Scripts for creating a database to organize the learning video I've watched.
These scripts requires latest youtube-dl.py
from http://youtube-dl.org/.
The first version (create-videodb.py
) was writted in Python, but my Python-fu is elementary and the schema design was a blunder.
The current version is writted in Node and migration scripts are available.
Terminology:
- source (extractor, video_id) tuple, can be mapped to the clip URL.
.info.json
Description for a video clip created byyoutube-dl.py
.
It's schema changed so2-update-infojson
was to update existing.info.json
.videodb
(.json
/videodb.json
) videodb created bycreate-videodb.py
.
Uses source as key for lookup (which is not orthogonal).dbjson
(.db.json
) videodb created by3-create-dbjson
.
- download info.json for all videos in videodb
- update existing info.json
- convert info.json to dbjson
- migrate tag and rating from videodb to dbjson
corresponding fields in info.json in brackets
{
[_id]
video_id (id)
title
extractor
duration
url (webpage_url)
uploader (uploader_id)
upload_date
thumbnail
tags
rating
}
- {presentor/organiztion}
- [playlist/series/conference]
dbjson
can be queryed by scripts or json or JSONSelect