python310Packages.ocrmypdf: 14.4.0 -> 15.0.1

Diff: https://github.com/ocrmypdf/OCRmyPDF/compare/v14.4.0...v15.0.1

Changelog: https://github.com/ocrmypdf/OCRmyPDF/blob/v15.0.1/docs/release_notes.rst

pkgs/development/python-modules/ocrmypdf/default.nix

@@ -25,14 +25,15 @@
 , tqdm
 , typing-extensions
 , unpaper
+, wheel
 , installShellFiles
 }:
 
 buildPythonPackage rec {
   pname = "ocrmypdf";
-  version = "14.4.0";
+  version = "15.0.1";
 
-  disabled = pythonOlder "3.8";
+  disabled = pythonOlder "3.9";
 
   format = "pyproject";
 
@@ -46,7 +47,7 @@ buildPythonPackage rec {
     postFetch = ''
       rm "$out/.git_archival.txt"
     '';
-    hash = "sha256-i1ZUBKR8dJXZkALUFwkzYcjtZ5Li66DfD2fupCGRQC4=";
+    hash = "sha256-WTUe4ivSEU/4KMRtKhf1MDXN1segIe/wrXYVA5zJrZw=";
   };
 
   SETUPTOOLS_SCM_PRETEND_VERSION = version;
@@ -65,6 +66,7 @@ buildPythonPackage rec {
   nativeBuildInputs = [
     setuptools
     setuptools-scm
+    wheel
     installShellFiles
   ];
 
@@ -78,9 +80,6 @@ buildPythonPackage rec {
     pluggy
     reportlab
     rich
-    tqdm
-  ] ++ lib.optionals (pythonOlder "3.9") [
-    importlib-resources
   ] ++ lib.optionals (pythonOlder "3.10") [
     typing-extensions
   ];

pkgs/development/python-modules/ocrmypdf/paths.patch

@@ -1,69 +1,61 @@
 diff --git a/src/ocrmypdf/_exec/ghostscript.py b/src/ocrmypdf/_exec/ghostscript.py
-index 9e21c33c..21f134b3 100644
+index e28d23e9..183cd918 100644
 --- a/src/ocrmypdf/_exec/ghostscript.py
 +++ b/src/ocrmypdf/_exec/ghostscript.py
-@@ -32,15 +32,7 @@ log = logging.getLogger(__name__)
+@@ -67,7 +67,7 @@ log.addFilter(DuplicateFilter(log))
- # Most reliable what to get the bitness of Python interpreter, according to Python docs
- _IS_64BIT = sys.maxsize > 2**32
  
--_GSWIN = None
+ 
--if os.name == 'nt':
+ # Ghostscript executable - gswin32c is not supported
--    if _IS_64BIT:
+-GS = 'gswin64c' if os.name == 'nt' else 'gs'
--        _GSWIN = 'gswin64c'
--    else:
--        _GSWIN = 'gswin32c'
--
--GS = _GSWIN if _GSWIN else 'gs'
--del _GSWIN
 +GS = '@gs@'
  
  
- def version():
+ def version() -> Version:
 diff --git a/src/ocrmypdf/_exec/jbig2enc.py b/src/ocrmypdf/_exec/jbig2enc.py
-index 0f8f7392..db792b10 100644
+index 28d3dd1c..477a7d6c 100644
 --- a/src/ocrmypdf/_exec/jbig2enc.py
 +++ b/src/ocrmypdf/_exec/jbig2enc.py
-@@ -12,7 +12,7 @@ from ocrmypdf.subprocess import get_version, run
+@@ -14,7 +14,7 @@ from ocrmypdf.subprocess import get_version, run
  
  
- def version():
+ def version() -> Version:
--    return get_version('jbig2', regex=r'jbig2enc (\d+(\.\d+)*).*')
+-    return Version(get_version('jbig2', regex=r'jbig2enc (\d+(\.\d+)*).*'))
-+    return get_version('@jbig2@', regex=r'jbig2enc (\d+(\.\d+)*).*')
++    return Version(get_version('@jbig2@', regex=r'jbig2enc (\d+(\.\d+)*).*'))
  
  
  def available():
-@@ -25,7 +25,7 @@ def available():
+@@ -27,7 +27,7 @@ def available():
  
- def convert_group(*, cwd, infiles, out_prefix):
+ def convert_group(*, cwd, infiles, out_prefix, threshold):
      args = [
 -        'jbig2',
 +        '@jbig2@',
          '-b',
          out_prefix,
-         '-s',  # symbol mode (lossy)
+         '--symbol-mode',  # symbol mode (lossy)
-@@ -44,7 +44,7 @@ def convert_group_mp(args):
+@@ -50,7 +50,7 @@ def convert_group_mp(args):
  
  
- def convert_single(*, cwd, infile, outfile):
+ def convert_single(*, cwd, infile, outfile, threshold):
--    args = ['jbig2', '-p', infile]
+-    args = ['jbig2', '--pdf', '-t', str(threshold), infile]
-+    args = ['@jbig2@', '-p', infile]
++    args = ['@jbig2@', '--pdf', '-t', str(threshold), infile]
      with open(outfile, 'wb') as fstdout:
          proc = run(args, cwd=cwd, stdout=fstdout, stderr=PIPE)
      proc.check_returncode()
 diff --git a/src/ocrmypdf/_exec/pngquant.py b/src/ocrmypdf/_exec/pngquant.py
-index 64e91139..ab5b9c49 100644
+index 8425caec..a027f041 100644
 --- a/src/ocrmypdf/_exec/pngquant.py
 +++ b/src/ocrmypdf/_exec/pngquant.py
-@@ -17,7 +17,7 @@ from ocrmypdf.subprocess import get_version, run
+@@ -18,7 +18,7 @@ from ocrmypdf.subprocess import get_version, run
  
  
- def version():
+ def version() -> Version:
--    return get_version('pngquant', regex=r'(\d+(\.\d+)*).*')
+-    return Version(get_version('pngquant', regex=r'(\d+(\.\d+)*).*'))
-+    return get_version('@pngquant@', regex=r'(\d+(\.\d+)*).*')
++    return Version(get_version('@pngquant@', regex=r'(\d+(\.\d+)*).*'))
  
  
  def available():
-@@ -44,7 +44,7 @@ def input_as_png(input_file: Path):
+@@ -45,7 +45,7 @@ def input_as_png(input_file: Path):
  def quantize(input_file: Path, output_file: Path, quality_min: int, quality_max: int):
      with input_as_png(input_file) as input_stream:
          args = [
@@ -73,19 +65,19 @@ index 64e91139..ab5b9c49 100644
              '--skip-if-larger',
              '--quality',
 diff --git a/src/ocrmypdf/_exec/tesseract.py b/src/ocrmypdf/_exec/tesseract.py
-index ad98836a..a12d3002 100644
+index 4eac3470..61315d14 100644
 --- a/src/ocrmypdf/_exec/tesseract.py
 +++ b/src/ocrmypdf/_exec/tesseract.py
-@@ -111,7 +111,7 @@ class TesseractVersion(Version):
+@@ -114,7 +114,7 @@ class TesseractVersion(Version):
  
  
- def version() -> str:
+ def version() -> Version:
--    return get_version('tesseract', regex=r'tesseract\s(.+)')
+-    return TesseractVersion(get_version('tesseract', regex=r'tesseract\s(.+)'))
-+    return get_version('@tesseract@', regex=r'tesseract\s(.+)')
++    return TesseractVersion(get_version('@tesseract@', regex=r'tesseract\s(.+)'))
  
  
- def has_user_words() -> bool:
+ def has_thresholding() -> bool:
-@@ -138,7 +138,7 @@ def get_languages() -> set[str]:
+@@ -132,7 +132,7 @@ def get_languages() -> set[str]:
          msg += output
          return msg
  
@@ -94,7 +86,7 @@ index ad98836a..a12d3002 100644
      try:
          proc = run(
              args_tess,
-@@ -160,7 +160,7 @@ def get_languages() -> set[str]:
+@@ -154,7 +154,7 @@ def get_languages() -> set[str]:
  
  
  def tess_base_args(langs: list[str], engine_mode: int | None) -> list[str]:
@@ -104,19 +96,19 @@ index ad98836a..a12d3002 100644
          args.extend(['-l', '+'.join(langs)])
      if engine_mode is not None:
 diff --git a/src/ocrmypdf/_exec/unpaper.py b/src/ocrmypdf/_exec/unpaper.py
-index d7f24265..d14f85de 100644
+index 2944b4f4..55a36140 100644
 --- a/src/ocrmypdf/_exec/unpaper.py
 +++ b/src/ocrmypdf/_exec/unpaper.py
-@@ -66,7 +66,7 @@ class UnpaperImageTooLargeError(Exception):
+@@ -69,7 +69,7 @@ class UnpaperImageTooLargeError(Exception):
  
  
- def version() -> str:
+ def version() -> Version:
--    return get_version('unpaper')
+-    return Version(get_version('unpaper'))
-+    return get_version('@unpaper@')
++    return Version(get_version('@unpaper@'))
  
  
  SUPPORTED_MODES = {'1', 'L', 'RGB'}
-@@ -120,7 +120,7 @@ def _setup_unpaper_io(input_file: Path) -> Iterator[tuple[Path, Path, Path]]:
+@@ -123,7 +123,7 @@ def _setup_unpaper_io(input_file: Path) -> Iterator[tuple[Path, Path, Path]]:
  def run_unpaper(
      input_file: Path, output_file: Path, *, dpi: DecFloat, mode_args: list[str]
  ) -> None: