Nothing particularly surprising. In PDF individual font glyphs are often positioned one by one, not as complete words or sentences. SO when extracting text from PDF, such as for copying, programs have to guess what are word boundaries based on positioning, they sometimes guess wrong.
|