fix(pypdfium2): Fix OCR bounding box misalignment caused by mismatched rotation metadata (#2039)

* Fix OCR bounding box misalignment caused by rotation metadata

Signed-off-by: AndrewTsai0406 <tsai247365@gmail.com>

* Add rotation-mismatch scanned pdf test case

Signed-off-by: AndrewTsai0406 <tsai247365@gmail.com>

* add ground truth for ocr_test_rotation_mismatch.pdf

Signed-off-by: AndrewTsai0406 <tsai247365@gmail.com>

* add ground truth for ocr_test_rotation_mismatch.pdf

Signed-off-by: AndrewTsai0406 <tsai247365@gmail.com>

* Updated test GT and merged from main

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

* Fix OCR test by excluding mismatched rotation example

Signed-off-by: Christoph Auer <cau@zurich.ibm.com>

---------

Signed-off-by: AndrewTsai0406 <tsai247365@gmail.com>
Signed-off-by: Christoph Auer <cau@zurich.ibm.com>
Co-authored-by: Christoph Auer <cau@zurich.ibm.com>

tests/test_backend_pdfium.py

@@ -9,6 +9,8 @@ from docling.backend.pypdfium2_backend import (
 )
 from docling.datamodel.base_models import InputFormat
 from docling.datamodel.document import InputDocument
+from docling.datamodel.pipeline_options import PdfPipelineOptions
+from docling.document_converter import DocumentConverter, PdfFormatOption
 
 
 @pytest.fixture
@@ -27,6 +29,23 @@ def _get_backend(pdf_doc):
     return doc_backend
 
 
+def test_get_text_from_rect_rotated():
+    pdf_doc = Path("./tests/data_scanned/sample_with_rotation_mismatch.pdf")
+    pipeline_options = PdfPipelineOptions()
+    pipeline_options.do_ocr = True
+
+    doc_converter = DocumentConverter(
+        format_options={
+            InputFormat.PDF: PdfFormatOption(
+                pipeline_options=pipeline_options, backend=PyPdfiumDocumentBackend
+            )
+        }
+    )
+    conv_res = doc_converter.convert(pdf_doc)
+
+    assert "1972" in conv_res.document.export_to_markdown()
+
+
 def test_text_cell_counts():
     pdf_doc = Path("./tests/data/pdf/redp5110_sampled.pdf")
 

tests/test_e2e_ocr_conversion.py

@@ -31,7 +31,8 @@ def get_pdf_paths():
     directory = Path("./tests/data_scanned")
 
     # List all PDF files in the directory and its subdirectories
-    pdf_files = sorted(directory.rglob("*.pdf"))
+    pdf_files = sorted(directory.rglob("ocr_test*.pdf"))
+
     return pdf_files