is empty include allowed?

[freddrake]

Both the 1.0 and 1.1 versions of the specification state:

The value of the include parameter MUST be a comma-separated (U+002C COMMA, “,”)
list of relationship paths. A relationship path is a dot-separated (U+002E FULL-STOP, “.”)
list of relationship names.

It is unclear whether an empty value for include represents an empty list of relationship paths or a single (invalid) relationship path.

[jelhan]

This seems to be unclear from the specification indeed. I think it should be considered as supported because otherwise a client would not be able to request to not include any resource.

The specification allows a server to include related resources by default:

An endpoint MAY return resources related to the primary data by default.

It allows a client to tell the server, which resources should be included:

An endpoint MAY also support an include request parameter to allow the client to customize which related resources should be returned.

[...]

If an endpoint supports the include parameter and a client supplies it, the server MUST NOT include unrequested resource objects in the included section of the compound document.

Using an empty include query parameter a client can tell the server to not include any related resources. Even if the server would do so by default.

For Sparse Fieldsets the meaning of an empty list is explicitly specified:

Sparse Fieldsets
[...]
The value of any fields[TYPE] parameter MUST be a comma-separated (U+002C COMMA, “,”) list that refers to the name(s) of the fields to be returned. An empty value indicates that no fields should be returned.

I think such a clarification should be added for include query parameter as well. At least if my interpretation of the specification is correct.