|
SQL Server 2014/2016, using SMO.
Want backup database 'dbName' of server 'serverName' to server folder 'backupFolder'.
1.Backup remote database, not backup local database.
2.The 'backupFolder' is on remote server, e.g. "D:\SQLBK\Mon".
public int BackupDB(string serverName, string dbName, string backupFolder)
{
Server oServer = new Server(new ServerConnection(ServerName));
Backup oBackup = new Backup();
oBackup.Action = BackupActionType.Database;
oBackup.Database = dbName;
oBackup.Initialize = false;
oBackup.Incremental = false;
oBackup.LogTruncation = BackupTruncateLogType.Truncate;
string sBackupFilename = Path.Combine(backupFolder, dbName + ".bak");
oBackup.Devices.AddDevice(sBackupFilename, DeviceType.File);
oBackup.SqlBackup(oServer);
return 0;
}
The 'backupFolder' is folder on 'serverName', not local folder.
1.If 'backupFolder' is not exist on 'serverName', the 'SqlBackup' command will cause exception on 'folder not exist'.
2.If use 'Directory.CreateDirectory(backupFolder)', then local folder will be created, not remote folder.
So I want:
1. The 'SqlBackup' command can auto create specified backup folder. How to do it?
2. If above can't be done, then how to create 'backupFolder' on 'serverName' by SMO or other method?
|
|
|
|
|
Hi, I have a report which will produce different numbers of columns depending on the client that is selected. There may be a dozen clients and the columns might be named differently so it's all completely dynamic. I'm not quite sure how to arrange this in an SSRS report or even if it is possible. Can anybody please shed light on how I can do this? Thanks
modified 25-May-17 6:21am.
|
|
|
|
|
SQL server 2014/16.
Want create a table. The table has some columns that need support:
1.Can contain multiple nulls.
2.If not null, then must be unique.
3.The columns has no relationship.
4.Not entire row is unique. Each column is unique.
How to make the CREATE TABLE command?
It seems SQL server can only support one null when using UNIQUE.
Below example is what I wanted:
CREATE TABLE UniqueTest (
col1 int,
col2 int unique null
);
INSERT INTO UniqueTest VALUES (1, 1);
-- SUCCESS
INSERT INTO UniqueTest VALUES (2, 2);
-- SUCCESS
INSERT INTO UniqueTest VALUES (3, 2);
-- FAIL
INSERT INTO UniqueTest VALUES (4, NULL);
-- SUCCESS
INSERT INTO UniqueTest VALUES (5, NULL);
-- SUCCESS
I had searched internet, lots of articles are discussed in old SQL server version, e.g. 2005/2008.
I think the SQL server 2014/2016 has a new CREATE TABLE option to meet my requirement, but I don't know.
The solution from internet would be:
create index on specific column.
But:
1.My table will has 10~30 columns that need unique and nulls. If each column create an index, then will be 10~30 indexes in a table. Is it possible?
2.Lots indexes will lower efficiency. Is it?
|
|
|
|
|
|
That question can't be replied, so I post it here.
From the answer, it is unique on 'col1+col2'. But my requirement is that col1 and col2 have their self unique constraint.
Below example is what I wanted:
CREATE TABLE UniqueTest (
col1 int,
col2 int unique null
);
INSERT INTO UniqueTest VALUES (1, 1);
-- SUCCESS
INSERT INTO UniqueTest VALUES (2, 2);
-- SUCCESS
INSERT INTO UniqueTest VALUES (3, 2);
-- FAIL
INSERT INTO UniqueTest VALUES (4, NULL);
-- SUCCESS
INSERT INTO UniqueTest VALUES (5, NULL);
-- SUCCESS
|
|
|
|
|
Filtered Indexes and IS NOT NULL - Brent Ozar Unlimited®[^]
CREATE TABLE UniqueTest (
col1 int,
col2 int null
);
CREATE UNIQUE INDEX UX_UniqueTest_col2
ON UniqueTest (col2)
WHERE col2 Is Not Null;
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Below is my concern:
1.My table will has 10~30 columns that need unique and nulls. If each column create an index, then will be 10~30 indexes in a table. Is it possible?
2.Lots indexes will lower efficiency. Is it?
|
|
|
|
|
1. Yes.
2. Lots of indexes will potentially decrease the performance of inserts and updates, but no more so that any other method of enforcing your requirements, and with less chance of making a mistake.
A table with 30 columns might be a candidate for further normalization. But you would need to examine the data to determine that, and test what effect that might have on the performance and complexity of your queries.
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Just use a trigger and move on. You have custom logic that SQL does not support.
There are two kinds of people in the world: those who can extrapolate from incomplete data.
There are only 10 types of people in the world, those who understand binary and those who don't.
|
|
|
|
|
RyanDev wrote: You have custom logic that SQL does not support.
Are you sure about that? 
A filtered unique index can get you around the problem, like this:
CREATE UNIQUE INDEX uq_UsersAllowNulls_DisplayName on dbo.UsersAllowNulls ( DisplayName )
WHERE DisplayName IS NOT NULL;
GO
"These people looked deep within my soul and assigned me a number based on the order in which I joined."
- Homer
|
|
|
|
|
Cool. 
There are two kinds of people in the world: those who can extrapolate from incomplete data.
There are only 10 types of people in the world, those who understand binary and those who don't.
|
|
|
|
|
MS Access,Oracle,MySQL both support unique-nulls on column, but MS SQL server don't.
It is confusing.
The 'CREATE UNIQUE INDEX' solution is traditional. We need a simpler, better and easier way.
So I think SQL server has new keyword option to support unique-nulls column on latest version, e.g. UNIQUE_NULLS in SQL server 2014/1026. But I don't know exact settings on CREATE TABLE. Maybe anyone knows, maybe the new option is not exist at all.
|
|
|
|
|
|
Hi, I have a small program that is for my own use that has a small database. In my program i have a DataGrid to display the information. Right now when i search i have to search via columns, what i want is to search for data contained in the column rows and only display the data searched. Below is the code i have that will return information when a "Column" is entered into the textbox "tbSearch" but I need to be able to search the column rows. I have nine columns that will contain data and i need to search those column rows.
private void button_Click(object sender, RoutedEventArgs e)
{
using (SqlConnection conn = new SqlConnection())
{
int result = 0;
string searchOut = tbSearch.Text;
conn.ConnectionString = " Data Source=(LocalDB)\\MSSQLLocalDB;Database=cloudyhamdb.mdf;Trusted_Connection=True;AttachDbFilename =|DataDirectory|cloudyhamdb.mdf; Integrated Security = True;";
SqlCommand MyCommand = new SqlCommand("INSERT INTO clouddata " + " (First_Name, Last_Name, Grid_Square, Country, State, Call_Sign, Date_Time, Mode, Power)" + " Values (@First_Name, @Last_Name, @Grid_Square, @Country, @State, @Call_Sign, @Date_Time, @Mode, @Power)", conn);
MyCommand.Parameters.Add("@First_Name", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Last_Name", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Grid_Square", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Country", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@State", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Call_Sign", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Date_Time", System.Data.SqlDbType.SmallDateTime);
MyCommand.Parameters.Add("@Power", System.Data.SqlDbType.Text);
MyCommand.Parameters.Add("@Mode", System.Data.SqlDbType.Text);
MyCommand.Parameters["@First_Name"].Value = Convert.ToString(tbFirstName.Text);
MyCommand.Parameters["@Last_Name"].Value = Convert.ToString(tbLastName.Text);
MyCommand.Parameters["@Grid_Square"].Value = Convert.ToString(tbGridSquare.Text);
MyCommand.Parameters["@Country"].Value = Convert.ToString(tbCountry.Text);
MyCommand.Parameters["@State"].Value = Convert.ToString(tbState.Text);
MyCommand.Parameters["@Call_Sign"].Value = Convert.ToString(tbCallSign.Text);
MyCommand.Parameters["@Date_Time"].Value = Convert.ToDateTime(DateTime.Now);
MyCommand.Parameters["@Power"].Value = Convert.ToString(tbPower.Text);
MyCommand.Parameters["@Mode"].Value = Convert.ToString(tbMode.Text);
conn.Open();
MyCommand.ExecuteNonQuery();
SqlDataAdapter da = new SqlDataAdapter();
da = new SqlDataAdapter("Select * FROM clouddata", conn);
DataSet ds = new DataSet();
if (tbSearch.Text.Length >= 1)
{
da.Fill(ds, "MyDataBinding");
DataRow[] returnRows = ds.Tables[0].Select("Last_Name" + searchOut);
returnRows = ds.Tables[0].Select("First_Name=" + searchOut);
result = returnRows.Length;
dataGrid2.DataContext = ds.Tables[0];
int results = MyCommand.ExecuteNonQuery();
}
conn.Close();
}
}
modified 23-May-17 13:26pm.
|
|
|
|
|
I may have missed something, but that is an INSERT statement. Surely a search would involve a SELECT from a pre-existing table. Something along the lines of
SqlCommand MyCommand = new SqlCommand("SELECT First_Name, Last_Name, Grid_Square, Country, State, Call_Sign, Date_Time, Mode, Power FROM clouddata WHERE First_Name = @First_Name", conn);
MyCommand.Parameters.AddWithValue("@First_Name", searchOut);
If the column you are going to search on can differ then you will need two textboxes or similar, one to name the column to search on and one to give the value to search on.
|
|
|
|
|
Can you provide an example for using 2 textboxes?
|
|
|
|
|
Here is an example using a ComboBox and a TextBox. I've used a ComboBox with fixed entries for the user to choose from. I don't want them to enter the column name for a few reasons: They might misspell it, they might introduce other errors and because I'm using string concatenation to get the column name (only the name, NOT the value) it provides a further protection against SQL Injection.
private void button1_Click(object sender, EventArgs e)
{
if (cmbColumns.SelectedIndex == -1)
{
MessageBox.Show("You must select a column");
return;
}
if (string.IsNullOrEmpty(tbSearch.Text))
{
MessageBox.Show("You must enter a value to search for");
return;
}
var conString = ConfigurationManager.ConnectionStrings["ConnectToDB"].ConnectionString;
using (SqlConnection conn = new SqlConnection(conString))
{
conn.Open();
using (var myCommand = new SqlCommand())
{
myCommand.Connection = conn;
var sb = new StringBuilder(
"SELECT First_Name, Last_Name, Grid_Square, Country, State, Call_Sign, Date_Time, Mode, Power ");
sb.Append("FROM clouddata WHERE ");
sb.Append(cmbColumns.SelectedItem);
sb.Append("=@searchValue");
myCommand.CommandText = sb.ToString();
myCommand.Parameters.AddWithValue("@searchValue", tbSearch.Text);
using (var adapter = new SqlDataAdapter(myCommand))
{
var myTable = new DataTable();
adapter.Fill(myTable);
dataGridView1.DataSource = myTable;
}
}
}
}
|
|
|
|
|
Team need help on this.
I have an SSRS Report:
In the matrix table:
I have a column name called [DER_CW_FLAG] which provides output in 3 columns named: CW,Other_Week,PW.
The outputs for these columns are obtained using expression "=sum(Fields!TotalApproved.Value)/SUM(Fields!Total.Value)"
I Need to calculate the difference between PW and CW in another column.
|
|
|
|
|
You can't work out that you need to create an expression in a new field that calculates the difference? Give it up and look for another career.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
Hi, I have a problem generating SQL to return a single row for a client. My table structure is the following...
Clients - ClientID, ClientName
Statuses - StatusID, StatusDescription, ClientID
StatusMappings - StatusID, ClientID
The statuses can be different depending on the client, so for example, clients table:
1, Barclays
2, Halifax
Statuses table:
1, Draft Letter, 1
2, Complete, 1
3, Draft Letter, 2
4, Awaiting Response, 2
5, Complete, 2
StatusMappings table:
1, 1 (Draft Letter for Barclays)
3, 2 (Draft Letter for Halifax)
4, 2 (Awaiting Response for Halifax)
How I want to display the data in a report (based on client selected), for example :
Client | Draft Letter | Awaiting Response | Complete |
-------------------------------------------------------------
Halifax Y Y
I'm not sure how I would best write the SQL for this. Also I would like to produce SSRS reports based on it. Can anybody please advise? Thanks
modified 22-May-17 10:30am.
|
|
|
|
|
I'm not sure about the example you are presenting ... Are the following statements true?
- The Statuses table shows the possible statuses for each Client?
- The StatusMapping table shows the actual statuses for each client.
If that is the case then with the data you presented you cannot possibly get the results because for Client "Halifax" Status 'Draft Letter' is Id 3 and there is nothing on the StatusMappings table for that Id. Similarly 'Awaiting Response' would be Id 4.
|
|
|
|
|
You're quite right sorry, I've amended the data. Thanks!
|
|
|
|
|
Assuming you get the data sorted out then something like the following will work
SELECT * FROM
(
SELECT C.ClientID, ClientName, StatusDescription,
CASE WHEN SM.StatusID IS NOT NULL THEN 'Y' ELSE 'N' END AS Actual
FROM @Clients C
LEFT OUTER JOIN @Statuses S ON C.ClientID = S.ClientID
LEFT OUTER JOIN @StatusMappings SM on SM.StatusID = S.StatusID
) qry
PIVOT
(
MAX(Actual) FOR StatusDescription in ([Draft Letter],[Awaiting Response],[Complete])
) pvt
WHERE ClientName = 'Barclays' If you want to use it in an SSRS report then put the query into a Stored Procedure, passing the Client name or id as a parameter.
If the Status Descriptions are not standard or consistent (i.e. not the ones listed) then you will have to use some dynamic sql.
There are articles here on CodeProject on all the topics above
|
|
|
|
|
Thanks CHill60, yes I will need to use dynamic SQL. I've seen various articles using cursors, XML, etc. I just wondered which was the best approach to take.
|
|
|
|
|
Cursor is a definite "No!"  (See my article Processing Loops in SQL Server[^] )
XML is a definite possibility, but the code I've shown using PIVOT is quite compact. There is an example of generating dynamic sql for a pivot in the article.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
