Like on BigQuery DWH manage Json arrays structures by offering on SQL (T SQL) tables schemas , this new column type ==> RECORD

Like on BigQuery DWH manage Json arrays structures by offering on SQL (T SQL) tables schemas , this new column type ==> RECORD

According

• to deal with JSON arrays and their children columns (perhaps new arrays)
• avoid to store json data on a nvarchar column
• to deal on RAW / Staging / Bronze layers with JSON formats (from files & API results)
• to unnest / flatten after on silv
• Something more relevant the nvarchar column
• Avoiding to unnest/flatten on these tools
• er & gold layers , thanks to UNNEST SQL instructions

ETL/ ingestions tools schema mapping will appreciate that