Omitting Endpoints

Whenever you define a new service in your Templefile, the following 4 endpoints are automatically generated:

  1. CREATE: Create a new object with the provided attributes
  2. READ: Read an existing object for a given ID
  3. UPDATE: Update an existing object with the new provided attributes, using the ID
  4. DELETE: Delete an object from the datastore, using the ID

Sometimes you might not want a particular endpoint at all, or might want to define your own custom implementation, as is shown in the Adding Endpoints guide. To remove the endpoint from the generated code, you can include additional metadata within your Templefile.

Removing specific endpoints

In the example from the Getting Started guide, we defined an ExampleService as follows:

ExampleService: service {
  foo: string;
  bar: int;
}

If we decide that we don't want the DELETE endpoint any more, we can simply omit it:

ExampleService: service {
  foo: string;
  bar: int;
  #omit([delete]);
}

The argument to the #omit metadata tag is a list of endpoints, from the options create, read, update or delete, meaning you are able to exclude multiple endpoints in a single command.

Furthermore, we allow the outer parentheses to be dropped, if you find this to be more readable: 

ExampleService: service {
  foo: string;
  bar: int;
  #omit[update, delete];
}

In addition to the CRUD endpoints that we generate, you are able to define your own endpoints for enumerating though multiple stored items, in a single query. More information about these list endpoints can be found in the Enumeration guide.

Finally, an additional endpoint, identify, may be generated if your service contains the #auth metadata. Please note that with the current release, it is not possible to exclude the identify endpoint from your service. More information about the identify endpoint can be found in the Authentication guide.

Edit this page