# Nijitrack

Historical subscriber count tracker for any subset of YouTube channels. Flask backend and NextJS frontend for visualization.

This code is current deployed as [PhaseTracker](https://phase-tracker.com). Collecting the subscriber data for members of Phase Connect.

# Backend:
The code in this repo is the frontend. The backend API is available as a submodule. You can find the repo [here](https://github.com/pinapelz/NijiTrack-API)

Below are the steps to set up the flask backend for data collection and serving data through a web server (`/backend` folder of repo)

### Dependencies
- PostgresSQL
- Python 3.11+
- Flask 2.1.2+ (Optional)

```python
pip install -r requirements.txt
```
### Usage
a. Add the environment variables in the `.env.template`
  - B2API fields are uncessary if you are not auto upload to Backblaze B2
  - YouTube API not necessary unless you plan on tracking a non-Holodex subset of channels

b. Specify trace colors in `member_color.py` (based on Channel Name)
#### For collection using a set of channels belonging to a Virtual YouTuber organization listed on Holodex
1. Edit main.py and edit `HOLODEX_ORG` to the organization name on Holodex and `ORG_MEMBER_COUNT` to the number of members that organization has
  - Overshooting the member count may lead to additional loop iterations, but in general there will be no problems
```
HOLODEX_ORG = "Phase%20Connect"
ORG_MEMBER_COUNT = 75
```
2. Add channels to be excluded from data collection in `data/excluded_channel.txt`
3. Execute with `python nijitrack.py`
#### For collection using a custom set of channels configured in `data/channels.txt`
1. Configure channels in data/channels.txt in the format `YOUTUBE_CHANNEL_ID,CHANNEL_NAME` for each line
2. Execute with `python nijitrack.py ytapi`

A basic graph will be created using plotly and saved to index.html in the root directory

You can use `app.py` to host your statistics through Flask or use any other services


Webpage design inspired by [TrackHololive](https://trackholo.live/)