Consider splitting the request into two parts.
The first API request GETs the token and saves the token into a variable.
The second API call uses that variable in the data load request. Note: I'm using two different data connections in the below example. Something like this:
First part
LIB CONNECT TO 'API_TOKEN_REFRESH';
TokenRefresh:
SQL SELECT
"token",
"__KEY_root"
FROM JSON (wrap on) "root" PK "__KEY_root";
[root]:
LOAD [token],
[__KEY_root]
RESIDENT TokenRefresh
WHERE NOT IsNull([__KEY_root]);
// Variable placeholder for autogenerated token
LET vToken = peek('token', 0, 'root');
DROP TABLE root;
Second Part where the important thing is in the 'WITH CONNECTION' part. I think there is a in-depth documentation somewhere on the Qlik site.
LIB CONNECT TO 'API_GET_DATA_REQUEST';
ExampleDataTable:
SQL SELECT
"id",
"SomeField",
"AnotherField"
FROM "SomeTable" PK "__KEY_root" )
FROM JSON (wrap on) "root" PK "__KEY_root"
WITH CONNECTION (
URL "https://api.yourservice.com/your-end-point",
QUERY "dataType" "good",
HTTPHEADER "auth" "$(vToken)"
);
// The rest of the your JSON tables structure follows here...
Hope it works for you -
Cheers
.png "Former Employee"){kind=icon}
