"""Download UFO/UAP records from war.gov DOW UFO release.
Bypasses Akamai TLS fingerprint check using curl_cffi browser impersonation.
Reads metadata CSV, then downloads PDFs, images, thumbnails, and videos
(via DVIDS API) in parallel.
"""
from __future__ import annotations
import csv
import json
import os
import re
import sys
from concurrent.futures import ThreadPoolExecutor, as_completed
from pathlib import Path
from urllib.parse import urlparse
from curl_cffi import requests
ROOT = Path(__file__).parent / "ufo_data"
CSV_URL = "https://www.war.gov/Portals/1/Interactive/2026/UFO/uap-csv.csv"
DVIDS_API_KEY = "key-68bb60d16b35e" # Same key the page uses publicly
IMPERSONATE = "chrome120"
def safe_name(s: str) -> str:
s = s.strip().replace("/", "_")
s = re.sub(r"[^A-Za-z0-9._-]+", "_", s)
return s.strip("._-") or "file"
def download(url: str, dest: Path, session: requests.Session) -> tuple[str, str]:
if dest.exists() and dest.stat().st_size > 0:
return ("skip", str(dest))
dest.parent.mkdir(parents=True, exist_ok=True)
try:
r = session.get(url, impersonate=IMPERSONATE, timeout=120)
if r.status_code != 200:
return (f"err {r.status_code}", url)
tmp = dest.with_suffix(dest.suffix + ".part")
tmp.write_bytes(r.content)
tmp.rename(dest)
return ("ok", f"{dest.name} ({len(r.content):,}B)")
except Exception as e:
return ("exc", f"{url}: {e}")
def fetch_csv() -> list[dict]:
s = requests.Session()
r = s.get(CSV_URL, impersonate=IMPERSONATE)
r.raise_for_status()
(ROOT / "csv").mkdir(parents=True, exist_ok=True)
(ROOT / "csv" / "uap-csv.csv").write_bytes(r.content)
text = r.content.decode("utf-8-sig")
return list(csv.DictReader(text.splitlines()))
_MP4_RE = re.compile(r'https?://[^"\'\s<>]+\.mp4[^"\'\s<>]*')
_TITLE_RE = re.compile(r"
([^<]+)")
def get_dvids_video(video_id: str, session: requests.Session) -> dict | None:
"""Scrape the public DVIDS video page for the direct mp4 URL."""
url = f"https://www.dvidshub.net/video/{video_id}"
r = session.get(url, impersonate=IMPERSONATE, timeout=60)
if r.status_code != 200:
return None
mp4 = _MP4_RE.search(r.text)
title = _TITLE_RE.search(r.text)
return {
"id": video_id,
"page": url,
"title": title.group(1).strip() if title else "",
"video": mp4.group(0) if mp4 else "",
}
def main() -> None:
rows = fetch_csv()
print(f"Records: {len(rows)}")
pdf_dir = ROOT / "pdfs"
img_dir = ROOT / "images"
thumb_dir = ROOT / "thumbnails"
vid_dir = ROOT / "videos"
meta_dir = ROOT / "metadata"
for d in (pdf_dir, img_dir, thumb_dir, vid_dir, meta_dir):
d.mkdir(parents=True, exist_ok=True)
jobs: list[tuple[str, Path]] = []
video_rows: list[dict] = []
manifest: list[dict] = []
for row in rows:
title = (row.get("Title") or "").strip()
rtype = (row.get("Type") or "").strip()
link = (row.get("PDF | Image Link") or "").strip()
thumb = (row.get("Modal Image") or "").strip()
dvid_id = (row.get("DVIDS Video ID") or "").strip()
agency = (row.get("Agency") or "").strip()
item = {
"title": title,
"type": rtype,
"agency": agency,
"incident_date": row.get("Incident Date", ""),
"incident_location": row.get("Incident Location", ""),
"description": row.get("Description Blurb", ""),
"link": link,
"thumbnail": thumb,
"dvids_video_id": dvid_id,
"files": [],
}
if link:
ext = Path(urlparse(link).path).suffix or ".bin"
sub = pdf_dir if rtype.startswith("PDF") else img_dir
dest = sub / f"{safe_name(title)}{ext}"
jobs.append((link, dest))
item["files"].append(str(dest.relative_to(ROOT)))
if thumb:
ext = Path(urlparse(thumb).path).suffix or ".jpg"
dest = thumb_dir / f"{safe_name(title)}{ext}"
jobs.append((thumb, dest))
item["files"].append(str(dest.relative_to(ROOT)))
if rtype == "VID" and dvid_id:
video_rows.append(row)
manifest.append(item)
# Download PDFs/images/thumbnails in parallel
print(f"Downloading {len(jobs)} files (PDFs/images/thumbnails)...")
session = requests.Session()
ok = err = skip = 0
with ThreadPoolExecutor(max_workers=8) as ex:
futs = {ex.submit(download, u, d, session): (u, d) for u, d in jobs}
for i, fut in enumerate(as_completed(futs), 1):
status, info = fut.result()
if status == "ok":
ok += 1
elif status == "skip":
skip += 1
else:
err += 1
print(f" [{status}] {info}")
if i % 25 == 0:
print(f" progress: {i}/{len(jobs)} (ok={ok} skip={skip} err={err})")
print(f"Static files: ok={ok} skip={skip} err={err}")
# Resolve and download DVIDS videos
print(f"Resolving {len(video_rows)} DVIDS videos...")
video_meta = []
video_jobs: list[tuple[str, Path]] = []
for row in video_rows:
vid = row["DVIDS Video ID"].strip()
meta = get_dvids_video(vid, session)
if not meta:
print(f" [warn] no meta for video {vid}")
continue
video_meta.append({"row_title": row.get("Title", ""), "dvids": meta})
# Pick highest-res mp4 from `video_files` if present
video_url = meta.get("video") or meta.get("source") or ""
if not video_url:
continue
title = safe_name(row.get("Title") or meta.get("title") or vid)
ext = Path(urlparse(video_url).path).suffix or ".mp4"
dest = vid_dir / f"{title}{ext}"
video_jobs.append((video_url, dest))
(meta_dir / "dvids_videos.json").write_text(json.dumps(video_meta, indent=2))
print(f"Downloading {len(video_jobs)} videos...")
ok = err = skip = 0
with ThreadPoolExecutor(max_workers=4) as ex:
futs = {ex.submit(download, u, d, session): (u, d) for u, d in video_jobs}
for i, fut in enumerate(as_completed(futs), 1):
status, info = fut.result()
if status == "ok":
ok += 1
elif status == "skip":
skip += 1
else:
err += 1
print(f" [{status}] {info}")
print(f" video {i}/{len(video_jobs)}: {status} {info}")
print(f"Videos: ok={ok} skip={skip} err={err}")
(meta_dir / "manifest.json").write_text(json.dumps(manifest, indent=2))
print(f"Wrote manifest with {len(manifest)} records to {meta_dir/'manifest.json'}")
if __name__ == "__main__":
main()