Use AWS CLI and JSON to Load Multiple Items into DynamoDB

Use AWS CLI and JSON to Load Multiple Items into DynamoDB

·

4 min read

I found it surprisingly difficult to find documentation that demonstrates how to add MULTIPLE items to a dynamo DB from a single JSON file. Well here is it, plain and simple. I've included a few different types of examples.

Here's the template:

{
  "DDBTableName":[
  {
    "PutRequest":{
      "Item":{
          "keyName":{"DataType":"Value1"}
        }
      }
    },
  {
    "PutRequest":{
      "Item":{
          "keyName":{"DataType":"Value2"}
        }
      }
    }
  ]
}

Here's a sample table called HotelInfoAccessibility. It has a single key called "text". Real fancy, I know.

{
    "HotelInfoAccessibility": [
        {
            "PutRequest": {
                "Item": {
                    "text": { "S": "Grab bars on tub walls"}
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "text": { "S":  "Shower chairs"}
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "text": { "S": "Telephones with volume control"}
                }
            }
        }
    ]
}

If you copy and paste to quickly fill in another table, DON'T FORGET TO CHANGE THE TABLE NAME. I did it... it wasn't fun.

This next example is also simple. It's a two key table called MenuLinks. The primary key is href, the sort key is text. I also include a class key cause that's probably a thing I need.

{
    "MenuLinks": [
        {
            "PutRequest": {
                "Item": {
                    "class": {
                        "S": "info"
                    },
                    "href": {
                        "S": "#hotelinfo"
                    },
                    "text": {
                        "S": "info"
                    }
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "class": {
                        "S": "rooms"
                    },
                    "href": {
                        "S": "#rooms"
                    },
                    "text": {
                        "S": "rooms"
                    }
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "class": {
                        "S": "dining"
                    },
                    "href": {
                        "S": "#dining"
                    },
                    "text": {
                        "S": "dining"
                    }
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "class": {
                        "S": "events"
                    },
                    "href": {
                        "S": "#events"
                    },
                    "text": {
                        "S": "events"
                    }
                }
            }
        },
        {
            "PutRequest": {
                "Item": {
                    "class": {
                        "S": "attractions"
                    },
                    "href": {
                        "S": "#attractions"
                    },
                    "text": {
                        "S": "attractions"
                    }
                }
            }
        }
    ]
}

And if you want to get super fancy, you can nest your json, should work the same way.

Once you've got your data properly formatted and saved to a json file, you can run the aws cli to write these items to the table.

aws dynamodb batch-write-item --request-items file://menu_links.json

Make sure you point to the correct file location, obviously. The response will be:

{
    "UnprocessedItems": {}
}

Login to the DynamoDB console, or use the cli or another program to query your data.

image.png

That's what I've been working on today. Some days are more productive than other :)

-AJ