FAILURE MAP
← Case archive

FA-59486 / Subscription proration billing / Open access

Cancellation refund for unused days: refund rounding direction · case 01

Refunds sometimes exceed the exact unused value by a cent.

Verified by executionVariant 1 · 8 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The refund is rounded half-up although the contract floors it so it never over-refunds.

THE FAILURE

The refund is rounded half-up although the contract floors it so it never over-refunds.

Unsuccessful approach: The attempt rounds up (ceiling), over-refunding whenever there is any fraction.

Case contract

Input {price, setup, period_days, used_days, min_days, mode immediate|period_end}. Unused = max(0, period_days - used_days). period_end: no refund, access continues for the unused days. immediate: charged days = max(used_days, min_days); refund = price*max(0, period_days - charged)/period_days floored (never over-refund); the setup fee is never refunded; access ends. Return [refund, access_days].

Why this case matters

Refund math must respect minimum-charge terms and exclude non-refundable fees.

1 / The failure

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(x):
    unused = max(0, x['period_days'] - x['used_days'])
    if x['mode'] == 'period_end':
        return [0, unused]
    charged_days = max(x['used_days'], x['min_days'])
    refundable_days = max(0, x['period_days'] - charged_days)
    refund = (x['price'] * refundable_days * 2 + x['period_days']) // (2 * x['period_days'])
    return [refund, 0]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression', {'price': 4900, 'setup': 2500, 'period_days': 90, 'used_days': 1, 'min_days': 7, 'mode': 'immediate'}, [4518, 0]), ('regression', {'price': 999, 'setup': 2500, 'period_days': 30, 'used_days': 18, 'min_days': 14, 'mode': 'immediate'}, [399, 0]), ('partial-repair probe', {'price': 4970, 'setup': 0, 'period_days': 90, 'used_days': 0, 'min_days': 7, 'mode': 'immediate'}, [4583, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 30, 'used_days': 22, 'min_days': 14, 'mode': 'immediate'}, [31973, 0]), ('normal control', {'price': 4900, 'setup': 9483, 'period_days': 31, 'used_days': 13, 'min_days': 0, 'mode': 'period_end'}, [0, 18]), ('normal control', {'price': 32440, 'setup': 0, 'period_days': 90, 'used_days': 25, 'min_days': 14, 'mode': 'period_end'}, [0, 65]), ('normal control', {'price': 28584, 'setup': 2500, 'period_days': 90, 'used_days': 93, 'min_days': 14, 'mode': 'immediate'}, [0, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 66, 'min_days': 0, 'mode': 'period_end'}, [0, 24])], [('regression', {'price': 999, 'setup': 2500, 'period_days': 30, 'used_days': 18, 'min_days': 14, 'mode': 'immediate'}, [399, 0]), ('regression', {'price': 86963, 'setup': 0, 'period_days': 90, 'used_days': 71, 'min_days': 14, 'mode': 'immediate'}, [18358, 0]), ('partial-repair probe', {'price': 4970, 'setup': 0, 'period_days': 90, 'used_days': 0, 'min_days': 7, 'mode': 'immediate'}, [4583, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 30, 'used_days': 22, 'min_days': 14, 'mode': 'immediate'}, [31973, 0]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 365, 'used_days': 45, 'min_days': 30, 'mode': 'period_end'}, [0, 320]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 31, 'used_days': 10, 'min_days': 30, 'mode': 'period_end'}, [0, 21]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 31, 'used_days': 17, 'min_days': 14, 'mode': 'period_end'}, [0, 14]), ('normal control', {'price': 119900, 'setup': 0, 'period_days': 90, 'used_days': 22, 'min_days': 7, 'mode': 'period_end'}, [0, 68])], [('regression', {'price': 86963, 'setup': 0, 'period_days': 90, 'used_days': 71, 'min_days': 14, 'mode': 'immediate'}, [18358, 0]), ('regression', {'price': 93530, 'setup': 0, 'period_days': 30, 'used_days': 29, 'min_days': 14, 'mode': 'immediate'}, [3117, 0]), ('partial-repair probe', {'price': 999, 'setup': 0, 'period_days': 30, 'used_days': 19, 'min_days': 0, 'mode': 'immediate'}, [366, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 365, 'used_days': 32, 'min_days': 0, 'mode': 'immediate'}, [109388, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 55, 'min_days': 14, 'mode': 'period_end'}, [0, 35]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 90, 'used_days': 3, 'min_days': 14, 'mode': 'period_end'}, [0, 87]), ('normal control', {'price': 119900, 'setup': 5059, 'period_days': 90, 'used_days': 7, 'min_days': 0, 'mode': 'period_end'}, [0, 83]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 365, 'used_days': 323, 'min_days': 7, 'mode': 'period_end'}, [0, 42])], [('regression', {'price': 93530, 'setup': 0, 'period_days': 30, 'used_days': 29, 'min_days': 14, 'mode': 'immediate'}, [3117, 0]), ('regression', {'price': 4900, 'setup': 1208, 'period_days': 365, 'used_days': 78, 'min_days': 14, 'mode': 'immediate'}, [3852, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 365, 'used_days': 32, 'min_days': 0, 'mode': 'immediate'}, [109388, 0]), ('partial-repair probe', {'price': 999, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'immediate'}, [483, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 30, 'used_days': 27, 'min_days': 0, 'mode': 'period_end'}, [0, 3]), ('normal control', {'price': 92433, 'setup': 0, 'period_days': 90, 'used_days': 5, 'min_days': 30, 'mode': 'period_end'}, [0, 85]), ('normal control', {'price': 83250, 'setup': 7374, 'period_days': 30, 'used_days': 17, 'min_days': 30, 'mode': 'immediate'}, [0, 0]), ('normal control', {'price': 999, 'setup': 6401, 'period_days': 31, 'used_days': 18, 'min_days': 30, 'mode': 'period_end'}, [0, 13])], [('regression', {'price': 4900, 'setup': 1208, 'period_days': 365, 'used_days': 78, 'min_days': 14, 'mode': 'immediate'}, [3852, 0]), ('regression', {'price': 119900, 'setup': 1279, 'period_days': 30, 'used_days': 14, 'min_days': 7, 'mode': 'immediate'}, [63946, 0]), ('partial-repair probe', {'price': 999, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'immediate'}, [483, 0]), ('partial-repair probe', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 53, 'min_days': 30, 'mode': 'immediate'}, [2014, 0]), ('normal control', {'price': 4900, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'period_end'}, [0, 15]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 31, 'used_days': 10, 'min_days': 30, 'mode': 'period_end'}, [0, 21]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 30, 'used_days': 0, 'min_days': 0, 'mode': 'immediate'}, [999, 0]), ('normal control', {'price': 219, 'setup': 2500, 'period_days': 365, 'used_days': 164, 'min_days': 7, 'mode': 'period_end'}, [0, 201])]]
for i, (label, args, expected) in enumerate(fixtures[N-1]):
    check("%s %d" % (label, i), solve(args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression 0[4519, 0][4518, 0]Failed
regression 1[400, 0][399, 0]Failed
partial-repair probe 2[4583, 0][4583, 0]Passed
partial-repair probe 3[31973, 0][31973, 0]Passed
normal control 4[0, 18][0, 18]Passed
normal control 5[0, 65][0, 65]Passed
normal control 6[0, 0][0, 0]Passed
normal control 7[0, 24][0, 24]Passed

SHA-256 / 727df43664765e4ffcf8c8793dd9a0000dc2cd789e846dd87bf6318e85e6a2c1

2 / The unsuccessful fix

Exit 1
"""Failure Map reference implementation. Python standard library only."""
import json

N = 1
observations = []
def solve(x):
    unused = max(0, x['period_days'] - x['used_days'])
    if x['mode'] == 'period_end':
        return [0, unused]
    charged_days = max(x['used_days'], x['min_days'])
    refundable_days = max(0, x['period_days'] - charged_days)
    refund = -(-x['price'] * refundable_days // x['period_days'])
    return [refund, 0]
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression', {'price': 4900, 'setup': 2500, 'period_days': 90, 'used_days': 1, 'min_days': 7, 'mode': 'immediate'}, [4518, 0]), ('regression', {'price': 999, 'setup': 2500, 'period_days': 30, 'used_days': 18, 'min_days': 14, 'mode': 'immediate'}, [399, 0]), ('partial-repair probe', {'price': 4970, 'setup': 0, 'period_days': 90, 'used_days': 0, 'min_days': 7, 'mode': 'immediate'}, [4583, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 30, 'used_days': 22, 'min_days': 14, 'mode': 'immediate'}, [31973, 0]), ('normal control', {'price': 4900, 'setup': 9483, 'period_days': 31, 'used_days': 13, 'min_days': 0, 'mode': 'period_end'}, [0, 18]), ('normal control', {'price': 32440, 'setup': 0, 'period_days': 90, 'used_days': 25, 'min_days': 14, 'mode': 'period_end'}, [0, 65]), ('normal control', {'price': 28584, 'setup': 2500, 'period_days': 90, 'used_days': 93, 'min_days': 14, 'mode': 'immediate'}, [0, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 66, 'min_days': 0, 'mode': 'period_end'}, [0, 24])], [('regression', {'price': 999, 'setup': 2500, 'period_days': 30, 'used_days': 18, 'min_days': 14, 'mode': 'immediate'}, [399, 0]), ('regression', {'price': 86963, 'setup': 0, 'period_days': 90, 'used_days': 71, 'min_days': 14, 'mode': 'immediate'}, [18358, 0]), ('partial-repair probe', {'price': 4970, 'setup': 0, 'period_days': 90, 'used_days': 0, 'min_days': 7, 'mode': 'immediate'}, [4583, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 30, 'used_days': 22, 'min_days': 14, 'mode': 'immediate'}, [31973, 0]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 365, 'used_days': 45, 'min_days': 30, 'mode': 'period_end'}, [0, 320]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 31, 'used_days': 10, 'min_days': 30, 'mode': 'period_end'}, [0, 21]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 31, 'used_days': 17, 'min_days': 14, 'mode': 'period_end'}, [0, 14]), ('normal control', {'price': 119900, 'setup': 0, 'period_days': 90, 'used_days': 22, 'min_days': 7, 'mode': 'period_end'}, [0, 68])], [('regression', {'price': 86963, 'setup': 0, 'period_days': 90, 'used_days': 71, 'min_days': 14, 'mode': 'immediate'}, [18358, 0]), ('regression', {'price': 93530, 'setup': 0, 'period_days': 30, 'used_days': 29, 'min_days': 14, 'mode': 'immediate'}, [3117, 0]), ('partial-repair probe', {'price': 999, 'setup': 0, 'period_days': 30, 'used_days': 19, 'min_days': 0, 'mode': 'immediate'}, [366, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 365, 'used_days': 32, 'min_days': 0, 'mode': 'immediate'}, [109388, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 55, 'min_days': 14, 'mode': 'period_end'}, [0, 35]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 90, 'used_days': 3, 'min_days': 14, 'mode': 'period_end'}, [0, 87]), ('normal control', {'price': 119900, 'setup': 5059, 'period_days': 90, 'used_days': 7, 'min_days': 0, 'mode': 'period_end'}, [0, 83]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 365, 'used_days': 323, 'min_days': 7, 'mode': 'period_end'}, [0, 42])], [('regression', {'price': 93530, 'setup': 0, 'period_days': 30, 'used_days': 29, 'min_days': 14, 'mode': 'immediate'}, [3117, 0]), ('regression', {'price': 4900, 'setup': 1208, 'period_days': 365, 'used_days': 78, 'min_days': 14, 'mode': 'immediate'}, [3852, 0]), ('partial-repair probe', {'price': 119900, 'setup': 0, 'period_days': 365, 'used_days': 32, 'min_days': 0, 'mode': 'immediate'}, [109388, 0]), ('partial-repair probe', {'price': 999, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'immediate'}, [483, 0]), ('normal control', {'price': 4900, 'setup': 0, 'period_days': 30, 'used_days': 27, 'min_days': 0, 'mode': 'period_end'}, [0, 3]), ('normal control', {'price': 92433, 'setup': 0, 'period_days': 90, 'used_days': 5, 'min_days': 30, 'mode': 'period_end'}, [0, 85]), ('normal control', {'price': 83250, 'setup': 7374, 'period_days': 30, 'used_days': 17, 'min_days': 30, 'mode': 'immediate'}, [0, 0]), ('normal control', {'price': 999, 'setup': 6401, 'period_days': 31, 'used_days': 18, 'min_days': 30, 'mode': 'period_end'}, [0, 13])], [('regression', {'price': 4900, 'setup': 1208, 'period_days': 365, 'used_days': 78, 'min_days': 14, 'mode': 'immediate'}, [3852, 0]), ('regression', {'price': 119900, 'setup': 1279, 'period_days': 30, 'used_days': 14, 'min_days': 7, 'mode': 'immediate'}, [63946, 0]), ('partial-repair probe', {'price': 999, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'immediate'}, [483, 0]), ('partial-repair probe', {'price': 4900, 'setup': 0, 'period_days': 90, 'used_days': 53, 'min_days': 30, 'mode': 'immediate'}, [2014, 0]), ('normal control', {'price': 4900, 'setup': 2500, 'period_days': 31, 'used_days': 16, 'min_days': 0, 'mode': 'period_end'}, [0, 15]), ('normal control', {'price': 119900, 'setup': 2500, 'period_days': 31, 'used_days': 10, 'min_days': 30, 'mode': 'period_end'}, [0, 21]), ('normal control', {'price': 999, 'setup': 0, 'period_days': 30, 'used_days': 0, 'min_days': 0, 'mode': 'immediate'}, [999, 0]), ('normal control', {'price': 219, 'setup': 2500, 'period_days': 365, 'used_days': 164, 'min_days': 7, 'mode': 'period_end'}, [0, 201])]]
for i, (label, args, expected) in enumerate(fixtures[N-1]):
    check("%s %d" % (label, i), solve(args), expected)
print(json.dumps({"observations": observations, "passed": all(x["passed"] for x in observations)}, ensure_ascii=False))
raise SystemExit(0 if all(x["passed"] for x in observations) else 1)
Boundary fixtureActualExpectedOutcome
regression 0[4519, 0][4518, 0]Failed
regression 1[400, 0][399, 0]Failed
partial-repair probe 2[4584, 0][4583, 0]Failed
partial-repair probe 3[31974, 0][31973, 0]Failed
normal control 4[0, 18][0, 18]Passed
normal control 5[0, 65][0, 65]Passed
normal control 6[0, 0][0, 0]Passed
normal control 7[0, 24][0, 24]Passed

SHA-256 / 60403f279be63c4a4ea9e28431f1be5ab920093a781e9127bdb4660a78d895a6

HELD IN THE MEMBER ARCHIVE

The verified repair and its recorded checks are member-only.

This mechanism has 8 recorded checks per implementation. The open-access tier publishes the failure and the unsuccessful fix; the repaired source that passes every check, and the observations that prove it, are available to members.

Every case sharing this mechanism uses the same contract and the same repair, so this one record is held back for all of them.

Member access is invitation-based. Sign in with your invited account to inspect the repair.

Sign in to the archive ↗

Verification & scope

A deterministic teaching model of a stipulated billing rule. It makes no claim to reproduce any billing provider's exact behaviour and is not billing software. This reproducer isolates one failure mechanism. Results cover the supplied fixtures. Variants within a family share a test contract and should remain grouped when constructing evaluation splits. Related mechanisms with a shared evaluation_group must also remain together; these controlled models are not independent production incidents.

Observations recorded using Python 3.12.14 at 2026-09-29T14:46:36.846797+00:00.

Case digest / cafc95cdba4e06c7c069eafe510bdf3b3a4b3aa5886cef50b2f93d4687873745