test: final source recheck for V1 jet CV and V3 Taylor-Green runs

Both runners now re-derive the recorded source identity after the run
(source_provenance.final_source_check) and only then mark the receipt
complete. A changed source, or a recheck that fails, rewrites the receipt
as invalid and raises:
- V1: in-run receipts carry run_status=INCOMPLETE; finish_receipt sets
  COMPLETE or INVALID_SOURCE_CHANGED (+ source_recheck/source_after).
- V3: completed stays False, invalid=SOURCE_CHANGED_DURING_RUN, no
  screening_pass; the assessor requires source_recheck UNCHANGED (frozen
  mode still accepts receipts from before the field existed).
Interrupted runs keep INCOMPLETE / completed=False. Host tests cover
unchanged, changed and failed rechecks and incomplete runs. No threshold
or solver change.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
ClaudeandClaude Opus 5.5 committed 2026-10-20 18:44:23 +00:00
1 parent 8cae58cdf1
commit df0f586331
7 files changed
+159 -15

No files matched your search

+36 -11
View File
@@ -244,6 +244,37 @@ def current_checkout(repo):
for f in files})
def source_identity(repo, current):
"""Recorded source of a run; recomputed after the run by finish_receipt."""
if current:
provenance = dict(**checkout_provenance(repo), **current_checkout(repo))
if provenance["checkout_dirty"]:
raise RuntimeError("Current-checkout runs require a clean checkout")
return provenance
source = verify(repo)
return dict(source_experiment=source["experiment"], source_mode="pinned",
**checkout_provenance(repo))
def finish_receipt(path, payload, provenance, recheck):
"""Mark the receipt COMPLETE only if the source is unchanged; otherwise INVALID.
Receipts written during the run carry run_status="INCOMPLETE", so an interrupted
run never leaves a complete-looking receipt.
"""
from source_provenance import final_source_check
fields = final_source_check(provenance, recheck)
unchanged = fields["source_recheck"] == "UNCHANGED"
payload = dict(payload, **fields,
run_status="COMPLETE" if unchanged else "INVALID_SOURCE_CHANGED")
temporary = Path(str(path) + ".tmp")
temporary.write_text(json.dumps(payload, indent=2) + "\n")
temporary.replace(path)
if not unchanged:
raise RuntimeError("Source changed during the run; receipt marked invalid")
return payload
def run_case(c, ctx, output, provenance, continuation=False):
from CelerisLab.body.geometry.circle import CircleGeometry
from CelerisLab.body.objects import SimObject
@@ -556,6 +587,7 @@ def run_case(c, ctx, output, provenance, continuation=False):
case=c,
**{"source_revision": PIN, **provenance},
status="PILOT_DIAGNOSTIC_ONLY",
run_status="INCOMPLETE", # final source recheck in main() completes it
windows=rows,
physics_validation_pass=False,
precision="FP32 evolution, FP64 host diagnostics",
@@ -638,16 +670,7 @@ def main():
):
p.error("Root must select one verified idle GPU via CUDA_VISIBLE_DEVICES")
repo = a.repo.resolve()
if a.current_checkout:
provenance = dict(**checkout_provenance(repo), **current_checkout(repo))
if provenance["checkout_dirty"]:
raise RuntimeError("Current-checkout runs require a clean checkout")
else:
source = verify(repo)
provenance = dict(
source_experiment=source["experiment"], source_mode="pinned",
**checkout_provenance(repo)
)
provenance = source_identity(repo, a.current_checkout)
sys.path.insert(0, str(repo / "src"))
import CelerisLab
@@ -663,7 +686,9 @@ def main():
free, total = cuda.mem_get_info()
if free < max(1024, 2 * plan["estimated_peak_device_mib"]) * 1024**2:
raise RuntimeError("Insufficient free device memory")
run_case(c, ctx, a.output, provenance, continuation=a.continuation)
payload = run_case(c, ctx, a.output, provenance, continuation=a.continuation)
finish_receipt(a.output / f"re{c['re_d']}_s{c['scale']}.json", payload, provenance,
lambda: source_identity(repo, a.current_checkout))
print(
json.dumps(
dict(status="COMPLETE_DIAGNOSTIC_ONLY", physics_validation_pass=False)