python3Packages.ocrmypdf: 16.13.0 -> 17.0.1 (#485418)

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

@@ -3,6 +3,7 @@
   buildPythonPackage,
   deprecation,
   fetchFromGitHub,
+  fpdf2,
   ghostscript_headless,
   hatch-vcs,
   hatchling,
@@ -16,19 +17,22 @@
   pillow,
   pluggy,
   pngquant,
+  pydantic,
+  pypdfium2,
   pytest-xdist,
   pytestCheckHook,
   rich,
   reportlab,
   replaceVars,
   tesseract,
+  uharfbuzz,
   unpaper,
   installShellFiles,
 }:
 
 buildPythonPackage rec {
   pname = "ocrmypdf";
-  version = "16.13.0";
+  version = "17.0.1";
   pyproject = true;
 
   src = fetchFromGitHub {
@@ -41,7 +45,7 @@ buildPythonPackage rec {
     postFetch = ''
       rm "$out/.git_archival.txt"
     '';
-    hash = "sha256-xxVtncIQ72echi0VogfgqwfB8IA7JEKVUV2lmL1coeU=";
+    hash = "sha256-wqauXBdy5kCCsuuM+3HKztl2jqwzF4BNFvYUElRzFm0=";
   };
 
   patches = [
@@ -64,6 +68,7 @@ buildPythonPackage rec {
 
   dependencies = [
     deprecation
+    fpdf2
     img2pdf
     packaging
     pdfminer-six
@@ -71,7 +76,10 @@ buildPythonPackage rec {
     pikepdf
     pillow
     pluggy
+    pydantic
+    pypdfium2
     rich
+    uharfbuzz
   ];
 
   nativeCheckInputs = [

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

@@ -1,8 +1,8 @@
 diff --git a/src/ocrmypdf/_exec/ghostscript.py b/src/ocrmypdf/_exec/ghostscript.py
-index eaa48117..30201d97 100644
+index b48f2080..beabcf4b 100644
 --- a/src/ocrmypdf/_exec/ghostscript.py
 +++ b/src/ocrmypdf/_exec/ghostscript.py
-@@ -31,7 +31,7 @@ COLOR_CONVERSION_STRATEGIES = frozenset(
+@@ -35,7 +35,7 @@ COLOR_CONVERSION_STRATEGIES = frozenset(
      ]
  )
  # Ghostscript executable - gswin32c is not supported
@@ -12,7 +12,7 @@ index eaa48117..30201d97 100644
  
  log = logging.getLogger(__name__)
 diff --git a/src/ocrmypdf/_exec/jbig2enc.py b/src/ocrmypdf/_exec/jbig2enc.py
-index 1c6dd5fe..b689a091 100644
+index 736de67e..d8c24ebd 100644
 --- a/src/ocrmypdf/_exec/jbig2enc.py
 +++ b/src/ocrmypdf/_exec/jbig2enc.py
 @@ -15,7 +15,7 @@ from ocrmypdf.subprocess import get_version, run
@@ -24,16 +24,7 @@ index 1c6dd5fe..b689a091 100644
      except CalledProcessError as e:
          # TeX Live for Windows provides an incompatible jbig2.EXE which may
          # be on the PATH.
-@@ -33,7 +33,7 @@ def available():
- 
- def convert_group(cwd, infiles, out_prefix, threshold):
-     args = [
--        'jbig2',
-+        '@jbig2@',
-         '-b',
-         out_prefix,
-         '--symbol-mode',  # symbol mode (lossy)
-@@ -50,7 +50,7 @@ def convert_group(cwd, infiles, out_prefix, threshold):
+@@ -32,7 +32,7 @@ def available():
  
  
  def convert_single(cwd, infile, outfile, threshold):
@@ -65,10 +56,10 @@ index 5b8600d0..fcad771b 100644
              '--skip-if-larger',
              '--quality',
 diff --git a/src/ocrmypdf/_exec/tesseract.py b/src/ocrmypdf/_exec/tesseract.py
-index 102bdab8..bfef4400 100644
+index d41a0af7..b189b0de 100644
 --- a/src/ocrmypdf/_exec/tesseract.py
 +++ b/src/ocrmypdf/_exec/tesseract.py
-@@ -95,7 +95,7 @@ class TesseractVersion(Version):
+@@ -116,7 +116,7 @@ class TesseractVersion(Version):
  
  
  def version() -> Version:
@@ -77,7 +68,7 @@ index 102bdab8..bfef4400 100644
  
  
  def has_thresholding() -> bool:
-@@ -113,7 +113,7 @@ def get_languages() -> set[str]:
+@@ -134,7 +134,7 @@ def get_languages() -> set[str]:
          msg += output
          return msg
  
@@ -86,7 +77,7 @@ index 102bdab8..bfef4400 100644
      try:
          proc = run(
              args_tess,
-@@ -135,7 +135,7 @@ def get_languages() -> set[str]:
+@@ -156,7 +156,7 @@ def get_languages() -> set[str]:
  
  
  def tess_base_args(langs: list[str], engine_mode: int | None) -> list[str]: