Skip to content

Table Storage

Persist and query structured NoSQL data with Azure Table storage bindings in PowerShell.

Input Binding

Read entities filtered by partition key:

function.json:

{
  "bindings": [
    {
      "name": "Request",
      "type": "httpTrigger",
      "direction": "in",
      "methods": ["get"]
    },
    {
      "name": "Response",
      "type": "http",
      "direction": "out"
    },
    {
      "name": "Orders",
      "type": "table",
      "direction": "in",
      "tableName": "Orders",
      "partitionKey": "{Query.customerId}",
      "connection": "AzureWebJobsStorage"
    }
  ]
}

run.ps1:

param($Request, $Orders, $TriggerMetadata)

Write-Information "Retrieved $($Orders.Count) order(s)"
Push-OutputBinding -Name Response -Value ([HttpResponseContext]@{
    StatusCode = [System.Net.HttpStatusCode]::OK
    Body       = ($Orders | ConvertTo-Json)
})

Output Binding

Write one or more entities. Each must include PartitionKey and RowKey:

{
  "name": "OutputOrder",
  "type": "table",
  "direction": "out",
  "tableName": "Orders",
  "connection": "AzureWebJobsStorage"
}

Push-OutputBinding -Name OutputOrder -Value @{
    PartitionKey = "customer-42"
    RowKey       = [guid]::NewGuid().ToString()
    Total        = 99.5
    Status       = "created"
}

Choosing keys

Pick a PartitionKey that distributes writes evenly and groups the entities you query together. A monotonic RowKey (such as a reverse-timestamp) keeps recent items first.

See Also

Sources