On, Thu Sep 27, 2007, RR4CLB wrote:
>
>
> Marcus,
>
> The menu comment is when I left click on it. They may be because
> the button does nothing.
>
> I read the documentation for programming a screen. It monitors
> light-bars, any change in colors. If a part of a screen changes it can
> read that area by setting a window for only that area. Either by text
> or color. So even the print statements if sent to only one location
> and I set a window to read just that area it would say what is
> there. Like the print statement you use, have it go to just a single
> location. But that is OK for this kind of test but can be a problem if
> normal Window conventions are not used; I.E. MSAA configuration.
Is there any online documentation or something like that regarding that
type of screen reader? It's somewhat difficult to understand for me.
I try to sum up what I understood so far:
* The screen reader tracks screen changes for certain windows
* On color changes it informs you and you can let it give you further
details about that area.
* Then it tries to read the textual information from that area in some
way and provides them using speech sythesis or a braille line
Is that correct?
>
> I still had it read only one button name when labeling the
> button. I think that is because the icon is the same. But I did
> not have success in doing the same line button. Instead of having
> down 40 I placed it across 40 to see if I could see another
> graphic symbol there and could not. Unless the other print
> statement over-wrote it. But that does not make sense.
If I get this right, it means that the reader distinguishes different
elements by using some way of image recognition. If they do not differ
by at least XY percent, they are recognized as "still the same element,
no need to notify the user"?
>
> I am not sure what is going on, but will have to get my son to
> read the new version and see what happens. Windows does want
> standard image size and if to small may not read it. O r to large
> may not read it. But something else seems to be going on here and
> if the mouse position values are not being sent back that could be
> the problem, but that to makes no sense.
I'm not sure at which point Windows tries to intercept the mouse
messages, but possibly the SDL hooks them up in a way, that Windows does
not see any need to deal with them later on.
> It just reads everything on one line with one graphic symbol at
> the beginning and 3 on the other side. Min, max, close
> The symbol on the left are the 2 buttons, or at least one of the
> 2, the first one displayed.
Once more for a better understanding: the reader splits the screen or
window into certain areas, which are read one after each other and tries
to determine what is displayed on them?
>
> Now, the 3 on the right seem to be no problem. The screen readers
> reads them, I am allowed to label each of them, and reads them
> everytime I go to that window, including the one graphic for our
> test buttons. But it just does not go to the second one, the
> screen reader doesn't. I will test it again with my son here.
What does that mean exactly? The screen reader describes what is
displayed and you can make up some label for that area, so you can
access it easier later on (by referring to the label name)?
Regards
Marcus
Attachment:
pgpdMemRw3mBG.pgp
Description: PGP signature