🚀 go-pugleaf

Your best bet is to render the page (or a portion of it) to an offscreen bitmap and then work from that.

Hi Leonard,

Thanks for the quick reply! Let me break down the areas I'm still unclear on into a couple of questions:

(1) What is the best way to allow the user to draw a rectangular selection? Are there some callback functions I should implement to allow the user to click and drag a marquee, or do I have to handle the code to draw the rectangle myself? I'm wondering, for example, if I need to implement the drawing code as part of a custom AVTool DoClickProcType callback, or if there is a better, built in way to get a rectangular area that a user has drawn (given that this operation is common to many tools).

(2) Once I have the rectangular selection coordinates from #1 above, how do I render it to an offscreen bitmap as you suggested? Is there a function in the API that allows me to do this?

Thanks again!

Steve

Yes, custom AVTool with DoClickProc. There is an AV API that will handle the tracking for you (don't recall name offhand).

Look at the PDPageDrawXXX APIs for rendering to offscreen...

Thanks for the help! I have this working close to the way I need.

I am now using AVPageViewDragOutNewRect to allow the user to select a rectangular region, then AVPageViewDrawRectOutline to draw the outline of the user's selection to the screen. I want to allow the user to select multiple rectangular regions, perhaps on different pages. However, AVPageViewDrawRectOutline uses device space coordinates, so all selected rectangles are drawn on every page, regardless of the page they were originally selected from by the user. Is there a way to draw rectangular areas only on specific pages? These rectangles are supposed to be temporary and not a modification to the document, so I want to avoid annotations.

Steve

Never mind. I have determined that Acrobat will set the page number in the current AVPageView before drawing each page. In the callback function registered for page view drawing events, simply comparing the page number returned by AVPageViewGetPageNum with the page of the rectangle to be drawn appears to work correctly (if they are not equal, don't draw the shape). Of course, this means the original page number of each drawing needs to be saved along with the rectangle, which I am accomplishing using a vector of structs.

Thanks again for all the help to date!

Steve

I am having some issues implementing Leonard's suggestion above. It appears that PDPageDrawContentsToMemory() is only available from the PDF Library, not the standard Acrobat SDK. I'm trying to accomplish the equivalent of the built-in snapshot tool--that is, copy the image of a rectangular area on a page to the clipboard (or a jpeg, or whatever). Does the SDK provide any functions for accomplishing this?

Thanks!

Steve

There are a set of PDPageDrawContentsXXX APIs that are available for plugins...

Thanks again for the help! I now have this nearly working.

I'm using PDPageDrawContentsToWindow to draw to an offscreen DC. This works fine. However, I want to limit what is drawn to the rectangle the user selected, which I store as an ASFixedRect. When I pass this to PDPageDrawContentsToWindow as the updateRect parameter, the entire page is still drawn with the same origin. I've also experimented with passing some invalid ASFixedRect values, and still the entire page is drawn. It seems like the updateRect parameter is being ignored (the default when this parameter is NULL).

Any advice?

Thanks!

Steve

Are you sure you are passing the coordinates in rotated user-space and not device-space? Have you also tried PDPageDrawContentsToWindowEx?

The coordinates are certainly in user space, though I don't know what is meant by "rotated". The same ASFixedRect is also used to select text within the chosen area, and the selected text corresponds to the expected rectangular area. Even if my coordinates are wrong, I would expect only objects within the (incorrect) rectangle to be drawn per the documentation--not the entire page. As a test, I created an ASFixedRect with no dimension (zero for all values). Still, the entire page was drawn.

PDPageDrawContentsToWindowEx just calls PDPageDrawContentsToWindow, but allows control of the display flags. The result of calling this function is the same.

Any other thoughts?

Steve