Method _remove_overlapped_lines

pdf2docx/layout/Blocks.py:387–403

Delete overlapped lines. NOTE: Don't run this method until floating images are excluded.

(self, line_overlap_threshold:float)

        return self

    def _remove_overlapped_lines(self, line_overlap_threshold:float):
        '''Delete overlapped lines. 
        NOTE: Don't run this method until floating images are excluded.
        '''
        # group lines by overlap
        fun = lambda a, b: a.get_main_bbox(b, threshold=line_overlap_threshold)
        groups = self.group(fun)
        
        # delete overlapped lines
        for group in filter(lambda group: len(group)>1, groups):
            # keep only the line with largest area
            sorted_lines = sorted(group, key=lambda line: line.bbox.get_area())
            for line in sorted_lines[:-1]:
                logging.warning('Ignore Line "%s" due to overlap', line.text)
                line.update_bbox((0,0,0,0))

        return self