BUG: NameObject doesn't handle all values correctly

pypdf/generic/_base.py

@@ -43,7 +43,12 @@
     read_until_regex,
     str_,
 )
-from ..errors import STREAM_TRUNCATED_PREMATURELY, PdfReadError, PdfStreamError
+from ..errors import (
+    STREAM_TRUNCATED_PREMATURELY,
+    PdfReadError,
+    PdfStreamError,
+    PyPdfError,
+)
 
 __author__ = "Mathieu Fenniak"
 __author_email__ = "biziqe@mathieu.fenniak.net"
@@ -580,15 +585,14 @@ def write_to_stream(
 
 
 class NameObject(str, PdfObject):  # noqa: SLOT000
+    encoding = "utf8"
     delimiter_pattern = re.compile(rb"\s+|[\(\)<>\[\]{}/%]")
     surfix = b"/"
-    renumber_table: ClassVar[Dict[str, bytes]] = {
-        "#": b"#23",
-        "(": b"#28",
-        ")": b"#29",
-        "/": b"#2F",
-        "%": b"#25",
-        **{chr(i): f"#{i:02X}".encode() for i in range(33)},
+    renumber_table: ClassVar[Dict[int, bytes]] = {
+        **{i: f"#{i:02X}".encode() for i in range(0x21)},
+        **{i: bytes([i]) for i in range(0x21, 0x7F)},
+        **{i: f"#{i:02X}".encode() for i in b"#()<>[]{}/%"},
+        **{i: f"#{i:02X}".encode() for i in range(0x7F, 0x100)},
     }
 
     def clone(
@@ -610,24 +614,19 @@ def write_to_stream(
             deprecate_no_replacement(
                 "the encryption_key parameter of write_to_stream", "5.0.0"
             )
+        if "\0" in self:
+            raise PyPdfError("Null character is not allowed in NameObject")
         stream.write(self.renumber())
 
     def renumber(self) -> bytes:
-        out = self[0].encode("utf-8")
-        if out != b"/":
+        out = self.surfix
+        if self[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"):
-                    out += f"#{x:02X}".encode()
-            else:
-                try:
-                    out += self.renumber_table[c]
-                except KeyError:
-                    out += c.encode("utf-8")
+        for c in self[1:].encode(self.encoding):
+            out += self.renumber_table[c]
         return out
 
     @staticmethod
@@ -643,38 +642,19 @@ 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)
         if name != NameObject.surfix:
             raise PdfReadError("name read error")
         name += read_until_regex(stream, NameObject.delimiter_pattern)
+        name = NameObject.unnumber(name)
         try:
-            # Name objects should represent irregular characters
-            # with a '#' followed by the symbol's hex number
-            name = NameObject.unnumber(name)
-            for enc in NameObject.CHARSETS:
-                try:
-                    ret = name.decode(enc)
-                    return NameObject(ret)
-                except Exception:
-                    pass
-            raise UnicodeDecodeError("", name, 0, 0, "Code Not Found")
-        except (UnicodeEncodeError, UnicodeDecodeError) as e:
-            if not pdf.strict:
-                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 NameObject ({name!r}). "
-                    "You may need to adjust NameObject.CHARSETS.",
-                ) from e
+            name = NameObject(name.decode(NameObject.encoding))
+        except UnicodeDecodeError:
+            name = NameObject(name.decode("latin1"))
+            name.encoding = "latin1"
+        return name
 
 
 def encode_pdfdocencoding(unicode_string: str) -> bytes:

tests/test_generic.py

@@ -9,7 +9,7 @@
 
 from pypdf import PdfMerger, PdfReader, PdfWriter
 from pypdf.constants import CheckboxRadioButtonAttributes
-from pypdf.errors import PdfReadError, PdfStreamError
+from pypdf.errors import PdfReadError, PdfStreamError, PyPdfError
 from pypdf.generic import (
     AnnotationBuilder,
     ArrayObject,
@@ -185,17 +185,6 @@ def test_name_object(caplog):
     with pytest.raises(PdfReadError) as exc:
         NameObject.read_from_stream(stream, None)
     assert exc.value.args[0] == "name read error"
-    assert (
-        NameObject.read_from_stream(
-            BytesIO(b"/A;Name_With-Various***Characters?"), None
-        )
-        == "/A;Name_With-Various***Characters?"
-    )
-    assert (
-        NameObject.read_from_stream(BytesIO(b"/paired#28#29parentheses"), None)
-        == "/paired()parentheses"
-    )
-    assert NameObject.read_from_stream(BytesIO(b"/A#42"), None) == "/AB"
 
     assert (
         NameObject.read_from_stream(
@@ -223,7 +212,6 @@ def test_name_object(caplog):
     NameObject("/hello").write_to_stream(b)
     assert bytes(b.getbuffer()) == b"/hello"
 
-    caplog.clear()
     b = BytesIO()
     with pytest.raises(DeprecationWarning):
         NameObject("hello").write_to_stream(b)
@@ -239,6 +227,52 @@ def test_name_object(caplog):
     assert bytes(b.getbuffer()) == b"/#E4#BD#A0#E5#A5#BD#E4#B8#96#E7#95#8C#20#28#25#29"
     assert caplog.text == ""
 
+    # ISO/DIS 32000-2 Table 4: Examples of literal names
+    for b, n in (
+        (b"/Name1", "Name1"),
+        (b"/ASomewhatLongerName", "ASomewhatLongerName"),
+        (b"/A;Name_With-Various***Characters?", "A;Name_With-Various***Characters?"),
+        (b"/1.2", "1.2"),
+        (b"/$$", "$$"),
+        (b"/@pattern", "@pattern"),
+        (b"/.notdef", ".notdef"),
+        (b"/Lime#20Green", "Lime Green"),
+        (b"/paired#28#29parentheses", "paired()parentheses"),
+        (b"/The_Key_of_F#23_Minor", "The_Key_of_F#_Minor"),
+    ):
+        assert (NameObject.read_from_stream(BytesIO(b), None)) == "/" + n
+        bio = BytesIO()
+        NameObject("/" + n).write_to_stream(bio)
+        assert bio.getbuffer() == b
+    assert (NameObject.read_from_stream(BytesIO(b"/A#42"), None)) == "/" + "AB"
+
+    with pytest.raises(PyPdfError):
+        NameObject("/\0").write_to_stream(BytesIO())
+
+    # testing all allowed values
+    NameObject.encoding = "latin1"
+    value = "/" + bytes(range(1, 0x100)).decode("latin1")
+    bio = BytesIO()
+    NameObject(value).write_to_stream(bio)
+    bio.seek(0)
+    assert (
+        bio.read()
+        == b'/#01#02#03#04#05#06#07#08#09#0A#0B#0C#0D#0E#0F#10#11#12#13#14#15#16#17#18#19#1A#1B#1C#1D#1E#1F#20!"#23$#25'
+        b"&'#28#29*+,-.#2F0123456789:;#3C=#3E?@ABCDEFGHIJKLMNOPQRSTUVWXYZ#5B\\#5D^_`abcdefghijklmnopqrstuvwxyz#7B|#7D~#"
+        b"7F#80#81#82#83#84#85#86#87#88#89#8A#8B#8C#8D#8E#8F#90#91#92#93#94#95#96#97#98#99#9A#9B#9C#9D#9E#9F#A0#A1#A2#A"
+        b"3#A4#A5#A6#A7#A8#A9#AA#AB#AC#AD#AE#AF#B0#B1#B2#B3#B4#B5#B6#B7#B8#B9#BA#BB#BC#BD#BE#BF#C0#C1#C2#C3#C4#C5#C6#C7"
+        b"#C8#C9#CA#CB#CC#CD#CE#CF#D0#D1#D2#D3#D4#D5#D6#D7#D8#D9#DA#DB#DC#DD#DE#DF#E0#E1#E2#E3#E4#E5#E6#E7#E8#E9#EA#EB#"
+        b"EC#ED#EE#EF#F0#F1#F2#F3#F4#F5#F6#F7#F8#F9#FA#FB#FC#FD#FE#FF"
+    )
+    bio.seek(0)
+    assert NameObject.read_from_stream(bio, None) == value
+    NameObject.encoding = "utf8"
+    bio.seek(0)
+    name = NameObject.read_from_stream(bio, None)
+    assert name == value
+    assert name.encoding == "latin1"
+    assert NameObject.encoding == "utf8"
+
 
 def test_destination_fit_r():
     d = Destination(