Data Engineering

gamechanger-data focuses on the data engineering work of gamechanger. To see all repositories gamechanger

Important Note!

• Configuration of repo is reliant on being able to hit advana-data-zone's s3 bucket. If you do not have access to advana-data-zone's s3 bucket, you will need to fill in your own values in config script; like topic_models (for ML features) and configure_app (ElasticSearch, Postgres, and Neo4j)
• Once venv is set up, set DEPLOYMENT_ENV variable and run ./paasJobs/configure_repo.sh or paasJobs/configure_repo.bat
  Example DEPLOYMENT_ENV=local ./paasJobs/configure_repo.sh or set DEPLOYMENT_ENV=local \paasJobs\configure_repo.bat

(Linux) Dev/Prod Deployment Instructions

• Clone fresh gamechanger-data repo
• Setup python3.8 venv with packages in requirements.txt.
  • Create python3.8 venv, e.g. python3 -m venv /opt/gc-venv-20210613
  • Before installing packages, update pip/wheel/setuptools, e.g. <venv>/bin/pip install --upgrade pip setuptools wheel
  • Install packages from requirements.txt, with no additional dependencies, e.g. <venv>/bin/pip install --no-deps -r requirements.txt
• Set up symlink /opt/gc-venv-current to the freshly created venv, e.g. ln -s /opt/gc-venv-20210613 /opt/gc-venv-current
• Pull in other dependencies and configure repo with env SCRIPT_ENV=<prod|dev> <repo>/paasJobs/configure_repo.sh
  • Config script will let you know if everything was configured correctly and if all backends can be reached.

How to Setup Local Env for Development

MacOS / Linux

• (Linux Only) Follow instruction appropriate to repo to install ocrmypdf and its dependencies: https://ocrmypdf.readthedocs.io/en/latest/installation.html#installing-on-linux
• (MacOS Only) Install "brew" then use it to install tesseract brew install tesseract-lang
• Install Miniconda or Anaconda (Miniconda is much smaller)
  • https://docs.conda.io/en/latest/miniconda.html
• Create gamechanger python3.8 environment, like so:
  • conda create -n gc python=3.8
• Clone the repo and change into that dir git clone ...; cd gamechanger
• Activate conda environment and install requirements:
  • :bangbang: reeeealy important - make sure you change into repo directory
  • conda activate gc
  • pip install --upgrade pip setuptools wheel
  • pip install -e '.[dev]' (quoting around .[dev] is important)
• That's it.

Windows (WSL Version)

• Setup Windows Subsystem for Linux (WSL) environment
  • https://docs.microsoft.com/en-us/windows/wsl/install-win10
• (In WSL)
  • Install ocrmypdf dependencies following ubuntu instructions here: https://ocrmypdf.readthedocs.io/en/latest/installation.html#installing-on-linux
  • Install Miniconda or Anaconda (Miniconda is much smaller)
    • https://docs.conda.io/en/latest/miniconda.html
  • Create gamechanger python3.8 environment, like so:
    • conda create -n gc python=3.8
  • Clone the repo and change into that dir git clone ...; cd gamechanger-data
  • Activate conda environment and install requirements:
    • :bangbang: reeeealy important - make sure you change into repo directory
    • conda activate gc
    • pip install --upgrade pip setuptools wheel
    • pip install -e '.[dev]' (quoting around .[dev] is important)
  • That's it, just activate that conda env if you want to use it inside the terminal.

Windows

Create venv python -m venv [venv-name] Activate \[venv-name]\Scripts\activate Update venv python -m pip install --upgrade pip setuptools wheel Install requirements.txt pip install --no-deps -r dev_tools\requirements\gc-venv-current.txt

Run Configure Repo, Steps at the top of this README

To-Do:

• convert .sh scripts to .bat to support window users

Docker

docker build -t gc-data --no-cache .
docker rm -f gc-data-test || true
docker run -it --name gc-data gc-data

Configure Repo

IDE SETUP

How to Setup PyCharm IDE

Note: If you're using containerized env, you'll need Pro version of PyCharm and separate set of instructions - here

• Create new project by opening directory where you cloned the repository. PyCharm will tell you that it sees existing repo there, just accept that and proceed.
• With your gc conda environment all good to go, change your "Preferences -> Project -> Python Interpreter" to the EXISTING gc conda env you created. https://www.jetbrains.com/help/pycharm/conda-support-creating-conda-virtual-environment.html
• Now, change your "Preferences -> Build, Execution, Deployment -> Console -> Python Console interpreter" to your gc conda interpreter env that you added earlier.
• That's it, you will now have correct env in Terminal, Python Console, and elsewhere in the IDE.

How to Setup Visual Studio Code IDE

Note: if you're using containerized env, you'll need setup like this

• Open the cloned dir in new workspace and make sure to set your conda gc venv as the python venv https://code.visualstudio.com/docs/python/environments
• That's it, when you start new integrated terminals, they'll activate the right environment and the syntax highlighting/autocompletion is going to work as it's supposed to.

Common Issues

My venv is broken somehow!

• Delete the old conda environment and create a new one, follow steps above to reinstall it.

License & Contributions

See LICENSE.md (including licensing intent - INTENT.md) and CONTRIBUTING.md