From f9917c23ed0cd0b9a57404abd4d71aa015aa14a3 Mon Sep 17 00:00:00 2001
From: Cody Maloney <cmaloney@theoreticalchaos.com>
Date: Sat, 25 Oct 2025 21:09:20 -0700
Subject: [PATCH] gh-104607: Validate returned byte count in RawIOBase.read

While `RawIOBase.readinto` should return a count of bytes between 0 and
the length of the given buffer, it is not required to. Add validation
inside RawIOBase.read that the returned byte count is reasonable.
---
 Lib/_pyio.py                                    |  2 ++
 Lib/test/test_io/test_general.py                | 17 +++++++++++++++++
 ...25-10-25-21-04-00.gh-issue-104607.oOZGxS.rst |  2 ++
 Modules/_io/iobase.c                            | 13 ++++++++++---
 4 files changed, 31 insertions(+), 3 deletions(-)
 create mode 100644 Misc/NEWS.d/next/Library/2025-10-25-21-04-00.gh-issue-104607.oOZGxS.rst

diff --git a/Lib/_pyio.py b/Lib/_pyio.py
index 9ae72743919a32..8ae2de51184708 100644
--- a/Lib/_pyio.py
+++ b/Lib/_pyio.py
@@ -617,6 +617,8 @@ def read(self, size=-1):
         n = self.readinto(b)
         if n is None:
             return None
+        if n < 0 or n > len(b):
+            raise ValueError(f"readinto returned '{n}' outside buffer size '{len(b)}'")
         del b[n:]
         return bytes(b)
 
diff --git a/Lib/test/test_io/test_general.py b/Lib/test/test_io/test_general.py
index ac9c5a425d7ea2..6a502f14e536ba 100644
--- a/Lib/test/test_io/test_general.py
+++ b/Lib/test/test_io/test_general.py
@@ -592,6 +592,23 @@ def test_RawIOBase_read(self):
         self.assertEqual(rawio.read(2), None)
         self.assertEqual(rawio.read(2), b"")
 
+    def test_RawIOBase_read_bounds_checking(self):
+        # Make sure a `.readinto` call which returns a value oustside
+        # (0, len(buffer)) raises.
+        class Misbehaved(io.RawIOBase):
+            def __init__(self, readinto_return) -> None:
+                self._readinto_return = readinto_return
+            def readinto(self, b):
+                return self._readinto_return
+
+        with self.assertRaises(ValueError) as cm:
+            Misbehaved(2).read(1)
+        self.assertEqual(str(cm.exception), "readinto returned '2' oustside buffer size '1'")
+        self.assertRaises(ValueError, Misbehaved(2147483647).read)
+        self.assertRaises(ValueError, Misbehaved(sys.maxsize).read)
+        self.assertRaises(ValueError, Misbehaved(-1).read)
+        self.assertRaises(ValueError, Misbehaved(-1000).read)
+
     def test_types_have_dict(self):
         test = (
             self.IOBase(),
diff --git a/Misc/NEWS.d/next/Library/2025-10-25-21-04-00.gh-issue-104607.oOZGxS.rst b/Misc/NEWS.d/next/Library/2025-10-25-21-04-00.gh-issue-104607.oOZGxS.rst
new file mode 100644
index 00000000000000..eb0e542488f8dd
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2025-10-25-21-04-00.gh-issue-104607.oOZGxS.rst
@@ -0,0 +1,2 @@
+Inside :meth:`io.RawIOBase.read` validate that the count of bytes returned by
+:meth:`io.RawIOBase.readinto` is reasonable (inside the provided buffer).
diff --git a/Modules/_io/iobase.c b/Modules/_io/iobase.c
index acadbcc4d59c38..e62dbeded16950 100644
--- a/Modules/_io/iobase.c
+++ b/Modules/_io/iobase.c
@@ -939,14 +939,21 @@ _io__RawIOBase_read_impl(PyObject *self, Py_ssize_t n)
         return res;
     }
 
-    n = PyNumber_AsSsize_t(res, PyExc_ValueError);
+    Py_ssize_t bytes_filled = PyNumber_AsSsize_t(res, PyExc_ValueError);
     Py_DECREF(res);
-    if (n == -1 && PyErr_Occurred()) {
+    if (readinto_len == -1 && PyErr_Occurred()) {
         Py_DECREF(b);
         return NULL;
     }
+    if (bytes_filled < 0 || bytes_filled > n) {
+        Py_DECREF(b);
+        PyErr_Format(PyExc_ValueError,
+                        "readinto returned '%zd' oustside buffer size '%zd'",
+                        bytes_filled, n);
+        return NULL;
+    }
 
-    res = PyBytes_FromStringAndSize(PyByteArray_AsString(b), n);
+    res = PyBytes_FromStringAndSize(PyByteArray_AsString(b), bytes_filled);
     Py_DECREF(b);
     return res;
 }