Ensure that ocrmypdf stops and reports an error if Ghostscript fails

Past behavior was to continue and let ruffus puke eventually
This commit is contained in:
James R. Barlow
2017-05-01 15:44:21 -07:00
parent cf643c9f43
commit b9b12e2879
5 changed files with 103 additions and 7 deletions
+1 -1
View File
@@ -48,7 +48,7 @@ class InputFileError(ExitCodeException):
class SubprocessOutputError(ExitCodeException):
exit_code = ExitCode.other_error
exit_code = ExitCode.child_process_error
class EncryptedPdfError(ExitCodeException):
+13 -6
View File
@@ -6,8 +6,9 @@ from subprocess import Popen, PIPE, STDOUT, check_call, CalledProcessError, \
check_output
from shutil import copy
from functools import lru_cache
import re
from . import get_program
from ..pdfa import SRGB_ICC_PROFILE
from ..exceptions import SubprocessOutputError
@lru_cache(maxsize=1)
@@ -28,6 +29,10 @@ def version():
return version.strip()
def _gs_error_reported(stream):
return re.search(r'error', stream, flags=re.IGNORECASE)
def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
pageno=1):
with NamedTemporaryFile(delete=True) as tmp:
@@ -48,15 +53,16 @@ def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
p = Popen(args_gs, close_fds=True, stdout=PIPE, stderr=STDOUT,
universal_newlines=True)
stdout, _ = p.communicate()
if 'error' in stdout:
log.error(stdout) # Ghostscript puts errors in stdout
if _gs_error_reported(stdout):
log.error(stdout)
else:
log.debug(stdout)
if p.returncode == 0:
copy(tmp.name, output_file)
else:
log.error('Ghostscript rendering failed')
log.error('Ghostscript rasterizing failed')
raise SubprocessOutputError()
def generate_pdfa(pdf_pages, output_file, log, threads=1):
@@ -81,7 +87,7 @@ def generate_pdfa(pdf_pages, output_file, log, threads=1):
universal_newlines=True)
stdout, _ = p.communicate()
if 'error' in stdout or 'ERROR' in stdout:
if _gs_error_reported(stdout):
log.error(stdout)
elif 'overprint mode not set' in stdout:
# Unless someone is going to print PDF/A documents on a
@@ -99,4 +105,5 @@ def generate_pdfa(pdf_pages, output_file, log, threads=1):
# PDF/A - check PDF/A status elsewhere
copy(gs_pdf.name, output_file)
else:
log.error('Ghostscript PDF/A failed')
log.error('Ghostscript PDF/A rendering failed')
raise SubprocessOutputError()
+31
View File
@@ -0,0 +1,31 @@
#!/usr/bin/env python3
import sys
import os
"""Replicate Ghostscript raster failure while allowing rendering"""
def real_ghostscript(argv):
gs_args = ['gs'] + argv[1:]
os.execvp("gs", gs_args)
return # Not reachable
def main():
if '--version' in sys.argv:
print('9.20')
print('SPOOFED: ' + os.path.basename(__filename__))
sys.exit(0)
# For any rendering calls (device == pdfwrite) call real ghostscript
if '-sDEVICE=pdfwrite' in sys.argv:
real_ghostscript(sys.argv)
return
# Fail
print("ERROR: Ghost story archive not found")
sys.exit(1)
if __name__ == '__main__':
main()
+31
View File
@@ -0,0 +1,31 @@
#!/usr/bin/env python3
import sys
import os
"""Replicate Ghostscript render failure while allowing rasterizing"""
def real_ghostscript(argv):
gs_args = ['gs'] + argv[1:]
os.execvp("gs", gs_args)
return # Not reachable
def main():
if '--version' in sys.argv:
print('9.20')
print('SPOOFED: ' + os.path.basename(__filename__))
sys.exit(0)
# For any rasterize calls (device != pdfwrite) call real ghostscript
if '-sDEVICE=pdfwrite' not in sys.argv:
real_ghostscript(sys.argv)
return
# Fail
print("ERROR: Casper is not a friendly ghost")
sys.exit(1)
if __name__ == '__main__':
main()
+27
View File
@@ -51,6 +51,16 @@ def spoof_no_tess_pdfa_warning():
return spoof(tesseract='tesseract_noop.py', gs='gs_feature_elision.py')
@pytest.fixture
def spoof_no_tess_gs_render_fail():
return spoof(tesseract='tesseract_noop.py', gs='gs_render_failure.py')
@pytest.fixture
def spoof_no_tess_gs_raster_fail():
return spoof(tesseract='tesseract_noop.py', gs='gs_raster_failure.py')
@pytest.fixture
def spoof_qpdf_always_error():
return spoof(qpdf='qpdf_dummy_return2.py')
@@ -843,4 +853,21 @@ def test_skip_big_with_no_images(spoof_tesseract_noop, resources, outpdf):
check_ocrmypdf(resources / 'blank.pdf', outpdf,
'--skip-big', '5',
'--force-ocr',
env=spoof_tesseract_noop)
def test_gs_render_failure(spoof_no_tess_gs_render_fail, resources, outpdf):
p, out, err = run_ocrmypdf(
resources / 'blank.pdf', outpdf,
env=spoof_no_tess_gs_render_fail)
print(err)
assert p.returncode == ExitCode.child_process_error
def test_gs_raster_failure(spoof_no_tess_gs_raster_fail, resources, outpdf):
p, out, err = run_ocrmypdf(
resources / 'ccitt.pdf', outpdf,
env=spoof_no_tess_gs_raster_fail)
print(err)
assert p.returncode == ExitCode.child_process_error
env=spoof_tesseract_noop)