Terraform AWS Lambda's redeployment - amazon-web-services

Question

We have a list of lambda functions that get updated quite often and new Lambda's are being introduced occasionally. The artifacts for lambdas are stored in AWS S3 bucket. We use Terraform AWS module to deploy lambdas in the following way:
module "lambdas" {
count = length(local.lambdas_list)
source = "terraform-aws-modules/lambda/aws"
function_name = "${local.lamdas_list[count.index]}"
lambda_role = module.create_network.lambda_role_arn
handler = "com.box.processing.Handler::handleRequest"
runtime = "java11"
memory_size = "1024"
timeout = "10"
vpc_subnet_ids = module.create_network.private_subnet_ids
vpc_security_group_ids = [module.create_test2_server.server_sg]
create_role = false
create_package = false
publish = false
ignore_source_code_hash = true
create_current_version_allowed_triggers = false
s3_existing_package = {
bucket = "lambdas_artifacts_bucket"
key = "${local.lambdas_list[count.index]}/${local.lambdas_list[count.index]}.jar"
}
}
lambdas_list variable looks like this:
lambdas_list = ["lambda1", "lambda2", "lambda3"]
===============================================================================================
The problem with this approach is that module is being instantiated as many times as there are elements in the lambdas_list variable, in case above - 3 times.
The biggest problem that we have is when somebody creates a new lambda and asks for it to be deployed. This means that we will be expanding the lambdas_list variable in the following way:
lambdas_list = ["lambda1", "lambda2", "lambda3", "lambda4"]
If I do a "terraform apply" after updating the lambdas_list variable in the way shown above, the code will deploy a new lambda (lambda4) and that's exactly what we want. The problem is when the lambdas_list variable is updated in the following way:
lambdas_list = ["lambda1", "lambda4", "lambda2", "lambda3"]
Note that "lambda4" is now at index 1 instead of index 3. In this case, the TF code will fail because the ordering of lambdas is changed and it the situation with lambda deployment looks like this:
INDEX OLD_STATE NEW_STATE
0 lambda1 lambda1
1 lambda2 lambda4
2 lambda3 lambda2
3 lambda3
The TF code will try to replace indexes 1, 2 and 3 with new lambdas. One of the things where it will fail is on the creation of Lambda Log Groups. When creating lambda with index 2 (lambda2), it will say that Log group already exists.
===============================================================================================
Is there a way to easily add/update list of lambdas to be deployed and not to have this problem every time? One of the important things is also that when a new lambda is added, we try to keep alphabetically ordered list of lambdas which causes this shift in the indexes.