Fivetrex.Destinations (Fivetrex v0.2.1)

View Source

Functions for managing Fivetran Destinations.

A Destination configures the target data warehouse where Fivetran will load synced data. Each group has exactly one destination. Supported destination types include Snowflake, BigQuery, Redshift, Databricks, and many others.

Overview

Destinations are the "where" of Fivetran - they define where your data lands. Each destination:

  • Belongs to a single group
  • Has a service type (e.g., "snowflake", "big_query")
  • Contains credentials and connection information
  • Has a region for data processing

Why No list/2 or stream/2?

Unlike Groups and Connectors, this module does not provide list/2 or stream/2 functions. This is intentional:

Fivetran's 1:1 Relationship: Each Fivetran group has exactly one destination, and the destination ID is the same as its group ID. This means:

  • To "list all destinations", simply list all groups and use their IDs
  • There's no separate /destinations endpoint that returns multiple items

How to iterate over destinations:

# Stream all destinations via their groups
client
|> Fivetrex.Groups.stream()
|> Stream.map(fn group ->
  case Fivetrex.Destinations.get(client, group.id) do
    {:ok, destination} -> destination
    {:error, _} -> nil
  end
end)
|> Stream.reject(&is_nil/1)
|> Enum.to_list()

Common Operations

Get a Destination

{:ok, destination} = Fivetrex.Destinations.get(client, "destination_id")

Create a Destination

{:ok, destination} = Fivetrex.Destinations.create(client, %{
  group_id: "group_id",
  service: "snowflake",
  region: "US",
  time_zone_offset: "-5",
  config: %{...}
})

Test Connection

{:ok, result} = Fivetrex.Destinations.test(client, "destination_id")

Supported Services

Common destination services include:

  • "snowflake" - Snowflake Data Cloud
  • "big_query" - Google BigQuery
  • "redshift" - Amazon Redshift
  • "databricks" - Databricks Lakehouse
  • "postgres" - PostgreSQL
  • "azure_sql_database" - Azure SQL Database

See Fivetran's documentation for the full list and configuration options.

Security Note

Destination configurations contain sensitive credentials. Always:

  • Store credentials securely (environment variables, secrets manager)
  • Use least-privilege database users
  • Rotate credentials periodically

See Also

Summary

Functions

create(client, params)

Creates a new destination.

delete(client, destination_id)

Deletes a destination.

get(client, destination_id)

Gets a destination by its ID.

test(client, destination_id)

Runs connection tests for a destination.

update(client, destination_id, params)

Updates an existing destination.

Functions

create(client, params)

@spec create(Fivetrex.Client.t(), map()) ::
  {:ok, Fivetrex.Models.Destination.t()} | {:error, Fivetrex.Error.t()}

Creates a new destination.

The configuration options are highly dependent on the destination service type. See Fivetran's documentation for service-specific options.

Parameters

  • client - The Fivetrex client
  • params - A map containing:
    • :group_id - Required. The group to create the destination in.
    • :service - Required. The destination type (e.g., "snowflake", "big_query").
    • :region - Required. Data processing location (e.g., "US", "EU").
    • :time_zone_offset - Required. Timezone offset as string (e.g., "-5", "+1").
    • :config - Required. Service-specific configuration.

Returns

  • {:ok, Destination.t()} - The created destination
  • {:error, Fivetrex.Error.t()} - On failure

Examples

Create a Snowflake destination:

{:ok, destination} = Fivetrex.Destinations.create(client, %{
  group_id: "group_id",
  service: "snowflake",
  region: "US",
  time_zone_offset: "-5",
  config: %{
    host: "myaccount.snowflakecomputing.com",
    port: 443,
    database: "ANALYTICS",
    auth: "PASSWORD",
    user: "FIVETRAN_USER",
    password: System.get_env("SNOWFLAKE_PASSWORD")
  }
})

Create a BigQuery destination:

{:ok, destination} = Fivetrex.Destinations.create(client, %{
  group_id: "group_id",
  service: "big_query",
  region: "US",
  time_zone_offset: "-8",
  config: %{
    project_id: "my-gcp-project",
    data_set_location: "US"
  }
})

delete(client, destination_id)

@spec delete(Fivetrex.Client.t(), String.t()) :: :ok | {:error, Fivetrex.Error.t()}

Deletes a destination.

Warning: You cannot delete a destination that has connectors. Delete all connectors first, or delete the entire group.

Parameters

  • client - The Fivetrex client
  • destination_id - The ID of the destination to delete

Returns

  • :ok - On successful deletion
  • {:error, Fivetrex.Error.t()} - On failure

Examples

:ok = Fivetrex.Destinations.delete(client, "destination_id")

get(client, destination_id)

@spec get(Fivetrex.Client.t(), String.t()) ::
  {:ok, Fivetrex.Models.Destination.t()} | {:error, Fivetrex.Error.t()}

Gets a destination by its ID.

Parameters

  • client - The Fivetrex client
  • destination_id - The unique identifier of the destination

Returns

  • {:ok, Destination.t()} - The destination
  • {:error, Fivetrex.Error.t()} - On failure

Examples

{:ok, destination} = Fivetrex.Destinations.get(client, "destination_id")
IO.puts("Service: #{destination.service}")
IO.puts("Region: #{destination.region}")

test(client, destination_id)

@spec test(Fivetrex.Client.t(), String.t()) ::
  {:ok, map()} | {:error, Fivetrex.Error.t()}

Runs connection tests for a destination.

This validates that Fivetran can connect to your destination warehouse with the provided credentials. Use this after creating or updating a destination to verify the configuration is correct.

Parameters

  • client - The Fivetrex client
  • destination_id - The ID of the destination to test

Returns

  • {:ok, map()} - Test results including:

    • "setup_status" - Overall status (e.g., "connected", "incomplete")
    • "tests" - List of individual test results

  • {:error, Fivetrex.Error.t()} - On failure

Examples

{:ok, result} = Fivetrex.Destinations.test(client, "destination_id")

case result["setup_status"] do
  "connected" ->
    IO.puts("Destination is properly configured!")

  status ->
    IO.puts("Setup status: #{status}")
    IO.inspect(result["tests"], label: "Test results")
end

update(client, destination_id, params)

@spec update(Fivetrex.Client.t(), String.t(), map()) ::
  {:ok, Fivetrex.Models.Destination.t()} | {:error, Fivetrex.Error.t()}

Updates an existing destination.

Use this to modify destination configuration, such as updating credentials or changing connection settings.

Parameters

  • client - The Fivetrex client
  • destination_id - The ID of the destination to update
  • params - A map with fields to update:
    • :region - Updated region
    • :time_zone_offset - Updated timezone offset
    • :config - Updated configuration (merged with existing)

Returns

  • {:ok, Destination.t()} - The updated destination
  • {:error, Fivetrex.Error.t()} - On failure

Examples

Update credentials:

{:ok, destination} = Fivetrex.Destinations.update(client, "destination_id", %{
  config: %{
    password: System.get_env("NEW_PASSWORD")
  }
})

Change region:

{:ok, destination} = Fivetrex.Destinations.update(client, "destination_id", %{
  region: "EU"
})