Merge branch 'develop' into release/v3.2

.gitignore

@@ -1,26 +1,31 @@
-tmp/
-log/
+# Development environment
 *.pyc
-tests/output/
-.ruffus_history.sqlite
 *.sublime-*
-/*.pdf
-build/
-dist/
-*.egg-info/
-venv/
 venv-3.4/
 venv-3.5/
-*/test/output
-bin/
-include/
-lib/
-pip-selfcheck.json
+venv/
 pyvenv.cfg
-htmlcov/
-.coverage
+
+# Package building
+*.egg-info/
 .cache/
+.eggs/
+build/
+dist/
+
+# Automatically generated files
+ocrmypdf/lib/_*.py
+ocrmypdf/version.py
+
+# Code coverage
+.coverage
+htmlcov/
+
+# Testing
+log/
+/*.pdf
 .ipynb_checkpoints/
 tests/cache/
+tests/output/
 tests/resources/private
-ocrmypdf/version.py
+tmp/

README.rst

@@ -11,12 +11,7 @@ Main features
    `PDF/A <https://en.wikipedia.org/?title=PDF/A>`__ file from a regular PDF
 -  Places OCR text accurately below the image to ease copy / paste
 -  Keeps the exact resolution of the original embedded images
-
-   -  or if requested oversamples the images before OCRing so as to get
-      better results
-
--  When possible, inserts OCR information as a "lossless" operation without transcoding
-   images or rendering vector information
+-  When possible, inserts OCR information as a "lossless" operation without rendering vector information
 -  Keeps file size about the same
 -  If requested deskews and/or cleans the image before performing OCR
 -  Validates input and output files

RELEASE_NOTES.rst

@@ -12,13 +12,21 @@ v3.2:
 New features
 ------------
 
--  Lossless reconstruction: when possible, OCRmyPDF will inject text layers without transcoding
-   images or otherwise manipulating the content and layout of a PDF page. The overall PDF is still reconstructed.
+-  Lossless reconstruction: when possible, OCRmyPDF will inject text layers without 
+   otherwise manipulating the content and layout of a PDF page. For example, a PDF containing a mix
+   of vector and raster content would see the vector content preserved. Images may still be transcoded
+   during PDF/A conversion.  (``--deskew`` and ``--clean-final`` disable this mode, necessarily.)
 -  New argument ``--tesseract-pagesegmode`` allows you to pass page segmentation arguments to Tesseract OCR.
    This helps for two column text and other situations that confuse Tesseract.
 -  Added a new "polyglot" version of the Docker image, that generates Tesseract with all languages packs installed,
    for the polyglots among us. It is much larger.
 
+Changes
+-------
+
+-  JPEG transcoding quality is now 95 instead of the default 75. Bigger file sizes for less degradation.
+
+
 
 v3.1.1:
 =======

ocrmypdf/ghostscript.py

@@ -45,6 +45,7 @@ def generate_pdfa(pdf_pages, output_file, threads=1):
             "-sDEVICE=pdfwrite",
             "-sColorConversionStrategy=/RGB",
             "-sProcessColorModel=DeviceRGB",
+            "-dJPEGQ=95",
             "-dPDFA=2",
             "-sPDFACompatibilityPolicy=2",
             "-sOutputICCProfile=srgb.icc",

tests/spoof/tesseract_noop.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python3
 import sys
 import img2pdf
+from PIL import Image
 
 
 VERSION_STRING = '''tesseract 3.04.00
@@ -40,9 +41,12 @@ def main():
         print('List of available languages (1):\neng', file=sys.stderr)
         sys.exit(0)
     elif sys.argv[-1] == 'hocr':
+        inputf = sys.argv[-3]
         output = sys.argv[-2]
-        with open(output + '.hocr', 'w', encoding='utf-8') as f:
-            f.write(HOCR_TEMPLATE.format('1000', '1000'))
+        with Image.open(inputf) as im, \
+                open(output + '.hocr', 'w', encoding='utf-8') as f:
+            w, h = im.size
+            f.write(HOCR_TEMPLATE.format(str(w), str(h)))
     elif sys.argv[-1] == 'pdf':
         inputf = sys.argv[-3]
         output = sys.argv[-2]