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
@pytest.mark.skipif('8.0.0' <= qpdf.version() <= '8.0.1',
reason="qpdf regression on pages with no contents")
def test_no_contents(spoof_tesseract_noop, resources, outpdf):
check_ocrmypdf(resources / 'no_contents.pdf', outpdf, '--force-ocr',
env=spoof_tesseract_noop)
@pytest.mark.parametrize('image', [
'baiona.png',
'baiona_gray.png',
'baiona_alpha.png',
'congress.jpg'
])
def test_compression_preserved(spoof_tesseract_noop, ocrmypdf_exec,
resources, image, outpdf):
input_file = str(resources / image)
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
@pytest.mark.skipif(
'8.0.0' <= qpdf.version() <= '8.0.1',
reason="qpdf regression on pages with no contents",
)
def test_no_contents(spoof_tesseract_noop, resources, outpdf):
check_ocrmypdf(resources / 'no_contents.pdf',
outpdf,
'--force-ocr',
env=spoof_tesseract_noop)
@pytest.mark.parametrize(
'image',
['baiona.png', 'baiona_gray.png', 'baiona_alpha.png', 'congress.jpg'])
def test_compression_preserved(spoof_tesseract_noop, ocrmypdf_exec, resources,
image, outpdf):
input_file = str(resources / image)
# © 2017 James R. Barlow: github.com/jbarlow83
import logging
import resource
import pytest
from ocrmypdf.exec import ghostscript, tesseract, qpdf
from ocrmypdf.pdfinfo import PdfInfo
@pytest.mark.skipif(
qpdf.version() < '7.0.0',
reason="negzero.pdf crashes earlier versions")
def test_qpdf_negative_zero(resources, outpdf):
negzero = resources / 'negzero.pdf'
hugemono = resources / 'hugemono.pdf'
# raises exception on err
qpdf.merge([str(negzero), str(hugemono)], outpdf, log=logging.getLogger())
@pytest.mark.timeout(15)
@pytest.mark.parametrize('max_files,skip', [
(2, 0), # Can we merge correctly without opening more than 2 files at once?
(16, 0), # And does this work properly when we can one-shot it?
(2, 1), # Or playing with even/odd
(3, 0) # Or odd step size
])
def test_qpdf_merge_correctness(resources, outpdf, max_files, skip):
# All of these must be only one page long
inputs = [
'2400dpi.pdf', 'aspect.pdf', 'blank.pdf', 'ccitt.pdf',
# © 2017 James R. Barlow: github.com/jbarlow83
import logging
import resource
import pytest
from ocrmypdf.exec import ghostscript, tesseract, qpdf
from ocrmypdf.pdfinfo import PdfInfo
@pytest.mark.skipif(qpdf.version() < '7.0.0',
reason="negzero.pdf crashes earlier versions")
def test_qpdf_negative_zero(resources, outpdf):
negzero = resources / 'negzero.pdf'
hugemono = resources / 'hugemono.pdf'
# raises exception on err
qpdf.merge([str(negzero), str(hugemono)], outpdf, log=logging.getLogger())
@pytest.mark.timeout(15)
@pytest.mark.parametrize(
'max_files,skip',
[
(
2, 0
), # Can we merge correctly without opening more than 2 files at once?
(16, 0), # And does this work properly when we can one-shot it?
(2, 1), # Or playing with even/odd
(3, 0) # Or odd step size
])
def test_qpdf_merge_correctness(resources, outpdf, max_files, skip):
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
@pytest.mark.skipif('8.0.0' <= qpdf.version() <= '8.0.1',
reason="qpdf regression")
def test_no_contents(spoof_tesseract_noop, resources, outpdf):
check_ocrmypdf(resources / 'no_contents.pdf', outpdf, '--force-ocr',
env=spoof_tesseract_noop)
@pytest.mark.parametrize('image', [
'baiona.png',
'baiona_gray.png',
'congress.jpg'
])
def test_compression_preserved(spoof_tesseract_noop, ocrmypdf_exec,
resources, image, outpdf):
from PIL import Image