FA-78086 / Subtitle cue timing / Open access

Two-point subtitle synchronisation: empty cue drop · case 01

Cues squeezed to zero length by the mapping remain in the output.

Verified by executionVariant 1 · 9 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

The drop test keeps cues whose mapped end equals the mapped start.

VERIFIED REPAIR

Keep only cues whose mapped end is strictly after the mapped start.

Unsuccessful approach: Testing only for a positive end keeps collapsed cues that end after zero.

Case contract

anchors [[a1,b1],[a2,b2]] (a1 != a2, b1 != b2) map subtitle time a to video time b linearly through both points. Each time is mapped exactly, rounded half up, clamped at 0; cues whose mapped end is not after the mapped start are dropped.

Why this case matters

Subtitle timing defects shift, hide or overlap captions that viewers depend on for comprehension and accessibility.

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(cues, anchors):
    (a1,b1),(a2,b2)=anchors
    k=Fraction(b2-b1,a2-a1)
    def m(t):
        return math.floor(b1+(t-a1)*k+Fraction(1,2))
    out=[]
    for s,e in cues:
        ns,ne=max(0,m(s)),max(0,m(e))
        if ne>=ns:
            out.append([ns,ne])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: empty cue drop', [[[0, 1000]], [[1000, 0], [3000, 1000]]], []), ('regression variant: empty cue drop', [[[0, 2]], [[10000, 10500], [11000, 12000]]], []), ('partial repair probe: empty cue drop', [[[250, 251], [8492, 8493]], [[5000, 5500], [9000, 7000]]], []), ('partial repair variant: empty cue drop', [[[250, 252]], [[10000, 10500], [11000, 10501]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[999, 1002], [999, 1999], [7777, 8277]], [[1000, 1500], [2000, 2000]]], [[1500, 1501], [1500, 2000], [4889, 5139]]), ('normal control', [[[1920, 2920]], [[0, 1000], [2000, 0]]], []), ('normal control', [[[999, 1999], [1, 2501], [7777, 8277], [250, 253]], [[0, -1000], [3, 503]]], [[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]])], [('regression: empty cue drop', [[[7777, 10277]], [[5000, 6000], [5003, 4503]]], []), ('regression variant: empty cue drop', [[[1, 501], [7777, 7778]], [[1000, 0], [11000, 12500]]], [[8471, 8473]]), ('partial repair probe: empty cue drop', [[[3031, 3034], [250, 251], [4360, 4860], [0, 1000]], [[3000, 2500], [1000, 1000]]], [[2523, 2526], [3520, 3895], [250, 1000]]), ('partial repair variant: empty cue drop', [[[1, 4]], [[1003, 1003], [1000, 2000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 1001]], [[0, -1000], [10000, 9500]]], [[0, 51]]), ('normal control', [[[1, 1001], [1500, 1503]], [[1000, 500], [0, 0]]], [[1, 501], [750, 752]]), ('normal control', [[[0, 2500], [250, 750], [0, 3]], [[10000, 10500], [20000, 20500]]], [[500, 3000], [750, 1250], [500, 503]])], [('regression: empty cue drop', [[[250, 2750], [0, 3]], [[5000, 4000], [6000, 5500]]], [[0, 625]]), ('regression variant: empty cue drop', [[[1500, 1502], [7286, 7786]], [[10000, 10000], [11000, 12000]]], [[4572, 5572]]), ('partial repair probe: empty cue drop', [[[999, 1499], [7777, 10277]], [[10000, 10500], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[7777, 8777], [3213, 3214], [1500, 1501], [250, 750]], [[5000, 6000], [5003, 5003]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[16012, 18512], [0, 2], [999, 1002], [1, 4]], [[10000, 10000], [11000, 10500]]], [[13006, 14256], [5000, 5001], [5500, 5501], [5001, 5002]]), ('normal control', [[[1, 2501], [5729, 5732], [250, 252]], [[7000, 8500], [5000, 6000]]], [[0, 2876], [6911, 6915], [63, 65]]), ('normal control', [[[1, 2], [1500, 4000]], [[15000, 15000], [5000, 5000]]], [[1, 2], [1500, 4000]])], [('regression: empty cue drop', [[[999, 1001], [7777, 7779], [1500, 2000], [250, 251]], [[3000, 1000], [1000, 0]]], [[0, 1], [3389, 3390], [250, 500]]), ('regression variant: empty cue drop', [[[0, 2500], [0, 3]], [[1000, 1500], [0, -3000]]], [[0, 8250]]), ('partial repair probe: empty cue drop', [[[7777, 7779], [1500, 1503]], [[1000, 1000], [2000, 0]]], []), ('partial repair variant: empty cue drop', [[[7777, 7779], [999, 1000]], [[6000, 4000], [5000, 6000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 2], [999, 3499]], [[10000, 10000], [12000, 11500]]], [[2501, 2502], [3249, 5124]]), ('normal control', [[[7777, 8777], [7777, 7780], [1500, 2000], [999, 1002]], [[0, 1000], [4000, 4500]]], [[7805, 8680], [7805, 7808], [2313, 2750], [1874, 1877]]), ('normal control', [[[250, 750], [0, 1]], [[5000, 6000], [9000, 8500]]], [[3031, 3344], [2875, 2876]])], [('regression: empty cue drop', [[[7782, 10282], [7777, 7780], [250, 252], [1500, 2000]], [[1000, 2000], [3000, 2500]]], [[3696, 4321], [3694, 3695], [2125, 2250]]), ('regression variant: empty cue drop', [[[5033, 5036], [999, 1000]], [[9000, 9000], [5000, 4000]]], [[4041, 4045]]), ('partial repair probe: empty cue drop', [[[13786, 13787], [7777, 8277], [1, 2501], [250, 251]], [[10000, 11000], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[1500, 4000], [6431, 6931], [1, 1001], [862, 1862]], [[2000, 1500], [1000, 2000]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[7777, 8277], [7777, 8277]], [[10000, 9500], [0, -1000]]], [[7166, 7691], [7166, 7691]]), ('normal control', [[[0, 2]], [[10000, 10000], [20000, 20000]]], [[0, 2]]), ('normal control', [[[7777, 10277], [0, 2500], [1, 1001]], [[1000, 1000], [11000, 12000]]], [[8455, 11205], [0, 2650], [0, 1001]])]]
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
regression: empty cue drop	`[[0, 0]]`	`[]`	Failed
regression variant: empty cue drop	`[[0, 0]]`	`[]`	Failed
partial repair probe: empty cue drop	`[[3719, 3719], [6810, 6810]]`	`[]`	Failed
partial repair variant: empty cue drop	`[[10490, 10490]]`	`[]`	Failed
boundary control	`[[2000, 4000]]`	`[[2000, 4000]]`	Passed
boundary control	`[[1, 2]]`	`[[1, 2]]`	Passed
normal control	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	Passed
normal control	`[]`	`[]`	Passed
normal control	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	Passed

SHA-256 / 8eb0ba8be706f2e0a124dba30e73d8f7a14638323f74465c9424b6053a6b52d7

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(cues, anchors):
    (a1,b1),(a2,b2)=anchors
    k=Fraction(b2-b1,a2-a1)
    def m(t):
        return math.floor(b1+(t-a1)*k+Fraction(1,2))
    out=[]
    for s,e in cues:
        ns,ne=max(0,m(s)),max(0,m(e))
        if ne>0:
            out.append([ns,ne])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: empty cue drop', [[[0, 1000]], [[1000, 0], [3000, 1000]]], []), ('regression variant: empty cue drop', [[[0, 2]], [[10000, 10500], [11000, 12000]]], []), ('partial repair probe: empty cue drop', [[[250, 251], [8492, 8493]], [[5000, 5500], [9000, 7000]]], []), ('partial repair variant: empty cue drop', [[[250, 252]], [[10000, 10500], [11000, 10501]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[999, 1002], [999, 1999], [7777, 8277]], [[1000, 1500], [2000, 2000]]], [[1500, 1501], [1500, 2000], [4889, 5139]]), ('normal control', [[[1920, 2920]], [[0, 1000], [2000, 0]]], []), ('normal control', [[[999, 1999], [1, 2501], [7777, 8277], [250, 253]], [[0, -1000], [3, 503]]], [[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]])], [('regression: empty cue drop', [[[7777, 10277]], [[5000, 6000], [5003, 4503]]], []), ('regression variant: empty cue drop', [[[1, 501], [7777, 7778]], [[1000, 0], [11000, 12500]]], [[8471, 8473]]), ('partial repair probe: empty cue drop', [[[3031, 3034], [250, 251], [4360, 4860], [0, 1000]], [[3000, 2500], [1000, 1000]]], [[2523, 2526], [3520, 3895], [250, 1000]]), ('partial repair variant: empty cue drop', [[[1, 4]], [[1003, 1003], [1000, 2000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 1001]], [[0, -1000], [10000, 9500]]], [[0, 51]]), ('normal control', [[[1, 1001], [1500, 1503]], [[1000, 500], [0, 0]]], [[1, 501], [750, 752]]), ('normal control', [[[0, 2500], [250, 750], [0, 3]], [[10000, 10500], [20000, 20500]]], [[500, 3000], [750, 1250], [500, 503]])], [('regression: empty cue drop', [[[250, 2750], [0, 3]], [[5000, 4000], [6000, 5500]]], [[0, 625]]), ('regression variant: empty cue drop', [[[1500, 1502], [7286, 7786]], [[10000, 10000], [11000, 12000]]], [[4572, 5572]]), ('partial repair probe: empty cue drop', [[[999, 1499], [7777, 10277]], [[10000, 10500], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[7777, 8777], [3213, 3214], [1500, 1501], [250, 750]], [[5000, 6000], [5003, 5003]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[16012, 18512], [0, 2], [999, 1002], [1, 4]], [[10000, 10000], [11000, 10500]]], [[13006, 14256], [5000, 5001], [5500, 5501], [5001, 5002]]), ('normal control', [[[1, 2501], [5729, 5732], [250, 252]], [[7000, 8500], [5000, 6000]]], [[0, 2876], [6911, 6915], [63, 65]]), ('normal control', [[[1, 2], [1500, 4000]], [[15000, 15000], [5000, 5000]]], [[1, 2], [1500, 4000]])], [('regression: empty cue drop', [[[999, 1001], [7777, 7779], [1500, 2000], [250, 251]], [[3000, 1000], [1000, 0]]], [[0, 1], [3389, 3390], [250, 500]]), ('regression variant: empty cue drop', [[[0, 2500], [0, 3]], [[1000, 1500], [0, -3000]]], [[0, 8250]]), ('partial repair probe: empty cue drop', [[[7777, 7779], [1500, 1503]], [[1000, 1000], [2000, 0]]], []), ('partial repair variant: empty cue drop', [[[7777, 7779], [999, 1000]], [[6000, 4000], [5000, 6000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 2], [999, 3499]], [[10000, 10000], [12000, 11500]]], [[2501, 2502], [3249, 5124]]), ('normal control', [[[7777, 8777], [7777, 7780], [1500, 2000], [999, 1002]], [[0, 1000], [4000, 4500]]], [[7805, 8680], [7805, 7808], [2313, 2750], [1874, 1877]]), ('normal control', [[[250, 750], [0, 1]], [[5000, 6000], [9000, 8500]]], [[3031, 3344], [2875, 2876]])], [('regression: empty cue drop', [[[7782, 10282], [7777, 7780], [250, 252], [1500, 2000]], [[1000, 2000], [3000, 2500]]], [[3696, 4321], [3694, 3695], [2125, 2250]]), ('regression variant: empty cue drop', [[[5033, 5036], [999, 1000]], [[9000, 9000], [5000, 4000]]], [[4041, 4045]]), ('partial repair probe: empty cue drop', [[[13786, 13787], [7777, 8277], [1, 2501], [250, 251]], [[10000, 11000], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[1500, 4000], [6431, 6931], [1, 1001], [862, 1862]], [[2000, 1500], [1000, 2000]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[7777, 8277], [7777, 8277]], [[10000, 9500], [0, -1000]]], [[7166, 7691], [7166, 7691]]), ('normal control', [[[0, 2]], [[10000, 10000], [20000, 20000]]], [[0, 2]]), ('normal control', [[[7777, 10277], [0, 2500], [1, 1001]], [[1000, 1000], [11000, 12000]]], [[8455, 11205], [0, 2650], [0, 1001]])]]
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
regression: empty cue drop	`[]`	`[]`	Passed
regression variant: empty cue drop	`[]`	`[]`	Passed
partial repair probe: empty cue drop	`[[3719, 3719], [6810, 6810]]`	`[]`	Failed
partial repair variant: empty cue drop	`[[10490, 10490]]`	`[]`	Failed
boundary control	`[[2000, 4000]]`	`[[2000, 4000]]`	Passed
boundary control	`[[1, 2]]`	`[[1, 2]]`	Passed
normal control	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	Passed
normal control	`[]`	`[]`	Passed
normal control	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	Passed

SHA-256 / 83d4a3a727a8013aeb9cf4a765c8dc791d915fc1f12ab0296824d3e462d1bbf1

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(cues, anchors):
    (a1,b1),(a2,b2)=anchors
    k=Fraction(b2-b1,a2-a1)
    def m(t):
        return math.floor(b1+(t-a1)*k+Fraction(1,2))
    out=[]
    for s,e in cues:
        ns,ne=max(0,m(s)),max(0,m(e))
        if ne>ns:
            out.append([ns,ne])
    return out
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
fixtures = [[('regression: empty cue drop', [[[0, 1000]], [[1000, 0], [3000, 1000]]], []), ('regression variant: empty cue drop', [[[0, 2]], [[10000, 10500], [11000, 12000]]], []), ('partial repair probe: empty cue drop', [[[250, 251], [8492, 8493]], [[5000, 5500], [9000, 7000]]], []), ('partial repair variant: empty cue drop', [[[250, 252]], [[10000, 10500], [11000, 10501]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[999, 1002], [999, 1999], [7777, 8277]], [[1000, 1500], [2000, 2000]]], [[1500, 1501], [1500, 2000], [4889, 5139]]), ('normal control', [[[1920, 2920]], [[0, 1000], [2000, 0]]], []), ('normal control', [[[999, 1999], [1, 2501], [7777, 8277], [250, 253]], [[0, -1000], [3, 503]]], [[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]])], [('regression: empty cue drop', [[[7777, 10277]], [[5000, 6000], [5003, 4503]]], []), ('regression variant: empty cue drop', [[[1, 501], [7777, 7778]], [[1000, 0], [11000, 12500]]], [[8471, 8473]]), ('partial repair probe: empty cue drop', [[[3031, 3034], [250, 251], [4360, 4860], [0, 1000]], [[3000, 2500], [1000, 1000]]], [[2523, 2526], [3520, 3895], [250, 1000]]), ('partial repair variant: empty cue drop', [[[1, 4]], [[1003, 1003], [1000, 2000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 1001]], [[0, -1000], [10000, 9500]]], [[0, 51]]), ('normal control', [[[1, 1001], [1500, 1503]], [[1000, 500], [0, 0]]], [[1, 501], [750, 752]]), ('normal control', [[[0, 2500], [250, 750], [0, 3]], [[10000, 10500], [20000, 20500]]], [[500, 3000], [750, 1250], [500, 503]])], [('regression: empty cue drop', [[[250, 2750], [0, 3]], [[5000, 4000], [6000, 5500]]], [[0, 625]]), ('regression variant: empty cue drop', [[[1500, 1502], [7286, 7786]], [[10000, 10000], [11000, 12000]]], [[4572, 5572]]), ('partial repair probe: empty cue drop', [[[999, 1499], [7777, 10277]], [[10000, 10500], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[7777, 8777], [3213, 3214], [1500, 1501], [250, 750]], [[5000, 6000], [5003, 5003]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[16012, 18512], [0, 2], [999, 1002], [1, 4]], [[10000, 10000], [11000, 10500]]], [[13006, 14256], [5000, 5001], [5500, 5501], [5001, 5002]]), ('normal control', [[[1, 2501], [5729, 5732], [250, 252]], [[7000, 8500], [5000, 6000]]], [[0, 2876], [6911, 6915], [63, 65]]), ('normal control', [[[1, 2], [1500, 4000]], [[15000, 15000], [5000, 5000]]], [[1, 2], [1500, 4000]])], [('regression: empty cue drop', [[[999, 1001], [7777, 7779], [1500, 2000], [250, 251]], [[3000, 1000], [1000, 0]]], [[0, 1], [3389, 3390], [250, 500]]), ('regression variant: empty cue drop', [[[0, 2500], [0, 3]], [[1000, 1500], [0, -3000]]], [[0, 8250]]), ('partial repair probe: empty cue drop', [[[7777, 7779], [1500, 1503]], [[1000, 1000], [2000, 0]]], []), ('partial repair variant: empty cue drop', [[[7777, 7779], [999, 1000]], [[6000, 4000], [5000, 6000]]], []), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('normal control', [[[1, 2], [999, 3499]], [[10000, 10000], [12000, 11500]]], [[2501, 2502], [3249, 5124]]), ('normal control', [[[7777, 8777], [7777, 7780], [1500, 2000], [999, 1002]], [[0, 1000], [4000, 4500]]], [[7805, 8680], [7805, 7808], [2313, 2750], [1874, 1877]]), ('normal control', [[[250, 750], [0, 1]], [[5000, 6000], [9000, 8500]]], [[3031, 3344], [2875, 2876]])], [('regression: empty cue drop', [[[7782, 10282], [7777, 7780], [250, 252], [1500, 2000]], [[1000, 2000], [3000, 2500]]], [[3696, 4321], [3694, 3695], [2125, 2250]]), ('regression variant: empty cue drop', [[[5033, 5036], [999, 1000]], [[9000, 9000], [5000, 4000]]], [[4041, 4045]]), ('partial repair probe: empty cue drop', [[[13786, 13787], [7777, 8277], [1, 2501], [250, 251]], [[10000, 11000], [10003, 9503]]], []), ('partial repair variant: empty cue drop', [[[1500, 4000], [6431, 6931], [1, 1001], [862, 1862]], [[2000, 1500], [1000, 2000]]], []), ('boundary control', [[[1000, 2000]], [[0, 0], [1000, 2000]]], [[2000, 4000]]), ('boundary control', [[[1, 3]], [[0, 0], [2, 1]]], [[1, 2]]), ('normal control', [[[7777, 8277], [7777, 8277]], [[10000, 9500], [0, -1000]]], [[7166, 7691], [7166, 7691]]), ('normal control', [[[0, 2]], [[10000, 10000], [20000, 20000]]], [[0, 2]]), ('normal control', [[[7777, 10277], [0, 2500], [1, 1001]], [[1000, 1000], [11000, 12000]]], [[8455, 11205], [0, 2650], [0, 1001]])]]
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
regression: empty cue drop	`[]`	`[]`	Passed
regression variant: empty cue drop	`[]`	`[]`	Passed
partial repair probe: empty cue drop	`[]`	`[]`	Passed
partial repair variant: empty cue drop	`[]`	`[]`	Passed
boundary control	`[[2000, 4000]]`	`[[2000, 4000]]`	Passed
boundary control	`[[1, 2]]`	`[[1, 2]]`	Passed
normal control	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	`[[1500, 1501], [1500, 2000], [4889, 5139]]`	Passed
normal control	`[]`	`[]`	Passed
normal control	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	`[[499499, 1000499], [0, 1252001], [3895277, 4145777], [124250, 125753]]`	Passed

SHA-256 / 433bc9331cc0f07763c8047861f0963a7d6da20963f1d934d5785fa64c4a7d6b

Verification & scope

A deterministic bounded teaching model with a stipulated toy contract; it does not claim conformance to any subtitle standard. 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 2026-09-29T14:49:32.014237+00:00.

Case digest / b95d435c7df68f4201249dd78a156b00cec9078ead419af8c1764571ae6b0c1d