Need to find differences between 2 equally structured sql tables

I need to find the differences between 2 identically structured sql tables Each table is loaded from the third patch tool into the sqlserver database.

Table structure:

Issue ID-status-Who

      

The error id will not be repeated in the table, although it is not explicitly defined as a primary key

There can be additions / deletions / updates between any two tables.

What I need

Number of rows added & their details
Number of rows deleted & their details
Number of rows updates & their details

      

How to do it

1) it is better to use sql 2) or use datatables

+2


a source to share


6 answers


You can do this using two left joins and a regular update join. This will show you, in relation to TableA, which rows are added, removed and updated. Note that they can be combined into one result set.

select b.*, 'added'
from tableb b
   left outer join tablea a on b.IssueID = a.IssueID
where a.IssueID is null

select a.*, 'deleted'
from tablea a
    left outer join tableb b on a.IssueID = b.IssueID
where b.IssueID is null

select a.*, 'updated'
from tablea a
    join tableb b on a.IssueID = b.IssueID
where a.Status <> b.Status or a.Who <> b.Who

      



A note for the latter, if you need to handle null values, I think you need to customize the where clause.

If the tables are large and this is a permanent operation, you should consider adding indexes to the join columns.

+3


a source


if you don't want to copy all the contents of each table in C #, then do it in the database.

All missing TableA or TableB tables will be found here, as well as any changes:



;WITH AllPKs AS
(
SELECT ID FROM TableA
UNION ID FROM TableB
)
SELECT
    z.ID,a.*, b.*
    FROM AllPKs                z
        LEFT OUTER JOIN TableA a ON z.ID=a.ID
        LEFT OUTER JOIN Tableb b ON z.ID=b.ID
    WHERE A.ID IS NULL OR B.ID IS NULL OR a.Col1!=b.Col1 OR a.Col2!=b.Col2 OR...

      

0


a source


I've heard good things about Redgate SQL Data Compare

0


a source


To find updates , you will need to compare the value of each column with another table.

select 'updated',
   a.IssueID as IssueID_A, b.IssueID as IssueID_B
   a.Status as Status_A, b.Status as Status_B,
   a.Who as Who_A, b.Who as Who_B
from tablea a
inner join tableb b 
   on a.IssueID = b.IssueID 
where a.Status <> b.Status or a.Who <> b.Who

      

0


a source


In general, I would recommend a product like Redgate SQL Data Compare , but as one, you can use a script like this:

-- Create some tables and data for testing purposes
USE [tempdb]
SET NOCOUNT ON
GO

DROP TABLE [Issues1]
DROP TABLE [Issues2]
GO

CREATE TABLE [Issues1] ([IssueID] int, [Status] varchar(max), [Who] varchar(max))
CREATE TABLE [Issues2] ([IssueID] int, [Status] varchar(max), [Who] varchar(max))
GO

INSERT [Issues1] VALUES (1, 'aaa', 'bbb')
INSERT [Issues1] VALUES (2, 'ccc', 'ddd')
INSERT [Issues1] VALUES (3, 'eee', 'fff')
GO

INSERT [Issues2] VALUES (1, 'aaa', 'bbb')
INSERT [Issues2] VALUES (3, 'ggg', 'hhh')
INSERT [Issues2] VALUES (4, 'iii', 'iii')
GO

-- **** START OF ANSWER PROPER ****

-- Create some temporary variables to store the change details
DECLARE @Inserts TABLE ([IssueID] int, [Status] varchar(max), [Who] varchar(max))
DECLARE @Updates TABLE ([IssueID] int, [OldStatus] varchar(max), [NewStatus] varchar(max), [OldWho] varchar(max), [NewWho] varchar(max))
DECLARE @Deletes TABLE ([IssueID] int, [Status] varchar(max), [Who] varchar(max))

-- Find all rows that exist in Issues2 but do not exist in Issues1
-- (matching on ID)
INSERT @Inserts
SELECT *
FROM [Issues2]
WHERE [IssueID] NOT IN
(
    SELECT
        [IssueID]
    FROM [Issues1]
)

-- Find all rows existing in both Issues1 and Issues2 (matching on ID)
-- and where either Status or Who has changed
INSERT @Updates
SELECT
    [Issues1].[IssueID],
    [Issues1].[Status],
    [Issues2].[Status],
    [Issues1].[Who],
    [Issues2].[Who]
FROM [Issues1]
    INNER JOIN [Issues2] ON [Issues2].[IssueID] = [Issues1].[IssueID]
        AND
        (
            [Issues2].[Status] != [Issues1].[Status]
                OR [Issues2].[Who] != [Issues1].[Who]
        )

-- Find all rows that exist in Issues1 but do not exist in Issues2
-- (matching on ID)
INSERT @Deletes
SELECT *
FROM [Issues1]
WHERE [IssueID] NOT IN
(
    SELECT
        [IssueID]
    FROM [Issues2]
)

-- Output the results
SELECT
    (SELECT COUNT(*) FROM @Inserts) AS [Number Inserted],
    (SELECT COUNT(*) FROM @Updates) AS [Number Updated],
    (SELECT COUNT(*) FROM @Deletes) AS [Number Deleted]

SELECT 'INSERTED', * FROM @Inserts
SELECT 'UPDATED', * FROM @Updates
SELECT 'DELETED', * FROM @Deletes
GO

      

0


a source


If you want to avoid listing all columns for comparison, and if you want to include a record for identical rows, you can do the following:

--Data setup
d r o p table tableA;
d r o p table tableB;

create table tableA as (
   select rownum-1 ID, chr(rownum-1+70) bb, chr(rownum-1+100) cc 
      from dual connect by rownum<=4
);

create table tableB as (
   select rownum ID, chr(rownum+70) data1, chr(rownum+100) cc from dual
   UNION ALL
   select rownum+2 ID, chr(rownum+70) data1, chr(rownum+100) cc 
      from dual connect by rownum<=3
);

      

~

--View Tables.
select * from tableA;
select * from tableB;

      

~

--Solution.
with UnionedRows As
(
   select * from tableA 
   UNION 
   select * from tableB
)
select ID, sum(MyCount),  
   case 
      when sum(MyCount) = 12 then 'In Table A and Table B - Identical.'
      when sum(MyCount) =  2 then 'In Table A.'
      when sum(MyCount) = 13 then 'In Table A and Table B - Different.'
      when sum(MyCount) = 11 then 'In Table B.'
   end Status
from
(
   select ID, count(*) MyCount from UnionedRows group by ID
   UNION ALL
   select ID, 1 from tableA
   UNION ALL 
   select ID, 10 from tableB
) group by ID order by ID;

      

0


a source







All Articles