FA-60496 / Loyalty points ledgers / Open access

Pending points hold period: repeat refund guard · case 01

A second refund message for the same order is processed again instead of being ignored.

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

ROOT CAUSE

The refund guard only checks that the order exists, not that its earn is still live.

VERIFIED REPAIR

Ignore refunds for unknown orders and for earns already cancelled or clawed.

Unsuccessful approach: Ignoring only cancelled earns still reprocesses refunds of clawed earns.

Case contract

Events: ["earn",day,order,pts] starts pending; points post (become available) on day >= earn day + hold. ["refund",day,order] cancels a live earn if day < earn day + hold ("cancelled"), otherwise claws it back ("clawed"; available may go negative); unknown or already refunded orders are "ignored". ["redeem",day,pts] spends available (posted - spent) points or returns "insufficient". ["query",day] returns [pending, available]. Return results list.

Why this case matters

Member point balances, statements and the program liability depend on this ledger rule.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(x):
    earns = {}
    spent = 0
    out = []
    posted = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day >= e[0] + x['hold'])
    pending = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day < e[0] + x['hold'])
    for ev in x['events']:
        k, day = ev[0], ev[1]
        if k == 'earn':
            earns[ev[2]] = [day, ev[3], 'live']
            out.append('pending')
        elif k == 'refund':
            e = earns.get(ev[2])
            if e is None:
                out.append('ignored')
                continue
            if day < e[0] + x['hold']:
                e[2] = 'cancelled'
            else:
                e[2] = 'clawed'
            out.append(e[2])
        elif k == 'redeem':
            if ev[2] > posted(day) - spent:
                out.append('insufficient')
                continue
            spent += ev[2]
            out.append('ok')
        else:
            out.append([pending(day), posted(day) - spent])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 32]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 41], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 41]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 23]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 31], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 151], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 12], ['query', 5], ['query', 12]]}, ['pending', [12, 0], [0, 12]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 26, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 21], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 33]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 42], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 42]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 24]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 32], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 152], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 13], ['query', 5], ['query', 12]]}, ['pending', [13, 0], [0, 13]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 27, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 22], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 34]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 43], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 43]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 25]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 33], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 153], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 14], ['query', 5], ['query', 12]]}, ['pending', [14, 0], [0, 14]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 28, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 23], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 35]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 44], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 44]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 26]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 34], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 154], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 15], ['query', 5], ['query', 12]]}, ['pending', [15, 0], [0, 15]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 29, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 24], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 36]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 45], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 45]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 27]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 35], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 155], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 16], ['query', 5], ['query', 12]]}, ['pending', [16, 0], [0, 16]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 25, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 25], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]]]
for label, args, expected in fixtures[N-1]:
    check(label, 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 fixture	Actual	Expected	Outcome
posts after hold	`['pending', [100, 0], [0, 100], [0, 100]]`	`['pending', [100, 0], [0, 100], [0, 100]]`	Passed
refund in hold	`['pending', 'pending', 'cancelled', [0, 41]]`	`['pending', 'pending', 'cancelled', [0, 41]]`	Passed
refund after post	`['pending', 'ok', 'clawed', [0, -50]]`	`['pending', 'ok', 'clawed', [0, -50]]`	Passed
double refund	`['pending', 'cancelled', 'cancelled', 'ignored']`	`['pending', 'cancelled', 'ignored', 'ignored']`	Failed
redeem pending	`['pending', 'pending', 'insufficient', 'ok']`	`['pending', 'pending', 'insufficient', 'ok']`	Passed
same day earn query	`['pending', [12, 0], [0, 12]]`	`['pending', [12, 0], [0, 12]]`	Passed
refund near posting	`['pending', 'cancelled', [0, 0]]`	`['pending', 'cancelled', [0, 0]]`	Passed
double refund after posting	`['pending', 'clawed', 'clawed']`	`['pending', 'clawed', 'ignored']`	Failed

SHA-256 / d94157aa779d39e14a7363e3a58b93a39b7d92d5cd10040698b2d4625e75e5d5

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(x):
    earns = {}
    spent = 0
    out = []
    posted = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day >= e[0] + x['hold'])
    pending = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day < e[0] + x['hold'])
    for ev in x['events']:
        k, day = ev[0], ev[1]
        if k == 'earn':
            earns[ev[2]] = [day, ev[3], 'live']
            out.append('pending')
        elif k == 'refund':
            e = earns.get(ev[2])
            if e is None or e[2] == 'cancelled':
                out.append('ignored')
                continue
            if day < e[0] + x['hold']:
                e[2] = 'cancelled'
            else:
                e[2] = 'clawed'
            out.append(e[2])
        elif k == 'redeem':
            if ev[2] > posted(day) - spent:
                out.append('insufficient')
                continue
            spent += ev[2]
            out.append('ok')
        else:
            out.append([pending(day), posted(day) - spent])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 32]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 41], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 41]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 23]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 31], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 151], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 12], ['query', 5], ['query', 12]]}, ['pending', [12, 0], [0, 12]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 26, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 21], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 33]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 42], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 42]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 24]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 32], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 152], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 13], ['query', 5], ['query', 12]]}, ['pending', [13, 0], [0, 13]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 27, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 22], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 34]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 43], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 43]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 25]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 33], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 153], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 14], ['query', 5], ['query', 12]]}, ['pending', [14, 0], [0, 14]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 28, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 23], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 35]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 44], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 44]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 26]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 34], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 154], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 15], ['query', 5], ['query', 12]]}, ['pending', [15, 0], [0, 15]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 29, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 24], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 36]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 45], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 45]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 27]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 35], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 155], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 16], ['query', 5], ['query', 12]]}, ['pending', [16, 0], [0, 16]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 25, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 25], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]]]
for label, args, expected in fixtures[N-1]:
    check(label, 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 fixture	Actual	Expected	Outcome
posts after hold	`['pending', [100, 0], [0, 100], [0, 100]]`	`['pending', [100, 0], [0, 100], [0, 100]]`	Passed
refund in hold	`['pending', 'pending', 'cancelled', [0, 41]]`	`['pending', 'pending', 'cancelled', [0, 41]]`	Passed
refund after post	`['pending', 'ok', 'clawed', [0, -50]]`	`['pending', 'ok', 'clawed', [0, -50]]`	Passed
double refund	`['pending', 'cancelled', 'ignored', 'ignored']`	`['pending', 'cancelled', 'ignored', 'ignored']`	Passed
redeem pending	`['pending', 'pending', 'insufficient', 'ok']`	`['pending', 'pending', 'insufficient', 'ok']`	Passed
same day earn query	`['pending', [12, 0], [0, 12]]`	`['pending', [12, 0], [0, 12]]`	Passed
refund near posting	`['pending', 'cancelled', [0, 0]]`	`['pending', 'cancelled', [0, 0]]`	Passed
double refund after posting	`['pending', 'clawed', 'clawed']`	`['pending', 'clawed', 'ignored']`	Failed

SHA-256 / 0fef00603604b2704c8d082805031c5cbc7cd636d9fd14902ab2eb862c613181

3 / The verified repair

Exit 0

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

N = 1
observations = []
def solve(x):
    earns = {}
    spent = 0
    out = []
    posted = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day >= e[0] + x['hold'])
    pending = lambda day: sum(e[1] for e in earns.values() if e[2] == 'live' and day < e[0] + x['hold'])
    for ev in x['events']:
        k, day = ev[0], ev[1]
        if k == 'earn':
            earns[ev[2]] = [day, ev[3], 'live']
            out.append('pending')
        elif k == 'refund':
            e = earns.get(ev[2])
            if e is None or e[2] != 'live':
                out.append('ignored')
                continue
            if day < e[0] + x['hold']:
                e[2] = 'cancelled'
            else:
                e[2] = 'clawed'
            out.append(e[2])
        elif k == 'redeem':
            if ev[2] > posted(day) - spent:
                out.append('insufficient')
                continue
            spent += ev[2]
            out.append('ok')
        else:
            out.append([pending(day), posted(day) - spent])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 32]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 41], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 41]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 23]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 31], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 151], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 12], ['query', 5], ['query', 12]]}, ['pending', [12, 0], [0, 12]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 26, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 21], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 33]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 42], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 42]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 24]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 32], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 152], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 13], ['query', 5], ['query', 12]]}, ['pending', [13, 0], [0, 13]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 27, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 22], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 34]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 43], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 43]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 25]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 33], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 153], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 14], ['query', 5], ['query', 12]]}, ['pending', [14, 0], [0, 14]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 28, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 23], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 35]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 44], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 44]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 26]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 34], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 154], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 15], ['query', 5], ['query', 12]]}, ['pending', [15, 0], [0, 15]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 29, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 24], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]], [['posts after hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['query', 29], ['query', 30], ['query', 36]]}, ['pending', [100, 0], [0, 100], [0, 100]]], ['refund in hold', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 2, 'o2', 45], ['refund', 10, 'o1'], ['query', 40]]}, ['pending', 'pending', 'cancelled', [0, 45]]], ['refund after post', {'hold': 14, 'events': [['earn', 0, 'o1', 70], ['redeem', 20, 50], ['refund', 21, 'o1'], ['query', 27]]}, ['pending', 'ok', 'clawed', [0, -50]]], ['double refund', {'hold': 10, 'events': [['earn', 0, 'o1', 35], ['refund', 3, 'o1'], ['refund', 4, 'o1'], ['refund', 5, 'zz']]}, ['pending', 'cancelled', 'ignored', 'ignored']], ['redeem pending', {'hold': 30, 'events': [['earn', 0, 'o1', 100], ['earn', 25, 'o2', 100], ['redeem', 40, 155], ['redeem', 40, 100]]}, ['pending', 'pending', 'insufficient', 'ok']], ['same day earn query', {'hold': 7, 'events': [['earn', 5, 'o1', 16], ['query', 5], ['query', 12]]}, ['pending', [16, 0], [0, 16]]], ['refund near posting', {'hold': 30, 'events': [['earn', 0, 'o1', 60], ['refund', 25, 'o1'], ['query', 60]]}, ['pending', 'cancelled', [0, 0]]], ['double refund after posting', {'hold': 5, 'events': [['earn', 0, 'o9', 25], ['refund', 9, 'o9'], ['refund', 10, 'o9']]}, ['pending', 'clawed', 'ignored']]]]
for label, args, expected in fixtures[N-1]:
    check(label, 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 fixture	Actual	Expected	Outcome
posts after hold	`['pending', [100, 0], [0, 100], [0, 100]]`	`['pending', [100, 0], [0, 100], [0, 100]]`	Passed
refund in hold	`['pending', 'pending', 'cancelled', [0, 41]]`	`['pending', 'pending', 'cancelled', [0, 41]]`	Passed
refund after post	`['pending', 'ok', 'clawed', [0, -50]]`	`['pending', 'ok', 'clawed', [0, -50]]`	Passed
double refund	`['pending', 'cancelled', 'ignored', 'ignored']`	`['pending', 'cancelled', 'ignored', 'ignored']`	Passed
redeem pending	`['pending', 'pending', 'insufficient', 'ok']`	`['pending', 'pending', 'insufficient', 'ok']`	Passed
same day earn query	`['pending', [12, 0], [0, 12]]`	`['pending', [12, 0], [0, 12]]`	Passed
refund near posting	`['pending', 'cancelled', [0, 0]]`	`['pending', 'cancelled', [0, 0]]`	Passed
double refund after posting	`['pending', 'clawed', 'ignored']`	`['pending', 'clawed', 'ignored']`	Passed

SHA-256 / 40721e7739ccbb479a5544049f4700c965d54bc2387aecefa97213353ac8bd19

Verification & scope

Stipulated bounded loyalty-program model with explicit toy rules; not any real program terms. 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 2026-09-29T14:46:46.105351+00:00.

Case digest / 4d8958f4894bef83eb6bd7e098a0097e2d75832639e0f5d8fbc7fe966f9e8927