This seems like it might be a bad idea, but I can't figure out why:
I have a class, cXYZ, with properties A, B and C. It also has a method 'sGetData' that loads those three properties from the database, and a method 'sSaveData' which saves it back.
class cXYZ
public property A as string...
public property B as string...
public property B as string..
public sub sGetData()...
public sub sSaveData()...
end class
A webform has the following property:
private property xyz() as cXYZ
get
return session("myXYZ")
end get
set (value as cXYZ)
session("myXYZ")=value
end set
end property
And the following events:
Protected Sub Page_Load(ByVal sender As Object, ByVal e As System.EventArgs) Handles Me.Load
if not ispostback() then
xyz=new cXYZ()
end if
end sub
Protected Sub ButtonLoad_Click(ByVal sender As Object, ByVal e As System.EventArgs) Handles ButtonLoad.Click
//Can now reference the class
txtA.text=xyz.A
txtB.text=xyz.B
txtC.text=xyz.C
end sub
Protected Sub ButtonSave_Click(ByVal sender As Object, ByVal e As System.EventArgs) Handles ButtonSave.Click
//Can now reference the class
xyz.A=txtA.text
xyz.B=txtA.text
xyz.C=txtC.text
xyz.sSaveData()
end sub
I can see some overhead with serializing/deserializing for each property reference- it might be worth doing this:
Protected Sub ButtonSave_Click(ByVal sender As Object, ByVal e As System.EventArgs) Handles ButtonSave.Click
dim localxyz as cXYZ=xyz
localxyz .A=txtA.text
localxyz .B=txtA.text
localxyz .C=txtC.text
xyz=localxyz
end sub
Other than that, views on why this is good or bad? The class is not large, it maintains the form state. Webforms suck, etc is not very useful..