How do I run mutlple SQL statements in an AppSync resolver? - amazon-web-services

Question

I have two tables, cuts and holds. In response to an event in my application I want to be able to move the values from a hold entry with a given id across to the cuts table. The naïve way is to do an INSERT and then a DELETE.
How do I run multiple sql statements in an AppSync resolver to achieve that result? I have tried the following (replacing sql by statements and turning it into an array) without success.
{
"version" : "2017-02-28",
"operation": "Invoke",
#set($id = $util.autoId())
"payload": {
"statements": [
"INSERT INTO cuts (id, rollId, length, reason, notes, orderId) SELECT '$id', rollId, length, reason, notes, orderId FROM holds WHERE id=:ID",
"DELETE FROM holds WHERE id=:ID"
],
"variableMapping": {
":ID": "$context.arguments.id"
},
"responseSQL": "SELECT * FROM cuts WHERE id = '$id'"
}
}

Answer 1

If you're using the "AWS AppSync Using Amazon Aurora as a Data Source via AWS Lambda" found here https://github.com/aws-samples/aws-appsync-rds-aurora-sample, you won't be able to send multiple statements in the sql field
If you are using the AWS AppSync integration with the Aurora Serverless Data API, you can pass up to 2 statements in a statements array such as in the example below:
{
"version": "2018-05-29",
"statements": [
"select * from Pets WHERE id='$ctx.args.input.id'",
"delete from Pets WHERE id='$ctx.args.input.id'"
]
}

Answer 2

You will be able to do it as the following.If you're using the "AWS AppSync Using Amazon Aurora as a Data Source via AWS Lambda" found here https://github.com/aws-samples/aws-appsync-rds-aurora-sample.
In the resolver, add the "sql0" & "sql1" field (you can name them what ever you want) :
{
"version" : "2017-02-28",
"operation": "Invoke",
#set($id = $util.autoId())
"payload": {
"sql":"INSERT INTO cuts (id, rollId, length, reason, notes, orderId)",
"sql0":"SELECT '$id', rollId, length, reason, notes, orderId FROM holds WHERE id=:ID",
"sql1":"DELETE FROM holds WHERE id=:ID",
"variableMapping": {
":ID": "$context.arguments.id"
},
"responseSQL": "SELECT * FROM cuts WHERE id = '$id'"
}
}
In your lambda, add the following piece of code:
if (event.sql0) {
const inputSQL0 = populateAndSanitizeSQL(event.sql0, event.variableMapping, connection);
await executeSQL(connection, inputSQL0);
}
if (event.sql1) {
const inputSQL1 = populateAndSanitizeSQL(event.sql1, event.variableMapping, connection);
await executeSQL(connection, inputSQL1);
}
With this approach, you can send to your lambda as much sql statements as you want,and then your lambda will execute them.