SQL Server: Replace values in field via lookup other table - sql

Question

I am currently running into a challenge that probably has a easy solution, but somehow I am not able to come up with it.
I have Table A with two fields that are formatted as follows:
[ID] [Codes]
1 A;B
2 D
3 A;C
And table B formatted as follows:
[ID] [Codes]
A Apple
B Orange
C Pear
D Strawberry
What I would like to do is a Lookup / Replace in order to generate the following output
a.[ID] a.[Parsed_Codes]
1 Apple;Orange
2 Strawberry
3 Apple;Pear
In short I want to replace the codes in table A, with the values associated with those codes in table B.
Of course I could just write a long replace statement (in my case there are several 100 codes), but that seems like an extremely inefficient method.
Thanks!

Answer 1

Simple way is by converting the Table A [Codes] column data(csv) into separate rows.
Then join with the Table B to get the respective codes. Finally convert the rows to CSV to get result. Try this.
CREATE TABLE #tablea
([ID] INT,[Codes] VARCHAR(100))
INSERT INTO #tablea
VALUES (1,'A;B'),(2,'D' ),(3,'A;C')
CREATE TABLE #tableB
([ID] VARCHAR(100),[Codes] VARCHAR(100))
INSERT INTO #tableb
VALUES ('A','Apple'),( 'B','Orange' ),
('C','Pear'),('D','Strawberry')
SELECT a.id,
a.Codes old_code,
b.Codes Parsed_Codes
INTO #final
FROM #tableb b
JOIN (SELECT id,
codes,
Split.a.value('.', 'VARCHAR(100)') [new_Codes]
FROM (SELECT id,
[Codes],
Cast ('<M>' + Replace([Codes], ';', '</M><M>')
+ '</M>' AS XML) AS Data
FROM #tablea) AS A
CROSS APPLY Data.nodes ('/M') AS Split(a)) a
ON a.new_Codes = b.id
SELECT t1.ID,
old_code,
Stuff((SELECT '; ' + CONVERT(VARCHAR, Parsed_Codes)
FROM #final b
WHERE b.ID = t1.ID
FOR XML PATH('')), 1, 2, '')
FROM #final t1
GROUP BY t1.id,
old_code
OUTPUT
ID old_code Parsed_Codes
-- -------- ------------
1 A;B Apple; Orange
2 D Strawberry
3 A;C Apple; Pear
(Note: temp table can be avoided to avoid code confusion i used temp table)

Answer 2

First create this function.. It's used to split a delimited string, into a variable table. Then we would be able to use this function to determine the codes from the other table and return them as one string using the STUFF function.
CREATE FUNCTION [dbo].[fnSplitString]
(
#string NVarchar(MAX)
,#delimiter Char(1) = ','
)
RETURNS #t TABLE (string NVarchar(MAX))
AS
BEGIN
DECLARE #pos Int
DECLARE #piece Varchar(500)
IF RIGHT(RTRIM(#string), 1) <> #delimiter
SET #string = #string + #delimiter
SET #pos = PATINDEX('%' + #delimiter + '%', #string)
WHILE #pos <> 0
BEGIN
SET #piece = LEFT(#string, #pos - 1)
INSERT #t
SELECT #piece
SET #string = STUFF(#string, 1, #pos, '')
SET #pos = PATINDEX('%' + #delimiter + '%', #string)
END
RETURN
END
Then run the following query...
DECLARE #result TABLE
(
[ID] Int
,[CommaDelimitedCodes] Varchar(500)
)
DECLARE #codes TABLE
(
[ID] Varchar(500)
,[FullNames] Varchar(500)
)
INSERT INTO #result
SELECT 1
,'A;B'
INSERT INTO #result
SELECT 2
,'D'
INSERT INTO #result
SELECT 3
,'A;C'
INSERT INTO #codes
SELECT 'A'
,'Apple'
INSERT INTO #codes
SELECT 'B'
,'Orange'
INSERT INTO #codes
SELECT 'C'
,'Pear'
INSERT INTO #codes
SELECT 'D'
,'Strawberry'
SELECT *
,STUFF((
SELECT ', ' + [FullNames]
FROM #codes t
WHERE id IN (SELECT *
FROM dbo.fnSplitString(r.[CommaDelimitedCodes], ';'))
FOR
XML PATH('')
), 1, 2, '') AS taglist
FROM #result AS r
I created two variable tables to test - but obviously in your case you'd need to replace those with actual field names in your tables.