STY: Update to mypy 1.16.0 (#3300)

Closes #3299.

Author: stefan6419846

.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]

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"]
         ):

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

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,

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:

pypdf/_writer.py

@@ -501,7 +501,7 @@ def _add_page(
         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!")
@@ -1103,10 +1103,11 @@ 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

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
     )

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):

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:

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
@@ -200,7 +201,7 @@ 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 +245,7 @@ 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:

requirements/ci-3.11.txt

@@ -24,7 +24,7 @@ fpdf2==2.8.1
     # via -r requirements/ci.in
 iniconfig==2.0.0
     # via pytest
-mypy==1.15.0
+mypy==1.16.0
     # via -r requirements/ci.in
 mypy-extensions==1.0.0
     # via mypy

tests/test_doc_common.py

@@ -9,12 +9,13 @@
 import pytest
 
 from pypdf import PdfReader, PdfWriter
-from pypdf.generic import EmbeddedFile, NullObject
+from pypdf.generic import EmbeddedFile, NullObject, ViewerPreferences
 from tests import get_data_from_url
 
 TESTS_ROOT = Path(__file__).parent.resolve()
 PROJECT_ROOT = TESTS_ROOT.parent
 SAMPLE_ROOT = PROJECT_ROOT / "sample-files"
+RESOURCES_ROOT = PROJECT_ROOT / "resources"
 
 PDFATTACH_BINARY = shutil.which("pdfattach")
 
@@ -155,3 +156,15 @@ def test_byte_encoded_named_destinations():
             "/Zoom": NullObject()
         }
     }
+
+
+def test_viewer_preferences__indirect_reference():
+    input_path = RESOURCES_ROOT / "git.pdf"
+    reader = PdfReader(input_path)
+    assert (0, 24) not in reader.resolved_objects
+    viewer_preferences = reader.viewer_preferences
+    assert isinstance(viewer_preferences, ViewerPreferences)
+    assert viewer_preferences == {"/DisplayDocTitle": True}
+    assert (0, 24) in reader.resolved_objects
+    assert id(viewer_preferences) == id(reader.viewer_preferences)
+    assert id(viewer_preferences) == id(reader.resolved_objects[(0, 24)])