Hi suppose i have the following dataset:
first, last, height, phone, email,date
Test, Test1, ,,,1/1/2001
Test, Test1, ,,,1/1/2004
Test2, Test2,5 4,,,1/1/2007
Test2, Test2,5 4 ,,,1/1/2010
Test2, Test2,5 4 ,,,1/1/2003
Test3, Test3,,123456789,,1/1/2020
Test3, Test3,,123456789,,1/1/2001
Test4, Test4,,,test@test.com,1/1/2013
Test4, Test4,,,test@test.com,1/1/2014
Test4, Test4,,,test@test.com,1/1/2018
I know i can get the count by doing having count(first) > X but my goal is to get the newest record so the final result should be:
first, last, height, phone, email,date
Test, Test1, ,,,1/1/2004
Test2, Test2,5 4 ,,,1/1/2010
Test3, Test3,,123456789,,1/1/2020
Test4, Test4,,,test@test.com,1/1/2018
I'm having a hard time figuring out a query to get the wanted results.
Thanks,