I have as source a simple json dictionary like this that I want to process with Azure Data Factory. In this example there are two objects with dynamic names: Type1 and Type2, but in the real case there could be much more
{
"TYPE": {
"Type1": {
"ADD": 5,
"UPDATE": 0,
"REMOVE": 2
},
"Type2": {
"ADD": 2,
"UPDATE": 2,
"REMOVE": 1
}
}
}
With an Azure Data Flow transformation, I would like to obtain a tabular structure like this:
| TYPE | ADD | UPDATE | REMOVE |
|---|---|---|---|
| Type1 | 5 | 0 | 2 |
| Type2 | 2 | 2 | 1 |
Parse or Flatten transformation doesn't seem to fit in this case.
I Tried with a derived column, using this expression:
byPath('TYPE.Type1')
This brings a partial result, it works only for 1 object. What I really need is to obtain the properties for all the objects without knowing in advance their names

Maybe if there is a grouping function to sum up the properties Add, Update and Remove of all the objects to obtain something like this, this will solve my issue as well.
| ADD | UPDATE | REMOVE |
|---|---|---|
| 7 | 2 | 3 |
Thanks in advance for any answer.