It is without doubt an excellent job.
But I speak as a Clipper user text mode.
I have no idea what would be the behavior of Windows, if this is easy or
difficult to do, if this is possible, or if final job will be slow.
Let's put object aside and think about the screen as ... a screen.
The screen with squares, which may contain text or graphics.
@ 5, 5 say "the text"
@ 6, 5 to 7,10 graphic "image.jpg"
Once the graph on the screen, become part of the screen, not to be treated
separately.
@ 5, 5 graphicprompt "image.jpg"
@ 6, 5 graphicprompt "image2.jpg"
menu to nOpc
As with text, a graphic will always be drawn on top of each other, and erase
earlier.
For the harbor, everything would be handled by Inkey () as is done for
screen text.
Just look at row / col of the mouse click. whether it is text or graphic
does not matter, just import a region where the click was made, as is done
in text.
If SaveScreen / RestoreScreen can be used to power, will be better.
This is what I imagined for an environment intermediate between console and
GUI.
No matter what's on screen, only when drawing need to know if text or
graphic.
As I said earlier, I am speaking only as a former user Clipper, on how this
was made before Windows.
Do not know if this is not possible, do not know if this will be extremely
slow, do not know if objects is the fatest way.
Perhaps the use of objects is so common nowadays that nobody think of doing
in the old way.
Escuse me if english errors, and if this text is an stupid question, but I
don't know if could be done in this way, without isolated objects.
Jos� M. C. Quintas