Replace leptonica deskew with tesseract find skew and pillow rotate

Also rebuild the cache.

src/ocrmypdf/_exec/tesseract.py

@@ -13,7 +13,7 @@ from distutils.version import StrictVersion
 from os import fspath
 from pathlib import Path
 from subprocess import PIPE, STDOUT, CalledProcessError, TimeoutExpired
-from typing import List, Optional
+from typing import Dict, Iterator, List, Optional
 
 from PIL import Image
 
@@ -126,6 +126,17 @@ def tess_base_args(langs: List[str], engine_mode: Optional[int]) -> List[str]:
     return args
 
 
+def _parse_tesseract_output(binary_output: bytes) -> Dict[str, str]:
+    def g():
+        for line in binary_output.decode().splitlines():
+            line = line.strip()
+            parts = line.split(':', maxsplit=2)
+            if len(parts) == 2:
+                yield parts[0].strip(), parts[1].strip()
+
+    return {k: v for k, v in g()}
+
+
 def get_orientation(
     input_file: Path, engine_mode: Optional[int], timeout: float
 ) -> OrientationConfidence:
@@ -138,7 +149,6 @@ def get_orientation(
 
     try:
         p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
-        stdout = p.stdout
     except TimeoutExpired:
         return OrientationConfidence(angle=0, confidence=0.0)
     except CalledProcessError as e:
@@ -150,19 +160,41 @@ def get_orientation(
         ):
             return OrientationConfidence(0, 0)
         raise SubprocessOutputError() from e
-    else:
-        osd = {}
-        for line in stdout.decode().splitlines():
-            line = line.strip()
-            parts = line.split(':', maxsplit=2)
-            if len(parts) == 2:
-                osd[parts[0].strip()] = parts[1].strip()
 
-        angle = int(osd.get('Orientation in degrees', 0))
-        oc = OrientationConfidence(
-            angle=angle, confidence=float(osd.get('Orientation confidence', 0))
-        )
-        return oc
+    osd = _parse_tesseract_output(p.stdout)
+    angle = int(osd.get('Orientation in degrees', 0))
+    oc = OrientationConfidence(
+        angle=angle, confidence=float(osd.get('Orientation confidence', 0))
+    )
+    return oc
+
+
+def get_deskew(
+    input_file: Path, languages: List[str], engine_mode: Optional[int], timeout: float
+) -> float:
+    """Gets angle to deskew this page, in radians."""
+    args_tesseract = tess_base_args(languages, engine_mode) + [
+        '--psm',
+        '2',
+        fspath(input_file),
+        'stdout',
+    ]
+
+    try:
+        p = run(args_tesseract, stdout=PIPE, stderr=STDOUT, timeout=timeout, check=True)
+    except TimeoutExpired:
+        return 0.0
+    except CalledProcessError as e:
+        tesseract_log_output(e.stdout)
+        tesseract_log_output(e.stderr)
+        if b'Empty page!!' in e.output:  # Not enough info for a skew angle
+            return 0.0
+
+        raise SubprocessOutputError() from e
+
+    parsed = _parse_tesseract_output(p.stdout)
+    deskew = float(parsed.get('Deskew angle', 0))
+    return deskew
 
 
 def tesseract_log_output(stream):

src/ocrmypdf/_pipeline.py

@@ -6,6 +6,7 @@
 
 
 import logging
+import math
 import os
 import re
 import sys
@@ -476,7 +477,20 @@ def preprocess_remove_background(input_file: Path, page_context: PageContext):
 def preprocess_deskew(input_file: Path, page_context: PageContext):
     output_file = page_context.get_path('pp_deskew.png')
     dpi = get_page_square_dpi(page_context.pageinfo, page_context.options)
-    leptonica.deskew(input_file, output_file, dpi.x)
+
+    ocr_engine = page_context.plugin_manager.hook.get_ocr_engine()
+    deskew_angle = ocr_engine.get_deskew(input_file, page_context.options)
+
+    deskew_angle_degrees = deskew_angle * 180.0 / math.pi
+
+    with Image.open(input_file) as im:
+        # According to Pillow docs, .rotate() will automatically use Image.NEAREST
+        # resampling if image is mode '1' or 'P'
+        deskewed = im.rotate(
+            deskew_angle_degrees, resample=Image.BICUBIC, fillcolor='white'
+        )
+        deskewed.save(output_file, dpi=dpi)
+
     return output_file
 
 

src/ocrmypdf/builtin_plugins/tesseract_ocr.py

@@ -142,6 +142,15 @@ class TesseractOcrEngine(OcrEngine):
             timeout=options.tesseract_timeout,
         )
 
+    @staticmethod
+    def get_deskew(input_file, options) -> float:
+        return tesseract.get_deskew(
+            input_file,
+            languages=options.languages,
+            engine_mode=options.tesseract_oem,
+            timeout=options.tesseract_timeout,
+        )
+
     @staticmethod
     def generate_hocr(input_file, output_hocr, output_text, options):
         tesseract.generate_hocr(

src/ocrmypdf/pluginspec.py

@@ -366,6 +366,11 @@ class OcrEngine(ABC):
     def get_orientation(input_file: Path, options: Namespace) -> OrientationConfidence:
         """Returns the orientation of the image."""
 
+    @staticmethod
+    def get_deskew(input_file: Path, options: Namespace) -> float:
+        """Returns the deskew angle of the image, in radians."""
+        return 0.0
+
     @staticmethod
     @abstractmethod
     def generate_hocr(