Spaces:
Runtime error
Runtime error
Update main.py
Browse files
main.py
CHANGED
|
@@ -27,9 +27,7 @@ TEMP_DIR = "/tmp/fb_cache"
|
|
| 27 |
if not os.path.exists(TEMP_DIR):
|
| 28 |
os.makedirs(TEMP_DIR)
|
| 29 |
|
| 30 |
-
|
| 31 |
-
RAPID_API_KEY = "7...f" # <-- Bro ရဲ့ Key ကို ပြန်ထည့်ပါ
|
| 32 |
-
RAPID_API_HOST = "facebook-media-downloader1.p.rapidapi.com"
|
| 33 |
|
| 34 |
def cleanup(file_path: str):
|
| 35 |
if os.path.exists(file_path):
|
|
@@ -43,129 +41,123 @@ def clean_fb_url(url: str):
|
|
| 43 |
except: pass
|
| 44 |
return urllib.parse.unquote(url).replace('&', '&')
|
| 45 |
|
| 46 |
-
def download_temp_file(url: str, suffix: str):
|
| 47 |
-
if not url: return None
|
| 48 |
-
path = os.path.join(TEMP_DIR, f"{uuid.uuid4()}{suffix}")
|
| 49 |
-
try:
|
| 50 |
-
headers = {'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36'}
|
| 51 |
-
r = requests.get(url, stream=True, timeout=20, headers=headers)
|
| 52 |
-
if r.status_code == 200:
|
| 53 |
-
with open(path, 'wb') as f:
|
| 54 |
-
for chunk in r.iter_content(chunk_size=1024*1024):
|
| 55 |
-
f.write(chunk)
|
| 56 |
-
if os.path.getsize(path) > 1024:
|
| 57 |
-
return path
|
| 58 |
-
if os.path.exists(path): os.remove(path)
|
| 59 |
-
except Exception as e:
|
| 60 |
-
print(f"Download failed: {e}")
|
| 61 |
-
if os.path.exists(path): os.remove(path)
|
| 62 |
-
return None
|
| 63 |
-
|
| 64 |
-
def get_stream_type(file_path):
|
| 65 |
-
try:
|
| 66 |
-
cmd = ['ffprobe', '-v', 'error', '-show_entries', 'stream=codec_type', '-of', 'json', file_path]
|
| 67 |
-
result = subprocess.run(cmd, capture_output=True, text=True)
|
| 68 |
-
data = json.loads(result.stdout)
|
| 69 |
-
return [s['codec_type'] for s in data.get('streams', [])]
|
| 70 |
-
except: return []
|
| 71 |
-
|
| 72 |
@app.get("/")
|
| 73 |
def root():
|
| 74 |
-
return {"message": "Bumiz
|
| 75 |
|
| 76 |
-
|
| 77 |
-
|
|
|
|
| 78 |
html_source = payload.get("html", "")
|
| 79 |
-
logs = ["Scraper: Starting..."]
|
| 80 |
-
|
| 81 |
if not html_source:
|
| 82 |
-
return {"success": False, "error": "HTML Source is empty"
|
| 83 |
|
| 84 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
audio_url = None
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
|
| 87 |
-
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
|
| 91 |
-
|
| 92 |
-
|
| 93 |
-
]
|
| 94 |
-
for p in v_patterns:
|
| 95 |
-
match = re.search(p, html_source)
|
| 96 |
-
if match:
|
| 97 |
-
url = clean_fb_url(match.group(1))
|
| 98 |
-
if url and "fbcdn.net" in url and "/v/" in url:
|
| 99 |
-
video_url = url
|
| 100 |
-
logs.append(f"Found Video: {url[:80]}...")
|
| 101 |
-
break
|
| 102 |
-
|
| 103 |
-
a_patterns = [
|
| 104 |
-
r'"mime_type":"audio/[^"]+".*?"base_url":"([^"]+)"',
|
| 105 |
-
r'"audio_channel_configuration".*?"base_url":"([^"]+)"'
|
| 106 |
-
]
|
| 107 |
-
for p in a_patterns:
|
| 108 |
-
match = re.search(p, html_source)
|
| 109 |
-
if match:
|
| 110 |
-
url = clean_fb_url(match.group(1))
|
| 111 |
-
if url and "fbcdn.net" in url:
|
| 112 |
-
audio_url = url
|
| 113 |
-
logs.append(f"Found Audio: {url[:80]}...")
|
| 114 |
-
break
|
| 115 |
-
|
| 116 |
-
if not video_url:
|
| 117 |
-
return {"success": False, "error": "Video link not found", "logs": logs}
|
| 118 |
-
|
| 119 |
-
logs.append("Downloading streams...")
|
| 120 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 121 |
file_id = str(uuid.uuid4())
|
| 122 |
-
|
| 123 |
-
v_tmp = None
|
| 124 |
-
a_tmp = None
|
| 125 |
|
| 126 |
try:
|
| 127 |
-
v_tmp =
|
| 128 |
-
|
|
|
|
|
|
|
|
|
|
| 129 |
|
| 130 |
-
|
| 131 |
-
if audio_url
|
|
|
|
|
|
|
|
|
|
| 132 |
|
| 133 |
-
#
|
| 134 |
-
logs.append("Starting HQ Encoding...")
|
| 135 |
cmd = ['ffmpeg', '-y', '-i', v_tmp]
|
| 136 |
-
if
|
| 137 |
cmd += ['-i', a_tmp, '-map', '0:v:0', '-map', '1:a:0']
|
| 138 |
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
| 142 |
-
|
| 143 |
-
|
| 144 |
|
| 145 |
-
|
| 146 |
-
if not os.path.exists(final_video) or os.path.getsize(final_video) < 1000:
|
| 147 |
-
raise Exception("Encoding failed: FFmpeg returned an empty file.")
|
| 148 |
|
| 149 |
-
if v_tmp: os.remove(v_tmp)
|
| 150 |
-
if a_tmp: os.remove(a_tmp)
|
| 151 |
|
| 152 |
space_id = os.getenv('SPACE_ID', '').replace("/", "-")
|
| 153 |
-
return {
|
| 154 |
-
"success": True,
|
| 155 |
-
"method": "Private (Full Processing)",
|
| 156 |
-
"video_url": f"https://{space_id}.hf.space/get_file?id={file_id}.mp4",
|
| 157 |
-
"logs": logs
|
| 158 |
-
}
|
| 159 |
|
| 160 |
except Exception as e:
|
| 161 |
-
|
| 162 |
-
|
| 163 |
-
|
| 164 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 165 |
|
| 166 |
@app.get("/get_file")
|
| 167 |
async def get_file(id: str):
|
| 168 |
file_path = os.path.join(TEMP_DIR, id)
|
| 169 |
if os.path.exists(file_path):
|
| 170 |
-
return FileResponse(file_path, media_type="video/mp4", filename=f"
|
| 171 |
-
raise HTTPException(status_code=404
|
|
|
|
| 27 |
if not os.path.exists(TEMP_DIR):
|
| 28 |
os.makedirs(TEMP_DIR)
|
| 29 |
|
| 30 |
+
MAX_FILE_AGE = 1800
|
|
|
|
|
|
|
| 31 |
|
| 32 |
def cleanup(file_path: str):
|
| 33 |
if os.path.exists(file_path):
|
|
|
|
| 41 |
except: pass
|
| 42 |
return urllib.parse.unquote(url).replace('&', '&')
|
| 43 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
@app.get("/")
|
| 45 |
def root():
|
| 46 |
+
return {"message": "Bumiz Facebook Engine: High-Quality Direct Scraper Active"}
|
| 47 |
|
| 48 |
+
# --- ၁။ SOURCE CODE ကို ANALYZE လုပ်ပြီး RESOLUTION အားလုံး ရှာဖွေခြင်း ---
|
| 49 |
+
@app.post("/analyze-private")
|
| 50 |
+
async def analyze_private_fb(payload: dict = Body(...)):
|
| 51 |
html_source = payload.get("html", "")
|
|
|
|
|
|
|
| 52 |
if not html_source:
|
| 53 |
+
return {"success": False, "error": "HTML Source is empty"}
|
| 54 |
|
| 55 |
+
found_qualities = []
|
| 56 |
+
# Video Representations ရှာဖွေခြင်း
|
| 57 |
+
matches = re.findall(r'"height":(\d+),.*?"base_url":"([^"]+)"', html_source)
|
| 58 |
+
|
| 59 |
+
seen_urls = set()
|
| 60 |
+
for height, url in matches:
|
| 61 |
+
h = int(height)
|
| 62 |
+
u = clean_fb_url(url)
|
| 63 |
+
if u and u not in seen_urls and "fbcdn.net" in u:
|
| 64 |
+
found_qualities.append({
|
| 65 |
+
"label": f"{h}p Quality",
|
| 66 |
+
"url": u,
|
| 67 |
+
"height": h,
|
| 68 |
+
"size_mb": 0,
|
| 69 |
+
"type": "hd" if h >= 720 else "sd"
|
| 70 |
+
})
|
| 71 |
+
seen_urls.add(u)
|
| 72 |
+
|
| 73 |
+
found_qualities.sort(key=lambda x: x['height'], reverse=True)
|
| 74 |
+
|
| 75 |
+
# Audio Link ရှာဖွေခြင်း
|
| 76 |
audio_url = None
|
| 77 |
+
a_match = re.search(r'"mime_type":"audio/[^"]+","base_url":"([^"]+)"', html_source)
|
| 78 |
+
if a_match:
|
| 79 |
+
audio_url = clean_fb_url(a_match.group(1))
|
| 80 |
+
|
| 81 |
+
if not found_qualities:
|
| 82 |
+
return {"success": False, "error": "No videos found in source code."}
|
| 83 |
|
| 84 |
+
return {"success": True, "qualities": found_qualities, "audio_url": audio_url}
|
| 85 |
+
|
| 86 |
+
# --- ၂။ ရွေးချယ်လိုက်သော QUALITY ကို အကြည်ဆုံး ဖြစ်အောင် PROCESSING လုပ်ခြင်း ---
|
| 87 |
+
@app.post("/process-private")
|
| 88 |
+
async def process_private_fb(background_tasks: BackgroundTasks, payload: dict = Body(...)):
|
| 89 |
+
# request လာတိုင်း အဟော��်းတွေကို ရှင်းမယ်
|
| 90 |
+
background_tasks.add_task(lambda: [os.remove(os.path.join(TEMP_DIR, f)) for f in os.listdir(TEMP_DIR) if time.time() - os.stat(os.path.join(TEMP_DIR, f)).st_mtime > MAX_FILE_AGE])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 91 |
|
| 92 |
+
video_url = payload.get("video_url")
|
| 93 |
+
audio_url = payload.get("audio_url")
|
| 94 |
+
compress_mode = payload.get("compress_mode")
|
| 95 |
+
|
| 96 |
file_id = str(uuid.uuid4())
|
| 97 |
+
final_video_path = os.path.join(TEMP_DIR, f"{file_id}.mp4") # NameError မဖြစ်အောင် အပေါ်မှာ ကြိုသတ်မှတ်သည်
|
|
|
|
|
|
|
| 98 |
|
| 99 |
try:
|
| 100 |
+
v_tmp = os.path.join(TEMP_DIR, f"{file_id}_v.mp4")
|
| 101 |
+
a_tmp = os.path.join(TEMP_DIR, f"{file_id}_a.mp3")
|
| 102 |
+
|
| 103 |
+
headers = {'User-Agent': 'Mozilla/5.0'}
|
| 104 |
+
with open(v_tmp, 'wb') as f: f.write(requests.get(video_url, headers=headers, timeout=60).content)
|
| 105 |
|
| 106 |
+
has_audio = False
|
| 107 |
+
if audio_url:
|
| 108 |
+
with open(a_tmp, 'wb') as f: f.write(requests.get(audio_url, headers=headers, timeout=60).content)
|
| 109 |
+
if os.path.exists(a_tmp) and os.path.getsize(a_tmp) > 1000:
|
| 110 |
+
has_audio = True
|
| 111 |
|
| 112 |
+
# FFmpeg Command Build
|
|
|
|
| 113 |
cmd = ['ffmpeg', '-y', '-i', v_tmp]
|
| 114 |
+
if has_audio:
|
| 115 |
cmd += ['-i', a_tmp, '-map', '0:v:0', '-map', '1:a:0']
|
| 116 |
|
| 117 |
+
if compress_mode and compress_mode != "original":
|
| 118 |
+
cmd += ['-vf', f"scale=w='trunc(oh*a/2)*2':h={compress_mode}"]
|
| 119 |
+
|
| 120 |
+
# Force High Quality Settings
|
| 121 |
+
cmd += ['-c:v', 'libx264', '-crf', '20', '-preset', 'superfast', '-c:a', 'aac', '-b:a', '128k', '-movflags', '+faststart', final_video_path]
|
| 122 |
|
| 123 |
+
subprocess.run(cmd, check=True, capture_output=True)
|
|
|
|
|
|
|
| 124 |
|
| 125 |
+
if os.path.exists(v_tmp): os.remove(v_tmp)
|
| 126 |
+
if os.path.exists(a_tmp): os.remove(a_tmp)
|
| 127 |
|
| 128 |
space_id = os.getenv('SPACE_ID', '').replace("/", "-")
|
| 129 |
+
return {"success": True, "video_url": f"https://{space_id}.hf.space/get_file?id={file_id}.mp4"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 130 |
|
| 131 |
except Exception as e:
|
| 132 |
+
return {"success": False, "error": str(e)}
|
| 133 |
+
|
| 134 |
+
# --- ၃။ PUBLIC DOWNLOAD LOGIC (No RapidAPI) ---
|
| 135 |
+
@app.get("/download")
|
| 136 |
+
async def download_public(url: str = Query(...)):
|
| 137 |
+
file_id = hashlib.md5(url.encode()).hexdigest()
|
| 138 |
+
video_path = os.path.join(TEMP_DIR, f"{file_id}.mp4")
|
| 139 |
+
|
| 140 |
+
if os.path.exists(video_path):
|
| 141 |
+
space_id = os.getenv('SPACE_ID', '').replace("/", "-")
|
| 142 |
+
return {"success": True, "video_url": f"get_file?id={file_id}.mp4"}
|
| 143 |
+
|
| 144 |
+
temp_raw = os.path.join(TEMP_DIR, f"{file_id}_raw.mp4")
|
| 145 |
+
try:
|
| 146 |
+
ydl_opts = {'format': 'best', 'outtmpl': temp_raw, 'quiet': True, 'socket_timeout': 20}
|
| 147 |
+
with yt_dlp.YoutubeDL(ydl_opts) as ydl:
|
| 148 |
+
ydl.download([url])
|
| 149 |
+
|
| 150 |
+
if os.path.exists(temp_raw):
|
| 151 |
+
subprocess.run(['ffmpeg', '-y', '-i', temp_raw, '-c:v', 'libx264', '-crf', '22', '-preset', 'ultrafast', '-c:a', 'aac', '-movflags', '+faststart', video_path], check=True)
|
| 152 |
+
os.remove(temp_raw)
|
| 153 |
+
space_id = os.getenv('SPACE_ID', '').replace("/", "-")
|
| 154 |
+
return {"success": True, "video_url": f"get_file?id={file_id}.mp4"}
|
| 155 |
+
except: pass
|
| 156 |
+
return {"success": False, "error": "Public failed. Try Private Mode."}
|
| 157 |
|
| 158 |
@app.get("/get_file")
|
| 159 |
async def get_file(id: str):
|
| 160 |
file_path = os.path.join(TEMP_DIR, id)
|
| 161 |
if os.path.exists(file_path):
|
| 162 |
+
return FileResponse(file_path, media_type="video/mp4", filename=f"bumiz_fb_{id}", background=BackgroundTask(cleanup, file_path))
|
| 163 |
+
raise HTTPException(status_code=404)
|