gh-99254: remove all unused consts from code objects (GH-99255)

3 files changed, 49 insertions, 14 deletions

diff --git a/Lib/importlib/_bootstrap_external.py b/Lib/importlib/_bootstrap_external.py
index 8cbc962..f4dbbeb 100644
--- a/Lib/importlib/_bootstrap_external.py
+++ b/Lib/importlib/_bootstrap_external.py
@@ -425,6 +425,7 @@ _code_type = type(_write_atomic.__code__)
 #     Python 3.12a1 3509 (Conditional jumps only jump forward)
 #     Python 3.12a1 3510 (FOR_ITER leaves iterator on the stack)
 #     Python 3.12a1 3511 (Add STOPITERATION_ERROR instruction)
+#     Python 3.12a1 3512 (Remove all unused consts from code objects)
 
 #     Python 3.13 will start with 3550
 
@@ -437,7 +438,7 @@ _code_type = type(_write_atomic.__code__)
 # Whenever MAGIC_NUMBER is changed, the ranges in the magic_values array
 # in PC/launcher.c must also be updated.
 
-MAGIC_NUMBER = (3511).to_bytes(2, 'little') + b'\r\n'
+MAGIC_NUMBER = (3512).to_bytes(2, 'little') + b'\r\n'
 
 _RAW_MAGIC_NUMBER = int.from_bytes(MAGIC_NUMBER, 'little')  # For import.c
 
diff --git a/Lib/test/test_compile.py b/Lib/test/test_compile.py
index f7847a3..a14509a 100644
--- a/Lib/test/test_compile.py
+++ b/Lib/test/test_compile.py
@@ -670,7 +670,7 @@ if 1:
         self.assertIs(f1.__code__.co_linetable, f2.__code__.co_linetable)
 
     @support.cpython_only
-    def test_strip_unused_consts(self):
+    def test_remove_unused_consts(self):
         def f():
             "docstring"
             if True:
@@ -679,7 +679,41 @@ if 1:
                 return "unused"
 
         self.assertEqual(f.__code__.co_consts,
-                         ("docstring", True, "used"))
+                         ("docstring", "used"))
+
+    @support.cpython_only
+    def test_remove_unused_consts_no_docstring(self):
+        # the first item (None for no docstring in this case) is
+        # always retained.
+        def f():
+            if True:
+                return "used"
+            else:
+                return "unused"
+
+        self.assertEqual(f.__code__.co_consts,
+                         (None, "used"))
+
+    @support.cpython_only
+    def test_remove_unused_consts_extended_args(self):
+        N = 1000
+        code = ["def f():\n"]
+        code.append("\ts = ''\n")
+        code.append("\tfor i in range(1):\n")
+        for i in range(N):
+            code.append(f"\t\tif True: s += 't{i}'\n")
+            code.append(f"\t\tif False: s += 'f{i}'\n")
+        code.append("\treturn s\n")
+
+        code = "".join(code)
+        g = {}
+        eval(compile(code, "file.py", "exec"), g)
+        exec(code, g)
+        f = g['f']
+        expected = tuple([None, '', 1] + [f't{i}' for i in range(N)])
+        self.assertEqual(f.__code__.co_consts, expected)
+        expected = "".join(expected[3:])
+        self.assertEqual(expected, f())
 
     # Stripping unused constants is not a strict requirement for the
     # Python semantics, it's a more an implementation detail.
diff --git a/Lib/test/test_dis.py b/Lib/test/test_dis.py
index 5640bf2..950af3c 100644
--- a/Lib/test/test_dis.py
+++ b/Lib/test/test_dis.py
@@ -168,13 +168,13 @@ dis_bug1333982 = """\
 %3d        RESUME                   0
 
 %3d        LOAD_ASSERTION_ERROR
-           LOAD_CONST               2 (<code object <listcomp> at 0x..., file "%s", line %d>)
+           LOAD_CONST               1 (<code object <listcomp> at 0x..., file "%s", line %d>)
            MAKE_FUNCTION            0
            LOAD_FAST                0 (x)
            GET_ITER
            CALL                     0
 
-%3d        LOAD_CONST               3 (1)
+%3d        LOAD_CONST               2 (1)
 
 %3d        BINARY_OP                0 (+)
            CALL                     0
@@ -1446,9 +1446,9 @@ def jumpy():
 # End fodder for opinfo generation tests
 expected_outer_line = 1
 _line_offset = outer.__code__.co_firstlineno - 1
-code_object_f = outer.__code__.co_consts[3]
+code_object_f = outer.__code__.co_consts[1]
 expected_f_line = code_object_f.co_firstlineno - _line_offset
-code_object_inner = code_object_f.co_consts[3]
+code_object_inner = code_object_f.co_consts[1]
 expected_inner_line = code_object_inner.co_firstlineno - _line_offset
 expected_jumpy_line = 1
 
@@ -1485,21 +1485,21 @@ expected_opinfo_outer = [
   Instruction(opname='MAKE_CELL', opcode=135, arg=0, argval='a', argrepr='a', offset=0, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='MAKE_CELL', opcode=135, arg=1, argval='b', argrepr='b', offset=2, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='RESUME', opcode=151, arg=0, argval=0, argrepr='', offset=4, starts_line=1, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=7, argval=(3, 4), argrepr='(3, 4)', offset=6, starts_line=2, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=5, argval=(3, 4), argrepr='(3, 4)', offset=6, starts_line=2, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=0, argval='a', argrepr='a', offset=8, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=1, argval='b', argrepr='b', offset=10, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='BUILD_TUPLE', opcode=102, arg=2, argval=2, argrepr='', offset=12, starts_line=None, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=3, argval=code_object_f, argrepr=repr(code_object_f), offset=14, starts_line=None, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=1, argval=code_object_f, argrepr=repr(code_object_f), offset=14, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='MAKE_FUNCTION', opcode=132, arg=9, argval=9, argrepr='defaults, closure', offset=16, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='STORE_FAST', opcode=125, arg=2, argval='f', argrepr='f', offset=18, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_GLOBAL', opcode=116, arg=1, argval='print', argrepr='NULL + print', offset=20, starts_line=7, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_DEREF', opcode=137, arg=0, argval='a', argrepr='a', offset=32, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_DEREF', opcode=137, arg=1, argval='b', argrepr='b', offset=34, starts_line=None, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=4, argval='', argrepr="''", offset=36, starts_line=None, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=5, argval=1, argrepr='1', offset=38, starts_line=None, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=2, argval='', argrepr="''", offset=36, starts_line=None, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=3, argval=1, argrepr='1', offset=38, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='BUILD_LIST', opcode=103, arg=0, argval=0, argrepr='', offset=40, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='BUILD_MAP', opcode=105, arg=0, argval=0, argrepr='', offset=42, starts_line=None, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=6, argval='Hello world!', argrepr="'Hello world!'", offset=44, starts_line=None, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=4, argval='Hello world!', argrepr="'Hello world!'", offset=44, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='CALL', opcode=171, arg=7, argval=7, argrepr='', offset=46, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='POP_TOP', opcode=1, arg=None, argval=None, argrepr='', offset=56, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_FAST', opcode=124, arg=2, argval='f', argrepr='f', offset=58, starts_line=8, is_jump_target=False, positions=None),
@@ -1511,13 +1511,13 @@ expected_opinfo_f = [
   Instruction(opname='MAKE_CELL', opcode=135, arg=0, argval='c', argrepr='c', offset=2, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='MAKE_CELL', opcode=135, arg=1, argval='d', argrepr='d', offset=4, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='RESUME', opcode=151, arg=0, argval=0, argrepr='', offset=6, starts_line=2, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=4, argval=(5, 6), argrepr='(5, 6)', offset=8, starts_line=3, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=2, argval=(5, 6), argrepr='(5, 6)', offset=8, starts_line=3, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=3, argval='a', argrepr='a', offset=10, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=4, argval='b', argrepr='b', offset=12, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=0, argval='c', argrepr='c', offset=14, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_CLOSURE', opcode=136, arg=1, argval='d', argrepr='d', offset=16, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='BUILD_TUPLE', opcode=102, arg=4, argval=4, argrepr='', offset=18, starts_line=None, is_jump_target=False, positions=None),
-  Instruction(opname='LOAD_CONST', opcode=100, arg=3, argval=code_object_inner, argrepr=repr(code_object_inner), offset=20, starts_line=None, is_jump_target=False, positions=None),
+  Instruction(opname='LOAD_CONST', opcode=100, arg=1, argval=code_object_inner, argrepr=repr(code_object_inner), offset=20, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='MAKE_FUNCTION', opcode=132, arg=9, argval=9, argrepr='defaults, closure', offset=22, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='STORE_FAST', opcode=125, arg=2, argval='inner', argrepr='inner', offset=24, starts_line=None, is_jump_target=False, positions=None),
   Instruction(opname='LOAD_GLOBAL', opcode=116, arg=1, argval='print', argrepr='NULL + print', offset=26, starts_line=5, is_jump_target=False, positions=None),