From 06a1f8c3a44f97d7196faea7511ba0693f987476 Mon Sep 17 00:00:00 2001
From: JustAnotherArchivist <JustAnotherArchivist@users.noreply.github.com>
Date: Tue, 17 May 2022 04:03:31 +0000
Subject: [PATCH] Initial commit

---
 Dockerfile     |  8 +++++++
 README.md      | 12 +++++++++++
 telegram-dl.py | 57 ++++++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 77 insertions(+)
 create mode 100644 Dockerfile
 create mode 100644 README.md
 create mode 100644 telegram-dl.py

diff --git a/Dockerfile b/Dockerfile
new file mode 100644
index 0000000..32789e5
--- /dev/null
+++ b/Dockerfile
@@ -0,0 +1,8 @@
+FROM python:3.7-alpine3.14
+RUN \
+	apk add --no-cache gcc libc-dev libffi-dev \
+	&& pip install --no-cache-dir telethon 'cryptg<0.3'
+COPY telegram-dl.py /
+VOLUME ["/data/"]
+WORKDIR /data
+ENTRYPOINT ["python3", "/telegram-dl.py"]
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..1002687
--- /dev/null
+++ b/README.md
@@ -0,0 +1,12 @@
+telegram-dl is a little script for retrieving content on Telegram.
+
+Usage: `telegram-dl.py URL [URL...]`
+
+All URLs must be messages in the same channel. The output consists of writing each message to a file `{channelName}_{messageId}.json`. If the message has any attachment (photo, video, file, etc.), it gets downloaded as well. The filename for that is chosen by Telethon; if it's a file attachment, the literal path as presented by Telegram is used, and if it's a photo or video, a name is made up.
+
+The only dependency of the script is Telethon (though you probably want cryptg for performance reasons). For authentication (always required), specify the `TELEGRAM_API_ID`, `TELEGRAM_API_HASH`, and `TELEGRAM_BOT_TOKEN` environment variables. The session is stored in a file `telegram-dl.session`; keeping this between executions is not critical but recommended.
+
+A `Dockerfile` is provided for running with Docker. Simply provide the URL(s) as arguments on running the image. Data is written to the `/data` volume, which you may want to mount from the host machine instead.
+
+    docker build -t telegram-dl:latest .
+    docker run --rm -v "$(pwd)"/data:/data -e TELEGRAM_API_ID=... -e TELEGRAM_API_HASH=... -e TELEGRAM_BOT_TOKEN=... telegram-dl https://t.me/telegram/178
diff --git a/telegram-dl.py b/telegram-dl.py
new file mode 100644
index 0000000..30bfe52
--- /dev/null
+++ b/telegram-dl.py
@@ -0,0 +1,57 @@
+#!/usr/bin/env python3
+import asyncio
+import base64
+import datetime
+import json
+import os
+import re
+import sys
+import telethon
+
+
+API_ID = os.environ['TELEGRAM_API_ID']
+API_HASH = os.environ['TELEGRAM_API_HASH']
+BOT_TOKEN = os.environ['TELEGRAM_BOT_TOKEN']
+URL_PATTERN = re.compile(r'^https?://t\.me/(?:s/)?(?P<channel>[^/]+)/(?P<message>\d+)$')
+
+
+def stuff_to_json(o):
+	if isinstance(o, datetime.datetime):
+		return o.isoformat()
+	if isinstance(o, bytes):
+		return f'binary data: {base64.b64encode(o).decode("ascii")}'
+	raise TypeError(f'Object of type {type(o)} is not JSON serializable')
+
+
+async def main():
+	# Parse URLs
+	targets = []
+	for url in sys.argv[1:]:
+		m = URL_PATTERN.match(url)
+		if not m:
+			print(f'Error: {url} is not a recognised Telegram URL', file = sys.stderr)
+			sys.exit(1)
+		targets.append((m['channel'], int(m['message'])))
+	if not targets:
+		print(f'Usage: telegram-dl.py URL [URL...]', file = sys.stderr)
+		sys.exit(1)
+	channelName = targets[0][0]
+	if not all(x[0] == channelName for x in targets[1:]):
+		print(f'Error: all URLs must be of the same channel', file = sys.stderr)
+		sys.exit(1)
+	ids = [x[1] for x in targets]
+
+	# Let's go...
+	client = telethon.TelegramClient('telegram-dl', API_ID, API_HASH)
+	await client.start(bot_token = BOT_TOKEN)
+	messages = await client.get_messages(channelName, ids = ids)
+	for message in messages:
+		if not message:
+			continue
+		with open(f'{channelName}_{message.id}.json', 'x') as fp:
+			json.dump(message.to_dict(), fp, default = stuff_to_json)
+
+		await client.download_media(message)
+
+
+asyncio.run(main())