main: consolidate integrated changes after v0.1.54

addon/plugins/aniworld_plugin.py

@@ -13,7 +13,8 @@ import hashlib
 import json
 import re
 import time
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, TypeAlias
+from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple
+from urllib.parse import quote
 
 try:  # pragma: no cover - optional dependency
     import requests
@@ -43,8 +44,8 @@ if TYPE_CHECKING:  # pragma: no cover
     from requests import Session as RequestsSession
     from bs4 import BeautifulSoup as BeautifulSoupT  # type: ignore[import-not-found]
 else:  # pragma: no cover
-    RequestsSession: TypeAlias = Any
-    BeautifulSoupT: TypeAlias = Any
+    RequestsSession = Any
+    BeautifulSoupT = Any
 
 
 SETTING_BASE_URL = "aniworld_base_url"
@@ -69,6 +70,16 @@ HEADERS = {
 SESSION_CACHE_TTL_SECONDS = 300
 SESSION_CACHE_PREFIX = "viewit.aniworld"
 SESSION_CACHE_MAX_TITLE_URLS = 800
+ProgressCallback = Optional[Callable[[str, Optional[int]], Any]]
+
+
+def _emit_progress(callback: ProgressCallback, message: str, percent: Optional[int] = None) -> None:
+    if not callable(callback):
+        return
+    try:
+        callback(str(message or ""), None if percent is None else int(percent))
+    except Exception:
+        return
 
 
 @dataclass
@@ -126,7 +137,7 @@ def _latest_episodes_url() -> str:
 
 
 def _search_url(query: str) -> str:
-    return f"{_get_base_url()}/search?q={query}"
+    return f"{_get_base_url()}/search?q={quote((query or '').strip())}"
 
 
 def _search_api_url() -> str:
@@ -289,37 +300,56 @@ def _get_soup(url: str, *, session: Optional[RequestsSession] = None) -> Beautif
     _ensure_requests()
     _log_visit(url)
     sess = session or get_requests_session("aniworld", headers=HEADERS)
+    response = None
     try:
         response = sess.get(url, headers=HEADERS, timeout=DEFAULT_TIMEOUT)
         response.raise_for_status()
     except Exception as exc:
         _log_error(f"GET {url} failed: {exc}")
         raise
-    if response.url and response.url != url:
-        _log_url(response.url, kind="REDIRECT")
-    _log_response_html(url, response.text)
-    if _looks_like_cloudflare_challenge(response.text):
-        raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
-    return BeautifulSoup(response.text, "html.parser")
+    try:
+        final_url = (response.url or url) if response is not None else url
+        body = (response.text or "") if response is not None else ""
+        if final_url != url:
+            _log_url(final_url, kind="REDIRECT")
+        _log_response_html(url, body)
+        if _looks_like_cloudflare_challenge(body):
+            raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
+        return BeautifulSoup(body, "html.parser")
+    finally:
+        if response is not None:
+            try:
+                response.close()
+            except Exception:
+                pass
 
 
 def _get_html_simple(url: str) -> str:
     _ensure_requests()
     _log_visit(url)
     sess = get_requests_session("aniworld", headers=HEADERS)
+    response = None
     try:
         response = sess.get(url, headers=HEADERS, timeout=DEFAULT_TIMEOUT)
         response.raise_for_status()
     except Exception as exc:
         _log_error(f"GET {url} failed: {exc}")
         raise
-    if response.url and response.url != url:
-        _log_url(response.url, kind="REDIRECT")
-    body = response.text
-    _log_response_html(url, body)
-    if _looks_like_cloudflare_challenge(body):
-        raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
-    return body
+    try:
+        final_url = (response.url or url) if response is not None else url
+        body = (response.text or "") if response is not None else ""
+        if final_url != url:
+            _log_url(final_url, kind="REDIRECT")
+        _log_response_html(url, body)
+        if _looks_like_cloudflare_challenge(body):
+            raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
+        return body
+    finally:
+        if response is not None:
+            try:
+                response.close()
+            except Exception:
+                pass
 
 
 def _get_soup_simple(url: str) -> BeautifulSoupT:
@@ -351,17 +381,27 @@ def _post_json(url: str, *, payload: Dict[str, str], session: Optional[RequestsS
     _ensure_requests()
     _log_visit(url)
     sess = session or get_requests_session("aniworld", headers=HEADERS)
-    response = sess.post(url, data=payload, headers=HEADERS, timeout=DEFAULT_TIMEOUT)
-    response.raise_for_status()
-    if response.url and response.url != url:
-        _log_url(response.url, kind="REDIRECT")
-    _log_response_html(url, response.text)
-    if _looks_like_cloudflare_challenge(response.text):
-        raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
+    response = None
     try:
-        return response.json()
-    except Exception:
-        return None
+        response = sess.post(url, data=payload, headers=HEADERS, timeout=DEFAULT_TIMEOUT)
+        response.raise_for_status()
+        final_url = (response.url or url) if response is not None else url
+        body = (response.text or "") if response is not None else ""
+        if final_url != url:
+            _log_url(final_url, kind="REDIRECT")
+        _log_response_html(url, body)
+        if _looks_like_cloudflare_challenge(body):
+            raise RuntimeError("Cloudflare-Schutz erkannt. requests reicht ggf. nicht aus.")
+        try:
+            return response.json()
+        except Exception:
+            return None
+    finally:
+        if response is not None:
+            try:
+                response.close()
+            except Exception:
+                pass
 
 
 def _extract_canonical_url(soup: BeautifulSoupT, fallback: str) -> str:
@@ -555,10 +595,18 @@ def resolve_redirect(target_url: str) -> Optional[str]:
     _log_visit(normalized_url)
     session = get_requests_session("aniworld", headers=HEADERS)
     _get_soup(_get_base_url(), session=session)
-    response = session.get(normalized_url, headers=HEADERS, timeout=DEFAULT_TIMEOUT, allow_redirects=True)
-    if response.url:
-        _log_url(response.url, kind="RESOLVED")
-    return response.url if response.url else None
+    response = None
+    try:
+        response = session.get(normalized_url, headers=HEADERS, timeout=DEFAULT_TIMEOUT, allow_redirects=True)
+        if response.url:
+            _log_url(response.url, kind="RESOLVED")
+        return response.url if response.url else None
+    finally:
+        if response is not None:
+            try:
+                response.close()
+            except Exception:
+                pass
 
 
 def fetch_episode_hoster_names(episode_url: str) -> List[str]:
@@ -629,11 +677,12 @@ def fetch_episode_stream_link(
     return resolved
 
 
-def search_animes(query: str) -> List[SeriesResult]:
+def search_animes(query: str, *, progress_callback: ProgressCallback = None) -> List[SeriesResult]:
     _ensure_requests()
     query = (query or "").strip()
     if not query:
         return []
+    _emit_progress(progress_callback, "AniWorld API-Suche", 15)
     session = get_requests_session("aniworld", headers=HEADERS)
     try:
         session.get(_get_base_url(), headers=HEADERS, timeout=DEFAULT_TIMEOUT)
@@ -643,7 +692,9 @@ def search_animes(query: str) -> List[SeriesResult]:
     results: List[SeriesResult] = []
     seen: set[str] = set()
     if isinstance(data, list):
-        for entry in data:
+        for idx, entry in enumerate(data, start=1):
+            if idx == 1 or idx % 50 == 0:
+                _emit_progress(progress_callback, f"API auswerten {idx}/{len(data)}", 35)
             if not isinstance(entry, dict):
                 continue
             title = _strip_html((entry.get("title") or "").strip())
@@ -665,10 +716,16 @@ def search_animes(query: str) -> List[SeriesResult]:
             seen.add(key)
             description = (entry.get("description") or "").strip()
             results.append(SeriesResult(title=title, description=description, url=url))
+        _emit_progress(progress_callback, f"API-Treffer: {len(results)}", 85)
         return results
 
-    soup = _get_soup_simple(_search_url(requests.utils.quote(query)))
-    for anchor in soup.select("a[href^='/anime/stream/'][href]"):
+    _emit_progress(progress_callback, "HTML-Suche (Fallback)", 55)
+    soup = _get_soup_simple(_search_url(query))
+    anchors = soup.select("a[href^='/anime/stream/'][href]")
+    total_anchors = max(1, len(anchors))
+    for idx, anchor in enumerate(anchors, start=1):
+        if idx == 1 or idx % 100 == 0:
+            _emit_progress(progress_callback, f"HTML auswerten {idx}/{total_anchors}", 70)
         href = (anchor.get("href") or "").strip()
         if not href or "/staffel-" in href or "/episode-" in href:
             continue
@@ -686,6 +743,7 @@ def search_animes(query: str) -> List[SeriesResult]:
             continue
         seen.add(key)
         results.append(SeriesResult(title=title, description="", url=url))
+    _emit_progress(progress_callback, f"HTML-Treffer: {len(results)}", 85)
     return results
 
 
@@ -696,6 +754,7 @@ class AniworldPlugin(BasisPlugin):
     def __init__(self) -> None:
         self._anime_results: Dict[str, SeriesResult] = {}
         self._title_url_cache: Dict[str, str] = self._load_title_url_cache()
+        self._title_meta: Dict[str, tuple[str, str]] = {}
         self._genre_names_cache: Optional[List[str]] = None
         self._season_cache: Dict[str, List[SeasonInfo]] = {}
         self._season_links_cache: Dict[str, List[SeasonInfo]] = {}
@@ -760,8 +819,135 @@ class AniworldPlugin(BasisPlugin):
                 changed = True
         if changed and persist:
             self._save_title_url_cache()
+        if description:
+            old_plot, old_poster = self._title_meta.get(title, ("", ""))
+            self._title_meta[title] = (description.strip() or old_plot, old_poster)
         return changed
 
+    def _store_title_meta(self, title: str, *, plot: str = "", poster: str = "") -> None:
+        title = (title or "").strip()
+        if not title:
+            return
+        old_plot, old_poster = self._title_meta.get(title, ("", ""))
+        merged_plot = (plot or old_plot or "").strip()
+        merged_poster = (poster or old_poster or "").strip()
+        self._title_meta[title] = (merged_plot, merged_poster)
+
+    @staticmethod
+    def _is_series_image_url(url: str) -> bool:
+        value = (url or "").strip().casefold()
+        if not value:
+            return False
+        blocked = (
+            "/public/img/facebook",
+            "/public/img/logo",
+            "aniworld-logo",
+            "favicon",
+            "/public/img/german.svg",
+            "/public/img/japanese-",
+        )
+        return not any(marker in value for marker in blocked)
+
+    @staticmethod
+    def _extract_style_url(style_value: str) -> str:
+        style_value = (style_value or "").strip()
+        if not style_value:
+            return ""
+        match = re.search(r"url\((['\"]?)(.*?)\1\)", style_value, flags=re.IGNORECASE)
+        if not match:
+            return ""
+        return (match.group(2) or "").strip()
+
+    def _extract_series_metadata(self, soup: BeautifulSoupT) -> tuple[str, str, str]:
+        if not soup:
+            return "", "", ""
+        plot = ""
+        poster = ""
+        fanart = ""
+
+        root = soup.select_one("#series") or soup
+
+        description_node = root.select_one("p.seri_des")
+        if description_node is not None:
+            full_text = (description_node.get("data-full-description") or "").strip()
+            short_text = (description_node.get_text(" ", strip=True) or "").strip()
+            plot = full_text or short_text
+
+        if not plot:
+            for selector in ("meta[property='og:description']", "meta[name='description']"):
+                node = soup.select_one(selector)
+                if node is None:
+                    continue
+                content = (node.get("content") or "").strip()
+                if content:
+                    plot = content
+                    break
+        if not plot:
+            for selector in (".series-description", ".seri_des", ".description", "article p"):
+                node = soup.select_one(selector)
+                if node is None:
+                    continue
+                text = (node.get_text(" ", strip=True) or "").strip()
+                if text:
+                    plot = text
+                    break
+
+        cover = root.select_one("div.seriesCoverBox img[itemprop='image'], div.seriesCoverBox img")
+        if cover is not None:
+            for attr in ("data-src", "src"):
+                value = (cover.get(attr) or "").strip()
+                if value:
+                    candidate = _absolute_url(value)
+                    if self._is_series_image_url(candidate):
+                        poster = candidate
+                        break
+
+        if not poster:
+            for selector in ("meta[property='og:image']", "meta[name='twitter:image']"):
+                node = soup.select_one(selector)
+                if node is None:
+                    continue
+                content = (node.get("content") or "").strip()
+                if content:
+                    candidate = _absolute_url(content)
+                    if self._is_series_image_url(candidate):
+                        poster = candidate
+                        break
+        if not poster:
+            for selector in ("img.seriesCoverBox", ".seriesCoverBox img"):
+                image = soup.select_one(selector)
+                if image is None:
+                    continue
+                value = (image.get("data-src") or image.get("src") or "").strip()
+                if value:
+                    candidate = _absolute_url(value)
+                    if self._is_series_image_url(candidate):
+                        poster = candidate
+                        break
+
+        backdrop_node = root.select_one("section.title .backdrop, .SeriesSection .backdrop, .backdrop")
+        if backdrop_node is not None:
+            raw_style = (backdrop_node.get("style") or "").strip()
+            style_url = self._extract_style_url(raw_style)
+            if style_url:
+                candidate = _absolute_url(style_url)
+                if self._is_series_image_url(candidate):
+                    fanart = candidate
+
+        if not fanart:
+            for selector in ("meta[property='og:image']",):
+                node = soup.select_one(selector)
+                if node is None:
+                    continue
+                content = (node.get("content") or "").strip()
+                if content:
+                    candidate = _absolute_url(content)
+                    if self._is_series_image_url(candidate):
+                        fanart = candidate
+                        break
+
+        return plot, poster, fanart
+
     @staticmethod
     def _season_links_cache_name(series_url: str) -> str:
         digest = hashlib.sha1((series_url or "").encode("utf-8")).hexdigest()[:20]
@@ -893,6 +1079,43 @@ class AniworldPlugin(BasisPlugin):
 
         return None
 
+    def metadata_for(self, title: str) -> tuple[dict[str, str], dict[str, str], list[object] | None]:
+        title = (title or "").strip()
+        if not title:
+            return {}, {}, None
+
+        info: dict[str, str] = {"title": title}
+        art: dict[str, str] = {}
+        cached_plot, cached_poster = self._title_meta.get(title, ("", ""))
+        if cached_plot:
+            info["plot"] = cached_plot
+        if cached_poster:
+            art = {"thumb": cached_poster, "poster": cached_poster}
+        if "plot" in info and art:
+            return info, art, None
+
+        series = self._find_series_by_title(title)
+        if series is None or not series.url:
+            return info, art, None
+        if series.description and "plot" not in info:
+            info["plot"] = series.description
+
+        try:
+            soup = _get_soup(series.url, session=get_requests_session("aniworld", headers=HEADERS))
+            plot, poster, fanart = self._extract_series_metadata(soup)
+        except Exception:
+            plot, poster, fanart = "", "", ""
+
+        if plot:
+            info["plot"] = plot
+        if poster:
+            art = {"thumb": poster, "poster": poster}
+        if fanart:
+            art["fanart"] = fanart
+            art["landscape"] = fanart
+        self._store_title_meta(title, plot=info.get("plot", ""), poster=poster)
+        return info, art, None
+
     def _ensure_popular(self) -> List[SeriesResult]:
         if self._popular_cache is not None:
             return list(self._popular_cache)
@@ -1151,7 +1374,7 @@ class AniworldPlugin(BasisPlugin):
             return self._episode_label_cache.get(cache_key, {}).get(episode_label)
         return None
 
-    async def search_titles(self, query: str) -> List[str]:
+    async def search_titles(self, query: str, progress_callback: ProgressCallback = None) -> List[str]:
         query = (query or "").strip()
         if not query:
             self._anime_results.clear()
@@ -1163,7 +1386,8 @@ class AniworldPlugin(BasisPlugin):
         if not self._requests_available:
             raise RuntimeError("AniworldPlugin kann ohne requests/bs4 nicht suchen.")
         try:
-            results = search_animes(query)
+            _emit_progress(progress_callback, "AniWorld Suche startet", 10)
+            results = search_animes(query, progress_callback=progress_callback)
         except Exception as exc:  # pragma: no cover
             self._anime_results.clear()
             self._season_cache.clear()
@@ -1178,6 +1402,7 @@ class AniworldPlugin(BasisPlugin):
         self._season_cache.clear()
         self._season_links_cache.clear()
         self._episode_label_cache.clear()
+        _emit_progress(progress_callback, f"Treffer aufbereitet: {len(results)}", 95)
         return [result.title for result in results]
 
     def _ensure_seasons(self, title: str) -> List[SeasonInfo]:
@@ -1213,6 +1438,18 @@ class AniworldPlugin(BasisPlugin):
             _log_url(link, kind="FOUND")
         return link
 
+    def episode_url_for(self, title: str, season: str, episode: str) -> str:
+        cache_key = (title, season)
+        cached = self._episode_label_cache.get(cache_key)
+        if cached:
+            info = cached.get(episode)
+            if info and info.url:
+                return info.url
+        episode_info = self._lookup_episode(title, season, episode)
+        if episode_info and episode_info.url:
+            return episode_info.url
+        return ""
+
     def available_hosters_for(self, title: str, season: str, episode: str) -> List[str]:
         if not self._requests_available:
             raise RuntimeError("AniworldPlugin kann ohne requests/bs4 keine Hoster laden.")