Currently to download file i'm using
from telethon import TelegramClient
client = TelegramClient('anon', api_id, api_hash)
async def main():
async for dialog in client.iter_messages(entity=peer_channel):
await dialog.download_media("file....")
def bot():
with client:
client.loop.run_until_complete(main())
client.start()
client.run_until_disconnected()
if __name__ == "__main__":
bot()
But sometimes I lost connection, because of my net or telegram disconnection.Or maybe because I need to restart... Here a log.
INFO:root:Download total: 99% 1048.50 mb/1057.82 mb tmp/Evil/2x12 Evil.rar
INFO:telethon.network.mtprotosender:Disconnecting from xxx.xxx.xxx.xxx:443/TcpFull...
INFO:telethon.network.mtprotosender:Disconnection from xxx.xxx.xxx.xxx:443/TcpFull complete!
INFO:telethon.network.mtprotosender:Connecting to xxx.xxx.xxx.xxx:443/TcpFull...
INFO:telethon.network.mtprotosender:Connection to xxx.xxx.xxx.xxx:443/TcpFull complete!
INFO:root:[DEBUG] DOWNLOADING /export/nasty/tmp/Evil/2x12 Evil.rar
INFO:telethon.client.downloads:Starting direct file download in chunks of 524288 at 0, stride 524288
"INFO:root" are messages writed by me using logging.info(....)
It's disgusting... file was at 99% but connection failed and must to restart download from zero.
Reading documentation I found this client.iter_download I tried:
async def main():
async for dialog in client.iter_messages(entity=peer_channel):
# Filename should be generated by dialog.media, is a example
with open('file.rar', 'wb') as fd:
async for chunk in client.iter_download(dialog.media):
fd.write(chunk)
But same result if I stop script, download start from zero
iter_download is the correct way, however, you need to manually specify the resume offset (and you should open the file in append mode):
import os
file = 'file.rar'
try:
offset = os.path.getsize(file)
except OSError:
offset = 0
with open(file, 'ab') as fd:
# ^ append
async for chunk in client.iter_download(dialog.media, offset=offset):
# ^~~~~~~~~~~~~ resume from offset
fd.write(chunk)