minnesota

Sample project to demonstrate how to use Azure Search Service with .NET Core.

Using wildcards in Azure Search queries is a common requirement. But it's not as straightforward as it seems.

I created this project to demonstrate a few options to use wildcards in Azure Search queries.

Spoiler alert: None of them works as expected 😢

First: Deploying Azure Search Service

After loggin in to Azure CLI, run the following command to create a new Azure Search Service.

cd src
source ../deploy_infra.sh

The deploy_infra.sh script will create a new Resource Group and an Azure Search Service instance. It will also set the SEARCH_ENDPOINT and SEARCH_ADMIN_KEY environment variables.

PROJECT_PREFIX=$(cat /dev/urandom | tr -dc 'a-z0-9' | fold -w 6 | head -n 1)
LOCATION="westus"

if [ -f .minnesota.json ]; then
  PROJECT_PREFIX=$(jq -r '.project_prefix' .minnesota.json)
  LOCATION=$(jq -r '.location' .minnesota.json)
else
  echo "{\"project_prefix\": \"$PROJECT_PREFIX\", \"location\": \"$LOCATION\"}" > .minnesota.json
fi

az group create --name "${PROJECT_PREFIX}-rg" --location "${LOCATION}"
az search service create --resource-group "${PROJECT_PREFIX}-rg" --name "${PROJECT_PREFIX}-search" --location "${LOCATION}" --sku "basic" --semantic-search "standard"

export SEARCH_ENDPOINT="https://${PROJECT_PREFIX}-search.search.windows.net"
export SEARCH_ADMIN_KEY=$(az search admin-key show --resource-group "${PROJECT_PREFIX}-rg" --service-name "${PROJECT_PREFIX}-search" --query "primaryKey" -o "tsv")

Second: Populating the Index

After creating the Azure Search Service, you can run the following command to populate the index with some sample data (./src/data.json).

cd src
dotnet run

The Program.cs file contains the code to create the index and populate it with some sample data (./src/data.json).

After Search Index is created and populated with _sample data+, Program.cs is going in a loop to ask for a search query, run the query and display the results.