2024-01-04 16:56:37 +01:00

315 lines
10 KiB
Python

# 4.08.2023 -> 14.09.2023 -> 17.09.2023 -> 3.12.2023
# Import
import re, os, sys, glob, time, requests, shutil, ffmpeg, subprocess
from functools import partial
from multiprocessing.dummy import Pool
from tqdm.rich import tqdm
import moviepy.editor as mp
from Stream.util.platform import *
# Class import
#from Stream.util.console import console
from Stream.util.console import console
# Disable warning
import warnings
from tqdm import TqdmExperimentalWarning
warnings.filterwarnings("ignore", category=TqdmExperimentalWarning)
# Variable
main_out_folder = "videos"
os.makedirs("videos", exist_ok=True)
# [ decoder ] -> costant = ou_ts
class Video_Decoder(object):
iv = ""
uri = ""
method = ""
def __init__(self, x_key, uri):
self.method = x_key["METHOD"] if "METHOD" in x_key.keys() else ""
self.uri = uri
self.iv = x_key["IV"].lstrip("0x") if "IV" in x_key.keys() else ""
def decode_aes_128(self, video_fname: str):
if is_platform_linux():
frame_name = video_fname.split("/")[-1].split("-")[0] + ".ts"
else:
frame_name = video_fname.split("\\")[-1].split("-")[0] + ".ts"
res_cmd = subprocess.run(["openssl","aes-128-cbc","-d","-in", video_fname,"-out", "ou_ts/"+frame_name,"-nosalt","-iv", self.iv,"-K", self.uri ], capture_output=True)
res_cmd_str = res_cmd.stderr.decode("utf-8")
res_cmd_fix = res_cmd_str.replace("b", "").replace("\n", "").replace("\r", "")
if "lengthad" in res_cmd_fix:
console.log("[red]Wrong m3u8 key or remove key from input !!!")
sys.exit(0)
def decode_ext_x_key(key_str: str):
key_str = key_str.replace('"', '').lstrip("#EXT-X-KEY:")
v_list = re.findall(r"[^,=]+", key_str)
key_map = {v_list[i]: v_list[i+1] for i in range(0, len(v_list), 2)}
return key_map
# [ util ]
def save_in_part(folder_ts, merged_mp4, file_extension = ".ts"):
# Get list of ts file in order
os.chdir(folder_ts)
# Order all ts file
try: ordered_ts_names = sorted(glob.glob(f"*{file_extension}"), key=lambda x:float(re.findall("(\d+)", x.split("_")[1])[0]))
except:
try: ordered_ts_names = sorted(glob.glob(f"*{file_extension}"), key=lambda x:float(re.findall("(\d+)", x.split("-")[1])[0]))
except: ordered_ts_names = sorted(glob.glob(f"*{file_extension}"))
open("concat.txt", "wb")
open("part_list.txt", "wb")
# Variable for download
list_mp4_part = []
part = 0
start = 0
end = 200
# Create mp4 from start ts to end
def save_part_ts(start, end, part):
#console.log(f"[blue]Process part [green][[red]{part}[green]]")
list_mp4_part.append(f"{part}.mp4")
with open(f"{part}_concat.txt", "w") as f:
for i in range(start, end):
f.write(f"file {ordered_ts_names[i]} \n")
ffmpeg.input(f"{part}_concat.txt", format='concat', safe=0).output(f"{part}.mp4", c='copy', loglevel="quiet").run()
# Save first part
save_part_ts(start, end, part)
# Save all other part
for _ in range(start, end):
# Increment progress ts file
start+= 200
end += 200
part+=1
# Check if end or not
if(end < len(ordered_ts_names)):
save_part_ts(start, end, part)
else:
save_part_ts(start, len(ordered_ts_names), part)
break
# Merge all part
console.log(f"[purple]Merge all: {file_extension} file")
with open("part_list.txt", 'w') as f:
for mp4_fname in list_mp4_part:
f.write(f"file {mp4_fname}\n")
ffmpeg.input("part_list.txt", format='concat', safe=0).output(merged_mp4, c='copy', loglevel="quiet").run()
def download_ts_file(ts_url: str, store_dir: str, headers):
# Get ts name and folder
ts_name = ts_url.split('/')[-1].split("?")[0]
ts_dir = store_dir + "/" + ts_name
# Check if exist
if(not os.path.isfile(ts_dir)):
# Download
ts_res = requests.get(ts_url, headers=headers)
if(ts_res.status_code == 200):
with open(ts_dir, 'wb+') as f:
f.write(ts_res.content)
else:
print(f"Failed to download streaming file: {ts_name}.")
time.sleep(0.5)
def download_vvt_sub(content, language, folder_id):
# Get content of vvt
url_main_sub = ""
vvt = content.split("\n")
# Find main url or vvt
for i in range(len(vvt)):
line = str(vvt[i])
if line.startswith("#EXTINF"):
url_main_sub = vvt[i+1]
# Save subtitle to main folder out
path = os.path.join(main_out_folder, str(folder_id))
os.makedirs(path, exist_ok=True)
open(os.path.join(path, "sub_"+str(language)+".vtt"), "wb").write(requests.get(url_main_sub).content)
# [ donwload ]
def dw_m3u8(m3u8_link, m3u8_content, m3u8_headers="", decrypt_key="", merged_mp4="test.mp4"):
# Reading the m3u8 file
m3u8_http_base = m3u8_link.rstrip(m3u8_link.split("/")[-1])
m3u8 = m3u8_content.split('\n')
ts_url_list = []
ts_names = []
x_key_dict = dict()
is_encryped = False
# Parsing the content in m3u8 with creation of url_list with url of ts file
for i_str in range(len(m3u8)):
line_str = m3u8[i_str]
if "AES-128" in str(line_str):
is_encryped = True
if line_str.startswith("#EXT-X-KEY:"):
x_key_dict = decode_ext_x_key(line_str)
if line_str.startswith("#EXTINF"):
ts_url = m3u8[i_str+1]
ts_names.append(ts_url.split('/')[-1])
if not ts_url.startswith("http"):
ts_url = m3u8_http_base + ts_url
ts_url_list.append(ts_url)
#console.log(f"[blue]Find [white]=> [red]{len(ts_url_list)}[blue] ts file to download")
#console.log(f"[green]Is m3u8 encryped => [red]{is_encryped}")
if is_encryped and decrypt_key == "":
console.log(f"[red]M3U8 Is encryped")
sys.exit(0)
if is_encryped:
#console.log(f"[blue]Use decrypting")
video_decoder = Video_Decoder(x_key=x_key_dict, uri=decrypt_key)
os.makedirs("ou_ts", exist_ok=True)
# Using multithreading to download all ts file
os.makedirs("temp_ts", exist_ok=True)
pool = Pool(15)
gen = pool.imap(partial(download_ts_file, store_dir="temp_ts", headers=m3u8_headers), ts_url_list)
for _ in tqdm(gen, total=len(ts_url_list), unit="bytes", unit_scale=True, unit_divisor=1024, desc="[yellow]Download"):
pass
pool.close()
pool.join()
if is_encryped:
for ts_fname in tqdm(glob.glob("temp_ts/*.ts"), desc="[yellow]Decoding"):
video_decoder.decode_aes_128(ts_fname)
# Start to merge all *.ts files
save_in_part("ou_ts", merged_mp4)
else:
save_in_part("temp_ts", merged_mp4)
# Clean temp file
os.chdir("..")
console.log("[green]Clean")
if is_platform_linux():
if is_encryped:
shutil.move("ou_ts/"+merged_mp4 , main_out_folder+"/")
else:
shutil.move("temp_ts/"+merged_mp4 , main_out_folder+"/")
else:
if is_encryped:
shutil.move("ou_ts\\"+merged_mp4 , main_out_folder+"\\")
else:
shutil.move("temp_ts\\"+merged_mp4 , main_out_folder+"\\")
shutil.rmtree("ou_ts", ignore_errors=True)
shutil.rmtree("temp_ts", ignore_errors=True)
def dw_aac(m3u8_link, m3u8_content, m3u8_headers, merged_mp3):
# Reading the m3u8 file
url_base = m3u8_link.rstrip(m3u8_link.split("/")[-1])
m3u8 = m3u8_content.split('\n')
ts_url_list = []
ts_names = []
# Parsing the content in m3u8 with creation of url_list with url of ts file
for i in range(len(m3u8)):
line = m3u8[i]
if line.startswith("#EXTINF"):
ts_url = m3u8[i+1]
ts_names.append(ts_url.split('/')[-1])
if not ts_url.startswith("http"):
ts_url = url_base + ts_url
ts_url_list.append(ts_url)
console.log(f"[blue]Find [white]=> [red]{len(ts_url_list)}[blue] ts file to download")
# Using multithreading to download all ts file
os.makedirs("temp_ts", exist_ok=True)
pool = Pool(15)
gen = pool.imap(partial(download_ts_file, store_dir="temp_ts", headers=m3u8_headers), ts_url_list)
for _ in tqdm(gen, total=len(ts_url_list), unit="bytes", unit_scale=True, unit_divisor=1024, desc="[yellow]Download"):
pass
pool.close()
pool.join()
save_in_part("temp_ts", merged_mp3, file_extension=".aac")
# Clean temp file
os.chdir("..")
console.log("[green]Clean")
shutil.move("temp_ts\\"+merged_mp3 , ".")
shutil.rmtree("ou_ts", ignore_errors=True)
shutil.rmtree("temp_ts", ignore_errors=True)
def dw_vvt_sub(url, headers, folder_id) -> (None):
print(url, headers, folder_id)
# Get content of m3u8 vvt
req = requests.get(url, headers=headers)
vvts = req.text.split('\n')
vvt_data = []
# Parsing the content in m3u8 of vvt with creation of url_list with url and name of language
for line in vvts:
line = line.split(",")
if line[0] == "#EXT-X-MEDIA:TYPE=SUBTITLES":
vvt_data.append({
'language': line[2].split("=")[1].replace('"', ""),
'url': line[-1].split("URI=")[1].replace('"', "")
})
# Check array is not empty
if len(vvt_data) > 0:
# Download all subtitle
for i in range(len(vvts)):
console.log(f"[blue]Download [red]sub => [green]{vvt_data[i]['language']}")
download_vvt_sub(requests.get(vvt_data[i]['url']).text, vvt_data[i]['language'], folder_id)
else:
console.log("[red]Cant find info of subtitle [SKIP]")
def join_audio_to_video(audio_path, video_path, out_path):
audio = mp.AudioFileClip(audio_path)
video1 = mp.VideoFileClip(video_path)
final = video1.set_audio(audio)
final.write_videofile(out_path)