From fc30cb8903c9c91ea64e0d0ee9302dc8ebb8c178 Mon Sep 17 00:00:00 2001
From: "James R. Barlow" <james@purplerock.ca>
Date: Tue, 23 Dec 2025 15:03:51 -0800
Subject: [PATCH] Revert "Work around Ghostscript 10.6.0 JPEG encoding issue by
 forcing optimization."

This reverts commit f4c6c8121ba8178ff3a1cb8f70037bbc3a31391b.

The issue is now resolved by correcting the encoidng issue directly.
---
 src/ocrmypdf/optimize.py | 23 ++++++++++-------------
 1 file changed, 10 insertions(+), 13 deletions(-)

diff --git a/src/ocrmypdf/optimize.py b/src/ocrmypdf/optimize.py
index 41a1a385..a7c6cded 100644
--- a/src/ocrmypdf/optimize.py
+++ b/src/ocrmypdf/optimize.py
@@ -17,7 +17,6 @@ from typing import Any, NamedTuple, NewType
 from zlib import compress
 
 import img2pdf
-from packaging.version import Version
 from pikepdf import (
     Dictionary,
     Name,
@@ -33,7 +32,7 @@ from pikepdf.models.image import HifiPrintImageNotTranscodableError
 from PIL import Image
 
 from ocrmypdf._concurrent import Executor, SerialExecutor
-from ocrmypdf._exec import ghostscript, jbig2enc, pngquant
+from ocrmypdf._exec import jbig2enc, pngquant
 from ocrmypdf._jobcontext import PdfContext
 from ocrmypdf._progressbar import ProgressBar
 from ocrmypdf.exceptions import OutputFileAccessError
@@ -191,16 +190,6 @@ def extract_image_jbig2(
     return None
 
 
-def _should_optimize_jpeg(options, filtdp):
-    if options.optimize >= 2:
-        return True
-    if options.optimize < 2 and ghostscript.version() >= Version('10.6.0'):
-        # Ghostscript 10.6.0+ introduced some sort of JPEG encoding issue.
-        # To resolve this, re-optimize the JPEG anyway.
-        return True
-    return False
-
-
 def extract_image_generic(
     *, pdf: Pdf, root: Path, image: Stream, xref: Xref, options
 ) -> XrefExt | None:
@@ -214,7 +203,15 @@ def extract_image_generic(
     if pim.bits_per_component == 1:
         return None
 
-    if filtdp[0] == Name.DCTDecode and _should_optimize_jpeg(options, filtdp):
+    if filtdp[0] == Name.DCTDecode and options.optimize >= 2:
+        # This is a simple heuristic derived from some training data, that has
+        # about a 70% chance of guessing whether the JPEG is high quality,
+        # and possibly recompressible, or not. The number itself doesn't mean
+        # anything.
+        # bytes_per_pixel = int(raw_jpeg.Length) / (w * h)
+        # jpeg_quality_estimate = 117.0 * (bytes_per_pixel ** 0.213)
+        # if jpeg_quality_estimate < 65:
+        #     return None
         try:
             imgname = root / f'{xref:08d}'
             with imgname.open('wb') as f: