Take a stab at writing test case for autorotate

ocrmypdf/ghostscript.py

@@ -7,7 +7,8 @@ from shutil import copy
 from . import get_program
 
 
-def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log):
+def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
+                  pageno=1):
     with NamedTemporaryFile(delete=True) as tmp:
         args_gs = [
             get_program('gs'),
@@ -15,6 +16,8 @@ def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log):
             '-dBATCH',
             '-dNOPAUSE',
             '-sDEVICE=%s' % raster_device,
+            '-dFirstPage=%i' % pageno,
+            '-dLastPage=%i' % pageno,
             '-o', tmp.name,
             '-r{0}x{1}'.format(str(xres), str(yres)),
             input_file

test_requirements.txt

@@ -1 +1 @@
-pytest>=2.7.2
+pytest>=2.7.2

tests/test_main.py

@@ -238,6 +238,28 @@ def test_argsfile(spoof_tesseract_noop):
                    env=spoof_tesseract_noop)
 
 
+@pytest.mark.parametrize('renderer', [
+    'hocr',
+    'tesseract',
+    ])
+def test_autorotate(spoof_tesseract_cache, renderer):
+    import ocrmypdf.ghostscript as ghostscript
+    import logging
+
+    gslog = logging.getLogger()
+
+    out = check_ocrmypdf('cardinal.pdf', 'test_autorotate_%s.pdf' % renderer,
+                         '-r', '-v', '1')
+    for n in range(1, 4+1):
+        ghostscript.rasterize_pdf(
+            out, _make_output('cardinal-%i.png' % n), xres=100, yres=100,
+            raster_device='pngmono', log=gslog, pageno=n)
+
+    ghostscript.rasterize_pdf(
+            _make_input('cardinal.pdf'), _make_output('reference.png'),
+            xres=100, yres=100, raster_device='pngmono', log=gslog, pageno=1)
+
+
 @pytest.mark.parametrize('renderer', [
     'hocr',
     'tesseract',