Project Name: SodaTracker #Stella-Sora-Pull-History

SodaTracker is a Stella Sora pull history tracker that uses OCR to convert screenshots into a searchable JSON database.
You can say: "Kinda make things difficult application to track your gacha pull history on Stella Sora Game."
at least for now.

---

🚀 Features

• Feature 1: Automatic screenshot cropping to make your files ready to ocread.
• Feature 2: Convert your cropped image to json string.
• Feature 3: Be Creative Myself. Go find yourself a method to serve your fresh json pull string.

🛠️ Installation (windows)

To get this project running locally, follow these steps:

1. Clone the repository:

   git clone [https://github.com/Retoi-II/SodaTracker.git](https://github.com/Retoi-II/SodaTracker.git)
   
   

2. Install Python (if you haven't or you don't know how to host the data you captured)

   For reference, i am using 3.12.0 version, so it's 3.12.0++ version :)

3. Install Package Manager

   I'd prefer chocolatey as my package manager, you can use whatever you have and perform the installation with equivalent command.

4. Install Magick and Tesseract-OCR

   choco install imagemagick -y
   choco install tesseract-ocr -y
   
   

5. Collect the dependency in requirements.txt and library

   py -m pip install -r requirements.txt
   py -m pip install pillow
   py -m pip install flask
   py -m pip install pytesseract

   Flask is just a lightweight web server, it's just one of the countless framework to serve json data as table or other method, stick with this if you don't know what to do anyway.

6. How do you actually collect your pulls history

   Pay attention to this

   • Make sure you only make batch screenshot on one type of banner, don't mix multiple banner tipe in one batch.
   • Start your screenshot from the first page as starting point up to last page as end point, you can choose whatever range you want or even starting point itself.
   • To make things easier make sure you have snipping tools (if you are using windows) which will saves your screenshot automatically.

   The capture step x step

   • Make sure your game on 1280x720 windowed instance.
   • Press Win+Shift+S > choose "Window Mode" > point Stella Sora instance as capture area.
   • Do it repeatedly until the point you wish.
   • Copy all the captured image to .\preprocess\ocr\screenshots
   • Execute the crop.bat
   • Open the .preprocess\ocr\screenshot\cropped folder and move all the cropped to above directory, replace the uncropped images.
   • Navigate to .preprocess\ocr directory and run the python file py read_gacha_to_json.py
   • Check the folder and you will find gacha_history.json

   Footnote

   • It doesn't give you "ItemType" and "Star/Rarity" line yet. You can ask AI to do that, find the prompt on .\preprocess folder.
   • Do whatever you want with the collected data, as advise you can just copy and pasting the content to flask template i have shared on the root folder.
   • You can customize the content served by edit the index.html on .\template directory

7. Load your captured json with flask app

   I said you can use my flask template, but there are countless of possibility. You can use whatever method you know to serve a data yourself. In case you decided to just using my flask template this is the command line.

   py -m flask --app flask_template.py run -p 5000
   

Future plan with this repository

I have plan to extend the feature of this project in the future, but i have no time or just lazy for now. If you interested to develop this application, then feel free to contribute yourself.