I am developing an web app using AWS amplify.
I want to delete multiple items that satisfies certain condition using a query like the following:
mutation delete {
deletePostTag(condition: {title: {eq: "Hello"}}) {
id
}
}
However, having tried to run the above query on AWS AppSync console, it complains that input field is missing, but unfortunately input only accepts id.
It seems that the resolver generated by amplify cli does not support deleting multiple items at once.
Do I have to implement a custom resolver?
You can delete multiple items in a batch. Example below and read more here.
Schema:
type Mutation {
batchDelete(ids: [ID]): [Post]
}
Query:
mutation delete {
batchDelete(ids:[1,2]){ id }
}
Not 100% sure if conditions are supported here, but hopefully you can test it. If, as I suspect, they are not supported then simply issue a query with those same conditions to retrieve matching items and then supply the resulting array of item keys to batchDelete.
Related
I am using AWS proxy with AWS API Gateway to interact with a DynamoDB table. I have an API resource, under which I have a GET method with the below configuration:
The API uses the Scan action as seen above to fetch all the items from the DynamoDB table. I also have the following request integration mapping template;
{
"TableName": tableName
}
Its really simple. But my problem is that I would like to add another GET method to get each item by their id, which will be supplied in the URL as a param. However, since I have already setup one GET method, I am not able to setup another to fetch only a single item. I am aware I can use mapping templates and Scan as given in the docs to conditionally fetch items if a param is given, but that would mean scanning the entire table, which is a waste each time I want to fetch a single item.
Is there any other way to do this?
I am trying to learn how to use AppSync and its DynamoDB integrations.
I have successfully created an AppSync GraphQL API and linked a resolver to a getter on the primary key and thought I understood what is happening. However, I can not get a putItem resolver to work at all and am struggling to find a useful way to debug the logic.
There is a cdk repository here which will deploy the app. Lines 133-145 have a hand written schema which I thought should work however that receives the error
One or more parameter values were invalid: Type mismatch for key food_name expected: S actual: NULL (Service: DynamoDb, Status Code: 400
I also have attempted to wrap the expressions in quotes but receive errors.
Where should I go from here?
The example data creates a table with keys
food_name
scientific_name
group
sub_group
with food_name as the primary key.
https://github.com/AG-Labs/AppSyncTask
Today I have attempted to reimplement the list resolver as
{
"version" : "2017-02-28",
"operation" : "Scan",
## Add 'limit' and 'nextToken' arguments to this field in your schema to implement pagination. **
"limit": $util.defaultIfNull(${ctx.args.limit}, 20),
"nextToken": $util.toJson($util.defaultIfNullOrBlank($ctx.args.nextToken, null))
}
with a response mapping of
$util.toJson($ctx.result.items)
In cloud watch I can see a list of results under log type ResponseMapping (albeit not correctly filtered but i'll ignore that for now) but these do not get returned to the querier. That result is simply
{
"data": {
"listGenericFoods": {
"items": null
}
}
}
I don't understand where this is going wrong.
The problem was that the resolvers were nested.
Writing a handwritten schema fixed the issue but resulted in a poorer API. Going back a few steps and will implement from the ground up slowly adding more resolvers.
The CloudWatch Logs once turned on helped somewhat but still required a lot of changing the resolvers ever so slightly and retrying.
Is there a simple way to retrieve all items from a DynamoDB table using a mapping template in an API Gateway endpoint? I usually use a lambda to process the data before returning it but this is such a simple task that a Lambda seems like an overkill.
I have a table that contains data with the following format:
roleAttributeName roleHierarchyLevel roleIsActive roleName
"admin" 99 true "Admin"
"director" 90 true "Director"
"areaManager" 80 false "Area Manager"
I'm happy with getting the data, doesn't matter the representation as I can later transform it further down in my code.
I've been looking around but all tutorials explain how to get specific bits of data through queries and params like roles/{roleAttributeName} but I just want to hit roles/ and get all items.
All you need to do is
create a resource (without curly braces since we dont need a particular item)
create a get method
use Scan instead of Query in Action while configuring the integration request.
Configurations as follows :
enter image description here
now try test...you should get the response.
to try it out on postman deploy the api first and then use the provided link into postman followed by your resource name.
API Gateway allows you to Proxy DynamoDB as a service. Here you have an interesting tutorial on how to do it (you can ignore the part related to index to make it work).
To retrieve all the items from a table, you can use Scan as the action in API Gateway. Keep in mind that DynamoDB limits the query sizes to 1MB either for Scan and Query actions.
You can also limit your own query before it is automatically done by using the Limit parameter.
AWS DynamoDB Scan Reference
I'm using aws appsync with react native and there is transaction happening offline and I want to know if mydata that is being transact offline is already saved in my db online.
The fetch policy I'm using is already network-only, but "network-only" policy is not working because it can still catch data if it is offline.
If you are using DynamoDB with AppSync, you can add a condition expression to your mutation resolver request mapping template. DynamoDB conditions are used to validate whether the mutation should succeed or not.
Many people use versions with DynamoDB condition checks to validate that a record hasn't already been updated, but you can add additional fields to keep track of whether the transaction has already been made.
Here is an example condition expression that you can add to your request mapping template to validate the incoming mutation:
"condition" : {
"expression" : "version = :expectedVersion",
"expressionValues" : {
":expectedVersion" : { "N" : ${context.arguments.expectedVersion} }
}
}
Here is is an overly comprehensive guide to using DynamoDB resolvers:
https://docs.aws.amazon.com/appsync/latest/devguide/tutorial-dynamodb-resolvers.html#modifying-the-updatepost-resolver-dynamodb-updateitem
Is there any way of connecting appsync to dynamodb with option to put multiple items at once? I'm asking about dynamodb reuqest mapping.
Let's say we have saveVotes mutation
type Mutation {
saveVotes(votes: [VoteInput]): [Vote]!
}
How should I design dynamo request template to get each vote saved as separate object in the dynamodb? Each VoteInput has ID. I'm sending 5 VoteInput, and I want to have 5 separate object, with separate ID in the dynamodb.
In the AWS docs there are examples just for single putItem which is not enough for me
https://docs.aws.amazon.com/appsync/latest/devguide/resolver-mapping-template-reference-dynamodb.html#aws-appsync-resolver-mapping-template-reference-dynamodb-putitem
You should take a look at using Batch Resolvers at https://docs.aws.amazon.com/appsync/latest/devguide/tutorial-dynamodb-batch.html
As I understand it, you can construct your GraphQL queries to accept a collection of items and then implement the associated resolvers to perform batch operations in the way you describe.
So the best solution for me is to make query this way
mutation saveVotes {
vote1: saveVotes(author: "AUTHORNAME", value: 1 ) { author, value }
vote2: saveVotes(author: "AUTHORNAME", value: 3 ) { author, value }
}