Fix KeyError: 'dpi' when using --threshold on image to PDF

Fixes #607
2020-08-07 02:21:02 -07:00
parent 4fa28d7e74
commit 9b641055e1
2 changed files with 17 additions and 1 deletions
@@ -530,7 +530,9 @@ def create_ocr_image(image: Path, page_context: PageContext):
        if options.threshold:
            pix = leptonica.Pix.frompil(im)
            pix = pix.masked_threshold_on_background_norm()
-            im = pix.topil()
+            im_pix = pix.topil()
+            im_pix.info['dpi'] = im.info['dpi']
+            im = im_pix

        del draw

@@ -877,3 +877,17 @@ def test_image_dpi_not_image(caplog, resources, outpdf):
        'tests/plugins/tesseract_noop.py',
    )
    assert '--image-dpi is being ignored' in caplog.text
+
+
+def test_image_dpi_threshold(resources, outpdf):
+    check_ocrmypdf(
+        resources / 'typewriter.png',
+        outpdf,
+        '--threshold',
+        '--image-dpi=170',
+        '--output-type=pdf',
+        '--optimize=0',
+        '--plugin',
+        'tests/plugins/tesseract_noop.py',
+    )
+    assert outpdf.exists()