root
3afad41d46
This commit implements a complete reorganization of the downloader system with a clear distinction between anime streaming sites and video hosting services. ## Structure Changes **New Organization:** - `app/downloaders/anime_sites/` - Anime streaming sites (catalogs + metadata) - `app/downloaders/video_players/` - Video hosting services (file downloads) **Base Classes:** - `BaseAnimeSite` - For anime providers (search, episodes, metadata) - `BaseVideoPlayer` - For video players (download link extraction) **Migrated Downloaders:** Anime Sites (4): - AnimeSama, NekoSama, AnimeUltime, Vostfree Video Players (8): - Doodstream, Sibnet, VidMoly, SendVid, Lpayer, 1fichier, Uptobox, Rapidfile ## Key Improvements 1. **Clear Separation**: Distinct base classes for different use cases 2. **Preserved Functionality**: All existing features maintained - VidMoly: M3U8 support, Playwright, multi-domains, target_filename param - SendVid: target_filename parameter support - All others: No behavioral changes 3. **Better Organization**: - Anime sites: search_anime(), get_episodes(), get_anime_metadata() - Video players: get_download_link(url, target_filename=None) 4. **Fixed Imports**: Updated cross-imports in AnimeSama - from ..video_players.vidmoly import - from ..video_players.sendvid import - from ..video_players.sibnet import - from ..video_players.lpayer import 5. **Updated Tests**: All test imports use new structure 6. **Updated Providers**: Added 4 missing file hosts to providers.py ## Backward Compatibility ✅ Main API unchanged: get_downloader() works identically ✅ All 23 tests passing ✅ Frontend fully functional ✅ No breaking changes for users ## Documentation - RESTRUCTURATION_SUMMARY.md - Technical details - FIX_IMPORT_ERROR.md - Import error resolution - IMPORT_VERIFICATION_REPORT.md - Complete import verification - FRONTEND_VERIFICATION_FINAL.md - Frontend validation 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>
436 lines
18 KiB
Python
436 lines
18 KiB
Python
from .base import BaseAnimeSite
|
|
from bs4 import BeautifulSoup
|
|
import re
|
|
import httpx
|
|
from urllib.parse import urljoin
|
|
|
|
|
|
class AnimeUltimeDownloader(BaseAnimeSite):
|
|
"""Downloader for anime-ultime.net"""
|
|
|
|
BASE_DOMAINS = ["anime-ultime.com", "anime-ultime.net", "www.anime-ultime.net"]
|
|
|
|
def can_handle(self, url: str) -> bool:
|
|
return any(domain in url.lower() for domain in self.BASE_DOMAINS)
|
|
|
|
async def get_download_link(self, url: str) -> tuple[str, str]:
|
|
"""
|
|
Extract download link from anime-ultime URL
|
|
Anime-Ultime stores video links in og:video meta tags
|
|
"""
|
|
try:
|
|
# Follow redirects
|
|
response = await self.client.get(url, follow_redirects=True)
|
|
final_url = str(response.url)
|
|
|
|
# Parse the page
|
|
soup = BeautifulSoup(response.text, 'lxml')
|
|
|
|
# Method 0: Look for og:video meta tag (most reliable for anime-ultime)
|
|
og_video = soup.find('meta', property='og:video')
|
|
if og_video and og_video.get('content'):
|
|
video_url = og_video['content']
|
|
if video_url.endswith('.mp4'):
|
|
filename = self._generate_filename(final_url)
|
|
print(f"[ANIME-ULTIME] Found og:video link: {video_url}")
|
|
return video_url, filename
|
|
|
|
# Method 1: Look for direct download links (DDL)
|
|
# Anime-Ultime often uses links to file hosts
|
|
download_links = soup.find_all('a', href=True)
|
|
for link in download_links:
|
|
href = link['href']
|
|
text = link.get_text().lower()
|
|
|
|
# Look for download buttons/links
|
|
if any(keyword in text for keyword in ['télécharger', 'download', 'ddl', 'mega', 'google', 'drive']):
|
|
# Check if it's a direct link or to a file host
|
|
if any(host in href.lower() for host in ['mega.nz', 'drive.google.com', 'uptobox.com', '1fichier.com']):
|
|
filename = self._generate_filename(final_url)
|
|
return href, filename
|
|
|
|
# Method 2: Look for iframe with video player
|
|
iframes = soup.find_all('iframe')
|
|
for iframe in iframes:
|
|
src = iframe.get('src', '')
|
|
if src and any(provider in src for provider in ['video', 'player', 'stream', 'play']):
|
|
if src.startswith('http'):
|
|
filename = self._generate_filename(final_url)
|
|
return src, filename
|
|
|
|
# Method 3: Look for video tags
|
|
videos = soup.find_all('video')
|
|
for video in videos:
|
|
src = video.get('src', '')
|
|
if src:
|
|
filename = self._generate_filename(final_url)
|
|
return src, filename
|
|
|
|
# Check source tags
|
|
sources = video.find_all('source')
|
|
for source in sources:
|
|
src = source.get('src', '')
|
|
if src:
|
|
filename = self._generate_filename(final_url)
|
|
return src, filename
|
|
|
|
# Method 4: Look in scripts for video URLs
|
|
scripts = soup.find_all('script')
|
|
for script in scripts:
|
|
if script.string:
|
|
# Look for common video patterns
|
|
patterns = [
|
|
r'(https?://[^"\'>\s]+\.(?:mp4|m3u8|mkv)(?:\?[^"\'>\s]*)?)',
|
|
r'"url":"([^"]+)"',
|
|
r'"video":"([^"]+)"',
|
|
r'"file":"([^"]+)"',
|
|
r'file:\s*"([^"]+)"',
|
|
]
|
|
|
|
for pattern in patterns:
|
|
matches = re.findall(pattern, script.string)
|
|
for match in matches:
|
|
# Clean up escaped characters
|
|
match = match.replace('\\/', '/').replace('\\', '')
|
|
if any(ext in match for ext in ['mp4', 'm3u8', 'mkv']):
|
|
filename = self._generate_filename(final_url)
|
|
return match, filename
|
|
|
|
# Look for anime-ultime specific patterns
|
|
# They sometimes store links in JavaScript variables
|
|
ddl_match = re.search(r'ddl["\']?\s*:\s*["\']([^"\']+)["\']', script.string)
|
|
if ddl_match:
|
|
ddl_url = ddl_match.group(1)
|
|
if ddl_url.startswith('http'):
|
|
filename = self._generate_filename(final_url)
|
|
return ddl_url, filename
|
|
|
|
# Method 5: Look for links with specific classes or IDs
|
|
# Anime-Ultime might use specific class names for download links
|
|
potential_links = soup.find_all('a', class_=re.compile(r'download|ddl|episode', re.I))
|
|
for link in potential_links:
|
|
href = link.get('href', '')
|
|
if href and href.startswith('http'):
|
|
filename = self._generate_filename(final_url)
|
|
return href, filename
|
|
|
|
# If nothing found, raise error
|
|
raise Exception("Could not find download link on page")
|
|
|
|
except Exception as e:
|
|
raise Exception(f"Error extracting Anime-Ultime link: {str(e)}")
|
|
|
|
def _generate_filename(self, url: str) -> str:
|
|
"""Generate filename from URL"""
|
|
# Extract anime name and episode from URL
|
|
# URL formats:
|
|
# - info-0-1/30200
|
|
# - info-0-1/30200/Naruto-OAV-01-vostfr
|
|
# - file-0-1/2991-Naruto-OAV
|
|
|
|
anime_name = "Anime"
|
|
episode = "01"
|
|
|
|
# Format: info-0-1/EPISODE_ID or info-0-1/EPISODE_ID/NAME-EP-vostfr
|
|
if 'info-0-1/' in url:
|
|
# Extract episode ID
|
|
ep_match = re.search(r'info-0-1/(\d+)', url)
|
|
if ep_match:
|
|
ep_id = ep_match.group(1)
|
|
|
|
# Try to get anime name from URL path
|
|
name_match = re.search(r'info-0-1/\d+/([^/]+)', url)
|
|
if name_match:
|
|
raw_name = name_match.group(1)
|
|
# Extract episode number
|
|
ep_num_match = re.search(r'-(\d+)-vostfr$', raw_name, re.I)
|
|
if ep_num_match:
|
|
episode = ep_num_match.group(1).zfill(2)
|
|
# Remove episode number and suffix from name
|
|
anime_name = re.sub(r'-\d+-vostfr$', '', raw_name, flags=re.I).replace('-', ' ')
|
|
else:
|
|
# Just use the ID
|
|
anime_name = f"Episode {ep_id}"
|
|
else:
|
|
anime_name = f"Episode {ep_id}"
|
|
|
|
elif 'file-0-1/' in url:
|
|
# Extract from file-0-1/ID-NAME format
|
|
file_match = re.search(r'file-0-1/\d+-(.+)$', url)
|
|
if file_match:
|
|
anime_name = file_match.group(1).replace('-', ' ')
|
|
|
|
# Sanitize filename
|
|
anime_name = anime_name.replace('/', ' ').strip()
|
|
filename = f"{anime_name} - Episode {episode}.mp4"
|
|
return filename.title()
|
|
|
|
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
|
|
start = time.time()
|
|
print(f"[ANIME-ULTIME] Searching for '{query}' ({lang})...")
|
|
|
|
# Anime-Ultime uses POST for search
|
|
search_url = "https://www.anime-ultime.net/search-0-1"
|
|
|
|
response = await self.client.post(search_url, data={'search': query})
|
|
soup = BeautifulSoup(response.text, 'lxml')
|
|
|
|
elapsed = time.time() - start
|
|
print(f"[ANIME-ULTIME] Got response {response.status_code} in {elapsed:.2f}s")
|
|
|
|
results = []
|
|
|
|
# Look for search result links - better parsing
|
|
# Search results use file-0-1/ pattern, not info-
|
|
search_results = soup.find_all('a', href=re.compile(r'file-0-1/'))
|
|
|
|
seen_urls = set()
|
|
for result in search_results[:10]: # Limit to 10 results
|
|
href = result.get('href', '')
|
|
raw_title = result.get_text().strip()
|
|
|
|
# Skip if no href
|
|
if not href:
|
|
continue
|
|
|
|
# Skip duplicates
|
|
if href in seen_urls:
|
|
continue
|
|
seen_urls.add(href)
|
|
|
|
# Extract better title from URL or parent elements
|
|
better_title = raw_title
|
|
|
|
# If raw_title is just "Télécharger" or similar, try to find better title
|
|
if len(raw_title) < 5 or raw_title.lower() in ['télécharger', 'download', 'ddl']:
|
|
# Try to extract from URL (file-0-1/ID-Title format)
|
|
url_match = re.search(r'file-0-1/\d+-(.+)$', href)
|
|
if url_match:
|
|
better_title = url_match.group(1).replace('-', ' ').title()
|
|
|
|
# If still no good title, look at parent/row elements
|
|
if len(better_title) < 5:
|
|
# Check parent row (table structure)
|
|
row = result.find_parent(['tr', 'td', 'div'])
|
|
if row:
|
|
# Look for text in the row that's not the link text
|
|
row_text = row.get_text().strip()
|
|
# Remove the link text from row text
|
|
if raw_title in row_text:
|
|
row_text = row_text.replace(raw_title, '').strip()
|
|
if len(row_text) > 5 and len(row_text) < 100:
|
|
better_title = row_text
|
|
|
|
# Make URL absolute
|
|
if not href.startswith('http'):
|
|
href = urljoin("https://www.anime-ultime.net/", href)
|
|
|
|
result_item = {
|
|
'title': better_title,
|
|
'url': href,
|
|
'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
|
|
|
|
except Exception as e:
|
|
print(f"[ANIME-ULTIME] Error: {e}")
|
|
return []
|
|
|
|
async def get_episodes(self, anime_url: str, lang: str = "vostfr") -> list[dict]:
|
|
"""
|
|
Get list of episodes for an anime
|
|
Returns list of episode numbers and their URLs
|
|
"""
|
|
try:
|
|
response = await self.client.get(anime_url)
|
|
soup = BeautifulSoup(response.text, 'lxml')
|
|
|
|
episodes = []
|
|
|
|
# Look for episode links - anime-ultime uses info-XXXXX-Name-XX-vostfr format
|
|
# The URL pattern is info-0-1/ID-Anime-Name-XX-vostfr where XX is episode number
|
|
episode_links = soup.find_all('a', href=re.compile(r'info-0-1/\d+'))
|
|
|
|
for link in episode_links:
|
|
href = link.get('href', '')
|
|
text = link.get_text().strip()
|
|
|
|
# Extract episode number from URL pattern
|
|
# Matches: info-0-1/30200/Naruto-OAV-01-vostfr
|
|
match = re.search(r'-(\d+)-vostfr$', href, re.I)
|
|
if not match:
|
|
# Try other patterns
|
|
match = re.search(r'Episode[-\s]?(\d+)', href, re.I)
|
|
if not match:
|
|
# Try to extract from text
|
|
match = re.search(r'(\d+)', text)
|
|
|
|
if match:
|
|
episode_num = match.group(1).zfill(2) # Pad with zero
|
|
|
|
# Extract the episode ID from href and build correct URL
|
|
# href might be "info-0-1/30200" or "info-0-1/30200/..."
|
|
# We need: https://www.anime-ultime.net/info-0-1/30200
|
|
ep_id_match = re.search(r'info-0-1/(\d+)', href)
|
|
if ep_id_match:
|
|
ep_id = ep_id_match.group(1)
|
|
# Build the correct episode URL
|
|
episode_url = f"https://www.anime-ultime.net/info-0-1/{ep_id}"
|
|
else:
|
|
# Fallback to making URL absolute
|
|
if not href.startswith('http'):
|
|
href = urljoin(anime_url, href)
|
|
episode_url = href
|
|
|
|
episodes.append({
|
|
'episode': episode_num,
|
|
'url': episode_url,
|
|
'title': text
|
|
})
|
|
|
|
# Remove duplicates and sort
|
|
seen = set()
|
|
unique_episodes = []
|
|
for ep in episodes:
|
|
if ep['episode'] not in seen:
|
|
seen.add(ep['episode'])
|
|
unique_episodes.append(ep)
|
|
|
|
unique_episodes.sort(key=lambda x: int(x['episode']))
|
|
|
|
return unique_episodes
|
|
|
|
except Exception as e:
|
|
print(f"Error getting episodes: {e}")
|
|
return []
|