TEST: Add tests for triangle.py.

[genedan]

Summary of Changes

Add tests for uncovered lines in triangle.py.

@henrydingliu, could you submit a PR testing for these lines for the disposal rate method? I think you'd be able to create a better test than I can:

chainladder-python/chainladder/core/triangle.py

Lines 1193 to 1195 in c45d10b

	if not obj.is_full:
	obj = obj[obj.valuation <= obj.valuation_date]
	if hasattr(obj, "disposal_w_"):

Related GitHub Issue(s)

Additional Context for Reviewers

I left any lines dealing with dask/cupy uncovered.

• I passed tests locally for both code (uv run pytest) and documentation changes (uv run jb build docs --builder=custom --custom-builder=doctest)

---

Note

Low Risk
Test-only changes with no modifications to triangle.py or runtime behavior; risk is limited to CI/runtime of the expanded test suite.

Overview
Adds a large batch of pytest coverage in test_triangle.py for previously untested branches in triangle.py (dask/cupy paths are explicitly left out per the PR description).

New tests lock in link_ratio behavior: pattern metadata (is_pattern, shape shrink), zero ratios converted to NaN (GH#181), and identity when called on an existing pattern triangle.

index / set_index are covered for DataFrame assignment (including slicer rebuild), length/type validation, and inplace vs copy semantics.

sort_axis tests assert that sorting columns, origin, or development permutes values with labels, not just axis metadata, and is a no-op when already sorted.

shift and trend gain tests for zero-period identity, invalid-axis errors, and unsupported trending axes.

dev_to_val on an already valuation triangle returns self when inplace=True.

Triangle.__init__ with AUTO_SPARSE disabled is tested for defaulting array_backend from cl.options.ARRAY_BACKEND and routing through set_backend when a backend is requested.

^{Reviewed by Cursor Bugbot for commit cce1937. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

Pyright Type Completeness

View the full pyright --verifytypes output for this commit

Project (full chainladder package, at this PR's head): 14.8% of exported symbols fully typed (188 / 1269)

	Known	Ambiguous	Unknown	Total
Project (head)	188	106	975	1269

Other symbols referenced but not exported by chainladder: 13

	Known	Ambiguous	Unknown	Total
Other (head)	3	1	9	13

Symbols without documentation:

• Functions without docstring: 313
• Functions without default param: 0
• Classes without docstring: 10

Patch (exported symbols added or changed by this PR): 23.5% fully typed (4 / 17)

	Known	Ambiguous	Unknown	Total
Patch	4	0	13	17

Patch symbol details

Symbol	Status	Change
chainladder.core.tests.test_triangle.test_dev_to_val_inplace_on_val_tri_returns_self	❌ unknown	new
chainladder.core.tests.test_triangle.test_index_setter_length_mismatch_raises	❌ unknown	new
chainladder.core.tests.test_triangle.test_index_setter_non_dataframe_raises	❌ unknown	new
chainladder.core.tests.test_triangle.test_index_setter_with_dataframe	❌ unknown	new
chainladder.core.tests.test_triangle.test_init_calls_set_backend_when_auto_sparse_disabled	✅ known	new
chainladder.core.tests.test_triangle.test_init_defaults_array_backend_to_option	✅ known	new
chainladder.core.tests.test_triangle.test_link_ratio_converts_zero_ratios_to_nan	✅ known	new
chainladder.core.tests.test_triangle.test_link_ratio_on_pattern_returns_self	❌ unknown	new
chainladder.core.tests.test_triangle.test_link_ratio_sets_pattern_metadata	❌ unknown	new
chainladder.core.tests.test_triangle.test_set_index_inplace	❌ unknown	new
chainladder.core.tests.test_triangle.test_set_index_not_inplace	❌ unknown	new
chainladder.core.tests.test_triangle.test_shift_invalid_axis_raises	❌ unknown	new
chainladder.core.tests.test_triangle.test_shift_zero_periods_returns_self	❌ unknown	new
chainladder.core.tests.test_triangle.test_sort_axis_columns_reorders_values	✅ known	new
chainladder.core.tests.test_triangle.test_sort_axis_development_reorders_values	❌ unknown	new
chainladder.core.tests.test_triangle.test_sort_axis_origin_reorders_values	❌ unknown	new
chainladder.core.tests.test_triangle.test_trend_invalid_axis_raises	❌ unknown	new

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.46%. Comparing base (55f9524) to head (cce1937).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1088      +/-   ##
==========================================
+ Coverage   90.18%   90.46%   +0.28%     
==========================================
  Files          91       91              
  Lines        5409     5287     -122     
  Branches      696      671      -25     
==========================================
- Hits         4878     4783      -95     
+ Misses        375      361      -14     
+ Partials      156      143      -13     

Flag	Coverage Δ
unittests	90.46% <ø> (+0.28%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[henrydingliu]

would a stronger test be tri.loc["A"] == clrd.iloc[:1]?

[henrydingliu]

inplace=False?

[henrydingliu]

why does this test feel like a deja vu?

[henrydingliu]

the implementation is kinda messy (maybe a small refactor is in order). i think we can benefit from a more consolidated test, i.e.

#the parameter supplied to the constructor will always override everything else (no idea if this is true, just making up something
assert cl.Triangle(backend = 'numpy').array_backend == 'numpy'

#if parameter is left out, cl.options.AUTO_SPARSE takes precedence
assert ...