Spaces:
Running
Running
File size: 11,533 Bytes
dbf7313 | 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 | from __future__ import annotations
import tempfile
from datetime import UTC, datetime
from pathlib import Path
from typing import Any
from huggingface_hub import HfApi
from slop_farmer.config import DatasetStatusOptions
from slop_farmer.data.hf_dataset_repo import (
list_remote_paths,
load_remote_file,
load_remote_json_file,
stable_snapshot_candidates,
)
from slop_farmer.data.parquet_io import read_json
from slop_farmer.data.snapshot_paths import (
CONTRIBUTOR_ARTIFACT_FILENAMES,
CURRENT_ANALYSIS_MANIFEST_PATH,
PR_SCOPE_CLUSTERS_FILENAME,
SNAPSHOTS_LATEST_PATH,
load_current_analysis_manifest,
repo_relative_path_to_local,
)
def _coerce_datetime(value: Any) -> datetime | None:
if not isinstance(value, str) or not value:
return None
try:
return datetime.fromisoformat(value.replace("Z", "+00:00"))
except ValueError:
return None
def _age_summary(value: str | None) -> dict[str, Any]:
timestamp = _coerce_datetime(value)
if timestamp is None:
return {"seconds": None, "summary": "unknown", "staleness": "unknown"}
age_seconds = max(int((datetime.now(tz=UTC) - timestamp).total_seconds()), 0)
if age_seconds <= 6 * 3600:
staleness = "fresh"
elif age_seconds <= 24 * 3600:
staleness = "aging"
else:
staleness = "stale"
if age_seconds < 3600:
summary = f"{age_seconds // 60}m"
elif age_seconds < 24 * 3600:
summary = f"{age_seconds // 3600}h"
else:
summary = f"{age_seconds // 86400}d"
return {"seconds": age_seconds, "summary": summary, "staleness": staleness}
def _local_status(output_dir: Path) -> dict[str, Any] | None:
latest_path = output_dir.resolve() / "snapshots" / "latest.json"
if not latest_path.exists():
return None
payload = read_json(latest_path)
snapshot_dir_raw = payload.get("snapshot_dir")
manifest: dict[str, Any] = {}
snapshot_dir: Path | None = None
if isinstance(snapshot_dir_raw, str) and snapshot_dir_raw:
snapshot_dir = Path(snapshot_dir_raw).resolve()
manifest_path = snapshot_dir / "manifest.json"
if manifest_path.exists():
manifest = read_json(manifest_path)
current_analysis = _local_current_analysis(snapshot_dir)
return {
"latest_path": str(latest_path),
"latest_pointer": payload,
"snapshot_dir": snapshot_dir_raw,
"snapshot_id": manifest.get("snapshot_id") or payload.get("latest_snapshot_id"),
"current_analysis": current_analysis,
}
def _local_current_analysis(snapshot_dir: Path | None) -> dict[str, Any]:
if snapshot_dir is None:
return {"present": False}
manifest_path = repo_relative_path_to_local(snapshot_dir, CURRENT_ANALYSIS_MANIFEST_PATH)
if not manifest_path.exists():
return {"present": False}
try:
manifest = load_current_analysis_manifest(manifest_path)
except ValueError as exc:
return {"present": True, "valid": False, "detail": str(exc)}
return {
"present": True,
"valid": True,
"snapshot_id": manifest["snapshot_id"],
"analysis_id": manifest["analysis_id"],
"variant": manifest["variant"],
"published_at": manifest["published_at"],
}
def _remote_status(repo_id: str, revision: str | None) -> dict[str, Any]:
api = HfApi()
with tempfile.TemporaryDirectory(prefix="slop-farmer-dataset-status-") as tmp:
root = Path(tmp)
remote_paths = list_remote_paths(api, repo_id, revision=revision)
latest_pointer = load_remote_json_file(
api,
repo_id,
SNAPSHOTS_LATEST_PATH,
root,
revision=revision,
)
watermark = load_remote_json_file(
api,
repo_id,
"state/watermark.json",
root,
revision=revision,
)
manifest = None
if latest_pointer is not None:
for candidate in stable_snapshot_candidates(latest_pointer, "manifest.json"):
downloaded = load_remote_file(
api,
repo_id,
candidate,
root,
revision=revision,
)
if downloaded is None:
continue
manifest = read_json(downloaded)
break
current_analysis = _remote_current_analysis(
api,
repo_id,
root,
revision=revision,
remote_paths=remote_paths,
latest_pointer=latest_pointer,
)
latest_snapshot_id = (
str(latest_pointer.get("latest_snapshot_id"))
if isinstance(latest_pointer, dict)
else None
)
archived_run_manifests = sorted(
path
for path in remote_paths
if path.startswith("snapshots/")
and "/analysis-runs/" in path
and path.endswith("/manifest.json")
)
current_snapshot_run_count = 0
if latest_snapshot_id:
current_snapshot_run_count = sum(
1
for path in archived_run_manifests
if path.startswith(f"snapshots/{latest_snapshot_id}/analysis-runs/")
)
extracted_at = manifest.get("extracted_at") if manifest else None
return {
"dataset_id": repo_id,
"revision": revision,
"latest_pointer": latest_pointer,
"watermark": watermark,
"manifest": manifest,
"cheap_artifacts": {
"pr_scope_clusters": _remote_has_latest_artifact(
remote_paths,
latest_pointer,
PR_SCOPE_CLUSTERS_FILENAME,
),
"contributors": all(
_remote_has_latest_artifact(remote_paths, latest_pointer, filename)
for filename in CONTRIBUTOR_ARTIFACT_FILENAMES
),
},
"current_analysis": current_analysis,
"archived_analysis_runs": {
"count": len(archived_run_manifests),
"current_snapshot_count": current_snapshot_run_count,
},
"remote_path_count": len(remote_paths),
"age": _age_summary(extracted_at),
}
def _remote_current_analysis(
api: HfApi,
repo_id: str,
root: Path,
*,
revision: str | None,
remote_paths: set[str],
latest_pointer: dict[str, Any] | None,
) -> dict[str, Any]:
if CURRENT_ANALYSIS_MANIFEST_PATH not in remote_paths:
return {"present": False}
downloaded = load_remote_file(
api,
repo_id,
CURRENT_ANALYSIS_MANIFEST_PATH,
root,
revision=revision,
)
if downloaded is None:
return {"present": False}
try:
manifest = load_current_analysis_manifest(downloaded)
except ValueError as exc:
return {"present": True, "valid": False, "detail": str(exc)}
latest_snapshot_id = (
str(latest_pointer.get("latest_snapshot_id")) if isinstance(latest_pointer, dict) else None
)
return {
"present": True,
"valid": True,
"snapshot_id": manifest["snapshot_id"],
"analysis_id": manifest["analysis_id"],
"variant": manifest["variant"],
"published_at": manifest["published_at"],
"matches_latest_snapshot": manifest["snapshot_id"] == latest_snapshot_id,
"artifact_count": len(manifest["artifacts"]),
}
def _remote_has_latest_artifact(
remote_paths: set[str],
latest_pointer: dict[str, Any] | None,
filename: str,
) -> bool:
candidates = stable_snapshot_candidates(latest_pointer, filename)
return any(candidate in remote_paths for candidate in candidates)
def get_dataset_status(options: DatasetStatusOptions) -> dict[str, Any]:
remote = _remote_status(options.hf_repo_id, options.hf_revision) if options.hf_repo_id else None
local = _local_status(options.output_dir)
repo = options.repo
if repo is None and remote and remote.get("manifest"):
repo = remote["manifest"].get("repo")
if repo is None and local and isinstance(local.get("latest_pointer"), dict):
repo = local["latest_pointer"].get("repo")
return {
"repo": repo,
"dataset_id": options.hf_repo_id,
"remote": remote,
"local": local,
}
def format_dataset_status(status: dict[str, Any]) -> str:
remote = status.get("remote") or {}
local = status.get("local") or {}
manifest = remote.get("manifest") or {}
watermark = remote.get("watermark") or {}
latest_pointer = remote.get("latest_pointer") or {}
age = remote.get("age") or {}
current_analysis = remote.get("current_analysis") or {}
cheap_artifacts = remote.get("cheap_artifacts") or {}
archived_runs = remote.get("archived_analysis_runs") or {}
lines = [
f"Repo: {status.get('repo') or '?'}",
f"Dataset: {status.get('dataset_id') or 'not configured'}",
]
if remote:
lines.extend(
[
f"Remote latest snapshot: {manifest.get('snapshot_id') or latest_pointer.get('latest_snapshot_id') or '?'}",
f"Remote extracted at: {manifest.get('extracted_at') or '?'}",
f"Remote next_since: {watermark.get('next_since') or latest_pointer.get('next_since') or '?'}",
f"PR scope artifact: {'yes' if cheap_artifacts.get('pr_scope_clusters') else 'no'}",
f"Contributor artifacts: {'yes' if cheap_artifacts.get('contributors') else 'no'}",
]
)
if current_analysis.get("present"):
if current_analysis.get("valid") is False:
lines.append(f"Current analysis: invalid ({current_analysis.get('detail')})")
else:
lines.append(
"Current analysis: "
f"snapshot={current_analysis.get('snapshot_id')} "
f"analysis_id={current_analysis.get('analysis_id')}"
)
lines.append(
"Current analysis matches latest snapshot: "
f"{'yes' if current_analysis.get('matches_latest_snapshot') else 'no'}"
)
else:
lines.append("Current analysis: none")
lines.append(
"Archived analysis runs: "
f"{archived_runs.get('count', 0)} total, {archived_runs.get('current_snapshot_count', 0)} for latest snapshot"
)
lines.append(
f"Freshness: {age.get('summary') or 'unknown'} ({age.get('staleness') or 'unknown'})"
)
if local:
lines.extend(
[
f"Local latest pointer: {local.get('latest_path')}",
f"Local snapshot id: {local.get('snapshot_id') or '?'}",
]
)
local_current_analysis = local.get("current_analysis") or {}
if local_current_analysis.get("present"):
lines.append(
"Local current analysis: "
f"snapshot={local_current_analysis.get('snapshot_id')} "
f"analysis_id={local_current_analysis.get('analysis_id')}"
)
else:
lines.append("Local current analysis: none")
else:
lines.append("Local latest pointer: none")
return "\n".join(lines)
|