modal 0.62.16__py3-none-any.whl → 0.72.11__py3-none-any.whl

test/cli_test.py

@@ -1,659 +0,0 @@
-# Copyright Modal Labs 2022-2023
-import asyncio
-import contextlib
-import json
-import os
-import pytest
-import re
-import subprocess
-import sys
-import tempfile
-import traceback
-from typing import List, Optional
-from unittest import mock
-
-import click
-import click.testing
-import pytest_asyncio
-import toml
-
-from modal import Client
-from modal.cli.entry_point import entrypoint_cli
-from modal_proto import api_pb2
-
-from .supports.skip import skip_windows
-
-dummy_app_file = """
-import modal
-
-import other_module
-
-stub = modal.Stub("my_app")
-
-# Sanity check that the module is imported properly
-import sys
-mod = sys.modules[__name__]
-assert mod.stub == stub
-"""
-
-dummy_other_module_file = "x = 42"
-
-
-@pytest_asyncio.fixture
-async def set_env_client(client):
-    try:
-        Client.set_env_client(client)
-        yield
-    finally:
-        Client.set_env_client(None)
-
-
-def _run(args: List[str], expected_exit_code: int = 0, expected_stderr: Optional[str] = ""):
-    runner = click.testing.CliRunner(mix_stderr=False)
-    with mock.patch.object(sys, "argv", args):
-        res = runner.invoke(entrypoint_cli, args)
-    if res.exit_code != expected_exit_code:
-        print("stdout:", repr(res.stdout))
-        print("stderr:", repr(res.stderr))
-        traceback.print_tb(res.exc_info[2])
-        print(res.exception, file=sys.stderr)
-        assert res.exit_code == expected_exit_code
-    if expected_stderr is not None:
-        assert res.stderr == expected_stderr
-    return res
-
-
-def test_app_deploy_success(servicer, mock_dir, set_env_client):
-    with mock_dir({"myapp.py": dummy_app_file, "other_module.py": dummy_other_module_file}):
-        # Deploy as a script in cwd
-        _run(["deploy", "myapp.py"])
-
-        # Deploy as a module
-        _run(["deploy", "myapp"])
-
-        # Deploy as a script with an absolute path
-        _run(["deploy", os.path.abspath("myapp.py")])
-
-    assert "my_app" in servicer.deployed_apps
-
-
-def test_app_deploy_with_name(servicer, mock_dir, set_env_client):
-    with mock_dir({"myapp.py": dummy_app_file, "other_module.py": dummy_other_module_file}):
-        _run(["deploy", "myapp.py", "--name", "my_app_foo"])
-
-    assert "my_app_foo" in servicer.deployed_apps
-
-
-def test_secret_create(servicer, set_env_client):
-    # fail without any keys
-    _run(["secret", "create", "foo"], 2, None)
-
-    _run(["secret", "create", "foo", "bar=baz"])
-    assert len(servicer.secrets) == 1
-
-    # Creating the same one again should fail
-    _run(["secret", "create", "foo", "bar=baz"], expected_exit_code=1)
-
-    # But it should succeed with --force
-    _run(["secret", "create", "foo", "bar=baz", "--force"])
-
-
-def test_secret_list(servicer, set_env_client):
-    res = _run(["secret", "list"])
-    assert "dummy-secret-0" not in res.stdout
-
-    _run(["secret", "create", "foo", "bar=baz"])
-    _run(["secret", "create", "bar", "baz=buz"])
-    _run(["secret", "create", "eric", "baz=bu 123z=b\n\t\r #(Q)JO5️⃣5️⃣😤WMLE🔧:GWam "])
-
-    res = _run(["secret", "list"])
-    assert "dummy-secret-0" in res.stdout
-    assert "dummy-secret-1" in res.stdout
-    assert "dummy-secret-2" in res.stdout
-    assert "dummy-secret-3" not in res.stdout
-
-
-def test_app_token_new(servicer, set_env_client, server_url_env, modal_config):
-    with modal_config() as config_file_path:
-        _run(["token", "new", "--profile", "_test"])
-        assert "_test" in toml.load(config_file_path)
-
-
-def test_app_setup(servicer, set_env_client, server_url_env, modal_config):
-    with modal_config() as config_file_path:
-        _run(["setup", "--profile", "_test"])
-        assert "_test" in toml.load(config_file_path)
-
-
-def test_run(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    _run(["run", stub_file.as_posix()])
-    _run(["run", stub_file.as_posix() + "::stub"])
-    _run(["run", stub_file.as_posix() + "::foo"])
-    _run(["run", stub_file.as_posix() + "::bar"], expected_exit_code=1, expected_stderr=None)
-    file_with_entrypoint = test_dir / "supports" / "app_run_tests" / "local_entrypoint.py"
-    _run(["run", file_with_entrypoint.as_posix()])
-    _run(["run", file_with_entrypoint.as_posix() + "::main"])
-    _run(["run", file_with_entrypoint.as_posix() + "::stub.main"])
-
-
-def test_run_async(servicer, set_env_client, test_dir):
-    sync_fn = test_dir / "supports" / "app_run_tests" / "local_entrypoint.py"
-    res = _run(["run", sync_fn.as_posix()])
-    assert "called locally" in res.stdout
-
-    async_fn = test_dir / "supports" / "app_run_tests" / "local_entrypoint_async.py"
-    res = _run(["run", async_fn.as_posix()])
-    assert "called locally (async)" in res.stdout
-
-
-def test_run_generator(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "generator.py"
-    result = _run(["run", stub_file.as_posix()], expected_exit_code=1)
-    assert "generator functions" in str(result.exception)
-
-
-def test_help_message_unspecified_function(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "stub_with_multiple_functions.py"
-    result = _run(["run", stub_file.as_posix()], expected_exit_code=2, expected_stderr=None)
-
-    # should suggest available functions on the stub:
-    assert "foo" in result.stderr
-    assert "bar" in result.stderr
-
-    result = _run(
-        ["run", stub_file.as_posix(), "--help"], expected_exit_code=2, expected_stderr=None
-    )  # TODO: help should not return non-zero
-    # help should also available functions on the stub:
-    assert "foo" in result.stderr
-    assert "bar" in result.stderr
-
-
-def test_run_states(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    _run(["run", stub_file.as_posix()])
-    assert servicer.app_state_history["ap-1"] == [
-        api_pb2.APP_STATE_INITIALIZING,
-        api_pb2.APP_STATE_EPHEMERAL,
-        api_pb2.APP_STATE_STOPPED,
-    ]
-
-
-def test_run_detach(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    _run(["run", "--detach", stub_file.as_posix()])
-    assert servicer.app_state_history["ap-1"] == [api_pb2.APP_STATE_INITIALIZING, api_pb2.APP_STATE_DETACHED]
-
-
-def test_run_quiet(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    # Just tests that the command runs without error for now (tests end up defaulting to `show_progress=False` anyway,
-    # without a TTY).
-    _run(["run", "--quiet", stub_file.as_posix()])
-
-
-def test_deploy(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    _run(["deploy", "--name=deployment_name", stub_file.as_posix()])
-    assert servicer.app_state_history["ap-1"] == [api_pb2.APP_STATE_INITIALIZING, api_pb2.APP_STATE_DEPLOYED]
-
-
-def test_run_custom_stub(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "custom_stub.py"
-    res = _run(["run", stub_file.as_posix() + "::stub"], expected_exit_code=1, expected_stderr=None)
-    assert "Could not find" in res.stderr
-    res = _run(["run", stub_file.as_posix() + "::stub.foo"], expected_exit_code=1, expected_stderr=None)
-    assert "Could not find" in res.stderr
-
-    _run(["run", stub_file.as_posix() + "::foo"])
-
-
-def test_run_aiofunc(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "async_stub.py"
-    _run(["run", stub_file.as_posix()])
-    assert len(servicer.client_calls) == 1
-
-
-def test_run_local_entrypoint(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "local_entrypoint.py"
-
-    res = _run(["run", stub_file.as_posix() + "::stub.main"])  # explicit name
-    assert "called locally" in res.stdout
-    assert len(servicer.client_calls) == 2
-
-    res = _run(["run", stub_file.as_posix()])  # only one entry-point, no name needed
-    assert "called locally" in res.stdout
-    assert len(servicer.client_calls) == 4
-
-
-def test_run_local_entrypoint_invalid_with_stub_run(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "local_entrypoint_invalid.py"
-
-    res = _run(["run", stub_file.as_posix()], expected_exit_code=1)
-    assert "app is already running" in str(res.exception.__cause__).lower()
-    assert "unreachable" not in res.stdout
-    assert len(servicer.client_calls) == 0
-
-
-def test_run_parse_args_entrypoint(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "cli_args.py"
-    res = _run(["run", stub_file.as_posix()], expected_exit_code=2, expected_stderr=None)
-    assert "You need to specify a Modal function or local entrypoint to run" in res.stderr
-
-    valid_call_args = [
-        (
-            [
-                "run",
-                f"{stub_file.as_posix()}::stub.dt_arg",
-                "--dt",
-                "2022-10-31",
-            ],
-            "the day is 31",
-        ),
-        (["run", f"{stub_file.as_posix()}::dt_arg", "--dt=2022-10-31"], "the day is 31"),
-        (["run", f"{stub_file.as_posix()}::int_arg", "--i=200"], "200 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::default_arg"], "10 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::unannotated_arg", "--i=2022-10-31"], "'2022-10-31' <class 'str'>"),
-        (["run", f"{stub_file.as_posix()}::unannotated_default_arg"], "10 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::optional_arg", "--i=20"], "20 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::optional_arg"], "None <class 'NoneType'>"),
-        (["run", f"{stub_file.as_posix()}::optional_arg_postponed"], "None <class 'NoneType'>"),
-    ]
-    if sys.version_info >= (3, 10):
-        valid_call_args.extend(
-            [
-                (["run", f"{stub_file.as_posix()}::optional_arg_pep604", "--i=20"], "20 <class 'int'>"),
-                (["run", f"{stub_file.as_posix()}::optional_arg_pep604"], "None <class 'NoneType'>"),
-            ]
-        )
-    for args, expected in valid_call_args:
-        res = _run(args)
-        assert expected in res.stdout
-        assert len(servicer.client_calls) == 0
-
-    if sys.version_info >= (3, 10):
-        res = _run(["run", f"{stub_file.as_posix()}::unparseable_annot", "--i=20"], expected_exit_code=1)
-        assert "Parameter `i` has unparseable annotation: typing.Union[int, str]" in str(res.exception)
-
-    if sys.version_info <= (3, 10):
-        res = _run(["run", f"{stub_file.as_posix()}::optional_arg_pep604"], expected_exit_code=1)
-        assert "Unable to generate command line interface for app entrypoint." in str(res.exception)
-
-
-def test_run_parse_args_function(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "cli_args.py"
-    res = _run(["run", stub_file.as_posix()], expected_exit_code=2, expected_stderr=None)
-    assert "You need to specify a Modal function or local entrypoint to run" in res.stderr
-
-    # HACK: all the tests use the same arg, i.
-    @servicer.function_body
-    def print_type(i):
-        print(repr(i), type(i))
-
-    valid_call_args = [
-        (["run", f"{stub_file.as_posix()}::int_arg_fn", "--i=200"], "200 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::ALifecycle.some_method", "--i=hello"], "'hello' <class 'str'>"),
-        (["run", f"{stub_file.as_posix()}::ALifecycle.some_method_int", "--i=42"], "42 <class 'int'>"),
-        (["run", f"{stub_file.as_posix()}::optional_arg_fn"], "None <class 'NoneType'>"),
-    ]
-    for args, expected in valid_call_args:
-        res = _run(args)
-        assert expected in res.stdout
-
-
-def test_run_user_script_exception(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "raises_error.py"
-    res = _run(["run", stub_file.as_posix()], expected_exit_code=1)
-    assert res.exc_info[1].user_source == str(stub_file.resolve())
-
-
-@pytest.fixture
-def fresh_main_thread_assertion_module(test_dir):
-    modules_to_unload = [n for n in sys.modules.keys() if "main_thread_assertion" in n]
-    assert len(modules_to_unload) <= 1
-    for mod in modules_to_unload:
-        sys.modules.pop(mod)
-    yield test_dir / "supports" / "app_run_tests" / "main_thread_assertion.py"
-
-
-def test_no_user_code_in_synchronicity_run(servicer, set_env_client, test_dir, fresh_main_thread_assertion_module):
-    pytest._did_load_main_thread_assertion = False  # type: ignore
-    _run(["run", fresh_main_thread_assertion_module.as_posix()])
-    assert pytest._did_load_main_thread_assertion  # type: ignore
-    print()
-
-
-def test_no_user_code_in_synchronicity_deploy(servicer, set_env_client, test_dir, fresh_main_thread_assertion_module):
-    pytest._did_load_main_thread_assertion = False  # type: ignore
-    _run(["deploy", "--name", "foo", fresh_main_thread_assertion_module.as_posix()])
-    assert pytest._did_load_main_thread_assertion  # type: ignore
-
-
-def test_serve(servicer, set_env_client, server_url_env, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "webhook.py"
-    _run(["serve", stub_file.as_posix(), "--timeout", "3"], expected_exit_code=0)
-
-
-@pytest.fixture
-def mock_shell_pty():
-    def mock_get_pty_info(shell: bool) -> api_pb2.PTYInfo:
-        rows, cols = (64, 128)
-        return api_pb2.PTYInfo(
-            enabled=True,
-            winsz_rows=rows,
-            winsz_cols=cols,
-            env_term=os.environ.get("TERM"),
-            env_colorterm=os.environ.get("COLORTERM"),
-            env_term_program=os.environ.get("TERM_PROGRAM"),
-            pty_type=api_pb2.PTYInfo.PTY_TYPE_SHELL,
-        )
-
-    captured_out = []
-    fake_stdin = [b"echo foo\n", b"exit\n"]
-
-    async def write_to_fd(fd: int, data: bytes):
-        nonlocal captured_out
-        captured_out.append((fd, data))
-
-    @contextlib.asynccontextmanager
-    async def fake_stream_from_stdin(handle_input, use_raw_terminal=False):
-        async def _write():
-            message_index = 0
-            while True:
-                if message_index == len(fake_stdin):
-                    break
-                data = fake_stdin[message_index]
-                await handle_input(data, message_index)
-                message_index += 1
-
-        write_task = asyncio.create_task(_write())
-        yield
-        write_task.cancel()
-
-    with mock.patch("rich.console.Console.is_terminal", True), mock.patch(
-        "modal._pty.get_pty_info", mock_get_pty_info
-    ), mock.patch("modal.runner.get_pty_info", mock_get_pty_info), mock.patch(
-        "modal._utils.shell_utils.stream_from_stdin", fake_stream_from_stdin
-    ), mock.patch("modal._sandbox_shell.write_to_fd", write_to_fd):
-        yield fake_stdin, captured_out
-
-
-@skip_windows("modal shell is not supported on Windows.")
-def test_shell(servicer, set_env_client, test_dir, mock_shell_pty):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    webhook_stub_file = test_dir / "supports" / "app_run_tests" / "webhook.py"
-    fake_stdin, captured_out = mock_shell_pty
-
-    fake_stdin.clear()
-    fake_stdin.extend([b'echo "Hello World"\n', b"exit\n"])
-
-    # Function is explicitly specified
-    _run(["shell", stub_file.as_posix() + "::foo"])
-
-    shell_prompt = servicer.sandbox_shell_prompt.encode("utf-8")
-
-    # first captured message is the empty message the mock server sends
-    assert captured_out == [(1, shell_prompt), (1, b"Hello World\n")]
-    captured_out.clear()
-
-    # Function is explicitly specified
-    _run(["shell", webhook_stub_file.as_posix() + "::foo"])
-    assert captured_out == [(1, shell_prompt), (1, b"Hello World\n")]
-    captured_out.clear()
-
-    # Function must be inferred
-    _run(["shell", webhook_stub_file.as_posix()])
-    assert captured_out == [(1, shell_prompt), (1, b"Hello World\n")]
-    captured_out.clear()
-
-
-@skip_windows("modal shell is not supported on Windows.")
-def test_shell_cmd(servicer, set_env_client, test_dir, mock_shell_pty):
-    stub_file = test_dir / "supports" / "app_run_tests" / "default_stub.py"
-    _, captured_out = mock_shell_pty
-    _run(["shell", "--cmd", "pwd", stub_file.as_posix() + "::foo"])
-    expected_output = subprocess.run(["pwd"], capture_output=True, check=True).stdout
-    shell_prompt = servicer.sandbox_shell_prompt.encode("utf-8")
-    assert captured_out == [(1, shell_prompt), (1, expected_output)]
-
-
-def test_app_descriptions(servicer, server_url_env, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "prints_desc_stub.py"
-    _run(["run", "--detach", stub_file.as_posix() + "::foo"])
-
-    create_reqs = [s for s in servicer.requests if isinstance(s, api_pb2.AppCreateRequest)]
-    assert len(create_reqs) == 1
-    assert create_reqs[0].app_state == api_pb2.APP_STATE_DETACHED
-    description = create_reqs[0].description
-    assert "prints_desc_stub.py::foo" in description
-    assert "run --detach " not in description
-
-    _run(["serve", "--timeout", "0.0", stub_file.as_posix()])
-    create_reqs = [s for s in servicer.requests if isinstance(s, api_pb2.AppCreateRequest)]
-    assert len(create_reqs) == 2
-    description = create_reqs[1].description
-    assert "prints_desc_stub.py" in description
-    assert "serve" not in description
-    assert "--timeout 0.0" not in description
-
-
-def test_logs(servicer, server_url_env):
-    async def app_done(self, stream):
-        await stream.recv_message()
-        log = api_pb2.TaskLogs(data="hello\n", file_descriptor=api_pb2.FILE_DESCRIPTOR_STDOUT)
-        await stream.send_message(api_pb2.TaskLogsBatch(entry_id="1", items=[log]))
-        await stream.send_message(api_pb2.TaskLogsBatch(app_done=True))
-
-    with servicer.intercept() as ctx:
-        ctx.set_responder("AppGetLogs", app_done)
-        res = _run(["app", "logs", "ap-123"], expected_exit_code=0)
-        assert res.stdout == "hello\n"
-
-
-def test_nfs_get(set_env_client):
-    nfs_name = "my-shared-nfs"
-    _run(["nfs", "create", nfs_name])
-    with tempfile.TemporaryDirectory() as tmpdir:
-        upload_path = os.path.join(tmpdir, "upload.txt")
-        with open(upload_path, "w") as f:
-            f.write("foo bar baz")
-            f.flush()
-        _run(["nfs", "put", nfs_name, upload_path, "test.txt"])
-
-        _run(["nfs", "get", nfs_name, "test.txt", tmpdir])
-        with open(os.path.join(tmpdir, "test.txt"), "r") as f:
-            assert f.read() == "foo bar baz"
-
-
-def test_volume_cli(set_env_client):
-    _run(["volume", "--help"])
-
-
-def test_volume_get(servicer, set_env_client):
-    vol_name = "my-test-vol"
-    _run(["volume", "create", vol_name])
-    file_path = b"test.txt"
-    file_contents = b"foo bar baz"
-    with tempfile.TemporaryDirectory() as tmpdir:
-        upload_path = os.path.join(tmpdir, "upload.txt")
-        with open(upload_path, "wb") as f:
-            f.write(file_contents)
-            f.flush()
-        _run(["volume", "put", vol_name, upload_path, file_path.decode()])
-
-        _run(["volume", "get", vol_name, file_path.decode(), tmpdir])
-        with open(os.path.join(tmpdir, file_path.decode()), "rb") as f:
-            assert f.read() == file_contents
-
-    with tempfile.TemporaryDirectory() as tmpdir2:
-        _run(["volume", "get", vol_name, "**", tmpdir2])
-        with open(os.path.join(tmpdir2, file_path.decode()), "rb") as f:
-            assert f.read() == file_contents
-
-
-def test_volume_put_force(servicer, set_env_client):
-    vol_name = "my-test-vol"
-    _run(["volume", "create", vol_name])
-    file_path = "test.txt"
-    file_contents = b"foo bar baz"
-    with tempfile.TemporaryDirectory() as tmpdir:
-        upload_path = os.path.join(tmpdir, "upload.txt")
-        with open(upload_path, "wb") as f:
-            f.write(file_contents)
-            f.flush()
-
-        # File upload
-        _run(["volume", "put", vol_name, upload_path, file_path])  # Seed the volume
-        with servicer.intercept() as ctx:
-            _run(["volume", "put", vol_name, upload_path, file_path], expected_exit_code=2, expected_stderr=None)
-            assert ctx.pop_request("VolumePutFiles").disallow_overwrite_existing_files
-
-            _run(["volume", "put", vol_name, upload_path, file_path, "--force"])
-            assert not ctx.pop_request("VolumePutFiles").disallow_overwrite_existing_files
-
-        # Dir upload
-        _run(["volume", "put", vol_name, tmpdir])  # Seed the volume
-        with servicer.intercept() as ctx:
-            _run(["volume", "put", vol_name, tmpdir], expected_exit_code=2, expected_stderr=None)
-            assert ctx.pop_request("VolumePutFiles").disallow_overwrite_existing_files
-
-            _run(["volume", "put", vol_name, tmpdir, "--force"])
-            assert not ctx.pop_request("VolumePutFiles").disallow_overwrite_existing_files
-
-
-def test_volume_rm(servicer, set_env_client):
-    vol_name = "my-test-vol"
-    _run(["volume", "create", vol_name])
-    file_path = b"test.txt"
-    file_contents = b"foo bar baz"
-    with tempfile.TemporaryDirectory() as tmpdir:
-        upload_path = os.path.join(tmpdir, "upload.txt")
-        with open(upload_path, "wb") as f:
-            f.write(file_contents)
-            f.flush()
-        _run(["volume", "put", vol_name, upload_path, file_path.decode()])
-
-        _run(["volume", "get", vol_name, file_path.decode(), tmpdir])
-        with open(os.path.join(tmpdir, file_path.decode()), "rb") as f:
-            assert f.read() == file_contents
-
-        _run(["volume", "rm", vol_name, file_path.decode()])
-        _run(["volume", "get", vol_name, file_path.decode()], expected_exit_code=2, expected_stderr=None)
-
-
-@pytest.mark.parametrize("command", [["run"], ["deploy"], ["serve", "--timeout=1"], ["shell"]])
-@pytest.mark.usefixtures("set_env_client", "mock_shell_pty")
-@skip_windows("modal shell is not supported on Windows.")
-def test_environment_flag(test_dir, servicer, command):
-    @servicer.function_body
-    def nothing(
-        arg=None,
-    ):  # hacky - compatible with both argless modal run and interactive mode which always sends an arg...
-        pass
-
-    stub_file = test_dir / "supports" / "app_run_tests" / "app_with_lookups.py"
-    with servicer.intercept() as ctx:
-        ctx.add_response(
-            "MountGetOrCreate",
-            api_pb2.MountGetOrCreateResponse(
-                mount_id="mo-123",
-                handle_metadata=api_pb2.MountHandleMetadata(content_checksum_sha256_hex="abc123"),
-            ),
-            request_filter=lambda req: req.deployment_name.startswith("modal-client-mount")
-            and req.namespace == api_pb2.DEPLOYMENT_NAMESPACE_GLOBAL,
-        )  # built-in client lookup
-        ctx.add_response(
-            "SharedVolumeGetOrCreate",
-            api_pb2.SharedVolumeGetOrCreateResponse(shared_volume_id="sv-123"),
-            request_filter=lambda req: req.deployment_name == "volume_app" and req.environment_name == "staging",
-        )
-        _run(command + ["--env=staging", str(stub_file)])
-
-    app_create: api_pb2.AppCreateRequest = ctx.pop_request("AppCreate")
-    assert app_create.environment_name == "staging"
-
-
-@pytest.mark.parametrize("command", [["run"], ["deploy"], ["serve", "--timeout=1"], ["shell"]])
-@pytest.mark.usefixtures("set_env_client", "mock_shell_pty")
-@skip_windows("modal shell is not supported on Windows.")
-def test_environment_noflag(test_dir, servicer, command, monkeypatch):
-    monkeypatch.setenv("MODAL_ENVIRONMENT", "some_weird_default_env")
-
-    @servicer.function_body
-    def nothing(
-        arg=None,
-    ):  # hacky - compatible with both argless modal run and interactive mode which always sends an arg...
-        pass
-
-    stub_file = test_dir / "supports" / "app_run_tests" / "app_with_lookups.py"
-    with servicer.intercept() as ctx:
-        ctx.add_response(
-            "MountGetOrCreate",
-            api_pb2.MountGetOrCreateResponse(
-                mount_id="mo-123",
-                handle_metadata=api_pb2.MountHandleMetadata(content_checksum_sha256_hex="abc123"),
-            ),
-            request_filter=lambda req: req.deployment_name.startswith("modal-client-mount")
-            and req.namespace == api_pb2.DEPLOYMENT_NAMESPACE_GLOBAL,
-        )  # built-in client lookup
-        ctx.add_response(
-            "SharedVolumeGetOrCreate",
-            api_pb2.SharedVolumeGetOrCreateResponse(shared_volume_id="sv-123"),
-            request_filter=lambda req: req.deployment_name == "volume_app"
-            and req.environment_name == "some_weird_default_env",
-        )
-        _run(command + [str(stub_file)])
-
-    app_create: api_pb2.AppCreateRequest = ctx.pop_request("AppCreate")
-    assert app_create.environment_name == "some_weird_default_env"
-
-
-def test_cls(servicer, set_env_client, test_dir):
-    stub_file = test_dir / "supports" / "app_run_tests" / "cls.py"
-
-    _run(["run", stub_file.as_posix(), "--x", "42", "--y", "1000"])
-    _run(["run", f"{stub_file.as_posix()}::AParametrized.some_method", "--x", "42", "--y", "1000"])
-
-
-def test_profile_list(servicer, server_url_env, modal_config):
-    config = """
-    [test-profile]
-    token_id = "ak-abc"
-    token_secret = "as-xyz"
-
-    [other-profile]
-    token_id = "ak-123"
-    token_secret = "as-789"
-    active = true
-    """
-
-    with modal_config(config):
-        res = _run(["profile", "list"])
-        table_rows = res.stdout.split("\n")
-        assert re.search("Profile .+ Workspace", table_rows[1])
-        assert re.search("test-profile .+ test-username", table_rows[3])
-        assert re.search("other-profile .+ test-username", table_rows[4])
-
-        res = _run(["profile", "list", "--json"])
-        json_data = json.loads(res.stdout)
-        assert json_data[0]["name"] == "test-profile"
-        assert json_data[0]["workspace"] == "test-username"
-        assert json_data[1]["name"] == "other-profile"
-        assert json_data[1]["workspace"] == "test-username"
-
-        orig_env_token_id = os.environ.get("MODAL_TOKEN_ID")
-        orig_env_token_secret = os.environ.get("MODAL_TOKEN_SECRET")
-        os.environ["MODAL_TOKEN_ID"] = "ak-abc"
-        os.environ["MODAL_TOKEN_SECRET"] = "as-xyz"
-        try:
-            res = _run(["profile", "list"])
-            assert "Using test-username workspace based on environment variables" in res.stdout
-        finally:
-            if orig_env_token_id:
-                os.environ["MODAL_TOKEN_ID"] = orig_env_token_id
-            else:
-                del os.environ["MODAL_TOKEN_ID"]
-            if orig_env_token_secret:
-                os.environ["MODAL_TOKEN_SECRET"] = orig_env_token_secret
-            else:
-                del os.environ["MODAL_TOKEN_SECRET"]