left join 2 tables into one field - sql-server

Question

I want to left join the 2 tables below (h and he) to the main table but so I only have one field [mpxn] rather than 2 [h.mpxn & he.mpxn]
Is there away to do this? there shouldn't be any instances where there is a result in both h and he tables.
SELECT distinct [Interaction ID]
,[BP Number]
,he.mpxn
FROM [DOMCustomers].[Bart].[OpenComplaints] as c
Left join DOMCustomers.Bart.MeterHealth as h on c.[BP Number]=h.[Business Partner]
Left join DOMCustomers.Bart.ENAMeterHealth as he on c.[BP Number]=he.[Business Partner]

Answer 1

Not sure if you are asking for the 2 column values to be concatenated together, then joined to the main table - or
if you are trying to combine the results of the 2 tables (h, he) into one result in one column, then join that to the table?
Anyway, here is some examples of how to achieve what I think you may be asking:
CREATE TABLE h
(
mpxn VARCHAR(100)
,[Business Partner] VARCHAR(100)
);
CREATE TABLE he
(
mpxn VARCHAR(100)
,[Business Partner] VARCHAR(100)
);
CREATE TABLE OpenComplaints
(
[Interaction ID] VARCHAR(100)
,[BP Number] VARCHAR(100)
);
INSERT INTO h VALUES ('h',1);
INSERT INTO he VALUES ('he',1);
INSERT INTO OpenComplaints VALUES ('1',1);
Now for some queries.
This one does a union, then joins to the main table:
;WITH a AS (
SELECT
h.mpxn
,h.[Business Partner]
FROM h
UNION ALL
SELECT
he.mpxn
,he.[Business Partner]
FROM he
)
SELECT
c.[Interaction ID]
,c.[BP Number]
,h.mpxn
FROM OpenComplaints c
left join a AS h on c.[BP Number]=h.[Business Partner]
[
This one concatenates the columns together, then joins:
;WITH b AS (
SELECT
h.mpxn+' '+he.mpxn AS mpxn
,h.[Business Partner]
FROM h
JOIN he ON h.[Business Partner] =he.[Business Partner] )
SELECT
c.[Interaction ID]
,c.[BP Number]
,h.mpxn
FROM OpenComplaints c
left join b AS h on c.[BP Number]=h.[Business Partner]
Here is a SQL Fiddle example

Answer 2

You could UNION the two tables together prior to the LEFT JOIN
with mpxn_cte([BP Number], mpxn) as (
select [BP Number], mpxn from DOMCustomers.Bart.MeterHealth
union
select [BP Number], mpxn from DOMCustomers.Bart.ENAMeterHealth)
select distinct c.[Interaction ID]
,c.[BP Number]
m.mpxn
from M [DOMCustomers].[Bart].[OpenComplaints] as c
left jion mpxn_cte m on c.[BP Number]=m.[BP Number];

Answer 3

With the assumption that there are never overlapping entries in MeterHealth and ENAMeterHealth this is probably the query result expected:
SELECT
opencomplaint.[Interaction ID],
opencomplaint.[BP Number],
Coalesce( health.mxpn, enahealth.mpxn, 0) AS mxpn
FROM
[DOMCustomers].[Bart].[OpenComplaints] AS opencomplaint
LEFT OUTER JOIN DOMCustomers.Bart.MeterHealth AS health
ON opencomplaint.[BP Number] = health.[Business Partner]
LEFT OUTER JOIN DOMCustomers.Bart.ENAMeterHealth AS enahealth
ON opencomplaint.[BP Number] = enahealth.[Business Partner]
GROUP BY
opencomplaint.[Interaction ID],
opencomplaint.[BP Number],
Coalesce( health.mxpn, enahealth.mpxn, 0) AS mxpn
;
Couple of points:
Using column names with spaces is counterproductive for SQL code reading
GROUP BY is always more secure than DISTINCT
Using one or two letter aliases might work up to three tables, after that and with more complicated transformations reading code becomes an exercise in "I do not like my colleagues at all"
Use semicolons to terminate SQL statements - again this makes code easier to read (caveat emptor: some tools just can not handle standard ANSI SQL after 20 years!)
Keep joins at same level on same level of indentation
The Coalesce works best with a default fallback value that instinctively tells the user that no value from the outer joins came back - 0 in this case
Coalesce introduces a first come, first serve transformation rule - best comment your business logic above the use of the function why it was written like it is