|
Thank you for your response.Its much appreciated.
The back_up table is used with some application for reporting purposes.Instead of saying back_up i should probably say another copy of live table every 2 hour.
In other words, To have two tables will have the same data as it updates every two hours.Within two hours live table will have more data until the update is done.
Thank you
|
|
|
|
|
Does the data in the production table change at all (UPDATE/DELETE)? Or is it always new data (INSERT)?
One answer is to use Red Gate's SQL Data Compare. It's pretty cheap and you can write a script to sync the tables every two hours.
You could dump/refresh the backup table.
Method 1:
This would work much like using the Red Gate tool.
1) Add a timestamp column to the livedata table and a binary column to the backupdata table
2) Compare and update existing data (UPDATE b SET column = live.column FROM backupdata b INNER JOIN livedata live ON live.ID = B.ID WHERE live.stamp <> b.stampcopy)
3) Insert all data from livedata that doesn't exist in backupdata
Method 2:
1) Create a logging table that records which records in livedata have changed and the change that was made (INSERT, UPDATE, DELETE).
2) Create a trigger on livedata would add a record to the logging table including ID, Action, ActionTime. (ActionTime could be a timestamp or datetime).
3) Write a script that reads the log table, and syncs the backupdata table on a schedule based on the ID and the action with the oldest ActionTime.
Method 3:
TRUNCATE TABLE backupdata<br />
INSERT INTO backupdata (columns)<br />
SELECT columns FROM livedata
Method 4:
CREATE TABLE backupdata_temp (<br />
columns.....<br />
)<br />
<br />
INSERT INTO backupdata_temp (columns)<br />
SELECT columns FROM livedata<br />
<br />
DROP TABLE backupdata<br />
<br />
EXEC sp_rename ......
Either way make sure you wrap the whole thing in a transaction and set the logging mode to BULK LOGGED before you do it and then change the logging mode back when you're done. Also, if you use method 3, make sure you recreate any indexes you had that were on the backup table.
The choice of methods will depend on the availability requirements for the backup table and the amount of data you're working with.
|
|
|
|
|
Hi Mark,
Thank you once again Mark.Its has been very helpfull.I am only INSERTING new records to the live table.No Update or Delete.
Which method will best fit this case?
Thank you.
|
|
|
|
|
Hi Mark,
I would like to use the First step.Does it mean that i have to create a relationship between Live Table and Backup_table.As you have mentioned INNER JOIN.
Thank you.
|
|
|
|
|
INNER JOIN does not require an explicit relationship, so no you have an implicit relationship due to the fact that backuptable will have the same primary key as livetable. You will be able to join the table using the primary key from each table. But if you are only doing inserts then there are two solutions:
1)
INSERT INTO backuptable (columns)<br />
SELECT columns FROM livetable L WHERE NOT EXISTS(SELECT * FROM backuptable WHERE ID = L.ID)<br />
2) If you are using an IDENTITY (aka AutoIncrement) column as your primary key you could also do this
INSERT INTO backuptable (columns)<br />
SELECT columns FROM livetable WHERE ID > (SELECT MAX(ID) FROM backuptable)
The second will be faster as long as there is an index on ID for both tables (which if you have created them correctly and assigned them as primary keys then there will be).
|
|
|
|
|
Hi All,
I a page in ASP.net i want this page to run every two hours with out any manual intervation.
How could i do it please?
Inshort,i want the application to execute itself every two hours.
Thank you.
|
|
|
|
|
Hi all,
I'm developing an application in Visual Studio 2005 using VB.NET. Now I would like to store the variables in a SQL database. Each member of the databse has roughly 1.000 values to store. But most of them are in arrays like this:
dim results(6, 50) as string
The VB-programme generates the values and thus I thought it would be possible to declare an array in the Database declaration like Results(index1, index2) and then index1.max = 6 and index2.max = 50, something like that.
And then I could add it like this:
myCommand.Parameters.AddWithValue("@mResults", Results)
I'm concerned for how to describe the input variables in the SQL Database. As far as I could tell it seems not possible?
I'm thinking that I will still have to give each and everyone of the 300 values a description, if it will allow nulls etc.
If I still have to do it that way then I guess a loop like this will also do the work:
for i as integer = 1 to 6
for j as integer = 1 to 50
myCommand.Parameters.AddWithValue("@mResults" & i.ToString & j.ToString, Results(i, j))
next
next
An other idea is of course to save the array as a text with delimiters between the values.
Or is there any smart way that I'm unaware of to store the arrays?
Any advice would be appriciated. I see a lot of work to first declare the 1.000 values and its equivalent code and so on. Of course, I'm planning to describe the arrays as seperate tables with the identical primary key.
Best, Per
|
|
|
|
|
There are a few ways to save arrays to SQL Server, but none of them built in (that will change for SQL 2008 with table-value parameters).
It would be easier to answer your question if I knew what your target schema is. You mention separate tables and a large number of variables, so I'm picturing (based on your array declaration) that we're looking at 6 tables with 50 columns each (plus primary key??).
You can submit a delimited array to SQL Server as a varchar parameter (varchar(max) if you're using 2005) and then write a user defined function to parse it in the database (I'd recommend writing a CLR function if you're using SQL 2005 as it would be much, much faster).
Here's a link to an article I wrote on passing arrays to SQL Server. http://www.codeproject.com/KB/database/TableValuedFnsAsArrays.aspx[^]
As for parsing the delimited string the article only addresses arrays with a single dimension. You'll need to alter the code to handle multiple dimensions, but it can be done without too much trouble. Or you can submit the data as 6 separate queries (one for each of the major dimensions).
A method not mentioned in the article is sending XML instead of a delimited string, I can't imagine it would be faster than the methods in the article, but it's an alternative nonetheless. I know I've seen an article about it here on codeproject, but I don't remember which one.
Without more information it's difficult to be more specific.
|
|
|
|
|
Dear Mark,
Many thanks for your kind help! Great. If I make an upgrade to Visual Studio 2008 then the simple handling is included.
The situation is that data in the VB-programme are stored in 10 arrays, in 2 of them it's (6, 50) while the other 8 are single columns of 50 values each. And I can of course re-organise the data in anyway to suit the SQL system.
Thanks again for your help!
Best wishes, Per
|
|
|
|
|
I've been using SQL for several production applications, but I have to say, my knowledge level of SQL is still at the beginner level, maybe just scraping the intermediate knowledge level.
I need some help from those of you who are more experienced with SQL code.
I want to take a table with several records that have identical file numbers and merge one of the fields in the table with the other records. The consolidated record can be written into another table.
Not all the records have duplicates, and I want to just write those records to the second table also.
For example the table I would be reading from would contain the following:
FileNo FlagInfo
1234 A
1234 B
1234 C
2345 A
3456 A
3456 B
The table being written to would contain the following:
FileNo FlagInfo
1234 A,B,C
2345 C
3456 A,B
Can this be done with a stored procedure or should I just write VB code to read the file in and write out the single records?
Any suggestions? 
Lost in the vast sea of .NET
|
|
|
|
|
I can think of two possibilities:
1) PIVOT if you are using SQL Server 2005
2) Use a user defined function like this:
ALTER FUNCTION dbo.NormalizeChildren<br />
( @FileNo INT )<br />
returns VARCHAR(8000)<br />
AS<br />
BEGIN<br />
DECLARE @result VARCHAR(8000)<br />
<br />
<br />
SELECT @result = ISNULL(@result, '') + CASE WHEN @result IS NULL THEN '' ELSE ',' END + FlagNo<br />
FROM FlagTable WHERE FileNo = @FileNo<br />
<br />
RETURN @result<br />
END<br />
GO<br />
<br />
--Call the function like this<br />
SELECT FileNo, dbo.NormalizeChildren(FileNo) AS Children<br />
FROM FileTable
|
|
|
|
|
Hi,
iam doing c#(desktop application)
my database is ms-access database.
how can i save and retrive images from ms-access database.
is it possible to save images in ms-access database.
murali krishna
|
|
|
|
|
Try searching the code project articles or google. Here[^] is one article that I have used for help in the past.
Keywords that may aid your search:
blob
clob
.mdb
Hope this gets you started.
|
|
|
|
|
Hi,
iam going to develop c#(desktop application).iam searching for embedded database for visual c#.
please give me the details of what are the embedded databases available for c#.net
Thanks in Advance.
murali krishna
|
|
|
|
|
Some options:
VistaDB
SQLite
Also, I think SQL Server CE can be embedded now.
|
|
|
|
|
Hello there guys.
I have a problem with an update statement.Im have an intranet site from work that I copied and tried to upload on my home machine.The problem is when I try to login....I get this error on the update quesry.
Whenever someone logs in, the database is updated to see who logged in at what time.
I get this error "Error #214767259 Operation must use an Updateable Query"
SQL = UPDATE tbl_employees SET lastlogin = "" Where employeelogin = ""
The problem was may have been write rights to write onto the database, but the files are on my Inetput folder, so it cant be that.I checked the rights and its not that.
What else could be the problem.
Thanx in advance guys.
kagiso
|
|
|
|
|
|
How to retrive the transactions back using log files
G. Satish
|
|
|
|
|
Have you looked up RESTORE LOG in BOL?
|
|
|
|
|
I have one doubt.
Can stored procedure retun bulk data through procedure itself.
I have studied in net that using OUTPUT parameters we can get multiple values from stored proceudre...is there any other way to get the data??
G. Satish
|
|
|
|
|
There are 2 ways to return data from a SP.
1) OUTPUT params
2) SELECT statements (of which you can have multiple select statements in the same sp)
|
|
|
|
|
|
SELECT distinct u.domain,u.username, count(d.downloadid) as DocumentCount
FROM primaryusertable AS u inner join tbluseraccessinfo a on(u.userid=a.userid and a.courntryid=51 )
right join download d on(u.userid = d.userid and CDate( d.downloadutctime) <now())
group by u.username, u.domain
it is not working giving syntax error.
i could not use inner and right join together in access database
|
|
|
|
|
Do you need a space between your on and your bracket?
|
|
|
|
|
SELECT DISTINCT u.domain, u.username, count(d.downloadid) AS DocumentCount
FROM primaryusertable AS u RIGHT JOIN download AS d ON u.userid=d.userid
WHERE CDate([d.downloadutctime])>= CDate(#2/2/2008#) And CDate([d.downloadutctime])<=CDate(#2/15/2008#)
GROUP BY u.username, u.domain;
by this query i am not getting date equal including
it is only > and < no =.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
