Cannot set a property

[yosiat]

Hi,
I am trying to set a property (on CefSharp 1, Stable) and the property isn't changing it's value.

public class BoundObject
{
     public string Name { get; set; }

     public BoundObject()
     {
          Name  = "yosy";
     }
}

bound.name //=> returns "yosy"
bound.name = "othername";
bound.name //=> returns "yosy"

and after when inspecting the clr object the value is still "yosy".

[chillitom]

bound.Name

On 22 Oct 2013, at 06:15 pm, Yosy Attias notifications@github.com wrote:

Hi,
I am trying to set a property (on CefSharp 1, Stable) and the property isn't changing it's value.

public class BoundObject
{
public string Name { get; set; }

 public BoundObject()
 {
      Name  = "yosy";
 }

}
bound.name //=> returns "yosy"
bound.name = "othername";
bound.name //=> returns "yosy"
and after when inspecting the clr object the value is still "yosy".

—
Reply to this email directly or view it on GitHub.

[yosiat]

I debugged the PropertyAccessor and found that on Set the property isn't found, because the conversion from javascript name to clr name isn't correct, the value we are looking for is "name" (and not "Name").

[yosiat]

@chillitom bound.Name dosen't exist.. bound.name exist.
or should I use "bound.Name" when I set properties? if so it dosen't make any sense..

[perlun]

@yosy - this is a bug introduced by me, I believe... I fixed name conversion a while ago, so that you can use CLR names (foo.MyBar) on the CLR side and make it visible using Javascript-standard naming on the JS side (foo.myBar). However, this does not seem to be fully reliable. If you have time, please make a pull request out of this. The proper fix would be to make the property setter know that the name should be converted back... (of course, that won't cover all cases but it's at least a lot better than what we now have.)

(I've only used the property bindings to retrieve values, i.e. call the property getter, not update them.)

[yosiat]

@perlun - the fix will be changing "toClr" implementation.

BTW - a little question, it's a performance penalty to get/set the values of objects using reflection, what do you think about generating code at runtime for each javascript wrapper using expressions that will get/set methods?

[perlun]

It could be done. I don't know that we are doing that performance critical things here, so it sounds a bit overkill at first glance. 😄 If you really need optimal performance, you should perhaps not use Javascript (or even .NET) in the first place. 😉

Having that said, I won't reject PR:s coming that way. Just make sure to follow the rules discussed in #191 so it doesn't get unmaintainable. Thanks.

[perlun]

This will not be fixed with CefSharp1, but possibly CefSharp3. Leaving open for some time more.

[amaitland]

@yosiat If your still interested please try using 33.1.0-pre01 nuget package

[amaitland]

Closing due to lack of feedback from OP, re-open if required 😄