https://github.com/yashindane/web-sudoku

:white_square_button: A flask webapp to solve sudoku from image input
https://github.com/yashindane/web-sudoku

aws-textract flask

Last synced: 1 day ago
JSON representation

:white_square_button: A flask webapp to solve sudoku from image input

• Host: GitHub
• URL: https://github.com/yashindane/web-sudoku
• Owner: YashIndane
• License: mit
• Created: 2021-05-24T10:42:50.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2022-12-12T09:46:05.000Z (almost 2 years ago)
• Last Synced: 2023-03-05T07:43:14.717Z (over 1 year ago)
• Topics: aws-textract, flask
• Language: CSS
• Homepage:
• Size: 816 KB
• Stars: 14
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
![](https://img.shields.io/badge/python-3-orange?style=flat-square)

# web-sudoku

A flask webapp to solve sudoku from image input

![solved1](https://user-images.githubusercontent.com/53041219/207013634-3a80d9ab-0e3d-43d5-a4da-628be5626c9a.png)

[Demo link](https://www.linkedin.com/posts/yash-indane-aa6534179_aws-flask-python-activity-6803238011279548416-jjeQ)

## AWS CLI configuration

Have user ready in AWS with admin and power user access. This user will be used to call the AWS Textract service.

Create a S3 bucket in this users account.

# Usage

```

$ sudo docker run -dit -p :1453  --name  yashindane/websudoku:v1 --aak="" --ask="" --region="" --bucketname=""

```

# On arm64 v8 machines

```

$ sudo docker run --platform linux/arm64/v8 -dit -p :1453  --name  yashindane/websudoku:arm64v8 --aak="" --ask="" --region="" --bucketname=""

```

# On arm64 v8 machines using podman

```

$ sudo podman run --platform linux/arm64/v8 -dit -p :1453  --name  docker.io/yashindane/websudoku:arm64v8 --aak="" --ask="" --region="" --bucketname=""

```

## How it works

The user first clicks a pic of there sudoku board and submits it. The `data_uri` of the image goes to backend `app2.py` script through the form on user click.

Using `base64`, `ìo` and `PIL` libraries, the image is saved in the server and then uploaded to `S3` bucket. Now using `AWS Textract`, the digits from the image in bucket are extracted and send back to the `app2.py`. This digits in a double array are fed to the function `sudoku3.solve()`, which returns the solution to the problem. This diits are then rendered on the `output.html` page.

Pull requests and new Ideas are welcome!