I have a Pipeline in Synapse with (currently) just a Copy activity, which has a SQL Server Source and an ADSL Gen 2 Parquet file Sink.
It uses a large query with about 140 columns and 20 tables.
The activity works fine with no partitioning, if very slow.
I am trying the Dynamic Range Partition option.
As per the instructions, I added ?AdfDynamicRangePartitionCondition to the WHERE clause.
In the SELECT clause, the column appears as bi.INVOICE_NO 'Billing_Item_No'
In the FROM clause, the table is PRODUCTION.dbo.BILLING_INVOICE bi
For the Partition Column Name, I have tried the following, all of which cause the activity to fail with an Invalid column name message.
- bi.INVOICE_NO (Message=Invalid column name 'bi.INVOICE_NO')
- Billing_Item_No (Message=Invalid column name 'Billing_Item_No')
- [Billing_Item_No] (Message=Invalid column name 'Billing_Item_No')
- BILLING_ITEM_NO (Message=Invalid column name 'BILLING_ITEM_NO')
- INVOICE_NO (Message=Ambiguous column name 'INVOICE_NO')
- PRODUCTION.dbo.BILLING_INVOICE.INVOICE_NO (Message=Invalid column name 'PRODUCTION.dbo.BILLING_INVOICE.INVOICE_NO')
How do I reference the partition column?
Thanks,
Johnny