Clojurians Log v2

idents are more of a legacy feature that still supported, but nowadays is much better to always use the input entity data to provide context data to the query. for instance, instead of: [{[:user/id 1] [:user/name]}] you can go with: {:pathom/entity {:user/id 1} :pathom/eql [:user/name]}

(the later example is supported when you use the boundary interace)

you can also provide more complex entities (many attributes) or even many entities, some examples:

{:pathom/entity {:some-users [{:user/id 1} {:user/id 2}]
                 :complex-input {:user/first-name "Foo" :user/last-name "bar"}}
 :pathom/eql [{:some-users [:user/name]}
              {:complex-input [:user/full-name]}]}

But still {:user/id 1} cannot be {:user/id :my-specific/id} , as the boundary interface is :root that implies I cannot nest this in a larger query, correct?

not sure if I understand the question, what you mean by boundary interface is :root?

From the https://pathom3.wsscode.com/docs/eql#boundary-interface, it says “Support provision of root entity data”

yes, but you can have complex data on root entity, like on the last example I sent you here

you can make "containers" in the input data with segments, and query inside those

I guess I’m missing something “Foo” and “bar” are constants in your example…

:complex-input {:user/first-name "Foo" :user/last-name "bar"}

its an example sending any arbirtrary data, could be ids, or anything else

So then “Foo” can be :my-pathom-attribute/that-can-be-resolved and it’ll sub that value in?

yes

the entity data provides the information you know about, and you can provide anything, them the query you express what you wanna find out, considering that available input