FA-10531 / Raster compositing / Open access
Straight alpha source over channel · case 01
A straight source channel is treated as already premultiplied.
ROOT CAUSE
A straight source channel is treated as already premultiplied.
VERIFIED REPAIR
Use the stage contract: Return a normalized straight source channel composited over an opaque destination channel. All three inputs lie in [0,1].
Unsuccessful approach: Attenuating the source without attenuating the destination double-counts covered background.
Case contract
Return a normalized straight source channel composited over an opaque destination channel. All three inputs lie in [0,1].
Why this case matters
A deterministic software graphics stage with explicit channel and coordinate conventions; no hardware, device profile or API behavior is inferred.
1 / The failure
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(source, alpha, destination):
return source+destination*(1-alpha)
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('fixture 1', solve(*(0.8, 0.5, 0.2)), 0.5)
check('fixture 2', solve(*(1, 0, 0.25)), 0.25)
check('fixture 3', solve(*(0.25, 1, 0.8)), 0.25)
check('fixture 4', solve(*(0, 0.5, 1)), 0.5)
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 |
|---|---|---|---|
| fixture 1 | 0.9 | 0.5 | Failed |
| fixture 2 | 1.25 | 0.25 | Failed |
| fixture 3 | 0.25 | 0.25 | Passed |
| fixture 4 | 0.5 | 0.5 | Passed |
SHA-256 / 0973a9f51c4d44832e114951374ec010f4725a512eaf130dc9f9639b1b4f645e
2 / The unsuccessful fix
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(source, alpha, destination):
return source*alpha+destination
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('fixture 1', solve(*(0.8, 0.5, 0.2)), 0.5)
check('fixture 2', solve(*(1, 0, 0.25)), 0.25)
check('fixture 3', solve(*(0.25, 1, 0.8)), 0.25)
check('fixture 4', solve(*(0, 0.5, 1)), 0.5)
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 |
|---|---|---|---|
| fixture 1 | 0.6000000000000001 | 0.5 | Failed |
| fixture 2 | 0.25 | 0.25 | Passed |
| fixture 3 | 1.05 | 0.25 | Failed |
| fixture 4 | 1.0 | 0.5 | Failed |
SHA-256 / 20ddbe3cf3262a0d018d670d6b08eef08b60dd8476c78bdf89d7f0fe97828f03
3 / The verified repair
Exit 0"""Failure Map reference implementation. Python standard library only."""
import json
import math
N = 1
observations = []
def solve(source, alpha, destination):
return source*alpha+destination*(1-alpha)
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('fixture 1', solve(*(0.8, 0.5, 0.2)), 0.5)
check('fixture 2', solve(*(1, 0, 0.25)), 0.25)
check('fixture 3', solve(*(0.25, 1, 0.8)), 0.25)
check('fixture 4', solve(*(0, 0.5, 1)), 0.5)
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 |
|---|---|---|---|
| fixture 1 | 0.5 | 0.5 | Passed |
| fixture 2 | 0.25 | 0.25 | Passed |
| fixture 3 | 0.25 | 0.25 | Passed |
| fixture 4 | 0.5 | 0.5 | Passed |
SHA-256 / d64e0fc59f1c154fdfbb72cc53cefb716307862f26742a3920eb6122094c26a2
Verification & scope
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 2026-09-29T14:38:39.898946+00:00.
Case digest / 67d7ae6732fc1c5efc8a6bbaa9b98feac51d36a97c12522c24e69adc8283dfb9