viewit/ViewIT
1
0
Fork 0
Code Issues 2 Pull Requests Actions Packages Projects Releases 37 Wiki Activity

nightly: fix movie search flow and add source metadata fallbacks

Browse Source
This commit is contained in:
itdrui.de
2026-02-23 17:52:44 +01:00
parent d414fac022
commit d5a1125e03
4 changed files with 445 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

91
addon/plugins/aniworld_plugin.py
View File

@@ -754,6 +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]] = {}
@@ -818,8 +819,64 @@ 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)
def _extract_series_metadata(self, soup: BeautifulSoupT) -> tuple[str, str]:
if not soup:
return "", ""
plot = ""
poster = ""
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
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:
poster = _absolute_url(content)
break
if not poster:
for selector in ("img.seriesCoverBox", ".seriesCoverBox img", "img[alt][src]"):
image = soup.select_one(selector)
if image is None:
continue
value = (image.get("data-src") or image.get("src") or "").strip()
if value:
poster = _absolute_url(value)
break
return plot, poster
@staticmethod
def _season_links_cache_name(series_url: str) -> str:
digest = hashlib.sha1((series_url or "").encode("utf-8")).hexdigest()[:20]
@@ -951,6 +1008,40 @@ 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 = self._extract_series_metadata(soup)
except Exception:
plot, poster = "", ""
if plot:
info["plot"] = plot
if poster:
art = {"thumb": poster, "poster": poster}
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)

100
addon/plugins/filmpalast_plugin.py
View File

@@ -244,6 +244,7 @@ class FilmpalastPlugin(BasisPlugin):
def __init__(self) -> None:
self._title_to_url: Dict[str, str] = {}
self._title_meta: Dict[str, tuple[str, str]] = {}
self._series_entries: Dict[str, Dict[int, Dict[int, EpisodeEntry]]] = {}
self._hoster_cache: Dict[str, Dict[str, str]] = {}
self._genre_to_url: Dict[str, str] = {}
@@ -722,6 +723,59 @@ class FilmpalastPlugin(BasisPlugin):
return hit.url
return ""
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)
def _extract_detail_metadata(self, soup: BeautifulSoupT) -> tuple[str, str]:
if not soup:
return "", ""
plot = ""
poster = ""
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 (".toggle-content .coverDetails", ".entry-content p", "article p"):
node = soup.select_one(selector)
if node is None:
continue
text = (node.get_text(" ", strip=True) or "").strip()
if text and len(text) > 40:
plot = text
break
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:
poster = _absolute_url(content)
break
if not poster:
for selector in ("img.cover", "article img", ".entry-content img"):
image = soup.select_one(selector)
if image is None:
continue
value = (image.get("data-src") or image.get("src") or "").strip()
if value:
poster = _absolute_url(value)
break
return plot, poster
def remember_series_url(self, title: str, series_url: str) -> None:
title = (title or "").strip()
series_url = (series_url or "").strip()
@@ -742,6 +796,52 @@ class FilmpalastPlugin(BasisPlugin):
return _series_hint_value(series_key)
return ""
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
detail_url = self._ensure_title_url(title)
if not detail_url:
series_key = self._series_key_for_title(title) or self._ensure_series_entries_for_title(title)
if series_key:
seasons = self._series_entries.get(series_key, {})
first_entry: Optional[EpisodeEntry] = None
for season_number in sorted(seasons.keys()):
episodes = seasons.get(season_number, {})
for episode_number in sorted(episodes.keys()):
first_entry = episodes.get(episode_number)
if first_entry is not None:
break
if first_entry is not None:
break
detail_url = first_entry.url if first_entry is not None else ""
if not detail_url:
return info, art, None
try:
soup = _get_soup(detail_url, session=get_requests_session("filmpalast", headers=HEADERS))
plot, poster = self._extract_detail_metadata(soup)
except Exception:
plot, poster = "", ""
if plot:
info["plot"] = plot
if poster:
art = {"thumb": poster, "poster": poster}
self._store_title_meta(title, plot=info.get("plot", ""), poster=poster)
return info, art, None
def is_movie(self, title: str) -> bool:
title = (title or "").strip()
if not title:

112
addon/plugins/topstreamfilm_plugin.py
View File

@@ -97,6 +97,7 @@ class SearchHit:
title: str
url: str
description: str = ""
poster: str = ""
def _normalize_search_text(value: str) -> str:
@@ -149,6 +150,7 @@ class TopstreamfilmPlugin(BasisPlugin):
self._season_to_episode_numbers: Dict[tuple[str, str], List[int]] = {}
self._episode_title_by_number: Dict[tuple[str, int, int], str] = {}
self._detail_html_cache: Dict[str, str] = {}
self._title_meta: Dict[str, tuple[str, str]] = {}
self._popular_cache: List[str] | None = None
self._default_preferred_hosters: List[str] = list(DEFAULT_PREFERRED_HOSTERS)
self._preferred_hosters: List[str] = list(self._default_preferred_hosters)
@@ -429,6 +431,7 @@ class TopstreamfilmPlugin(BasisPlugin):
continue
seen.add(hit.title)
self._title_to_url[hit.title] = hit.url
self._store_title_meta(hit.title, plot=hit.description, poster=hit.poster)
titles.append(hit.title)
if titles:
self._save_title_url_cache()
@@ -487,6 +490,69 @@ class TopstreamfilmPlugin(BasisPlugin):
except Exception:
return ""
def _pick_image_from_node(self, node: Any) -> str:
if node is None:
return ""
image = node.select_one("img")
if image is None:
return ""
for attr in ("data-src", "src"):
value = (image.get(attr) or "").strip()
if value and "lazy_placeholder" not in value.casefold():
return self._absolute_external_url(value, base=self._get_base_url())
srcset = (image.get("data-srcset") or image.get("srcset") or "").strip()
if srcset:
first = srcset.split(",")[0].strip().split(" ", 1)[0].strip()
if first:
return self._absolute_external_url(first, base=self._get_base_url())
return ""
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)
def _extract_detail_metadata(self, soup: BeautifulSoupT) -> tuple[str, str]:
if not soup:
return "", ""
plot = ""
poster = ""
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:
candidates: list[str] = []
for paragraph in soup.select("article p, .TPost p, .Description p, .entry-content p"):
text = (paragraph.get_text(" ", strip=True) or "").strip()
if len(text) >= 60:
candidates.append(text)
if candidates:
plot = max(candidates, key=len)
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:
poster = self._absolute_external_url(content, base=self._get_base_url())
break
if not poster:
for selector in ("article", ".TPost", ".entry-content"):
poster = self._pick_image_from_node(soup.select_one(selector))
if poster:
break
return plot, poster
def _clear_stream_index_for_title(self, title: str) -> None:
for key in list(self._season_to_episode_numbers.keys()):
if key[0] == title:
@@ -721,7 +787,17 @@ class TopstreamfilmPlugin(BasisPlugin):
continue
if is_movie_hint:
self._movie_title_hint.add(title)
hits.append(SearchHit(title=title, url=self._absolute_url(href), description=""))
description_tag = item.select_one(".TPMvCn .Description, .Description, .entry-summary")
description = (description_tag.get_text(" ", strip=True) or "").strip() if description_tag else ""
poster = self._pick_image_from_node(item)
hits.append(
SearchHit(
title=title,
url=self._absolute_url(href),
description=description,
poster=poster,
)
)
return hits
def is_movie(self, title: str) -> bool:
@@ -794,6 +870,7 @@ class TopstreamfilmPlugin(BasisPlugin):
continue
seen.add(hit.title)
self._title_to_url[hit.title] = hit.url
self._store_title_meta(hit.title, plot=hit.description, poster=hit.poster)
titles.append(hit.title)
if titles:
self._save_title_url_cache()
@@ -905,7 +982,8 @@ class TopstreamfilmPlugin(BasisPlugin):
self._movie_title_hint.add(title)
description_tag = item.select_one(".TPMvCn .Description")
description = description_tag.get_text(" ", strip=True) if description_tag else ""
hit = SearchHit(title=title, url=self._absolute_url(href), description=description)
poster = self._pick_image_from_node(item)
hit = SearchHit(title=title, url=self._absolute_url(href), description=description, poster=poster)
if _matches_query(query, title=hit.title, description=hit.description):
hits.append(hit)
@@ -918,11 +996,41 @@ class TopstreamfilmPlugin(BasisPlugin):
continue
seen.add(hit.title)
self._title_to_url[hit.title] = hit.url
self._store_title_meta(hit.title, plot=hit.description, poster=hit.poster)
titles.append(hit.title)
self._save_title_url_cache()
_emit_progress(progress_callback, f"Fertig: {len(titles)} Treffer", 95)
return titles
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
soup = self._get_detail_soup(title)
if soup is None:
return info, art, None
plot, poster = self._extract_detail_metadata(soup)
if plot:
info["plot"] = plot
if poster:
art = {"thumb": poster, "poster": poster}
self._store_title_meta(title, plot=plot, poster=poster)
return info, art, None
def genres(self) -> List[str]:
if not REQUESTS_AVAILABLE or BeautifulSoup is None:
return []