ROB: Capture UnicodeDecodeError at PdfReader.pdf_header (#1768)

Fixes #1758

pypdf/_reader.py

@@ -32,7 +32,7 @@
 import struct
 import zlib
 from datetime import datetime
-from io import BytesIO
+from io import BytesIO, UnsupportedOperation
 from pathlib import Path
 from typing import (
     Any,
@@ -360,7 +360,7 @@ def pdf_header(self) -> str:
         #       but that needs a deprecation
         loc = self.stream.tell()
         self.stream.seek(0, 0)
-        pdf_file_version = self.stream.read(8).decode("utf-8")
+        pdf_file_version = self.stream.read(8).decode("utf-8", "backslashreplace")
         self.stream.seek(loc, 0)  # return to where it was
         return pdf_file_version
 
@@ -1541,19 +1541,22 @@ def read(self, stream: StreamType) -> None:
 
     def _basic_validation(self, stream: StreamType) -> None:
         """Ensure file is not empty. Read at most 5 bytes."""
-        # start at the end:
-        stream.seek(0, os.SEEK_END)
-        if not stream.tell():
-            raise EmptyFileError("Cannot read an empty file")
-        if self.strict:
-            stream.seek(0, os.SEEK_SET)
+        stream.seek(0, os.SEEK_SET)
+        try:
             header_byte = stream.read(5)
-            if header_byte != b"%PDF-":
+        except UnicodeDecodeError:
+            raise UnsupportedOperation("cannot read header")
+        if header_byte == b"":
+            raise EmptyFileError("Cannot read an empty file")
+        elif header_byte != b"%PDF-":
+            if self.strict:
                 raise PdfReadError(
                     f"PDF starts with '{header_byte.decode('utf8')}', "
                     "but '%PDF-' expected"
                 )
-            stream.seek(0, os.SEEK_END)
+            else:
+                logger_warning(f"invalid pdf header: {header_byte}", __name__)
+        stream.seek(0, os.SEEK_END)
 
     def _find_eof_marker(self, stream: StreamType) -> None:
         """
@@ -1567,7 +1570,10 @@ def _find_eof_marker(self, stream: StreamType) -> None:
         line = b""
         while line[:5] != b"%%EOF":
             if stream.tell() < HEADER_SIZE:
-                raise PdfReadError("EOF marker not found")
+                if self.strict:
+                    raise PdfReadError("EOF marker not found")
+                else:
+                    logger_warning("EOF marker not found", __name__)
             line = read_previous_line(stream)
 
     def _find_startxref_pos(self, stream: StreamType) -> int:

tests/test_reader.py

@@ -460,10 +460,16 @@ def test_read_empty():
     assert exc.value.args[0] == "Cannot read an empty file"
 
 
-def test_read_malformed_header():
+def test_read_malformed_header(caplog):
     with pytest.raises(PdfReadError) as exc:
         PdfReader(io.BytesIO(b"foo"), strict=True)
     assert exc.value.args[0] == "PDF starts with 'foo', but '%PDF-' expected"
+    caplog.clear()
+    try:
+        PdfReader(io.BytesIO(b"foo"), strict=False)
+    except Exception:
+        pass
+    assert caplog.messages[0].startswith("invalid pdf header")
 
 
 def test_read_malformed_body():
@@ -1352,3 +1358,37 @@ def test_iss1710():
     name = "irbookonlinereading.pdf"
     in_pdf = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
     in_pdf.outline
+
+
+def test_broken_file_header():
+    pdf_data = (
+        b"%%PDF-\xa0sd\n"
+        b"1 0 obj << /Count 1 /Kids [4 0 R] /Type /Pages >> endobj\n"
+        b"2 0 obj << >> endobj\n"
+        b"3 0 obj << >> endobj\n"
+        b"4 0 obj << /Contents 3 0 R /CropBox [0.0 0.0 2550.0 3508.0]"
+        b" /MediaBox [0.0 0.0 2550.0 3508.0] /Parent 1 0 R"
+        b" /Resources << /Font << >> >>"
+        b" /Rotate 0 /Type /Page >> endobj\n"
+        b"5 0 obj << /Pages 1 0 R /Type /Catalog >> endobj\n"
+        b"xref 1 5\n"
+        b"%010d 00000 n\n"
+        b"%010d 00000 n\n"
+        b"%010d 00000 n\n"
+        b"%010d 00000 n\n"
+        b"%010d 00000 n\n"
+        b"trailer << %s/Root 5 0 R /Size 6 >>\n"
+        b"startxref %d\n"
+        b"%%%%EOF"
+    )
+    with_prev_0 = True
+    pdf_data = pdf_data % (
+        pdf_data.find(b"1 0 obj"),
+        pdf_data.find(b"2 0 obj"),
+        pdf_data.find(b"3 0 obj"),
+        pdf_data.find(b"4 0 obj"),
+        pdf_data.find(b"5 0 obj"),
+        b"/Prev 0 " if with_prev_0 else b"",
+        pdf_data.find(b"xref") - 1,
+    )
+    PdfReader(io.BytesIO(pdf_data))