The best way to ask my question is to present an example:
I have a view called List_Of_Active_Combined_Re
spondents_
With_Count
s
It has columns and sample data as follows:
Respondent_Email Name Password Resondent_Type Count_Of_Things
john.smith@blah.com Smith, John AAAAAAAAAA B 3
jane.doe@blah.com Doe, Jane BBBBBBBBBB B 2
jane.doe@blah.com Doe, Jane CCCCCCCCC T 7
bob.johnson@blah.com Johnson, Bob DDDDDDDDD T 1
There are never more than 2 record for the same email address, and if there are 2 records, one will be Respondent_Type B and the other will be Respondent_Type T.
I am using the following SQL Syntax:
SELECT Respondent_Email,
SUBSTRING(Respondent_Name,
CHARINDEX(',',Name) + 2, 1000) AS First_Name,
SUBSTRING(Respondent_Name,
1,CHARINDEX(',',Name) -1) AS Last_Name,
Password,
Respondent_Type,
Count_Of_Things
FROM List_Of_Active_Combined_Re
spondents_
With_Count
s
The results I get are:
john.smith@blah.com John Smith AAAAAAAAAA B 3
jane.doe@blah.com Jane Doe BBBBBBBBBB B 2
jane.doe@blah.com Jane Doe CCCCCCCCC T 7
bob.johnson@blah.com Bob Johnson DDDDDDDDD T 1
What I want is to get this:
john.smith@blah.com John Smith AAAAAAAAAA B 3
jane.doe@blah.com Jane Doe BBBBBBBBBB C 9
bob.johnson@blah.com Bob Johnson DDDDDDDDD T 1
I want only one record per email address. If that record represents a source record "doublet", then it should contain the email address, the first and last names, EITHER of the 2 passwords (I do not care which), a "C" to indicate the result was derived from a combination of B and T records in the source, and Count_Of_Things should be the combined values of the 2 source records.
All records that are singlets in the source, should come into the results just as they do now.
What should my SQL syntax be?
Start Free Trial