File size: 4,007 Bytes
a1af790 | 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 | """Acquire current official 5 fps annotations without local account credentials."""
from pathlib import Path
import concurrent.futures
import hashlib
import json
import sys
import time
import urllib.request
ROOT = Path(__file__).resolve().parents[1]
sys.path.insert(0, str(ROOT / "code/vendor"))
import gdown
import ijson
import requests.sessions
# Public downloads only: do not consult a user's .netrc or stored cookies.
requests.sessions.get_netrc_auth = lambda *args, **kwargs: None
def acquire(item):
dest = ROOT / item["local_path"]
dest.parent.mkdir(parents=True, exist_ok=True)
print("START", item["local_path"], flush=True)
try:
gdown.download(
id=item["drive_id"], output=str(dest), quiet=True,
use_cookies=False, resume=True, timeout=(30, 120), retries=5,
)
except gdown.exceptions.FileURLRetrievalError:
# The public direct endpoint can work when Drive's HTML link resolver fails.
url = ("https://drive.usercontent.google.com/download?id=" + item["drive_id"]
+ "&export=download&confirm=t")
request = urllib.request.Request(url, headers={"User-Agent": "dataset-download-audit"})
partial = dest.with_suffix(dest.suffix + ".direct_partial")
with urllib.request.urlopen(request, timeout=120) as response, partial.open("wb") as stream:
if response.status != 200:
raise RuntimeError("Unexpected partial response from direct annotation URL")
expected = response.headers.get("Content-Length")
count = 0
for block in iter(lambda: response.read(4 * 1024 * 1024), b""):
stream.write(block)
count += len(block)
if expected is not None and count != int(expected):
raise RuntimeError("Annotation download has incorrect length")
partial.replace(dest)
sha = hashlib.sha256()
with dest.open("rb") as stream:
for block in iter(lambda: stream.read(8 * 1024 * 1024), b""):
sha.update(block)
record = dict(item, bytes=dest.stat().st_size, sha256=sha.hexdigest())
if dest.suffix == ".json":
# Consume every token, so a truncated response cannot pass as complete.
counts = {"images": 0, "annotations": 0, "captures": 0}
with dest.open("rb") as stream:
for prefix, event, value in ijson.parse(stream):
if prefix == "images.item" and event == "start_map":
counts["images"] += 1
elif prefix == "annotations.item" and event == "start_map":
counts["annotations"] += 1
elif prefix == "" and event == "map_key" and value.isdigit():
counts["captures"] += 1
record["json_counts"] = counts
print("OK", item["local_path"], record["bytes"], flush=True)
return record
def main():
items = json.loads((ROOT / "audit/annotations_manifest.json").read_text())
results, failures = [], []
with concurrent.futures.ThreadPoolExecutor(max_workers=4) as pool:
pending = {pool.submit(acquire, item): item for item in items}
for future in concurrent.futures.as_completed(pending):
item = pending[future]
try:
results.append(future.result())
except Exception as exc:
failures.append({"path": item["local_path"], "error": str(exc)})
print("FAILED", item["local_path"], type(exc).__name__, flush=True)
report = {"complete": len(results) == len(items), "updated_unix": time.time(),
"verified": sorted(results, key=lambda x: x["local_path"]), "failures": failures}
temp = ROOT / "audit/annotations_verification.json.tmp"
temp.write_text(json.dumps(report, indent=2) + "\n")
temp.replace(ROOT / "audit/annotations_verification.json")
if failures:
raise SystemExit(1)
if __name__ == "__main__":
main()
|