I'm trying to convert a very large JSON file to JSON-LD. When I import the example code below into GraphDB, db:metadata
becomes a property/predicate connecting each GeneSet:setX
node with an automatically created blank node _:bX
(gray in figure), from which the properties set_name
and set_id
then point to their corresponding values.
{
"@context":{
"xsd": "http://www.w3.org/2001/XMLSchema#",
"db": "http://example.com/setDB#",
"_id": "@id",
"GeneSet": {"@id": "db:GeneSet"},
"data": {"@id": "db:metadata"},
"set_name": {
"@id": "db:set_name",
"@type": "xsd:string"
},
"set_id": {
"@id": "db:set_id",
"@type": "xsd:string"
}
},
"@graph": [
{
"_id": "GeneSet:set1",
"data": {
"set_name": "set1",
"set_id": "M123"}},
{
"_id": "GeneSet:set2",
"data": {
"set_name": "set2",
"set_id": "M456"}},
{
"_id": "GeneSet:set3",
"data": {
"set_name": "set3",
"set_id": "M789"}
}]
}
I would like to convert these blank nodes into actual URIs, preferably with the unique form db:setX/data
, but the more generic db:data
could also work. I know I can achieve the former by adding "@id": "db:data/setX"
within the curly brackets of each "data": {...}
definition, but since my complete data has over 100,000 of these GeneSets I would like to achieve this by simply modifying the @context
.
Is this possible?
The sort of "templated" ID generation your after is (deliberately) not possible via the JSON-LD
@context
mechanism. The@context
is for mapping tree data into graph data, not transforming it.Consequently, looking into injecting those ID values programmatically into the JSON-LD before you process it is one option.
You might also want to ask on the JSON-LD WG's mailing list to get further options.