gh-118908: Limit exposed globals from internal imports and definitions on new REPL startup

Lib/_pyrepl/main.py

@@ -0,0 +1,45 @@
+import os
+import sys
+
+IS_SUPPORTED_PYREPL_PLATFORM = sys.platform != "win32"
+IS_USING_PYREPL = False
+
+
+def interactive_console():
+    global IS_USING_PYREPL
+    if not IS_SUPPORTED_PYREPL_PLATFORM:
+        return sys._baserepl()
+
+    startup_path = os.getenv("PYTHONSTARTUP")
+    if startup_path:
+        import tokenize
+        with tokenize.open(startup_path) as f:
+            startup_code = compile(f.read(), startup_path, "exec")
+            exec(startup_code)
+
+    # set sys.{ps1,ps2} just before invoking the interactive interpreter. This
+    # mimics what CPython does in pythonrun.c
+    if not hasattr(sys, "ps1"):
+        sys.ps1 = ">>> "
+    if not hasattr(sys, "ps2"):
+        sys.ps2 = "... "
+
+    run_interactive = None
+    try:
+        import errno
+        if not os.isatty(sys.stdin.fileno()):
+            raise OSError(errno.ENOTTY, "tty required", "stdin")
+        from .simple_interact import check
+        if err := check():
+            raise RuntimeError(err)
+        from .simple_interact import run_multiline_interactive_console
+        run_interactive = run_multiline_interactive_console
+    except Exception as e:
+        from .trace import trace
+        msg = f"warning: can't use pyrepl: {e}"
+        trace(msg)
+        print(msg, file=sys.stderr)
+    if run_interactive is None:
+        return sys._baserepl()
+    IS_USING_PYREPL = True
+    return run_interactive()

Lib/_pyrepl/simple_interact.py

@@ -27,13 +27,20 @@
 
 import _sitebuiltins
 import linecache
+import builtins
 import sys
 import code
 from types import ModuleType
 
 from .console import InteractiveColoredConsole
 from .readline import _get_reader, multiline_input
 
+TYPE_CHECKING = False
+
+if TYPE_CHECKING:
+    from typing import Any
+
+
 _error: tuple[type[Exception], ...] | type[Exception]
 try:
     from .unix_console import _error
@@ -73,20 +80,28 @@ def _clear_screen():
     "clear": _clear_screen,
 }
 
+DEFAULT_NAMESPACE: dict[str, Any] = {
+    '__name__': '__main__',
+    '__doc__': None,
+    '__package__': None,
+    '__loader__': None,
+    '__spec__': None,
+    '__annotations__': {},
+    '__builtins__': builtins,
+}
 
 def run_multiline_interactive_console(
     mainmodule: ModuleType | None = None,
     future_flags: int = 0,
     console: code.InteractiveConsole | None = None,
 ) -> None:
-    import __main__
     from .readline import _setup
     _setup()
 
-    mainmodule = mainmodule or __main__
+    namespace = mainmodule.__dict__ if mainmodule else DEFAULT_NAMESPACE
     if console is None:
         console = InteractiveColoredConsole(
-            mainmodule.__dict__, filename="<stdin>"
+            namespace, filename="<stdin>"
         )
     if future_flags:
         console.compile.compiler.flags |= future_flags

Lib/site.py

@@ -525,14 +525,10 @@ def register_readline():
             pass
 
         def write_history():
-            try:
-                # _pyrepl.__main__ is executed as the __main__ module
-                from __main__ import CAN_USE_PYREPL
-            except ImportError:
-                CAN_USE_PYREPL = False
+            from _pyrepl.main import IS_USING_PYREPL
 
             try:
-                if os.getenv("PYTHON_BASIC_REPL") or not CAN_USE_PYREPL:
+                if os.getenv("PYTHON_BASIC_REPL") or not IS_USING_PYREPL:
                     readline.write_history_file(history)
                 else:
                     _pyrepl.readline.write_history_file(history)

Lib/test/test_pyrepl/test_pyrepl.py

@@ -1,9 +1,14 @@
-import itertools
+import importlib
 import io
+import itertools
 import os
 import rlcompleter
+import select
+import subprocess
+import sys
 from unittest import TestCase
 from unittest.mock import patch
+from test.support import force_not_colorized
 
 from .support import (
     FakeConsole,
@@ -828,3 +833,56 @@ def test_bracketed_paste_single_line(self):
         reader = self.prepare_reader(events)
         output = multiline_input(reader)
         self.assertEqual(output, input_code)
+
+
+class TestMain(TestCase):
+    @force_not_colorized
+    def test_exposed_globals_in_repl(self):
+        expected_output = (
+            '["__annotations__", "__builtins__", "__doc__", "__loader__", '
+            '"__name__", "__package__", "__spec__"]'
+        )
+        output, exit_code = self.run_repl(["sorted(dir())", "exit"])
+        self.assertEqual(exit_code, 0)
+        output = output.replace("\'", '"')
+        self.assertIn(expected_output, output)
+
+    def test_dumb_terminal_exits_cleanly(self):
+        env = os.environ.copy()
+        env.update({"TERM": "dumb"})
+        output, exit_code = self.run_repl("exit()\n", env=env)
+        self.assertEqual(exit_code, 0)
+        self.assertIn("warning: can\'t use pyrepl", output)
+        self.assertNotIn("Exception", output)
+        self.assertNotIn("Traceback", output)
+
+    def run_repl(self, repl_input: str | list[str], env: dict | None = None) -> tuple[str, int]:
+        try:
+            import pty
+        except ImportError:
+            self.skipTest("pty module not available")
+        master_fd, slave_fd = pty.openpty()
+        process = subprocess.Popen(
+            [sys.executable, "-i", "-u"],
+            stdin=slave_fd,
+            stdout=slave_fd,
+            stderr=slave_fd,
+            text=True,
+            close_fds=True,
+            env=env if env else os.environ,
+       )
+        if isinstance(repl_input, list):
+            repl_input = "\n".join(repl_input) + "\n"
+        os.write(master_fd, repl_input.encode("utf-8"))
+
+        output = []
+        while select.select([master_fd], [], [], 0.5)[0]:
+            data = os.read(master_fd, 1024).decode("utf-8")
+            if not data:
+                break
+            output.append(data)
+
+        os.close(master_fd)
+        os.close(slave_fd)
+        exit_code = process.wait()
+        return "\n".join(output), exit_code

Lib/test/test_repl.py

@@ -1,9 +1,9 @@
 """Test the interactive interpreter."""
 
-import sys
 import os
-import unittest
 import subprocess
+import sys
+import unittest
 from textwrap import dedent
 from test import support
 from test.support import cpython_only, has_subprocess_support, SuppressCrashReport
@@ -199,7 +199,6 @@ def test_asyncio_repl_is_ok(self):
         assert_python_ok("-m", "asyncio")
 
 
-
 class TestInteractiveModeSyntaxErrors(unittest.TestCase):
 
     def test_interactive_syntax_error_correct_line(self):

Misc/NEWS.d/next/Library/2024-05-25-10-40-38.gh-issue-118908.XcZiq4.rst

@@ -0,0 +1,2 @@
+Limit exposed globals from internal imports and definitions on new REPL
+startup