inexact finite-extreme case as None.
Both signs and the midpoint-to-infinity case follow the pinned malachite implementation's ties-to-even rounding. The VM's existing infinity-to-OverflowError conversion remains in place.
Validation
Independently source-reviewed against CPython 3.15 long_true_divide, integer-to-double conversion, and the pinned malachite 0.14.0 implementations. In particular, the dependency implementation was checked rather than relying on its description of overflow.
Normal commit hooks, prek run --files crates/common/src/int.rs, and git diff --check passed.
Native RustPython compilation and tests were not run locally. On the prior exact head, hosted Linux, macOS and Windows each reported LongTest.test_float_conversion as an unexpected success in both the initial run and rerun. The update_lib auto-mark function removed only that expected-failure decorator; the IEEE guard and assertions are unchanged. New-head CI remains pending. Other expected-failure markers are unchanged.
Scope
#8624 independently handles the negative-zero result for zero divided by a negative integer in the same function. Both fixes need to be preserved and tested when the branches are combined. The only Lib change is the verified expected-failure marker removal described above.
Summary by CodeRabbit
Bug Fixes
Improved rounding for rational results and conversion of large integers to finite floating-point values.
Preserve the nearest finite result until the actual halfway overflow threshold for integer division and integer-to-float conversion.
Assisted-by: OpenAI Codex:model version not exposed
No actionable comments were generated in the recent review.
Walkthrough
The integer conversion code now returns the nearest-rounded result from true_div. When BigInt::to_f64() returns None, bigint_to_finite_float uses true_div and returns the result only if it is finite.
Changes: Integer-to-float conversion
Layer / File(s)
Summary
Rounding and conversion fallback crates/common/src/int.rs
true_div returns the nearest-rounded value without converting finite extrema to infinity. If to_f64() returns None, bigint_to_finite_float falls back to true_div(int, 1) and returns the result only if it is finite.
RustPython/RustPython#9520: Routes math.ldexp() integer arguments through ArgIntoFloat and tests oversized integer conversion.
Merge Risk: Minimal · up to 4fd58
No merge-blocking issue is established for the overflow-boundary rounding change. Merge after normal checks pass, including the pending native tests.
Pre-merge checks: 5
Check name
Status
Explanation
Description Check
Check skipped - CodeRabbit’s high-level summary is enabled.
Title check
The title clearly and concisely describes the main change: correcting the integer-to-float overflow rounding boundary.
Docstring Coverage
No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentat…
Linked Issues check
Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check
Check skipped because no linked issues were found for this pull request.
Full details: Docstring Coverage
Explanation
No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 1 files.
Finishing Touches
Generate unit tests (beta)
Create a new PR
Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
Reviewing files that changed from the base of the PR and between 18fc82b and 4fd589a.
Files ignored due to path filters: 1
Lib/test/test_long.py is excluded by !Lib/**
Files selected for processing: 1
crates/common/src/int.rs
Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 9 remain after this review.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
float(int): retain the existing fast conversion and retry only whenmalachite-bigintreports the inexact finite-extreme case asNone.OverflowErrorconversion remains in place.Validation
long_true_divide, integer-to-double conversion, and the pinned malachite 0.14.0 implementations. In particular, the dependency implementation was checked rather than relying on its description of overflow.prek run --files crates/common/src/int.rs, andgit diff --checkpassed.LongTest.test_float_conversionas an unexpected success in both the initial run and rerun. The update_lib auto-mark function removed only that expected-failure decorator; the IEEE guard and assertions are unchanged. New-head CI remains pending. Other expected-failure markers are unchanged.Scope
#8624 independently handles the negative-zero result for zero divided by a negative integer in the same function. Both fixes need to be preserved and tested when the branches are combined. The only Lib change is the verified expected-failure marker removal described above.
Summary by CodeRabbit