"""
Author: [.2C52.69AD.]
Key (auth= fng^k): 10101 01010 - then -> 01010
:)

just automates the manual labor of testing the JE-files that are given as .PDFs but are actually media
like mp4 and avi files. Save for later investigation if desired.

You can change headless to True down a little to not disp browser.
"""
from playwright.sync_api import sync_playwright, TimeoutError as PlaywrightTimeoutError
import time
from urllib.parse import urlencode
import random

SEARCH_TERM = "no images produced"
BASE_URL = "https://www.justice.gov"
SEARCH_PATH = "/multimedia-search"

START_PAGE = 1
MAX_PAGES = 10
SKIP_DATASETS = ['DataSet 9']           # skip media testing on these
SKIP_COLLECTING_SKIP_DATASETS = True
MIN_VALID_FILE = 2000          # bytes — filter out junk? probably just rm this + associated code.

all_links = []                  # collected PDF links
valid_media = []                # discovered media URLs

# JSON capture
captured_json = None

with sync_playwright() as p:
    browser = p.chromium.launch(headless=False, slow_mo=150)
    context = browser.new_context(
        viewport={"width": 620, "height": 400},
        user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/132.0.0.0 Safari/537.36",
        locale="en-US",
        timezone_id="America/New_York",
    )
    page = context.new_page()

    print("Visiting main page to establish session / pass Akamai + gates")
    page.goto(BASE_URL + "/epstein", wait_until="networkidle", timeout=60000)

    # Robot challenge
    try:
        robot = page.wait_for_selector(
            'input[type="button"][class="usa-button"][value="I am not a robot"][onclick*="reauth()"]',
            timeout=12000
        )
        robot.click()
        page.wait_for_load_state("networkidle", timeout=15000)
        print("Robot challenge passed")
    except PlaywrightTimeoutError:
        pass
    except Exception as e:
        print(f"Robot button failed: {e}")

    # Age gate
    try:
        page.wait_for_selector('text=/are you 18|age verification/i', timeout=12000)
        page.click('text=Yes', timeout=8000)
        page.wait_for_load_state("networkidle", timeout=25000)
        print("Age gate passed?")
    except PlaywrightTimeoutError:
        pass
    except Exception as e:
        print(f"Age gate failed: {e}")

    time.sleep(2)

    def on_response(response):
        global captured_json
        ct = response.headers.get("content-type", "").lower()
        if "application/json" in ct:
            try:
                data = response.json()
                if "hits" in data and "hits" in data["hits"]:
                    captured_json = data
                    print(f"Captured JSON — {len(data['hits']['hits'])} hits")
            except:
                pass

    page.on("response", on_response)

    # Collect links
    for pg in range(START_PAGE, START_PAGE + MAX_PAGES):
        captured_json = None
        params = {"keys": SEARCH_TERM, "page": pg}
        url = f"{BASE_URL}{SEARCH_PATH}?{urlencode(params)}"
        print(f"Loading page {pg}: {url}")

        page.goto(url, wait_until="networkidle", timeout=60000)
        time.sleep(0.2)

        if captured_json:
            for hit in captured_json["hits"]["hits"]:
                src = hit.get("_source", {})
                uri = src.get("ORIGIN_FILE_URI")
                if uri:
                    if SKIP_COLLECTING_SKIP_DATASETS and not any(ds in uri for ds in SKIP_DATASETS):
                        clean = uri.replace("\\/", "/")
                        all_links.append(clean)
                        print("  ->", clean)
                    else:
                        print(f"Ignoring skip-list item? -> {uri}")
                else:
                    name = src.get("ORIGIN_FILE_NAME")
                    if name:
                        print(f"Warning: No URI for {name} — skipped")
        else:
            print("No JSON — snippet:", page.content()[:300])

    # Media variant testing
    print("\nTesting media variants...")
    EXTENSIONS = ['.mp4', '.avi', '.wav', '.mov']

    for pdf_url in all_links:
        if any(ds in pdf_url for ds in SKIP_DATASETS):
            continue

        base = pdf_url.rsplit('.', 1)[0]
        found = False

        for ext in EXTENSIONS:
            test_url = base + ext
            try:
                result = page.evaluate('''async (url) => {
                    try {
                        const res = await fetch(url, {method: 'HEAD', redirect: 'follow'});
                        return {
                            ok: res.ok,
                            status: res.status,
                            contentType: res.headers.get('content-type') || '',
                            length: parseInt(res.headers.get('content-length') || '0', 10)
                        };
                    } catch (e) {
                        return {ok: false};
                    }
                }''', test_url)

                if result.get('ok') and result['status'] == 200:
                    ct = result['contentType'].lower()
                    length = result['length']
                    if length > MIN_VALID_FILE and ('video/' in ct or 'audio/' in ct or 'octet-stream' in ct):
                        print(f"✓ VALID: {test_url} ({ct}, {length:,} bytes)")
                        valid_media.append(test_url)
                        found = True
                        break
            except:
                pass

        if not found:
            print(f"No media: {pdf_url}")

        time.sleep(0.8 + random.random() * 1.1)

    browser.close()

# Save results
if all_links:
    with open("epstein_no_images_links.txt", "w", encoding="utf-8") as f:
        for link in sorted(set(all_links)):
            f.write(link + "\n")
    print(f"\nSaved {len(all_links)} PDF links → epstein_no_images_links.txt")

if valid_media:
    with open("epstein_valid_media.txt", "w", encoding="utf-8") as f:
        for url in sorted(set(valid_media)):
            f.write(url + "\n")
    print(f"Saved {len(valid_media)} media files → epstein_valid_media.txt")
else:
    print("\nNo valid media found.")

print("Done.")