Group by Max

[marc_frei]

Hi,

My aim is to show only the Highest Phasen_ID by each Project. I tried Group by which i'm not succeeding to get working. Can anyone help me?

@Anonymous 

[Anonymous · Accepted solution]

You are not grouping by ONLY the Phase _ID, though. Your code says you are grouping by several other columns as well, so your grouping will return the combination of unique values of the fields you selected to group.

If you have a list of the 12 months, with each month having 30 rows for the date. If you group by month ONLY, you will get 12 unique rows. If you group by month and date, you will get 30 groups for each month. If you need to have no duplicates, group ONLY on Phasen_ID, and add an "All Rows" aggregation. You can still do your List.Max as well. Any other values you might need are in the nested tables.

--Nate

[Anonymous · Accepted solution]

You would still Group by Project, and then choose the aggregation Max for the Phazen column. Use the Group By GUI dialog.

--Nate

[marc_frei]

thanks for your response, I see the point.

But how can I achieve now, that it shows me only each Project with the highest Phasen ID? How can I access the table based on the highest Phasen_ID by Project Nr?

My target is to have the initial table, after the Pivot, but just showing the highest/max phasen_id.

[Anonymous · Accepted solution]

You are not grouping by ONLY the Phase _ID, though. Your code says you are grouping by several other columns as well, so your grouping will return the combination of unique values of the fields you selected to group.

If you have a list of the 12 months, with each month having 30 rows for the date. If you group by month ONLY, you will get 12 unique rows. If you group by month and date, you will get 30 groups for each month. If you need to have no duplicates, group ONLY on Phasen_ID, and add an "All Rows" aggregation. You can still do your List.Max as well. Any other values you might need are in the nested tables.

--Nate

[marc_frei · Accepted solution]

I could solve it with table.buffer, sort descending and then remove duplicates. 👍

[Anonymous · Accepted solution]

You would still Group by Project, and then choose the aggregation Max for the Phazen column. Use the Group By GUI dialog.

--Nate