Expand from nested table does not match with nested view

[JanSoltys]

Dear comunity, 

please help me with following issue. 
1) I have a merge with match by selected parameters. In nested table view, I see that for the row 1, all records are with VersionNo "007", which is correct. 

2) When I do Expand of any columns from the nested table, the original VersionNo (and other columns) are changed, even if that was the matching parameter from the merge in previous step. 

3) When I use Add Column with Table.Column([BMWConditions], "VersionNo"){0}, I get the result correctly. 

What am I doing wrong that I get the NOK result when expanding the nested table?
I cannot use the Table.Column([BMWConditions], "VersionNo"){0} as there are multiple columns I need to expand, and it would be time consuming. 

Thank you very much for your comments / suggestions! 
Have a great day, Jan

[JanSoltys]

Hello All, here is updated screenshots, I just filtered one row only, for better understanding. 
1) I have a merge with match by selected parameters. In nested table view, I see that for the row 1, all records are with VersionNo "005", which is correct. 

2) When I do Expand of any columns from the nested table (for example the VersionNo), the original VersionNo (and other columns) are changed, even if that was the matching parameter from the merge in previous step. 

[v-tejrama · Accepted solution]

Hi @JanSoltys ,

Thanks for reaching out to the Microsoft fabric community forum.

The behavior you are observing in Power Query is expected. When merging tables, the result is a nested table column, which retains all matching rows within that nested table. If you expand this column, Power Query duplicates the parent row for each matching row in the child table, which may appear as if the original values are changing, but they are actually being repeated for each match. Using expressions like Table.Column([BMWConditions], "VersionNo"){0} or Table.FirstValue([BMWConditions][VersionNo]) retrieves a single value directly from the nested table without expansion. Therefore, the merge itself is functioning correctly; the difference lies in how expansion is handled. If you want to bring in all matching rows, expanding is appropriate. If only one value is needed, extracting it directly is the best approach. Alternatively, you can expand and then use Group By to return to one row per key if needed.

Best Regards,

Tejaswi.

Community Support