"""X/Twitter без платного API.

Три пути:
1. X_BEARER есть -> официальный /2/tweets/search/recent.
2. Иначе -> ищем ссылки на твиты через поисковики, текст достаём из publish.twitter.com/oembed (без ключей).
3. Профиль целиком -> syndication.twitter.com/srv/timeline-profile (отдаёт JSON в HTML, без ключей).
"""
import json
import re
import urllib.parse

from .. import config, net
from ..model import Item

TWEET_RE = re.compile(r"https?://(?:www\.)?(?:twitter\.com|x\.com)/([A-Za-z0-9_]{1,15})/status/(\d+)")
UA = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) Chrome/141.0.0.0 Safari/537.36"}


def available(name="x"):
    if config.KEYS.get("X_BEARER"):
        return True, "официальный API v2 (X_BEARER)"
    return True, "без ключей: oembed + syndication timeline"


def _oembed(url):
    api = "https://publish.twitter.com/oembed?omit_script=1&dnt=1&url=" + urllib.parse.quote(url, safe="")
    r = net.get(api, headers=UA, json_mode=True)
    if r.status != 200:
        return None
    d = r.json() or {}
    html = d.get("html") or ""
    text = re.sub(r"<[^>]+>", " ", html)
    text = re.sub(r"\s+", " ", text).replace("—", "-").strip()
    return {"author": d.get("author_name") or "", "text": text,
            "url": d.get("url") or url, "handle": (d.get("author_url") or "").rstrip("/").split("/")[-1]}


def search_x(query, limit=15, handles=None):
    items = []
    bearer = config.KEYS.get("X_BEARER")
    if bearer:
        q = urllib.parse.quote(f"{query} -is:retweet lang:ru OR lang:en")
        url = (f"https://api.x.com/2/tweets/search/recent?query={q}"
               f"&max_results={min(max(limit, 10), 100)}&tweet.fields=created_at,public_metrics,lang")
        r = net.get(url, headers={"Authorization": f"Bearer {bearer}"}, json_mode=True)
        for t in ((r.json() or {}).get("data") or []):
            m = t.get("public_metrics") or {}
            it = Item("x", "tweet", url=f"https://x.com/i/status/{t['id']}", title=t["text"][:120],
                      text=t["text"], author="", published=t.get("created_at", ""),
                      extra={"likes": m.get("like_count"), "rt": m.get("retweet_count")})
            it.score = (m.get("like_count") or 0) ** 0.5
            items.append(it)
    else:
        # путь 2: находим ссылки поисковиками, текст тянем oembed-ом
        from . import searchers
        found = []
        for eng in ("ddg", "mojeek", "bing", "tavily"):
            try:
                res = getattr(searchers, "search_" + eng)(f"{query} site:x.com OR site:twitter.com", limit=limit)
            except Exception:
                res = []
            for r0 in res:
                for m in TWEET_RE.finditer(r0.url or ""):
                    found.append(m.group(0))
            if len(found) >= limit:
                break
        seen = set()
        for u in found[:limit * 2]:
            if u in seen:
                continue
            seen.add(u)
            d = _oembed(u)
            if not d or len(d["text"]) < 20:
                continue
            it = Item("x", "tweet", url=d["url"], title=d["text"][:120], text=d["text"],
                      author=d["author"], source="x.com/" + (d["handle"] or ""),
                      extra={"handle": d["handle"]})
            items.append(it)

    # путь 3: конкретные профили
    for h in (handles or [])[:4]:
        r = net.get(f"https://syndication.twitter.com/srv/timeline-profile/screen-name/{h}",
                    headers=UA)
        if r.status != 200:
            continue
        m = re.search(r'<script id="__NEXT_DATA__" type="application/json">(.*?)</script>', r.text, re.S)
        if not m:
            continue
        try:
            data = json.loads(m.group(1))
        except Exception:
            continue
        entries = (((data.get("props") or {}).get("pageProps") or {}).get("timeline") or {}).get("entries") or []
        for e in entries[:12]:
            tw = ((e.get("content") or {}).get("tweet") or {})
            txt = tw.get("full_text") or tw.get("text") or ""
            if not txt:
                continue
            tid = tw.get("id_str") or ""
            it = Item("x", "tweet", url=f"https://x.com/{h}/status/{tid}" if tid else f"https://x.com/{h}",
                      title=txt[:120], text=txt, author=h, source="x.com/" + h,
                      published=tw.get("created_at", ""),
                      extra={"likes": tw.get("favorite_count"), "rt": tw.get("retweet_count")})
            it.score = float(tw.get("favorite_count") or 0) ** 0.5
            items.append(it)
    return items[:limit * 2]