fix: Respect document_timeout in new threaded StandardPdfPipeline (#2653)

* fix: Respect document_timeout in new threaded StandardPdfPipeline

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

* add test case to test_options

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

* fix: Make sure unprocessed pages are not getting into assemble_document

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

---------

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

tests/test_options.py

@@ -15,6 +15,7 @@ from docling.datamodel.pipeline_options import (
     TableFormerMode,
 )
 from docling.document_converter import DocumentConverter, PdfFormatOption
+from docling.pipeline.legacy_standard_pdf_pipeline import LegacyStandardPdfPipeline
 
 
 @pytest.fixture
@@ -118,6 +119,33 @@ def test_page_range(test_doc_path):
     assert doc_result.status == ConversionStatus.FAILURE
 
 
+def test_document_timeout(test_doc_path):
+    converter = DocumentConverter(
+        format_options={
+            InputFormat.PDF: PdfFormatOption(
+                pipeline_options=PdfPipelineOptions(document_timeout=1)
+            )
+        }
+    )
+    result = converter.convert(test_doc_path)
+    assert result.status == ConversionStatus.PARTIAL_SUCCESS, (
+        "Expected document timeout to be used"
+    )
+
+    converter = DocumentConverter(
+        format_options={
+            InputFormat.PDF: PdfFormatOption(
+                pipeline_options=PdfPipelineOptions(document_timeout=1),
+                pipeline_cls=LegacyStandardPdfPipeline,
+            )
+        }
+    )
+    result = converter.convert(test_doc_path)
+    assert result.status == ConversionStatus.PARTIAL_SUCCESS, (
+        "Expected document timeout to be used"
+    )
+
+
 def test_ocr_coverage_threshold(test_doc_path):
     pipeline_options = PdfPipelineOptions()
     pipeline_options.do_ocr = True