ENH: Add support for pathlib.Path in PdfMerger.merge (#1190)

Replace many os.path usages with pathlib

Author: MartinThoma

PyPDF2/_merger.py

@@ -26,6 +26,7 @@
 # POSSIBILITY OF SUCH DAMAGE.
 
 from io import BytesIO, FileIO, IOBase
+from pathlib import Path
 from typing import Any, Dict, Iterable, List, Optional, Tuple, Union, cast
 
 from ._encryption import Encryption
@@ -99,7 +100,7 @@ def __init__(self, strict: bool = False) -> None:
     def merge(
         self,
         position: int,
-        fileobj: Union[StrByteType, PdfReader],
+        fileobj: Union[Path, StrByteType, PdfReader],
         outline_item: Optional[str] = None,
         pages: Optional[PageRangeSpec] = None,
         import_outline: bool = True,
@@ -184,7 +185,7 @@ def merge(
         self.pages[position:position] = srcpages
 
     def _create_stream(
-        self, fileobj: Union[StrByteType, PdfReader]
+        self, fileobj: Union[Path, StrByteType, PdfReader]
     ) -> Tuple[IOBase, bool, Optional[Encryption]]:
         # This parameter is passed to self.inputs.append and means
         # that the stream used was created in this method.
@@ -198,7 +199,7 @@ def _create_stream(
         # If fileobj is none of the above types, it is not modified
         encryption_obj = None
         stream: IOBase
-        if isinstance(fileobj, str):
+        if isinstance(fileobj, (str, Path)):
             stream = FileIO(fileobj, "rb")
             my_file = True
         elif isinstance(fileobj, PdfReader):
@@ -224,7 +225,7 @@ def _create_stream(
     @deprecate_bookmark(bookmark="outline_item", import_bookmarks="import_outline")
     def append(
         self,
-        fileobj: Union[StrByteType, PdfReader],
+        fileobj: Union[StrByteType, PdfReader, Path],
         outline_item: Optional[str] = None,
         pages: Union[None, PageRange, Tuple[int, int], Tuple[int, int, int]] = None,
         import_outline: bool = True,

tests/bench.py

@@ -1,17 +1,18 @@
 import os
+from pathlib import Path
 
 import PyPDF2
 from PyPDF2 import PdfReader, Transformation
 from PyPDF2.generic import Destination
 
-TESTS_ROOT = os.path.abspath(os.path.dirname(__file__))
-PROJECT_ROOT = os.path.dirname(TESTS_ROOT)
-RESOURCE_ROOT = os.path.join(PROJECT_ROOT, "resources")
-SAMPLE_ROOT = os.path.join(PROJECT_ROOT, "sample-files")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
+SAMPLE_ROOT = PROJECT_ROOT / "sample-files"
 
 
 def page_ops(pdf_path, password):
-    pdf_path = os.path.join(RESOURCE_ROOT, pdf_path)
+    pdf_path = RESOURCE_ROOT / pdf_path
 
     reader = PdfReader(pdf_path)
 
@@ -50,10 +51,10 @@ def test_page_operations(benchmark):
 
 
 def merge():
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
-    outline = os.path.join(RESOURCE_ROOT, "pdflatex-outline.pdf")
-    pdf_forms = os.path.join(RESOURCE_ROOT, "pdflatex-forms.pdf")
-    pdf_pw = os.path.join(RESOURCE_ROOT, "libreoffice-writer-password.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
+    outline = RESOURCE_ROOT / "pdflatex-outline.pdf"
+    pdf_forms = RESOURCE_ROOT / "pdflatex-forms.pdf"
+    pdf_pw = RESOURCE_ROOT / "libreoffice-writer-password.pdf"
 
     file_merger = PyPDF2.PdfMerger()
 
@@ -126,5 +127,5 @@ def text_extraction(pdf_path):
 
 
 def test_text_extraction(benchmark):
-    file_path = os.path.join(SAMPLE_ROOT, "009-pdflatex-geotopo/GeoTopo.pdf")
+    file_path = SAMPLE_ROOT / "009-pdflatex-geotopo/GeoTopo.pdf"
     benchmark(text_extraction, file_path)

tests/test_basic_features.py

@@ -1,14 +1,15 @@
 import os
+from pathlib import Path
 
 from PyPDF2 import PdfReader, PdfWriter
 
-TESTS_ROOT = os.path.abspath(os.path.dirname(__file__))
-PROJECT_ROOT = os.path.dirname(TESTS_ROOT)
-RESOURCE_ROOT = os.path.join(PROJECT_ROOT, "resources")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
 
 
 def test_basic_features():
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     reader = PdfReader(pdf_path)
     writer = PdfWriter()
 

tests/test_encryption.py

@@ -1,4 +1,4 @@
-import os
+from pathlib import Path
 
 import pytest
 
@@ -14,9 +14,9 @@
 except ImportError:
     HAS_PYCRYPTODOME = False
 
-TESTS_ROOT = os.path.abspath(os.path.dirname(__file__))
-PROJECT_ROOT = os.path.dirname(TESTS_ROOT)
-RESOURCE_ROOT = os.path.join(PROJECT_ROOT, "resources")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
 
 
 @pytest.mark.parametrize(
@@ -51,7 +51,7 @@
     ],
 )
 def test_encryption(name, requres_pycryptodome):
-    inputfile = os.path.join(RESOURCE_ROOT, "encryption", name)
+    inputfile = RESOURCE_ROOT / "encryption" / name
     if requres_pycryptodome and not HAS_PYCRYPTODOME:
         with pytest.raises(DependencyError) as exc:
             ipdf = PyPDF2.PdfReader(inputfile)
@@ -61,7 +61,7 @@ def test_encryption(name, requres_pycryptodome):
         return
     else:
         ipdf = PyPDF2.PdfReader(inputfile)
-        if inputfile.endswith("unencrypted.pdf"):
+        if str(inputfile).endswith("unencrypted.pdf"):
             assert not ipdf.is_encrypted
         else:
             assert ipdf.is_encrypted
@@ -91,7 +91,7 @@ def test_encryption(name, requres_pycryptodome):
 def test_both_password(name, user_passwd, owner_passwd):
     from PyPDF2 import PasswordType
 
-    inputfile = os.path.join(RESOURCE_ROOT, "encryption", name)
+    inputfile = RESOURCE_ROOT / "encryption" / name
     ipdf = PyPDF2.PdfReader(inputfile)
     assert ipdf.is_encrypted
     assert ipdf.decrypt(user_passwd) == PasswordType.USER_PASSWORD
@@ -113,7 +113,7 @@ def test_get_page_of_encrypted_file_new_algorithm(pdffile, password):
     This is a regression test for issue 327:
     IndexError for get_page() of decrypted file
     """
-    path = os.path.join(RESOURCE_ROOT, pdffile)
+    path = RESOURCE_ROOT / pdffile
     PyPDF2.PdfReader(path, password=password).pages[0]
 
 
@@ -133,7 +133,7 @@ def test_get_page_of_encrypted_file_new_algorithm(pdffile, password):
 @pytest.mark.skipif(not HAS_PYCRYPTODOME, reason="No pycryptodome")
 def test_encryption_merge(names):
     pdf_merger = PyPDF2.PdfMerger()
-    files = [os.path.join(RESOURCE_ROOT, "encryption", x) for x in names]
+    files = [RESOURCE_ROOT / "encryption" / x for x in names]
     pdfs = [PyPDF2.PdfReader(x) for x in files]
     for pdf in pdfs:
         if pdf.is_encrypted:
@@ -157,7 +157,7 @@ def test_encrypt_decrypt_class(cryptcls):
 
 
 def test_decrypt_not_decrypted_pdf():
-    path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    path = RESOURCE_ROOT / "crazyones.pdf"
     with pytest.raises(PdfReadError) as exc:
         PdfReader(path, password="nonexistant")
     assert exc.value.args[0] == "Not encrypted file"

tests/test_generic.py

@@ -1,5 +1,6 @@
 import os
 from io import BytesIO
+from pathlib import Path
 from unittest.mock import patch
 
 import pytest
@@ -33,9 +34,9 @@
 
 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")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
 
 
 def test_float_object_exception():
@@ -395,7 +396,7 @@ def test_remove_child_not_in_tree():
 
 
 def test_remove_child_in_tree():
-    pdf = os.path.join(RESOURCE_ROOT, "form.pdf")
+    pdf = RESOURCE_ROOT / "form.pdf"
 
     tree = TreeObject()
     reader = PdfReader(pdf)
@@ -503,7 +504,7 @@ def test_issue_997(mock_logger_warning):
 
 def test_annotation_builder_free_text():
     # Arrange
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     reader = PdfReader(pdf_path)
     page = reader.pages[0]
     writer = PdfWriter()
@@ -533,7 +534,7 @@ def test_annotation_builder_free_text():
 
 def test_annotation_builder_line():
     # Arrange
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     reader = PdfReader(pdf_path)
     page = reader.pages[0]
     writer = PdfWriter()
@@ -558,7 +559,7 @@ def test_annotation_builder_line():
 
 def test_annotation_builder_link():
     # Arrange
-    pdf_path = os.path.join(RESOURCE_ROOT, "outline-without-title.pdf")
+    pdf_path = RESOURCE_ROOT / "outline-without-title.pdf"
     reader = PdfReader(pdf_path)
     page = reader.pages[0]
     writer = PdfWriter()

tests/test_javascript.py

@@ -1,18 +1,18 @@
-import os
+from pathlib import Path
 
 import pytest
 
 from PyPDF2 import PdfReader, PdfWriter
 
 # Configure path environment
-TESTS_ROOT = os.path.abspath(os.path.dirname(__file__))
-PROJECT_ROOT = os.path.dirname(TESTS_ROOT)
-RESOURCE_ROOT = os.path.join(PROJECT_ROOT, "resources")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
 
 
 @pytest.fixture()
 def pdf_file_writer():
-    reader = PdfReader(os.path.join(RESOURCE_ROOT, "crazyones.pdf"))
+    reader = PdfReader(RESOURCE_ROOT / "crazyones.pdf")
     writer = PdfWriter()
     writer.append_pages_from_reader(reader)
     return writer

tests/test_merger.py

@@ -1,6 +1,7 @@
 import os
 import sys
 from io import BytesIO
+from pathlib import Path
 
 import pytest
 
@@ -10,18 +11,18 @@
 
 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")
+TESTS_ROOT = Path(__file__).parent.resolve()
+PROJECT_ROOT = TESTS_ROOT.parent
+RESOURCE_ROOT = PROJECT_ROOT / "resources"
 
-sys.path.append(PROJECT_ROOT)
+sys.path.append(str(PROJECT_ROOT))
 
 
 def test_merge():
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
-    outline = os.path.join(RESOURCE_ROOT, "pdflatex-outline.pdf")
-    pdf_forms = os.path.join(RESOURCE_ROOT, "pdflatex-forms.pdf")
-    pdf_pw = os.path.join(RESOURCE_ROOT, "libreoffice-writer-password.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
+    outline = RESOURCE_ROOT / "pdflatex-outline.pdf"
+    pdf_forms = RESOURCE_ROOT / "pdflatex-forms.pdf"
+    pdf_pw = RESOURCE_ROOT / "libreoffice-writer-password.pdf"
 
     merger = PyPDF2.PdfMerger()
 
@@ -122,7 +123,7 @@ def test_merge():
 
 def test_merge_page_exception():
     merger = PyPDF2.PdfMerger()
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     with pytest.raises(TypeError) as exc:
         merger.merge(0, pdf_path, pages="a:b")
     assert exc.value.args[0] == '"pages" must be a tuple of (start, stop[, step])'
@@ -131,14 +132,14 @@ def test_merge_page_exception():
 
 def test_merge_page_tuple():
     merger = PyPDF2.PdfMerger()
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     merger.merge(0, pdf_path, pages=(0, 1))
     merger.close()
 
 
 def test_merge_write_closed_fh():
     merger = PyPDF2.PdfMerger()
-    pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
+    pdf_path = RESOURCE_ROOT / "crazyones.pdf"
     merger.append(pdf_path)
 
     err_closed = "close() was called and thus the writer cannot be used anymore"
@@ -313,9 +314,7 @@ def test_iss1145():
 
 
 def test_deprecate_bookmark_decorator_warning():
-    reader = PdfReader(
-        os.path.join(RESOURCE_ROOT, "outlines-with-invalid-destinations.pdf")
-    )
+    reader = PdfReader(RESOURCE_ROOT / "outlines-with-invalid-destinations.pdf")
     merger = PdfMerger()
     with pytest.warns(
         UserWarning,
@@ -326,9 +325,7 @@ def test_deprecate_bookmark_decorator_warning():
 
 @pytest.mark.filterwarnings("ignore::UserWarning")
 def test_deprecate_bookmark_decorator_output():
-    reader = PdfReader(
-        os.path.join(RESOURCE_ROOT, "outlines-with-invalid-destinations.pdf")
-    )
+    reader = PdfReader(RESOURCE_ROOT / "outlines-with-invalid-destinations.pdf")
     merger = PdfMerger()
     merger.merge(0, reader, import_bookmarks=True)
     first_oi_title = 'Valid Destination: Action /GoTo Named Destination "section.1"'