Accessibility - Difference between the sample map and pattern editor controls

[Diamond]

Hello Saga,
You made a comment about the sample map being a bitmap rendered control.  I thought I would continue that discussion over here.  So my question is, if the sample map is a bitmapped control and so is the pattern editor, do you have any idea what the difference between the two might be?  I was not aware of this point.  However that being the case, I find it unusual that my screen reader is able to read content from the sample map, but not the pattern editor.  As far as I know, it should not be possible for any screen reader to access text from a bitmapped control.

[Saga Musix]

The pattern editor is drawn using a custom bitmap font, while the instrument map uses GDI's text drawing, as descibed on the issue tracker. The former method cannot be understood by screen readers, but in theory the calls to GDI could be intercepted by the screen reader, so that it has an idea of where the text would be rendered.

[Diamond]

That is interesting.  I did not believe it was possible for screen readers to access text rendered as part of an image.  At least in my mind, it is similar to the difference between a screen reader accessing a document saved in some kind of textual format, as opposed to a scanned/graphical PDF document.  The latter is not possible without using OCR software to extract/convert the document into a textual format.  Mind you, some newer versions of screen readers do come with built-in OCR capabilities, but I knew that was not involved in this case since OCR of a window/control has to be initiated manually and I have always been able to read the sample map with Jaws.  Long before OCR capabilities were added to any screen reader.  Also, I have tested the Jaws OCR feature on the pattern editor just out of curiosity, but it did not work.

[Saga Musix]

Well, as said, it's easy for the screen reader to access the text in this case. In the pattern editor, it would indeed need to do some kind of OCR, but in the sample map, as the Operating System's standard APIs are being used, there is no need to do that. Of course the screen reader will still not parse the image, but it will know at which place which string was printed.