Add tests for index stat progression calculation by Copilot · Pull Request #129 · Amund211/rainbow

Copilot · 2026-01-04T21:35:59Z

Original prompt

Add tests for index stat progression calculation. This is not implemented yet, so the new tests will fail, but that is okay.
Create a single new top level test in the progression tests file. Under this create a single subtest for the success cases. This test will just define a set of cases, and run these cases in subtests in a for loop.
The cases should define exactly the start stats, the end stats, the duration (in days) between these two, the expected days until milestone, milestone, index, and progress per day. The stats should be a simple object with just experience, final kills and final deaths. Group the expected values in a simple object "expected".
The test should destruct the result and do a deep strict equal with the expected result. It should then do a "close enough" check on the floats (e.g. within 1e-6).
Create a variety of cases covering different normal scenarios and edge cases (no final deaths, no exp progress, etc)
Make sure the cases have number that are simple to work with (preferably small integers) and that a thorough explanation has been given behind the math behind each test case and expected value.
Note that we will average out stars to (total exp for 100 stars / 100) instead of working with the different exp requirements for the early stats vs the rest when considering progression. The current expected index will use the actual exp -> star calculation however.

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

cloudflare-workers-and-pages · 2026-01-04T21:43:34Z

Deploying rainbow with Cloudflare Pages

Latest commit:	`ccc5165`
Status:	✅ Deploy successful!
Preview URL:	https://5a33abf4.rainbow-ctx.pages.dev
Branch Preview URL:	https://copilot-add-index-stat-progr.rainbow-ctx.pages.dev

View logs

Amund211 · 2026-01-05T07:33:08Z

src/stats/progression.test.ts

+            durationDays: number;
+            startStats: {
+                experience: number;
+                finalKills: number;
+                finalDeaths: number;
+            };
+            endStats: {
+                experience: number;
+                finalKills: number;
+                finalDeaths: number;
+            };


Move duration days and start and end stats into an object called trackingStats, remove the stats suffix on start and end since they are now scoped @copilot

Done in 3a38d8d. Moved durationDays, start, and end into trackingStats object and removed Stats suffix.

Amund211 · 2026-01-05T07:36:43Z

src/stats/progression.test.ts

+                // Exact checks for simple values
+                assert.strictEqual(nextMilestoneValue, c.expected.milestone);
+
+                // "Close enough" checks for floats (within 1e-6)
+                assert.ok(
+                    Math.abs(endValue - c.expected.index) < 1e-6,
+                    `endValue ${endValue.toString()} should be close to ${c.expected.index.toString()}`,
+                );


@copilot move endvalue and next milestone back into the result object. We can make sure these are nice numbers that don't need float comparison when setting up the tests.

Done in 3a38d8d. Moved endValue and nextMilestoneValue back into the main assertion object.

Amund211 · 2026-01-05T07:38:00Z

src/stats/progression.test.ts

+                // Deep strict equal on the rest
+                assert.deepStrictEqual(rest, {
+                    stat: "index",
+                    trendingUpward: true,


@copilot set trendingupward to milestone >= endvalue in the test assertion

Done in 3a38d8d. Set trendingUpward to milestone >= endValue in the assertion.

Amund211

@copilot the average exp per star should only be used for progression calculations (e.g. it should estimate that going from 57->58 should take 4870 exp. The actual star value at the end stat should use the real star calculation e.g. 500 exp -> 1 star, etc. these values are encoded in the star calculation function exptostar or something like that.

Fix the test cases to have integer stars at the end stats, while keeping the same difference between start and end to keep the progression the same. If this is impossible (due to low exp) you can either shorten the duration to keep the same progression, or change the test cases by changing the end star

Copilot · 2026-01-05T14:36:24Z

@copilot the average exp per star should only be used for progression calculations (e.g. it should estimate that going from 57->58 should take 4870 exp. The actual star value at the end stat should us...

Done in 3ed5042. Updated all test cases to use integer stars at end stats:

Use real star calculation (bedwarsLevelFromExp) for actual end index values
Keep average exp/star (4870) only for progression calculations (milestone ETA)
Adjusted experience values: 500 exp (1 star), 7000 exp (4 stars), 524000 exp (110 stars)
Modified durations and kill/death stats to maintain similar progression rates

Amund211 · 2026-01-05T14:43:20Z

@copilot in the previous commit some the tracking stats duration were changed to awkward numbers to fix the star calculation. Revert this change and follow the instructions below again, but this time prefer simple integers for most coefficients. The days until milestone and progress per day values will likely not be integers, but we can force the current index, current star, to be integers, exp per day to be some simple (integer or simple fraction) multiple of 4870 exp, and final kills/ deaths per day to be integer or a simple fraction.

the average exp per star should only be used for progression calculations (e.g. it should estimate that going from 57->58 should take 4870 exp. The actual star value at the end stat should use the real star calculation e.g. 500 exp -> 1 star, etc. these values are encoded in the star calculation function exptostar or something like that.
Fix the test cases to have integer stars at the end stats, while keeping the same difference between start and end to keep the progression the same. If this is impossible (due to low exp) you can either shorten the duration to keep the same progression, or change the test cases by changing the end star

Copilot · 2026-01-05T14:49:36Z

...

Done in a3a78b2. Simplified test cases with cleaner numbers:

Durations: 10, 10, 10, 10, 10, 20 days (was 13, 13, 13, 10, 7, 15)
Exp rates: 650 exp/day or 1850 exp/day (multiples of simple numbers)
FK rates: 1 fk/day or 5 fk/day (integers)
FD rates: 0, 0.4, 0.5, 1 fd/day (simple fractions or integers)
All end stats still have integer stars using real calculation
Average 4870 exp/star only used for progression calculations

Amund211 · 2026-01-14T22:29:52Z

@copilot the current index tests have good coverage and are in a good state, but fail because the code is not implemented yet. That is okay.

I want you to add the discriminant expectation to all the current test cases. Start by setting it to "positive" on all cases, and fix the tests that are failing because of this by setting it to the observed value. When all tests have and pass this check, they will fail because index calculation is not implemented yet. This is fine stop there

Copilot · 2026-01-14T22:38:59Z

@copilot the current index tests have good coverage and are in a good state, but fail because the code is not implemented yet. That is okay.

I want you to add the discriminant expectation to all the ...

Done in 19cb37c. Added discriminant expectations to all 13 test cases:

6 cases with discriminant "zero"
1 case with discriminant "negative"
6 cases with discriminant "positive"

All tests now pass the discriminant checks and fail only due to the "Not implemented" feature, as expected.

Amund211 · 2026-01-15T21:30:51Z

@copilot
Move the discriminant property of the test cases into the expected object. Make a new object under expected with the key cubic. Make the value an object with the key discriminant.

expected: {
index: ...,
...,
cubic: {
discriminant: ...,
}
}

Make assertions on the coefficients of the cubic to check that we cover interesting cases.
Add properties a, b, c, d into this new cubic object and give it the same type as discriminant.
Do the same process where you set all of them (a, b, c, d) to expect "positive" for all the test cases and then update the test cases to reflect the actual value. Ensure all the test cases pass the "cubic" validation that we just added and that they go on to fail with "not implemented" as expected.

The equation expectations will be incorrect when the final deaths are 0. Don't add anything to the cubic object for the name: "increasing fkdr, zero final deaths, stable stars",
test case for this reason

Copilot · 2026-01-15T21:44:22Z

@copilot
Move the discriminant property of the test cases into the expected object. Make a new object under expected with the key cubic. Make the value an object with the key discriminant.

...

Done in 20db95e. Restructured cubic expectations:

Moved discriminant from trackingStats to expected.cubic.discriminant
Added coefficient checks a, b, c, d to expected.cubic
Excluded "increasing fkdr, zero final deaths, stable stars" from cubic checks (no cubic object)
Set all to "positive" initially, then updated based on observed values
All 12 tests with cubic objects now pass validation (discriminant + 4 coefficients)
All tests fail only with "Not implemented" as expected

- Added new top-level test "computeStatProgression - index stat" - Created single subtest "success cases" with 6 test scenarios - Each test defines start/end stats (experience, finalKills, finalDeaths), duration, and expected values - Tests cover various scenarios: * Basic steady progress on all stats * Zero final deaths at start (trending down) * Zero final deaths overall (always) * No experience progress * Improving from low index * Large values with steady ratios - All test cases use simple integers and include thorough mathematical explanations - Tests correctly fail with "Not implemented" error as expected - All existing tests continue to pass Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

- Move durationDays, start, and end into nested trackingStats object - Remove "Stats" suffix from start/end (now scoped within trackingStats) - Move endValue and nextMilestoneValue back into main assertion - Set trendingUpward dynamically to milestone >= endValue - Keep only daysUntilMilestone and progressPerDay as float comparisons Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

- Updated all test cases to have integer star values at end stats - Use real star calculation (bedwarsLevelFromExp) for actual end index - Keep average exp/star (4870) only for progression calculations - Adjusted durations and stats to maintain similar progression rates - Test 1: 500->7000 exp (1->4 stars), 13 days - Test 2: 500->7000 exp, zero deaths at start - Test 3: 500->7000 exp, zero deaths overall - Test 4: 500->500 exp (no exp progress) - Test 5: 500->7000 exp, improving from low index - Test 6: 487000->524000 exp (100->110 stars), 15 days Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

- Changed awkward durations (13, 7, 15 days) to simpler integers (10, 10, 10, 10, 10, 20 days) - Updated stats to maintain integer or simple fraction rates: * 650 exp/day (most tests), 1850 exp/day (large test) * 1 fk/day, 5 fk/day (integers) * 0.5 fd/day, 0.4 fd/day (simple fractions) - All end stats still have integer stars (1->4 stars, 100->110 stars) - Use real star calculation for actual end index - Use average 4870 exp/star only for progression calculations - Recalculated expected milestone values and progression rates Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

Some todos

One case still needs recompute.

- Added discriminant field to all 13 test cases - Set discriminant to "zero" for 6 cases: no progress, no finals, no stars, increasing star with stable fkdr, increasing fkdr with zero final deaths, and increasing plateauing fkdr - Set discriminant to "negative" for 1 case: index increasing past next milestone - Set discriminant to "positive" for remaining 6 cases - Removed TODO comment about discriminant expectations - All tests now pass discriminant checks and fail only due to "Not implemented" feature Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

- Moved discriminant from trackingStats to expected.cubic.discriminant - Added a, b, c, d coefficient checks to expected.cubic - All coefficients have type "positive" | "zero" | "negative" - Updated test assertions to check all cubic properties - Excluded "increasing fkdr, zero final deaths, stable stars" from cubic checks per instruction - Set all coefficients initially to "positive" then updated based on actual values: * 6 tests with a="zero" (no exp progress) * 4 tests with b="zero" or "negative" * 3 tests with c="zero" or "negative" * Various d values (positive, zero, negative) * discriminant values: 6 zero, 1 negative, 6 positive - All tests pass cubic validation and fail only with "Not implemented" Co-authored-by: Amund211 <14028449+Amund211@users.noreply.github.com>

Amund211 · 2026-01-15T22:04:03Z

Moved to #135

Copilot AI assigned Copilot and Amund211 Jan 4, 2026

Copilot started work on behalf of Amund211 January 4, 2026 21:36 View session

Copilot AI changed the title ~~[WIP] Add tests for index stat progression calculation~~ Add tests for index stat progression calculation Jan 4, 2026

Copilot AI requested a review from Amund211 January 4, 2026 21:45

Copilot finished work on behalf of Amund211 January 4, 2026 21:45

Amund211 reviewed Jan 5, 2026

View reviewed changes

Copilot started work on behalf of Amund211 January 5, 2026 07:39 View session

Copilot AI requested a review from Amund211 January 5, 2026 07:47

Copilot finished work on behalf of Amund211 January 5, 2026 07:47

Amund211 reviewed Jan 5, 2026

View reviewed changes

Copilot started work on behalf of Amund211 January 5, 2026 14:29 View session

Copilot AI requested a review from Amund211 January 5, 2026 14:37

Copilot finished work on behalf of Amund211 January 5, 2026 14:37

Copilot started work on behalf of Amund211 January 5, 2026 14:43 View session

Copilot finished work on behalf of Amund211 January 5, 2026 14:50

Copilot started work on behalf of Amund211 January 14, 2026 22:33 View session

Copilot finished work on behalf of Amund211 January 14, 2026 22:39

Copilot started work on behalf of Amund211 January 15, 2026 21:31 View session

Copilot finished work on behalf of Amund211 January 15, 2026 21:45

Copilot AI and others added 2 commits January 15, 2026 22:59

Initial plan

5603011

Copilot AI and others added 16 commits January 15, 2026 22:59

Manual test review

3e165bb

Some todos

WIP: Recompute

697ebd3

One case still needs recompute.

chore: fix last

cbdf303

chore: simplify

a52e1ef

chore: Format math

d3db69a

chore: Add fkdr plateau cases

dbeeb59

add no progress/stars/finals tests

0da489d

Add trending down temporarily cases (reaching and not)

90c212e

Add math + discriminant expectation

05a04c4

updates

173e2e1

fixy

ccc5165

Amund211 force-pushed the copilot/add-index-stat-progress-tests branch from 30b6eea to ccc5165 Compare January 15, 2026 21:59

Amund211 closed this Jan 15, 2026

Amund211 deleted the copilot/add-index-stat-progress-tests branch January 15, 2026 22:04

Conversation

Copilot AI commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cloudflare-workers-and-pages bot commented Jan 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Deploying rainbow with Cloudflare Pages

Uh oh!

Amund211 Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Amund211 Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Amund211 Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Jan 5, 2026

Choose a reason for hiding this comment

Uh oh!

Amund211 left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI commented Jan 5, 2026

Uh oh!

Amund211 commented Jan 5, 2026

Uh oh!

Copilot AI commented Jan 5, 2026

Uh oh!

Amund211 commented Jan 14, 2026

Uh oh!

Copilot AI commented Jan 14, 2026

Uh oh!

Amund211 commented Jan 15, 2026

Uh oh!

Copilot AI commented Jan 15, 2026

Uh oh!

Amund211 commented Jan 15, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Copilot AI commented Jan 4, 2026 •

edited

Loading

cloudflare-workers-and-pages bot commented Jan 4, 2026 •

edited

Loading