STY: Same attributes between PdfReader and PdfWriter

pypdf/_page_labels.py

@@ -11,12 +11,12 @@
 Example 1
 ---------
 
->>> reader.trailer["/Root"]["/PageLabels"]["/Nums"]
+>>> reader.root_object["/PageLabels"]["/Nums"]
 [0, IndirectObject(18, 0, 139929798197504),
  8, IndirectObject(19, 0, 139929798197504)]
->>> reader.get_object(reader.trailer["/Root"]["/PageLabels"]["/Nums"][1])
+>>> reader.get_object(reader.root_object["/PageLabels"]["/Nums"][1])
 {'/S': '/r'}
->>> reader.get_object(reader.trailer["/Root"]["/PageLabels"]["/Nums"][3])
+>>> reader.get_object(reader.root_object["/PageLabels"]["/Nums"][3])
 {'/S': '/D'}
 
 Example 2
@@ -57,7 +57,7 @@
                            aa to zz for the next 26, and so on)
 """
 
-from typing import Iterator, Optional, Tuple
+from typing import Iterator, Optional, Tuple, cast
 
 from ._protocols import PdfReaderProtocol
 from ._utils import logger_warning
@@ -127,10 +127,10 @@ def index2label(reader: PdfReaderProtocol, index: int) -> str:
     Returns:
         The label of the page, e.g. "iv" or "4".
     """
-    root = reader.trailer["/Root"]
+    root = cast(DictionaryObject, reader.root_object)
     if "/PageLabels" not in root:
         return str(index + 1)  # Fallback
-    number_tree = root["/PageLabels"]
+    number_tree = cast(DictionaryObject, root["/PageLabels"])
     if "/Nums" in number_tree:
         # [Nums] shall be an array of the form
         #   [ key 1 value 1 key 2 value 2 ... key n value n ]
@@ -139,7 +139,7 @@ def index2label(reader: PdfReaderProtocol, index: int) -> str:
         # The keys shall be sorted in numerical order,
         # analogously to the arrangement of keys in a name tree
         # as described in 7.9.6, "Name Trees."
-        nums = number_tree["/Nums"]
+        nums = cast(ArrayObject, number_tree["/Nums"])
         i = 0
         value = None
         start_index = 0

pypdf/_protocols.py

@@ -59,6 +59,10 @@ def pages(self) -> List[Any]:
     def trailer(self) -> Dict[str, Any]:
         ...
 
+    @property
+    def root_object(self) -> PdfObjectProtocol:
+        ...
+
     def get_object(self, indirect_reference: Any) -> Optional[PdfObjectProtocol]:
         ...
 
@@ -83,3 +87,7 @@ def pages(self) -> List[Any]:
     @property
     def pdf_header(self) -> bytes:
         ...
+
+    @property
+    def root_object(self) -> PdfObjectProtocol:
+        ...

pypdf/_reader.py

@@ -329,8 +329,7 @@ def __init__(
             # Some documents may not have a /ID, use two empty
             # byte strings instead. Solves
             # https://github.com/py-pdf/pypdf/issues/608
-            id_entry = self.trailer.get(TK.ID)
-            id1_entry = id_entry[0].get_object().original_bytes if id_entry else b""
+            id1_entry = self._ID[0].get_object().original_bytes if self._ID else b""
             encrypt_entry = cast(
                 DictionaryObject, self.trailer[TK.ENCRYPT].get_object()
             )
@@ -348,6 +347,34 @@ def __init__(
         elif password is not None:
             raise PdfReadError("Not encrypted file")
 
+    @property
+    def root_object(self) -> DictionaryObject:
+        """Provide access to "/Root". standardized with PdfWriter."""
+        return cast(DictionaryObject, self.trailer[TK.ROOT])
+
+    @property
+    def _info(self) -> Optional[DictionaryObject]:
+        """
+        Provide access to "/Info". standardized with PdfWriter.
+
+        Returns:
+            /Info Dictionary ; None if the entry does not exists
+
+        """
+        info = self.trailer.get(TK.INFO, None)
+        return None if info is None else cast(DictionaryObject, info.get_object())
+
+    @property
+    def _ID(self) -> Optional[ArrayObject]:
+        """
+        Provide access to "/ID". standardized with PdfWriter.
+
+        Returns:
+            /ID array ; None if the entry does not exists
+        """
+        id = self.trailer.get(TK.ID, None)
+        return None if id is None else cast(ArrayObject, id.get_object())
+
     @property
     def pdf_header(self) -> str:
         """
@@ -375,9 +402,9 @@ def metadata(self) -> Optional[DocumentInformation]:
         """
         if TK.INFO not in self.trailer:
             return None
-        obj = self.trailer[TK.INFO]
+        obj = self._info
         retval = DocumentInformation()
-        if isinstance(obj, type(None)):
+        if obj is None:
             raise PdfReadError(
                 "trailer not found or does not point to document information directory"
             )
@@ -408,7 +435,7 @@ def xmp_metadata(self) -> Optional[XmpInformation]:
         """XMP (Extensible Metadata Platform) data."""
         try:
             self._override_encryption = True
-            return self.trailer[TK.ROOT].xmp_metadata  # type: ignore
+            return self.root_object.xmp_metadata  # type: ignore
         finally:
             self._override_encryption = False
 
@@ -446,7 +473,7 @@ def _get_num_pages(self) -> int:
         # the PDF file's page count is used in this case. Otherwise,
         # the original method (flattened page count) is used.
         if self.is_encrypted:
-            return self.trailer[TK.ROOT]["/Pages"]["/Count"]  # type: ignore
+            return self.root_object["/Pages"]["/Count"]  # type: ignore
         else:
             if self.flattened_pages is None:
                 self._flatten()
@@ -546,10 +573,9 @@ def get_fields(
         field_attributes.update(CheckboxRadioButtonAttributes.attributes_dict())
         if retval is None:
             retval = {}
-            catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
             # get the AcroForm tree
-            if CD.ACRO_FORM in catalog:
-                tree = cast(Optional[TreeObject], catalog[CD.ACRO_FORM])
+            if CD.ACRO_FORM in self.root_object:
+                tree = cast(Optional[TreeObject], self.root_object[CD.ACRO_FORM])
             else:
                 return None
         if tree is None:
@@ -739,13 +765,12 @@ def _get_named_destinations(
         """
         if retval is None:
             retval = {}
-            catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
 
             # get the name tree
-            if CA.DESTS in catalog:
-                tree = cast(TreeObject, catalog[CA.DESTS])
-            elif CA.NAMES in catalog:
-                names = cast(DictionaryObject, catalog[CA.NAMES])
+            if CA.DESTS in self.root_object:
+                tree = cast(TreeObject, self.root_object[CA.DESTS])
+            elif CA.NAMES in self.root_object:
+                names = cast(DictionaryObject, self.root_object[CA.NAMES])
                 if CA.DESTS in names:
                     tree = cast(TreeObject, names[CA.DESTS])
 
@@ -825,11 +850,10 @@ def _get_outline(
     ) -> OutlineType:
         if outline is None:
             outline = []
-            catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
 
             # get the outline dictionary and named destinations
-            if CO.OUTLINES in catalog:
-                lines = cast(DictionaryObject, catalog[CO.OUTLINES])
+            if CO.OUTLINES in self.root_object:
+                lines = cast(DictionaryObject, self.root_object[CO.OUTLINES])
 
                 if isinstance(lines, NullObject):
                     return outline
@@ -882,9 +906,8 @@ def threads(self) -> Optional[ArrayObject]:
         It's an array of dictionaries with "/F" and "/I" properties or
         None if there are no articles.
         """
-        catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
-        if CO.THREADS in catalog:
-            return cast("ArrayObject", catalog[CO.THREADS])
+        if CO.THREADS in self.root_object:
+            return cast("ArrayObject", self.root_object[CO.THREADS])
         else:
             return None
 
@@ -1097,9 +1120,8 @@ def page_layout(self) -> Optional[str]:
            * - /TwoPageRight
              - Show two pages at a time, odd-numbered pages on the right
         """
-        trailer = cast(DictionaryObject, self.trailer[TK.ROOT])
-        if CD.PAGE_LAYOUT in trailer:
-            return cast(NameObject, trailer[CD.PAGE_LAYOUT])
+        if CD.PAGE_LAYOUT in self.root_object:
+            return cast(NameObject, self.root_object[CD.PAGE_LAYOUT])
         return None
 
     def getPageLayout(self) -> Optional[str]:  # deprecated
@@ -1143,7 +1165,7 @@ def page_mode(self) -> Optional[PagemodeType]:
              - Show attachments panel
         """
         try:
-            return self.trailer[TK.ROOT]["/PageMode"]  # type: ignore
+            return self.root_object["/PageMode"]  # type: ignore
         except KeyError:
             return None
 
@@ -1183,7 +1205,7 @@ def _flatten(
         if pages is None:
             # Fix issue 327: set flattened_pages attribute only for
             # decrypted file
-            catalog = self.trailer[TK.ROOT].get_object()
+            catalog = self.root_object.get_object()
             pages = catalog["/Pages"].get_object()  # type: ignore
             self.flattened_pages = []
 
@@ -2073,12 +2095,13 @@ def isEncrypted(self) -> bool:  # deprecated
     def xfa(self) -> Optional[Dict[str, Any]]:
         tree: Optional[TreeObject] = None
         retval: Dict[str, Any] = {}
-        catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
 
-        if "/AcroForm" not in catalog or not catalog["/AcroForm"]:
+        if isinstance(
+            self.root_object.get("/AcroForm", None), (type(None), NullObject)
+        ):
             return None
 
-        tree = cast(TreeObject, catalog["/AcroForm"])
+        tree = cast(TreeObject, self.root_object["/AcroForm"])
 
         if "/XFA" in tree:
             fields = cast(ArrayObject, tree["/XFA"])
@@ -2103,13 +2126,12 @@ def add_form_topname(self, name: str) -> Optional[DictionaryObject]:
         Returns:
             The created object. ``None`` means no object was created.
         """
-        catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
-
-        if "/AcroForm" not in catalog or not isinstance(
-            catalog["/AcroForm"], DictionaryObject
+        if isinstance(
+            self.root_object.get("/AcroForm", None), (type(None), NullObject)
         ):
             return None
-        acroform = cast(DictionaryObject, catalog[NameObject("/AcroForm")])
+
+        acroform = cast(DictionaryObject, self.root_object["/AcroForm"])
         if "/Fields" not in acroform:
             # TODO: :No error returns but may be extended for XFA Forms
             return None
@@ -2145,13 +2167,12 @@ def rename_form_topname(self, name: str) -> Optional[DictionaryObject]:
         Returns:
             The modified object. ``None`` means no object was modified.
         """
-        catalog = cast(DictionaryObject, self.trailer[TK.ROOT])
-
-        if "/AcroForm" not in catalog or not isinstance(
-            catalog["/AcroForm"], DictionaryObject
+        if isinstance(
+            self.root_object.get("/AcroForm", None), (type(None), NullObject)
         ):
             return None
-        acroform = cast(DictionaryObject, catalog[NameObject("/AcroForm")])
+
+        acroform = cast(DictionaryObject, self.root_object["/AcroForm"])
         if "/Fields" not in acroform:
             return None
 
@@ -2178,14 +2199,14 @@ def _list_attachments(self) -> List[str]:
         Returns:
             list of filenames
         """
-        catalog = cast(DictionaryObject, self.trailer["/Root"])
-        # From the catalog get the embedded file names
         try:
             filenames = cast(
                 ArrayObject,
                 cast(
                     DictionaryObject,
-                    cast(DictionaryObject, catalog["/Names"])["/EmbeddedFiles"],
+                    cast(DictionaryObject, self.root_object["/Names"])[
+                        "/EmbeddedFiles"
+                    ],
                 )["/Names"],
             )
         except KeyError:
@@ -2220,14 +2241,15 @@ def _get_attachments(
             dictionary of filename -> Union[bytestring or List[ByteString]]
             if the filename exists multiple times a List of the different version will be provided
         """
-        catalog = cast(DictionaryObject, self.trailer["/Root"])
         # From the catalog get the embedded file names
         try:
             filenames = cast(
                 ArrayObject,
                 cast(
                     DictionaryObject,
-                    cast(DictionaryObject, catalog["/Names"])["/EmbeddedFiles"],
+                    cast(DictionaryObject, self.root_object["/Names"])[
+                        "/EmbeddedFiles"
+                    ],
                 )["/Names"],
             )
         except KeyError: