More tests

tests/test_merger.py

@@ -185,6 +185,18 @@ def test_zoom():
     os.remove("tmp-merger-do-not-commit.pdf")
 
 
+def test_zoom_xyz_no_left():
+    url = "https://corpora.tika.apache.org/base/docs/govdocs1/933/933322.pdf"
+    name = "tika-933322.pdf"
+    reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+    merger = PdfMerger()
+    merger.append(reader)
+    merger.write("tmp-merger-do-not-commit.pdf")
+
+    # cleanup
+    os.remove("tmp-merger-do-not-commit.pdf")
+
+
 def test_bookmark():
     url = "https://corpora.tika.apache.org/base/docs/govdocs1/997/997511.pdf"
     name = "tika-997511.pdf"

tests/test_writer.py

@@ -3,10 +3,12 @@
 
 import pytest
 
-from PyPDF2 import PdfReader, PdfWriter
+from PyPDF2 import PdfMerger, PdfReader, PdfWriter
 from PyPDF2.errors import PageSizeNotDefinedError
 from PyPDF2.generic import RectangleObject
 
+from . import get_pdf_from_url
+
 TESTS_ROOT = os.path.abspath(os.path.dirname(__file__))
 PROJECT_ROOT = os.path.dirname(TESTS_ROOT)
 RESOURCE_ROOT = os.path.join(PROJECT_ROOT, "resources")
@@ -455,3 +457,17 @@ def test_issue301():
         writer.append_pages_from_reader(reader)
         o = BytesIO()
         writer.write(o)
+
+
+def test_sweep_indirect_references_nullobject_exception():
+    # TODO: Check this more closely... this looks weird
+    url = "https://corpora.tika.apache.org/base/docs/govdocs1/924/924666.pdf"
+    name = "tika-924666.pdf"
+    reader = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+    merger = PdfMerger()
+    merger.append(reader)
+    with pytest.warns(UserWarning, match="returning NullObject instead"):
+        merger.write("tmp-merger-do-not-commit.pdf")
+
+    # cleanup
+    os.remove("tmp-merger-do-not-commit.pdf")