'group by' works good, but need another column

Question

Hi,
I am seeking for better construction for next issue, this code is not good-readable and seems I can't add another column to returned output:
;WITH q AS (
SELECT srvs.description as srv, apps.description as app
FROM srv_link_app as link LEFT JOIN servers as srvs
ON link.srv = srvs.id
LEFT JOIN apps
ON link.app = apps.id
SELECT
FROM (
SELECT DISTINCT app
FROM q
) qo CROSS APPLY
SELECT CASE ROW_NUMBER() OVER(ORDER BY srv) WHEN 1 THEN '' ELSE '; ' END + qi.srv
FROM q qi
WHERE qi.app = qo.app
ORDER BY
srv
FOR XML PATH ('')
) qi(srv)
Works perfect, It connects 3 tables: apps, servers, link and groupy by column servers. So,
link table is auxiliary connection.
Output is:
app1 | server1;server3
app16 | server6;server13
It's good but I need add another column app.long_info, for example (Output):
app1|srv1;srv2;srv5 | app.long_info here

Answer 1

Can you post DDL? What little you showed us is a mess. You have a magic generic “description” that transform from servers to applications in violation of ISO-11179 Standards, logic and data modeling. Likewise, you have the classic noob “id” that also morphs
from servers to applications. I will bet it can be a squid or an automobile, too! 
>> So, link_table is auxiliary connection. << 
Auxiliary connections?? I never head that term before and I think been around SQL for awhile :). The term “link” comes from 1970's network databases. It was a type of pointer structure. Pointers connect, references do not. We would never use it in RDBMS. We
have n:m, 1:m or n:1 relationships and the relationships have real names. Using OUTER JOINs makes no sense. 
Next, I see you hate Dr. Codd and First Normal Form (1NF). You use XML to make the code hard to port, hard to maintain an unpredictable. Good SQL programmers write and do not do display formatting in the query.  This is the principle of any tiered architecture,
not just SQL. 
Here is my guess, with clean up: 
CREATE TABLE Applications
(app_id CHAR(12) NOT NULL PRIMARY KEY,
 app_description VARCHAR(25) NOT NULL,
CREATE TABLE Servers
(server_id CHAR(12) NOT NULL PRIMARY KEY,
 server_description VARCHAR(25) NOT NULL,
CREATE TABLE Application_Assignments
(app_id CHAR(12) NOT NULL 
   REFERENCES Applications
  ON DELETE CASCADE
  ON UPDATE CASCADE,
 server_id CHAR(12) NOT NULL 
   REFERENCES Servers
  ON DELETE CASCADE
  ON UPDATE CASCADE,
PRIMARY KEY (app_id, server_id),
Since we have no sample data, no sample output, we can only guess. But your terms an approach are completely wrong. Have read any book on data modeling or RDBMS? 
--CELKO-- Books in Celko Series for Morgan-Kaufmann Publishing: Analytics and OLAP in SQL / Data and Databases: Concepts in Practice Data / Measurements and Standards in SQL SQL for Smarties / SQL Programming Style / SQL Puzzles and Answers / Thinking
in Sets / Trees and Hierarchies in SQL