2016-01-15 14:14:08 -08:00
|
|
|
#!/usr/bin/env python3
|
|
|
|
# © 2015 James R. Barlow: github.com/jbarlow83
|
|
|
|
|
|
|
|
from ocrmypdf import hocrtransform
|
2016-12-10 15:34:00 -08:00
|
|
|
from ocrmypdf.exec.tesseract import HOCR_TEMPLATE
|
2017-01-26 16:38:59 -08:00
|
|
|
from ocrmypdf.exec import qpdf
|
2016-01-15 14:14:08 -08:00
|
|
|
from reportlab.pdfgen.canvas import Canvas
|
|
|
|
from PIL import Image
|
|
|
|
from tempfile import NamedTemporaryFile
|
|
|
|
from contextlib import suppress
|
2017-01-26 16:38:59 -08:00
|
|
|
from pathlib import Path
|
2016-01-15 14:14:08 -08:00
|
|
|
import os
|
|
|
|
import shutil
|
|
|
|
import pytest
|
|
|
|
import img2pdf
|
|
|
|
import pytest
|
|
|
|
import sys
|
|
|
|
|
|
|
|
|
2017-01-26 16:38:59 -08:00
|
|
|
@pytest.fixture
|
|
|
|
def blank_hocr(tmpdir):
|
|
|
|
filename = Path(tmpdir) / "blank.hocr"
|
|
|
|
with open(filename, 'w') as f:
|
2016-01-15 14:14:08 -08:00
|
|
|
f.write(HOCR_TEMPLATE)
|
2017-01-26 16:38:59 -08:00
|
|
|
return filename
|
2016-01-15 14:14:08 -08:00
|
|
|
|
|
|
|
|
2017-01-26 16:38:59 -08:00
|
|
|
def test_mono_image(blank_hocr, outdir):
|
2016-01-15 14:14:08 -08:00
|
|
|
im = Image.new('1', (8, 8), 0)
|
|
|
|
for n in range(8):
|
|
|
|
im.putpixel((n, n), 1)
|
2017-01-26 16:38:59 -08:00
|
|
|
im.save(outdir / 'mono.tif', format='TIFF')
|
2016-01-15 14:14:08 -08:00
|
|
|
|
2017-01-26 16:38:59 -08:00
|
|
|
hocr = hocrtransform.HocrTransform(blank_hocr, 300)
|
|
|
|
hocr.to_pdf(str(outdir / 'mono.pdf'), imageFileName=outdir / 'mono.tif')
|
2016-01-15 14:14:08 -08:00
|
|
|
|
2017-01-26 16:38:59 -08:00
|
|
|
qpdf.check(str(outdir / 'mono.pdf'))
|
2016-01-15 14:14:08 -08:00
|
|
|
|
|
|
|
|
|
|
|
|