From 18d2ec67612597837b2215ef4665ed9c81891eda Mon Sep 17 00:00:00 2001
From: Benoit Pierre <benoit.pierre@gmail.com>
Date: Mon, 11 Nov 2024 12:44:22 +0100
Subject: [PATCH] kopt: fix OCR segmentation mode (#12726)

Previously unused by `libk2pdfopt`, the `ocr_type` argument passed to `k2pdfopt_tocr_single_word`
and forwarded to `ocrtess_ocrwords_from_bmp8` now has a big impact for some languages (e.g. Arabic).
---
 frontend/document/koptinterface.lua | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/frontend/document/koptinterface.lua b/frontend/document/koptinterface.lua
index 4d84748aa..f64a28a1a 100644
--- a/frontend/document/koptinterface.lua
+++ b/frontend/document/koptinterface.lua
@@ -24,7 +24,7 @@ local KoptInterface = {
     -- in `$TESSDATA_PREFIX/` on more recent versions).
     tessocr_data = not os.getenv('TESSDATA_PREFIX') and DataStorage:getDataDir().."/data/tessdata" or nil,
     ocr_lang = "eng",
-    ocr_type = 3, -- default 0, for more accuracy use 3
+    ocr_type = -1, -- default: assume a single uniform block of text.
     last_context_size = nil,
     default_context_size = 1024*1024,
 }