When inserting columns into a data table within Spotfire, the user selects the column(s) to include in the last window. However, what is actually happening behind the scenes is an Exclude Columns transformation where the columns the user hasn't selected are excluded from being inserted.
This becomes a problem when the table from which the columns are being inserted changes, i.e. columns are added to that data table in the source. Since those new columns were not present during the creation of the exclusion step, they are now included in the data table to which the columns were added. This really becomes a problem when the user inserts columns from that data table multiple times as the new columns get added multiple times to the final data table.
The solution to this would be allowing the user to explicitly select ONLY the columns to include in the final data table.
Implemented in | 7.13 |
Adding data to your analysis by inserting columns (joins) is a core part of data preparation. With Spotfire 7.13 now available you now have the alternative to add columns by using Spotfire's web client Business Author in addition to the Windows client Spotfire Analyst. This means more users will be able to do their visual data discovery in their web browser without installing Spotfire's Windows client. If you discover issues with your joins while using any of the clients you can also edit them.
A new user interface based on Spotfire's recent Add Rows UI makes it really easy to understand the different join methods through illustrations. The result is previewed before finishing the edit. The UI is also smart. For example, it shortlists and recommends id and categorical columns that share name and datatype. This makes it quicker to find the columns you are most probably looking for.
Is basically the same idea as mine (SPF-I-5225) ;-)