EnforceConstraint on Datatable

On a dataset is a EnforceConstraints property indicating whether the constraints should b开发者_如何学编程e enabled. Althought a datatable also can have constraints i cannot disable the constraints for a datatable.

My case is that I have datatable which i use in memory with a uniqueconstraint on one of the columns. Sometimes i want to temporarily disable the unique constraint. How can I do this? The only thing I came up with is removing and re-adding the constraint. Is there a better way?

My solution is this

using (IDataReader reader = ExecuteReader(sql))
            {
                DataTable dt = new DataTable();
                using (DataSet ds = new DataSet() { EnforceConstraints = false })
                {
                    ds.Tables.Add(dt);
                    dt.Load(reader, LoadOption.OverwriteChanges);
                    ds.Tables.Remove(dt);
                }
                return dt;
            }

There is no public EnforceConstraints property on DataTable. I propose one of the following methods to disable constraints:

• To temporarily disable constraints use method DataTable.BeginLoadData() and then DataTable.EndLoadData() to reenable them again.
• Add that DataTable to a (dummy) DataSet and set property DataSet.EnforceConstraints to false.

Note: Disabling checking constraints also disables checking values regarding DataColumn.AllowDBNull and DataColumn.MaxLength.

Basically what i did is loop through the contrains and remove them. Do the action and re-add the constraints on the table

You can use constraints to enforce restrictions on the data in a DataTable, in order to maintain the integrity of the data. Constraints are enforced when the System.Data.DataSet.EnforceConstraints property of the DataSet is true.

There are two kinds of constraints in ADO.NET: the ForeignKeyConstraint and the UniqueConstraint. By default, both constraints are created automatically when you create a relationship between two or more tables by adding a DataRelation to the DataSet. However, you can disable this behavior by specifying createConstraints = fals when creating the relation.

This exceptions mostly occurs when you select key columns i.e. in a JOIN and the resulting table has more than one of the same value in the resulting table. Imagine you have a customer table with a CustID-PK and you join this table with an address table where the relation is 1..n. The customer can have more than one address. You end up with a resulting table where CustID is repeated for all adresses. The datatable loads the source schema and sees CustID is a PK, so it's unique, but the resulting table from the JOIN-command has a repetition of that value in the CustID-column. Then the exception occurs.

If you want handle and collect the error:

        using var tbl = new DataTable();
        using var reader = cmd.ExecuteReader();

        tbl.BeginLoadData();

        try
        {
            tbl.Load(reader);
        }
        catch (ConstraintException ex)
        {
            var sb = new StringBuilder();

            foreach (var row in tbl.GetErrors())
            {
                sb.AppendLine(row.RowError);

                foreach (var col in row.GetColumnsInError())
                {
                    var colError = col.ColumnName
                                   + ":" + row.GetColumnError(col);
                    sb.AppendLine(colError);
                }
            }

            reader.Close();
            tbl.Clear();
            tbl.Constraints.Clear();

            if (IgnoreErrors)
                tbl.Load(cmd.ExecuteReader());
            else
                throw new ConstraintException(sb.ToString(), ex);
        }
        finally
        {
            tbl.EndLoadData();
        }

"IgnoreErrors" is a property in your data layer or entity loader etc..

If you wanna avoid the error, rename the source column with "AS":

"SELECT A.*, C.CustID AS CustIDNew FROM Addresses AS A INNER JOIN Customer AS C ON C.CustID = A.lCustID"

Regards

If there is data in the datatable when you load it, an exception may be thrown. To play safe, clear your datatable before loading it like this:

MyDataTable.Clear();
MyDataTable.Load(MyDataReader, LoadOption.OverwriteChanges);

This worked for me :)