What did you suggest?
Enlarge your Firefox/Seamonkey window over 2 or more XPager pages, the more the better.
Enable screen capture of program window in PMView and activate it.
Here it flip around and trigger all pages in sequence that the window span over.
I'd be quite surprised if PMView is responsible for moving the Firefox main window as well hide/show all other open windows to achieve it! Calling/Triggering XPager is therefore more likely.
Or are you trying to say that you can capture all desktops, but that cropping and merging is the problem?
Yes, merging multiple captured images is the current problem now as I've adjusted the code to move the viewport around on the active desktop until it has captured all required parts. It doesn't handle other application windows that is in the way.
It use the full size needed for the desired window and move it around. The result is that each image will contain black area(s) and one part that was visible for that specific move. If one combine the non black parts, exclude the black area(s) that is, of the images, one get the complete one.
I don't know what you mean by "exclude black areas"? Surely you don't want me to write yours (no GCC DLLs, 1.5 supported different bitmap file formats, exactly 1 desktop size-related API in use), and I'd consider using NCONVERT or tools like that. If you cannot calculate which part of each bitmaps you need, then features like its crop or autocrop (cropping) and watermark (merging) may be helpful somehow. And I was just saying that ePager isn't involved, a.o. based on the desktops' background.
My Rexx image support dll
"rxImgSze" rely on gbm and handle enough file formats. Just need to add C/C++ code to it to either grab what I want somehow or stitch parts (see attached example images) together.
It would be a lot better if one could get the full image as I have the window handle and not just the visible portion.
How can one print or use windows that extend beyond the desktop to retrieve their bitmap representation?