UDP storing: add support for categories, returns, links, ...

[soxofaan]

(from #223)

openeo-python-client/openeo/rest/udp.py

Lines 63 to 75 in 3d62acc

	def store(
	self, process_graph: Union[dict, ProcessBuilderBase], parameters: List[Union[Parameter, dict]] = None,
	public: bool = False, summary: str = None, description: str = None
	):
	process = build_process_dict(
	process_graph=process_graph, parameters=parameters,
	summary=summary, description=description
	)
	# TODO: this "public" flag is not standardized yet EP-3609, https://github.com/Open-EO/openeo-api/issues/310
	process["public"] = public
	self._connection.put(path="/process_graphs/{}".format(self.user_defined_process_id), json=process)

add support for additional UDP properties: categories, returns, examples, links

[soxofaan]

done

[m-mohr]

@soxofaan Is there a way to enrich the metadata of a process without storing it? That's what I'm looking for in code generation...

[soxofaan]

I'm not sure what you mean, do you have an example/pseudo code what you want to do?

[m-mohr]

In JS (and soon in R) I can so something like:

let builder = await connection.buildProcess();

// Set the metadata for the process
builder.description = "# Moisture Stress Index\nCalculate MSI for an area specifying co...";
builder.id = "MSI";
builder.categories = ["foo", "bar"];
builder.summary = "MSI calculation";

let loadcollection1 = builder.load_collection(...);

print(builder.toJSON());

Which allows me to print the full metadata with all the additional metadata such as id, summary, categories. I'm not sure how I can set the metadata in Python though. When I was implementing code generation you said that's not possible yet and my hope was that this issue would tackle this. If that's not the case, I can open a new one... @soxofaan

[soxofaan]

there is openeo.rest.udp.build_process_dict to simplify building a dictionary with UDP fields
e.g.

openeo-python-client/tests/rest/test_udp.py

Lines 440 to 459 in 1cc75a3

	def test_build_process_dict_from_datacube(con100):
	data = Parameter.number("data")
	cube = con100.datacube_from_process("add", x=data, y=1)
	actual = build_process_dict(
	process_graph=cube,
	process_id="increment", summary="Increment value", description="Add 1 to input.",
	parameters=[data],
	returns={"schema": {"type": "number"}}
	)
	expected = {
	"id": "increment",
	"description": "Add 1 to input.",
	"summary": "Increment value",
	"process_graph": {
	"add1": {"arguments": {"x": {"from_parameter": "data"}, "y": 1}, "process_id": "add", "result": True}
	},
	"parameters": [{"name": "data", "description": "data", "schema": {"type": "number"}}],
	"returns": {"schema": {"type": "number"}},
	}
	assert actual == expected

It requires that you already have a process graph or data cube, but I think that's in your use case

[m-mohr]

Thanks, I'll try whether I can make that work.