I used to have 'Tableau Desktop', and my company moved to 'Desktop Professional' and now to server. I have made a range of workbooks with csv as the source and now try to load the data into our PostgreSQL DB and change the workbooks to connect to them. I also prototype workbooks using csv and then likewise try to change over the datasource. With this process I have come across several problems that give me the impression that Tableau handles this switch rather poorly. I have not found direct solutions, but rather workarounds. I will describe 3 situations in the hope that there may be some generic solutions:
- Sets: I made a set out of two dimensions. This caused an error when replacing data source. The reason for the error was apparently that the datasource [while it contained the same variables] did not have the same data combinations so couldn't recognise the set. The workaround was to also create the set in the source before switching.
- Changing Variable data type: as we know Tableau automatically sets what type a variable is. From the source CSV I changed this. When I went to change to DB connection this caused a bug. Workaround was to change the data type back before switching data source.
- Filter: I have a large workbook with 14 tabs. It has several filters like the one in the attached screenshot (a) (for a numeric dimension). When I try to change datasource I get the error (screenshot b). I have not yet resolved this, but I suspect that I will have to remove all the filters, change datasource, and then add them back in, this is really frustrating as there is 5-10 such errors.
Has anyone else experienced these sort of limitations when changing datasource? Does this only happen in my unique case with csv to PostgreSQL? Any help here would be really appreciated. It's difficult to post workbooks as they contain confidential data.
Looking forward to your replies.