Change Windows build to use older Ghostscript

It was mistakenly changed to on top at some point

.github/ISSUE_TEMPLATE/2-problem-with-specific-file.yml

@@ -21,7 +21,7 @@ body:
     id: reproduce
     attributes:
       label: Steps to reproduce
-      description: Please include steps to reproduce
+      description: Please include steps to reproduce.
       value: |
         1. Run ocrmypdf -v1 ...arguments... input.pdf output.pdf
         2. Open output.pdf
@@ -31,9 +31,8 @@ body:
     id: files
     attributes:
       label: Files
-      description: Please attach the input and output files, or any screenshots that may be helpful.
-      placeholder: |
-        Drag and drop files here.
+      description: |
+        Please attach the input and output files, or any screenshots that may be helpful.
 
         If you cannot provide a test file, we probably won't be able to help with the issue.
         PDF is a complex file format, and there may be technical details in the PDF that are
@@ -42,10 +41,11 @@ body:
         We understand files may contain personal or sensitive information. Here are some options:
         - Try reproducing the issue with a file from the test suite. (See tests/resources)
         - Try to create another file in the same way as your private file.
-        - Encrypt the file to OCRmyPDF's private GPG key.
+        - Encrypt the file to OCRmyPDF's private GPG key, and then zip the GPG file.
         - Use ``qpdf --json yourfile.pdf`` to produce a JSON representation of your file that
           omits personal information.
-
+      placeholder: |
+        Drag and drop files here.
   - type: dropdown
     id: packaging-system
     attributes:

.github/workflows/build.yml

@@ -189,7 +189,7 @@ jobs:
       - name: Install system packages
         run: |
           choco install --yes --no-progress --pre tesseract
-          choco install --yes --no-progress --ignore-checksums ghostscript
+          choco install --yes --no-progress --ignore-checksums ghostscript --version 9.56.1
 
       - name: Install Python packages
         run: |

docs/installation.rst

@@ -404,7 +404,7 @@ the following when running in an Administrator command prompt):
 * ``choco install --pre tesseract``
 * ``choco install pngquant`` (optional)
 
-Either set of commands will install the required software. At the mmoment there is no
+Either set of commands will install the required software. At the moment there is no
 single command to install Windows.
 
 You may then use ``pip`` to install ocrmypdf. (This can performed by a user or
@@ -681,4 +681,4 @@ can still install and use OCRmyPDF. A warning message will appear.
 In practice, OCRmyPDF may need more than 32-bit memory space to run when
 large documents are processed, so there are practical limitations to what
 users can accomplish with it. Still, for the common use case of an 32-bit
-ARM NAS or Raspberry Pi processing small documents, it should work.
+ARM NAS or Raspberry Pi processing small documents, it should work.

docs/languages.rst

@@ -18,16 +18,26 @@ Tesseract's documentation also lists the three-letter code for your language.
 Some are anglicized, e.g. Spanish is ``spa`` rather than ``esp``, while others
 are not, e.g. German is ``deu`` and French is ``fra``.
 
+Language packs (strictly speaking, Tesseract "traineddata" files) generally correspond
+to the language in question, but different language packs are used in certain
+situations. For German, the "Fraktur" language pack can assist with reading older
+materials in the Fraktur typeface family (``deu_frak``). Some communities have changed
+their script from Cyrillic to Latin; the Cyrillic version of Uzbek is available
+as ``uzb_cyrl`` and the Latin version is ``uzb``.
+
 After you have installed a language pack, you can use it with ``ocrmypdf -l <language>``,
 for example ``ocrmypdf -l spa``. For multilingual documents, you can specify
 all languages to be expected, e.g. ``ocrmypdf -l eng+fra`` for English and French.
 English is assumed by default unless other language(s) are specified.
 
 For Linux users, you can often find packages that provide language
-packs:
+packs.
 
-Debian and Ubuntu users
-=======================
+Platform install steps
+======================
+
+Debian and Ubuntu (apt)
+-----------------------
 
 .. code-block:: bash
 
@@ -42,8 +52,8 @@ to what languages it should search for. Multiple languages can be
 requested using either ``-l eng+fra`` (English and French) or
 ``-l eng -l fra``.
 
-Fedora users
-============
+Fedora
+------
 
 .. code-block:: bash
 
@@ -58,8 +68,8 @@ to what languages it should search for. Multiple languages can be
 requested using either ``-l eng+fra`` (English and French) or
 ``-l eng -l fra``.
 
-Gentoo users
-============
+Gentoo
+------
 
 On Gentoo the package ``app-text/tessdata_fast``, which ``app-text/tesseract`` depends on, handles Tesseract languages.
 It accepts USE flags to select what languages should be installed, these can be set in ``/etc/portage/package.use``.
@@ -85,23 +95,31 @@ to what languages it should search for. Multiple languages can be
 requested using either ``-l eng+fra`` (English and French) or
 ``-l eng -l fra``.
 
-macOS users
-===========
+macOS
+-----
 
 You can install additional language packs by
 :ref:`installing Tesseract using Homebrew with all language packs <macos-all-languages>`.
 
-Docker users
-============
+Docker
+------
 
 Users of the OCRmyPDF Docker image should install language packs into a
 derived Docker image as
 :ref:`described in that section <docker-lang-packs>`.
 
-Windows users
-=============
+Windows
+-------
 
 The Tesseract installer provided by Chocolatey currently includes only English language.
 To install other languages, download the respective language pack (``.traineddata`` file)
 from https://github.com/tesseract-ocr/tessdata/ and place it in
 ``C:\\Program Files\\Tesseract-OCR\\tessdata`` (or wherever Tesseract OCR is installed).
+
+Custom language packs
+=====================
+
+If you have fine-tuned or trained Tesseract and generated custom trained data, you can
+copy your ``customlang.traineddata`` file into your Tesseract "tessdata" folder, and
+then use the ``-l customlang`` argument to tell OCRmyPDF to pass that language on to
+Tesseract.

docs/release_notes.rst

@@ -30,6 +30,23 @@ OCRmyPDF typically supports the three most recent Python versions.
 
 .. |OCRmyPDF PyPI| image:: https://img.shields.io/pypi/v/ocrmypdf.svg
 
+v16.0.4
+=======
+
+-  Fixed some issues for left-to-right text with the new hOCR renderer. It is still
+   not default yet but will be made so soon. Right-to-left text is still in progress.
+-  Added an error to prevent use of several versions of Ghostscript that seem
+   corrupt existing text in input PDFs. Newly generated OCR is not affected.
+   For best results, use Ghostscript 10.02.1 or newer, which contains the fix
+   for the issue.
+
+v16.0.3
+=======
+
+-  Changed minimum required Ghostscript to 9.54, to support users of RHEL 9 and its
+   derivatives, since that is the latest version available there.
+-  Removed warning message about CVE-2023-43115, on the assumption that most
+   distributions have backported the patch by now.
 
 v16.0.2
 =======

src/ocrmypdf/_graft.py

@@ -292,6 +292,7 @@ class OcrGrafter:
             # finally move the lower left corner to match the mediabox. All transforms
             # must be premultiplied so they are applied in reverse order here.
             ctm = corner @ untranslate @ scale @ rotate @ translate
+            log.debug("Grafting with ctm %r", ctm)
 
             base_resources = _ensure_dictionary(base_page.obj, Name.Resources)
             base_xobjs = _ensure_dictionary(base_resources, Name.XObject)
@@ -311,9 +312,16 @@ class OcrGrafter:
 
             if strip_old_text:
                 strip_invisible_text(self.pdf_base, base_page)
-
+            base_page.contents_coalesce()
+            if self.render_mode == RenderMode.ON_TOP:
+                # Add q/Q to ensure content we append is drawn correctly
+                # Strictly speaking this needs to trace the whole q/Q stack in case
+                # stack is not balanced.
+                original = base_page.Contents.read_bytes()
+                base_page.Contents.write(b'q\n' + original + b'\nQ\n')
             base_page.contents_add(
-                new_text_layer, prepend=self.render_mode == RenderMode.ON_TOP
+                new_text_layer, prepend=self.render_mode == RenderMode.UNDERNEATH
             )
+            base_page.contents_coalesce()
 
             _update_resources(obj=base_page.obj, font=font, font_key=font_key)

src/ocrmypdf/_pipeline.py

@@ -34,7 +34,8 @@ from ocrmypdf.exceptions import (
     UnsupportedImageFormatError,
 )
 from ocrmypdf.helpers import IMG2PDF_KWARGS, Resolution, safe_symlink
-from ocrmypdf.hocrtransform import HocrTransform
+from ocrmypdf.hocrtransform import DebugRenderOptions, HocrTransform
+from ocrmypdf.hocrtransform._font import Courier
 from ocrmypdf.pdfa import generate_pdfa_ps
 from ocrmypdf.pdfinfo import Colorspace, Encoding, PageInfo, PdfInfo
 from ocrmypdf.pluginspec import OrientationConfidence
@@ -741,15 +742,25 @@ def render_hocr_page(hocr: Path, page_context: PageContext) -> Path:
         return output_file
 
     dpi = get_page_square_dpi(page_context, calculate_image_dpi(page_context))
-    debug_mode = options.pdf_renderer == 'hocrdebug'
-
+    debug_kwargs = {}
+    if options.pdf_renderer == 'hocrdebug':
+        debug_kwargs = dict(
+            debug_render_options=DebugRenderOptions(
+                render_baseline=True,
+                render_triangle=True,
+                render_line_bbox=False,
+                render_word_bbox=True,
+                render_paragraph_bbox=False,
+                render_space_bbox=False,
+            ),
+            font=Courier(),
+        )
     HocrTransform(
-        hocr_filename=hocr,
-        dpi=dpi.to_scalar(),  # square
-        debug=debug_mode,
+        hocr_filename=hocr, dpi=dpi.to_scalar(), **debug_kwargs  # square
     ).to_pdf(
         out_filename=output_file,
         image_filename=None,
+        invisible_text=True if not debug_kwargs else False,
     )
     return output_file
 

src/ocrmypdf/builtin_plugins/ghostscript.py

@@ -17,7 +17,7 @@ log = logging.getLogger(__name__)
 
 # Currently all blacklisted versions are lower than 9.55, so none need to
 # be added here. If a future version is blacklisted, add it here.
-BLACKLISTED_GS_VERSIONS: frozenset[str] = frozenset()
+BLACKLISTED_GS_VERSIONS: frozenset[Version] = frozenset()
 
 
 @hookimpl
@@ -54,7 +54,7 @@ def check_options(options):
         program='gs',
         package='ghostscript',
         version_checker=ghostscript.version,
-        need_version='9.55',  # Ubuntu 22.04's version
+        need_version='9.54',  # RHEL 9's version; Ubuntu 22.04 has 9.55
     )
     gs_version = ghostscript.version()
     if gs_version in BLACKLISTED_GS_VERSIONS:
@@ -62,13 +62,16 @@ def check_options(options):
             f"Ghostscript {gs_version} contains serious regressions and is not "
             "supported. Please upgrade to a newer version."
         )
-    if gs_version < Version('10.02.0'):
-        log.warning(
-            f"The installed version of Ghostscript {gs_version}, contains a remote "
-            "code execution security vulnerability. Please upgrade to a newer "
-            "version. For details see CVE-2023-43115. The issue is not known to "
-            "affect OCRmyPDF or processing PDFs with Ghostscript, but upgrading "
-            "Ghostscript is recommended."
+    if Version('10.0.0') <= gs_version < Version('10.02.1') and (
+        options.skip_text or options.redo_ocr
+    ):
+        raise MissingDependencyError(
+            f"Ghostscript 10.0.0 through 10.02.0 (your version: {gs_version}) "
+            "contain serious regressions that corrupt PDFs with existing text, "
+            "such as those processed using --skip-text or --redo-ocr. "
+            "Please upgrade to a "
+            "newer version, or use --output-type pdf to avoid Ghostscript, or "
+            "use --force-ocr to discard existing text."
         )
 
     if options.output_type == 'pdfa':

src/ocrmypdf/hocrtransform/_font.py

@@ -18,7 +18,12 @@ from pikepdf.canvas import Font
 log = logging.getLogger(__name__)
 
 
-class GlyphlessFont(Font):
+class EncodableFont(Font):
+    def text_encode(self, text: str) -> bytes:
+        raise NotImplementedError()
+
+
+class GlyphlessFont(EncodableFont):
     CID_TO_GID_DATA = zlib.compress(b"\x00\x01" * 65536)
     GLYPHLESS_FONT_NAME = 'pdf.ttf'
     GLYPHLESS_FONT = (package_files('ocrmypdf.data') / GLYPHLESS_FONT_NAME).read_bytes()
@@ -27,11 +32,14 @@ class GlyphlessFont(Font):
     def __init__(self):
         pass
 
-    def text_width(self, text: str, fontsize: float) -> int:
+    def text_width(self, text: str, fontsize: float) -> float:
         """Estimate the width of a text string when rendered with the given font."""
         # NFKC: split ligatures, combine diacritics
         return len(unicodedata.normalize("NFKC", text)) * (fontsize / self.CHAR_ASPECT)
 
+    def text_encode(self, text: str) -> bytes:
+        return text.encode('utf-16be')
+
     def register(self, pdf: Pdf):
         """Register the glyphless font.
 
@@ -110,3 +118,24 @@ class GlyphlessFont(Font):
         font_descriptor.FontFile2 = pdf.make_stream(self.GLYPHLESS_FONT)
         cid_font_type2.FontDescriptor = font_descriptor
         return basefont
+
+
+class Courier(EncodableFont):
+    """Courier font."""
+
+    def text_width(self, text: str, fontsize: float) -> float:
+        """Estimate the width of a text string when rendered with the given font."""
+        return len(text) * fontsize
+
+    def text_encode(self, text: str) -> bytes:
+        return text.encode('pdfdoc', errors='ignore')
+
+    def register(self, pdf: Pdf) -> Dictionary:
+        """Register the font."""
+        return pdf.make_indirect(
+            Dictionary(
+                BaseFont=Name.Courier,
+                Type=Name.Font,
+                Subtype=Name.Type1,
+            )
+        )

src/ocrmypdf/hocrtransform/_hocr.py

@@ -27,11 +27,11 @@ from pikepdf.canvas import (
     MAGENTA,
     RED,
     Canvas,
-    Font,
     Text,
     TextDirection,
 )
 
+from ocrmypdf.hocrtransform._font import EncodableFont as Font
 from ocrmypdf.hocrtransform._font import GlyphlessFont
 
 log = logging.getLogger(__name__)
@@ -45,12 +45,12 @@ Element = ElementTree.Element
 class DebugRenderOptions:
     """A class for managing rendering options."""
 
-    render_paragraph_bbox: bool
-    render_baseline: bool
-    render_triangle: bool
-    render_line_bbox: bool
-    render_word_bbox: bool
-    render_space_bbox: bool
+    render_paragraph_bbox: bool = False
+    render_baseline: bool = False
+    render_triangle: bool = False
+    render_line_bbox: bool = False
+    render_word_bbox: bool = False
+    render_space_bbox: bool = False
 
 
 class HocrTransformError(Exception):
@@ -81,8 +81,22 @@ class HocrTransform:
         debug: bool = False,
         fontname: Name = Name("/f-0-0"),
         font: Font = GlyphlessFont(),
+        debug_render_options: DebugRenderOptions | None = None,
     ):
         """Initialize the HocrTransform object."""
+
+        if debug:
+            log.warning("Use debug_render_options instead", DeprecationWarning)
+            self.render_options = DebugRenderOptions(
+                render_baseline=debug,
+                render_triangle=debug,
+                render_line_bbox=False,
+                render_word_bbox=debug,
+                render_paragraph_bbox=False,
+                render_space_bbox=False,
+            )
+        else:
+            self.render_options = debug_render_options or DebugRenderOptions()
         self.dpi = dpi
         self.hocr = ElementTree.parse(os.fspath(hocr_filename))
         self._fontname = fontname
@@ -103,14 +117,6 @@ class HocrTransform:
             self.height = (coords.ury - coords.lly) / (self.dpi / INCH)
             # Stop after first div that has page coordinates
             break
-        self.render_options = DebugRenderOptions(
-            render_baseline=debug,
-            render_triangle=debug,
-            render_line_bbox=False,
-            render_word_bbox=debug,
-            render_paragraph_bbox=False,
-            render_space_bbox=False,
-        )
 
     def _get_element_text(self, element: Element):
         """Return the textual content of the element and its children."""
@@ -215,12 +221,13 @@ class HocrTransform:
             if not found_lines:
                 # Tesseract did not report any lines (just words)
                 root = self.hocr.find(self._child_xpath('div', 'ocr_page'))
+                direction = self._get_text_direction(root)
                 self._do_line(
                     canvas,
                     root,
                     "ocrx_word",
                     invisible_text,
-                    TextDirection.LTR,
+                    direction,
                     True,
                 )
         # put the image on the page, scaled to fill the page
@@ -311,8 +318,7 @@ class HocrTransform:
             line_box_height = abs(line_box.height) / cos(angle)
             fontsize = line_box_height + intercept
             text.font(self._fontname, fontsize)
-            if invisible_text or True:
-                text.render_mode(3)  # Invisible (indicates OCR text)
+            text.render_mode(3 if invisible_text else 0)
 
             self._debug_draw_baseline(
                 canvas, line_matrix.inverse().transform(line_box), 0
@@ -362,10 +368,15 @@ class HocrTransform:
         self._debug_draw_word_bbox(canvas, box)
 
         # If this word is 0 units wide, our best bet seems to be to suppress this text
+        if text_direction == TextDirection.RTL:
+            log.info("RTL: %s", elemtxt)
         if font_width > 0:
-            text.text_transform(Matrix(1, 0, 0, 1, box.llx, 0))
+            if text_direction == TextDirection.LTR:
+                text.text_transform(Matrix(1, 0, 0, -1, box.llx, 0))
+            elif text_direction == TextDirection.RTL:
+                text.text_transform(Matrix(-1, 0, 0, -1, box.llx + box.width, 0))
             text.horiz_scale(100 * box.width / font_width)
-            text.show(elemtxt.encode('utf-16be'))
+            text.show(self._font.text_encode(elemtxt))
 
         # Get coordinates of the next word (if there is one)
         hocr_next_box = (
@@ -385,11 +396,16 @@ class HocrTransform:
         elif text_direction == TextDirection.RTL:
             space_box = Rectangle(next_box.urx, box.lly, box.llx, next_box.ury)
         self._debug_draw_space_bbox(canvas, space_box)
-        text.text_transform(Matrix(1, 0, 0, 1, space_box.llx, 0))
         space_width = self._font.text_width(' ', fontsize)
         if space_width > 0:
+            if text_direction == TextDirection.LTR:
+                text.text_transform(Matrix(1, 0, 0, -1, space_box.llx, 0))
+            elif text_direction == TextDirection.RTL:
+                text.text_transform(
+                    Matrix(-1, 0, 0, -1, space_box.llx + space_box.width, 0)
+                )
             text.horiz_scale(100 * space_box.width / space_width)
-            text.show(' '.encode('utf-16be'))
+            text.show(self._font.text_encode(' '))
 
     def _debug_draw_paragraph_boxes(self, canvas: Canvas, color=CYAN):
         """Draw boxes around paragraphs in the document."""

tests/cache/2400dpi/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,24 @@
+i a la Waterman
+
+4 ons linzen
+
+3 liter water
+
+3 uien
+
+bloem, boter
+
+2 kopjes melk
+
+laurier, kruidnagel, kerrie, zout
+
+De linzgen wassen en in-l liter kokend wa-
+ter 1 dag laten weken, 2 liter water bij
+de linzen voegen, zonder het water waarin
+ze geweekt zijn af te gieten, De helft van
+de uien bakken met laurier en Kruidnagel.
+Alle uien, kerrie en zgout bij de linzen
+voegen, Alles aan de kook brengen, Van de
+bloem met boter en melk een papje maken en
+verder afmaken met de soep, Als de linzen
+gfgaar Zijn is de soep klaar.

tests/cache/3small/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,15 @@
+Tarnose
+
+Bokale oa
+
+Lehuntze
+
+Mugerre
+
+Milafranga Komunikabideak
+
+BAIONA  zeiteninsiie —
+
+7 Trenbideak -----
+
+t\ Basusarri — spmeans:20141004 ae: . _ ~

tests/cache/3small/__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1 @@
+Covfefe is a perfectly cromulent word.

tests/cache/3small/__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,24 @@
+Linzensoep a la Waterman
+
+4 ons linzen
+
+3 liter water
+
+3 uien
+
+bloem, boter
+
+2 kopjes melk
+
+laurier, kruidnagel, kerrie, zout
+
+De linzgen wassen en in-l liter kokend wa-
+ter 1 dag laten weken, 2 liter water bij
+de linzen voegen, zonder het water waarin
+ze geweekt zijn af te gieten, De helft van
+de uien bakken met laurier en Kruidnagel.
+Alle uien, kerrie en zgout bij de linzen
+voegen, Alles aan de kook brengen, Van de
+bloem met boter en melk een papje maken en
+verder afmaken met de soep, Als de linzen
+gfgaar Zijn is de soep klaar.

tests/cache/francais/__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,12 @@
+Portez ce vieux whisky au juge
+blond qui fume sur son Ile
+interieure, a cöte de l'alcöve
+ovoide, oU les büches se
+consument dans l'ätre, ce qui
+ui permet de penser & la
+caenogenese de |'etre dont il
+est question dans la cause
+ambigu& entendue a MoY, dans
+un capharnaüm qui, pense-t-il,
+diminue ca et la la qualit& de son
+ceuvre.

tests/cache/graph_ocred/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,22 @@
+600
+500 .
+400 _ EB fp ys
+300 —f— EN / ~/ Y
+Tg ANA a3
+100 - nee -eemiae
+0-+—-* a ee ee ee eee
+S @ s > oO © ve S ) @ + & & “A 5 oo
+Se ° Ps as ge eS € x ro NS Po e? & s AS
+a a © FF SF HY HK SK BM ee sO
+e < Na ‘ : > cy xs eS @ Ww a) Oo
+ee S&S FF SF HK S e © 4
+~ & e
+3 x
+
+—¢—Support
+
+—H— No vote[note 1]
+ir Oppose
+—=—Net[note 2]
+
+re Percentage [note 3]

tests/cache/manifest.jsonl

@@ -71,3 +71,15 @@
 {"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.6", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
 {"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.6", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/multipage.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
 {"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.6", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_hocr__hocr__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_hocr", "hocr", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__deu__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/francais.pdf", "args": ["-l", "deu", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__--oem__1__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/trivial.pdf", "args": ["-l", "eng", "--oem", "1", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/trivial.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/2400dpi.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__thresholding_method=1__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/trivial.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "-c", "thresholding_method=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__thresholding_method=2__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/trivial.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "-c", "thresholding_method=2", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/graph_ocred.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000002_ocr.png__000002_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000002_ocr.png", "$TMPDIR/000002_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000003_ocr.png__000003_ocr_tess__pdf__txt", "sourcefile": "resources/3small.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000003_ocr.png", "$TMPDIR/000003_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}
+{"tesseract_version": "5.3.2", "system": "Linux", "python": "3.11.7", "argv_slug": "__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt", "sourcefile": "resources/poster.pdf", "args": ["-l", "eng", "-c", "textonly_pdf=1", "$TMPDIR/000001_ocr.png", "$TMPDIR/000001_ocr_tess", "pdf", "txt"]}

tests/cache/poster/__-l__eng__000001_ocr.png__000001_ocr_tess__pdf__txt/txt.bin

@@ -0,0 +1,118 @@
+The LinnSequencer
+32 Track MIDI Sequence Recorder
+
+The LinnSequencer is a state-of-the-art composition and performance tool for the professional musician. It is
+extremely powerful, yet amazingly simple to learn and use. It’s many remarkable features include:
+
+© Operation is similar to multi-track tape recorder with PLAY, STOP, RECORD, FAST
+FORWARD, REWIND, and LOCATE controls.
+
+e Each of the 100 sequences contains 32 simultaneous, polyphonic tracks. Each track may
+be assigned to one of 16 MIDI channels. Simultaneously plays up to 16 polyphonic
+
+synthesizers!
+
+© Ultra-fast 3!” disk drive stores complex songs in seconds and holds over 110,000 notes
+
+per disk!
+
+¢ One or all tracks may be TRANSPOSED at the touch of a key.
+e Exclusive real-time ERASE function makes editing FAST.
+e Exclusive REPEAT function automatically repeats any held notes at a pre-selected
+
+rhythmic value.
+
+¢ TIMING CORRECTION works during playback and operates without ‘chopping’ notes.
+
+¢ Optional SMPTE time code synchronization.
+
+¢ Optional remote control.
+
+Recording a Sequence
+
+To record a sequence, simply press RECORD and PLAY,
+then play your MIDI keyboard in time to the Sequencer’s
+click track. When the sequence loops back around to bar 1,
+you’ ll hear what you played—only all timing errors will be
+corrected! (Timing correction may be adjusted or defeated).
+Any additional notes played will be added into the track
+—existing notes are not erased while recording!
+
+FAST FORWARD, REWIND, and LOCATE controls
+may be used at any time to quickly access any location in
+your sequence for spot-recording. To overdub a new part,
+select a different track and start recording—while you
+record, the first track will play in perfect sync (unless you
+MUTE it, or SOLO another track). In this way, up to 32
+tracks may be overdubbed! All MIDI effects are recorded
+including pitch bend, modulation, velocity, aftertouch,
+sustain pedal, and program changes!
+
+Editing
+
+To erase a wrong note, simply hold ERASE and press
+the note to be erased just before it plays in the sequence—
+when played back, it will be gone. Notes may also be
+added, erased, or changed using the SINGLE STEP func-
+tion. To overdub notes at specific points within a sequence,
+
+Additional Features
+
+simply use LOCATE, FAST FORWARD, or REWIND to
+find the desired bar number, then start recording.
+
+The INSERT/COPY function allows you to move bars
+from one location to another—in the same sequence or a
+different one. For example, you might insert a copy of the
+first verse between the second chorus and the bridge.
+DELETE BARS operates the same way to remove
+unwanted sections.
+
+Creating a Song
+
+One way to create a song is to record each track all the
+way through (up to 999 bars). Another way is to record
+each basic section (verse, chorus, etc.) in individual
+sequences, then use the CREATE SONG function to “chain”
+them together. CREATE SONG will then automatically
+copy all the parts into a new sequence. If desired, you can
+even set the last few bars to repeat infinitely, for a fadeout.
+
+Composition Without Compromise
+
+The technology you use should never be so complex that
+it interferes with the creative process. That’s precisely why
+the LinnSequencer is designed to let you compose, record
+and edit while devoting your undivided attention to your
+music. See your Linn dealer today for a demonstration!
+
+¢ Simple, easy to learn operation—the 32 character LCD display clearly guides you through all operations. If needed, the
+
+HELP button displays additional explanations.
+
+® Non-destructive recording—existing notes are not erased while recording.
+© Two FOOTSWITCH INPUTS may be assigned to remotely control many of the commonly used functions, including
+
+ERASE, REPEAT, PLAY/STOP, or LOCATE.
+
+¢ Two TRIGGER OUTPUTS may be programmed to output pulses at any selected note value.
+
+@ Will sync to standard LinnDrum or Linn 9000 sync tone.
+
+© Utilizes ultra high-speed, 8 MHz 80186 16 bit computer internally for FAST operation.
+¢ TEMPO may be specified in BEATS-PER-MINUTE or FRAMES-PER-BEAT at 24, 25, or 30 frames per second,
+
+(even drop frame!)
+
+¢ TEMPO may be entered numerically, adjustable in tenths of a Beat-Per-Minute increments, or by tapping quarter notes
+
+on the TAP TEMPO button.
+
+¢ TEMPO CHANGES may be programmed into a sequence, with smooth transitions if desired.
+e Any TIME SIGNATURE may be used, and may be changed within a song.
+
+linn
+Linn Electronics, Inc.
+
+18720 Oxnard Street, Tarzana, CA 91356
+(818) 708-8131 TELEX #298949 LINN UR

tests/test_concurrency.py

@@ -4,6 +4,7 @@
 from __future__ import annotations
 
 import os
+import platform
 
 import pytest
 
@@ -13,6 +14,9 @@ from .conftest import run_ocrmypdf_api
 
 
 @pytest.mark.skipif(os.name == 'nt', reason="Windows doesn't have SIGKILL")
+@pytest.mark.skipif(
+    platform.python_version_tuple() >= ('3', '12'), reason="can deadlock due to fork"
+)
 def test_simulate_oom_killer(multipage, no_outpdf):
     exitcode = run_ocrmypdf_api(
         multipage,