Remove duplicate geographical distance

• akash_singh

  Ten Centuries

  Points: 1089

  More actions

  September 14, 2017 at 3:23 am

  #323505

  
If Object_Id('tempdb..#TravelDistance') Is Not Null
    Drop Table #TravelDistance

  Create table #TravelDistance
(
    StartPt        Varchar(200),
    EndPt        Varchar(200),
    Distance    Int
)

  Insert Into #TravelDistance
Select 'Mumbai','Delhi',2000
Union All
Select 'Delhi','Mumbai',2000
Union All
Select 'Varanasi','Mumbai',1500
Union All
Select 'Mumbai','Varanasi',1500
Union All
Select 'Delhi','Shimla',4000
Union All
Select 'Chennai','Shimla',4000
Union All
Select 'Pune','Mumbai',4000
Union All
Select 'Mumbai','Pune',4000
Union All
Select 'Thane','Banglore',4000

  In the above table, Mumbai to Pune and Pune to Mumbai distance are appearing twice. Could you please give a query to remove such duplicate entries across the columns. Also Thane to Banglore is appearing only once so it should also come in output. 

  Query output should be :

  Mumbai    Varanasi    1500
  Delhi    Mumbai    2000
  Mumbai    Pune    4000
  Chennai    Shimla    4000
  Delhi    Shimla    4000
  Thane    Banglore    4000

• Thom A

  SSC Guru

  Points: 99133

  More actions

  September 14, 2017 at 3:57 am

  #1959385

  This is one solution. you don't need to do it in 2 CTE's, you could do it in one, but split it for readability:
  --Put the Places in Alphabetical order. So Mumbai Delhi becomes Dehli Mumbai, but Dehli Mumbai remains the same
--This creates true duplicates. Good!
WITH Alphabetical AS (
  SELECT CASE WHEN StartPt < EndPt THEN StartPt ELSE EndPt END AS LowerAlpha,
        CASE WHEN StartPt > EndPt THEN StartPt ELSE EndPt END AS UpperAlpha,
        Distance
  FROM #TravelDistance),
--Now give them Row Numbers
RNs AS (
  SELECT *,
        ROW_NUMBER() OVER (PARTITION BY LowerAlpha, UpperAlpha, Distance ORDER BY Distance) AS RN
  FROM Alphabetical)
--Finally, delete the duplicates (those with an RN > 1) Using CTE "Magic".
DELETE FROM RNs
WHERE RN > 1;

  Thom~

  Excuse my typos and sometimes awful grammar. My fingers work faster than my brain does.
  Larnu.uk

• sgmunson

  SSC Guru

  Points: 110640

  More actions

  September 14, 2017 at 6:44 am

  #1959409

  Here's another alternative:
  If OBJECT_ID(N'tempdb..#TravelDistance', N'U') IS NOT NULL
    BEGIN
    DROP TABLE #TravelDistance;
    END;

  CREATE TABLE #TravelDistance (
    StartPt varchar(200),
    EndPt varchar(200),
    Distance int
);

  CREATE NONCLUSTERED INDEX [IX_#TravelDistance_Distance_StartPt_EndPt] ON #TravelDistance
    (
    Distance ASC,
    StartPt ASC,
    EndPt ASC
    );

  CREATE NONCLUSTERED INDEX [IX_#TravelDistance_StartPt_EndPt] ON #TravelDistance
    (
    StartPt ASC,
    EndPt ASC
    )
    INCLUDE (Distance);

  CREATE NONCLUSTERED INDEX [IX_#TravelDistance_EndPt_StartPt] ON #TravelDistance
    (
    EndPt ASC,
    StartPt ASC
    )
    INCLUDE (Distance);

  INSERT INTO #TravelDistance (StartPt, EndPt, Distance)
SELECT 'Mumbai','Delhi',2000 UNION ALL
SELECT 'Delhi','Mumbai',2000 UNION ALL
SELECT 'Varanasi','Mumbai',1500 UNION ALL
SELECT 'Mumbai','Varanasi',1500 UNION ALL
SELECT 'Delhi','Shimla',4000 UNION ALL
SELECT 'Chennai','Shimla',4000 UNION ALL
SELECT 'Pune','Mumbai',4000 UNION ALL
SELECT 'Mumbai','Pune',4000 UNION ALL
SELECT 'Thane','Banglore',4000;

  SELECT TD.StartPt, TD.EndPt, TD.Distance
FROM #TravelDistance AS TD WITH (INDEX([IX_#TravelDistance_Distance_StartPt_EndPt]))
    LEFT OUTER JOIN #TravelDistance AS T2 WITH (INDEX([IX_#TravelDistance_EndPt_StartPt]))
        ON TD.StartPt = T2.EndPt
        AND TD.EndPt = T2.StartPt
        AND TD.Distance = T2.Distance
WHERE T2.Distance IS NULL
    OR T2.StartPt > TD.StartPt
ORDER BY TD.Distance, TD.StartPt, TD.EndPt;

  DROP TABLE #TravelDistance;

  Steve (aka sgmunson) 🙂 🙂 🙂
  Rent Servers for Income (picks and shovels strategy)