Web object identification -unexpected recognition

[Krowka]

There is a strange thing I have observed.
I can see that QTP is able to recognize WebEdit object without any properties in repository.

On the page there are two WebEdit object. I have added these WebEdit objects to the repository then removed all the properties for these elements (turned off smart identification and ordinal identifier) and QTP can find both objects.

   

Anyone can explain it?
For standard windows environment it is impossible but for web somehow yes.
This is something like a 'QTP memory cache' or some bug?

Thanks in advance!

Tomek
QTP 10

[Saket]

If the properties that QTP learns to identify the object as Assistive properties are sufficient to identify the object without mandatory properties, then if you delete mandatory properties it identifies the object without Mandatory properties, but if the assistive properties are not sufficient to identify the object without mandatory properties then it would not identify the object. eg for a winedit object property 'nativeclass' is mandatory without which QTP will not able to identify the object at AUT, in case of webedit say you delete 'html tag' then QTP can identify using html id, name etc. we can take example of browser too where if you dont define any property then also it will be able to identify browser using creation time. but it will not be the case for Window.
I have tried to explain it, hope it helps. however I am also not fully satisfied with my explanation. will back to you if I find anything else on this.