quickfix ticks - don't round to zero in log scaled range

[lsabor]

don't try to round to zero if there is a zero point b/c it would be outside of the range

Summary by CodeRabbit

Bug Fixes

• Fixed an issue with number formatting where zero-value approximations were incorrectly applied in certain edge cases. The formatter now properly respects custom zero-point configurations, ensuring accurate number display across different formatting scenarios. This prevents unintended zero-approximation behavior when specific zero points are defined.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: fc03acfd-5708-4801-b486-65c99a173686

📥 Commits

Reviewing files that changed from the base of the PR and between 31f82db and 48449a6.

📒 Files selected for processing (1)

• front_end/src/utils/formatters/number.ts

---

📝 Walkthrough

Walkthrough

The abbreviatedNumber function in the number formatter utility has been modified to add an additional gating condition for the zero-reduction logic. It now only applies this fast-path when scaling.zero_point is nil, ensuring values near zero are not automatically abbreviated when an explicit zero point is defined in the scaling configuration.

Changes

Cohort / File(s)	Summary
Zero-Point Gating Logic front_end/src/utils/formatters/number.ts	Added condition to prevent zero-fast-path activation when scaling.zero_point is present, requiring both range_min and range_max to be defined AND zero_point to be nil before applying zero-reduction logic.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Possibly related PRs

• issue/3913/almost-0-ticks-can-display-badly #4373 — Propagates scaling into numeric display formatting functions (getForecastNumericDisplayValue, formatPredictionDisplayValue), affecting the same formatting behavior around scaling parameters.
• followup to not-quite-0 #4399 — Directly modifies zero-detection and zero-reduction logic in the same abbreviatedNumber function.

Suggested reviewers

• ncarazon

Poem

🐰 A zero point's a tricky thing,
When scaling spreads its wings,
Now we check before we abbreviate,
Preventing zeros to their fate! ✨

🚥 Pre-merge checks | ✅ 2

✅ Passed checks (2 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title 'quickfix ticks - don't round to zero in log scaled range' directly describes the main change: preventing zero-rounding behavior in log-scaled ranges with zero points.

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

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch quickfix/ticks

📝 Coding Plan

• Generate coding plan for human review comments

---

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

• X
• Mastodon
• Reddit
• LinkedIn

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

[github-actions]

Cleanup: Preview Environment Removed

The preview environment for this PR has been destroyed.

Resource	Status
🌐 Preview App	Deleted
🗄️ PostgreSQL Branch	Deleted
⚡ Redis Database	Deleted
🔧 GitHub Deployments	Removed
📦 Docker Image	Retained (auto-cleanup via GHCR policies)

---

Cleanup triggered by PR close at 2026-03-16T18:52:59Z