community[patch]: add to pypdf tests and run in CI (langchain-ai#26663)

libs/community/tests/integration_tests/document_loaders/test_pdf.py

@@ -1,4 +1,3 @@
-import re
 from pathlib import Path
 from typing import Sequence, Union
 
@@ -11,7 +10,6 @@
     PDFMinerPDFasHTMLLoader,
     PyMuPDFLoader,
     PyPDFium2Loader,
-    PyPDFLoader,
     UnstructuredPDFLoader,
 )
 
@@ -86,37 +84,6 @@ def test_pdfminer_pdf_as_html_loader() -> None:
     assert len(docs) == 1
 
 
-def test_pypdf_loader() -> None:
-    """Test PyPDFLoader."""
-    file_path = Path(__file__).parent.parent / "examples/hello.pdf"
-    loader = PyPDFLoader(str(file_path))
-    docs = loader.load()
-
-    assert len(docs) == 1
-
-    file_path = Path(__file__).parent.parent / "examples/layout-parser-paper.pdf"
-    loader = PyPDFLoader(str(file_path))
-
-    docs = loader.load()
-    assert len(docs) == 16
-
-
-def test_pypdf_loader_with_layout() -> None:
-    """Test PyPDFLoader with layout mode."""
-    file_path = Path(__file__).parent.parent / "examples/layout-parser-paper.pdf"
-    loader = PyPDFLoader(str(file_path), extraction_mode="layout")
-
-    docs = loader.load()
-    first_page = docs[0].page_content
-
-    expected = (
-        Path(__file__).parent.parent / "examples/layout-parser-paper-page-1.txt"
-    ).read_text(encoding="utf-8")
-    cleaned_first_page = re.sub(r"\x00", "", first_page)
-    cleaned_expected = re.sub(r"\x00", "", expected)
-    assert cleaned_first_page == cleaned_expected
-
-
 def test_pypdfium2_loader() -> None:
     """Test PyPDFium2Loader."""
     file_path = Path(__file__).parent.parent / "examples/hello.pdf"

libs/community/tests/unit_tests/document_loaders/test_pdf.py

@@ -0,0 +1,62 @@
+import re
+from pathlib import Path
+
+import pytest
+
+from langchain_community.document_loaders import PyPDFLoader
+
+path_to_simple_pdf = (
+    Path(__file__).parent.parent.parent / "integration_tests/examples/hello.pdf"
+)
+path_to_layout_pdf = (
+    Path(__file__).parent.parent
+    / "document_loaders/sample_documents/layout-parser-paper.pdf"
+)
+path_to_layout_pdf_txt = (
+    Path(__file__).parent.parent.parent
+    / "integration_tests/examples/layout-parser-paper-page-1.txt"
+)
+
+
+@pytest.mark.requires("pypdf")
+def test_pypdf_loader() -> None:
+    """Test PyPDFLoader."""
+    loader = PyPDFLoader(str(path_to_simple_pdf))
+    docs = loader.load()
+
+    assert len(docs) == 1
+
+    loader = PyPDFLoader(str(path_to_layout_pdf))
+
+    docs = loader.load()
+    assert len(docs) == 16
+    for page, doc in enumerate(docs):
+        assert doc.metadata["page"] == page
+        assert doc.metadata["source"].endswith("layout-parser-paper.pdf")
+        assert len(doc.page_content) > 10
+
+    first_page = docs[0].page_content
+    for expected in ["LayoutParser", "A Unified Toolkit"]:
+        assert expected in first_page
+
+
+@pytest.mark.requires("pypdf")
+def test_pypdf_loader_with_layout() -> None:
+    """Test PyPDFLoader with layout mode."""
+    loader = PyPDFLoader(str(path_to_layout_pdf), extraction_mode="layout")
+
+    docs = loader.load()
+    assert len(docs) == 16
+    for page, doc in enumerate(docs):
+        assert doc.metadata["page"] == page
+        assert doc.metadata["source"].endswith("layout-parser-paper.pdf")
+        assert len(doc.page_content) > 10
+
+    first_page = docs[0].page_content
+    for expected in ["LayoutParser", "A Unified Toolkit"]:
+        assert expected in first_page
+
+    expected = path_to_layout_pdf_txt.read_text(encoding="utf-8")
+    cleaned_first_page = re.sub(r"\x00", "", first_page)
+    cleaned_expected = re.sub(r"\x00", "", expected)
+    assert cleaned_first_page == cleaned_expected