unpaper: fix check for missing and old versions, add test case

This commit is contained in:
James R. Barlow
2016-03-10 15:37:09 -08:00
parent 3fed94bb79
commit 322085933b
4 changed files with 54 additions and 17 deletions
+8 -3
View File
@@ -234,13 +234,18 @@ if options.pdf_renderer == 'tesseract' and tesseract.version() < '3.04.01' \
"Some PDF viewers will fail to find searchable text.\n"
"--pdf-renderer=tesseract is not recommended.")
if any((options.deskew, options.clean, options.clean_final)):
if any((options.clean, options.clean_final)):
try:
from . import unpaper
except ImportError:
if unpaper.version() < '6.1':
complain(
"The installed 'unpaper' is not supported. "
"Install version 6.1 or newer.")
sys.exit(ExitCode.missing_dependency)
except FileNotFoundError:
complain(
"Install the 'unpaper' program to use --deskew or --clean.")
sys.exit(ExitCode.bad_args)
sys.exit(ExitCode.missing_dependency)
else:
unpaper = None
+15 -14
View File
@@ -3,7 +3,7 @@
# unpaper documentation:
# https://github.com/Flameeyes/unpaper/blob/master/doc/basic-concepts.md
from subprocess import Popen, PIPE
from subprocess import CalledProcessError, STDOUT, check_output, check_call
from tempfile import NamedTemporaryFile
import sys
import os
@@ -17,10 +17,9 @@ def version():
get_program('unpaper'),
'--version'
]
p_unpaper = Popen(args_unpaper, close_fds=True, universal_newlines=True,
stdout=PIPE, stderr=PIPE)
version, _ = p_unpaper.communicate(timeout=5)
version = check_output(
args_unpaper, close_fds=True, universal_newlines=True,
stderr=STDOUT, timeout=5)
return version.strip()
@@ -68,15 +67,17 @@ def run(input_file, output_file, dpi, log, mode_args):
os.unlink(output_pnm.name)
args_unpaper.extend([input_pnm.name, output_pnm.name])
p_unpaper = Popen(
args_unpaper, close_fds=True,
universal_newlines=True, stdout=PIPE, stderr=PIPE
)
out, err = p_unpaper.communicate()
log.debug(out)
log.debug(err)
Image.open(output_pnm.name).save(output_file)
try:
stdout = check_output(
args_unpaper, close_fds=True,
universal_newlines=True, stderr=STDOUT,
)
except CalledProcessError as e:
log.debug(e.output)
raise e from e
else:
log.debug(stdout)
Image.open(output_pnm.name).save(output_file)
def deskew(input_file, output_file, dpi, log):
+14
View File
@@ -0,0 +1,14 @@
#!/usr/bin/env python3
import sys
def main():
if sys.argv[1] == '--version':
print('0.5')
sys.exit(0)
print("Only supports --version")
sys.exit(1)
if __name__ == '__main__':
main()
+17
View File
@@ -503,3 +503,20 @@ def test_tesseract_image_too_big(renderer, spoof_tesseract_big_image_error):
check_ocrmypdf(
'hugemono.pdf', 'hugemono_%s.pdf' % renderer, '-r',
'--pdf-renderer', renderer, env=spoof_tesseract_big_image_error)
def test_no_unpaper():
env = os.environ.copy()
env['OCRMYPDF_UNPAPER'] = os.path.abspath('./spoof/no_unpaper_here.py')
sh, out, err = run_ocrmypdf_env(
'c02-22.pdf', 'wont_be_created.pdf', '--clean', env=env)
assert sh.returncode == ExitCode.missing_dependency
def test_old_unpaper():
env = os.environ.copy()
env['OCRMYPDF_UNPAPER'] = os.path.abspath('./spoof/unpaper_oldversion.py')
sh, out, err = run_ocrmypdf_env(
'c02-22.pdf', 'wont_be_created.pdf', '--clean', env=env)
assert sh.returncode == ExitCode.missing_dependency