We Keep Coding

C# with Npgsql - Can't write CLR type System.String with handler type Int32Handler

i have problem when im trying to add integer to Postgresql database. Varchars works fine, but when i use code:
var parameter = cmd.CreateParameter();
parameter.Value = Int32.Parse(x.Value.ToString());
Console.WriteLine(parameter.Value);
parameter.ParameterName = x.Key.ToString();
cmd.Parameters.Add(new NpgsqlParameter("confirmations",NpgsqlTypes.NpgsqlDbType.Integer));
for (int search=0; search != cmd.Parameters.Count; search++)
error in executing, where im using:
cmd.ExecuteNonQuery();
sounds like:
Database problem: System.InvalidCastException: Can't write CLR type System.String with handler type Int32Handler\
at lambda_method(Closure , NpgsqlTypeHandler , Object , NpgsqlLengthCache& , NpgsqlParameter )
at Npgsql.NpgsqlParameter.ValidateAndGetLength() in C:\projects\npgsql\src\Npgsql\NpgsqlParameter.cs:line 553
at Npgsql.NpgsqlCommand.ValidateParameters() in C:\projects\npgsql\src\Npgsql\NpgsqlCommand.cs:line 793
at Npgsql.NpgsqlCommand.ExecuteDbDataReader(CommandBehavior behavior, Boolean async, CancellationToken cancellationToken) in C:\projects\npgsql\src\Npgsql\NpgsqlCommand.cs:line 1141
at Npgsql.NpgsqlCommand.ExecuteNonQuery(Boolean async, CancellationToken cancellationToken) in C:\projects\npgsql\src\Npgsql\NpgsqlCommand.cs:line 1042
at Npgsql.NpgsqlCommand.ExecuteNonQuery() in C:\projects\npgsql\src\Npgsql\NpgsqlCommand.cs:line 1025
at BitcoinType.DatabaseManager.MakeInsert(Dictionary`2 requestData) in /Users/kamilkostrzewski/Projects/BitcoinType/BitcoinType/Program.cs:line 261
This did not helped me.

I had this problem and I solved like this. In my database I have a column that type Integer. If you want to add data this column, You should use NpgsqlDbType.Integer datatype. But when I use NpgsqlDbType.String datatype, I get that error.
So you should change your NpgsqlDBtType. I hope it helps you.

Assume you have the following table:
create table foo (
bar integer
)
If you prepare an insert into the table as follows:
NpgsqlCommand cmd = new NpgsqlCommand("insert into foo values (:BAR)", conn);
cmd.Parameters.Add(new NpgsqlParameter("BAR", NpgsqlDbType.Integer));
It's probably pretty obvious that this will throw an error:
cmd.Parameters[0].Value = "Hello";
cmd.ExecuteNonQuery();
For the same reason this would throw an error:
insert into foo values ('Hello')
This will clearly work, as it's an integer all around:
cmd.Parameters[0].Value = 5;
But what you may not know is the following will all still work, since they are compatible datatypes:
cmd.Parameters[0].Value = 5M;
cmd.Parameters[0].Value = 5f;
cmd.Parameters[0].Value = (short)5;
cmd.Parameters[0].Value = (long)5;
However, even though PostgreSQL is tolerant enough to accept this in raw SQL:
insert into foo values ('5')
Npgsql 4 will throw the error you mentioned when you try to do this:
cmd.Parameters[0].Value = "5";
cmd.ExecuteNonQuery();
Interestingly, I'm pretty sure it worked with Npgsql 3.x, but I consider the error an improvement.
Since it is expecting a numeric datatype, the simple solution is to be sure any .Value you assign is also a numeric datatype, ideally an integer:
cmd.Parameters[0].Value = Convert.ToInt32(yourObject);
cmd.ExecuteNonQuery();
With the standard warnings about error trapping and the use of int.TryParse.
From your code above it actually looks like you might be taking something that could already exist as an integer and convert it into a string. If you do away with all that, you may find it works out of the box. In any case, the .Value property needs to be numeric.

It's difficult to understand your code sample above - you create and populate parameter, but then add a completely different instance of NpgsqlParameter to the comment.
However, the error message clearly says that you are trying to write a string as an integer. This happens when you set NpgsqlParameter.Value to a string value, but set its NpgsqlDbType to Integer. Either omit setting NpgsqlDbType altogether (Npgsql will usually infer the correct type from the value) or make sure they are properly aligned.

I was having the same issue with the following code
var utctime = DateTimeOffset.UtcNow.ToString("yyyy-MM-dd HH:mm:ss");
writer.Write(utctime, NpgsqlDbType.TimestampTz);
So basically you are converting utctime to string but while writing to database using type as Timestamp which is not a string. Therefore it throws error as the data is not compatible with the data type.
The fix is to not cast the utctime to string just use it as is. so working solution is
var utctime = DateTimeOffset.UtcNow;
writer.Write(utctime, NpgsqlDbType.TimestampTz);

Error converting int to nvarchar

This error has me very confused. I have a stored procedure with two output parameters as defined below
#agentNumber int OUTPUT,
#currentAgentNum int OUTPUT,
They are then populated using the SELECT statements below
SELECT #agentNumber = AgentNumber
FROM AgentIdentification
WHERE AgentTaxId = #ssn
SELECT #currentAgentNum = AgentNumber
FROM UniqueAgentIdToAgentId
WHERE AgentId = #agentId
In the database, AgentNumber is defined in both tables, as an int. However, when I call this stored procedure in my C# code, I get a SQL exception stating:
Error converting data type int to nvarchar.
If I change the data types of the output parameters to nvarchar, the code will execute, however it will only return nothing more than the first digit of the whole number. Below is how the variables are defined in the C# code
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", "");//Output parameter - leave blank
SqlParameter outCurrentAgentNumber = new SqlParameter("#currentAgentNum", "");//Output parameter - leave blank
outNewAgentNumber.Direction = ParameterDirection.Output;
outCurrentAgentNumber.Direction = ParameterDirection.Output;
I add these parameters to a SqlCommand object, specify the appropriate database and commandType, then use .ExecuteNonQuery() to call the procedure. Again, what has me really confused is the error message stating that I'm using nvarchar as a data type, which could only (to the best of my knowledge) be referring to something on the database side. However, as I said I've double/triple checked and both AgentNumber columns are of type int.
EDIT
Changing the sqlParameter declarations to a starting value of 0 has solved this issue. I'm now running into the same problem with two other parameters.
SqlParameter outOperator = new SqlParameter("#operator", "");//Output parameter - leave blank
SqlParameter outDate = new SqlParameter("#date", "");//Output parameter - leave blank

Change
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", "");//Output parameter - leave blank
SqlParameter outCurrentAgentNumber = new SqlParameter("#currentAgentNum", "");//Output parameter - leave blank
to
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", 0);//Output parameter - leave blank
SqlParameter outCurrentAgentNumber = new SqlParameter("#currentAgentNum", 0);//Output parameter - leave blank
In your code, you're initially specifying the value as a string because you are passing it a string, which maps to a varchar when going from the CLR type to the SQL Data Type.
Once the runtime sees it as a string, it retains the same type throughout code, so if you assign a numeric (int) value, it still sees it as a string. The trick is to assign it the correct data type in the first place.

These two lines are the culprits:
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", "");
SqlParameter outCurrentAgentNumber = new SqlParameter("#currentAgentNum", "");
By specifying an empty string ("") as the second parameter, the ADO.NET runtime assumes it's a string paramter. And since you didn't specify any lengths - it probably defaults to just one character length (that's why it's returning only the first digit).
So, my recommendation would be to always explicitly define the datatype (and if it's a string - also define a length!) for your SQL parameters:
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", SqlDbType.Int);
outNewAgentNumber.Direction = ParameterDirection.Output;
outNewAgentNumber.Value = 0;
or
SqlParameter outNewAgentNumber = new SqlParameter("#agentNumber", SqlDbType.NVarChar, 50);
outNewAgentNumber.Direction = ParameterDirection.Output;
outNewAgentNumber.Value = "";

Is a cursor parameter really to be declared differently than a "regular" parameter?

OK, nobody seems to know how to solve the problem I'm having looping through a cursor/result set for storage into a List, so I'm going to break it down into pieces and try to slog through it that way. So, first of all:
I add SQL Parameters to an OracleCommand object this way (works fine):
cmd.Parameters.Add("ABCID", _ABCID);
cmd.Parameters["ABCID"].Direction = ParameterDirection.Input;
cmd.Parameters["ABCID"].DbType = DbType.String;
IOW, when I add the param, I pass the name of the parameterized portion of the SQL ("ABCID" above) and a value to give it (_ABCID is a variable that has been assigned, let's say, "42").
However, when adding a Cursor (output) param, it seems to want, not a value (such as an initialized cursor object), but simply the data type:
cmd.Parameters.Add("cur", Devart.Data.Oracle.OracleDbType.Cursor);
cmd.Parameters["cur"].Direction = ParameterDirection.Output;
(I tried both ways, and neither one works, so...?)
Verily/thus, my question is: Is this really the correct way of declaring a Cursor parameter to be outputted back for traversal/access?
I'm using the brand new version of DevArt DotConnect components (6.80.332), VS 2010, .NET 4
Updated:
Here's the code in more context:
public void PopulateCurrentUserRoles(String AUserName, List<String> ACurrentUserRoles) {
_UserName = AUserName;
String query = "select roleid from ABCrole where ABCid = :ABCID";
Devart.Data.Oracle.OracleCommand cmd = new Devart.Data.Oracle.OracleCommand(query, con);
cmd.CommandType = CommandType.Text;
int _ABCID = GetABCIDForUserName();
cmd.Parameters.Add("cur", Devart.Data.Oracle.OracleDbType.Cursor);
cmd.Parameters["cur"].Direction = ParameterDirection.Output;
cmd.Parameters.Add("ABCID", _ABCID);
cmd.Parameters["ABCID"].Direction = ParameterDirection.Input;
cmd.Parameters["ABCID"].DbType = DbType.String;
//cmd.ExecuteNonQuery(); blows up: "illegal variable name/number"
//cmd.ExecuteCursor(); " "
//cmd.ExecuteReader(); " "
Devart.Data.Oracle.OracleCursor oraCursor =
(Devart.Data.Oracle.OracleCursor)cmd.Parameters["cur"].Value;
Devart.Data.Oracle.OracleDataReader odr = oraCursor.GetDataReader(); // "Object reference not set to an instance of an object"
while (odr.Read()) {
ACurrentUserRoles.Add(odr.GetString(0));
}
}

the following is from the Oracle Data Provider for .NET Developer's Guide. yes, I know, "Devart". Nonetheless, It suggests the following:
Be careful with your parameter typing declaration.
add that cursor/output parameter to the Parameters collection before
any others.
As a long shot... my guide shows a OracleDbType.RefCursor but not a OracleDbType.Cursor. If DevArt has RefCursor, try that. In visual studio, what type does .NET think that parameter is? This question is not as dumb as I used to think.
... On the other hand, if the parameter is set as an OracleDbType.Char type by setting the OracleDbType property, the output data is returned
as an OracleString type. If both DbType and OracleDbType properties
are set before the command execution, the last setting takes affect.
. . .
"An application should not bind a value for output parameters; it is
the responsibility of ODP.NET to create the value object and populate
the OracleParameter Value property with the object. When binding by
position (default) to a function, ODP.NET expects the return value to
be bound first, before any other parameters."
EDIT:
Based on #Clay's self-answer... so there is no parameter specified for the output, rather one simply does this: OracleDataReader odr = cmd.ExecuteReader();

Straight from the horse's mouth (the DevArt folks):
_UserName = AUserName;
// From the DevArtisans:
String query = "select roleid from ABCrole where ABCid = :ABCID";
Devart.Data.Oracle.OracleCommand cmd = new Devart.Data.Oracle.OracleCommand(query, con);
cmd.CommandType = CommandType.Text;
int _ABCID = GetABCIDForUserName();
cmd.Parameters.Add("ABCID", _ABCID);
cmd.Parameters["ABCID"].Direction = ParameterDirection.Input;
cmd.Parameters["ABCID"].DbType = DbType.String;
Devart.Data.Oracle.OracleDataReader odr = cmd.ExecuteReader();
while (odr.Read()) {
ACurrentUserRoles.Add(odr.GetString(0));
}
To quote Casey and the Sonshine Banned, "That's the way, Uh huh Uh huh, I like it, Uh huh Uh huh"; actually, I can't stand that crap, but I do kind of relate to that sentiment right about now.

Issue with mdmysql.Parameters.AddWithValue for string value

When I try to insert a string value using mdmysql.Parameters.AddWithValue it generates an input format string exception. Following is the code which produces the error:
cmdmysql.Parameters.AddWithValue("#p_mode", MySqlDbType.VarChar).Value = "ccc";
I tried with varchar, string, text but nothing is working. Also if I put null in place of "ccc" then the record gets inserted into the table. The variable type for p_mode in table is varchar. What is the reason for this exception?

The AddWithValue function takes only two inputs, 1. The parametername and 2. the value
cmdmysql.Parameters.AddWithValue("#p_mode", "ccc");
If you want to use the definition of types you must use the normal Add function:
cmdmysql.Parameters.Add("#p_mode", MySqlDbType.VarChar).Value = "ccc";
I would guess that the exception comes from your second input where you parse an MySqlDbType to the function and not the acutal value you want the returned mysqlparameter to have.

Try this
cmdmysql.Parameters.AddWithValue("#p_mode","ccc");

C# - ODP.NET and ora-01475 must reparse cursor to change bind variable datatype

I'm getting ora-01475 whenever I try to insert a null value in a column (of type DateTime) after some records have already been inserted that have real date values.
I'm using the OracleParameter constructor that takes the name and the value as an object (I assume the data type is then implied from the datatype of the object), but since sometimes the value of my parameter is null, it's being set as a String, therefore throwing this error.
I don't want to use the constructor that takes the datatype explicitly because I use reflection heavily to build the OracleCommand object and its parameters.
How can I reparse the cursor (as the error suggests) if I find this situation?
Has anyone else run into this and has a solution?

Have you tried to use nullable types?
DateTime? myDate;
//Code to set myDate value...
string sql = "[your SQL]"
using (OracleCommand command = new SqlCommand(sql, cn))
{
OracleParameter param = new OracleParameter(":Name",myDate);
command.Paerameters.add(param);
command.ExecuteNonQuery();
}