AWS + Amplify + Step Function + Lambda - Simple example tutorial

[@FoundationForLawandG]

From a friend who found your video helpful: "That video sent made me almost cry. The guy on that video. May Allah give him hidaya because he helped me; may he go to paradise. Ameen. If you watch the video he does exactly what we are trying to do." We are a small non-profit attempting to provide a free service to senior citizens. Thank You.

[@parrou3]

Hi, thank you for this awesome Video, I have a qyestion plz, I am developing a public event app using AWS amplify, data are stored in a dynamoDB table and connection between data and application is done with appsync/graphQl, an authenticated user can create a public event, but if no one is interested in it, after 24 hours the status (field in dynamo table) of the event must be switched to 'Expired' and a notification will be sent to the owner of event, Since I am newbie to AWS and amplify, is Lambda function is best service to implement this feature and How? thanks

[@ForDevelopers]

Hi, technically you could use TTL on the table, but most probably this is not what you want - to delete item. So then the solution could be really based on the Lambda function which can use the DynamoDbStream and listen for the Insert event, when the new item is inserted you always have some creation timestamp and base of that you can calculate the 24hours event. In my example I was using the Step Function and wait there, because step function can last long time and the execution is cheap. From that step function you could then trigger the notification (I think the example I showed in the video you can use as well). Another approach would be to have only one Lambda function, which would periodically check the dynamodb table, and if createTimestamp of the item is beyond the 24hours then if could change the state and send the notification. This of course means you need to decide on your schedule interval? 1 hour? 15min? Will the lambda function finish until the timeout? It depends how many items you have in the table, which fullfill the condition. The first approach is more reliable but there are many others, it really depends of the complexity, what you want to build, how fast it should be ...

[@parrou3]

@@ForDevelopers Thank you a lot, I appreciate all the time and effort you put in to answer me :)