Local_machine_setup.md

Cohort Manager

Prerequisites

• Visual Studio Code - VS Code is the chosen editor for the project due to VS for Mac being retired
• Git
• HomeBrew (Mac Only):
  /bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"
• .NET SDK (8.0)
• Member of the NHSDigital GitHub organisation
• Signed Git commits (if you would like to contribute): https://github.com/NHSDigital/software-engineering-quality-framework/blob/main/practices/guides/commit-signing.md

Set-up

1. Import the NHS DToS Profile

To easily install the required extensions and settings/configuration for VS Code, you can import the profile located in Set-up/NHS_DToS.code-profile

On the top toolbar of Visual Studio Code go to Code > Settings > Profiles > Import Profile > click on Select File... and select the file NHS_DToS.code-profile

2. Azure Functions Core Tools

Azure Function Core Tools lets you develop and test your functions on your local computer. To install, press ctrl/ command + shift + P and enter Azure Functions: Install or Update Azure Functions Core Tools

3. Azure Data Studio & Storage Explorer

Azure Data Studio & Storage Explorer are the GUI tools we are using to manually interact with the database & Azure Storage respectively.

• Install Azure Data Studio
• (Optional) Install Azure Storage Explorer

Use the Intel Chip/ x64 installer if you have and Intel Chip in your Mac. Otherwise, use the Apple Silicon/ ARM64 installer.

Note: to check which version you are using, you can click on the Apple icon of your machine > About this Mac and a new window will appear. You can see the Chip your machine. Intel will have Intel in it, Apple Silicon will have something like Apple M1.

Running the Application Locally (Windows)

Download docker engine using these instructions

First, copy the .env.example file, rename it to just ".env", and follow the instructions inside the file to add the variables.

The docker compose has now been split into 4 files due to the size of the application being too large to build in one go. There are now 4 files:

• compose.deps.yaml - contains the database, azurite and setup containers, this must be run before the other files
• compose.core.yaml - contains the application minus cohort distribution
• compose.cohort-distribution.yaml - cohort distribution
• compose.yaml - imports the core and cohort distribution files so they can be interacted with together

To build and run the system locally:

# Build the functions
docker compose -f compose.core.yaml build
docker compose -f compose.cohort-distribution.yaml build

docker compose -f compose.deps.yaml up --build -d # Run the deps before the rest of the functions
docker compose up # Run the functions

Other useful commands:

docker compose down    # Stop the functions
docker compose up/ down <service-name>   # Start a particular function or dependency

# Example:
docker compose up receive-caas-file

docker ps -a   # List all of the containers
docker logs <container-name>   # View the logs of the container

*Note anything in local.settings.json will not apply to the containerised functions

Alternatively, you can run an individual function locally with func start

Running the Application Locally (Mac)

The full containerised solution does not work on Macs so you will have to run the application manually, but you can run the dependencies with docker

Dependencies

Download Colima using these instructions

Add the environment variables using the instructions in the windows setup

You can then run and setup the dependencies using docker: docker compose -f compose.deps.yaml up

Functions

*Someone using a Mac please update this with instructions

Appendix A: Storage

The send-sample-file.py script

There is a script in application/CohortManager/Set-up/azurite that alllows you to send sample files to azurite.

Before your run the script you must download the sample files from confluence (you can see which files you need to download by running the command without arguments) run the following command: pip install azure-storage-blob python-dotenv

Run the file without arguments (pyhton send-sample-file.py) to see the help page

Set-up Azure Storage Explorer

Alternatively, you can use the storage explorer to send files to azurite

Open the Azure Storage Explorer and in the Explorer, you will see Azurite (Key). Expand that and you will see 1) Blob Container 2) Queues and 3) Tables. Right click on the Blob Container and click on Create Blob Container.

On Azure Storage Explorer, collapse Emulator & Attached > Storage Accounts > Azurite (Key) and right click on Blob containers and select Create Blob Container and type in inbound to create a container with that name.

Once created, use the sample csv files upload it to that new inbound container.

Back in VS Code you should see the logs of the functions running locally, once it's complete, you can refresh the database again to see the changes made by the CSV files.

*Note: Sample Data and Scripts to create the database are provided by the Data team. The latest files can be found in the dtos-data-modes repository in the NHS Digital GitHub https://github.com/NHSDigital/dtos-data-models