Avoid division by 0

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "pdf-document-layout-analysis"
-version = "2024.12.17.5"
+version = "2024.12.17.7"
 description = "This tool is for PDF document layout analysis"
 license = { file = "LICENSE" }
 authors = [{ name = "HURIDOCS" }]

src/pdf_features/Rectangle.py

@@ -25,6 +25,9 @@ def from_poppler_tag_etree(tag: ElementBase) -> "Rectangle":
         x_max = x_min + int(tag.attrib["width"])
         y_max = y_min + int(tag.attrib["height"])
 
+        if len(content) <= 1:
+            return Rectangle(x_min, y_min, x_max, y_max)
+
         one_character_length = max(int((x_max - x_min) / len(content)), 2)
         if content[0] == " ":
             x_min += one_character_length

src/test_end_to_end.py

@@ -54,7 +54,7 @@ def test_regular_pdf(self):
             self.assertEqual(expected_content, results_list[0]["text"])
             self.assertEqual(157, results_list[0]["left"])
             self.assertEqual(105, results_list[0]["top"])
-            self.assertEqual(283, results_list[0]["width"])
+            self.assertEqual(282, results_list[0]["width"])
             self.assertEqual(36, results_list[0]["height"])
             self.assertEqual(1, results_list[0]["page_number"])
             self.assertEqual(595, results_list[0]["page_width"])
@@ -112,7 +112,7 @@ def test_regular_pdf_fast(self):
         self.assertEqual(expected_content, results_list[0]["text"])
         self.assertEqual(157, results_list[0]["left"])
         self.assertEqual(106, results_list[0]["top"])
-        self.assertEqual(284, results_list[0]["width"])
+        self.assertEqual(278, results_list[0]["width"])
         self.assertEqual(24, results_list[0]["height"])
         self.assertEqual(1, results_list[0]["page_number"])
         self.assertEqual(595, results_list[0]["page_width"])