Object initializing with db null check

2019-09-04 07:45发布

I am trying to initialize an object with data from database return in a dataset as below:

Example Dataset

Class Pdf
    Public FileId As Integer
    Public AccountNumber As Integer
    Public DateSaved As DateTime
    Public FileName As String
    Public DateImported As DateTime

Scenerio 1 I can intialize the object like this:

Dim pdf = New Pdf With {.FileId = ds.Tables(0).Rows(i)("fileid"),
                        .AccountNumber = ds.Tables(0).Rows(i)("accountnumber"),
                        .DateSaved = ds.Tables(0).Rows(i)("datesaved"),
                        .FileName = ds.Tables(0).Rows(i)("filename"),
                        .DateImported = ds.Tables(0).Rows(i)("dateimported")
                        }

But this is not working, because column data can be null and I am not if how to do a db null check in this approach.

Then I have scenerio 2:

Dim pdf As New pdf
If Not IsDBNull(ds.Tables(0).Rows(i)("fileid")) Then
    PdfFileId = ds.Tables(0).Rows(i)("fileid")
Else
    PdfFileId = 0
End If

If Not IsDBNull(ds.Tables(0).Rows(i)("accountnumber")) Then
    pdf.AccountNumber = ds.Tables(0).Rows(i)("accountnumber")
Else
    pdf.AccountNumber = 0
End If

If Not IsDBNull(ds.Tables(0).Rows(i)("datesaved")) Then
    pdf.DateSaved = Format(ds.Tables(0).Rows(i)("datesaved"), "yyyy-MM-dd")
Else
    pdf.DateSaved = Nothing
End If

If Not IsDBNull(ds.Tables(0).Rows(i)("dateimported")) Then
    pdf.DateImported= Format(ds.Tables(0).Rows(i)("dateimported"), "yyyy-MM-dd")
Else
    pdf.DateImported= Nothing
End If

How can I do this to avoid doing so many If statements below. This way seems inefficient to me, can anyone suggest an better approach to initializing the object in scenario one or two? If the question is unclear, please do let me know, I will try and explain.

Please note this is sample data.

2条回答
霸刀☆藐视天下
2楼-- · 2019-09-04 07:51

Its not inefficient but a lot of code, so maybe there is a more readable/maintainable approach.

The If operator is one, you can also use DataRow.IsNull:

For i As Int32 = 0 To ds.Tables(0).Rows.Count - 1
    Dim row = ds.Tables(0).Rows(i)
    Dim pdf = New Pdf With {
        .FileId = If(row.IsNull("fileid"), 0, row.Field(Of Int32)("fileid")),
        .AccountNumber = If(row.IsNull("accountnumber"), 0, row.Field(Of Int32)("accountnumber")),
        .DateSaved = If(row.IsNull("datesaved"), Date.MinValue, row.Field(Of Date)("datesaved")),
        .FileName = If(row.IsNull("filename"), "", row.Field(Of String)("filename")),
        .DateImported = If(row.IsNull("dateimported"), Date.MinValue, row.Field(Of Date)("dateimported"))
    }
Next

(changed the name of your class from Object to Pdf)

You should also set Option Strict to On, then you need to code type safe but you gain compile time safety and avoid unwanted conversions.

查看更多
成全新的幸福
3楼-- · 2019-09-04 07:53

From reading T Field<T>(this DataRow row, string columnName), I believe that there is a check for the DBNull.Value for both reference and value types, returning a default value if DBNull.Value is passed.

So you can use it instead of checking for DBNull.Value each time:

.FileName = ds.Tables(0).Rows(i).Field(Of String)("FileName")

If the value of the specified DataColumn is null and T is a reference type or nullable type, the return type will be null. The Field method will not return Value.

DataRowExtensions.Field

Since you cant use this then @TimSchmelter provided an answer which you could build upon:

.FileId = If(row.IsNull("fileid"), 0, Convert.ToInt32(row("fileid"))

查看更多
登录 后发表回答