Lazy init constant_limits with LazyAllocator #970

Alan-Jowett · 2026-01-12T20:46:57Z

Resolves: #972

Problem

The constant_limits static variable in ebpf_domain.cpp was initialized at static initialization time, before thread_local_options and variable_registry were properly set up. In release builds, this caused static initialization order issues where EbpfDomain::calculate_constant_limits() would capture uninitialized or default state from these thread-local variables, leading to incorrect constraint limits being cached and reused throughout the program's lifetime.

Additionally, even with lazy initialization, the cached limits could become stale when threads are reused with different configurations (e.g., in test suites where thread_local_options is reassigned between test cases), as the cache was never invalidated during thread-local state cleanup.

Solution

Lazy initialization: Replace static initialization with LazyAllocator<EbpfDomain, EbpfDomain::calculate_constant_limits> to defer computation until first use, ensuring thread_local_options and variable_registry are populated before calculate_constant_limits() captures their state.
Cache invalidation: Add EbpfDomain::clear_thread_local_state() method that clears the constant_limits cache, and call it from ebpf_verifier_clear_thread_local_state() to prevent stale cached limits when threads are reused with different configurations.
Soundness documentation: Expand comment to explicitly document reentrancy safety (no recursion possible since calculate_constant_limits doesn't call get_constant_limits() or widen(to_constants=true)) and soundness of intersection-based clamping (only shrinks domain, preserving over-approximation).

Changes

Changed constant_limits from static initialization to thread_local LazyAllocator
Added get_constant_limits() accessor function
Added EbpfDomain::clear_thread_local_state() to invalidate the cache
Updated widen() to use get_constant_limits() accessor
Added comprehensive comment documenting soundness and safety

Widening clamping behavior remains identical; this only fixes the initialization timing and cache lifetime issues.

Testing

Existing tests pass. The fix ensures correct behavior in release builds and when threads are reused across different verification contexts.

Summary by CodeRabbit

Chores
- Optimized constant limits initialization through lazy-loading mechanism for improved performance and resource management.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

Summary by CodeRabbit

Refactor
- Domain constants now initialize lazily to reduce startup cost and memory use.
- Public control added to reset domain thread-local state for better lifecycle handling.
Bug Fixes
- Verifier teardown now clears domain-specific thread-local state too, preventing stale state across runs.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

coderabbitai · 2026-01-12T20:47:10Z

Walkthrough

Introduce thread-local lazy initialization for EbpfDomain constant limits via a thread-local LazyAllocator and accessor; add EbpfDomain::clear_thread_local_state() and call it from verifier shutdown; tidy duplicate #include <ranges> and update includes.

Changes

Cohort / File(s)	Summary
EbpfDomain implementation `src/crab/ebpf_domain.cpp`	Add `thread_local LazyAllocator` and `get_constant_limits()` for lazy initialization; replace direct static constant usage in `widen(... to_constants=true)` with the lazy accessor; include `crab_utils/lazy_allocator.hpp`; preserve `calculate_constant_limits()` and add explanatory comments about TLS/non-reentrancy.
EbpfDomain header `src/crab/ebpf_domain.hpp`	Add public static method `static void clear_thread_local_state()` (placed after `calculate_constant_limits()`).
Verifier shutdown / includes cleanup `src/fwd_analyzer.cpp`	Remove duplicate `#include <ranges>` and add `EbpfDomain::clear_thread_local_state()` call to `ebpf_verifier_clear_thread_local_state()` to clear thread-local lazy state on shutdown.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: replacing static initialization with LazyAllocator for constant_limits to defer computation until first use.
Linked Issues check	✅ Passed	All requirements from #972 are met: constant_limits uses LazyAllocator for lazy initialization, clear_thread_local_state() was added to invalidate cache, and integration into ebpf_verifier_clear_thread_local_state() ensures cache lifetime correctness.
Out of Scope Changes check	✅ Passed	All changes are directly scoped to fixing the static initialization ordering issue: LazyAllocator integration, clear_thread_local_state() method, and duplicate #include removal are all necessary for the fix.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

📝 Generate docstrings

📜 Recent review details

Configuration used: Repository UI

Review profile: ASSERTIVE

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 3ef0772 and c73b885.

📒 Files selected for processing (3)

src/crab/ebpf_domain.cpp
src/crab/ebpf_domain.hpp
src/fwd_analyzer.cpp

🧰 Additional context used

📓 Path-based instructions (4)

src/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

src/**/*.{cpp,hpp,h}: Prove invariants when possible — encode the assumptions an analysis relies on (preconditions, lattice properties, monotonicity) directly in code comments, assertions, or type-level checks before trusting experiments.
Optimise for auditability — choose designs that are easy to step through and review by hand (even if marginally slower or more verbose) so future engineers can re-establish the soundness argument quickly.
Favour explicit error handling and early exits to surface problems instead of deferring to implicit behaviour.

Files:

src/crab/ebpf_domain.hpp
src/crab/ebpf_domain.cpp
src/fwd_analyzer.cpp

src/{asm_*,cfg,arith,crab,linux}/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

Narrate the reasoning — any change that affects analysis results should spell out the argument for soundness: what inputs are assumed, what invariants are maintained, and how the change preserves them.

Files:

src/crab/ebpf_domain.hpp
src/crab/ebpf_domain.cpp

**/*.{hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

Use #pragma once for static includes in header files; follow existing patterns within each subdirectory.

Files:

src/crab/ebpf_domain.hpp

**/*.{cpp,c,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

**/*.{cpp,c,hpp,h}: Ensure new C/C++ sources include the standard SPDX license header; validate with ./scripts/check-license.sh.
Run ./scripts/format-code --staged before committing to enforce clang-format standards; this mirrors the installed Git hook.

Files:

src/crab/ebpf_domain.hpp
src/crab/ebpf_domain.cpp
src/fwd_analyzer.cpp

🧠 Learnings (3)

📚 Learning: 2025-12-09T20:31:07.030Z

Learnt from: CR
Repo: vbpf/prevail PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-09T20:31:07.030Z
Learning: Applies to src/**/*.{cpp,hpp,h} : Prove invariants when possible — encode the assumptions an analysis relies on (preconditions, lattice properties, monotonicity) directly in code comments, assertions, or type-level checks before trusting experiments.

Applied to files:

src/crab/ebpf_domain.cpp

📚 Learning: 2025-12-09T20:31:07.030Z

Learnt from: CR
Repo: vbpf/prevail PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-09T20:31:07.030Z
Learning: Applies to src/{asm_*,cfg,arith,crab,linux}/**/*.{cpp,hpp,h} : Narrate the reasoning — any change that affects analysis results should spell out the argument for soundness: what inputs are assumed, what invariants are maintained, and how the change preserves them.

Applied to files:

src/crab/ebpf_domain.cpp
src/fwd_analyzer.cpp

📚 Learning: 2025-12-09T20:31:07.030Z

Learnt from: CR
Repo: vbpf/prevail PR: 0
File: AGENTS.md:0-0
Timestamp: 2025-12-09T20:31:07.030Z
Learning: Core verifier is implemented in modern C++20 — see CMakeLists.txt for build configuration details.

Applied to files:

src/fwd_analyzer.cpp

🧬 Code graph analysis (1)

src/fwd_analyzer.cpp (1)

src/crab/ebpf_domain.cpp (2)

clear_thread_local_state (167-167)

clear_thread_local_state (167-167)

🪛 Clang (14.0.6)

src/crab/ebpf_domain.cpp

[warning] 163-163: variable 'constant_limits' is non-const and globally accessible, consider making it const

(cppcoreguidelines-avoid-non-const-global-variables)

[warning] 165-165: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 169-169: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 169-169: method 'widen' can be made static

(readability-convert-member-functions-to-static)

src/fwd_analyzer.cpp

[error] 3-3: 'ranges' file not found

(clang-diagnostic-error)

🪛 Cppcheck (2.19.0)

src/crab/ebpf_domain.cpp

[style] 164-164: The function 'urem' is never used.

(unusedFunction)

[style] 173-173: The function 'copy' is never used.

(unusedFunction)

src/fwd_analyzer.cpp

[information] 3-3: Include file

(missingIncludeSystem)

[information] 30-30: Include file

(missingIncludeSystem)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (10)

GitHub Check: build_ubuntu (Debug, library)
GitHub Check: build_windows (Debug, library)
GitHub Check: build_ubuntu (Debug, tests)
GitHub Check: build_windows (Release, library)
GitHub Check: build_windows (Release, tests)
GitHub Check: build_ubuntu (Release, tests)
GitHub Check: build_windows (Debug, tests)
GitHub Check: build_ubuntu (Debug)
GitHub Check: build_ubuntu (Release)
GitHub Check: Analyze (cpp)

🔇 Additional comments (6)

src/crab/ebpf_domain.hpp (1)

70-71: LGTM!

The clear_thread_local_state() declaration is appropriately placed alongside calculate_constant_limits() and follows the existing pattern for static cleanup methods.

src/crab/ebpf_domain.cpp (3)

17-17: LGTM!

Include added for LazyAllocator template.

157-167: Well-documented soundness and safety invariants.

The comment correctly narrates:

Initialization timing: Deferred until thread-local state is ready

Non-reentrancy: calculate_constant_limits() doesn't recurse into get_constant_limits() or widen(to_constants=true)

Soundness of clamping: Intersection only shrinks domains, preserving over-approximation

Per coding guidelines, this satisfies the requirement to spell out invariants and the reasoning for soundness.

169-176: LGTM!

The accessor get_constant_limits() correctly replaces direct access to constant_limits. First invocation of widen(..., true) will trigger lazy initialization after thread-local state is set.

src/fwd_analyzer.cpp (2)

3-3: LGTM!

The <ranges> header is required for std::views::values at line 129. The static analysis warning about missing header is a false positive—the codebase targets C++20 where <ranges> is standard.

25-31: LGTM!

Adding EbpfDomain::clear_thread_local_state() to the cleanup sequence ensures the lazily-initialized constant_limits is invalidated when thread-local state is reset. This prevents stale cached values when threads are reused with different configurations (e.g., in test scenarios).

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.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

coderabbitai

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In @src/crab/ebpf_domain.cpp:
- Around line 157-160: The LazyAllocator instance constant_limits caches results
from EbpfDomain::calculate_constant_limits which capture thread_local_options
and variable_registry; add a call to constant_limits.clear() inside
ebpf_verifier_clear_thread_local_state() (in the cleanup routine in
src/fwd_analyzer.cpp) so the LazyAllocator is invalidated when thread-local
state is reset, preventing stale cached limits across reused threads.

📜 Review details

Configuration used: Repository UI

Review profile: ASSERTIVE

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 1ff4ace and 07e1a06.

📒 Files selected for processing (1)

src/crab/ebpf_domain.cpp

🧰 Additional context used

📓 Path-based instructions (3)

src/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

src/**/*.{cpp,hpp,h}: Prove invariants when possible — encode the assumptions an analysis relies on (preconditions, lattice properties, monotonicity) directly in code comments, assertions, or type-level checks before trusting experiments.
Optimise for auditability — choose designs that are easy to step through and review by hand (even if marginally slower or more verbose) so future engineers can re-establish the soundness argument quickly.
Favour explicit error handling and early exits to surface problems instead of deferring to implicit behaviour.

Files:

src/crab/ebpf_domain.cpp

src/{asm_*,cfg,arith,crab,linux}/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

Narrate the reasoning — any change that affects analysis results should spell out the argument for soundness: what inputs are assumed, what invariants are maintained, and how the change preserves them.

Files:

src/crab/ebpf_domain.cpp

**/*.{cpp,c,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

**/*.{cpp,c,hpp,h}: Ensure new C/C++ sources include the standard SPDX license header; validate with ./scripts/check-license.sh.
Run ./scripts/format-code --staged before committing to enforce clang-format standards; this mirrors the installed Git hook.

Files:

src/crab/ebpf_domain.cpp

🧬 Code graph analysis (1)

src/crab/ebpf_domain.cpp (1)

src/crab/ebpf_domain.hpp (10)

other (59-59)

other (60-60)

other (61-61)

other (62-62)

other (63-63)

other (64-64)

other (65-65)

other (66-66)

other (67-67)

other (68-68)

🪛 Clang (14.0.6)

src/crab/ebpf_domain.cpp

[warning] 158-158: variable 'constant_limits' is non-const and globally accessible, consider making it const

(cppcoreguidelines-avoid-non-const-global-variables)

[warning] 160-160: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 162-162: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 162-162: method 'widen' can be made static

(readability-convert-member-functions-to-static)

🪛 Cppcheck (2.19.0)

src/crab/ebpf_domain.cpp

[style] 164-164: The function 'urem' is never used.

(unusedFunction)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (10)

GitHub Check: build_windows (Debug, library)
GitHub Check: build_ubuntu (Release, library)
GitHub Check: build_windows (Release, library)
GitHub Check: build_windows (Debug, tests)
GitHub Check: build_ubuntu (Release, tests)
GitHub Check: build_ubuntu (Debug, tests)
GitHub Check: build_windows (Release, tests)
GitHub Check: build_ubuntu (Release)
GitHub Check: build_ubuntu (Debug)
GitHub Check: Analyze (cpp)

🔇 Additional comments (2)

src/crab/ebpf_domain.cpp (2)

17-17: LGTM!

Required include for the LazyAllocator template.

162-169: LGTM!

Usage of get_constant_limits() correctly replaces the former static reference. The widening logic remains intact.

(The static-analysis hint that widen can be made static is a false positive—it accesses this->rcp and stack.)

coderabbitai

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In @src/crab/ebpf_domain.cpp:
- Around line 157-160: Update the comment above the LazyAllocator instantiation
to explicitly document why clamping to constant limits via intersection is sound
and safe: state that LazyAllocator<EbpfDomain,
EbpfDomain::calculate_constant_limits> is used so thread_local_options is
populated first, note that LazyAllocator::get() has no reentrancy guard but
calculate_constant_limits does not call get_constant_limits() or
widen(to_constants=true) so recursion cannot occur, and explain that the
intersection operation only shrinks the domain (element-wise tightening of
constraints) and therefore preserves the over-approximation required for
soundness; reference LazyAllocator, EbpfDomain::calculate_constant_limits,
get_constant_limits, and widen(to_constants=true) in the comment.

📜 Review details

Configuration used: Repository UI

Review profile: ASSERTIVE

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 07e1a06 and 9f3f25f.

📒 Files selected for processing (1)

src/crab/ebpf_domain.cpp

🧰 Additional context used

📓 Path-based instructions (3)

src/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

src/**/*.{cpp,hpp,h}: Prove invariants when possible — encode the assumptions an analysis relies on (preconditions, lattice properties, monotonicity) directly in code comments, assertions, or type-level checks before trusting experiments.
Optimise for auditability — choose designs that are easy to step through and review by hand (even if marginally slower or more verbose) so future engineers can re-establish the soundness argument quickly.
Favour explicit error handling and early exits to surface problems instead of deferring to implicit behaviour.

Files:

src/crab/ebpf_domain.cpp

src/{asm_*,cfg,arith,crab,linux}/**/*.{cpp,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

Narrate the reasoning — any change that affects analysis results should spell out the argument for soundness: what inputs are assumed, what invariants are maintained, and how the change preserves them.

Files:

src/crab/ebpf_domain.cpp

**/*.{cpp,c,hpp,h}

📄 CodeRabbit inference engine (AGENTS.md)

**/*.{cpp,c,hpp,h}: Ensure new C/C++ sources include the standard SPDX license header; validate with ./scripts/check-license.sh.
Run ./scripts/format-code --staged before committing to enforce clang-format standards; this mirrors the installed Git hook.

Files:

src/crab/ebpf_domain.cpp

🧬 Code graph analysis (1)

src/crab/ebpf_domain.cpp (1)

src/crab/ebpf_domain.hpp (10)

other (59-59)

other (60-60)

other (61-61)

other (62-62)

other (63-63)

other (64-64)

other (65-65)

other (66-66)

other (67-67)

other (68-68)

🪛 Clang (14.0.6)

src/crab/ebpf_domain.cpp

[warning] 158-158: variable 'constant_limits' is non-const and globally accessible, consider making it const

(cppcoreguidelines-avoid-non-const-global-variables)

[warning] 160-160: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 162-162: use a trailing return type for this function

(modernize-use-trailing-return-type)

[warning] 162-162: method 'widen' can be made static

(readability-convert-member-functions-to-static)

🪛 Cppcheck (2.19.0)

src/crab/ebpf_domain.cpp

[style] 164-164: The function 'urem' is never used.

(unusedFunction)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (8)

GitHub Check: build_windows (Release, library)
GitHub Check: build_ubuntu (Release, library)
GitHub Check: build_ubuntu (Debug, library)
GitHub Check: build_ubuntu (Release, tests)
GitHub Check: build_ubuntu (Debug, tests)
GitHub Check: Analyze (cpp)
GitHub Check: build_ubuntu (Release)
GitHub Check: build_ubuntu (Debug)

🔇 Additional comments (2)

src/crab/ebpf_domain.cpp (2)

17-17: LGTM!

Include is correctly placed and necessary for the LazyAllocator template.

162-169: LGTM!

The widening implementation correctly uses get_constant_limits() to clamp the result when to_constants is true. The intersection operation preserves soundness by narrowing to the domain of representable constants.

src/crab/ebpf_domain.cpp

Alan-Jowett · 2026-01-12T21:05:14Z

✅ Review comment addressed. I've added:

EbpfDomain::clear_thread_local_state() static method to clear the constant_limits LazyAllocator
Called this method from ebpf_verifier_clear_thread_local_state() alongside the other cleanup operations

This ensures the cached constant limits are invalidated when thread-local state is reset, preventing stale cached values when threads are reused with different configurations.

coveralls · 2026-01-12T21:39:37Z

Pull Request Test Coverage Report for Build 20967734412

Details

5 of 5 (100.0%) changed or added relevant lines in 2 files are covered.
No unchanged relevant lines lost coverage.
Overall coverage increased (+0.05%) to 86.664%

Totals
Change from base Build 20882100451:	0.05%
Covered Lines:	9033
Relevant Lines:	10423

💛 - Coveralls

elazarg · 2026-01-13T05:59:23Z

@Alan-Jowett Could you just update the sign-off to use your signature instead of Copilot's?

Alan-Jowett · 2026-01-13T06:00:43Z

Whoops. Will do.

Signed-off-by: Alan Jowett <alanjo@microsoft.com>

coderabbitai bot reviewed Jan 12, 2026

View reviewed changes

Alan-Jowett mentioned this pull request Jan 12, 2026

Release build hang in bpf2c due to static init ordering of constant_limits #972

Closed

Alan-Jowett force-pushed the lazy-allocator-constant-limits branch from 07e1a06 to 9f3f25f Compare January 12, 2026 20:52

coderabbitai bot reviewed Jan 12, 2026

View reviewed changes

src/crab/ebpf_domain.cpp Outdated Show resolved Hide resolved

Alan-Jowett force-pushed the lazy-allocator-constant-limits branch from 173f260 to bdc575b Compare January 12, 2026 23:41

elazarg approved these changes Jan 13, 2026

View reviewed changes

Alan-Jowett force-pushed the lazy-allocator-constant-limits branch from bdc575b to 3ef0772 Compare January 13, 2026 16:41

Lazy init constant_limits using LazyAllocator

c73b885

Signed-off-by: Alan Jowett <alanjo@microsoft.com>

Alan-Jowett force-pushed the lazy-allocator-constant-limits branch from 3ef0772 to c73b885 Compare January 13, 2026 18:19

elazarg merged commit 67479e2 into vbpf:main Jan 14, 2026
16 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Lazy init constant_limits with LazyAllocator #970

Lazy init constant_limits with LazyAllocator #970

Uh oh!

Alan-Jowett commented Jan 12, 2026 •

edited by coderabbitai bot

Loading

Uh oh!

coderabbitai bot commented Jan 12, 2026 •

edited

Loading

Uh oh!

coderabbitai bot left a comment

Uh oh!

coderabbitai bot left a comment

Uh oh!

Uh oh!

Alan-Jowett commented Jan 12, 2026

Uh oh!

coveralls commented Jan 12, 2026 •

edited

Loading

Uh oh!

elazarg commented Jan 13, 2026

Uh oh!

Alan-Jowett commented Jan 13, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Lazy init constant_limits with LazyAllocator #970

Lazy init constant_limits with LazyAllocator #970

Uh oh!

Conversation

Alan-Jowett commented Jan 12, 2026 • edited by coderabbitai bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Problem

Solution

Changes

Testing

Summary by CodeRabbit

Summary by CodeRabbit

Uh oh!

coderabbitai bot commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Walkthrough

Changes

Estimated code review effort

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

coderabbitai bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Alan-Jowett commented Jan 12, 2026

Uh oh!

coveralls commented Jan 12, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Pull Request Test Coverage Report for Build 20967734412

Details

💛 - Coveralls

Uh oh!

elazarg commented Jan 13, 2026

Uh oh!

Alan-Jowett commented Jan 13, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Alan-Jowett commented Jan 12, 2026 •

edited by coderabbitai bot

Loading

coderabbitai bot commented Jan 12, 2026 •

edited

Loading

coveralls commented Jan 12, 2026 •

edited

Loading