ENH: Allow multiple charsets for NameObject.read_from_stream (#2585)

Closes #2323

Author: pubpub-zz

pypdf/generic/_base.py

@@ -615,7 +615,10 @@ def write_to_stream(
     def renumber(self) -> bytes:
         out = self[0].encode("utf-8")
         if out != b"/":
-            deprecate_no_replacement(f"Incorrect first char in NameObject, should start with '/': ({self})", "6.0.0")
+            deprecate_no_replacement(
+                f"Incorrect first char in NameObject, should start with '/': ({self})",
+                "6.0.0",
+            )
         for c in self[1:]:
             if c > "~":
                 for x in c.encode("utf-8"):
@@ -640,6 +643,8 @@ def unnumber(sin: bytes) -> bytes:
                 i = i + 1
         return sin
 
+    CHARSETS = ("utf-8", "gbk", "latin1")
+
     @staticmethod
     def read_from_stream(stream: StreamType, pdf: Any) -> "NameObject":  # PdfReader
         name = stream.read(1)
@@ -650,7 +655,7 @@ def read_from_stream(stream: StreamType, pdf: Any) -> "NameObject":  # PdfReader
             # Name objects should represent irregular characters
             # with a '#' followed by the symbol's hex number
             name = NameObject.unnumber(name)
-            for enc in ("utf-8", "gbk"):
+            for enc in NameObject.CHARSETS:
                 try:
                     ret = name.decode(enc)
                     return NameObject(ret)
@@ -659,11 +664,16 @@ def read_from_stream(stream: StreamType, pdf: Any) -> "NameObject":  # PdfReader
             raise UnicodeDecodeError("", name, 0, 0, "Code Not Found")
         except (UnicodeEncodeError, UnicodeDecodeError) as e:
             if not pdf.strict:
-                logger_warning(f"Illegal character in Name Object ({name!r})", __name__)
+                logger_warning(
+                    f"Illegal character in NameObject ({name!r}), "
+                    "you may need to adjust NameObject.CHARSETS",
+                    __name__,
+                )
                 return NameObject(name.decode("charmap"))
             else:
                 raise PdfReadError(
-                    f"Illegal character in Name Object ({name!r})"
+                    f"Illegal character in NameObject ({name!r}). "
+                    "You may need to adjust NameObject.CHARSETS.",
                 ) from e
 
 

tests/test_generic.py

@@ -1,5 +1,6 @@
 """Test the pypdf.generic module."""
 
+from copy import deepcopy
 from io import BytesIO
 from pathlib import Path
 from unittest.mock import patch
@@ -212,6 +213,11 @@ def test_name_object(caplog):
         )
     ) == "/你好世界"
 
+    # to test latin-1 aka stdencoding
+    assert (
+        NameObject.read_from_stream(BytesIO(b"/DocuSign\xae"), None)
+    ) == "/DocuSign®"
+
     # test write
     b = BytesIO()
     NameObject("/hello").write_to_stream(b)
@@ -1036,16 +1042,20 @@ def test_checkboxradiobuttonattributes_opt():
 
 
 def test_name_object_invalid_decode():
-    stream = BytesIO(b"/\x80\x02\x03")
-
-    # strict:
-    with pytest.raises(PdfReadError) as exc:
-        NameObject.read_from_stream(stream, ReaderDummy(strict=True))
-    assert "Illegal character in Name Object" in exc.value.args[0]
-
-    # non-strict:
-    stream.seek(0)
-    NameObject.read_from_stream(stream, ReaderDummy(strict=False))
+    charsets = deepcopy(NameObject.CHARSETS)
+    try:
+        NameObject.CHARSETS = ("utf-8",)
+        stream = BytesIO(b"/\x80\x02\x03")
+        # strict:
+        with pytest.raises(PdfReadError) as exc:
+            NameObject.read_from_stream(stream, ReaderDummy(strict=True))
+        assert "Illegal character in NameObject " in exc.value.args[0]
+
+        # non-strict:
+        stream.seek(0)
+        NameObject.read_from_stream(stream, ReaderDummy(strict=False))
+    finally:
+        NameObject.CHARSETS = charsets
 
 
 def test_indirect_object_invalid_read():