1# ... 2# 3# Copyright (c) 2019 Philippe Mathieu-Daudé <f4bug@amsat.org> 4# 5# This work is licensed under the terms of the GNU GPL, version 2 or 6# later. See the COPYING file in the top-level directory. 7 8import logging 9from subprocess import run 10 11 12def tesseract_ocr(image_path, tesseract_args=''): 13 console_logger = logging.getLogger('console') 14 console_logger.debug(image_path) 15 proc = run(['tesseract', image_path, 'stdout'], 16 capture_output=True, encoding='utf8') 17 if proc.returncode: 18 return None 19 lines = [] 20 for line in proc.stdout.split('\n'): 21 sline = line.strip() 22 if len(sline): 23 console_logger.debug(sline) 24 lines += [sline] 25 return lines 26