Merge branch 'main' into feat/add-brotli-decode

Author: ash01ish

pypdf/_doc_common.py

@@ -978,13 +978,13 @@ def _build_outline_item(self, node: DictionaryObject) -> Optional[Destination]:
             title = ""
 
         if "/A" in node:
-            # Action, PDFv1.7 Section 12.6 (only type GoTo supported)
+            # Action, PDF 1.7 and PDF 2.0 §12.6 (only type GoTo supported)
             action = cast(DictionaryObject, node["/A"])
             action_type = cast(NameObject, action[GoToActionArguments.S])
             if action_type == "/GoTo":
                 dest = action[GoToActionArguments.D]
         elif "/Dest" in node:
-            # Destination, PDFv1.7 Section 12.3.2
+            # Destination, PDF 1.7 and PDF 2.0 §12.3.2
             dest = node["/Dest"]
             # if array was referenced in another object, will be a dict w/ key "/D"
             if isinstance(dest, DictionaryObject) and "/D" in dest:
@@ -994,7 +994,7 @@ def _build_outline_item(self, node: DictionaryObject) -> Optional[Destination]:
             outline_item = self._build_destination(title, dest)
         elif isinstance(dest, str):
             # named destination, addresses NameObject Issue #193
-            # TODO : keep named destination instead of replacing it ?
+            # TODO: Keep named destination instead of replacing it?
             try:
                 outline_item = self._build_destination(
                     title, self._named_destinations[dest].dest_array
@@ -1028,7 +1028,7 @@ def _build_outline_item(self, node: DictionaryObject) -> Optional[Destination]:
                 # absolute value = num. visible children
                 # with positive = open/unfolded, negative = closed/folded
                 outline_item[NameObject("/Count")] = node["/Count"]
-            #  if count is 0 we will consider it as open ( in order to have always an is_open to simplify
+            #  if count is 0 we will consider it as open (to have available is_open)
             outline_item[NameObject("/%is_open%")] = BooleanObject(
                 node.get("/Count", 0) >= 0
             )
@@ -1428,8 +1428,8 @@ def _repr_mimebundle_(
 
 
 class LazyDict(Mapping[Any, Any]):
-    def __init__(self, *args: Any, **kw: Any) -> None:
-        self._raw_dict = dict(*args, **kw)
+    def __init__(self, *args: Any, **kwargs: Any) -> None:
+        self._raw_dict = dict(*args, **kwargs)
 
     def __getitem__(self, key: str) -> Any:
         func, arg = self._raw_dict.__getitem__(key)

pypdf/_page.py

@@ -144,7 +144,7 @@ class Transformation:
     Represent a 2D transformation.
 
     The transformation between two coordinate systems is represented by a 3-by-3
-    transformation matrix matrix with the following form::
+    transformation matrix with the following form::
 
         a b 0
         c d 0
@@ -467,7 +467,7 @@ def __getitem__(
         len_self = len(lst)
         if index < 0:
             # support negative indexes
-            index = len_self + index
+            index += len_self
         if index < 0 or index >= len_self:
             raise IndexError("Sequence index out of range")
         return self.get_function(lst[index])
@@ -575,7 +575,7 @@ def create_blank_page(
         """
         page = PageObject(pdf)
 
-        # Creates a new page (cf PDF Reference  7.7.3.3)
+        # Creates a new page (cf PDF Reference §7.7.3.3)
         page.__setitem__(NameObject(PG.TYPE), NameObject("/Page"))
         page.__setitem__(NameObject(PG.PARENT), NullObject())
         page.__setitem__(NameObject(PG.RESOURCES), DictionaryObject())
@@ -907,7 +907,7 @@ def _merge_resources(
             pdf = self.indirect_reference.pdf
             is_pdf_writer = hasattr(
                 pdf, "_add_object"
-            )  # ---------- expect isinstance(pdf,PdfWriter)
+            )  # expect isinstance(pdf, PdfWriter)
         except (AssertionError, AttributeError):
             pdf = None
             is_pdf_writer = False
@@ -927,8 +927,8 @@ def compute_unique_key(base_key: str) -> Tuple[str, bool]:
 
             """
             value = page2res.raw_get(base_key)
-            # TODO : possible improvement : in case of writer, the indirect_reference
-            # can not be found because translated : this may be improved
+            # TODO: a possible improvement for writer, the indirect_reference
+            # cannot be found because translated
 
             # try the current key first (e.g. "foo"), but otherwise iterate
             # through "foo-0", "foo-1", etc. new_res can contain only finitely
@@ -1097,7 +1097,7 @@ def replace_contents(
             assert content is not None, "mypy"
             content.indirect_reference = self[
                 PG.CONTENTS
-            ].indirect_reference  # TODO: in a future may required generation management
+            ].indirect_reference  # TODO: in the future may require generation management
             try:
                 self.indirect_reference.pdf._objects[
                     content.indirect_reference.idnum - 1  # type: ignore
@@ -1147,7 +1147,7 @@ def _merge_page(
             assert isinstance(self.indirect_reference, IndirectObject)
             if hasattr(
                 self.indirect_reference.pdf, "_add_object"
-            ):  # ---------- to detect PdfWriter
+            ):  # to detect PdfWriter
                 return self._merge_page_writer(
                     page2, page2transformation, ctm, over, expand
                 )
@@ -1683,7 +1683,7 @@ def page_number(self) -> Optional[int]:
         Read-only property which returns the page number within the PDF file.
 
         Returns:
-            int : page number; None if the page is not attached to a PDF.
+            Page number; None if the page is not attached to a PDF.
 
         """
         if self.indirect_reference is None:
@@ -1837,7 +1837,7 @@ def _extract_text(
         try:
             objr = obj
             while NameObject(PG.RESOURCES) not in objr:
-                # /Resources can be inherited sometimes so we look to parents
+                # /Resources can be inherited so we look to parents
                 objr = objr["/Parent"].get_object()
                 # If no parents then no /Resources will be available,
                 # so an exception will be raised
@@ -1900,8 +1900,8 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
             nonlocal char_scale, space_scale, _space_width, TL, font_size, cmap
             nonlocal orientations, rtl_dir, visitor_text, output, text, _actual_str_size
 
-            check_crlf_space: bool = False
             str_widths: float = 0.0
+
             # Table 5.4 page 405
             if operator == b"BT":
                 tm_matrix = [1.0, 0.0, 0.0, 1.0, 0.0, 0.0]
@@ -1919,6 +1919,7 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
                 text = ""
                 memo_cm = cm_matrix.copy()
                 memo_tm = tm_matrix.copy()
+
             # Table 4.7 "Graphics state operators", page 219
             # cm_matrix calculation is reserved for later
             elif operator == b"q":
@@ -1957,6 +1958,7 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
                 )
                 memo_cm = cm_matrix.copy()
                 memo_tm = tm_matrix.copy()
+
             # Table 5.2 page 398
             elif operator == b"Tz":
                 char_scale = float(operands[0]) / 100 if operands else 1.0
@@ -2005,7 +2007,6 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
                     pass  # keep previous size
             # Table 5.5 page 406
             elif operator == b"Td":
-                check_crlf_space = True
                 # A special case is a translating only tm:
                 # tm = [1, 0, 0, 1, e, f]
                 # i.e. tm[4] += tx, tm[5] += ty.
@@ -2015,18 +2016,15 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
                 str_widths = compute_str_widths(_actual_str_size["str_widths"])
                 _actual_str_size["str_widths"] = 0.0
             elif operator == b"Tm":
-                check_crlf_space = True
                 tm_matrix = [float(operand) for operand in operands[:6]]
                 str_widths = compute_str_widths(_actual_str_size["str_widths"])
                 _actual_str_size["str_widths"] = 0.0
             elif operator == b"T*":
-                check_crlf_space = True
                 tm_matrix[4] -= TL * tm_matrix[2]
                 tm_matrix[5] -= TL * tm_matrix[3]
                 str_widths = compute_str_widths(_actual_str_size["str_widths"])
                 _actual_str_size["str_widths"] = 0.0
             elif operator == b"Tj":
-                check_crlf_space = True
                 text, rtl_dir, _actual_str_size = self._handle_tj(
                     text,
                     operands,
@@ -2043,7 +2041,7 @@ def process_operation(operator: bytes, operands: List[Any]) -> None:
             else:
                 return
 
-            if check_crlf_space:
+            if operator in {b"Td", b"Tm", b"T*", b"Tj"}:
                 try:
                     text, output, cm_prev, tm_prev = crlf_space_check(
                         text,
@@ -2421,7 +2419,7 @@ def _get_fonts(self) -> Tuple[Set[str], Set[str]]:
         Get the names of embedded fonts and unembedded fonts.
 
         Returns:
-            A tuple (Set of embedded fonts, set of unembedded fonts)
+            A tuple (set of embedded fonts, set of unembedded fonts)
 
         """
         obj = self.get_object()
@@ -2518,7 +2516,7 @@ def __getitem__(
         len_self = len(self)
         if index < 0:
             # support negative indexes
-            index = len_self + index
+            index += len_self
         if index < 0 or index >= len_self:
             raise IndexError("Sequence index out of range")
         return self.get_function(index)
@@ -2537,8 +2535,8 @@ def __delitem__(self, index: Union[int, slice]) -> None:
         len_self = len(self)
         if index < 0:
             # support negative indexes
-            index = len_self + index
-        if index < 0 or index >= len_self:
+            index += len_self
+        if not (0 <= index < len_self):
             raise IndexError("Index out of range")
         ind = self[index].indirect_reference
         assert ind is not None

pypdf/_writer.py

@@ -2054,7 +2054,7 @@ def remove_objects_from_page(
             text_filters: Properties of text to be deleted, if applicable. Optional.
                 This is a Python dictionary with the following properties:
 
-                * font_ids: List of font IDs (such as /F1 or /T1_0) to be deleted.
+                * font_ids: List of font resource IDs (such as /F1 or /T1_0) to be deleted.
 
         """
         if isinstance(to_delete, (list, tuple)):
@@ -2119,8 +2119,9 @@ def clean(
                     )
                 ):
                     if (
-                      not to_delete & ObjectDeletionFlag.TEXT
-                      or (not font_ids_to_delete or font_id in font_ids_to_delete)
+                        not to_delete & ObjectDeletionFlag.TEXT
+                        or (to_delete & ObjectDeletionFlag.TEXT and not text_filters)
+                        or (to_delete & ObjectDeletionFlag.TEXT and font_id in font_ids_to_delete)
                     ):
                         del content.operations[i]
                     else:
@@ -2246,16 +2247,49 @@ def remove_text(self, font_names: Optional[List[str]] = None) -> None:
             font_names = []
 
         for page in self.pages:
-            font_ids = []
-            fonts = page.get("/Resources", {}).get("/Font", {})
-            for font_id, font_info in fonts.items():
-                font_name = font_info.get("/BaseFont", "").split("+")[-1]
-                if font_name in font_names:
-                    font_ids.append(font_id)
-
-            text_filters = {
-                "font_ids": font_ids,
-            }
+            resource_ids_to_remove = []
+
+            # Content streams reference fonts and other resources with names like "/F1" or "/T1_0"
+            # Font names need to be converted to resource names/IDs for easier removal
+            if font_names:
+                # Recursively loop through page objects to gather font info
+                def get_font_info(
+                    obj: Any,
+                    font_info: Optional[Dict[str, Any]] = None,
+                    key: Optional[str] = None
+                ) -> Dict[str, Any]:
+                    if font_info is None:
+                        font_info = {}
+                    if isinstance(obj, IndirectObject):
+                        obj = obj.get_object()
+                    if isinstance(obj, dict):
+                        if obj.get("/Type") == "/Font":
+                            font_name = obj.get("/BaseFont", "")
+                            # Normalize font names like "/RRXFFV+Palatino-Bold" to "Palatino-Bold"
+                            normalized_font_name = font_name.lstrip("/").split("+")[-1]
+                            if normalized_font_name not in font_info:
+                                font_info[normalized_font_name] = {
+                                    "normalized_font_name": normalized_font_name,
+                                    "resource_ids": [],
+                                }
+                            if key not in font_info[normalized_font_name]["resource_ids"]:
+                                font_info[normalized_font_name]["resource_ids"].append(key)
+                        for k in obj:
+                            font_info = get_font_info(obj[k], font_info, k)
+                    elif isinstance(obj, (list, ArrayObject)):
+                        for child_obj in obj:
+                            font_info = get_font_info(child_obj, font_info)
+                    return font_info
+
+                # Add relevant resource names for removal
+                font_info = get_font_info(page.get("/Resources"))
+                for font_name in font_names:
+                    if font_name in font_info:
+                        resource_ids_to_remove.extend(font_info[font_name]["resource_ids"])
+
+            text_filters = {}
+            if font_names:
+                text_filters["font_ids"] = resource_ids_to_remove
             self.remove_objects_from_page(page, ObjectDeletionFlag.TEXT, text_filters=text_filters)
 
     def add_uri(