Fix issue #194 - --sidecar creates blank txt file

[ci skip]

.travis.yml

@@ -85,6 +85,7 @@ deploy:
   script: /usr/bin/true
   on:
     branch: master
+    tags: true
     condition: $TRAVIS_OS_NAME == "osx"
 
 after_deploy: |

.travis/autobrew.py

@@ -30,6 +30,7 @@ class Ocrmypdf < Formula
   depends_on "libjpeg"
   depends_on "webp"
   depends_on "little-cms2"
+  depends_on "zlib"
 
 ${resources}
   def install

.travis/linux_before_install.sh

@@ -25,14 +25,6 @@ sudo apt-get install -y --no-install-recommends \
 		tesseract-ocr-fra \
 		tesseract-ocr-deu
 
-# Workaround for https://github.com/tesseract-ocr/tesseract/issues/1167
-# Replace tesseract data installed by tesseract-ocr-eng with a known good version
-sudo apt-get install -y wget
-wget -q https://github.com/tesseract-ocr/tessdata/raw/master/eng.traineddata -O packages/eng.traineddata
-sha1sum packages/eng.traineddata
-sha1sum /usr/share/tesseract-ocr/4.00/tessdata/eng.traineddata
-sudo cp packages/eng.traineddata /usr/share/tesseract-ocr/4.00/tessdata/eng.traineddata
-
 pip install --upgrade pip
 mkdir -p packages
 [ -f packages/unpaper_6.1-1.deb ] || wget -q 'https://www.dropbox.com/s/vaq0kbwi6e6au80/unpaper_6.1-1.deb?raw=1' -O packages/unpaper_6.1-1.deb

docs/installation.rst

@@ -105,6 +105,12 @@ In this worked example, the current working directory contains an input file cal
 
 Note that ``ocrmypdf`` has its own separate ``-v VERBOSITYLEVEL`` argument to control debug verbosity. All Docker arguments should before the ``ocrmypdf`` image name and all arguments to ``ocrmypdf`` should be listed after.
 
+In some environments the permissions associated with Docker can be complex to configure. The process that executes Docker may end up not having the permissions to write the specified file system. In that case one can stream the file into and out of the Docker process and avoid all permission hassles, using ``-`` as the input and output filename:
+
+.. code-block:: bash
+
+    docker run --rm -i   ocrmypdf <other arguments to ocrmypdf> - - <input.pdf >output.pdf
+
 For convenience, a shell alias can hide the docker command:
 
 .. code-block:: bash

docs/release_notes.rst

@@ -5,9 +5,21 @@ OCRmyPDF uses `semantic versioning <http://semver.org/>`_ for its command line i
 
 The OCRmyPDF package itself does not contain a public API, although it is fairly stable and breaking changes are usually timed with a major release. A future release will clearly define the stable public API.
 
-v5.4
+v5.4.2
 ------
 
+-   Fixed a regression from v5.4.1 that caused sidecar files to be created as empty files
+
+
+v5.4.1
+------
+
+-   Add workaround for Tesseract v4.00alpha crash when trying to obtain orientation and the latest language packs are installed
+
+
+v5.4
+----
+
 -   Change wording of a deprecation warning to improve clarity
 -   Added option to generate PDF/A-1b output if desired (``--output-type pdfa-1``); default remains PDF/A-2b generation
 -   Update documentation

ocrmypdf/exec/tesseract.py

@@ -126,7 +126,7 @@ def tess_base_args(langs, engine_mode):
 
 def get_orientation(input_file, language: list, engine_mode, timeout: float,
                     log):
-    args_tesseract = tess_base_args(language, engine_mode) + [
+    args_tesseract = tess_base_args(['osd'], engine_mode) + [
         psm(), '0',
         input_file,
         'stdout'

ocrmypdf/pipeline.py

@@ -945,7 +945,7 @@ def merge_sidecars(
             if page_num != 0:
                 stream.write('\f')  # Form feed between pages
             if txt_file:
-                with open(txt_file, 'r') as in_:
+                with open(txt_file, 'r', encoding="utf-8") as in_:
                     txt = in_.read()
                     # Tesseract v4 alpha started adding form feeds in 
                     # commit aa6eb6b
@@ -953,9 +953,9 @@ def merge_sidecars(
                     # for consistency just ignore its form feeds and insert our
                     # own
                     if txt.endswith('\f'):
-                        stream.write(in_.read()[:-1])
+                        stream.write(txt[:-1])
                     else:
-                        stream.write(in_.read())
+                        stream.write(txt)
             else:
                 stream.write('[OCR skipped on page {}]'.format(
                         page_num + 1))
@@ -964,7 +964,7 @@ def merge_sidecars(
         write_pages(sys.stdout)
         sys.stdout.flush()
     else:
-        with open(output_file, 'w', encoding='utf-8') as out:
+        with open(output_file, 'w', encoding="utf-8") as out:
             write_pages(out)
 
 

tests/test_main.py

@@ -805,7 +805,6 @@ def test_user_words(resources, outdir):
             '--image-dpi', 150,
             '--sidecar', sidecar_before
         )
-
         assert 'cromulent' not in sidecar_before.open().read()
 
     with word_list.open('w') as f:
@@ -984,6 +983,19 @@ def test_sidecar_pagecount(spoof_tesseract_cache, resources, outpdf):
         "Sidecar page count does not match PDF page count"
 
 
+def test_sidecar_nonempty(spoof_tesseract_cache, resources, outpdf):
+    sidecar = outpdf + '.txt'
+    check_ocrmypdf(
+        resources / 'ccitt.pdf', outpdf,
+        '--sidecar', sidecar,
+        env=spoof_tesseract_cache
+    )
+
+    with open(sidecar, 'r') as f:
+        ocr_text = f.read()
+    assert 'the' in ocr_text
+
+
 def test_pdfa_1(spoof_tesseract_cache, resources, outpdf):
     check_ocrmypdf(
         resources / 'ccitt.pdf', outpdf,