0đź‘Ť
This is very interesting question but I think the truth depends on your whole architecture, implementation and approaches you use.
Say, if you worry about the “separation of concern” you wouldn’t give a popup any access to API because its work is to show you some data as a popup, return data, and that’s it.
On the other hand, how are you handling errors? What if an error occurs when user works in popup? Where do you show error?
Another question is the usability. For example, if error occurs when you save data, if it’s done by the main window, you are going to:
-
Show the error message
-
Make user to click some button again to show the popup
-
Fix a problem and click the Save button in it.
But if you would access API right from popup you would avoid first 2 steps. Another concern is how you handle wrong data.
If you are still in the popup you easily can validate the data and cancel saving (or disable this button at all) but if it’s done after the popup is closed it may be too late.