I have tried both Join and Join Multiple tool, and haven't find the best solution for my need yet.
I made an example to show what I am trying to achieve. I have a database looks like the following
| Date | spec | value |
| 1-Jul | a | 10 |
| 1-Jul | b | 20 |
| 1-Jul | c | 30 |
| 2-Jul | b | 40 |
| 2-Jul | c | 50 |
| 2-Jul | d | 60 |
I am trying to compare the value difference between the two dates according to spec. Basically I want to create a table looks like the following:
| Date1 | spec | value1 | date2 | value2 |
| 1-Jul | a | 10 | null | null |
| 1-Jul | b | 20 | 2-Jul | 40 |
| 1-Jul | c | 30 | 2-Jul | 50 |
| null | d | null | 2-Jul | 60 |
Obviously I need a full outer join, and I will need to rename the columns except the spec. I have tried both the Join tool and multiple join tool. and they both have some limitation to achieve my desired result.
For the Join tool, it will only rename the columns for the J output, but not for L and R output, which makes the following union result look strange (I need to union the L, J, R output), it has misplaced date column in the end of the columns.
For the Join Multiple tool, it will either omit the left input or the right input when it can't find the match, which means my spec value is gone.
attached is my workflow.
