From 91b42cbfa855972112492699297a44c6806c9804 Mon Sep 17 00:00:00 2001
From: "James R. Barlow" <jim@purplerock.ca>
Date: Tue, 9 Jan 2018 00:17:53 -0800
Subject: [PATCH] Fix issue in sandwich renderer when skipping OCR on a rotated
 and deskewed page

If OCR is skipped due to --tesseract-timeout or similar, and the skip page is rotated with /Rotate, and the skip page was deskewed or had other image processing, then the skip page was created with the wrong dimensions causing the output page to be cropped.
---
 ocrmypdf/exec/ghostscript.py |  2 +-
 ocrmypdf/exec/tesseract.py   |  4 ++++
 tests/test_main.py           | 23 +++++++++++++++++++++++
 3 files changed, 28 insertions(+), 1 deletion(-)

diff --git a/ocrmypdf/exec/ghostscript.py b/ocrmypdf/exec/ghostscript.py
index b94c56fb..d77efecc 100644
--- a/ocrmypdf/exec/ghostscript.py
+++ b/ocrmypdf/exec/ghostscript.py
@@ -37,7 +37,7 @@ def rasterize_pdf(input_file, output_file, xres, yres, raster_device, log,
     :param yres: 
     :param raster_device: 
     :param log: 
-    :param pageno: page number to rasterize
+    :param pageno: page number to rasterize (beginning at page 1)
     :param page_dpi: resolution tuple (x, y) overriding output image DPI 
     :return: 
     """
diff --git a/ocrmypdf/exec/tesseract.py b/ocrmypdf/exec/tesseract.py
index e4cf9600..1ead0c6b 100644
--- a/ocrmypdf/exec/tesseract.py
+++ b/ocrmypdf/exec/tesseract.py
@@ -279,6 +279,10 @@ def use_skip_page(text_only, skip_pdf, output_pdf, output_text):
     with open(output_pdf, 'wb') as out:
         pdf_out = pypdf.PdfFileWriter()
         w, h = page0.mediaBox.getWidth(), page0.mediaBox.getHeight()
+        # If skip page has a /Rotate key, replicate the rotation
+        rotation = int(page0.get('/Rotate', 0))
+        if rotation % 180 == 90:
+            w, h = h, w
         pdf_out.addBlankPage(w, h)
         pdf_out.write(out)
 
diff --git a/tests/test_main.py b/tests/test_main.py
index c21f2253..b6c652b0 100644
--- a/tests/test_main.py
+++ b/tests/test_main.py
@@ -1037,3 +1037,26 @@ def test_bad_utf8(spoof_tess_bad_utf8, renderer, resources, no_outpdf):
     assert p.returncode != 0
     assert 'not utf-8' in err, "should whine about utf-8"
     assert '\\x96' in err, 'should repeat backslash encoded output'
+
+
+@pytest.mark.skipif(
+    not tesseract.has_textonly_pdf(), 
+    reason="issue only affects sandwich")
+def test_rotate_deskew_timeout(resources, outdir):
+    check_ocrmypdf(
+        resources / 'rotated_skew.pdf',
+        outdir / 'deskewed.pdf',
+        '--deskew',
+        '--tesseract-timeout', '0',
+        '--pdf-renderer', 'sandwich'
+    )
+
+    correlation = check_monochrome_correlation(
+        outdir,
+        reference_pdf=resources / 'ccitt.pdf',
+        reference_pageno=1,
+        test_pdf=outdir / 'deskewed.pdf',
+        test_pageno=1)
+
+    # Confirm that the page still got deskewed
+    assert correlation > 0.50
\ No newline at end of file