CURIE maybe not working?

[mortbauer]

I have the impression, that CURIE does actually not work, they only expand to the correct url if you have a "base_uri" specified which prefixes everything.

[wharris]

Can you give an example of CURIE expansion not working. It seems to work for me.

Here, a document is being loaded without any base_uri:

import dougrain
obj = {"_links": {"curies": [{'name': "app",
                              'href': "/rels/{rel}",
                              'templated': True}],
                  "self": {'href': "/1"},
                  "app:home": {'href': "/"},
                  "app:spam": {'href': "/main"},
                  "app:eggs": {'href': "/eggs/1"},
                  "/rels/eggs": {'href': "/eggs/2"}}}
doc = dougrain.Document.from_object(obj)
assert doc.links["/rels/spam"].url() == "/main"
assert doc.links["app:spam"].url() == "/main"
assert len(doc.links["app:eggs"]) == 2
assert doc.links["app:eggs"] == doc.links["/rels/eggs"]

[mortbauer]

If that is how it should work, then I just didn't get what a CURIE should do. I thought that:

doc.links["app:spam"].url() 

should expand to

"/rels/main"

I'm sorry for taking your time, probably should have studied the CURIE's better.

[wharris]

You can expand a CURIE from a document using the expand_curie method. In the example above, doc.expand_curie("app:spam") would return "/rels/spam".