Spaces:
Running
Running
Update main.py
Browse files
main.py
CHANGED
|
@@ -9,7 +9,7 @@ import urllib.parse
|
|
| 9 |
import mimetypes
|
| 10 |
from bs4 import BeautifulSoup
|
| 11 |
from fastapi import FastAPI, HTTPException, Request
|
| 12 |
-
from fastapi.responses import StreamingResponse
|
| 13 |
from fake_useragent import UserAgent
|
| 14 |
import uvicorn
|
| 15 |
|
|
@@ -26,12 +26,12 @@ ua = UserAgent(fallback='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/5
|
|
| 26 |
|
| 27 |
# MediaFire Cache
|
| 28 |
MEDIAFIRE_CACHE = {}
|
| 29 |
-
CACHE_TTL =
|
| 30 |
|
| 31 |
client = httpx.AsyncClient(
|
| 32 |
timeout=httpx.Timeout(60.0, read=None),
|
| 33 |
follow_redirects=True,
|
| 34 |
-
limits=httpx.Limits(max_connections=
|
| 35 |
)
|
| 36 |
|
| 37 |
@app.get("/")
|
|
@@ -63,7 +63,7 @@ def get_clean_filename(url):
|
|
| 63 |
decoded_url = urllib.parse.unquote(url)
|
| 64 |
name = decoded_url.split('/')[-1].split('?')[0]
|
| 65 |
if not name or '.' not in name:
|
| 66 |
-
|
| 67 |
return name
|
| 68 |
|
| 69 |
@app.get("/download")
|
|
@@ -78,45 +78,39 @@ async def download_proxy(request: Request, url: str, key: str = None):
|
|
| 78 |
|
| 79 |
# --- MediaFire Section ---
|
| 80 |
if "mediafire.com" in clean_url:
|
| 81 |
-
cached_data = MEDIAFIRE_CACHE.get(clean_url)
|
| 82 |
-
|
| 83 |
target_link = None
|
| 84 |
-
|
| 85 |
-
|
| 86 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
try:
|
| 88 |
-
|
| 89 |
-
async with httpx.AsyncClient(headers=
|
| 90 |
-
|
| 91 |
-
if
|
| 92 |
-
|
| 93 |
-
|
| 94 |
-
|
| 95 |
-
|
| 96 |
-
|
| 97 |
-
|
| 98 |
-
|
| 99 |
-
|
| 100 |
-
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
if target_link:
|
| 105 |
-
if target_link.startswith("//"): target_link = f"https:{target_link}"
|
| 106 |
-
elif target_link.startswith("/"): target_link = f"https://www.mediafire.com{target_link}"
|
| 107 |
-
MEDIAFIRE_CACHE[clean_url] = {'link': target_link, 'time': current_time}
|
| 108 |
except Exception as e:
|
| 109 |
-
print(f"
|
| 110 |
-
raise HTTPException(status_code=500, detail=str(e))
|
| 111 |
|
| 112 |
if target_link:
|
| 113 |
-
|
| 114 |
-
# ဒါဟာ MediaFire ဆီ Request ရောက်တဲ့အခါ query parameter အနေနဲ့ပဲ မြင်သွားမှာမို့ link မပျက်ပါဘူး
|
| 115 |
-
separator = "&" if "?" in target_link else "?"
|
| 116 |
-
hinted_link = f"{target_link}{separator}ext={filename}"
|
| 117 |
-
return RedirectResponse(url=hinted_link)
|
| 118 |
else:
|
| 119 |
-
raise HTTPException(status_code=404, detail="
|
| 120 |
|
| 121 |
# --- Google Drive Section ---
|
| 122 |
elif "drive.google.com" in clean_url:
|
|
@@ -133,13 +127,14 @@ async def download_proxy(request: Request, url: str, key: str = None):
|
|
| 133 |
headers = {"Authorization": f"Bearer {token}"}
|
| 134 |
if range_header: headers['Range'] = range_header
|
| 135 |
try:
|
| 136 |
-
|
| 137 |
-
r = await client.
|
| 138 |
if r.status_code in [200, 206]:
|
| 139 |
return await process_response(r, filename)
|
| 140 |
await r.aclose()
|
| 141 |
except: continue
|
| 142 |
-
|
|
|
|
| 143 |
public_url = f"https://drive.google.com/uc?export=download&id={file_id}"
|
| 144 |
return await stream_file(public_url, range_header, filename)
|
| 145 |
|
|
@@ -149,34 +144,54 @@ async def download_proxy(request: Request, url: str, key: str = None):
|
|
| 149 |
async def stream_file(target_url, range_header, filename):
|
| 150 |
headers = {'User-Agent': ua.random}
|
| 151 |
if range_header: headers['Range'] = range_header
|
|
|
|
| 152 |
try:
|
| 153 |
-
|
| 154 |
-
r = await client.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 155 |
return await process_response(r, filename)
|
|
|
|
|
|
|
| 156 |
except Exception as e:
|
|
|
|
| 157 |
raise HTTPException(status_code=500, detail=str(e))
|
| 158 |
|
| 159 |
async def process_response(r, filename):
|
| 160 |
mime_type, _ = mimetypes.guess_type(filename)
|
| 161 |
if not mime_type or 'application' in mime_type:
|
| 162 |
-
mime_type = 'video/
|
| 163 |
-
|
| 164 |
-
|
| 165 |
-
res_headers = {
|
| 166 |
-
|
| 167 |
-
|
|
|
|
|
|
|
|
|
|
| 168 |
|
| 169 |
-
|
| 170 |
-
|
|
|
|
|
|
|
| 171 |
|
| 172 |
async def stream_generator():
|
| 173 |
try:
|
| 174 |
-
async for chunk in r.aiter_bytes(chunk_size=
|
| 175 |
yield chunk
|
| 176 |
finally:
|
| 177 |
await r.aclose()
|
| 178 |
|
| 179 |
-
return StreamingResponse(
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
|
| 181 |
if __name__ == "__main__":
|
| 182 |
uvicorn.run(app, host="0.0.0.0", port=7860)
|
|
|
|
| 9 |
import mimetypes
|
| 10 |
from bs4 import BeautifulSoup
|
| 11 |
from fastapi import FastAPI, HTTPException, Request
|
| 12 |
+
from fastapi.responses import StreamingResponse
|
| 13 |
from fake_useragent import UserAgent
|
| 14 |
import uvicorn
|
| 15 |
|
|
|
|
| 26 |
|
| 27 |
# MediaFire Cache
|
| 28 |
MEDIAFIRE_CACHE = {}
|
| 29 |
+
CACHE_TTL = 3600
|
| 30 |
|
| 31 |
client = httpx.AsyncClient(
|
| 32 |
timeout=httpx.Timeout(60.0, read=None),
|
| 33 |
follow_redirects=True,
|
| 34 |
+
limits=httpx.Limits(max_connections=200, max_keepalive_connections=50)
|
| 35 |
)
|
| 36 |
|
| 37 |
@app.get("/")
|
|
|
|
| 63 |
decoded_url = urllib.parse.unquote(url)
|
| 64 |
name = decoded_url.split('/')[-1].split('?')[0]
|
| 65 |
if not name or '.' not in name:
|
| 66 |
+
return "video.mp4"
|
| 67 |
return name
|
| 68 |
|
| 69 |
@app.get("/download")
|
|
|
|
| 78 |
|
| 79 |
# --- MediaFire Section ---
|
| 80 |
if "mediafire.com" in clean_url:
|
|
|
|
|
|
|
| 81 |
target_link = None
|
| 82 |
+
|
| 83 |
+
# 1. Check Cache
|
| 84 |
+
cached = MEDIAFIRE_CACHE.get(clean_url)
|
| 85 |
+
if cached and (current_time - cached['time']) < CACHE_TTL:
|
| 86 |
+
target_link = cached['link']
|
| 87 |
+
|
| 88 |
+
# 2. Fetch New Link if not cached
|
| 89 |
+
if not target_link:
|
| 90 |
try:
|
| 91 |
+
headers = {'User-Agent': ua.random}
|
| 92 |
+
async with httpx.AsyncClient(headers=headers, follow_redirects=True, timeout=30.0) as temp_client:
|
| 93 |
+
r = await temp_client.get(clean_url)
|
| 94 |
+
if r.status_code == 200:
|
| 95 |
+
# Regex for direct link
|
| 96 |
+
match = re.search(r"https?://download[0-9]+\.mediafire\.com/[a-zA-Z0-9_-]+/[a-zA-Z0-9_-]+/[^\s'\"]+", r.text)
|
| 97 |
+
if match:
|
| 98 |
+
target_link = match.group(0).replace('"', '').replace("'", "")
|
| 99 |
+
else:
|
| 100 |
+
soup = BeautifulSoup(r.text, 'html.parser')
|
| 101 |
+
btn = soup.find('a', {'id': 'downloadButton'}) or soup.find('a', {'class': 'input_btn_p'})
|
| 102 |
+
if btn: target_link = btn.get('href')
|
| 103 |
+
|
| 104 |
+
if target_link:
|
| 105 |
+
if target_link.startswith("//"): target_link = f"https:{target_link}"
|
| 106 |
+
MEDIAFIRE_CACHE[clean_url] = {'link': target_link, 'time': current_time}
|
|
|
|
|
|
|
|
|
|
|
|
|
| 107 |
except Exception as e:
|
| 108 |
+
print(f"Scraper Error: {e}")
|
|
|
|
| 109 |
|
| 110 |
if target_link:
|
| 111 |
+
return await stream_file(target_link, range_header, filename)
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
else:
|
| 113 |
+
raise HTTPException(status_code=404, detail="MediaFire link could not be resolved.")
|
| 114 |
|
| 115 |
# --- Google Drive Section ---
|
| 116 |
elif "drive.google.com" in clean_url:
|
|
|
|
| 127 |
headers = {"Authorization": f"Bearer {token}"}
|
| 128 |
if range_header: headers['Range'] = range_header
|
| 129 |
try:
|
| 130 |
+
# Use stream=True to handle large video files
|
| 131 |
+
r = await client.get(api_link, headers=headers, stream=True)
|
| 132 |
if r.status_code in [200, 206]:
|
| 133 |
return await process_response(r, filename)
|
| 134 |
await r.aclose()
|
| 135 |
except: continue
|
| 136 |
+
|
| 137 |
+
# Public Fallback
|
| 138 |
public_url = f"https://drive.google.com/uc?export=download&id={file_id}"
|
| 139 |
return await stream_file(public_url, range_header, filename)
|
| 140 |
|
|
|
|
| 144 |
async def stream_file(target_url, range_header, filename):
|
| 145 |
headers = {'User-Agent': ua.random}
|
| 146 |
if range_header: headers['Range'] = range_header
|
| 147 |
+
|
| 148 |
try:
|
| 149 |
+
# Request with stream=True
|
| 150 |
+
r = await client.get(target_url, headers=headers, stream=True)
|
| 151 |
+
|
| 152 |
+
# If MediaFire returns HTML (Error), raise exception to retry or show error
|
| 153 |
+
if "text/html" in r.headers.get("Content-Type", "").lower():
|
| 154 |
+
await r.aclose()
|
| 155 |
+
raise HTTPException(status_code=415, detail="MediaFire blocked the request. Try again.")
|
| 156 |
+
|
| 157 |
return await process_response(r, filename)
|
| 158 |
+
except HTTPException:
|
| 159 |
+
raise
|
| 160 |
except Exception as e:
|
| 161 |
+
print(f"Stream Error: {e}")
|
| 162 |
raise HTTPException(status_code=500, detail=str(e))
|
| 163 |
|
| 164 |
async def process_response(r, filename):
|
| 165 |
mime_type, _ = mimetypes.guess_type(filename)
|
| 166 |
if not mime_type or 'application' in mime_type:
|
| 167 |
+
mime_type = 'video/mp4' if filename.lower().endswith('.mp4') else 'video/x-matroska'
|
| 168 |
+
|
| 169 |
+
# Essential Headers for Resume and Seeking
|
| 170 |
+
res_headers = {
|
| 171 |
+
'Content-Type': mime_type,
|
| 172 |
+
'Accept-Ranges': 'bytes',
|
| 173 |
+
'Content-Disposition': f'inline; filename="{urllib.parse.quote(filename)}"',
|
| 174 |
+
'Cache-Control': 'no-cache'
|
| 175 |
+
}
|
| 176 |
|
| 177 |
+
if 'content-length' in r.headers:
|
| 178 |
+
res_headers['Content-Length'] = r.headers['content-length']
|
| 179 |
+
if 'content-range' in r.headers:
|
| 180 |
+
res_headers['Content-Range'] = r.headers['content-range']
|
| 181 |
|
| 182 |
async def stream_generator():
|
| 183 |
try:
|
| 184 |
+
async for chunk in r.aiter_bytes(chunk_size=131072): # 128KB chunks
|
| 185 |
yield chunk
|
| 186 |
finally:
|
| 187 |
await r.aclose()
|
| 188 |
|
| 189 |
+
return StreamingResponse(
|
| 190 |
+
stream_generator(),
|
| 191 |
+
status_code=r.status_code,
|
| 192 |
+
headers=res_headers,
|
| 193 |
+
media_type=mime_type
|
| 194 |
+
)
|
| 195 |
|
| 196 |
if __name__ == "__main__":
|
| 197 |
uvicorn.run(app, host="0.0.0.0", port=7860)
|