image data

[Jeanne A. E. DeVoto]

At 11:13 PM +0200 4/10/2007, Wilhelm Sanke wrote:
>I would very much like to try out to come up with a routine to
>manipulate pixels on the basis of the "text of image" property (to
>compare it with the possibilities the imagedata format offers).
>Where could I find background information for such an approach?

I warn you ahead of time that it will be ugly and you will probably 
pull out much of your hair. ;-) That being said, I think the best 
approach is to choose a format to work with, then consult the 
standards documentation for that format.

For example, you might choose to work with JPEG images. I think John 
Craig mentioned <http://www.jpeg.org> as a good URL to start with, 
although I think the Wikipedia page on JPEG is mmore informative 
<http://en.wikipedia.org/wiki/JPEG>. The text property of the image 
is in the format specified by the image's paintCompression, so for 
practical purposes you treat this property as though it were a JPEG 
file, decoding the data and then performing whatever transformation 
you need on it, then re-encoding it and putting the result into the 
image. (This is essentially what the engine is already doing when it 
displays a JPEG image, allows you to change it using paint tools or 
imageData, and saves the changed image.)
-- 
jeanne a. e. devoto ~ revolution at jaedworks.com
http://www.jaedworks.com