From 0436ef95937185c8221acea84f00d2f3913d074e Mon Sep 17 00:00:00 2001
From: AuroraWright <AuroraWright@users.noreply.github.com>
Date: Mon, 6 Oct 2025 02:07:44 +0200
Subject: [PATCH] Add optional line text field to coordinate format (helps with
 text output format in some languages/corner cases)

---
 owocr/ocr.py |  4 ++++
 owocr/run.py | 15 +++++++++------
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/owocr/ocr.py b/owocr/ocr.py
index e744c4b..43a1235 100644
--- a/owocr/ocr.py
+++ b/owocr/ocr.py
@@ -110,6 +110,7 @@ class Line:
     """Represents a single line of text, composed of words."""
     bounding_box: BoundingBox
     words: List[Word] = field(default_factory=list)
+    text: Optional[str] = None
 
 @dataclass
 class Paragraph:
@@ -567,6 +568,7 @@ class Bing:
                             words.append(word)
 
                         line = Line(
+                            text=l.get('text', ''),
                             bounding_box=self._quad_to_center_bbox(l['boundingBox']),
                             words=words
                         )
@@ -840,6 +842,7 @@ class AppleLiveText:
                 
                 l_bbox = l.quad().boundingBox()
                 line = Line(
+                    text=l.string(),
                     bounding_box=BoundingBox(
                         width=l_bbox.size.width,
                         height=l_bbox.size.height,
@@ -996,6 +999,7 @@ class OneOCR:
                 words.append(word)
             
             line = Line(
+                text=l.get('text', ''),
                 bounding_box=self._pixel_quad_to_center_bbox(l['bounding_rect'], img_width, img_height),
                 words=words
             )
diff --git a/owocr/run.py b/owocr/run.py
index d4c2d0c..ca11c47 100644
--- a/owocr/run.py
+++ b/owocr/run.py
@@ -729,12 +729,15 @@ class OutputResult:
         full_text_parts = []
         for p in result_data.paragraphs:
             for l in p.lines:
-                for w in l.words:
-                    full_text_parts.append(w.text)
-                    if w.separator != None:
-                        full_text_parts.append(w.separator)
-                    else:
-                        full_text_parts.append(' ')
+                if l.text != None:
+                    full_text_parts.append(l.text)
+                else:
+                    for w in l.words:
+                        full_text_parts.append(w.text)
+                        if w.separator != None:
+                            full_text_parts.append(w.separator)
+                        else:
+                            full_text_parts.append(' ')
                 full_text_parts.append('\n')
         return "".join(full_text_parts)