FA-27531 / HTTP ranges / Open access
A zero-copy range sender accounts for partial progress reported with syscall errors: a clean zero-byte return with remaining bytes is truncation · case 01
A zero-copy range sender accounts for partial progress reported with syscall errors: a clean zero-byte return with remaining bytes is truncation.
ROOT CAUSE
The sendfile-range-events-zero-success-eof decision uses offset=x['offset'] remaining=x['remaining'] state='running' opened=True out=[] for accepted,error in x['events']: if state in ('complete','aborted','truncated'): out.append([offset,remaining,state,opened]) continue if accepted<0 or accepted>remaining: state='aborted' opened=False out.append([offset,remaining,state,opened]) continue offset+=accepted remaining-=accepted if error=='fatal': state='aborted' elif remaining==0: state='complete' elif error=='again': state='blocked' elif error=='interrupt': state='running' elif accepted==0: state='running' else: state='running' opened=state not in ('complete','aborted','truncated') out.append([offset,remaining,state,opened]) return out.
THE FAILURE
The sendfile-range-events-zero-success-eof decision uses offset=x['offset'] remaining=x['remaining'] state='running' opened=True out=[] for accepted,error in x['events']: if state in ('complete','aborted','truncated'): out.append([offset,remaining,state,opened]) continue if accepted<0 or accepted>remaining: state='aborted' opened=False out.append([offset,remaining,state,opened]) continue offset+=accepted remaining-=accepted if error=='fatal': state='aborted' elif remaining==0: state='complete' elif error=='again': state='blocked' elif error=='interrupt': state='running' elif accepted==0: state='running' else: state='running' opened=state not in ('complete','aborted','truncated') out.append([offset,remaining,state,opened]) return out.
Unsuccessful approach: The partial repair uses offset=x['offset'] remaining=x['remaining'] state='running' opened=True out=[] for accepted,error in x['events']: if state in ('complete','aborted','truncated'): out.append([offset,remaining,state,opened]) continue if accepted<0 or accepted>remaining: state='aborted' opened=False out.append([offset,remaining,state,opened]) continue offset+=accepted remaining-=accepted if error=='fatal': state='aborted' elif remaining==0: state='complete' elif error=='again': state='blocked' elif error=='interrupt': state='running' elif accepted==0: state='blocked' else: state='running' opened=state not in ('complete','aborted','truncated') out.append([offset,remaining,state,opened]) return out, which still violates the stated contract.
Case contract
x contains absolute offset, remaining promised bytes, and syscall events [accepted,error], where error is none,again,interrupt,fatal. Counts outside0..remaining abort without advancing. Valid accepted bytes advance even when again or interrupt accompanies them. Fatal errors abort; zero remaining completes; again blocks; interrupt reschedules; zero bytes with none truncates. Terminal states ignore later events. Return state snapshots [offset,remaining,state,handle-open] after each event.
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):
offset=x['offset']
remaining=x['remaining']
state='running'
opened=True
out=[]
for accepted,error in x['events']:
if state in ('complete','aborted','truncated'):
out.append([offset,remaining,state,opened])
continue
if accepted<0 or accepted>remaining:
state='aborted'
opened=False
out.append([offset,remaining,state,opened])
continue
offset+=accepted
remaining-=accepted
if error=='fatal': state='aborted'
elif remaining==0: state='complete'
elif error=='again': state='blocked'
elif error=='interrupt': state='running'
elif accepted==0: state='running'
else: state='running'
opened=state not in ('complete','aborted','truncated')
out.append([offset,remaining,state,opened])
return out
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('zero-success-eof fixture 0', json.loads(json.dumps(solve({'offset':10,'remaining':5,'events':[[2,'interrupt'],[1,'again'],[2,'none']]}))), json.loads(json.dumps([[12,3,'running',True],[13,2,'blocked',True],[15,0,'complete',False]])))
check('zero-success-eof fixture 1', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[0,'again'],[0,'interrupt'],[0,'none'],[1,'none']]}))), json.loads(json.dumps([[0,2,'blocked',True],[0,2,'running',True],[0,2,'truncated',False],[0,2,'truncated',False]])))
check('zero-success-eof fixture 2', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[1,'fatal'],[1,'none']]}))), json.loads(json.dumps([[1,1,'aborted',False],[1,1,'aborted',False]])))
check('zero-success-eof fixture 3', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[-1,'none']]}))), json.loads(json.dumps([[0,2,'aborted',False]])))
check('zero-success-eof fixture 4', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[3,'none']]}))), json.loads(json.dumps([[0,2,'aborted',False]])))
check('zero-success-eof fixture 5', json.loads(json.dumps(solve({'offset':N,'remaining':N,'events':[[N,'again'],[0,'fatal']]}))), json.loads(json.dumps([[2*N,0,'complete',False],[2*N,0,'complete',False]])))
check('zero-success-eof fixture 6', json.loads(json.dumps(solve({'offset':0,'remaining':0,'events':[[0,'none']]}))), json.loads(json.dumps([[0,0,'complete',False]])))
check('zero-success-eof fixture 7', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[2,'fatal']]}))), json.loads(json.dumps([[2,0,'aborted',False]])))
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 |
|---|---|---|---|
| zero-success-eof fixture 0 | [[12, 3, 'running', True], [13, 2, 'blocked', True], [15, 0, 'complete', False]] | [[12, 3, 'running', True], [13, 2, 'blocked', True], [15, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 1 | [[0, 2, 'blocked', True], [0, 2, 'running', True], [0, 2, 'running', True], [1, 1, 'running', True]] | [[0, 2, 'blocked', True], [0, 2, 'running', True], [0, 2, 'truncated', False], [0, 2, 'truncated', False]] | Failed |
| zero-success-eof fixture 2 | [[1, 1, 'aborted', False], [1, 1, 'aborted', False]] | [[1, 1, 'aborted', False], [1, 1, 'aborted', False]] | Passed |
| zero-success-eof fixture 3 | [[0, 2, 'aborted', False]] | [[0, 2, 'aborted', False]] | Passed |
| zero-success-eof fixture 4 | [[0, 2, 'aborted', False]] | [[0, 2, 'aborted', False]] | Passed |
| zero-success-eof fixture 5 | [[2, 0, 'complete', False], [2, 0, 'complete', False]] | [[2, 0, 'complete', False], [2, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 6 | [[0, 0, 'complete', False]] | [[0, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 7 | [[2, 0, 'aborted', False]] | [[2, 0, 'aborted', False]] | Passed |
SHA-256 / e2c8e8bcb7091d525399cb8db07917ea830e474bc9f2788e05bf8b5122b3561d
2 / The unsuccessful fix
Exit 1"""Failure Map reference implementation. Python standard library only."""
import json
N = 1
observations = []
def solve(x):
offset=x['offset']
remaining=x['remaining']
state='running'
opened=True
out=[]
for accepted,error in x['events']:
if state in ('complete','aborted','truncated'):
out.append([offset,remaining,state,opened])
continue
if accepted<0 or accepted>remaining:
state='aborted'
opened=False
out.append([offset,remaining,state,opened])
continue
offset+=accepted
remaining-=accepted
if error=='fatal': state='aborted'
elif remaining==0: state='complete'
elif error=='again': state='blocked'
elif error=='interrupt': state='running'
elif accepted==0: state='blocked'
else: state='running'
opened=state not in ('complete','aborted','truncated')
out.append([offset,remaining,state,opened])
return out
def check(label, actual, expected):
observations.append({"check": label, "actual": actual, "expected": expected, "passed": actual == expected})
check('zero-success-eof fixture 0', json.loads(json.dumps(solve({'offset':10,'remaining':5,'events':[[2,'interrupt'],[1,'again'],[2,'none']]}))), json.loads(json.dumps([[12,3,'running',True],[13,2,'blocked',True],[15,0,'complete',False]])))
check('zero-success-eof fixture 1', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[0,'again'],[0,'interrupt'],[0,'none'],[1,'none']]}))), json.loads(json.dumps([[0,2,'blocked',True],[0,2,'running',True],[0,2,'truncated',False],[0,2,'truncated',False]])))
check('zero-success-eof fixture 2', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[1,'fatal'],[1,'none']]}))), json.loads(json.dumps([[1,1,'aborted',False],[1,1,'aborted',False]])))
check('zero-success-eof fixture 3', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[-1,'none']]}))), json.loads(json.dumps([[0,2,'aborted',False]])))
check('zero-success-eof fixture 4', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[3,'none']]}))), json.loads(json.dumps([[0,2,'aborted',False]])))
check('zero-success-eof fixture 5', json.loads(json.dumps(solve({'offset':N,'remaining':N,'events':[[N,'again'],[0,'fatal']]}))), json.loads(json.dumps([[2*N,0,'complete',False],[2*N,0,'complete',False]])))
check('zero-success-eof fixture 6', json.loads(json.dumps(solve({'offset':0,'remaining':0,'events':[[0,'none']]}))), json.loads(json.dumps([[0,0,'complete',False]])))
check('zero-success-eof fixture 7', json.loads(json.dumps(solve({'offset':0,'remaining':2,'events':[[2,'fatal']]}))), json.loads(json.dumps([[2,0,'aborted',False]])))
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 |
|---|---|---|---|
| zero-success-eof fixture 0 | [[12, 3, 'running', True], [13, 2, 'blocked', True], [15, 0, 'complete', False]] | [[12, 3, 'running', True], [13, 2, 'blocked', True], [15, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 1 | [[0, 2, 'blocked', True], [0, 2, 'running', True], [0, 2, 'blocked', True], [1, 1, 'running', True]] | [[0, 2, 'blocked', True], [0, 2, 'running', True], [0, 2, 'truncated', False], [0, 2, 'truncated', False]] | Failed |
| zero-success-eof fixture 2 | [[1, 1, 'aborted', False], [1, 1, 'aborted', False]] | [[1, 1, 'aborted', False], [1, 1, 'aborted', False]] | Passed |
| zero-success-eof fixture 3 | [[0, 2, 'aborted', False]] | [[0, 2, 'aborted', False]] | Passed |
| zero-success-eof fixture 4 | [[0, 2, 'aborted', False]] | [[0, 2, 'aborted', False]] | Passed |
| zero-success-eof fixture 5 | [[2, 0, 'complete', False], [2, 0, 'complete', False]] | [[2, 0, 'complete', False], [2, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 6 | [[0, 0, 'complete', False]] | [[0, 0, 'complete', False]] | Passed |
| zero-success-eof fixture 7 | [[2, 0, 'aborted', False]] | [[2, 0, 'aborted', False]] | Passed |
SHA-256 / 990a7f225a4c5acbd0a2994969e716c6294f29cedd4cbecade7a4fb6f0426c02
HELD IN THE MEMBER ARCHIVE
The verified repair and its recorded checks are member-only.
This mechanism has 8 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 2026-09-29T14:41:25.873226+00:00.
Case digest / 0e453990f4418a0b850251f47f3fe818e96619cb07c0c80b9a5d6cdb42c88556