Recent work with unit tests and the console gui strongly suggests that Leo 
could have been designed differently.

Don't panic.  I have no plans to turn Leo's internal design on its head!

*Aha*

The basis for *all* Leo guis could have been string widgets. These widgets 
could interact with the actual gui as needed.

In fact, this is pretty much how the console gui works! Caveat: the code is 
so complicated that I wouldn't be my life on it!  But the fact that all 
unit tests pass despite recently-uncovered problems with headlines strongly 
suggests that this is so.

*Gotchas?*

This sounds like a super elegant approach, but guis are never simple.  In 
particular, handling qt headlines is wretchedly difficult because the 
widgets don't produce notifications on each keystroke.  There *might* be 
workaround.  I don't know.

In any event, handling keystrokes from the "real" gui is always going to be 
fraught with gui-dependent complications.

*Summary*

Basing Leo's gui code on string widgets is a cute idea, but it does not 
seem to have any practical benefits.

I've written this post so I can forget about it ;-)

Edward

-- 
You received this message because you are subscribed to the Google Groups 
"leo-editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/leo-editor.
For more options, visit https://groups.google.com/d/optout.

Reply via email to