import argparse
import arrow
import hashlib
import html
import io
import json
import logging
import re
import httpx
import time
import os
import subprocess
import uuid
import random
from urllib.parse import urlparse
from dotenv import load_dotenv
from atproto import Client, client_utils, models
from playwright.sync_api import sync_playwright
from moviepy import VideoFileClip
from bs4 import BeautifulSoup
from PIL import Image
import grapheme

# --- Configuration ---
LOG_PATH = "tiktok2bsky.log"
STATE_PATH = "tiktok2bsky_state.json"
DEDUPE_BSKY_LIMIT = 30
VIDEO_MAX_AGE_DAYS = 3
BSKY_TEXT_MAX_LENGTH = 300
DEFAULT_BSKY_LANGS = ["ca"]

TIKTOK_PAGE_LOAD_WAIT_S = 5.0    # was 3.0 — increased for slower grid render
TIKTOK_MAX_SCROLLS      = 8      # was 5 — more scrolls = more videos discovered
SCRAPE_VIDEO_LIMIT      = 30     # was 15


VIDEO_MAX_DURATION_SECONDS = 179
MAX_VIDEO_UPLOAD_SIZE_MB = 45

BSKY_IMAGE_MAX_BYTES = 950 * 1024
BSKY_IMAGE_MAX_DIMENSION = 2000
BSKY_IMAGE_MIN_JPEG_QUALITY = 45

EXTERNAL_THUMB_MAX_BYTES = 950 * 1024
EXTERNAL_THUMB_MAX_DIMENSION = 1200
EXTERNAL_THUMB_MIN_JPEG_QUALITY = 40

BSKY_BLOB_UPLOAD_MAX_RETRIES = 5
BSKY_BLOB_UPLOAD_BASE_DELAY = 10
BSKY_BLOB_UPLOAD_MAX_DELAY = 300
BSKY_BLOB_TRANSIENT_ERROR_RETRIES = 3
BSKY_BLOB_TRANSIENT_ERROR_DELAY = 15

BSKY_SEND_POST_MAX_RETRIES = 3
BSKY_SEND_POST_BASE_DELAY = 5
BSKY_SEND_POST_MAX_DELAY = 60

BSKY_LOGIN_MAX_RETRIES = 4
BSKY_LOGIN_BASE_DELAY = 10
BSKY_LOGIN_MAX_DELAY = 600
BSKY_LOGIN_JITTER_MAX = 1.5

MEDIA_DOWNLOAD_TIMEOUT = 30
LINK_METADATA_TIMEOUT = 10
SUBPROCESS_TIMEOUT_SECONDS = 180
FFPROBE_TIMEOUT_SECONDS = 15
DEFAULT_BSKY_BASE_URL = "https://bsky.social"
SESSION_FILE_PERMISSIONS = 0o600

TIKTOK_SCROLL_PAUSE_S = 2.5      # pause between scrolls to let videos load
TIKTOK_PAGE_LOAD_WAIT_S = 3.0    # initial wait after profile page loads
DYNAMIC_ALT_MAX_LENGTH = 150
TRUNCATE_MIN_PREFIX_CHARS = 20
ORPHAN_DIGIT_MAX_DIGITS = 3

# --- Logging Setup ---
logging.basicConfig(
    format="%(asctime)s [%(levelname)s] %(message)s",
    handlers=[
        logging.FileHandler(LOG_PATH, encoding="utf-8"),
        logging.StreamHandler(),
    ],
    level=logging.INFO,
)


# --- Per-run caches ---
class _RunCache:
    def __init__(self):
        self.url_validity: dict = {}
        self.video_hash_owner: dict = {}
        self.video_url_owner: dict = {}
        self.locale: str = "en-US"

    def clear(self):
        self.url_validity.clear()
        self.video_hash_owner.clear()
        self.video_url_owner.clear()


_cache = _RunCache()


def reset_caches():
    _cache.clear()


# --- Custom Classes ---
class ScrapedMedia:
    def __init__(self, url, media_type="video"):
        self.type = media_type
        self.media_url_https = url


class ScrapedTikTok:
    """Mirrors ScrapedTweet from twitter2bsky.py."""
    def __init__(self, created_on, text, video_url, post_url=None, thumbnail_url=None):
        self.created_on = created_on      # ISO8601 string or arrow-parseable
        self.text = text                  # caption / description
        self.post_url = post_url          # https://www.tiktok.com/@user/video/123
        self.thumbnail_url = thumbnail_url
        self.media = [ScrapedMedia(video_url, "video")] if video_url else []


# --- Helpers (shared with twitter2bsky.py pattern) ---
def sha256_file(path, chunk_size=1024 * 1024):
    h = hashlib.sha256()
    with open(path, "rb") as f:
        while True:
            chunk = f.read(chunk_size)
            if not chunk:
                break
            h.update(chunk)
    return h.hexdigest()


def grapheme_len(text):
    return grapheme.length(text)


def remove_file_quietly(path):
    if path and os.path.exists(path):
        try:
            os.remove(path)
            logging.info(f"🧹 Removed temp file: {path}")
        except Exception as e:
            logging.warning(f"⚠️ Could not remove temp file {path}: {e}")


def take_error_screenshot(page, label):
    timestamp = time.strftime("%Y%m%d_%H%M%S")
    name = f"screenshot_{label}_{timestamp}.png"
    try:
        page.screenshot(path=name)
        logging.info(f"📸 Screenshot saved: {name}")
    except Exception as e:
        logging.warning(f"⚠️ Could not save screenshot: {e}")


def clean_post_text(text):
    raw = (text or "").strip()
    raw = re.sub(r"\r", "\n", raw)
    raw = re.sub(r"\n{3,}", "\n\n", raw)
    return raw.strip()


def normalize_post_text(text):
    if not text:
        return ""
    text = clean_post_text(text)
    text = re.sub(r"\s+", " ", text).strip()
    return text.lower()


def truncate_text_safely(text, max_length=BSKY_TEXT_MAX_LENGTH):
    if grapheme_len(text) <= max_length:
        return text
    clusters = list(grapheme.graphemes(text))
    truncated = "".join(clusters[:max_length])
    last_space = truncated.rfind(" ")
    if last_space > TRUNCATE_MIN_PREFIX_CHARS:
        return truncated[:last_space]
    return truncated


def extract_tiktok_video_id(post_url):
    """Extract numeric video ID from a TikTok URL."""
    if not post_url:
        return None
    match = re.search(r"/video/(\d+)", post_url)
    return match.group(1) if match else None


def canonicalize_tiktok_url(url):
    """Normalize TikTok URL to a stable canonical form."""
    if not url:
        return None
    match = re.search(
        r"https?://(?:www\.)?tiktok\.com/@([^/]+)/video/(\d+)",
        url, re.IGNORECASE,
    )
    if match:
        return f"https://www.tiktok.com/@{match.group(1)}/video/{match.group(2)}"
    return url.strip()


def make_unique_video_temp_base(post_url=None):
    video_id = extract_tiktok_video_id(post_url) or "unknown"
    ts_ms = int(time.time() * 1000)
    rand = uuid.uuid4().hex[:8]
    base = f"temp_tiktok_{video_id}_{ts_ms}_{rand}"
    logging.info(f"🎞️ Using unique temp video base: {base}")
    return base


def build_media_fingerprint(tiktok):
    if not tiktok or not tiktok.media:
        return "no-media"
    parts = []
    for media in tiktok.media:
        media_url = getattr(media, "media_url_https", "") or ""
        stable = canonicalize_tiktok_url(tiktok.post_url) or media_url
        parts.append(f"video:{stable}")
    parts.sort()
    raw = "|".join(parts)
    return hashlib.sha256(raw.encode("utf-8")).hexdigest()


def build_bsky_media_fingerprint(post_view):
    try:
        embed = getattr(post_view, "embed", None)
        if not embed:
            return "no-media"
        parts = []
        video = getattr(embed, "video", None)
        if video:
            ref = getattr(video, "ref", None) or getattr(video, "cid", None) or str(video)
            parts.append(f"video:{ref}")
        if not parts:
            return "no-media"
        parts.sort()
        return hashlib.sha256("|".join(parts).encode("utf-8")).hexdigest()
    except Exception as e:
        logging.debug(f"Could not build Bluesky media fingerprint: {e}")
        return "no-media"


def build_text_media_key(normalized_text, media_fingerprint):
    return hashlib.sha256(
        f"{normalized_text}||{media_fingerprint}".encode("utf-8")
    ).hexdigest()


# --- Bluesky login / retry helpers (identical pattern to twitter2bsky.py) ---
def is_rate_limited_error(e):
    t = repr(e).lower()
    return "429" in t or "ratelimitexceeded" in t or "too many requests" in t

def is_auth_error(e):
    t = repr(e).lower()
    return "401" in t or "403" in t or "invalid identifier" in t

def is_transient_error(e):
    signals = ["InvokeTimeoutError","ReadTimeout","WriteTimeout",
                "RemoteProtocolError","ConnectError","503","502","504"]
    return any(s in repr(e) for s in signals)

def is_network_error(e):
    signals = ["ConnectError","RemoteProtocolError","ReadTimeout",
                "WriteTimeout","TimeoutException","503","502","504"]
    return any(s in repr(e) for s in signals)

def get_rate_limit_wait_seconds(e, default_delay):
    try:
        headers = getattr(e, "headers", None) or {}
        ra = headers.get("retry-after") or headers.get("Retry-After")
        if ra:
            return min(max(int(ra), 1), BSKY_LOGIN_MAX_DELAY)
    except Exception:
        pass
    return default_delay


def create_bsky_client(base_url, handle, password):
    normalized = (base_url or DEFAULT_BSKY_BASE_URL).strip().rstrip("/")
    client = Client(base_url=normalized)
    for attempt in range(1, BSKY_LOGIN_MAX_RETRIES + 1):
        try:
            client.login(handle, password)
            logging.info("✅ Bluesky login successful.")
            return client
        except Exception as e:
            if is_auth_error(e):
                raise
            if attempt < BSKY_LOGIN_MAX_RETRIES:
                wait = min(BSKY_LOGIN_BASE_DELAY * attempt, BSKY_LOGIN_MAX_DELAY)
                wait += random.uniform(0, BSKY_LOGIN_JITTER_MAX)
                logging.warning(f"⏳ Bluesky login retry {attempt} in {wait:.1f}s: {e}")
                time.sleep(wait)
                continue
            raise
    raise RuntimeError("Bluesky login failed after all retries.")


# --- State management (identical pattern) ---
def default_state():
    return {"version": 1, "posted_videos": {}, "posted_by_bsky_uri": {}, "updated_at": None}


def load_state(state_path=STATE_PATH):
    if not os.path.exists(state_path):
        return default_state()
    try:
        with open(state_path, "r", encoding="utf-8") as f:
            state = json.load(f)
        state.setdefault("version", 1)
        state.setdefault("posted_videos", {})
        state.setdefault("posted_by_bsky_uri", {})
        state.setdefault("updated_at", None)
        return state
    except Exception as e:
        logging.warning(f"⚠️ Could not load state: {e}. Reinitializing.")
        return default_state()


def save_state(state, state_path=STATE_PATH):
    try:
        state["updated_at"] = arrow.utcnow().isoformat()
        temp = f"{state_path}.tmp"
        with open(temp, "w", encoding="utf-8") as f:
            json.dump(state, f, ensure_ascii=False, indent=2, sort_keys=True)
        os.replace(temp, state_path)
        logging.info(f"💾 State saved to {state_path}")
    except Exception as e:
        logging.error(f"❌ Failed to save state: {e}")


def remember_posted_video(state, candidate, bsky_uri=None):
    key = candidate.get("canonical_post_url") or f"textmedia:{candidate['text_media_key']}"
    state["posted_videos"][key] = {
        "canonical_post_url": candidate.get("canonical_post_url"),
        "normalized_text": candidate["normalized_text"],
        "text_media_key": candidate["text_media_key"],
        "media_fingerprint": candidate["media_fingerprint"],
        "bsky_uri": bsky_uri,
        "video_created_on": candidate["tiktok"].created_on,
        "post_url": candidate["tiktok"].post_url,
        "video_id": candidate.get("video_id"),
        "posted_at": arrow.utcnow().isoformat(),
    }
    if bsky_uri:
        state["posted_by_bsky_uri"][bsky_uri] = key


def candidate_matches_state(candidate, state):
    canonical_url = candidate["canonical_post_url"]
    text_media_key = candidate["text_media_key"]
    normalized_text = candidate["normalized_text"]
    posted = state.get("posted_videos", {})

    if canonical_url and canonical_url in posted:
        return True, "state:post_url"
    for rec in posted.values():
        if rec.get("text_media_key") == text_media_key:
            return True, "state:text_media_fingerprint"
    for rec in posted.values():
        if rec.get("normalized_text") == normalized_text and normalized_text:
            return True, "state:normalized_text"
    return False, None


def prune_state(state, max_entries=5000):
    posted = state.get("posted_videos", {})
    if len(posted) <= max_entries:
        return state
    sortable = sorted(posted.items(), key=lambda x: x[1].get("posted_at", ""), reverse=True)
    keep = {k for k, _ in sortable[:max_entries]}
    state["posted_videos"] = {k: v for k, v in posted.items() if k in keep}
    state["posted_by_bsky_uri"] = {
        uri: k for uri, k in state.get("posted_by_bsky_uri", {}).items() if k in keep
    }
    return state


# --- Bluesky feed helpers ---
def get_recent_bsky_posts(client, handle, limit=30):
    recent = []
    try:
        timeline = client.get_author_feed(handle, limit=limit)
        for item in timeline.feed:
            try:
                if item.reason is not None:
                    continue
                record = item.post.record
                if getattr(record, "reply", None) is not None:
                    continue
                text = getattr(record, "text", "") or ""
                normalized = normalize_post_text(text)
                media_fp = build_bsky_media_fingerprint(item.post)
                recent.append({
                    "uri": getattr(item.post, "uri", None),
                    "normalized_text": normalized,
                    "media_fingerprint": media_fp,
                    "text_media_key": build_text_media_key(normalized, media_fp),
                })
            except Exception as e:
                logging.debug(f"Skipping feed item: {e}")
    except Exception as e:
        logging.warning(f"⚠️ Could not fetch recent Bluesky posts: {e}")
    return recent


def candidate_matches_existing_bsky(candidate, recent_bsky_posts):
    for existing in recent_bsky_posts:
        if candidate["text_media_key"] == existing["text_media_key"]:
            return True, "bsky:text_media_fingerprint"
        if candidate["normalized_text"] and candidate["normalized_text"] == existing["normalized_text"]:
            return True, "bsky:normalized_text"
    return False, None


# --- Upload / blob helpers (same as twitter2bsky.py) ---
def upload_blob_with_retry(client, binary_data, media_label="media"):
    last_exception = None
    transient_attempts = 0
    for attempt in range(1, BSKY_BLOB_UPLOAD_MAX_RETRIES + 1):
        try:
            result = client.upload_blob(binary_data)
            return result.blob
        except Exception as e:
            last_exception = e
            if "429" in str(e) or "RateLimitExceeded" in str(e):
                wait = min(BSKY_BLOB_UPLOAD_BASE_DELAY * (2 ** (attempt - 1)), BSKY_BLOB_UPLOAD_MAX_DELAY)
                if attempt < BSKY_BLOB_UPLOAD_MAX_RETRIES:
                    logging.warning(f"⏳ Blob upload rate-limited. Retry {attempt} after {wait}s.")
                    time.sleep(wait)
                    continue
                break
            if is_transient_error(e) and transient_attempts < BSKY_BLOB_TRANSIENT_ERROR_RETRIES:
                transient_attempts += 1
                wait = BSKY_BLOB_TRANSIENT_ERROR_DELAY * transient_attempts
                logging.warning(f"⏳ Transient blob upload error. Retry {transient_attempts} after {wait}s.")
                time.sleep(wait)
                continue
            logging.warning(f"Could not upload {media_label}: {repr(e)}")
            return None
    logging.warning(f"Could not upload {media_label}: {repr(last_exception)}")
    return None


def send_post_with_retry(client, **kwargs):
    last_exception = None
    for attempt in range(1, BSKY_SEND_POST_MAX_RETRIES + 1):
        try:
            return client.send_post(**kwargs)
        except Exception as e:
            last_exception = e
            if "429" in str(e) or "RateLimitExceeded" in str(e):
                wait = min(BSKY_SEND_POST_BASE_DELAY * (2 ** (attempt - 1)), BSKY_SEND_POST_MAX_DELAY)
                if attempt < BSKY_SEND_POST_MAX_RETRIES:
                    time.sleep(wait)
                    continue
                raise
            if is_transient_error(e) and attempt < BSKY_SEND_POST_MAX_RETRIES:
                time.sleep(BSKY_SEND_POST_BASE_DELAY * attempt)
                continue
            raise
    raise last_exception


def get_blob_from_file(file_path, client):
    try:
        if not os.path.exists(file_path):
            logging.warning(f"File not found: {file_path}")
            return None
        size_mb = os.path.getsize(file_path) / (1024 * 1024)
        if size_mb > MAX_VIDEO_UPLOAD_SIZE_MB:
            logging.warning(f"File too large: {size_mb:.2f} MB > {MAX_VIDEO_UPLOAD_SIZE_MB} MB")
            return None
        with open(file_path, "rb") as f:
            data = f.read()
        return upload_blob_with_retry(client, data, media_label=file_path)
    except Exception as e:
        logging.warning(f"Could not upload file {file_path}: {repr(e)}")
        return None


def build_video_embed(video_blob, alt_text):
    try:
        return models.AppBskyEmbedVideo.Main(video=video_blob, alt=alt_text)
    except AttributeError:
        logging.error("❌ atproto version does not support AppBskyEmbedVideo. Upgrade atproto.")
        return None


def build_dynamic_alt(text):
    alt = clean_post_text(text or "").replace("\n", " ").strip()
    alt = re.sub(r"(?:https?://|www\.)\S+", "", alt).strip()
    if not alt:
        alt = "TikTok video"
    return alt[:DYNAMIC_ALT_MAX_LENGTH]


def make_rich(content):
    """Build a Bluesky TextBuilder with hashtag and URL facets."""
    text_builder = client_utils.TextBuilder()
    content = clean_post_text(content)
    lines = content.splitlines()
    for line_idx, line in enumerate(lines):
        if not line.strip():
            if line_idx < len(lines) - 1:
                text_builder.text("\n")
            continue
        words = line.split(" ")
        for i, word in enumerate(words):
            if not word:
                if i < len(words) - 1:
                    text_builder.text(" ")
                continue
            if word.startswith("#") and len(word) > 1:
                tag = word[1:].rstrip(".,;:!?)'\"")
                if tag:
                    text_builder.tag(word, tag)
                else:
                    text_builder.text(word)
            elif word.startswith(("http://", "https://")):
                text_builder.link(word, word)
            else:
                text_builder.text(word)
            if i < len(words) - 1:
                text_builder.text(" ")
        if line_idx < len(lines) - 1:
            text_builder.text("\n")
    return text_builder


# --- TikTok Scraping ---
# --- TikTok Scraping ---
def scrape_tiktoks_via_playwright(target_handle: str, locale: str = "en-US") -> list:
    """
    Scrape recent TikTok videos from a public profile using Playwright.
    No login required for public profiles.
    Returns a list of ScrapedTikTok objects.

    Fixes applied:
      1. Aggressive GDPR/consent banner dismissal (Spanish + English)
      2. Stealth headers: timezone, locale, sec-ch-ua, webdriver flag hidden
      3. playwright-stealth applied before navigation
      4. Broader + longer grid selector wait (30s, more selectors)
    """
    tiktoks = []
    profile_url = f"https://www.tiktok.com/@{target_handle.lstrip('@')}"

    # playwright-stealth is optional but strongly recommended
    try:
        from playwright_stealth import stealth_sync
        USE_STEALTH = True
        logging.info("🥷 playwright-stealth available — stealth mode ON")
    except ImportError:
        USE_STEALTH = False
        logging.warning("⚠️ playwright-stealth not installed — running without stealth")

    with sync_playwright() as p:
        browser = p.chromium.launch(
            headless=True,
            args=[
                "--disable-blink-features=AutomationControlled",
                "--no-sandbox",
                "--disable-setuid-sandbox",
                "--disable-dev-shm-usage",
                "--disable-gpu",
                "--window-size=1366,768",
            ],
        )

        # FIX 2 — Fake a real Windows Chrome browser with Spanish locale + Madrid timezone
        context = browser.new_context(
            user_agent=(
                "Mozilla/5.0 (Windows NT 10.0; Win64; x64) "
                "AppleWebKit/537.36 (KHTML, like Gecko) "
                "Chrome/124.0.0.0 Safari/537.36"
            ),
            viewport={"width": 1366, "height": 768},
            locale="es-ES",
            timezone_id="Europe/Madrid",
            extra_http_headers={
                "Accept-Language": "es-ES,es;q=0.9,en;q=0.8",
                "Accept": (
                    "text/html,application/xhtml+xml,application/xml;"
                    "q=0.9,image/avif,image/webp,*/*;q=0.8"
                ),
                "Sec-Fetch-Dest": "document",
                "Sec-Fetch-Mode": "navigate",
                "Sec-Fetch-Site": "none",
                "Sec-Ch-Ua": '"Chromium";v="124", "Google Chrome";v="124", "Not-A.Brand";v="99"',
                "Sec-Ch-Ua-Mobile": "?0",
                "Sec-Ch-Ua-Platform": '"Windows"',
            },
        )

        page = context.new_page()

        # FIX 3 — Apply playwright-stealth before any navigation
        if USE_STEALTH:
            stealth_sync(page)
            logging.info("🥷 Stealth patches applied.")

        # FIX 2 — Hide webdriver flag + fake plugins/languages via init script
        page.add_init_script("""
            Object.defineProperty(navigator, 'webdriver', { get: () => undefined });
            Object.defineProperty(navigator, 'plugins', {
                get: () => [
                    { name: 'Chrome PDF Plugin' },
                    { name: 'Chrome PDF Viewer' },
                    { name: 'Native Client' }
                ]
            });
            Object.defineProperty(navigator, 'languages', {
                get: () => ['es-ES', 'es', 'en']
            });
            window.chrome = {
                runtime: {},
                loadTimes: function() {},
                csi: function() {},
                app: {}
            };
        """)

        try:
            logging.info(f"🌐 Navigating to TikTok profile: {profile_url}")
            page.goto(profile_url, wait_until="domcontentloaded", timeout=40000)

            # FIX 1 — Wait longer for initial page render (was 3.0s)
            time.sleep(TIKTOK_PAGE_LOAD_WAIT_S + 2)

            # FIX 1 — Aggressive GDPR/consent banner dismissal (Spanish + English)
            GDPR_SELECTORS = [
                'button:has-text("Entendido")',
                'button:has-text("Aceptar todo")',
                'button:has-text("Accept all")',
                'button:has-text("Got it")',
                'button:has-text("Decline optional")',
                '[data-e2e="cookie-banner-accept"]',
                '[id*="accept"]',
                '[class*="accept-btn"]',
            ]
            for selector in GDPR_SELECTORS:
                try:
                    btn = page.locator(selector).first
                    if btn.is_visible(timeout=3000):
                        btn.click()
                        logging.info(f"✅ Dismissed banner: {selector}")
                        time.sleep(2)
                        break
                except Exception:
                    pass

            # FIX 4 — Broader selector list + longer timeout (30s, was 20s)
            GRID_SELECTORS = (
                '[data-e2e="user-post-item"], '
                '[class*="DivItemContainerV2"], '
                'a[href*="/video/"], '
                '[class*="video-feed"], '
                'div[class*="VideoFeed"], '
                '[class*="DivVideoFeedV2"]'
            )
            try:
                page.wait_for_selector(GRID_SELECTORS, timeout=30000)
                logging.info("✅ TikTok video grid detected.")
            except Exception:
                # FIX 4 — Don't give up immediately: try scrolling anyway
                logging.warning(
                    "⚠️ Grid selector timed out — attempting scroll anyway "
                    "(grid may still be partially loaded)"
                )

            # Scroll to load more videos
            for scroll_i in range(TIKTOK_MAX_SCROLLS):
                page.evaluate("window.scrollBy(0, window.innerHeight * 2)")
                time.sleep(TIKTOK_SCROLL_PAUSE_S)
                logging.info(f"📜 Scroll {scroll_i + 1}/{TIKTOK_MAX_SCROLLS}")

            # Collect video links
            video_links = page.locator('a[href*="/video/"]').all()
            logging.info(
                f"📊 Found {len(video_links)} video links. "
                f"Parsing up to {SCRAPE_VIDEO_LIMIT}..."
            )

            if not video_links:
                take_error_screenshot(page, "tiktok_no_video_links")
                logging.error("❌ No video links found after scroll. TikTok may be blocking.")
                browser.close()
                return []

            seen_urls = set()
            for link in video_links:
                if len(tiktoks) >= SCRAPE_VIDEO_LIMIT:
                    break
                try:
                    href = link.get_attribute("href")
                    if not href:
                        continue

                    post_url = (
                        f"https://www.tiktok.com{href}"
                        if href.startswith("/")
                        else href
                    )
                    canonical = canonicalize_tiktok_url(post_url)
                    if not canonical or canonical in seen_urls:
                        continue
                    if "/video/" not in canonical:
                        continue
                    seen_urls.add(canonical)

                    # Try to get caption from the card itself
                    caption = ""
                    try:
                        card = link.locator("..").first
                        caption_el = card.locator(
                            '[data-e2e="video-desc"], '
                            '[class*="SpanUniqueId"], '
                            'p[class*="caption"]'
                        ).first
                        if caption_el.is_visible(timeout=1000):
                            caption = caption_el.inner_text()
                    except Exception:
                        pass

                    # Thumbnail
                    thumbnail_url = None
                    try:
                        img = link.locator("img").first
                        if img.is_visible(timeout=1000):
                            thumbnail_url = img.get_attribute("src")
                    except Exception:
                        pass

                    created_on = arrow.utcnow().isoformat()

                    tiktoks.append(
                        ScrapedTikTok(
                            created_on=created_on,
                            text=caption,
                            video_url=canonical,
                            post_url=canonical,
                            thumbnail_url=thumbnail_url,
                        )
                    )
                    logging.info(f"🎵 Scraped TikTok: {canonical}")

                except Exception as e:
                    logging.warning(f"⚠️ Failed to parse video card: {e}")
                    continue

        except Exception as e:
            take_error_screenshot(page, "tiktok_scrape_failed")
            logging.error(f"❌ Failed to scrape TikTok profile: {e}")

        browser.close()

    logging.info(f"✅ Scraped {len(tiktoks)} TikTok videos.")
    return tiktoks

# --- Video extraction ---
def extract_tiktok_video_url_isolated(browser, post_url: str, video_id: str = None) -> str | None:
    """
    Open a single TikTok video page in an isolated context and intercept
    the actual MP4/HLS stream URL from network responses.
    Mirrors extract_video_url_from_tweet_page_isolated() in twitter2bsky.py.
    """
    ctx = None
    page = None
    best_mp4_url = None
    best_m3u8_url = None
    seen_urls = set()

    def current_best():
        return best_mp4_url or best_m3u8_url

    def handle_response(response):
        nonlocal best_mp4_url, best_m3u8_url
        try:
            url = response.url
            if not url or url in seen_urls:
                return
            seen_urls.add(url)
            content_type = (response.headers.get("content-type") or "").lower()
            url_l = url.lower()

            # Skip audio-only and segment files
            if ".m4s" in url_l or "/aud/" in url_l or "mp4a" in url_l:
                return

            if ".m3u8" in url_l or "mpegurl" in content_type:
                if best_m3u8_url is None:
                    best_m3u8_url = url
                return

            if ".mp4" in url_l or "video/mp4" in content_type:
                if best_mp4_url is None:
                    best_mp4_url = url
                return
        except Exception as e:
            logging.debug(f"Response parse error: {e}")

    try:
        ctx = browser.new_context(
            user_agent=(
                "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) "
                "AppleWebKit/537.36 (KHTML, like Gecko) "
                "Chrome/145.0.7632.6 Safari/537.36"
            ),
            viewport={"width": 1920, "height": 1080},
        )
        page = ctx.new_page()
        page.on("response", handle_response)

        logging.info(f"[video_id={video_id}] 🎬 Opening TikTok video page: {post_url}")
        page.goto(post_url, wait_until="domcontentloaded", timeout=40000)
        time.sleep(2)

        # Try clicking the video player to trigger stream loading
        for selector in ['[data-e2e="video-player"]', "video", '[class*="Video"]']:
            try:
                player = page.locator(selector).first
                if player.count() > 0:
                    player.click(force=True, timeout=3000)
                    break
            except Exception:
                pass

        # Wait up to 10s for a stream URL to appear
        for _ in range(10):
            if current_best():
                break
            time.sleep(1)

        selected = current_best()
        logging.info(f"[video_id={video_id}] ✅ Resolved video URL: {selected}")
        return selected

    except Exception as e:
        logging.warning(f"[video_id={video_id}] ⚠️ Could not extract video URL: {e}")
        return None
    finally:
        try:
            if page:
                page.remove_listener("response", handle_response)
                page.close()
        except Exception:
            pass
        try:
            if ctx:
                ctx.close()
        except Exception:
            pass


# --- Video download + compress (same ffmpeg pipeline as twitter2bsky.py) ---
def _probe_video_duration(file_path):
    result = subprocess.run(
        ["ffprobe", "-v", "error", "-show_entries", "format=duration",
         "-of", "default=noprint_wrappers=1:nokey=1", file_path],
        capture_output=True, text=True, timeout=FFPROBE_TIMEOUT_SECONDS,
    )
    if result.returncode != 0:
        raise RuntimeError(f"ffprobe error: {result.stderr.strip()}")
    duration_str = result.stdout.strip()
    if not duration_str:
        raise RuntimeError("ffprobe returned empty duration")
    return float(duration_str)


def download_and_crop_video(video_url: str, output_path: str) -> str | None:
    """Identical ffmpeg pipeline to twitter2bsky.py."""
    temp_input = output_path.replace(".mp4", "_source.mp4")
    temp_trimmed = output_path.replace(".mp4", "_trimmed.mp4")
    temp_output = output_path.replace(".mp4", "_compressed.mp4")

    try:
        logging.info(f"⬇️ Downloading TikTok video: {video_url}")
        url_l = video_url.lower()

        if ".m3u8" in url_l:
            download_cmd = [
                "ffmpeg", "-y",
                "-protocol_whitelist", "file,http,https,tcp,tls,crypto",
                "-allowed_extensions", "ALL",
                "-i", video_url, "-c", "copy", temp_input,
            ]
        else:
            download_cmd = [
                "ffmpeg", "-y", "-i", video_url, "-c", "copy", temp_input,
            ]

        result = subprocess.run(download_cmd, capture_output=True, text=True,
                                timeout=SUBPROCESS_TIMEOUT_SECONDS)
        if result.returncode != 0:
            logging.error(f"❌ ffmpeg download failed:\n{result.stderr}")
            return None

        if not os.path.exists(temp_input) or os.path.getsize(temp_input) == 0:
            logging.error("❌ Downloaded file is missing or empty.")
            return None

        duration = _probe_video_duration(temp_input)
        if duration <= 0:
            logging.error("❌ Invalid video duration.")
            return None

        end_time = min(VIDEO_MAX_DURATION_SECONDS, duration)
        end_time = min(end_time, duration - 0.05)
        end_time = max(end_time, 0.1)

        from moviepy import VideoFileClip
        video_clip = VideoFileClip(temp_input)
        try:
            if hasattr(video_clip, "subclipped"):
                cropped = video_clip.subclipped(0, end_time)
            else:
                cropped = video_clip.subclip(0, end_time)
            try:
                cropped.write_videofile(
                    temp_trimmed, codec="libx264", audio_codec="aac",
                    preset="veryfast", bitrate="1800k", audio_bitrate="128k", logger=None,
                )
            finally:
                cropped.close()
        finally:
            video_clip.close()

        if not os.path.exists(temp_trimmed) or os.path.getsize(temp_trimmed) == 0:
            logging.error("❌ Trimmed video is missing or empty.")
            return None

        compress_cmd = [
            "ffmpeg", "-y", "-i", temp_trimmed,
            "-vf", "scale='min(720,iw)':-2",
            "-c:v", "libx264", "-preset", "veryfast", "-crf", "30",
            "-maxrate", "1800k", "-bufsize", "3600k",
            "-c:a", "aac", "-b:a", "128k",
            "-movflags", "+faststart", temp_output,
        ]
        result = subprocess.run(compress_cmd, capture_output=True, text=True,
                                timeout=SUBPROCESS_TIMEOUT_SECONDS)
        if result.returncode != 0:
            logging.error(f"❌ ffmpeg compression failed:\n{result.stderr}")
            return None

        if not os.path.exists(temp_output) or os.path.getsize(temp_output) == 0:
            logging.error("❌ Compressed video is missing or empty.")
            return None

        os.replace(temp_output, output_path)
        size_mb = os.path.getsize(output_path) / (1024 * 1024)
        logging.info(f"✅ Video ready: {output_path} ({size_mb:.2f} MB)")
        return output_path

    except subprocess.TimeoutExpired:
        logging.error(f"❌ ffmpeg timed out after {SUBPROCESS_TIMEOUT_SECONDS}s")
        return None
    except Exception as e:
        logging.error(f"❌ Video processing error: {repr(e)}")
        return None
    finally:
        remove_file_quietly(temp_input)
        remove_file_quietly(temp_trimmed)
        remove_file_quietly(temp_output)


# --- Main sync logic ---
def sync_feeds(args):
    logging.info("🔄 Starting TikTok → Bluesky sync cycle...")
    dry_run = getattr(args, "dry_run", False)
    bsky_langs = getattr(args, "bsky_langs", None) or DEFAULT_BSKY_LANGS

    if dry_run:
        logging.info("🧪 DRY RUN MODE — no posts will be created on Bluesky.")

    try:
        state = load_state(STATE_PATH)
        state = prune_state(state, max_entries=5000)

        tiktoks = scrape_tiktoks_via_playwright(
            args.tiktok_handle,
            locale=bsky_langs[0] if bsky_langs else "en-US",
        )

        if not tiktoks:
            logging.warning("⚠️ No TikTok videos found. Skipping sync.")
            return

        bsky_client = None
        if not dry_run:
            bsky_client = create_bsky_client(
                args.bsky_base_url, args.bsky_handle, args.bsky_password,
            )

        recent_bsky_posts = []
        if not dry_run:
            recent_bsky_posts = get_recent_bsky_posts(
                bsky_client, args.bsky_handle, limit=DEDUPE_BSKY_LIMIT,
            )

        too_old_cutoff = arrow.utcnow().shift(days=-VIDEO_MAX_AGE_DAYS)

        # --- Build candidates ---
        candidates = []
        for tiktok in reversed(tiktoks):
            try:
                # TikTok grid doesn't expose timestamps reliably —
                # use state-based dedup as primary guard
                canonical_url = canonicalize_tiktok_url(tiktok.post_url)
                if canonical_url and canonical_url in state.get("posted_videos", {}):
                    logging.info(f"⚡ Early skip (already in state): {canonical_url}")
                    continue

                text = clean_post_text(tiktok.text or "")
                normalized_text = normalize_post_text(text)
                media_fp = build_media_fingerprint(tiktok)
                text_media_key = build_text_media_key(normalized_text, media_fp)
                video_id = extract_tiktok_video_id(tiktok.post_url)

                candidate = {
                    "tiktok": tiktok,
                    "raw_text": truncate_text_safely(text),
                    "normalized_text": normalized_text,
                    "media_fingerprint": media_fp,
                    "text_media_key": text_media_key,
                    "canonical_post_url": canonical_url,
                    "video_id": video_id,
                    "resolved_video_url": None,
                    "resolved_video_hash": None,
                }

                is_dup_state, reason = candidate_matches_state(candidate, state)
                if is_dup_state:
                    logging.info(f"⏭️ Skipping (state duplicate: {reason}): {canonical_url}")
                    continue

                is_dup_bsky, reason = candidate_matches_existing_bsky(candidate, recent_bsky_posts)
                if is_dup_bsky:
                    logging.info(f"⏭️ Skipping (Bluesky duplicate: {reason}): {canonical_url}")
                    continue

                candidates.append(candidate)

            except Exception as e:
                logging.warning(f"⚠️ Failed to prepare candidate: {e}")

        logging.info(f"📬 {len(candidates)} new TikTok videos to post after dedup.")

        if not candidates:
            logging.info("✅ Nothing new to post.")
            return

        # --- Pre-resolve video URLs ---
        with sync_playwright() as p_pre:
            pre_browser = p_pre.chromium.launch(
                headless=True,
                args=["--disable-blink-features=AutomationControlled"],
            )
            try:
                for c in candidates:
                    c["resolved_video_url"] = extract_tiktok_video_url_isolated(
                        pre_browser,
                        c["tiktok"].post_url,
                        video_id=c.get("video_id"),
                    )
            finally:
                pre_browser.close()

        # --- Post to Bluesky ---
        new_posts = 0
        for candidate in candidates:
            tiktok = candidate["tiktok"]
            raw_text = candidate["raw_text"]

            logging.info(
                f"📝 {'[DRY RUN] Would post' if dry_run else 'Posting'} "
                f"TikTok video: {tiktok.post_url}"
            )

            if dry_run:
                logging.info(f"  📄 Caption: {raw_text[:200]}")
                remember_posted_video(state, candidate, bsky_uri=f"dry-run:{arrow.utcnow().isoformat()}")
                save_state(state, STATE_PATH)
                new_posts += 1
                continue

            real_video_url = candidate.get("resolved_video_url")
            video_embed = None
            video_blob = None

            if real_video_url:
                temp_base = make_unique_video_temp_base(tiktok.post_url)
                temp_path = f"{temp_base}.mp4"
                try:
                    cropped_path = download_and_crop_video(real_video_url, temp_path)
                    if cropped_path:
                        video_hash = sha256_file(cropped_path)
                        candidate["resolved_video_hash"] = video_hash
                        owner = _cache.video_hash_owner.get(video_hash)
                        if owner and owner != candidate["video_id"]:
                            logging.warning(f"⚠️ Video hash owned by another video. Skipping.")
                        else:
                            _cache.video_hash_owner[video_hash] = candidate["video_id"]
                            video_blob = get_blob_from_file(cropped_path, bsky_client)
                            if video_blob:
                                alt = build_dynamic_alt(raw_text)
                                video_embed = build_video_embed(video_blob, alt)
                finally:
                    remove_file_quietly(temp_path)
                    remove_file_quietly(f"{temp_base}_source.mp4")
                    remove_file_quietly(f"{temp_base}_trimmed.mp4")
                    remove_file_quietly(f"{temp_base}_compressed.mp4")
            else:
                logging.warning(f"⚠️ Could not resolve video URL for {tiktok.post_url}")

            try:
                rich_text = make_rich(raw_text)

                if video_embed:
                    post_result = send_post_with_retry(
                        bsky_client, text=rich_text, embed=video_embed, langs=bsky_langs,
                    )
                    post_mode = "video"
                else:
                    # Fallback: post caption as text-only with link to TikTok
                    fallback_text = make_rich(
                        f"{raw_text}\n\n{tiktok.post_url}".strip()
                    )
                    post_result = send_post_with_retry(
                        bsky_client, text=fallback_text, langs=bsky_langs,
                    )
                    post_mode = "text_only_fallback"

                bsky_uri = getattr(post_result, "uri", None)
                remember_posted_video(state, candidate, bsky_uri=bsky_uri)
                state = prune_state(state, max_entries=5000)
                save_state(state, STATE_PATH)

                recent_bsky_posts.insert(0, {
                    "uri": bsky_uri,
                    "normalized_text": candidate["normalized_text"],
                    "media_fingerprint": candidate["media_fingerprint"],
                    "text_media_key": candidate["text_media_key"],
                })
                recent_bsky_posts = recent_bsky_posts[:DEDUPE_BSKY_LIMIT]

                new_posts += 1
                logging.info(f"✅ Posted TikTok to Bluesky [{post_mode}]: {raw_text[:80]}")
                time.sleep(5)

            except Exception as e:
                logging.error(f"❌ Failed to post to Bluesky: {e}")

        logging.info(f"✅ Sync complete. Posted {new_posts} new TikTok videos.")

    except Exception as e:
        logging.error(f"❌ Error during sync cycle: {e}")


def main():
    load_dotenv()

    parser = argparse.ArgumentParser(description="TikTok to Bluesky Sync")
    parser.add_argument("--tiktok-handle", help="TikTok account handle to scrape (without @)")
    parser.add_argument("--bsky-handle", help="Your Bluesky handle")
    parser.add_argument("--bsky-password", help="Your Bluesky app password")
    parser.add_argument("--bsky-base-url", help="Bluesky PDS base URL", default=None)
    parser.add_argument("--bsky-langs", help="Comma-separated language codes", default=None)
    parser.add_argument("--dry-run", action="store_true", default=False)

    args = parser.parse_args()

    args.tiktok_handle = args.tiktok_handle or os.getenv("TIKTOK_HANDLE")
    args.bsky_handle = args.bsky_handle or os.getenv("BSKY_HANDLE")
    args.bsky_password = args.bsky_password or os.getenv("BSKY_APP_PASSWORD")
    args.bsky_base_url = args.bsky_base_url or os.getenv("BSKY_BASE_URL") or DEFAULT_BSKY_BASE_URL

    raw_langs = args.bsky_langs or os.getenv("BSKY_LANGS")
    args.bsky_langs = (
        [l.strip() for l in raw_langs.split(",") if l.strip()]
        if raw_langs else DEFAULT_BSKY_LANGS
    )

    missing = []
    if not args.tiktok_handle:
        missing.append("--tiktok-handle / TIKTOK_HANDLE")
    if not args.bsky_handle:
        missing.append("--bsky-handle / BSKY_HANDLE")
    if not args.bsky_password:
        missing.append("--bsky-password / BSKY_APP_PASSWORD")
    if missing:
        logging.error(f"❌ Missing: {', '.join(missing)}")
        return

    logging.info(f"🤖 TikTok→Bluesky bot started. Scraping @{args.tiktok_handle}")
    reset_caches()
    sync_feeds(args)
    logging.info("🤖 Bot finished.")


if __name__ == "__main__":
    main()