Import the pathlib module in main.py instead of the individual Path class

[gioannidis]

First Check

• I added a very descriptive title here.
• I used the GitHub search to find a similar question and didn't find it.
• I searched the documentation, with the integrated search.
• I already searched in Google "How to X" and didn't find any information.
• I already read and followed all the tutorials in the docs and didn't find an answer.

Example Code

#!/usr/bin/env python3
# test_typer_pyfakefs.py
from __future__ import annotations

import pathlib
from typing import Annotated

import pytest
import typer
from typer import testing

app: typer.Typer = typer.Typer()


@app.command()
def echo(path: Annotated[pathlib.Path, typer.Argument()]) -> None:
    print(path)


@pytest.mark.small
@pytest.mark.parametrize(
    "should_use_fakefs",
    [True, False],
    ids=["FakeFs", "RealFs"],
)
def test_with_pyfakefs(
    request: pytest.FixtureRequest,
    should_use_fakefs: bool,
) -> None:
    if should_use_fakefs:
        _ = request.getfixturevalue("fs")

    runner: testing.CliRunner = testing.CliRunner()
    result: testing.Result = runner.invoke(app=app, args="/path/to/test.txt")

    assert "test" in result.stdout

Description

This discussion follows up to #1954 and is opened per the contribution guidelines. It is addressed to the development/contributing team, proposing a pull request.

Goal: run pytest with both real filesystems and pyfakefs for in-memory implementation. This uses stringified annotations with __future__.annotations, as discussed in #1954, to force eager annotation evaluation.

Problem: an exception is raised from get_click_type, namely:

RuntimeError: Type not yet supported: <class 'pyfakefs.fake_pathlib.FakePath'>

Root cause: an upstream bug in pyfakefs, which has been already reported in issue #1334. This is not expected to be fixed in the short-term due to the scale of involved refactoring.

Proposed fix: refactor typer/main.py to import the pathlib module itself, instead of directly importing Path. More concretely:

diff --git a/typer/main.py b/typer/main.py
index 700648d..114ca85 100644
--- a/typer/main.py
+++ b/typer/main.py
@@ -1,5 +1,6 @@
 import inspect
 import os
+import pathlib
 import platform
 import shutil
 import subprocess
@@ -9,7 +10,6 @@ from collections.abc import Callable, Sequence
 from datetime import datetime
 from enum import Enum
 from functools import update_wrapper
-from pathlib import Path
 from traceback import FrameSummary, StackSummary
 from types import TracebackType
 from typing import Annotated, Any

... and also update the call sites: Path -> pathlib.Path.

The proposed fix in module_import.patch.

Test cases: all tests pass. Windows tests have been skipped.

$ bash scripts/test.sh
...
Results:
    1380 passed
       2 xfailed
      21 skipped

Key takeaways:

• ⚙️ Refactors imports by importing the entire module: from pathlib import Path -> import pathlib
• ✔️ Is a refactor with no behavioral change
• 👍 Follows best practices, recommended by the Google and the Khan Academy style guides
• ➰ Allows working around the pyfakefs issue #1334

Operating System

Linux

Operating System Details

• Linux-7.1.13+deb14-amd64-x86_64-with-glibc2.43
• Python 3.14.6 (main, Jun 23 2026, 15:18:23) [Clang 22.1.3 ]
• pyfakefs 6.2.0
• pytest 9.1.1

Project Version

0.27.2

Python Version

3.14.6

Additional Context

No response

[lunar-me]

Why? typer/main.py already imports Path from pathlib.

[gioannidis]

Thanks for your comment. As suggested by the title, this is a proposed refactoring with no behavioral change; not a feature or a bug fix. Yes, typer/main.py already imports Path from pathlib, correct.

The proposed change is to import the module (import pathlib) rather than the class (from pathlib import Path). Call sites would be updated from Path to pathlib.Path.

Rationale: apart from adhering to best practices, it mitigates an upstream bug in pyfakefs. This bug emerges when testing CLI applications with in-memory filesystems.