|
Sure, I believe in standards too. In fact, my standards that I used on the last project are actually more strict. But unless someone can come up with a valid justification, this particular standard is a bad one.
|
|
|
|
|
Wait a second! Does this error handling strategy even work? When I execute the following stored procedure, I get an error on the RAISERROR statement.
CREATE PROCEDURE dbo.spErrorHandlingTest1<br />
AS <br />
BEGIN<br />
<br />
DECLARE @Error INT<br />
DECLARE @myInt INT<br />
<br />
SELECT @myInt = 1/0<br />
<br />
SELECT @Error = @@ERROR<br />
<br />
IF @Error > 0 GOTO EH<br />
<br />
RETURN 0 <br />
EH: <br />
RAISERROR(@Error, 16, 1)<br />
RETURN @Error<br />
<br />
END
Msg 8134, Level 16, State 1, Procedure spErrorHandlingTest1, Line 8
Divide by zero error encountered.
Msg 2732, Level 16, State 1, Procedure spErrorHandlingTest1, Line 16
Error number 8134 is invalid. The number must be from 13000 through 2147483647 and it cannot be 50000.
|
|
|
|
|
I think it works as expected. Since you are passing forward the original error (8134) as message id in a custom RAISERROR, you'll have trouble. From Books Online:
msg_id
Is a user-defined error message number stored in the sys.messages catalog view using sp_addmessage. Error numbers for user-defined error messages should be greater than 50000. When msg_id is not specified, RAISERROR raises an error message with an error number of 50000.
|
|
|
|
|
You mean RAISERROR is working as expected and the calling code is incorrect, right?
|
|
|
|
|
Had to re-read my post and yes that's exactly what I meant... and yes the answer was confusing since I didn't point out what I was referring to. Sorry about that 
But the problem is that you're using reserved error numbers in custom RAISERROR call. If you change the first argument to for example 50001 you'll see it behaves differently.
Mika
|
|
|
|
|
Sorry to revisit an old thread, but I happened to stumble across an article that relates to this topic. According to Microsoft's Design Guidelines for Exceptions:
"Do not overuse catch. Exceptions should often be allowed to propagate up the call stack."
http://msdn.microsoft.com/en-us/library/ms229005.aspx
Of course, this guideline is in reference to .NET code and not SQL, but it's still good advice.
|
|
|
|
|
I'm using DAAB (Enterprise Library) for accessing database. My problem is I can't read output parameters when using DataReader. That is I can only read stored procedure's output parameter only when I use ExecuteNonQuery or ExecuteScalar functions. When I use ExecuteReader function, the output parameter is 0. Do you know how I can read output parameter when using ExecuteReader function?
here is a sample code:
var db = new SqlDatabase(ConnectionString);
string sqlCommand = "ReadItems";
DbCommand dbCommand = db.GetStoredProcCommand(sqlCommand);
db.AddInParameter(dbCommand, "@Count", SqlDbType.Int, count);
db.AddOutParameter(dbCommand, "@Total", SqlDbType.Int, sizeof(Int32));
var reader = db.ExecuteReader(dbCommand);
while (reader.Read())
{
}
int total = Convert.ToInt32(dbCommand.Parameters["@Total"].Value);
|
|
|
|
|
You need to close the reader before you can get the output parameters. No idea why, it has always been this way.
Bob
Ashfield Consultants Ltd
|
|
|
|
|
Does anyone know if it is possible to supply multiple queries and tables to exp - so that the result is a single export file ?
e.g. something like
exp USER/passwd file=db.dmp tables=a,b,c query=[different restictions on tables a, b and c]
?
thanks !
|
|
|
|
|
Sounds like Oracle.
If it is, you can specify multiple tables for exp, but only single query. When used this way, the query must be applicable to all the tables specified.
If the version of Oracle you use is 10g or higher, you should use Data pump. It's capable of exporting several tables with different queries.
Mika
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
|
Dear all ,
I establish a Transaction with update and merge replication between two SQL 2005 servers. every thing is OK and working fine. but when the publisher/subscriber is broken, i tired to rebuild the replication because both DB (at publisher and at subscriber) are still OK. The rebuilding is successful and a two way replication is rebuilt it. but the data changed at the subscriber site are not replication to the publisher only data change at the publisher is replicated to the subscriber.
How can i solve this problem, i mean replicated data from subscriber to publisher after rebuilding broken replication.
best regards and thanks
thaar
|
|
|
|
|
thanks Abdul Aleem for ur suggestion
|
|
|
|
|
Hi all,
i have arequirement to import excel sheet to postgresql using C#.
i wrote a code:
try
{
connection = new OleDbConnection("Provider=Microsoft.Jet.OLEDB.4.0;" + @"Data Source=" + txtfilename.Text + ";" + "Extended Properties=Excel 8.0;");
mycommand = new OleDbDataAdapter(@"select * from [Ranjit$]", connection);
mydataset = new DataSet();
mycommand.Fill(mydataset, "ExcelInfo");
try
{
dataGridView1.DataSource = mydataset.Tables["ExcelInfo"].DefaultView;
}
catch (Exception ex)
{
//System.Diagnostics.Debug.WriteLine(ex.GetBaseException().ToString());
//MessageBox.Show(ex.GetBaseException().ToString());
throw new Exception(ex.Message);
}
So mydataset is displaying data in datagridview. this is working fine
No i have to send data to postgresql database
For that i wrote code
NpgsqlConnection strconn = new NpgsqlConnection(@"server=localhost;user id=postgres;password=thinksoft10@;database=Test;SyncNotification=true ");
strconn.Open();
NpgsqlCommand cmd = new NpgsqlCommand("COPY \"Test\" FROM STDIN", strconn);
NpgsqlCopyIn cin = new NpgsqlCopyIn(cmd, strconn);
cin.start();
but this is copying files to the database.
So please tell me how to copy that information in the postgresql database.
in my application i am able to show the excel sheet data in datagrid.
but i am not aable to send the excel sheet data to postgresql Test database test table.
Any help is appreciated!!!
thanks in anticipation
Ranjit.balu
|
|
|
|
|
Don't cross-post. It's considered rude.
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
|
In MS SQL SERVER, I found Primary key is numeric / auto generated in so many senerios. Is primary key is numeric or autogenerated rather VARCHAR to improve the performance?
or Is there any other reason to have primary key as numeric?
ch sriniw8z
|
|
|
|
|
Autogenerated (identity) columns are used as surrogate keys. Refer to: http://en.wikipedia.org/wiki/Surrogate_key[^]
It's a good practice to use surrogate key so that changes in actual data won't affect keys (both primary and foreign keys). Also when numeric datatype is used it's a bit more efficient than character types and also typically uses less space. But the main point is that it is not derived from data.
Mika
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
|
There are usually 2 reasons. First, performance - an int is quicker than a varchar/char as a primary key. Secondly, using an identity column (autogenerated) means no duplicate key checking is required.
Hope this clarifies things for you.
Bob
Ashfield Consultants Ltd
|
|
|
|
|
Another reason is if your primary key is also a clustered index, which it is by default, the rows are physically ordered by the key. If you use a character data type you can cause the database engine to have a lot of disk IO to re-order the data on inserts. This may not be that noticeable for tables without many inserts or when they are small, but for high-activity tables, especially if they get large (millions of rows) you'd want to stay away from it. Say for example you primary key column is varchar(10) and you started with 'a' as the first item's key value and went up through 'zzzzzzzzzz'. If you insert 'a' and then inserted 'zzzzzzzzzz', they would be stored physically in that order. Anything you insert afterwards may cause the database to move the rows around on disk in order to put the primary key columns in order. I say "may" because it will depend on how full a page is whether the engine has to create a new page, move data from one page to another, etc.
I worked on a project where the primary key columns were char(32) (a GUID without the dashes) and this situation, with the clustered indexes, wreaked so much havoc on performance that on almost every frequently-used table the primary key column's index was changed from clustered to non-clustered. Of course, that design was in place before I looked at it and I would highly discourage anyone from doing that. Personally, I've never found a reason not to use either int or bigint set as identity 1, 1 for the primary key.
Keep It Simple Stupid! (KISS)
|
|
|
|
|
Good point, its so long since I, like you, have used anything other than int/bigint that I had forgotten that horror.
Bob
Ashfield Consultants Ltd
|
|
|
|
|
Hi all,
I am using Visual Studio 2008 and Sql Server 2005
I am use some CLR Programming in my windows application and it goes fine.
But I got Some Error during Transaction in my application.
Error : Transaction (Process ID 55) was deadlocked on lock resources with another process and has been chosen as the deadlock victim. Rerun the transaction.
So, What is the Problem and How Can I Solve this Type of Problem.
Thanking you very much.
Arindam Banerjee
Sr. Software Developer
Rance Computer Pvt Ltd.
Kolkata (India)
|
|
|
|
|
For deadlock descriptoin, see:
- http://en.wikipedia.org/wiki/Deadlock[^]
- Deadlock[^]
In order to prevent a deadlock you must for example ensure that operations (and lock acquisitions) are executed in the same order, everywhere. This won't prevent all situations, but the amount will be much smaller.
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
|
In your code check the exception, if it is sql exception, then instead of skipping the control, stop the process for some time and call the same functionality once again, by the time, other process will finish it's work and release the data, then you process will continue
ch sriniw8z
|
|
|
|
|
I suppose you didn't actually want to answer my post, but the original question. When answering a question (or another post), press reply on the post which you want to answer or comment.
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
|
True Dat!
Easy to implement when the DB is new and there aren't that many queries, sps, etc.
On an existing database with alot of objects, little hard to implement. Lack of fore thought will kill you later on a large database with deadlocks/delaylocks.
Any suggestions, ideas, or 'constructive criticism' are always welcome.
"There's no such thing as a stupid question, only stupid people." - Mr. Garrison
|
|
|
|
|
That's true. One basic rule is to make operations top-down in relational order. It's quite simple rule of thumb, but works in many cases.
The need to optimize rises from a bad design.
My articles[ ^]
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
