Replace numpy with math

[vyagubov]

See #1962

• Replaced numpy with math (available in python by default).
• No functional changes to the product's behavior are expected.

Summary by CodeRabbit

• Refactor

  • Replaced reliance on an external numeric library with built-in Python math operations in utility routines to simplify behavior and reduce overhead.

• Chores

  • Removed an explicit constraint on a numeric dependency in project metadata to allow newer versions.

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

[github-actions]

👋 @vyagubov
Thank you for raising your pull request.
Please make sure to add tests and document all user-facing changes.
You can do this by editing the docs files in this pull request.

[coderabbitai]

Walkthrough

This PR removes numpy usage from a JSON utility by switching np.isinf/np.isnan to math.isinf/math.isnan and deletes the explicit numpy = "<2.0.0" constraint from pyproject.toml. No public API changes.

Changes

Cohort / File(s)	Summary
JSON utility change elementary/utils/json_utils.py	Removed numpy import; added math import; replaced np.isinf() and np.isnan() calls with math.isinf() and math.isnan() in inf_and_nan_to_str.
Project configuration pyproject.toml	Removed the explicit numpy = "<2.0.0" dependency constraint and updated the nearby comment to reflect that the constraint was removed.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

• Review focus: correctness of math.isinf/math.isnan usage (scalar inputs), and ensuring no remaining numpy references in the module.
• Files to double-check:
  • elementary/utils/json_utils.py — confirm behavior with edge-case values (e.g., float('nan'), float('inf')).
  • pyproject.toml — ensure dependency removal aligns with project's dependency policy.

Poem

🐰 I swapped out arrays for simpler math,
Quiet numbers now find a calmer path,
Fewer hops in the dependency stack,
A lighter step along the track,
Hooray — a simpler, hop-filled patch! 🥕✨

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 50.00% which is insufficient. The required threshold is 80.00%.	You can run @coderabbitai generate docstrings to improve docstring coverage.

✅ 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 'Replace numpy with math' directly and accurately summarizes the main changes: numpy usage has been replaced with Python's math module in json_utils.py and the numpy constraint removed from pyproject.toml.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

📜 Recent review details

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between b34a1b0 and f2eb9e7.

📒 Files selected for processing (1)

• elementary/utils/json_utils.py (2 hunks)

🚧 Files skipped from review as they are similar to previous changes (1)

• elementary/utils/json_utils.py

---

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.}

[coderabbitai]

Actionable comments posted: 0

🧹 Nitpick comments (1)

pyproject.toml (1)

16-16: Clarify or update the comment about Python version constraints.

The comment mentions "numpy 1.26.x(which was removed)" but still references it as a reason for Python version limits. Since numpy has been removed from dependencies, this is confusing. Either remove the numpy reference entirely or clarify that the Python version constraint is now solely due to dbt v1.9 compatibility.

Apply this diff to clarify the comment:

-# matches the versions supported by dbt as of v1.9, which are probably due to the limits of numpy 1.26.x(which was removed)
+# matches the versions supported by dbt as of v1.9

📜 Review details

Configuration used: CodeRabbit UI

Review profile: CHILL

Plan: Pro

Disabled knowledge base sources:

• Linear integration is disabled by default for public repositories

You can enable these sources in your CodeRabbit configuration.

📥 Commits

Reviewing files that changed from the base of the PR and between 272a40c and b34a1b0.

📒 Files selected for processing (2)

• elementary/utils/json_utils.py (2 hunks)
• pyproject.toml (1 hunks)

🔇 Additional comments (2)

elementary/utils/json_utils.py (2)

2-2: LGTM! Clean replacement of numpy with standard library.

The import change from numpy to Python's built-in math module is appropriate and eliminates an external dependency.

Also applies to: 5-5

---

87-90: LGTM! Functionally equivalent replacement.

The replacement of np.isinf() and np.isnan() with math.isinf() and math.isnan() is correct. Since line 86 confirms obj is a float, the standard library functions provide identical behavior to their numpy counterparts for single values. Verified that numpy has been completely removed from the codebase.

[vyagubov]

I will upgrade python version to 3.13 in the next PR, if that one goes well.

[NoyaOffer]

Hey @vyagubov ! We're taking a look, would be happy to get the contribution for Python 3.13 as well 🙌

[vyagubov]

@NoyaOffer
I fixed the pre-commit issue.
Could you pls trigger tests once again?