Fix issue #163, color and grayscale images JPEG compressed when not needed

ocrmypdf/exec/ghostscript.py

@@ -77,6 +77,10 @@ def generate_pdfa(pdf_pages, output_file, log, threads=1):
             "-dAutoRotatePages=/None",
             "-sColorConversionStrategy=/RGB",
             "-sProcessColorModel=DeviceRGB",
+            "-dAutoFilterColorImages=false",
+            "-sColorImageFilter=",
+            "-dAutoFilterGrayImages=false",
+            "-sGrayImageFilter=",
             "-dJPEGQ=95",
             "-dPDFA=2",
             "-dPDFACompatibilityPolicy=1",

ocrmypdf/pipeline.py

@@ -520,6 +520,8 @@ def select_visible_page_image(
     pageinfo = get_pageinfo(image, context)
     if pageinfo['images'] and \
             all(im['enc'] == 'jpeg' for im in pageinfo['images']):
+        log.debug('{:4d}: JPEG input -> JPEG output'.format(
+            page_number(page_pdf)))
         # If all images were JPEGs originally, produce a JPEG as output
         im = Image.open(image)
 

tests/resources/README.rst

@@ -34,6 +34,9 @@ In some cases they were converted from one image format to another without other
     *   - typewriter.png, 2400dpi.pdf
         - `Wikimedia: Triumph typewrtier text Linzensoep`_
         * Creative Commons BY-SA 2.5
+    *   - baiona.png
+        - `Wikimedia: Baionako udalerri mugakideak`_
+        - Creative Commons BY-SA 4.0
 
 
 Files generated for this project
@@ -118,4 +121,6 @@ These test resources are assemblies from other previously mentioned files, relea
 
 .. _`Linux (Wikipedia Article)`: https://de.wikipedia.org/wiki/Linux 
 
-.. _`Wikimedia: Triumph typewrtier text Linzensoep`: https://commons.wikimedia.org/wiki/File:Triumph.typewriter_text_Linzensoep.gif
+.. _`Wikimedia: Triumph typewrtier text Linzensoep`: https://commons.wikimedia.org/wiki/File:Triumph.typewriter_text_Linzensoep.gif
+
+.. _`Wikimedia: Baionako udalerri mugakideak`: https://commons.wikimedia.org/wiki/File:Baionako_udalerri_mugakideak.png

tests/test_main.py

@@ -874,4 +874,38 @@ def test_gs_raster_failure(spoof_no_tess_gs_raster_fail, resources, outpdf):
 
 def test_no_contents(spoof_tesseract_noop, resources, outpdf):
     check_ocrmypdf(resources / 'no_contents.pdf', outpdf, '--force-ocr',
-                   env=spoof_tesseract_noop)
+                   env=spoof_tesseract_noop)
+
+
+@pytest.mark.parametrize('image', [
+    'baiona.png',
+    'baiona_gray.png',
+    ])
+def test_lossless_to_lossless(spoof_tesseract_noop, ocrmypdf_exec,
+                              resources, image, outpdf):
+    from PIL import Image
+
+    input_file = str(resources / image)
+    output_file = str(outpdf)
+
+    im = Image.open(input_file)
+
+    # Runs: ocrmypdf - output.pdf < testfile
+    with open(input_file, 'rb') as input_stream:
+        p_args = ocrmypdf_exec + ['--image-dpi', '150', '-', output_file]
+        p = Popen(
+            p_args, close_fds=True, stdout=PIPE, stderr=PIPE,
+            stdin=input_stream, env=spoof_tesseract_noop)
+        out, err = p.communicate()
+
+        assert p.returncode == ExitCode.ok
+
+    pdfinfo = pdf_get_all_pageinfo(output_file)
+    assert pdfinfo[0]['images'][0]['enc'] != 'jpeg', \
+        "Lossless compression changed to lossy!"
+    if im.mode.startswith('RGB') or im.mode.startswith('BGR'):
+        assert pdfinfo[0]['images'][0]['color'] == 'rgb', \
+            "Colorspace changed"
+    elif im.mode.startswith('L'):
+        assert pdfinfo[0]['images'][0]['color'] == 'gray', \
+            "Colorspace changed"