Prior to MySQL 5.7, I was able to dedupe queries by using GROUP BY as long as the query results were done in such a way that the duplicates had the same data and were already in the sort order I wanted.
I got really good at that.
Now, with 5.7 and only_full_group_by mode, those queries are invalid. I've already been through the 5 stages of grief over it, and am working to update / improve my queries to be more standard.
the problem is: when I do a LEFT JOIN (in some cases) it produces rows and rows of (essentially) duplicate information. I get that they are not technically duplicates because some column may be different. And, I see a lot of questions about this in a general way. One such use case is "find the highest value for X in the day." You used to just use ORDER BY x DESC with a GROUP BY and poof you had it. Now, you have to do LEFT JOINs and look for NULLs. (Part of me still wonders "why is this better?" but that's off topic).
....but I want to know what best practice replaced using GROUP BY to remove these duplicates? SELECT DISTINCT?