initial commit

.gitignore

@@ -0,0 +1,3 @@
+__pycache__
+venv/
+**/*.mp4

README.md

@@ -0,0 +1,14 @@
+# FAU-TV Video Downloader
+
+Downloads all videos of a given course (even if the download has been restricted) to a local folder.
+This software is provided without warranty. Usage is discouraged! Usage of this software is at your own risk!
+
+## Usage
+
+```bash
+./dl '<fau course id>' '[output directory]' '[starter url]'
+```
+
+# Procedure
+
+Upon starting, a firefox instance is opened, prompting you to log in using IDM SSO. As soon as the url changes to something starting with `https://www.fau.tv`, download will be starting. If you do not want to provide your credentials, you can provide `https://www.fau.tv` as a starter url.

dl

@@ -0,0 +1,21 @@
+from .lib import *
+import sys
+import os
+
+
+def main():
+    course_id = sys.argv[1]
+    out_dir = sys.argv[2] if len(sys.argv) >= 3 else './out'
+    auth_url = sys.argv[3] if len(sys.argv) >= 4 else None
+
+    load_token(auth_url) if auth_url is not None else load_token()
+
+    os.makedirs(out_dir, exist_ok=True)
+
+    for index, clip_id in enumerate(get_course_clip_ids(course_id)):
+        print(f'downloading clip {clip_id}')
+        download_clip(clip_id, f'{out_dir}/{index+1: 04d}_{clip_id}.mp4')
+
+
+if __name__ == '__main__':
+    main()

lib.py

@@ -0,0 +1,154 @@
+from selenium import webdriver
+import time
+from dataclasses import dataclass
+import requests
+import re
+import subprocess
+import shutil
+
+
+_token: "Token" = None
+
+
+@dataclass
+class Token():
+    auth_token: str
+    session_id: str
+    session_ci: str
+
+    def cookies(self):
+        return {
+            "SimpleSAMLAuthToken": self.auth_token,
+            "SimpleSAMLSessionID": self.session_id,
+            "session_ci": self.session_ci,
+        }
+
+
+@dataclass
+class ClipDetails():
+    combined_media_id: str = None
+    combined_playlist_url: str = None
+    camera_media_id: str = None
+    camera_playlist_url: str = None
+    slides_media_id: str = None
+    slides_playlist_url: str = None
+
+
+def load_token(auth_url: str = "https://www.fau.tv/auth/sso"):
+    global _token
+
+    driver = webdriver.Firefox()
+
+    driver.get(auth_url)
+
+    while not driver.current_url.startswith('https://www.fau.tv/'):
+        time.sleep(0.5)
+
+    def get_value(cookie):
+        if cookie is None:
+            return None
+        return cookie.get("value")
+
+    _token = Token(
+        auth_token=get_value(driver.get_cookie("SimpleSAMLAuthToken")),
+        session_id=get_value(driver.get_cookie("SimpleSAMLSessionID")),
+        session_ci=get_value(driver.get_cookie("session_ci")),
+    )
+
+    print(_token)
+
+    driver.close()
+
+
+def get_course_clip_ids(course_id: str) -> list[str]:
+    global _token
+
+    regex = re.compile(r'(/clip/id/)([0-9]+)(\"\s*class=\"preview\")')
+
+    urls = []
+
+    url = f'https://www.fau.tv/course/id/{course_id}'
+    with requests.get(url, cookies=_token.cookies()) as r:
+        clip_matches = regex.findall(r.text)
+        for match in clip_matches:
+            urls.append(match[1])
+
+    return urls
+
+
+def get_clip_details(clip_id: str) -> ClipDetails:
+    global _token
+
+    url = f'https://www.fau.tv/clip/id/{clip_id}'
+    details = ClipDetails()
+
+    with requests.get(url, cookies=_token.cookies()) as r:
+        def get_details(keyword: str):
+            mediaid_re = re.compile(
+                r'(' + keyword + r'Sources[^,]*,\s+mediaid\:\s+\")([0-9]+)'
+            )
+
+            playlist_url_re = re.compile(
+                r'(file\:\s+\")([^\"]*' + keyword + r'\.smil[^\"]*)(\")'
+            )
+
+            mediaid_matches = mediaid_re.findall(r.text)
+            if len(mediaid_matches) > 0:
+                setattr(details, keyword + "_media_id", mediaid_matches[0][1])
+
+            playlist_url_matches = playlist_url_re.findall(r.text)
+            if len(playlist_url_matches) > 0:
+                setattr(details, keyword + "_playlist_url",
+                        playlist_url_matches[0][1])
+
+        get_details("combined")
+        get_details("camera")
+        get_details("slides")
+
+    return details
+
+
+def download_media(media_id: str, outfile_path: str):
+    global _token
+
+    url = f'https://itunes.video.uni-erlangen.de/get/file/' + \
+        str(media_id) + '?download=1'
+
+    with requests.get(url, stream=True, cookies=_token.cookies()) as r:
+        if (r.status_code != 200):
+            return False
+
+        with open(outfile_path, 'wb') as f:
+            shutil.copyfileobj(r.raw, f)
+
+    return True
+
+
+def download_playlist(playlist_url: str, outfile_path: str):
+    subprocess.run([
+        'ffmpeg',
+        '-i', playlist_url,
+        '-c', 'copy',
+        outfile_path,
+    ])
+
+
+def download_clip(clip_id: str, outfile_path: str):
+    details = get_clip_details(clip_id)
+
+    media_id = next(id for id in [
+        details.combined_media_id,
+        details.camera_media_id,
+        details.slides_media_id,
+    ] if id is not None)
+
+    if download_media(media_id, outfile_path):
+        return
+
+    url = next(url for url in [
+        details.combined_playlist_url,
+        details.camera_playlist_url,
+        details.slides_playlist_url,
+    ] if url is not None)
+
+    download_playlist(url, outfile_path)

pyvenv.cfg

@@ -0,0 +1,5 @@
+home = /usr/bin
+include-system-site-packages = false
+version = 3.12.4
+executable = /usr/bin/python3.12
+command = /usr/bin/python -m venv /home/ludwig/git/fau-tv-dl

requirements.txt

@@ -0,0 +1,2 @@
+selenium
+requests