FA-97006 / Recipe scaling and nutrition / Open access

Percent daily value formatter: unit normalization · case 01

0.5 g of sodium is reported as 0.02% instead of 22%.

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

ROOT CAUSE

The amount is divided by the DV number without reconciling g/mg/mcg units.

VERIFIED REPAIR

Convert amount and DV to a common unit before dividing.

Unsuccessful approach: Assuming every DV is stated in mg breaks fat, fiber and vitamin D.

Case contract

DVs: sodium 2300 mg, fat 78 g, fiber 28 g, calcium 1300 mg, vitamin d 20 mcg, added sugars 50 g, protein 50 g. Names are trimmed, case-insensitive; unknown -> "error: no daily value". Units g, mg, mcg convert exactly; "iu" is allowed only for vitamin d (40 IU = 1 mcg), else "error: unit". pct = amount/DV*100. Zero -> "0%". Calcium and vitamin d: <2 -> "<2%", <=10 -> nearest 2, <=50 -> nearest 5, else nearest 10. Others: <1 -> "<1%", else nearest 1. Half-up.

Why this case matters

Label software prints %DV with different increments for micronutrients and macronutrients.

1 / The failure

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import math
from fractions import Fraction
N = 1
observations = []
def solve(nutrient, amount, unit):
    DV = {'sodium': (2300, 'mg'), 'fat': (78, 'g'), 'fiber': (28, 'g'), 'calcium': (1300, 'mg'),
          'vitamin d': (20, 'mcg'), 'added sugars': (50, 'g'), 'protein': (50, 'g')}
    SCALE = {'g': 1000000, 'mg': 1000, 'mcg': 1}
    key = nutrient.strip().lower()
    if key not in DV:
        return 'error: no daily value'
    amt = Fraction(str(amount))
    if unit == 'iu':
        if key != 'vitamin d':
            return 'error: unit'
        amt, unit = amt / 40, 'mcg'
    if unit not in SCALE:
        return 'error: unit'
    dv, dvu = DV[key]
    p = amt / dv * 100
    def near(v, s):
        return math.floor(v / s + Fraction(1, 2)) * s
    if p == 0:
        return '0%'
    if key in ('calcium', 'vitamin d'):
        if p < 2:
            return '<2%'
        v = near(p, 2) if p <= 10 else near(p, 5) if p <= 50 else near(p, 10)
    else:
        if p < 1:
            return '<1%'
        v = near(p, 1)
    return '%d%%' % v
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'],
  ['regression: unit normalization', ['CALCIUM', 40, 'g'], '3080%'],
  ['repair check: unit normalization', ['fat', 130, 'mg'], '<1%'],
  ['generated control 1', ['fat', 0.39, 'iu'], 'error: unit'],
  ['generated control 2', ['CALCIUM', 1, 'mg'], '<2%'],
  ['generated control 3', [' Sodium', 1500, 'g'], '65217%']],
 [['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['regression: unit normalization', ['protein', 40, 'mg'], '<1%'],
  ['repair check: unit normalization', ['added sugars', 130, 'g'], '260%'],
  ['generated control 1', ['iron', 5, 'g'], 'error: no daily value'],
  ['generated control 2', ['sodium', 2, 'mg'], '<1%'],
  ['generated control 3', ['vitamin d', 5, 'g'], '25000000%']],
 [['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['zero calcium', ['calcium', 0, 'mg'], '0%'], ['regression: unit normalization', ['fat', 10, 'mg'], '<1%'],
  ['repair check: unit normalization', ['vitamin d', 2, 'mg'], '10000%'],
  ['generated control 1', ['CALCIUM', 7.8, 'mg'], '<2%'], ['generated control 2', ['fat', 0, 'mg'], '0%'],
  ['generated control 3', ['iron', 7.8, 'mg'], 'error: no daily value']],
 [['fat in grams', ['fat', 3, 'g'], '4%'], ['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['regression: unit normalization', [' Sodium', 650, 'g'], '28261%'],
  ['repair check: unit normalization', ['added sugars', 700, 'g'], '1400%'],
  ['generated control 1', ['protein', 0, 'iu'], 'error: unit'],
  ['generated control 2', [' Sodium', 10, 'mg'], '<1%'],
  ['generated control 3', ['sodium', 260, 'mg'], '11%']],
 [['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['regression: unit normalization', [' Sodium', 2300, 'g'], '100000%'],
  ['repair check: unit normalization', ['fat', 1500, 'mcg'], '<1%'],
  ['generated control 1', ['protein', 1500, 'mg'], '3%'],
  ['generated control 2', ['sodium', 300, 'g'], '13043%'],
  ['generated control 3', ['protein', 10, 'mcg'], '<1%']]]
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
sodium control	`20%`	`20%`	Passed
calcium exactly two percent	`2%`	`2%`	Passed
vitamin d in IU	`50%`	`50%`	Passed
regression: unit normalization	`4%`	`3080%`	Failed
repair check: unit normalization	`167%`	`<1%`	Failed
generated control 1	`error: unit`	`error: unit`	Passed
generated control 2	`<2%`	`<2%`	Passed
generated control 3	`65%`	`65217%`	Failed

SHA-256 / 96dfa76706579587810bcd6875fdb9254b4b90b037dcf7efe8ca664ef0c0b596

2 / The unsuccessful fix

Exit 1

"""Failure Map reference implementation. Python standard library only."""
import json
import math
from fractions import Fraction
N = 1
observations = []
def solve(nutrient, amount, unit):
    DV = {'sodium': (2300, 'mg'), 'fat': (78, 'g'), 'fiber': (28, 'g'), 'calcium': (1300, 'mg'),
          'vitamin d': (20, 'mcg'), 'added sugars': (50, 'g'), 'protein': (50, 'g')}
    SCALE = {'g': 1000000, 'mg': 1000, 'mcg': 1}
    key = nutrient.strip().lower()
    if key not in DV:
        return 'error: no daily value'
    amt = Fraction(str(amount))
    if unit == 'iu':
        if key != 'vitamin d':
            return 'error: unit'
        amt, unit = amt / 40, 'mcg'
    if unit not in SCALE:
        return 'error: unit'
    dv, dvu = DV[key]
    p = amt * SCALE[unit] / (dv * SCALE['mg']) * 100
    def near(v, s):
        return math.floor(v / s + Fraction(1, 2)) * s
    if p == 0:
        return '0%'
    if key in ('calcium', 'vitamin d'):
        if p < 2:
            return '<2%'
        v = near(p, 2) if p <= 10 else near(p, 5) if p <= 50 else near(p, 10)
    else:
        if p < 1:
            return '<1%'
        v = near(p, 1)
    return '%d%%' % v
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'],
  ['regression: unit normalization', ['CALCIUM', 40, 'g'], '3080%'],
  ['repair check: unit normalization', ['fat', 130, 'mg'], '<1%'],
  ['generated control 1', ['fat', 0.39, 'iu'], 'error: unit'],
  ['generated control 2', ['CALCIUM', 1, 'mg'], '<2%'],
  ['generated control 3', [' Sodium', 1500, 'g'], '65217%']],
 [['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['regression: unit normalization', ['protein', 40, 'mg'], '<1%'],
  ['repair check: unit normalization', ['added sugars', 130, 'g'], '260%'],
  ['generated control 1', ['iron', 5, 'g'], 'error: no daily value'],
  ['generated control 2', ['sodium', 2, 'mg'], '<1%'],
  ['generated control 3', ['vitamin d', 5, 'g'], '25000000%']],
 [['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['zero calcium', ['calcium', 0, 'mg'], '0%'], ['regression: unit normalization', ['fat', 10, 'mg'], '<1%'],
  ['repair check: unit normalization', ['vitamin d', 2, 'mg'], '10000%'],
  ['generated control 1', ['CALCIUM', 7.8, 'mg'], '<2%'], ['generated control 2', ['fat', 0, 'mg'], '0%'],
  ['generated control 3', ['iron', 7.8, 'mg'], 'error: no daily value']],
 [['fat in grams', ['fat', 3, 'g'], '4%'], ['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['regression: unit normalization', [' Sodium', 650, 'g'], '28261%'],
  ['repair check: unit normalization', ['added sugars', 700, 'g'], '1400%'],
  ['generated control 1', ['protein', 0, 'iu'], 'error: unit'],
  ['generated control 2', [' Sodium', 10, 'mg'], '<1%'],
  ['generated control 3', ['sodium', 260, 'mg'], '11%']],
 [['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['regression: unit normalization', [' Sodium', 2300, 'g'], '100000%'],
  ['repair check: unit normalization', ['fat', 1500, 'mcg'], '<1%'],
  ['generated control 1', ['protein', 1500, 'mg'], '3%'],
  ['generated control 2', ['sodium', 300, 'g'], '13043%'],
  ['generated control 3', ['protein', 10, 'mcg'], '<1%']]]
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
sodium control	`20%`	`20%`	Passed
calcium exactly two percent	`2%`	`2%`	Passed
vitamin d in IU	`<2%`	`50%`	Failed
regression: unit normalization	`3080%`	`3080%`	Passed
repair check: unit normalization	`167%`	`<1%`	Failed
generated control 1	`error: unit`	`error: unit`	Passed
generated control 2	`<2%`	`<2%`	Passed
generated control 3	`65217%`	`65217%`	Passed

SHA-256 / 457e90ee40b76c2465492e479f47992c7bca2d2eebb112be6f7064c1cc7e9172

3 / The verified repair

Exit 0

"""Failure Map reference implementation. Python standard library only."""
import json
import math
from fractions import Fraction
N = 1
observations = []
def solve(nutrient, amount, unit):
    DV = {'sodium': (2300, 'mg'), 'fat': (78, 'g'), 'fiber': (28, 'g'), 'calcium': (1300, 'mg'),
          'vitamin d': (20, 'mcg'), 'added sugars': (50, 'g'), 'protein': (50, 'g')}
    SCALE = {'g': 1000000, 'mg': 1000, 'mcg': 1}
    key = nutrient.strip().lower()
    if key not in DV:
        return 'error: no daily value'
    amt = Fraction(str(amount))
    if unit == 'iu':
        if key != 'vitamin d':
            return 'error: unit'
        amt, unit = amt / 40, 'mcg'
    if unit not in SCALE:
        return 'error: unit'
    dv, dvu = DV[key]
    p = amt * SCALE[unit] / (dv * SCALE[dvu]) * 100
    def near(v, s):
        return math.floor(v / s + Fraction(1, 2)) * s
    if p == 0:
        return '0%'
    if key in ('calcium', 'vitamin d'):
        if p < 2:
            return '<2%'
        v = near(p, 2) if p <= 10 else near(p, 5) if p <= 50 else near(p, 10)
    else:
        if p < 1:
            return '<1%'
        v = near(p, 1)
    return '%d%%' % v
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'],
  ['regression: unit normalization', ['CALCIUM', 40, 'g'], '3080%'],
  ['repair check: unit normalization', ['fat', 130, 'mg'], '<1%'],
  ['generated control 1', ['fat', 0.39, 'iu'], 'error: unit'],
  ['generated control 2', ['CALCIUM', 1, 'mg'], '<2%'],
  ['generated control 3', [' Sodium', 1500, 'g'], '65217%']],
 [['calcium exactly two percent', ['calcium', 26, 'mg'], '2%'],
  ['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['regression: unit normalization', ['protein', 40, 'mg'], '<1%'],
  ['repair check: unit normalization', ['added sugars', 130, 'g'], '260%'],
  ['generated control 1', ['iron', 5, 'g'], 'error: no daily value'],
  ['generated control 2', ['sodium', 2, 'mg'], '<1%'],
  ['generated control 3', ['vitamin d', 5, 'g'], '25000000%']],
 [['vitamin d in IU', ['vitamin d', 400, 'iu'], '50%'], ['fat in grams', ['fat', 3, 'g'], '4%'],
  ['zero calcium', ['calcium', 0, 'mg'], '0%'], ['regression: unit normalization', ['fat', 10, 'mg'], '<1%'],
  ['repair check: unit normalization', ['vitamin d', 2, 'mg'], '10000%'],
  ['generated control 1', ['CALCIUM', 7.8, 'mg'], '<2%'], ['generated control 2', ['fat', 0, 'mg'], '0%'],
  ['generated control 3', ['iron', 7.8, 'mg'], 'error: no daily value']],
 [['fat in grams', ['fat', 3, 'g'], '4%'], ['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['regression: unit normalization', [' Sodium', 650, 'g'], '28261%'],
  ['repair check: unit normalization', ['added sugars', 700, 'g'], '1400%'],
  ['generated control 1', ['protein', 0, 'iu'], 'error: unit'],
  ['generated control 2', [' Sodium', 10, 'mg'], '<1%'],
  ['generated control 3', ['sodium', 260, 'mg'], '11%']],
 [['zero calcium', ['calcium', 0, 'mg'], '0%'],
  ['iu for sodium rejected', ['sodium', 10, 'iu'], 'error: unit'],
  ['sodium control', ['sodium', 460, 'mg'], '20%'],
  ['regression: unit normalization', [' Sodium', 2300, 'g'], '100000%'],
  ['repair check: unit normalization', ['fat', 1500, 'mcg'], '<1%'],
  ['generated control 1', ['protein', 1500, 'mg'], '3%'],
  ['generated control 2', ['sodium', 300, 'g'], '13043%'],
  ['generated control 3', ['protein', 10, 'mcg'], '<1%']]]
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
sodium control	`20%`	`20%`	Passed
calcium exactly two percent	`2%`	`2%`	Passed
vitamin d in IU	`50%`	`50%`	Passed
regression: unit normalization	`3080%`	`3080%`	Passed
repair check: unit normalization	`<1%`	`<1%`	Passed
generated control 1	`error: unit`	`error: unit`	Passed
generated control 2	`<2%`	`<2%`	Passed
generated control 3	`65217%`	`65217%`	Passed

SHA-256 / bb13cfdd7b0b28cc3f961faf1c4523b4667e51b41fa9bc54bea955a647d6f7ba

Verification & scope

A deterministic toy contract stated in full here; it is a bounded teaching model, not an authoritative reference or standards implementation. 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 2026-09-29T14:52:27.966222+00:00.

Case digest / 85c355697d901fceb95bdbd7ad3b484cfc9bf17d85efa20986465ffd2bacfd8f