Datasets:
Formats:
parquet
Size:
100K - 1M
Tags:
population-genetics
demographic-inference
effective-population-size
coalescent
msprime
simulation
License:
File size: 14,128 Bytes
ee5b44c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 | from __future__ import annotations
import gzip
import hashlib
import json
from concurrent.futures import ThreadPoolExecutor
from pathlib import Path
import pytest
from simulator.config import Config
from simulator.dataset_state import SAMPLE_GENERATOR_SCHEMA_VERSION, generation_config_state
from simulator import replacement as v1
from simulator import replacement_pool as pool
GENERATION_PLAN_SHA = "9" * 64
def _canonical(value: object) -> str:
return hashlib.sha256(
json.dumps(
value,
ensure_ascii=False,
allow_nan=False,
sort_keys=True,
separators=(",", ":"),
).encode("utf-8")
).hexdigest()
def _write_json(path: Path, value: dict) -> None:
path.parent.mkdir(parents=True, exist_ok=True)
path.write_text(json.dumps(value), encoding="utf-8")
def _file_record(path: Path) -> dict:
return {
"size_bytes": path.stat().st_size,
"sha256": hashlib.sha256(path.read_bytes()).hexdigest(),
}
def _source_plan(root: Path, *, n_samples: int, seed: int) -> dict:
config = Config(
preset="full_v01",
n_samples=n_samples,
seed=seed,
shard_size=2,
compute_chunk_size=1,
out_dir=str(root),
workers=1,
)
state = generation_config_state(config)
layout = {
"n_samples": n_samples,
"shard_size": 2,
"compute_chunk_size": 1,
"n_shards": (n_samples + 1) // 2,
"n_chunks": n_samples,
"sample_id_rule": "sim_{global_sample_index:08d}",
"seed_rule": "config.seed + global_sample_index * 7919",
"partition_rule": "chunk_idx_mod_world_size",
"array_schema_version": SAMPLE_GENERATOR_SCHEMA_VERSION,
}
plan = {
"schema_version": 1,
"kind": "dlcoalsim_generation_plan",
"sample_generation_config": state,
"layout": layout,
"layout_sha256": _canonical(layout),
"empirical_maps_sha256": hashlib.sha256(b"maps").hexdigest(),
"empirical_ne_templates_sha256": hashlib.sha256(b"templates").hexdigest(),
"runtime_environment_sha256": hashlib.sha256(b"runtime").hexdigest(),
}
plan["plan_sha256"] = _canonical(plan)
_write_json(root / "samples" / "PLAN.json", plan)
return plan
def _source_fragment(root: Path, plan: dict, index: int) -> None:
shard = index // 2
relative = f"samples/.fragments/fragment_{index:08d}.npz"
npz = root / relative
metadata = npz.with_suffix(".jsonl.gz")
npz.parent.mkdir(parents=True, exist_ok=True)
npz.write_bytes(f"source-npz-{index}".encode())
metadata.write_bytes(f"source-meta-{index}".encode())
receipt = {
"schema_version": 1,
"plan_sha256": plan["plan_sha256"],
"result": {
"chunk_idx": index,
"shard_idx": shard,
"start": index,
"n_samples": 1,
"fragment_path": relative,
},
"files": {
"npz": _file_record(npz),
"metadata_jsonl_gz": _file_record(metadata),
},
}
receipt["receipt_sha256"] = _canonical(receipt)
_write_json(
root
/ "samples"
/ ".fragments"
/ f"fragment_{index:08d}.done.json",
receipt,
)
def _patch_generator(
monkeypatch: pytest.MonkeyPatch,
output: Path,
runtime_sha256: str,
) -> list[list[str]]:
calls: list[list[str]] = []
monkeypatch.setattr(
v1,
"_runtime_probe",
lambda root, python: {
"simulator_file": str(root / "simulator" / "__init__.py"),
"runtime_environment_sha256": runtime_sha256,
},
)
def fake_run(
root,
python,
arguments,
*,
module=True,
capture_output=True,
):
calls.append(list(arguments))
if "--prepare-only" in arguments:
_write_json(output / "samples" / "PLAN.json", {"test": True})
return type(
"Completed",
(),
{"returncode": 0, "stdout": "", "stderr": ""},
)()
monkeypatch.setattr(v1, "_run_generator_subprocess", fake_run)
monkeypatch.setattr(
v1,
"_validate_replacement_generation_plan",
lambda out, config, source: {
"plan_sha256": GENERATION_PLAN_SHA,
**dict(source),
},
)
return calls
def _prepare(
tmp_path: Path,
monkeypatch: pytest.MonkeyPatch,
*,
name: str = "case",
n_samples: int = 5,
completed: int = 2,
candidate_count: int = 5,
) -> tuple[Path, Path, dict, list[list[str]]]:
source = tmp_path / f"{name}-source"
source_plan = _source_plan(source, n_samples=n_samples, seed=100)
for index in range(completed):
_source_fragment(source, source_plan, index)
output = tmp_path / f"{name}-pool"
calls = _patch_generator(
monkeypatch, output, source_plan["runtime_environment_sha256"]
)
generator_root = tmp_path / f"{name}-generator"
(generator_root / "simulator").mkdir(parents=True)
(generator_root / "simulator" / "__init__.py").write_text("")
generator_python = tmp_path / f"{name}-python"
generator_python.write_text("")
plan = pool.prepare_pool(
[source],
target_total=n_samples,
candidate_count=candidate_count,
out_dir=output,
generator_root=generator_root,
generator_python=generator_python,
)
return output, source, plan, calls
def _candidate_fragment(
output: Path,
plan: dict,
index: int,
*,
seed: int | None = None,
) -> None:
expected = plan["candidates"][index]
npz = output / expected["npz_path"]
metadata = output / expected["metadata_path"]
receipt_path = output / expected["receipt_path"]
npz.parent.mkdir(parents=True, exist_ok=True)
npz.write_bytes(f"candidate-npz-{index}".encode())
with gzip.open(metadata, "wt", encoding="utf-8") as handle:
handle.write(
json.dumps(
{
"sample_id": expected["expected_sample_id"],
"sample_index": index,
"seed": expected["seed"] if seed is None else seed,
# These fields must never participate in pool selection.
"target": [999],
"n_variants": 999,
"demography_type": "ignored",
}
)
+ "\n"
)
receipt = {
"schema_version": 1,
"plan_sha256": plan["generation_plan_sha256"],
"result": {
"chunk_idx": index,
"shard_idx": expected["shard_index"],
"start": index,
"n_samples": 1,
"fragment_path": expected["npz_path"],
},
"files": {
"npz": _file_record(npz),
"metadata_jsonl_gz": _file_record(metadata),
},
}
receipt["receipt_sha256"] = _canonical(receipt)
_write_json(receipt_path, receipt)
def test_k3_c5_freezes_out_of_order_prefix_and_late_is_immutable(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, _ = _prepare(tmp_path, monkeypatch)
assert plan["required_k"] == 3
assert plan["candidate_count"] == 5
assert plan["candidate_config"]["n_samples"] == 5
assert plan["candidate_config"]["compute_chunk_size"] == 1
assert plan["candidate_config"]["workers"] == 1
assert plan["candidate_groups"]["primary"]["candidate_indices"] == [0, 1, 2]
assert plan["candidate_groups"]["standby"]["candidate_indices"] == [3, 4]
for index in (4, 1, 3):
_candidate_fragment(output, plan, index)
result = pool.register_candidate(output, index)
assert result["selection_frozen"] is True
selection_path = output / pool.SELECTION_NAME
before = selection_path.read_bytes()
selection = json.loads(before)
assert selection["selected_candidate_indices"] == [4, 1, 3]
assert selection["registration_cutoff_ordinal"] == 3
assert [item["registration_ordinal"] for item in selection["selected_candidates"]] == [
1,
2,
3,
]
assert selection["runtime_censoring"]["applied"] is True
assert (
selection["selection_rule"]["target_variant_demography_fields_read"]
is False
)
for item in selection["selected_candidates"]:
for key in ("receipt", "npz", "metadata_jsonl_gz"):
source_record = item["source_artifacts"][key]
sealed_record = item["sealed_artifacts"][key]
assert sealed_record["size_bytes"] == source_record["size_bytes"]
assert sealed_record["sha256"] == source_record["sha256"]
assert (output / sealed_record["path"]).samefile(
output / source_record["path"]
)
_candidate_fragment(output, plan, 0)
assert pool.register_candidate(output, 0)["status"] == "selection_frozen"
assert pool.register_candidate(output, 4)["status"] == "already_selected"
assert selection_path.read_bytes() == before
assert pool.pool_status(output)["selected_candidate_indices"] == [4, 1, 3]
def test_concurrent_registration_has_one_unique_k_prefix(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, _ = _prepare(tmp_path, monkeypatch, name="concurrent")
for index in range(5):
_candidate_fragment(output, plan, index)
with ThreadPoolExecutor(max_workers=5) as executor:
results = list(
executor.map(lambda index: pool.register_candidate(output, index), range(5))
)
assert all(result["status"] in {"registered", "selection_frozen"} for result in results)
selection = json.loads((output / pool.SELECTION_NAME).read_text())
selected = selection["selected_candidate_indices"]
assert len(selected) == 3
assert len(set(selected)) == 3
assert sorted(
item["registration_ordinal"]
for item in selection["selected_candidates"]
) == [1, 2, 3]
registry_before = (output / pool.REGISTRY_NAME).read_bytes()
pool.register_candidate(output, selected[0])
assert (output / pool.REGISTRY_NAME).read_bytes() == registry_before
@pytest.mark.parametrize("failure", ["receipt", "payload", "seed"])
def test_bad_receipt_hash_or_seed_fails_before_registration(
tmp_path: Path,
monkeypatch: pytest.MonkeyPatch,
failure: str,
) -> None:
output, _, plan, _ = _prepare(
tmp_path, monkeypatch, name=f"bad-{failure}"
)
_candidate_fragment(output, plan, 0)
expected = plan["candidates"][0]
if failure == "receipt":
receipt_path = output / expected["receipt_path"]
receipt = json.loads(receipt_path.read_text())
receipt["result"]["start"] = 1
_write_json(receipt_path, receipt)
elif failure == "payload":
(output / expected["npz_path"]).write_bytes(b"tampered")
else:
_candidate_fragment(output, plan, 0, seed=expected["seed"] + 1)
with pytest.raises(pool.ReplacementPoolError):
pool.register_candidate(output, 0)
assert not (output / pool.SELECTION_NAME).exists()
assert pool.pool_status(output)["registered_count"] == 0
def test_less_than_k_does_not_publish_and_missing_candidate_does_not_block(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, _ = _prepare(tmp_path, monkeypatch, name="missing")
for index in (1, 2):
_candidate_fragment(output, plan, index)
pool.register_candidate(output, index)
status = pool.pool_status(output)
assert status["registered_count"] == 2
assert status["remaining_to_freeze"] == 1
assert not (output / pool.SELECTION_NAME).exists()
# Candidate zero never produces a receipt. Candidate three is enough to
# freeze the exact K=3 prefix without waiting for it.
_candidate_fragment(output, plan, 3)
pool.register_candidate(output, 3)
assert pool.pool_status(output)["selected_candidate_indices"] == [1, 2, 3]
assert not (output / plan["candidates"][0]["receipt_path"]).exists()
def test_selected_payload_tamper_is_detected_through_sealed_hardlink(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, _ = _prepare(tmp_path, monkeypatch, name="tamper")
for index in (0, 1, 2):
_candidate_fragment(output, plan, index)
pool.register_candidate(output, index)
(output / plan["candidates"][1]["npz_path"]).write_bytes(b"changed")
with pytest.raises(pool.ReplacementPoolError, match="sealed artifact"):
pool.pool_status(output)
def test_generate_requires_one_rank_per_candidate(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, calls = _prepare(tmp_path, monkeypatch, name="generate")
with pytest.raises(pool.ReplacementPoolError, match="world_size"):
pool.generate_candidate(output, rank=0, world_size=4)
result = pool.generate_candidate(output, rank=4, world_size=5)
assert result["candidate_index"] == 4
assert result["seed"] == plan["candidates"][4]["seed"]
assert calls[-1][-5:] == [
"--generate-only",
"--rank",
"4",
"--world-size",
"5",
]
def test_k_zero_is_an_empty_frozen_noop(
tmp_path: Path, monkeypatch: pytest.MonkeyPatch
) -> None:
output, _, plan, calls = _prepare(
tmp_path,
monkeypatch,
name="zero",
n_samples=2,
completed=2,
candidate_count=0,
)
assert plan["required_k"] == 0
assert plan["candidate_config"] is None
assert not any("--prepare-only" in call for call in calls)
selection = json.loads((output / pool.SELECTION_NAME).read_text())
assert selection["selected_candidates"] == []
assert selection["runtime_censoring"]["applied"] is False
assert pool.generate_candidate(output) == {
"no_op": True,
"required_k": 0,
"candidate_count": 0,
}
status = pool.pool_status(output)
assert status["status"] == "complete"
assert status["no_op"] is True
|