Quote:
Originally Posted by foobnix
Maybe mdp can help us with mathematics how to detect text borders without watermark?
|
: D Ok, with pleasure, but what about
-- the automatic measurement guessing the margins determines four initial values [top, bottom, left, right];
-- the user has his floating form to modify those four margins at will.
I think this solves the problem of watermarks, adds freedom to the user and is flexible enough to work in unexpected page layout. PDF files with consistently different pages should probably be preprocessed...
But the idea of an algorithm to identify overflowinig watermarks is very interesting! Foobnix, could you please mail me (or post here) the full (hierarchical) name of the class currently recognizing the borders, I would like to read it?