Skip to content

tattle-made/dau-dashboard

Repository files navigation

DAU Dashboard

Deepfake Analysis Unit(DAU) is a collaborative space for analyzing deepfakes

Developing Locally

Prerequisites

  1. docker : 24.0.5
  2. aws cli : 2.2.25
  3. jq : 1.6
  4. elixir : 1.15.0
  5. Erlang/OTP : 26.0

1. Setup infrastructure dependencies**

  • Run source env.dev to setup enviroment variables.
  • docker-compose up Visit localhost:8080 to access adminer. Login with following fields
System PostgreSQL
Server db
Username tattle
Password weak_password

2. Fetch Credentials

Credentials are stored using AWS Secrets Manager. Please request @dennyabrain for SECRET_ARN and credentials for a suitable AWS IAM account

aws secretsmanager get-secret-value --secret-id $SECRET_ARN | jq '.SecretString' | jq -r | jq -r 'to_entries[]' | jq -r '"export " + .key + "=" + .value' > env.dev

3. Setup and Start your Phoenix server

  1. Load environment variables : source env.dev
  2. Install frontend Dependencies : cd assets && npm install
  3. To start your Phoenix server, first Run mix setup to install and setup dependencies.
  4. Start Phoenix endpoint with mix phx.server or inside IEx with iex -S mix phx.server

Now you can visit localhost:4000 from your browser.

Viewing Documentation

Run mix docs. This generates the documentation and puts the .html and .epub file in doc/. Run http-server doc/ and visit localhost:80801

Media during development

Copy Paste files in /priv/static/assets/media. These will be available to the app at http://localhost:4000/assets/media/${filename.extention}

Test media files

Since this app relies on media files so heavily, we have added some test files in the project specific s3 bucket. These are available with the prefix temp. Some example files are temp/audio-01.wav and temp/video-01.mp4. These are used throughout unit tests and integration tests. There are 10 audio files and 50 video files.

Seeding Data

  1. Upload test files to s3 : aws s3 sync . s3://staging.dau.tattle.co.in/temp
  2. Seed database : mix run priv/repo/seeds.exs

About

a collaborative space for analyzing deepfakes

Resources

Stars

Watchers

Forks

Packages

No packages published

Contributors 3

  •  
  •  
  •