Categories
- All Categories
- 75 Oracle Analytics News
- 7 Oracle Analytics Videos
- 14K Oracle Analytics Forums
- 5.2K Oracle Analytics Idea Labs
- Oracle Analytics User Groups
- 40 Oracle Analytics Trainings
- 59 Oracle Analytics Data Visualizations
- 2 Oracle Analytics Data Visualizations Challenge
- 3 Oracle Analytics Career
- 4 Oracle Analytics Industry
- Find Partners
- For Partners
Data Flow ability to change order of columns for proper use with Union Rows step
Description
Data Flow should have a step that allows to change the order of columns of a data set easily so that the data sets can be used for the Union Rows step.
Use Case and Business Need
Sometimes two data sets contain the same columns but in different order. If you want to merge these data sets using the Union Rows step you need to make sure both data sets have the same columns in the same order.
Today, you would need to either write SQL (if the source is a DB table) or play around with Select Columns or Add Columns. This is not easy to use and rather error prone.
Original Idea Number: 4a648b79fd
Comments
-
There seem to be two different ideas here. Let me know if this sounds right.
1. Allow unions to work regardless of column order as long as column names match.
2. Allow a way to reorder columns without going to a SQL statement.
Let us know if that makes sense.
0 -
Hi Bret,
Number 2 is certainly a yes.
However, I would not agree with number 1. When doing a union, I always want to have the control over the matching of columns.
How would you automatically match? Based on names and data types? What if the names are different, but the meaning is the same?Whatever I do in a data flow needs to be comprehensible. Automatic matching without a way to override the matches or without even having a way to see which columns are matched would make the flow incomprehensible.
Kind regards,
Philipp
0 -
Understood Philipp. Yes you would still have control. What I mean is that the UI could allow you to match up the columns without regard to order. Of course they would have to be able to be matched as you say -- data types, etc. There just might be two ways to do it... switch the columns around before the union or do it in the union experience.
0 -
Thanks, now I understand it. Both would be fine then.
0 -
I love both options here.
0