SQL - Return all Rows Where at Least one has value 'Y'

问题

My question is very similar to this previous post SQL Query, get a columns if another columns equal to x

The only difference is I am joining two tables and the previous solution does not seem to work. Basically I have two Columns once the tables have been joined. I want all rows of a name, where at least one row for that name has "Shasta" as a location. For example,

Column 1 = Name (From Table 1) Column 2 = Location (From Table 2)

Name   |  Location
-------------------
Bob    |   Shasta
Bob    |   Leaves
Sean   |   Leaves
Dylan  |   Shasta
Dylan  |   Redwood
Dylan  |   Leaves

Should Return:

Name   |   Location
--------------------
Bob    |   Shasta
Bob    |   Leaves
Dylan  |   Shasta
Dylan  |   Redwood
Dylan  |   Leaves

I tried the solution of the previous post

where x in
(
  select distinct x
  from table 1
  where y like 'Shasta'
)

Unfortunately, it only returned:

Name   |  Location
--------------------
Bob    |   Shasta
Dylan  |   Shasta

回答1:

You are looking for a WHERE EXISTS clause. As an illustration, let's assume you have the following query:

select a.Name, b.Location
from table1 a
join table2 b on a.TableBId = b.Id

You are looking to retrieve all rows in this query where there exists a row in the same query results with that Name and where Location = 'Shasta'. Therefore we can use the query as a derived table, match the Name and look for a Location = 'Shasta':

select a.Name, b.Location
from table1 a
join table2 b on a.TableBId = b.Id
where exists (
    select 1
    from
    (
        select a.Name, b.Location
        from table1 a
        join table2 b on a.TableBId = b.Id
    ) x --this is the same as the above query, as a derived table
    where x.Name = a.Name --correlate the queries by matching the Name
    and x.Location = 'Shasta' --look for the column value in question
)

Granted, you may be able to simplify this query and/or eliminate the derived table in the WHERE EXISTS clause, depending on the actual schema and what table1 and table2 represent.

回答2:

Simply use EXISTS to return a row if there exists a row with same name and location Shasta:

select name, location
from tablename t1
where exists (select 1 from tablename t2
              where t1.name = t2.name
                and t2.locaion = 'Shasta')

回答3:

I think you should do another subquery. I'll call the joinresult joinedTable as you didn't showed us the statement with which you build the table. But you can just switch the joinedTable for your statement. Then do

select * from joinedTable where name in (select name from joinedTable where location = 'Shasta');

Result is your wanted: 2x Bob, 3x Dylan

Heres a fiddle for it:

http://sqlfiddle.com/#!9/7584e/3

来源：https://stackoverflow.com/questions/30520964/sql-return-all-rows-where-at-least-one-has-value-y

标签

sql

distinct