file_parsing.py 文件源码

def file_parser(fname, pages=None):
    if magic.from_file(fname, mime=True) == 'application/pdf':
        try:
            text_array = []
            d = pdf.Document(fname)
            for i, p in enumerate(d, start=1):
                for f in p:
                    for b in f:
                        for l in b:
                            text_array.append(l.text.encode('UTF-8'))

                if i == pages:  # break after x pages
                    break

            print "Processed %i pages" % (i)
            return '\n'.join(text_array)
        except Exception as e:
            print "PDF Parser Exception: ", e
    else:
        try:
            content = parser.from_file(fname)['content']
            return (content or '').encode('UTF-8')
        except Exception as e:
            print "File Parser Exception: ", e