Selecting Distinct Rows

Or how the definition of the request can get us confused?

One of my team members asked me today to help her in a SQL query and although I hate SQL I tried to help her. She wanted to select a row by a distinct field – ColumnA and to view other columns in that row, so I started the query with the distinct keyword:

1. SELECT DISTINCT ColumnA
2. FROM SomeTable

But run into the problem that she wanted those other columns of the row which weren’t distinct.

So I Googled Distinct and got lost. After taking a coffee break I changed my search to “sql select id of first distinct” the first link was of StackOverFlow: SQL SELECT DISTINCT ID from copies ORDER BY CID

Which told me of course that I was using the wrong keyword and should have used the Group by keyword.

The end result query:

1. SELECT *
2. FROM SomeTable
3. WHERE OBJECTID IN ( SELECT MIN(OBJECTID)
4.                     FROM SomeTable
5.                     GROUP BY ColumnA)

Was what she wanted.

God bless coffee (and StackOverFlow)…

Resources:

StackOverFlow - SQL SELECT DISTINCT ID from copies ORDER BY CID