feat: Add anime metadata extraction and fix episode selection bug
Features: - Added rich metadata extraction for all anime providers (Anime-Sama, Neko-Sama, Anime-Ultime, Vostfree) - New AnimeMetadata model with synopsis, genres, rating, release year, studio, poster/banner images, episode count, and status - New /api/anime/metadata endpoint for fetching metadata of specific anime - Enhanced /api/anime/search endpoint with optional include_metadata parameter - Updated web interface with metadata display (expandable synopsis, genres, rating, year) - Added metadata toggle checkbox in search UI (disabled by default for performance) Bug Fixes: - Fixed episode selection bug where select would reset to default after any change - Removed onchange event from select element that was causing unwanted reloads - Fixed download button disappearing after episode download - Episodes can now be downloaded multiple times without page refresh Enhancements: - Metadata displayed with icons (📅 year, ⭐ rating, 🏷️ genres, 📺 episodes, 📡 status) - Expandable synopsis section for detailed descriptions - Better visual organization of anime information - Maintains backward compatibility (metadata is optional) Generated with [Claude Code](https://claude.ai/code) via [Happy](https://happy.engineering) Co-Authored-By: Claude <noreply@anthropic.com> Co-Authored-By: Happy <yesreply@happy.engineering>
This commit is contained in:
root
@@ -165,10 +165,124 @@ class AnimeUltimeDownloader(BaseDownloader):
|
||||
filename = f"{anime_name} - Episode {episode}.mp4"
|
||||
return filename.title()
|
||||
|
||||
async def search_anime(self, query: str, lang: str = "vostfr") -> list[dict]:
|
||||
async def get_anime_metadata(self, anime_url: str) -> dict:
|
||||
"""
|
||||
Extract rich metadata from anime page
|
||||
Returns synopsis, genres, rating, release year, studio, etc.
|
||||
"""
|
||||
try:
|
||||
print(f"[ANIME-ULTIME] Extracting metadata from: {anime_url}")
|
||||
response = await self.client.get(anime_url)
|
||||
soup = BeautifulSoup(response.text, 'lxml')
|
||||
|
||||
metadata = {
|
||||
'synopsis': None,
|
||||
'genres': [],
|
||||
'rating': None,
|
||||
'release_year': None,
|
||||
'studio': None,
|
||||
'poster_image': None,
|
||||
'banner_image': None,
|
||||
'total_episodes': None,
|
||||
'status': None,
|
||||
'alternative_titles': []
|
||||
}
|
||||
|
||||
# Extract synopsis
|
||||
synopsis_selectors = [
|
||||
'div.synopsis',
|
||||
'div.description',
|
||||
'div[class*="synopsis"]',
|
||||
'div[class*="synopsis"]',
|
||||
'p.synopsis',
|
||||
'.info',
|
||||
'div.texte'
|
||||
]
|
||||
|
||||
for selector in synopsis_selectors:
|
||||
synopsis_elem = soup.select_one(selector)
|
||||
if synopsis_elem:
|
||||
synopsis = synopsis_elem.get_text(strip=True)
|
||||
if len(synopsis) > 50:
|
||||
metadata['synopsis'] = synopsis
|
||||
break
|
||||
|
||||
# Extract genres from meta tags and page content
|
||||
page_text = soup.get_text()
|
||||
|
||||
# Look for genre in meta tags
|
||||
genre_meta = soup.find('meta', property='genre') or soup.find('meta', attrs={'name': 'genre'})
|
||||
if genre_meta:
|
||||
genres_text = genre_meta.get('content', '')
|
||||
if genres_text:
|
||||
metadata['genres'] = [g.strip() for g in genres_text.split(',')]
|
||||
|
||||
# Try to find genre links
|
||||
genre_links = soup.find_all('a', href=re.compile(r'genre|tag|type|cat', re.I))
|
||||
if genre_links:
|
||||
for link in genre_links[:5]:
|
||||
genre = link.get_text(strip=True)
|
||||
if genre and genre not in metadata['genres']:
|
||||
metadata['genres'].append(genre)
|
||||
|
||||
# Extract rating
|
||||
rating_selectors = [
|
||||
'span.rating',
|
||||
'div.rating',
|
||||
'span.score',
|
||||
'div.note',
|
||||
'.rating'
|
||||
]
|
||||
|
||||
for selector in rating_selectors:
|
||||
rating_elem = soup.select_one(selector)
|
||||
if rating_elem:
|
||||
rating_text = rating_elem.get_text(strip=True)
|
||||
rating_match = re.search(r'(\d+\.?\d*)\s*/\s*10', rating_text)
|
||||
if rating_match:
|
||||
metadata['rating'] = f"{rating_match.group(1)}/10"
|
||||
break
|
||||
rating_match = re.search(r'(\d+\.?\d*)\s*/\s*5', rating_text)
|
||||
if rating_match:
|
||||
rating_val = float(rating_match.group(1)) * 2
|
||||
metadata['rating'] = f"{rating_val:.1f}/10"
|
||||
break
|
||||
|
||||
# Extract release year
|
||||
year_match = re.search(r'\b(19\d{2}|20\d{2})\b', page_text)
|
||||
if year_match:
|
||||
import datetime
|
||||
current_year = datetime.datetime.now().year + 2
|
||||
year = int(year_match.group(1))
|
||||
if 1950 <= year <= current_year:
|
||||
metadata['release_year'] = year
|
||||
|
||||
# Extract poster image from og:image
|
||||
og_image = soup.find('meta', property='og:image')
|
||||
if og_image:
|
||||
metadata['poster_image'] = og_image.get('content')
|
||||
|
||||
# Extract total episodes
|
||||
episodes_count = len(await self.get_episodes(anime_url))
|
||||
if episodes_count > 0:
|
||||
metadata['total_episodes'] = episodes_count
|
||||
|
||||
print(f"[ANIME-ULTIME] Extracted metadata: {metadata}")
|
||||
return metadata
|
||||
|
||||
except Exception as e:
|
||||
print(f"[ANIME-ULTIME] Error extracting metadata: {e}")
|
||||
return {}
|
||||
|
||||
async def search_anime(self, query: str, lang: str = "vostfr", include_metadata: bool = False) -> list[dict]:
|
||||
"""
|
||||
Search for anime on anime-ultime
|
||||
Returns list of anime with title, url, and cover image
|
||||
|
||||
Args:
|
||||
query: Search query string
|
||||
lang: Language preference (vostfr, vf)
|
||||
include_metadata: Whether to fetch full metadata for each result (slower)
|
||||
"""
|
||||
try:
|
||||
import time
|
||||
@@ -231,11 +345,19 @@ class AnimeUltimeDownloader(BaseDownloader):
|
||||
if not href.startswith('http'):
|
||||
href = urljoin("https://www.anime-ultime.net/", href)
|
||||
|
||||
results.append({
|
||||
result_item = {
|
||||
'title': better_title,
|
||||
'url': href,
|
||||
'type': 'search_result'
|
||||
})
|
||||
'type': 'search_result',
|
||||
'metadata': None
|
||||
}
|
||||
|
||||
# Fetch metadata if requested
|
||||
if include_metadata:
|
||||
metadata = await self.get_anime_metadata(href)
|
||||
result_item['metadata'] = metadata
|
||||
|
||||
results.append(result_item)
|
||||
|
||||
print(f"[ANIME-ULTIME] Found {len(results)} results")
|
||||
return results
|
||||
|
||||
Reference in New Issue
Block a user