ENH: loosen XLS signature

doc/source/whatsnew/v1.3.0.rst

@@ -831,6 +831,7 @@ I/O
 - Bug in :meth:`DataFrame.to_string` misplacing the truncation column when ``index=False`` (:issue:`40907`)
 - Bug in :func:`read_orc` always raising ``AttributeError`` (:issue:`40918`)
 - Bug in the conversion from pyarrow to pandas (e.g. for reading Parquet) with nullable dtypes and a pyarrow array whose data buffer size is not a multiple of dtype size (:issue:`40896`)
+- Loosen XLS signatures used in :func:`read_excel` to determine if the `xlrd` engine should be used (:issue:`41225`)
 
 Period
 ^^^^^^

pandas/io/excel/_base.py

@@ -1014,16 +1014,22 @@ def close(self):
         return content
 
 
-XLS_SIGNATURE = b"\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1"
-ZIP_SIGNATURE = b"PK\x03\x04"
-PEEK_SIZE = max(len(XLS_SIGNATURE), len(ZIP_SIGNATURE))
+SIGNATURES = {
+    "biff2": b"\x09\x00\x04\x00",
+    "biff3": b"\x09\x02\x06\x00",
+    "biff4": b"\x09\x04\x06\x00",
+    "biff5": b"\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1",
+    "biff8": b"\xD0\xCF\x11\xE0\xA1\xB1\x1A\xE1",
+    "zip": b"PK\x03\x04",
+}
+PEEK_SIZE = max(map(len, SIGNATURES.values()))
 
 
 @doc(storage_options=_shared_docs["storage_options"])
 def inspect_excel_format(
     content_or_path: FilePathOrBuffer,
     storage_options: StorageOptions = None,
-) -> str:
+) -> str | None:
     """
     Inspect the path or content of an excel file and get its format.
 
@@ -1037,8 +1043,8 @@ def inspect_excel_format(
 
     Returns
     -------
-    str
-        Format of file.
+    str or None
+        Format of file (if it can be determined)
 
     Raises
     ------
@@ -1063,10 +1069,14 @@ def inspect_excel_format(
             peek = buf
         stream.seek(0)
 
-        if peek.startswith(XLS_SIGNATURE):
+        if any(
+            peek.startswith(signature)
+            for (file_format, signature) in SIGNATURES.items()
+            if file_format.startswith("biff")
+        ):
             return "xls"
-        elif not peek.startswith(ZIP_SIGNATURE):
-            raise ValueError("File is not a recognized excel file")
+        elif not peek.startswith(SIGNATURES["zip"]):
+            return None
 
         # ZipFile typing is overly-strict
         # https://github.com/python/typeshed/issues/4212
@@ -1174,8 +1184,12 @@ def __init__(
                 ext = inspect_excel_format(
                     content_or_path=path_or_buffer, storage_options=storage_options
                 )
+                if ext is None:
+                    raise ValueError(
+                        "Excel file format cannot be determined, you must specify "
+                        "an engine manually."
+                    )
 
-            # ext will always be valid, otherwise inspect_excel_format would raise
             engine = config.get_option(f"io.excel.{ext}.reader", silent=True)
             if engine == "auto":
                 engine = get_default_engine(ext, mode="reader")
@@ -1190,12 +1204,13 @@ def __init__(
                         path_or_buffer, storage_options=storage_options
                     )
 
-            if ext != "xls" and xlrd_version >= "2":
+            # Pass through if ext is None, otherwise check if ext valid for xlrd
+            if ext and ext != "xls" and xlrd_version >= "2":
                 raise ValueError(
                     f"Your version of xlrd is {xlrd_version}. In xlrd >= 2.0, "
                     f"only the xls format is supported. Install openpyxl instead."
                 )
-            elif ext != "xls":
+            elif ext and ext != "xls":
                 caller = inspect.stack()[1]
                 if (
                     caller.filename.endswith(

pandas/tests/io/excel/test_readers.py

@@ -718,9 +718,19 @@ def test_missing_file_raises(self, read_ext):
 
     def test_corrupt_bytes_raises(self, read_ext, engine):
         bad_stream = b"foo"
-        if engine is None or engine == "xlrd":
+        if engine is None:
             error = ValueError
-            msg = "File is not a recognized excel file"
+            msg = (
+                "Excel file format cannot be determined, you must "
+                "specify an engine manually."
+            )
+        elif engine == "xlrd":
+            from xlrd import XLRDError
+            error = XLRDError
+            msg = (
+                "Unsupported format, or corrupt file: Expected BOF "
+                "record; found b'foo'"
+            )
         else:
             error = BadZipFile
             msg = "File is not a zip file"