FA-12216 / Navigation route constraints / Open access

Roundabout instructions count entrances as exits · case 01

Roundabout instructions count entrances as exits in the controlled route-policy fixture.

Verified by executionVariant 1 · 7 checks per implementationDownload source bundle ↓JSON ↗

ROOT CAUSE

Every encountered arm increments the exit ordinal, including arms that cannot be legally exited.

THE FAILURE

Every encountered arm increments the exit ordinal, including arms that cannot be legally exited.

Unsuccessful approach: Counting only public exits omits legal private exits explicitly permitted in the modeled route.

Case contract

Arms are ordered after entry as (ID, may_exit, public). Return one-based legal-exit ordinal for a legal selected arm, or None for an absent or forbidden arm.

Why this case matters

Navigation policies depend on road context that a shortest-path cost alone does not encode.

1 / The failure

Exit 1

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

N = 1
observations = []
def solve(arms, selected):
    for i,(name,allowed,public) in enumerate(arms):
        if name == selected:
            return i+1
    return None
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
a,b,c,d = N,N+10,N+20,N+30
arms = [(a,False,True),(b,True,True),(c,True,False),(d,True,True)]
check('entrance excluded', solve(arms,b), 1)
check('private legal exit counted', solve(arms,c), 2)
check('later exit ordinal', solve(arms,d), 3)
check('forbidden selected arm', solve(arms,a), None)
check('unknown arm', solve(arms,N+40), None)
check('empty roundabout', solve([],a), None)
check('first arm legal', solve([(a,True,True)],a), 1)
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
entrance excluded	`2`	`1`	Failed
private legal exit counted	`3`	`2`	Failed
later exit ordinal	`4`	`3`	Failed
forbidden selected arm	`1`	`None`	Failed
unknown arm	`None`	`None`	Passed
empty roundabout	`None`	`None`	Passed
first arm legal	`1`	`1`	Passed

SHA-256 / 7d5d434c74494f5369074f170bc25edae827d5681d8a77c48f9aa8d766e972be

2 / The unsuccessful fix

Exit 1

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

N = 1
observations = []
def solve(arms, selected):
    count = 0
    for name,allowed,public in arms:
        count += bool(allowed and public)
        if name == selected:
            return count if allowed and public else None
    return None
def check(label, actual, expected):
    observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
a,b,c,d = N,N+10,N+20,N+30
arms = [(a,False,True),(b,True,True),(c,True,False),(d,True,True)]
check('entrance excluded', solve(arms,b), 1)
check('private legal exit counted', solve(arms,c), 2)
check('later exit ordinal', solve(arms,d), 3)
check('forbidden selected arm', solve(arms,a), None)
check('unknown arm', solve(arms,N+40), None)
check('empty roundabout', solve([],a), None)
check('first arm legal', solve([(a,True,True)],a), 1)
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
entrance excluded	`1`	`1`	Passed
private legal exit counted	`None`	`2`	Failed
later exit ordinal	`2`	`3`	Failed
forbidden selected arm	`None`	`None`	Passed
unknown arm	`None`	`None`	Passed
empty roundabout	`None`	`None`	Passed
first arm legal	`1`	`1`	Passed

SHA-256 / 420ca4168ef759cacd6fe09101cdfab7b619acbcc70dae435bf2a274b6f20805

HELD IN THE MEMBER ARCHIVE

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

This mechanism has 7 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.

Verification & scope

Simplified deterministic offline policy model; not a production router or authoritative road guidance. 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 2026-09-29T14:38:54.922605+00:00.

Case digest / aee9c945267f1261fc4fdbc7405547aae8de0a1d62574614baa6e28084765b6d