DATA COMPARE Question??

I have two tables as:
table1
tablename tablecount
a 1
b 2
c 3
d 4
.
.
table2
tablename tablecount
a 1
b 2
c 10
d 11
.
.
.
How do I write a SP to do a COMPARE and give me the table names that are
different along with count differences.
Thanks,
TomdThis can basically be done by joining table1 with table2 and selecting where
tablecount <> tablecount.
select table1.tablename, table1.tablecount - table2.tablecount as difference
from table1
join table2 on table2.tablename = table1.tablename
where table1.tablecount <> table2.tablecount
"tom d" <tomd@.discussions.microsoft.com> wrote in message
news:C2BB644B-1849-45AF-A72A-38A5E105F8F9@.microsoft.com...
>I have two tables as:
> table1
> tablename tablecount
> a 1
> b 2
> c 3
> d 4
> .
> .
> table2
> tablename tablecount
> a 1
> b 2
> c 10
> d 11
> .
> .
> .
> How do I write a SP to do a COMPARE and give me the table names that are
> different along with count differences.
> Thanks,
> Tomd
>|||Tom,
Try:
SELECT A.TABLENAME, A.TABLECOUNT AS 'COUNT OF TABLE1',B.TABLECOUNT AS
'COUNT OF TABLE2'
FROM TABLE1 A JOIN TABLE2 B
ON A.TABLENAME = B.TABLENAME
WHERE A.TABLECOUNT <> B.TABLECOUNT
HTH
Jerry
"tom d" <tomd@.discussions.microsoft.com> wrote in message
news:C2BB644B-1849-45AF-A72A-38A5E105F8F9@.microsoft.com...
>I have two tables as:
> table1
> tablename tablecount
> a 1
> b 2
> c 3
> d 4
> .
> .
> table2
> tablename tablecount
> a 1
> b 2
> c 10
> d 11
> .
> .
> .
> How do I write a SP to do a COMPARE and give me the table names that are
> different along with count differences.
> Thanks,
> Tomd
>

Data access From sqlserver 2000

Hello

i have one prob . i m using sql server 2000 . and i have write a store procedure to fetch data from multiple table using cursor . this query exculate in sql server . but i can't fetch this data from in our page . how to fetch data from using multiple table .plz help

thnx

Hi shruram.gore,

You don't need to use cursor for this unless you need to do some operations (processess) before displaying the data.
Cursor will not help you in displying the data, to display the data use SELECT and if you need to fetch the data from more than one table, use JOIN.

Example:

SELECT a.col1, a.col2, b.col2FROM table1 aINNERJOIN table2 bon (a.id = b.id)

Good luck.

|||

thnx , for suggestion

But dont have fix table name . in my prob my table name is variable plz check the query . this query we will use to bind treeView to show parent child releation . i have execute data from this query but we cant fetch data in dataset

thanx in advance

CREATE PROCEDURE [dbo].GetRootChild
AS

SET TRANSACTION ISOLATION LEVEL READ COMMITTED
declare @.parentId int
declare @.TableName varchar (50)
declare @.RelIDName char(10),@.s int,@.sql varchar(4000)

declare comments_list CURSOR FOR
select distinct [PcRelation].[ChID],[ObjectType].[ObjName],[ObjectType].[RelIdName] from PcRelation,ObjectType where [PcRelation].[ChObjectID]=[ObjectType].[ID] and [PcRelation].[PID]='0' and [PcRelation].[PObjectID]='0'
OPEN comments_list
FETCH NEXT FROM comments_list INTO @.parentId,@.TableName,@.RelIDName
WHILE @.@.FETCH_STATUS = 0
BEGIN
DECLARE product_cursor CURSOR FOR
select @.TableName as TableName
--select * from Company
select @.sql ='select * from [dbo].['+ @.TableName +'] where ['+ @.TableName +'].['+@.RelIDName+']= '+convert(char(10),@.parentId)+''
OPEN product_cursor
FETCH NEXT FROM product_cursor
WHILE @.@.FETCH_STATUS = 0
BEGIN
--select @.TableName as TableName
exec(@.sql)
FETCH NEXT FROM product_cursor
END
CLOSE product_cursor
DEALLOCATE product_cursor
FETCH NEXT FROM comments_list INTO @.parentId,@.TableName,@.RelIDName
END
CLOSE comments_list
DEALLOCATE comments_list
GO

and fetch data from sql

if cant image refer this url

http://shriramgore.ifastnet.com/?q=node/41

thnx

|||

Hi,

Based on my understing, since you didn't do any row-based operations on the table(after you have fetched each row, seems you didn't perform any operations), there is no need for you to use cursors. If what you want is just to fetch data from multi tables, a simpleJoin will work in your case. Use dynamic query if you cannot determine your table name until run-time. Like:

declare @.str nvarchar(100) ,@.str1 varchar(100)set @.str='table_1'set @.str1='table_2'exec ('select * from '+@.str+' as a inner join '+ @.str1+' as b on a.id=b.id')

Hope my suggestion helps|||

thnx for replyBo Chen

but i have following like ......... please check example ...

I have data from multiple table, those table's Names are store as Variable in a table . in this Case i have written 1 store procedure using Cursor . this query Execute in server like (Please check this Url -http://shriramgore.ifastnet.com/?q=node/41&size=_original ) . after executing storeprocedure i get data in different tabels this data i can't fetch in single dataset .if i will fetch all tables(results of storeprocedure) then it eaiser to me for next part .

e.g
DataSet ds = new DataSet();
ds = proc.GetDataSet();

for (int i= 0; i< ds.Tables.Count; i++)
{
for (int y= 0; y< ds.Tables [i].Rows.Count; y++)
{
String str=ds.tabels[i].rows[y]["ColumnName"].toString();
}
}

in this example i enable to get other tabels (dsParent1.Tables[int] ).

please help

thnx in advance

|||

Hi shriram.gore,

I can see your problem. In my opinion, i think you can try to revise your stored procedure, like this: pass three or four parameters(as you need) into your stored procedure and make them as output paramters. In your stored procedure, you can assign them the table-name values during sql clause execution. For example:

create procedure sp_get_tbl_name (@.str1 varchar(100) output,
@.str2 varchar(100) output, @.str3 varchar(100) output...)
as
select @.str1=tablename
from yourtablename
where .....
go
select @.str2=tablename
from yoursecondtablename
where...
go
..

Hope my suggestion helps

|||

Hello ,

thnx for Reply

Yes uropinion is correct . That query iwould like to use in for loop to fetch data from multiple table . m i rt . in my case i can't use for loop in my page bcoz we have 200 record's from 5 table (Varible table - those declare in other table).if i had use for loop then those query fire 200 time . m i rt .

excuse i m not using sqlserver 2005

thnx

shriram

|||

Yes uropinion is correct . That query iwould like to use in for loop to fetch data from multiple table . m i rt . in my case i can't use for loop in my page bcoz we have 200 record's from 5 table (Varible table - those declare in other table).if i had use for loop then those query fire 200 time . m i rt .

If i understand you correct, that's not a problem.

Since now you have got those table names, you can user your sldataadpater to perform a sql data retrieval against each data table name--after that you will get server datasets, say, ds1, ds2,ds3,ds4..etc. Generally speaking each dataset you get contains only one table(it depends on your select command)

Then you can create a new dataset named ds, run : ds.tables.add(ds1.table[0]); ds.tables.add(ds2.table[0]),ds.tables.add(ds3.table[0])...etc..

Now your dataset ds contains all the tables you need. You can perform your loop on it.

BTW, what does "m i rt. " stand for??

thanks

|||

HelloBo Chen ,

thank u very much . my prob is solved .

thnx

shriram

cycling through results of a select statement

I am new to stored procedures and T-SQL so please stick with me. I have a table that holds information about companies. I am trying to write a stored procedure that when run will query that table and find out if there are more than one entry of that company. All company names in that table must be unique (they can only occur once), if they occur more than once I need to flag it for reporting.

So what is the best way to go about this? Essentially what i was thinking was doing a select * on the table and then going from the first entry to the last and at each entry running a select * from table where companyname = @.nameofcompany. @.nameofcompany would be the name for that entry. If the select statement revealed more than one entry then i would know there was a problem.

Like I said I am new and this is probably very simple but i need a little help getting started

thanksSELECT *
FROM myTable99 o
WHERE EXISTS ( SELECT Company_Name
FROM myTable99 i
WHERE o.Company_Name = i.Company_Name
GROUP BY Company_Name
HAVING COUNT(*) > 1)

But you wouldn't have to do that if you defined the table like

CREATE TABLE myTable99(Company_Name varchar(50) UNIQUE)

EDIT: Where in Jersey? And what school?|||I am originally from Vernon (Mountain Creek). Went to school at Stevens Institute of Technology in Hoboken, NJ.

I didn't create the dll for this database so i just loaded the schema by the .sql file. Right now I am handed an excel template and i wrote somce vb code to go through that excel file pull out the information i want and then write it to a text file delimited with "#" and then i load it into SQL server with a bulk load command. If the user sends me a template that already has duplicate entries in it and i try to load the data into SQL column that has a unique indentifier what will happen? Will it throw an error? If this is the case then it would probably be better to get the data in the database and then decided whether or not it is a duplicate.
Redefining the table seems like the simplest way to go but i don't want to break functionality in the process

thanks|||I just tried to add this code and it is complaining on the second line in reference to the o. Here is the error: Error 170: Line 2: incorrect syntax near 'o'. As I said i am new to stored procedures and T-SQL do i need to declare the o and i as variables somewhere?|||I didn't test the code, so you gave me a start...but the code does work...

Where are you running this from? Do you have query analyzer and the other sql server client toools?

USE Northwind
GO

SET NOCOUNT ON
CREATE TABLE myTable99 (Company_Name varchar(50))
GO

INSERT INTO myTable99(Company_Name)
SELECT 'Vernon Valley' UNION ALL
SELECT 'Mountain Creek' UNION ALL
SELECT 'Hidden Valley' UNION ALL
SELECT 'Campgaw' UNION ALL
SELECT 'Break Neck Road' UNION ALL
SELECT 'High Point' UNION ALL
SELECT 'Octogon Lounge' UNION ALL
SELECT 'Great Gorge' UNION ALL
SELECT 'Mountain Creek'
GO

SELECT *
FROM myTable99 o
WHERE EXISTS ( SELECT Company_Name
FROM myTable99 i
WHERE o.Company_Name = i.Company_Name
GROUP BY Company_Name
HAVING COUNT(*) > 1)
GO

SET NOCOUNT OFF
DROP TABLE myTable99
GO|||sorry i am an a** i have an extra space floating in there. Man i am an idiot|||Hey...you're from Jersey...never apologize