FA-26956 / HTTP ranges / Open access
Partial cache admission publishes complete requested fragments atomically: a contradicting duplicate fragment invalidates admission · case 01
Partial cache admission publishes complete requested fragments atomically: a contradicting duplicate fragment invalidates admission.
ROOT CAUSE
The partial-admission-transaction-duplicate-conflict decision uses visible=dict(x[0]) pending=None reads=[] for event in x[1]: op=event[0] if op=='begin': if pending is not None: continue pending=[event[1],set(event[2]),{}] elif op=='put': if pending is None or event[1]!=pending[0]: continue at,char=event[2:] if at not in pending[1]: continue if at in pending[2] and pending[2][at]!=char: continue pending[2][at]=char elif op=='commit': if pending is None or event[1]!=pending[0]: continue if set(pending[2])!=pending[1]: continue visible=dict(pending[2]) pending=None elif op=='abort': if pending is not None and event[1]==pending[0]: pending=None elif op=='read': reads.append(sorted(visible.items())) return [reads,sorted(visible.items()),pending[0] if pending is not None else None].
THE FAILURE
The partial-admission-transaction-duplicate-conflict decision uses visible=dict(x[0]) pending=None reads=[] for event in x[1]: op=event[0] if op=='begin': if pending is not None: continue pending=[event[1],set(event[2]),{}] elif op=='put': if pending is None or event[1]!=pending[0]: continue at,char=event[2:] if at not in pending[1]: continue if at in pending[2] and pending[2][at]!=char: continue pending[2][at]=char elif op=='commit': if pending is None or event[1]!=pending[0]: continue if set(pending[2])!=pending[1]: continue visible=dict(pending[2]) pending=None elif op=='abort': if pending is not None and event[1]==pending[0]: pending=None elif op=='read': reads.append(sorted(visible.items())) return [reads,sorted(visible.items()),pending[0] if pending is not None else None].
Unsuccessful approach: The partial repair uses visible=dict(x[0]) pending=None reads=[] for event in x[1]: op=event[0] if op=='begin': if pending is not None: continue pending=[event[1],set(event[2]),{}] elif op=='put': if pending is None or event[1]!=pending[0]: continue at,char=event[2:] if at not in pending[1]: continue if at in pending[2] and pending[2][at]!=char: pending[2][at]=char continue pending[2][at]=char elif op=='commit': if pending is None or event[1]!=pending[0]: continue if set(pending[2])!=pending[1]: continue visible=dict(pending[2]) pending=None elif op=='abort': if pending is not None and event[1]==pending[0]: pending=None elif op=='read': reads.append(sorted(visible.items())) return [reads,sorted(visible.items()),pending[0] if pending is not None else None], which still violates the stated contract.
Case contract
x is [old-visible-cells,events]. Events begin(id,expected-offsets), put(id,offset,char), commit(id), abort(id), read. One transaction at a time; unknown or mismatching ids ignored. A conflicting duplicate byte invalidates the active transaction. Commit succeeds only when every expected cell is present and replaces the old generation. Readers see committed cells sorted by offset. Return [read-snapshots,final-cells,pending-id].
Why this case matters
Range responses combine representation identity, conditional requests, framing, and partial-object state.
1 / The failure
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
N = 1
observations = []
def solve(x):
visible=dict(x[0])
pending=None
reads=[]
for event in x[1]:
op=event[0]
if op=='begin':
if pending is not None: continue
pending=[event[1],set(event[2]),{}]
elif op=='put':
if pending is None or event[1]!=pending[0]: continue
at,char=event[2:]
if at not in pending[1]: continue
if at in pending[2] and pending[2][at]!=char:
continue
pending[2][at]=char
elif op=='commit':
if pending is None or event[1]!=pending[0]: continue
if set(pending[2])!=pending[1]: continue
visible=dict(pending[2])
pending=None
elif op=='abort':
if pending is not None and event[1]==pending[0]: pending=None
elif op=='read':
reads.append(sorted(visible.items()))
return [reads,sorted(visible.items()),pending[0] if pending is not None else None]
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('duplicate-conflict fixture 0', json.loads(json.dumps(solve([[[0,'o'],[9,'z']],[['begin','a',[0,1]],['put','a',1,'b'],['read'],['put','a',0,'a'],['commit','a'],['read']]]))), json.loads(json.dumps([[[[0,'o'],[9,'z']],[[0,'a'],[1,'b']]],[[0,'a'],[1,'b']],None])))
check('duplicate-conflict fixture 1', json.loads(json.dumps(solve([[],[['begin','a',[0,1]],['put','a',0,'a'],['commit','a'],['read']]]))), json.loads(json.dumps([[[]],[],'a'])))
check('duplicate-conflict fixture 2', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','wrong',0,'x'],['put','a',1,'y'],['commit','a'],['read']]]))), json.loads(json.dumps([[[]],[],'a'])))
check('duplicate-conflict fixture 3', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',1,'b'],['commit','a']]]))), json.loads(json.dumps([[],[[0,'a']],None])))
check('duplicate-conflict fixture 4', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',0,'b'],['commit','a']]]))), json.loads(json.dumps([[],[],None])))
check('duplicate-conflict fixture 5', json.loads(json.dumps(solve([[],[['begin','a',[0]],['begin','b',[]],['abort','b'],['put','a',0,'a'],['commit','a']]]))), json.loads(json.dumps([[],[[0,'a']],None])))
check('duplicate-conflict fixture 6', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',0,'a'],['commit','wrong'],['read'],['commit','a']]]))), json.loads(json.dumps([[[]],[[0,'a']],None])))
check('duplicate-conflict fixture 7', json.loads(json.dumps(solve([[],[['begin','a',[]],['commit','a'],['begin','b',[N]],['put','b',N,'p'],['commit','b']]]))), json.loads(json.dumps([[],[[N,'p']],None])))
check('duplicate-conflict fixture 8', json.loads(json.dumps(solve([[[0,'o']],[['begin','a',[0]],['put','a',0,'a'],['abort','a'],['read']]]))), json.loads(json.dumps([[[[0,'o']]],[[0,'o']],None])))
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 |
|---|---|---|---|
| duplicate-conflict fixture 0 | [[[[0, 'o'], [9, 'z']], [[0, 'a'], [1, 'b']]], [[0, 'a'], [1, 'b']], None] | [[[[0, 'o'], [9, 'z']], [[0, 'a'], [1, 'b']]], [[0, 'a'], [1, 'b']], None] | Passed |
| duplicate-conflict fixture 1 | [[[]], [], 'a'] | [[[]], [], 'a'] | Passed |
| duplicate-conflict fixture 2 | [[[]], [], 'a'] | [[[]], [], 'a'] | Passed |
| duplicate-conflict fixture 3 | [[], [[0, 'a']], None] | [[], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 4 | [[], [[0, 'a']], None] | [[], [], None] | Failed |
| duplicate-conflict fixture 5 | [[], [[0, 'a']], None] | [[], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 6 | [[[]], [[0, 'a']], None] | [[[]], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 7 | [[], [[1, 'p']], None] | [[], [[1, 'p']], None] | Passed |
| duplicate-conflict fixture 8 | [[[[0, 'o']]], [[0, 'o']], None] | [[[[0, 'o']]], [[0, 'o']], None] | Passed |
SHA-256 / 4d3397717a88cbbe98bc80e4a38c3e5f09879c99fb9bb9a3aace1821e0a307bf
2 / The unsuccessful fix
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
N = 1
observations = []
def solve(x):
visible=dict(x[0])
pending=None
reads=[]
for event in x[1]:
op=event[0]
if op=='begin':
if pending is not None: continue
pending=[event[1],set(event[2]),{}]
elif op=='put':
if pending is None or event[1]!=pending[0]: continue
at,char=event[2:]
if at not in pending[1]: continue
if at in pending[2] and pending[2][at]!=char:
pending[2][at]=char
continue
pending[2][at]=char
elif op=='commit':
if pending is None or event[1]!=pending[0]: continue
if set(pending[2])!=pending[1]: continue
visible=dict(pending[2])
pending=None
elif op=='abort':
if pending is not None and event[1]==pending[0]: pending=None
elif op=='read':
reads.append(sorted(visible.items()))
return [reads,sorted(visible.items()),pending[0] if pending is not None else None]
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('duplicate-conflict fixture 0', json.loads(json.dumps(solve([[[0,'o'],[9,'z']],[['begin','a',[0,1]],['put','a',1,'b'],['read'],['put','a',0,'a'],['commit','a'],['read']]]))), json.loads(json.dumps([[[[0,'o'],[9,'z']],[[0,'a'],[1,'b']]],[[0,'a'],[1,'b']],None])))
check('duplicate-conflict fixture 1', json.loads(json.dumps(solve([[],[['begin','a',[0,1]],['put','a',0,'a'],['commit','a'],['read']]]))), json.loads(json.dumps([[[]],[],'a'])))
check('duplicate-conflict fixture 2', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','wrong',0,'x'],['put','a',1,'y'],['commit','a'],['read']]]))), json.loads(json.dumps([[[]],[],'a'])))
check('duplicate-conflict fixture 3', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',1,'b'],['commit','a']]]))), json.loads(json.dumps([[],[[0,'a']],None])))
check('duplicate-conflict fixture 4', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',0,'b'],['commit','a']]]))), json.loads(json.dumps([[],[],None])))
check('duplicate-conflict fixture 5', json.loads(json.dumps(solve([[],[['begin','a',[0]],['begin','b',[]],['abort','b'],['put','a',0,'a'],['commit','a']]]))), json.loads(json.dumps([[],[[0,'a']],None])))
check('duplicate-conflict fixture 6', json.loads(json.dumps(solve([[],[['begin','a',[0]],['put','a',0,'a'],['put','a',0,'a'],['commit','wrong'],['read'],['commit','a']]]))), json.loads(json.dumps([[[]],[[0,'a']],None])))
check('duplicate-conflict fixture 7', json.loads(json.dumps(solve([[],[['begin','a',[]],['commit','a'],['begin','b',[N]],['put','b',N,'p'],['commit','b']]]))), json.loads(json.dumps([[],[[N,'p']],None])))
check('duplicate-conflict fixture 8', json.loads(json.dumps(solve([[[0,'o']],[['begin','a',[0]],['put','a',0,'a'],['abort','a'],['read']]]))), json.loads(json.dumps([[[[0,'o']]],[[0,'o']],None])))
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 |
|---|---|---|---|
| duplicate-conflict fixture 0 | [[[[0, 'o'], [9, 'z']], [[0, 'a'], [1, 'b']]], [[0, 'a'], [1, 'b']], None] | [[[[0, 'o'], [9, 'z']], [[0, 'a'], [1, 'b']]], [[0, 'a'], [1, 'b']], None] | Passed |
| duplicate-conflict fixture 1 | [[[]], [], 'a'] | [[[]], [], 'a'] | Passed |
| duplicate-conflict fixture 2 | [[[]], [], 'a'] | [[[]], [], 'a'] | Passed |
| duplicate-conflict fixture 3 | [[], [[0, 'a']], None] | [[], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 4 | [[], [[0, 'b']], None] | [[], [], None] | Failed |
| duplicate-conflict fixture 5 | [[], [[0, 'a']], None] | [[], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 6 | [[[]], [[0, 'a']], None] | [[[]], [[0, 'a']], None] | Passed |
| duplicate-conflict fixture 7 | [[], [[1, 'p']], None] | [[], [[1, 'p']], None] | Passed |
| duplicate-conflict fixture 8 | [[[[0, 'o']]], [[0, 'o']], None] | [[[[0, 'o']]], [[0, 'o']], None] | Passed |
SHA-256 / 9653faa2d4205f4435a04d84571ec4bb03c05cc43e8e439cbacab2e56971da94
HELD IN THE MEMBER ARCHIVE
The verified repair and its recorded checks are member-only.
This mechanism has 9 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.
Sign in to the archive ↗Verification & scope
Deterministic simplified range service, with stipulated local policies and already parsed trusted inputs; not a complete HTTP 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 2026-09-29T14:41:20.639618+00:00.
Case digest / c6893763ac9429c7efd2965865fabb2e15ee1169a9a273c0d9a7e870eb30c7e1