NoSQL to SQL adapter design pattern

I have 2 existing applications that I'd like to bridge somehow. Both have similar domains (Product Catalogs), but the first application uses a NoSQL document store for records, and the 2nd application uses SQL tables.

An example record from the first table looks something like:

{
  "productId": 123,
  "sku": "abc",
  "packageSizes": {
    "container": 20,
    "pallet": 50
  }
}

Whereas the same item in the 2nd domain would be 1 row in the ProductItem table:

| id  | sku |
| 123 | abc |

Then 2 rows in the ProductPackageSizes table:

| productId | type      | size |
| 123       | container | 20   |
| 123       | pallet    | 50   | 

The systems currently are completely independent, but I'd like it so that whenever a record is created in the NoSQL application to have the same item created in the SQL application.

I can write a one off script for this, that just creates it procedurally based on what the data looks like currently. However, I would be interested to know if there are any established design patterns to describe such transformations? Particularly if there are new packageSizes or other relations added in the future.