Add with ... as ... usage

PyPDF2/_merger.py

@@ -26,7 +26,7 @@
 # POSSIBILITY OF SUCH DAMAGE.
 
 from io import BytesIO, FileIO, IOBase
-from typing import Any, Dict, Iterable, List, Optional, Tuple, Union, cast
+from typing import Any, Dict, Iterable, List, Optional, Tuple, Union, cast, Type
 
 from ._encryption import Encryption
 from ._page import PageObject
@@ -51,6 +51,7 @@
 )
 from .pagerange import PageRange, PageRangeSpec
 from .types import FitType, LayoutType, OutlinesType, PagemodeType, ZoomArgType
+from types import TracebackType
 
 ERR_CLOSED_WRITER = "close() was called and thus the writer cannot be used anymore"
 
@@ -75,20 +76,34 @@ class PdfMerger:
     See the functions :meth:`merge()<merge>` (or :meth:`append()<append>`)
     and :meth:`write()<write>` for usage information.
 
+    :param fileobj: Output file. Can be a filename or any kind of
+            file-like object.
     :param bool strict: Determines whether user should be warned of all
             problems and also causes some correctable problems to be fatal.
             Defaults to ``False``.
     """
 
-    def __init__(self, strict: bool = False) -> None:
+    def __init__(self, fileobj: StrByteType = "", strict: bool = False) -> None:
         self.inputs: List[Tuple[Any, PdfReader, bool]] = []
         self.pages: List[Any] = []
         self.output: Optional[PdfWriter] = PdfWriter()
         self.bookmarks: OutlinesType = []
         self.named_dests: List[Any] = []
         self.id_count = 0
+        self.fileobj = fileobj
         self.strict = strict
 
+    # There is nothing to do.
+    def __enter__(self) -> "PdfMerger":
+        return self
+
+    def __exit__(self, exc_type: Optional[Type[BaseException]], exc: Optional[BaseException],
+                 traceback: Optional[TracebackType]) -> None:
+        """Write to the fileobj and close the merger."""
+        if self.fileobj:
+            self.write(self.fileobj)
+        self.close()
+
     def merge(
         self,
         position: int,
@@ -252,10 +267,6 @@ def write(self, fileobj: StrByteType) -> None:
         """
         if self.output is None:
             raise RuntimeError(ERR_CLOSED_WRITER)
-        my_file = False
-        if isinstance(fileobj, str):
-            fileobj = FileIO(fileobj, "wb")
-            my_file = True
 
         # Add pages to the PdfWriter
         # The commented out line below was replaced with the two lines below it
@@ -274,7 +285,7 @@ def write(self, fileobj: StrByteType) -> None:
         self._write_bookmarks()
 
         # Write the output to the file
-        self.output.write(fileobj)
+        my_file, fileobj = self.output.write(fileobj)
 
         if my_file:
             fileobj.close()

PyPDF2/_writer.py

@@ -47,6 +47,7 @@
     Tuple,
     Union,
     cast,
+    Type
 )
 
 from PyPDF2.errors import PdfReadWarning
@@ -59,6 +60,7 @@
     _get_max_pdf_version_header,
     b_,
     deprecate_with_replacement,
+    StrByteType
 )
 from .constants import AnnotationDictionaryAttributes
 from .constants import CatalogAttributes as CA
@@ -107,6 +109,9 @@
     ZoomArgsType,
     ZoomArgType,
 )
+from io import FileIO
+from types import TracebackType
+
 
 logger = logging.getLogger(__name__)
 
@@ -121,7 +126,7 @@ class PdfWriter:
     class (typically :class:`PdfReader<PyPDF2.PdfReader>`).
     """
 
-    def __init__(self) -> None:
+    def __init__(self, fileobj: StrByteType = "") -> None:
         self._header = b"%PDF-1.3"
         self._objects: List[Optional[PdfObject]] = []  # array of indirect objects
         self._idnum_hash: Dict[bytes, IndirectObject] = {}
@@ -158,6 +163,19 @@ def __init__(self) -> None:
         )
         self._root: Optional[IndirectObject] = None
         self._root_object = root
+        self.fileobj = fileobj
+        self.with_as_usage = False
+
+    # Let it know whether it is initialized by with ... as ... usage or not
+    def __enter__(self) -> "PdfWriter":
+        self.with_as_usage = True
+        return self
+
+    # Write to the fileobj.
+    def __exit__(self, exc_type: Optional[Type[BaseException]], exc: Optional[BaseException],
+                 traceback: Optional[TracebackType]) -> None:
+        if self.fileobj:
+            self.write(self.fileobj)
 
     @property
     def pdf_header(self) -> bytes:
@@ -763,13 +781,7 @@ def encrypt(
         self._encrypt = self._add_object(encrypt)
         self._encrypt_key = key
 
-    def write(self, stream: StreamType) -> None:
-        """
-        Write the collection of pages added to this object out as a PDF file.
-
-        :param stream: An object to write the file to.  The object must support
-            the write method and the tell method, similar to a file object.
-        """
+    def write_stream(self, stream: StreamType) -> None:
         if hasattr(stream, "mode") and "b" not in stream.mode:
             warnings.warn(
                 f"File <{stream.name}> to write to is not in binary mode. "  # type: ignore
@@ -794,6 +806,31 @@ def write(self, stream: StreamType) -> None:
         self._write_trailer(stream)
         stream.write(b_(f"\nstartxref\n{xref_location}\n%%EOF\n"))  # eof
 
+    def write(self, stream: StrByteType) -> None:
+        """
+        Write the collection of pages added to this object out as a PDF file.
+
+        :param stream: An object to write the file to.  The object can support
+            the write method and the tell method, similar to a file object, or
+            be a file path, just like the fileobj, just named it stream to keep
+            existing workflow.
+        """
+        my_file = False
+
+        if fileobj == "":
+            raise ValueError(f"Output(fileobj={fileobj}) is empty.")
+
+        if isinstance(fileobj, str):
+            fileobj = FileIO(fileobj, "wb")
+            my_file = True
+
+        self.write_stream(fileobj)
+
+        if self.with_as_usage:
+            fileobj.close()
+
+        return my_file, fileobj
+
     def _write_header(self, stream: StreamType) -> List[int]:
         object_positions = []
         stream.write(self.pdf_header + b"\n")

tests/test_merger.py

@@ -17,14 +17,12 @@
 sys.path.append(PROJECT_ROOT)
 
 
-def test_merge():
+def merger_operate(merger):
     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")
 
-    merger = PyPDF2.PdfMerger()
-
     # string path:
     merger.append(pdf_path)
     merger.append(outline)
@@ -78,10 +76,6 @@ def test_merge():
     merger.set_page_layout("/SinglePage")
     merger.set_page_mode("/UseThumbs")
 
-    tmp_path = "dont_commit_merged.pdf"
-    merger.write(tmp_path)
-    merger.close()
-
     # Check if bookmarks are correct
     reader = PyPDF2.PdfReader(tmp_path)
     assert [
@@ -102,7 +96,37 @@ def test_merge():
 
     # TODO: There seem to be no destinations for those links?
 
-    # Clean up
+
+tmp_path = "dont_commit_merged.pdf"
+
+
+def test_merger_operations_by_totally_traditional_usage():
+    merger = PdfMerger()
+
+    merger_operate(merger)
+
+    merger.write(tmp_path)
+    merger.close()
+
+    # cleanup
+    os.remove(tmp_path)
+
+
+def test_merger_operations_by_semi_traditional_usage():
+    with PdfMerger() as merger:
+        merger_operate(merger)
+
+        merger.write(tmp_path)
+
+    # cleanup
+    os.remove(tmp_path)
+
+
+def test_merger_operation_by_totally_new_usage():
+    with PdfMerger(tmp_path) as merger:
+        merger_operate(merger)
+
+    # cleanup
     os.remove(tmp_path)
 
 

tests/test_writer.py

@@ -24,20 +24,16 @@ def test_writer_clone():
     assert len(writer.pages) == 4
 
 
-def test_writer_operations():
+def writer_operate(writer):
     """
-    This test just checks if the operation throws an exception.
-
-    This should be done way more thoroughly: It should be checked if the
-    output is as expected.
+    To test the writer that initialized by each of the four usages.
     """
     pdf_path = os.path.join(RESOURCE_ROOT, "crazyones.pdf")
     pdf_outline_path = os.path.join(RESOURCE_ROOT, "pdflatex-outline.pdf")
 
     reader = PdfReader(pdf_path)
     reader_outline = PdfReader(pdf_outline_path)
 
-    writer = PdfWriter()
     page = reader.pages[0]
     with pytest.raises(PageSizeNotDefinedError) as exc:
         writer.add_blank_page()
@@ -80,17 +76,57 @@ def test_writer_operations():
 
     writer.add_attachment("foobar.gif", b"foobarcontent")
 
-    # finally, write "output" to PyPDF2-output.pdf
-    tmp_path = "dont_commit_writer.pdf"
-    with open(tmp_path, "wb") as output_stream:
-        writer.write(output_stream)
-
     # Check that every key in _idnum_hash is correct
     objects_hash = [o.hash_value() for o in writer._objects]
     for k, v in writer._idnum_hash.items():
         assert v.pdf == writer
         assert k in objects_hash, "Missing %s" % v
 
+
+tmp_path = "dont_commit_writer.pdf"
+
+
+def test_writer_operations_by_totally_traditional_usage():
+    writer = PdfWriter()
+
+    writer_operate(writer)
+
+    # finally, write "output" to PyPDF2-output.pdf
+    with open(tmp_path, "wb") as output_stream:
+        writer.write(output_stream)
+
+    # cleanup
+    os.remove(tmp_path)
+
+
+def test_writer_operations_by_semi_traditional_usage():
+    with PdfWriter() as writer:
+        writer_operate(writer)
+
+        # finally, write "output" to PyPDF2-output.pdf
+        with open(tmp_path, "wb") as output_stream:
+            writer.write(output_stream)
+
+    # cleanup
+    os.remove(tmp_path)
+
+
+def test_writer_operations_by_semi_new_traditional_usage():
+    with PdfWriter() as writer:
+        writer_operate(writer)
+
+        # finally, write "output" to PyPDF2-output.pdf
+        writer.write(tmp_path)
+
+    # cleanup
+    os.remove(tmp_path)
+
+
+def test_writer_operation_by_totally_new_usage():
+    # This includes write "output" to PyPDF2-output.pdf
+    with PdfWriter(tmp_path) as writer:
+        writer_operate(writer)
+
     # cleanup
     os.remove(tmp_path)