gh-154848: Add check_frames option to pickletools.dis() and genops()

Doc/library/pickletools.rst

@@ -88,7 +88,8 @@ Programmatic interface
 ----------------------
 
 
-.. function:: dis(pickle, out=None, memo=None, indentlevel=4, annotate=0)
+.. function:: dis(pickle, out=None, memo=None, indentlevel=4, annotate=0, *, \
+                  check_frames=False)
 
    Outputs a symbolic disassembly of the pickle to the file-like
    object *out*, defaulting to ``sys.stdout``.  *pickle* can be a
@@ -101,17 +102,30 @@ Programmatic interface
    a short description.  The value of *annotate* is used as a hint for
    the column where annotation should start.
 
+   Framing (:pep:`3154`) is ignored by default, as an unpickler is free to do.
+   If *check_frames* is true, an argument that straddles a frame boundary, or a
+   frame that begins before the previous one ends, raises a :exc:`ValueError`,
+   as in the standard unpickler.
+
    .. versionchanged:: 3.2
       Added the *annotate* parameter.
 
-.. function:: genops(pickle)
+   .. versionchanged:: next
+      Added the *check_frames* parameter.
+
+.. function:: genops(pickle, *, check_frames=False)
 
    Provides an :term:`iterator` over all of the opcodes in a pickle, returning a
    sequence of ``(opcode, arg, pos)`` triples.  *opcode* is an instance of an
    :class:`OpcodeInfo` class; *arg* is the decoded value, as a Python object, of
    the opcode's argument; *pos* is the position at which this opcode is located.
    *pickle* can be a string or a file-like object.
 
+   The *check_frames* argument has the same meaning as in :func:`dis`.
+
+   .. versionchanged:: next
+      Added the *check_frames* parameter.
+
 .. function:: optimize(picklestring)
 
    Returns a new equivalent pickle string after eliminating unused ``PUT``

Lib/pickletools.py

@@ -2293,7 +2293,38 @@ def assure_pickle_consistency(verbose=False):
 ##############################################################################
 # A pickle opcode generator.
 
-def _genops(data, yield_end_pos=False):
+class _FramedReader:
+    """Wrap a file object to enforce pickle frame boundaries (PEP 3154).
+
+    Reads are limited to the current frame, so an argument that would straddle
+    a frame boundary is reported as truncated data by the argument readers,
+    just like a genuinely truncated pickle.
+    """
+
+    def __init__(self, file):
+        self.file = file
+        self.frame_len = None   # bytes left in the current frame, or None
+
+    def read(self, n):
+        if self.frame_len is None:
+            return self.file.read(n)
+        data = self.file.read(min(n, self.frame_len))
+        self.frame_len -= len(data)
+        if self.frame_len == 0:
+            self.frame_len = None
+        return data
+
+    def readline(self):
+        if self.frame_len is None:
+            return self.file.readline()
+        data = self.file.readline(self.frame_len)
+        self.frame_len -= len(data)
+        if self.frame_len == 0:
+            self.frame_len = None
+        return data
+
+
+def _genops(data, yield_end_pos=False, check_frames=False):
     if isinstance(data, bytes_types):
         data = io.BytesIO(data)
 
@@ -2317,6 +2348,13 @@ def _genops(data, yield_end_pos=False):
             arg = None
         else:
             arg = opcode.arg.reader(data)
+        if check_frames and opcode.name == 'FRAME':
+            if not isinstance(data, _FramedReader):
+                data = _FramedReader(data)
+            elif data.frame_len is not None:
+                raise ValueError("beginning of a new frame before end of "
+                                 "current frame")
+            data.frame_len = arg or None
         if yield_end_pos:
             yield opcode, arg, pos, getpos()
         else:
@@ -2325,7 +2363,7 @@ def _genops(data, yield_end_pos=False):
             assert opcode.name == 'STOP'
             break
 
-def genops(pickle):
+def genops(pickle, *, check_frames=False):
     """Generate all the opcodes in a pickle.
 
     'pickle' is a file-like object, or string, containing the pickle.
@@ -2347,8 +2385,13 @@ def genops(pickle):
     it's wrapped in a BytesIO object, and the latter's tell() result is
     used.  Else (the pickle doesn't have a tell(), and it's not obvious how
     to query its current position) pos is None.
+
+    Framing (PEP 3154) is ignored by default, as an unpickler is free to do.
+    If 'check_frames' is true, an argument that straddles a frame boundary, or
+    a frame that begins before the previous one ends, raises a ValueError, as
+    it does in the standard unpickler.
     """
-    return _genops(pickle)
+    return _genops(pickle, check_frames=check_frames)
 
 ##############################################################################
 # A pickle optimizer.
@@ -2420,7 +2463,8 @@ def optimize(p):
 ##############################################################################
 # A symbolic pickle disassembler.
 
-def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0):
+def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0, *,
+        check_frames=False):
     """Produce a symbolic disassembly of a pickle.
 
     'pickle' is a file-like object, or string, containing a (at least one)
@@ -2457,6 +2501,9 @@ def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0):
     + A memo entry isn't referenced before it's defined.
 
     + The markobject isn't stored in the memo.
+
+    Framing (PEP 3154) is ignored by default.  If 'check_frames' is true,
+    frame boundaries are enforced as in the standard unpickler; see genops().
     """
 
     # Most of the hair here is for sanity checks, but most of it is needed
@@ -2472,7 +2519,7 @@ def dis(pickle, out=None, memo=None, indentlevel=4, annotate=0):
     errormsg = None
     annocol = annotate  # column hint for annotations
     t = get_theme(tty_file=out).pickletools
-    for opcode, arg, pos in genops(pickle):
+    for opcode, arg, pos in genops(pickle, check_frames=check_frames):
         if pos is not None:
             print(f"{t.position}{pos:5d}:{t.reset}", end=' ', file=out)
 

Lib/test/test_pickletools.py

@@ -142,6 +142,44 @@ def test_truncated_data(self):
                 'not enough data in stream to read int4'):
             next(it)
 
+    def test_check_frames(self):
+        # Valid framed pickles are disassembled the same way regardless of
+        # whether frame boundaries are checked.
+        valid = pickle.dumps([1, 2, 3], 4)
+        for check in (False, True):
+            with self.subTest(check_frames=check):
+                ops = [op.name for op, arg, pos in
+                       pickletools.genops(valid, check_frames=check)]
+                self.assertIn('FRAME', ops)
+                self.assertEqual(ops[-1], 'STOP')
+
+    def test_check_frames_straddle(self):
+        # An opcode argument that straddles a frame boundary is ignored by
+        # default, but rejected when check_frames is true.  See gh-154848.
+        # FRAME 3; SHORT_BINBYTES argument straddles the frame.
+        data = b'\x80\x04\x95\x03\x00\x00\x00\x00\x00\x00\x00C\x0ahelloworld.'
+        self.assertEqual(list(pickletools.genops(data))[-1][0].name, 'STOP')
+        with self.assertRaisesRegex(ValueError,
+                'expected 10 bytes in a bytes1, but only 1 remain'):
+            list(pickletools.genops(data, check_frames=True))
+
+        # FRAME 6; UNICODE argument (read by readline) straddles the frame.
+        data = b'\x80\x04\x95\x06\x00\x00\x00\x00\x00\x00\x00Vhelloworld\n.'
+        self.assertEqual(list(pickletools.genops(data))[-1][0].name, 'STOP')
+        with self.assertRaisesRegex(ValueError,
+                'no newline found when trying to read unicodestringnl'):
+            list(pickletools.genops(data, check_frames=True))
+
+    def test_check_frames_nested(self):
+        # A new frame beginning before the current one ends is rejected only
+        # when check_frames is true.
+        data = (b'\x80\x04\x95\x0c\x00\x00\x00\x00\x00\x00\x00'
+                b'N\x95\x00\x00\x00\x00\x00\x00\x00\x00NN.')
+        self.assertEqual(list(pickletools.genops(data))[-1][0].name, 'STOP')
+        with self.assertRaisesRegex(ValueError,
+                'beginning of a new frame before end of current frame'):
+            list(pickletools.genops(data, check_frames=True))
+
     def test_unknown_opcode(self):
         it = pickletools.genops(b'N\xff')
         item = next(it)

Misc/NEWS.d/next/Library/2026-07-30-10-00-00.gh-issue-154848.Pk7Fr8.rst

@@ -0,0 +1,5 @@
+:func:`pickletools.dis` and :func:`pickletools.genops` now accept a
+*check_frames* keyword argument.  When true, an argument that straddles a
+frame boundary, or a frame that begins before the previous one ends, raises
+:exc:`ValueError` instead of being read across the boundary (PEP 3154).
+Framing is still ignored by default, as an unpickler is free to do.