Hi Christoph
All fields of the parent and child documents must match a field or
dynamicField definition in the schema. Those field definitions are the same
for all types of documents, whether it is a parent or a child. This doesn't
only mean they have to be the same fieldType, it also means you can only
make a field required if it is required for all types of documents.
However, "indications" isn't actually a field, child documents just use the
same syntax as field values. This "pseudo field" doesn't need to be defined
in the schema, there is no "child document" field type.
Thomas
Op za 18 feb. 2023 om 21:10 schreef <christoph.jaeg...@swissmedic.ch.invalid
>:
> Hello everyone
>
>
> For quite some time I am struggling with a problem of indexing a nested
> structure in Solr and defining the Schema accordingly beforehand by using
> the Schema API.
>
>
> In particular let's consider the following document:
>
> {
> "id": "doc_1",
> "title": "Am I a banana or not?",
> "indications": [
> { "id": "ind_1", "title": "Yes you are", "text": "True story" },
> { "id": "ind_2", "title": "No you aren't", "text": "Because she
> said so"},
> ]
> }
>
>
> and have two main issues
>
> 1. First I don't know how the nested fields should be defined in the
> managed-schema.xml and if the field "indications" should also be defined
> 2. Is it possible to give a different definition for the top level
> "text" field with respect to the "text" field inside each "indication"?
>
> I really hope that someone can help me with that because I am really
> struggling with it.
>
> Thank you very much already
>
> Christopn
>
