Manage a collection of YouTube video archives
In order to use Accord's Library - YouTube-Archiver, you will need to download a few executables from other projects:
- https://github.com/ytdl-org/youtube-dl/releases
- https://github.com/wez/atomicparsley/releases
- https://ffmpeg.org/download.html
You'll also need to create a cookies.txt file that will provide credentials for youtube-dl to use. This is optional, but it is the only way to archive aged-restricted videos. To generate this cookies.txt file, you can use the extension called https://chrome.google.com/webstore/detail/get-cookiestxt/bgaddhkoddajcdgocldbbfleckgcbcid by Rahul Shaw. Simply log in YouTube and click on the extension icon to extract the cookies.
All the executables and the cookies.txt file needs to be included at the root of the repository.
- Simply write down video UIDs in the
video-append-uid.txtfile (i.e. dQw4w9WgXcQ). One UID by line, do not include the rest of the URL. - Run
script-list.py - The output video is located (by default) in
output/[UID]
The resulting archive is composed as follows:
- The video, in the highest quality possible for the audio and video (limited to codex compatible with the MP4 container format).
- Highest quality thumbnail, saved as JPG or WebP
- A
.descriptionfile containing the original description of the video - A
.info.jsonfile containing most if not all metadata publicly available about the video (duration, uploader, channel_id, tags, available qualities...) - A
.vttfile for each subtitle language available.
In addition to the archive, a collection.csv is generated. In includes the list of archived videos currently present in the output folder. For each archive, a few information is included: the uploader, the video title, its publication date, its url, and its duration.