Merge branch 'py-pdf:main' into issue-3290

larsga · web-flow · commit 235c7844d444 · 2025-06-18T10:05:38.000+02:00
diff --git a/.github/workflows/benchmark.yaml b/.github/workflows/benchmark.yaml
@@ -24,7 +24,7 @@ jobs:
       uses: actions/setup-python@v5
       with:
         python-version: ${{ matrix.python-version }}
-    - name: Install requirements (Python 3)
+    - name: Install requirements
       run: |
         pip install -r requirements/ci-3.11.txt
     - name: Install pypdf
@@ -46,4 +46,3 @@ jobs:
         alert-threshold: '200%'
         comment-on-alert: true
         fail-on-alert: true
-        alert-comment-cc-users: '@MartinThoma'
diff --git a/.github/workflows/github-ci.yaml b/.github/workflows/github-ci.yaml
@@ -49,7 +49,7 @@ jobs:
         python -c "from tests import download_test_pdfs; download_test_pdfs()"
     - name: Test with pytest
       run: |
-        python -m pytest tests --cov=pypdf --cov-append -n auto -vv
+        python -m pytest tests --cov=pypdf --cov-append -n auto -vv -p no:benchmark
 
 
   tests:
@@ -137,7 +137,7 @@ jobs:
         echo "    pass" >> $SITECUSTOMIZE_PATH
     - name: Test with pytest
       run: |
-        python -m pytest tests --cov=pypdf --cov-append -n auto -vv
+        python -m pytest tests --cov=pypdf --cov-append -n auto -vv -p no:benchmark
       env:
         COVERAGE_PROCESS_START: 'pyproject.toml'
     - name: Rename coverage data file
@@ -197,7 +197,7 @@ jobs:
       - uses: actions/checkout@v4
       - uses: actions/setup-python@v5
         with:
-          python-version: ${{env.PYTHON_LATEST}}
+          python-version: ${{ env.PYTHON_LATEST }}
 
       - run: python -m pip install flit check-wheel-contents
       - run: flit build
@@ -221,7 +221,7 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           # Use latest Python, so it understands all syntax.
-          python-version: ${{env.PYTHON_LATEST}}
+          python-version: ${{ env.PYTHON_LATEST }}
 
       - run: python -m pip install --upgrade coverage[toml]
 
diff --git a/.pre-commit-config.yaml b/.pre-commit-config.yaml
@@ -31,7 +31,7 @@ repos:
         args: [--py38-plus]
 
 -   repo: https://github.com/pre-commit/mirrors-mypy
-    rev: 'v1.15.0'
+    rev: 'v1.16.0'
     hooks:
       - id: mypy
         additional_dependencies: [types-Pillow==10.2.0.20240822]
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,5 +1,22 @@
 # CHANGELOG
 
+## Version 5.6.0, 2025-06-01
+
+### New Features (ENH)
+- Add basic support for JBIG2 by using jbig2dec (#3163)
+
+### Bug Fixes (BUG)
+- Fix crashes by removing unnecessary line (#3293)
+- Add delimiters to NameObject.renumber_table (#3286)
+
+### Robustness (ROB)
+- Handle DecodeParms being a NullObject (#3285)
+
+### Code Style (STY)
+- Update to mypy 1.16.0 (#3300)
+
+[Full Changelog](https://github.com/py-pdf/pypdf/compare/5.5.0...5.6.0)
+
 ## Version 5.5.0, 2025-05-11
 
 ### New Features (ENH)
diff --git a/pypdf/_cmap.py b/pypdf/_cmap.py
@@ -6,6 +6,7 @@
 from ._codecs import adobe_glyphs, charset_encoding
 from ._utils import logger_error, logger_warning
 from .generic import (
+    ArrayObject,
     DecodedStreamObject,
     DictionaryObject,
     StreamObject,
@@ -443,7 +444,7 @@ def build_font_width_map(
                 )
                 break
     elif "/Widths" in ft:
-        w = ft["/Widths"].get_object()
+        w = cast(ArrayObject, ft["/Widths"].get_object())
         if "/FontDescriptor" in ft and "/MissingWidth" in cast(
             DictionaryObject, ft["/FontDescriptor"]
         ):
diff --git a/pypdf/_doc_common.py b/pypdf/_doc_common.py
@@ -329,7 +329,7 @@ def viewer_preferences(self) -> Optional[ViewerPreferences]:
         o = o.get_object()
         if not isinstance(o, ViewerPreferences):
             o = ViewerPreferences(o)
-            if hasattr(o, "indirect_reference"):
+            if hasattr(o, "indirect_reference") and o.indirect_reference is not None:
                 self._replace_object(o.indirect_reference, o)
             else:
                 self.root_object[NameObject(CD.VIEWER_PREFERENCES)] = o
diff --git a/pypdf/_page.py b/pypdf/_page.py
@@ -1430,7 +1430,7 @@ def merge_transformed_page(
         self._merge_page(
             page2,
             lambda page2Content: PageObject._add_transformation_matrix(
-                page2Content, page2.pdf, cast(CompressedTransformationMatrix, ctm)
+                page2Content, page2.pdf, ctm
             ),
             ctm,
             over,
diff --git a/pypdf/_reader.py b/pypdf/_reader.py
@@ -254,6 +254,7 @@ def _info(self) -> Optional[DictionaryObject]:
         info = self.trailer.get(TK.INFO, None)
         if is_null_or_none(info):
             return None
+        assert info is not None, "mypy"
         info = info.get_object()
         if not isinstance(info, DictionaryObject):
             raise PdfReadError(
@@ -271,7 +272,10 @@ def _ID(self) -> Optional[ArrayObject]:
 
         """
         id = self.trailer.get(TK.ID, None)
-        return None if is_null_or_none(id) else cast(ArrayObject, id.get_object())
+        if is_null_or_none(id):
+            return None
+        assert id is not None, "mypy"
+        return cast(ArrayObject, id.get_object())
 
     @property
     def pdf_header(self) -> str:
diff --git a/pypdf/_utils.py b/pypdf/_utils.py
@@ -466,6 +466,7 @@ def rename_kwargs(
                     f"{old_term} is deprecated as an argument. Use {new_term} instead"
                 ),
                 category=DeprecationWarning,
+                stacklevel=3,
             )
 
 
diff --git a/pypdf/_version.py b/pypdf/_version.py
@@ -1 +1 @@
-__version__ = "5.5.0"
+__version__ = "5.6.0"
diff --git a/pypdf/_writer.py b/pypdf/_writer.py
@@ -544,7 +544,7 @@ def _create_link(ref: PdfObject, src: PdfReader)-> Optional[RefLink]:
         while not is_null_or_none(node):
             node = cast(DictionaryObject, node.get_object())
             node[NameObject(PA.COUNT)] = NumberObject(cast(int, node[PA.COUNT]) + 1)
-            node = node.get(PA.PARENT, None)
+            node = node.get(PA.PARENT, None)  # type: ignore[assignment]  # TODO: Fix.
             recurse += 1
             if recurse > 1000:
                 raise PyPdfError("Too many recursive calls!")
@@ -1155,10 +1155,11 @@ def update_page_form_field_values(
                     )
                 else:
                     parent_annotation[NameObject(FA.V)] = TextStringObject(value)
-                if parent_annotation.get(FA.FT) in ("/Btn"):
+                if parent_annotation.get(FA.FT) == "/Btn":
                     # Checkbox button (no /FT found in Radio widgets)
                     v = NameObject(value)
-                    if v not in annotation[NameObject(AA.AP)][NameObject("/N")]:
+                    ap = cast(DictionaryObject, annotation[NameObject(AA.AP)])
+                    if v not in cast(ArrayObject, ap[NameObject("/N")]):
                         v = NameObject("/Off")
                     # other cases will be updated through the for loop
                     annotation[NameObject(AA.AS)] = v
diff --git a/pypdf/_xobj_image_helpers.py b/pypdf/_xobj_image_helpers.py
@@ -13,7 +13,6 @@
     ArrayObject,
     DecodedStreamObject,
     EncodedStreamObject,
-    IndirectObject,
     NullObject,
     TextStringObject,
 )
@@ -57,49 +56,46 @@ def _get_imagemode(
         )
     if isinstance(color_space, NullObject):
         return "", False
+    color_space_str: str = ""
     if isinstance(color_space, str):
-        pass
+        color_space_str = color_space
     elif not isinstance(color_space, list):
         raise PdfReadError(
             "Cannot interpret color space", color_space
         )  # pragma: no cover
     elif color_space[0].startswith("/Cal"):  # /CalRGB and /CalGray
-        color_space = "/Device" + color_space[0][4:]
+        color_space_str = "/Device" + color_space[0][4:]
     elif color_space[0] == "/ICCBased":
         icc_profile = color_space[1].get_object()
         color_components = cast(int, icc_profile["/N"])
-        color_space = icc_profile.get("/Alternate", "")
+        color_space_str = icc_profile.get("/Alternate", "")
     elif color_space[0] == "/Indexed":
-        color_space = color_space[1].get_object()
+        color_space_str = color_space[1].get_object()
         mode, invert_color = _get_imagemode(
-            color_space, color_components, prev_mode, depth + 1
+            color_space_str, color_components, prev_mode, depth + 1
         )
         if mode in ("RGB", "CMYK"):
             mode = "P"
         return mode, invert_color
     elif color_space[0] == "/Separation":
-        color_space = color_space[2]
-        if isinstance(color_space, IndirectObject):
-            color_space = color_space.get_object()
+        color_space_str = color_space[2].get_object()
         mode, invert_color = _get_imagemode(
-            color_space, color_components, prev_mode, depth + 1
+            color_space_str, color_components, prev_mode, depth + 1
         )
         return mode, True
     elif color_space[0] == "/DeviceN":
         original_color_space = color_space
         color_components = len(color_space[1])
-        color_space = color_space[2]
-        if isinstance(color_space, IndirectObject):  # pragma: no cover
-            color_space = color_space.get_object()
-        if color_space == "/DeviceCMYK" and color_components == 1:
+        color_space_str = color_space[2].get_object()
+        if color_space_str == "/DeviceCMYK" and color_components == 1:
             if original_color_space[1][0] != "/Black":
                 logger_warning(
                     f"Color {original_color_space[1][0]} converted to Gray. Please share PDF with pypdf dev team",
                     __name__,
                 )
             return "L", True
         mode, invert_color = _get_imagemode(
-            color_space, color_components, prev_mode, depth + 1
+            color_space_str, color_components, prev_mode, depth + 1
         )
         return mode, invert_color
 
@@ -114,7 +110,7 @@ def _get_imagemode(
     }
 
     mode = (
-        mode_map.get(color_space)
+        mode_map.get(color_space_str)
         or list(mode_map.values())[color_components]
         or prev_mode
     )
diff --git a/pypdf/generic/_data_structures.py b/pypdf/generic/_data_structures.py
@@ -411,7 +411,7 @@ def _clone(
                         v.indirect_reference = None
                     vv = v.clone(pdf_dest, force_duplicate, ignore_fields)
                     assert vv.indirect_reference is not None
-                    self[k.clone(pdf_dest)] = vv.indirect_reference  # type: ignore[attr-defined]
+                    self[k.clone(pdf_dest)] = vv.indirect_reference
                 elif k not in self:
                     self[NameObject(k)] = (
                         v.clone(pdf_dest, force_duplicate, ignore_fields)
@@ -495,6 +495,7 @@ def xmp_metadata(self) -> Optional[XmpInformationProtocol]:
         metadata = self.get("/Metadata", None)
         if is_null_or_none(metadata):
             return None
+        assert metadata is not None, "mypy"
         metadata = metadata.get_object()
 
         if not isinstance(metadata, XmpInformation):
@@ -978,7 +979,7 @@ def hash_bin(self) -> int:
             Hash considering type and value.
 
         """
-        # use of _data to prevent errors on non decoded stream such as JBIG2
+        # Use _data to prevent errors on non-decoded streams.
         return hash((super().hash_bin(), self._data))
 
     def get_data(self) -> bytes:
diff --git a/pypdf/generic/_viewerpref.py b/pypdf/generic/_viewerpref.py
@@ -48,7 +48,7 @@ def __init__(self, obj: Optional[DictionaryObject] = None) -> None:
         except AttributeError:
             pass
 
-    def _get_bool(self, key: str, default: Optional[BooleanObject]) -> BooleanObject:
+    def _get_bool(self, key: str, default: Optional[BooleanObject]) -> Optional[BooleanObject]:
         return self.get(key, default)
 
     def _set_bool(self, key: str, v: bool) -> None:
@@ -64,7 +64,7 @@ def _set_name(self, key: str, lst: List[str], v: NameObject) -> None:
             raise ValueError(f"{v} is an unacceptable value")
         self[NameObject(key)] = NameObject(v)
 
-    def _get_arr(self, key: str, default: Optional[List[Any]]) -> NumberObject:
+    def _get_arr(self, key: str, default: Optional[List[Any]]) -> Optional[ArrayObject]:
         return self.get(key, None if default is None else ArrayObject(default))
 
     def _set_arr(self, key: str, v: Optional[ArrayObject]) -> None:
@@ -78,7 +78,7 @@ def _set_arr(self, key: str, v: Optional[ArrayObject]) -> None:
             raise ValueError("ArrayObject is expected")
         self[NameObject(key)] = v
 
-    def _get_int(self, key: str, default: Optional[NumberObject]) -> NumberObject:
+    def _get_int(self, key: str, default: Optional[NumberObject]) -> Optional[NumberObject]:
         return self.get(key, default)
 
     def _set_int(self, key: str, v: int) -> None:
diff --git a/pypdf/types.py b/pypdf/types.py
@@ -14,17 +14,18 @@
 from .generic._outline import OutlineItem
 
 BorderArrayType: TypeAlias = List[Union[NameObject, NumberObject, ArrayObject]]
+
 OutlineItemType: TypeAlias = Union[OutlineItem, Destination]
+
 FitType: TypeAlias = Literal[
     "/XYZ", "/Fit", "/FitH", "/FitV", "/FitR", "/FitB", "/FitBH", "/FitBV"
 ]
-# Those go with the FitType: They specify values for the fit
+# These go with the FitType, they specify values for the fit
 ZoomArgType: TypeAlias = Union[NumberObject, NullObject, float]
 ZoomArgsType: TypeAlias = List[ZoomArgType]
 
-# Recursive types like the following are not yet supported by mypy:
+# Recursive types like the following are not yet supported by Sphinx:
 #    OutlineType = List[Union[Destination, "OutlineType"]]
-# See https://github.com/python/mypy/issues/731
 # Hence use this for the moment:
 OutlineType = List[Union[Destination, List[Union[Destination, List[Destination]]]]]
 
@@ -37,6 +38,7 @@
     "/TwoPageLeft",
     "/TwoPageRight",
 ]
+
 PagemodeType: TypeAlias = Literal[
     "/UseNone",
     "/UseOutlines",
@@ -45,6 +47,7 @@
     "/UseOC",
     "/UseAttachments",
 ]
+
 AnnotationSubtype: TypeAlias = Literal[
     "/Text",
     "/Link",
diff --git a/pypdf/xmp.py b/pypdf/xmp.py
@@ -21,6 +21,7 @@
 from xml.dom.minidom import Element as XmlElement
 from xml.parsers.expat import ExpatError
 
+from ._protocols import XmpInformationProtocol
 from ._utils import StreamType, deprecate_no_replacement
 from .errors import PdfReadError
 from .generic import ContentStream, PdfObject
@@ -51,6 +52,9 @@
 # Schemas".
 PDFX_NAMESPACE = "http://ns.adobe.com/pdfx/1.3/"
 
+# PDF/A
+PDFAID_NAMESPACE = "http://www.aiim.org/pdfa/ns/id/"
+
 iso8601 = re.compile(
     """
         (?P<year>[0-9]{4})
@@ -200,7 +204,7 @@ def get(self: "XmpInformation") -> Optional[Any]:
     return get
 
 
-class XmpInformation(PdfObject):
+class XmpInformation(XmpInformationProtocol, PdfObject):
     """
     An object that represents Extensible Metadata Platform (XMP) metadata.
     Usually accessed by :py:attr:`xmp_metadata()<pypdf.PdfReader.xmp_metadata>`.
@@ -244,7 +248,7 @@ def get_nodes_in_namespace(self, about_uri: str, namespace: str) -> Iterator[Any
             if desc.getAttributeNS(RDF_NAMESPACE, "about") == about_uri:
                 for i in range(desc.attributes.length):
                     attr = desc.attributes.item(i)
-                    if attr.namespaceURI == namespace:
+                    if attr and attr.namespaceURI == namespace:
                         yield attr
                 for child in desc.childNodes:
                     if child.namespaceURI == namespace:
@@ -362,6 +366,12 @@ def _get_text(self, element: XmlElement) -> str:
     """An identifier for a specific incarnation of a document, updated each
     time a file is saved."""
 
+    pdfaid_part = property(_getter_single(PDFAID_NAMESPACE, "part"))
+    """The part of the PDF/A standard that the document conforms to (e.g., 1, 2, 3)."""
+
+    pdfaid_conformance = property(_getter_single(PDFAID_NAMESPACE, "conformance"))
+    """The conformance level within the PDF/A standard (e.g., 'A', 'B', 'U')."""
+
     @property
     def custom_properties(self) -> Dict[Any, Any]:
         """
diff --git a/pyproject.toml b/pyproject.toml
@@ -142,7 +142,6 @@ ignore = [
     "ARG001",  # Unused function argument
     "ARG002",  # Unused method argument
     "ARG004",  # Unused static method argument
-    "B028",    # No explicit `stacklevel` keyword argument found
     "B904",    # Within an `except` clause, raise exceptions with
     "B905",    # `zip()` without an explicit `strict=` parameter
     "BLE001",  # Do not catch blind exception: `Exception`
diff --git a/requirements/ci-3.11.txt b/requirements/ci-3.11.txt
diff --git a/requirements/dev.txt b/requirements/dev.txt
diff --git a/requirements/docs.txt b/requirements/docs.txt
diff --git a/tests/test_annotations.py b/tests/test_annotations.py
diff --git a/tests/test_doc_common.py b/tests/test_doc_common.py
diff --git a/tests/test_utils.py b/tests/test_utils.py
diff --git a/tests/test_xmp.py b/tests/test_xmp.py

Original file line number	Diff line number	Diff line change
`@@ -466,6 +466,7 @@ def rename_kwargs(`
`466`	`466`	`f"{old_term} is deprecated as an argument. Use {new_term} instead"`
`467`	`467`	`),`
`468`	`468`	`category=DeprecationWarning,`
	`469`	`+ stacklevel=3,`
`469`	`470`	`)`
`470`	`471`
`471`	`472`
Original file line number	Diff line number	Diff line change
`@@ -1 +1 @@`
`1`		`-__version__ = "5.5.0"`
	`1`	`+__version__ = "5.6.0"`