BUG: Missing spaces in extract_text() method (#1328) (#2868)

* BUG: Missing spaces in extract_text() method (#1328) * Revert "BUG: Missing spaces in extract_text() method (#1328)" This reverts commit 5400f5a. * BUG: Missing spaces in extract_text() method (#1328) * BUG: Missing spaces in extract_text() method (#1328) add test * Revert "BUG: Missing spaces in extract_text() method (#1328)" This reverts commit 5400f5a. BUG: Missing spaces in extract_text() method (#1328) BUG: Missing spaces in extract_text() method (#1328) add test * BUG: Missing spaces in extract_text() method (#1328) Convert font size comparison to ratio * Correction to new file URL. Co-authored-by: Stefan <96178532+stefan6419846@users.noreply.github.com> * BUG: Missing spaces in extract_text() method (#1328) calculation efficiency * BUG: Missing spaces in extract_text() method (#1328) Simplify the assertion process --------- Co-authored-by: Stefan <96178532+stefan6419846@users.noreply.github.com>
py-pdf · Sep 24, 2024 · 635a7c1 · 635a7c1
1 parent d974d5c
commit 635a7c1
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 1 deletion.
diff --git a/pypdf/_page.py b/pypdf/_page.py
@@ -1985,11 +1985,13 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
  process_operation(b"TL", [-operands[1]])
  process_operation(b"Td", operands)
  elif operator == b"TJ":
+ # The space width may be smaller than the font width, so the width should be 95%.
+ _confirm_space_width = _space_width * 0.95
  for op in operands[0]:
  if isinstance(op, (str, bytes)):
  process_operation(b"Tj", [op])
  if isinstance(op, (int, float, NumberObject, FloatObject)) and (
- (abs(float(op)) >= _space_width)
+ (abs(float(op)) >= _confirm_space_width)
  and (len(text) > 0)
  and (text[-1] != " ")
  ):

diff --git a/tests/test_text_extraction.py b/tests/test_text_extraction.py
@@ -189,3 +189,14 @@ def test_layout_mode_warnings(mock_logger_warning):
  mock_logger_warning.assert_called_with(
  "Argument visitor_text is ignored in layout mode", "pypdf._page"
  )
+
+
+@pytest.mark.enable_socket()
+def test_space_with_one_unit_smaller_than_font_width():
+ """Tests for #1328"""
+ url = "https://github.com/py-pdf/pypdf/files/9498481/0004.pdf"
+ name = "iss1328.pdf"
+ reader = PdfReader(BytesIO(get_data_from_url(url, name=name)))
+ page = reader.pages[0]
+ extracted = page.extract_text()
+ assert "Reporting crude oil leak.\n" in extracted