Prepare

10 minutes

You're updating an existing .NET solution that has a Blazor application as it's sole project. This project includes service classes for data storage and conversation that are implemented with placeholders. Here, we discuss the overall goals of the project and how you're going to implement the data store and conversational functionality. Additionally, we cover how to set up the Azure resources that you need.

Project overview

This project's goal is to implement two service classes for data storage and completions. For the data storage service class, use the Azure Cosmos DB for NoSQL .NET SDK to create, update, query, and delete multiple items in an existing container. For chat completion, use the Azure OpenAI .NET SDK to send requests and parse responses.

The key tasks you need to do are:

Deploy Azure Cosmos DB for NoSQL and Azure OpenAI resources to Azure.
Get the credentials for the deployed Azure Cosmos DB for NoSQL and Azure OpenAI resources.
Open the starter solution in Visual Studio Code.
Use the .NET SDK to implement Azure Cosmos DB for NoSQL as a data storage service.
Use the .NET SDK to implement Azure OpenAI as a conversational interface.
Check the application's operation.

Setup

To complete this project, you need an Azure Cosmos DB for NoSQL account and an Azure OpenAI account. To streamline this process, deploy a Bicep template to Azure with both of these accounts.

Configure dev environment

A development container environment is available with all dependencies required to complete every exercise in this project. GitHub Codespaces runs a development container managed by GitHub with Visual Studio Code for the Web as the user interface. For the most straightforward development environment, use GitHub Codespaces so that you have the correct developer tools and dependencies preinstalled to complete this training module.

Important

All GitHub accounts can use Codespaces for up to 60 hours free each month with 2 core instances. For more information, see GitHub Codespaces monthly included storage and core hours.

Start the process to create a new GitHub Codespace by selecting the Active Codespaces option at the beginning of this unit.
On the Create codespace page, review the codespace configuration settings, and then select Create new codespace.
Wait for the codespace to start. This startup process can take a few minutes.
Open a new terminal in the codespace.

Tip

You can use the main menu to navigate to the Terminal menu option and then select the New Terminal option.
Validate that the Azure CLI is installed in your environment.
```
az --version
```
Validate that .NET 8 is installed in your environment.
```
dotnet --list-sdks
```
Change to the src directory, where the project file is located.
```
cd src
```
Build the .NET project.
```
dotnet build
```
Close the terminal.
The remaining exercises in this project take place in the context of this development container.

Important

Leave the GitHub Codespace running for the remainder of this training module.

Deploy infrastructure from template

This project uses the azure-samples/cosmosdb-chatgpt GitHub repository for both the Azure Resource Manager templated deployment and the application code. In this step, create a new deployment using the Bicep template to deploy your resources using "infrastructure as code."

Tip

This "zero-touch" Bicep template also includes the final application running in Azure App Service. You can always check this out if you want to see a fully working solution.

Sign in to the Azure CLI.
```
az login --use-device-code
```
Create a new shell variable named resourceGroupName with the name of the Azure resource group that you create (mslearn-cosmos-openai).
```
resourceGroupName="mslearn-cosmos-openai"
```

Create a resource group using az group create.

az group create \
  --name $resourceGroupName \
  --location "eastus"

Deploy the azuredeploy.json template file to the resource group using az deployment group create.

az deployment group create \
  --resource-group $resourceGroupName \
  --name zero-touch-deployment \
  --template-uri https://raw.githubusercontent.com/Azure-Samples/cosmosdb-chatgpt/start/azuredeploy.json

Wait for the deployment to complete before proceeding with this project.

Note

On average, this deployment can take five to ten minutes.

Get Azure Cosmos DB for NoSQL and Azure OpenAI account credentials

Your template deployed Azure Cosmos DB for NoSQL and Azure OpenAI accounts and then stored their credentials in the Azure App Service web app's configuration. Now, you have the choice of using the Azure portal or Azure CLI to retrieve the credentials for each service. Use az webapp config from the Azure CLI to get the Azure OpenAI and Azure Cosmos DB credentials to use in your local development environment. In this section, you filter the output of the various commands to return your endpoint and connection string to use with the .NET libraries of both Azure OpenAI and Azure Cosmos DB for NoSQL.

First, get a list of all running web applications in your resource group using az webapp list.

az webapp list \
  --resource-group $resourceGroupName \
  --query "[?state=='Running'].name"

Now, run the same query again but only return the first result from the array in tab-separated value (tsv) format.
```
webAppName=$( \
  az webapp list \
    --resource-group $resourceGroupName \
    --query "[?state=='Running'].name | [0]" \
    --output tsv \
)
```
Tip

If you're not sure that your variable has the correct value, simply run echo $webAppName in the terminal to check the value.

Use az webapp config appsettings list to get a list of application settings for the web app.

az webapp config appsettings list \
  --name $webAppName \
  --resource-group $resourceGroupName

Here's a truncated example of the output assuming the name of the Azure Cosmos DB for NoSQL account is nybncrsna76fo-cosmos-nosql and the name of the Azure OpenAI account is nybncrsna76fo-openai. The keys in this example are fictitious.

[
  {
    "setting": "COSMOSDB__ENDPOINT",
    "value": "https://nybncrsna76fo-cosmos-nosql.documents.azure.com:443/"
  },
  {
    "setting": "COSMOSDB__KEY",
    "value": "Aj2OSQqWhXfFzkITIi4MIbocWAAoBCUWnSPzHlQbDYA3TiyZCJlk8A6R1l6WqD45BxAG5vrsyQ4SCXEkPS1pLh=="
  },
  ...
  {
    "setting": "OPENAI__ENDPOINT",
    "value": "https://nybncrsna76fo-openai.openai.azure.com/"
  },
  {
    "setting": "OPENAI__KEY",
    "value": "4bf98cb194cdf0f9001eae3259a76ed8"
  },
  ...
]