I am wondering if anyone knows best practices for handling Plaid webhooks with Java Springboot?
Does the Plaid SDK offer any easy way to convert the webhook request object to a model object for the given event type? I only see they have Node Express examples which seems to only deconstruct the JSON request object by key.
Also wondering if their is anyway to verify the incoming webhook request is actually from Plaid
#PostMapping(value = "/webhook/plaid", produces =
MediaType.APPLICATION_JSON_VALUE)
public ResponseEntity plaidWebhook(#RequestBody String payload) {
JSONParser parser = new JSONParser(payload);
JSONObject plaidWebhookRequest = null;
try {
plaidWebhookRequest = (JSONObject) parser.parse();
String webhookType = plaidWebhookRequest.has("webhook_type") ? (String) plaidWebhookRequest.get("webhook_type") : null;
String webhookCode = plaidWebhookRequest.has("webhook_code") ? (String) plaidWebhookRequest.get("webhook_code") : null;
String error = plaidWebhookRequest.has("error") ? (String) plaidWebhookRequest.get("error") : null;
String itemID = plaidWebhookRequest.has("item_id") ? (String) plaidWebhookRequest.get("item_id") : null;
if (webhookType != null && webhookCode != null && webhookType.equals(WebhookType.ITEM.name())) {
switch (webhookCode) {
case ERROR_WEBCODE:
log.info("Plaid webhook received: " + ERROR_WEBCODE);
break;
case PENDING_EXPIRATION:
log.info("Plaid webhook received: " + PENDING_EXPIRATION);
break;
case USER_PERMISSION_REVOKED:
log.info("Plaid webhook received: " + USER_PERMISSION_REVOKED);
break;
}
}
} catch (ParseException e) {
log.debug("Plaid webhook object failed to convert to JSONObject");
}
return ResponseEntity.status(HttpStatus.OK).body("");
}
I am not a Java expert but I can speak to some of the other parts of your question:
You can use the webhook verification endpoint to verify that the webhook is from Plaid: https://plaid.com/docs/api/webhooks/webhook-verification/ although I will admit the process is not as easy as most of the other things you can do with the Plaid API.
As an alternative option -- for a situation like this, you can always check the Item status by calling /item/get to confirm that the Item needs to be updated before sending the user through update mode. As a rule, Plaid doesn't ever send sensitive information in webhooks, and information in webhooks can be verified by calling endpoints that are free to call, so you should never need to "trust" a Plaid webhook without verifying it if you don't want to. This is generally smart to do anyway, for example: even if you got a webhook indicating that the Item is in an error state, the user may have resolved it or it may have self-healed in the interim.
Related
I am using AWS Java SDK v2 to list users using the code defined here on the AWS GitHub repo.
public static void listAllUsers(IamClient iam) {
try {
boolean done = false;
String newMarker = null;
while (!done) {
ListUsersResponse response;
ListUsersRequest request;
if (newMarker == null) {
request = ListUsersRequest.builder().build();
} else {
request = ListUsersRequest.builder()
.marker(newMarker).build();
}
response = iam.listUsers(request);
for (User user : response.users()) {
System.out.format("\n Retrieved user %s", user.userName());
System.out.println("\nPermission Boundary: " + user.permissionsBoundary());
}
if (!response.isTruncated()) {
done = true;
} else {
newMarker = response.marker();
}
}
} catch (IamException e) {
System.err.println(e);
System.exit(1);
}
}
It returns NULL for user.permissionsBoundary(). Here is the output for print statements in the above code.
Retrieved user jamshaid
Permission Boundary: null
Retrieved user luminadmin
Permission Boundary: null
Retrieved user test
Permission Boundary: null
When I run the following command in AWS CloudShell on AWS console, it returns the PermissionBoundary for the users it is defined.
aws iam get-user --user-name test
Here is the sample output from AWS CloudShell.
I am using the same account to make both requests.
I have confirmed this behavior by setting a permission boundary on an IAM user in the AWS Management Console. I changed the ListUsers example to include this code:
for(User user : response.users()) {
System.out.format("\n Retrieved user %s", user.userName());
AttachedPermissionsBoundary permissionsBoundary = user.permissionsBoundary();
if (permissionsBoundary != null)
System.out.format("\n Permissions boundary details %s", permissionsBoundary.permissionsBoundaryTypeAsString());
}
...
The permissionsBoundary() method does return null - even though the permission is set. This is a bug.
My advice here is to log a Github issue here:
https://github.com/aws/aws-sdk-java-v2
I also tested this with Kotlin SDK. Same result.
suspend fun listAllUsers() {
IamClient { region = "AWS_GLOBAL" }.use { iamClient ->
val response = iamClient.listUsers(ListUsersRequest { })
response.users?.forEach { user ->
println("Retrieved user ${user.userName}")
val permissionsBoundary = user.permissionsBoundary
if (permissionsBoundary != null)
println("Permissions boundary details ${permissionsBoundary.permissionsBoundaryType.toString()}")
}
}
}
I do not think it is an issue, but the programmed behavior. From the API docs:
IAM resource-listing operations return a subset of the available attributes
for the resource. For example, this operation does not return tags, even
though they are an attribute of the returned object. To view all of the
information for a user, see GetUser.
This is stated as well in the API javadocs.
In the console you are using get-user, not list-users, and this is why the command is returning all the information about the user, PermissionsBoundary within it.
Please, try instead using:
aws iam list-users
and check the output, it should match the result you obtained with the Java SDK, it will not contain PermissionsBoundary either.
If you want to obtain the same results that you are currently getting with the command aws iam get-user --user-name test from Java code, you can use the getUser method in IamClient. Try Something like:
GetUserRequest request = GetUserRequest.builder()
.userName("test")
.build()
;
GetUserResponse response = iam.getUser(request);
User user = response.user();
System.out.println("\nPermission Boundary: " + user.permissionsBoundary());
The User class is reused in both operations, get and list, but only in the former all the fields are populated.
I am trying to iterate over a paginated list of accounts but when I send a request using the value from "next_uri" I receive an error from the server:
{"errors":[{"id":"not_found","message":"Not found"}]}
I am correctly adding headers etc as all other API calls work fine, its just the request using the "next_uri" that is not working. I think I am following the api spec correctly so I am unsure what is the issue and how to fix it. Does anyone know what is wrong with the code / logic please?
Simplified code:
ArrayList<X> results = new ArrayList<>();
String uri = "/v2/accounts";
javax.ws.rs.client.Client client = getClient();
while(uri != null){
T response = client.target("https://api.coinbase.com")
.path(uri).request(MediaType.APPLICATION_JSON).get(responseType);
results.addAll(response.getData());
uri = response.getPagination()==null ? null :response.getPagination().getNextUri();
}
return results;
The results are this:
Request 1:
https://api.coinbase.com/v2/accounts
Response 1: pagination":
{"ending_before":null,"starting_after":null,"previous_ending_before":null,"next_starting_after":"ef35df6c-a45b-5858-b755-f12a709cf26e","limit":25,"order":"desc","previous_uri":null,"next_uri":"/v2/accounts?starting_after=ef35df6c-a45b-5858-b755-f12a709cf26e"},"data":[{....}]
Request 2:
https://api.coinbase.com/v2/accounts%3Fstarting_after=ef35df6c-a45b-5858-b755-f12a709cf26e
Response 2:
{"errors":[{"id":"not_found","message":"Not found"}]}
This was related to how the jax-rs library needs the query params adding. Just relying on the uri is not enough, the parameters also need adding specifically:
target = target.queryParam(e.getKey(), e.getValue());
so the final code is something like
WebTarget target = client.target(e"https://api.coinbase.com");
if(params !=null){
for(Map.Entry<String, String> e : params.entrySet()){
target = target.queryParam(e.getKey(), e.getValue());
}
}
target = target.path(path);
return target.request(MediaType.APPLICATION_JSON).get(responseType);
Could someone let me know why the below code only fetching few entries from the parameter store ?
GetParametersByPathRequest getParametersByPathRequest = new GetParametersByPathRequest();
getParametersByPathRequest.withPath("/").setRecursive(true);
getParametersByPathRequest.setWithDecryption(true);
GetParametersByPathResult result = client.getParametersByPath(getParametersByPathRequest);
result.getParameters().forEach(parameter -> {
System.out.println(parameter.getName() + " - > " + parameter.getValue());
});
GetParametersByPath is a paged operation. After each call you must retrieve NextToken from the result object, and if it's not null and not empty you must make another call with it added to the request.
Here's an example using DescribeParameters, which has the same behavior:
DescribeParametersRequest request = new DescribeParametersRequest();
DescribeParametersResult response;
do
{
response = client.describeParameters(request);
for (ParameterMetadata param : response.getParameters())
{
// do something with metadata
}
request.setNextToken(response.getNextToken());
}
while ((response.getNextToken() != null) && ! respose.getNextToken.isEmpty());
Here is the code, based on the code above, for the new 2.0 version of AWS SSM manager. Notice I have set the maxResults to 1 to prove out the loop. You will want to remove that. AWS has mentioned that in the new code they wanted to emphasize immutability.
Using this dependency:
<dependency>
<groupId>software.amazon.awssdk</groupId>
<artifactId>ssm</artifactId>
<version>2.10.32</version>
</dependency>
I came up with this code:
private void refreshCache() {
StopWatch sw = StopWatch.createStarted();
GetParametersByPathRequest request = GetParametersByPathRequest.builder()
.path(prefix)
.withDecryption(useDecryption)
.maxResults(1)
.build();
GetParametersByPathResponse response;
do {
response = ssm.getParametersByPath(request);
for (Parameter p : response.parameters()) {
//do something with the values.
}
request = GetParametersByPathRequest.builder()
.path(prefix)
.withDecryption(useDecryption)
.nextToken(response.nextToken())
.maxResults(1)
.build();
}
while (StringUtils.isNotBlank(response.nextToken()));
LOG.trace("Refreshed parameters in {}ms", sw.getTime());
}
private void getSsmParams() {
AWSSimpleSystemsManagement client = AWSSimpleSystemsManagementClientBuilder.defaultClient();
GetParametersByPathRequest request = new GetParametersByPathRequest();
request.withRecursive(true);
request.withPath('/your/path/parameterName').setWithDecryption(true);
GetParametersByPathResult response;
do {
response = client.getParametersByPath(request);
for (Parameter p : response.parameters()) {
//do something with the values. maybe add to a list
}
request.setNextToken(response.getNextToken())
}
while (StringUtils.isNotBlank(response.getNextToken()));
}
Above piece of code worked for me .ssm only sends 10 parameters at a time, so if you want to fetch more than 10 parameters from ssm parameter store programatically you will have to use multiple calls to fetch them. here the token is important , if there are more values in the path (request.withPath('/your/path/parameterName')) you have given, it will send a token indicating that there are more values in the given path ,and you will have to make the following request with the token received from the previous request in order to get the rest of the values.
I am attempting to create new Interactions programmatically on Genesys Platform SDK 8.5 for Java.
I use the example on the API reference
public void createInteraction(String ixnType, String ixnSubtype, String queue) throws Exception
{
RequestSubmit req = RequestSubmit.create();
req.setInteractionType(ixnType);
req.setInteractionSubtype(ixnSubtype);
req.setQueue(queue);
req.setMediaType("email");
Message response = mPMService.getProtocol("IxnSrv").request(req);
if(response == null || response.messageId() != EventAck.ID) {
// For this sample, no error handling is implemented
return;
}
EventAck event = (EventAck)response;
mInteractionId = event.getExtension().getString("InteractionId");
}
However, this gives me an Unsupported protocol element error.
'EventError' (126) attributes:
attr_error_desc [str] = "Unsupported protocol element"
attr_ref_id [int] = 2
attr_error_code [int] = 4
How do I create a new Interaction programmatically?
Interaction server should be connected with ClientType as either MediaServer or AgentApplication for this request(RequestSubmit).
First of all, you must open your protocol as Media Server. After that you must submit your interaction to interaction server.
Firstly your protocol config must be like this;
interactionServerConfiguration.ClientName = "TestClient";
interactionServerConfiguration.ClientType = InteractionClient.MediaServer;
// Register this connection configuration with Protocol Manager
protocolManagementService.Register(interactionServerConfiguration);
Note : You must have MediaServer type application definition on your Configuration Env., you must see it in CME.
After open you connection to ixn server. You can submit your interaction what you like. Even you can create new type interaction just like i do. I did for our coopate sms system. Its name is not important. We defined it on our bussiness attribute, so our agent can send coopate 3rd party sms system from their agent desktop. Without new extension or new license :) Just tricked it system. Also genesys allows it. i know it because we are genesys official support team in our country :) (But agent seat license may be required depends on agent head count).
RequestSubmit request = RequestSubmit.Create();
request.TenantId = 1;
request.MediaType = "email";
request.Queue = c_inboundQueue;
request.InteractionType = "Inbound";
request.InteractionSubtype = "InboundNew";
// Prepare the message to send. It is inserted in the request as UserData
KeyValueCollection userData =
new KeyValueCollection();
// Prepare the message to send
userData.Add("Subject", "subject goes here");
request.UserData = userData; protocolManagementService[c_interactionServerConfigurationIdentifier].Send(request);
Turns out I needed to set ClientType to InteractionClient.ReportingEngine.
I have an Android Application that is stored in the Cloud, with Google App Engine. I use Cloud Endpoints. My problem is that I cannot send the data from the server to my client(Android Device), or better said, so far, I have no idea how to do that.
So far, I have managed to insert data in the datastore, by creating an endpoint and calling the method that is in charge with adding a record in the database(that's located on the server side, in myProject - AppEngine) , using the following code (on the client):\
Noteendpoint.Builder endpointBuilder = new Noteendpoint.Builder(
AndroidHttp.newCompatibleTransport(),
new JacksonFactory(),
new HttpRequestInitializer() {
public void initialize(HttpRequest httpRequest) { }
});
Noteendpoint endpoint = CloudEndpointUtils.updateBuilder(
endpointBuilder).build();
try {
// Construct the note.
Note note = new Note().setDescription("Note DescriptionRoxana");
String noteID = new Date().toString();
note.setId(noteID);
note.setEmailAddress("E-Mail AddressRoxana");
// Insert the Note, by calling a method that's on the server side - insertNote();
Note result = endpoint.insertNote(note).execute();
} catch (IOException e) {
e.printStackTrace();
}
But I cannot see a way of retrieving data from the datastore and to display it on the server side. I tried to do the same, create an endpoint, that will call the method that retrieves all the records in the database (method that is located on the server), but my application crashes.
The code for the method that retrieves data from the datastore is the following:
public CollectionResponse<Note> listNote(
#Nullable #Named("cursor") String cursorString,
#Nullable #Named("limit") Integer limit) {
EntityManager mgr = null;
Cursor cursor = null;
List<Note> execute = null;
try {
mgr = getEntityManager();
Query query = mgr.createQuery("select from Note as Note");
if (cursorString != null && cursorString != "") {
cursor = Cursor.fromWebSafeString(cursorString);
query.setHint(JPACursorHelper.CURSOR_HINT, cursor);
}
if (limit != null) {
query.setFirstResult(0);
query.setMaxResults(limit);
}
execute = (List<Note>) query.getResultList();
cursor = JPACursorHelper.getCursor(execute);
if (cursor != null)
cursorString = cursor.toWebSafeString();
// Tight loop for fetching all entities from datastore and accomodate
// for lazy fetch.
for (Note obj : execute)
;
} finally {
mgr.close();
}
return CollectionResponse.<Note> builder().setItems(execute)
.setNextPageToken(cursorString).build();
}
You see, the returned type is collection response. You have access to this type of data, after performing the following import:
import com.google.api.server.spi.response.CollectionResponse;
I inferred that this is a data type characteristic to the server side, thus, I have no idea how I can cast it into a List, ArrayList, or any other type of collection, that can be used on the client side.
How should I do it then? Since adding data was so easy and so straight forward, I have assumed that retrieving data would be performed in the same manner, but apparently I am missing something essential for this matter.
Thank you in advance!
The classes you use in the backend are not the same as the classes you'll use in the client. Endpoints will generate a set of libraries for you, either on the command line or with tooling like Google Plugin for Eclipse. See Using Endpoints in an Android Client.
The generated class representing the collection of Notes in your example will be named something like NotesCollection. This object will have a method getItems, which provides you a List<Note> you can iterate on in your Android application.
Similar to having an endpoint for inserting data into a datastore model(methods of type POST), you need to have an endpoint for querying the data from the datastore model (methods of type GET). After you define both these methods, you need generate your discovery document and client library so that clients know about both these methods and can call them. If your speaking about displaying the data in the web itself then you can build a Javascript client by using the required client library.