fix: read UTF-8 from standard input on all systems (#515)

* Use Utf-8 when reading stdin

* Document the use of Utf-8 encoding

* Update changelog

* Fix tests using stdin

* Reformat conftest.py

* Update docs/users/changelog.md

* Update docs/users/style.md

* Update test_cli.py

* Update test_config_file.py

* Trigger pre-commit.ci

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: Taneli Hukkinen <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: 3 people

docs/users/changelog.md

@@ -3,6 +3,12 @@
 This log documents all Python API or CLI breaking backwards incompatible changes.
 Note that there is currently no guarantee for a stable Markdown formatting style across versions.
 
+## **unreleased**
+
+- Fixed
+  - Read UTF-8 from standard input on all systems.
+    Thank you, [Christopher Prohm](https://github.com/chmp), for the PR.
+
 ## 0.7.22
 
 - Performance

src/mdformat/_cli.py

@@ -124,7 +124,7 @@ def run(cli_args: Sequence[str], cache_toml: bool = True) -> int:  # noqa: C901
             original_str = path.read_bytes().decode()
         else:
             path_str = "-"
-            original_str = sys.stdin.read()
+            original_str = sys.stdin.buffer.read().decode()
 
         # Lazy import to improve module import time
         from mdformat.renderer import LOGGER as RENDERER_LOGGER

tests/conftest.py

@@ -0,0 +1,16 @@
+import io
+import sys
+
+import pytest
+
+
+@pytest.fixture
+def patch_stdin(monkeypatch):
+    """Fixture to patch sys.stdin for the running test."""
+
+    def set_stdin(text):
+        buffer = io.BytesIO(text.encode())
+        fobj = io.TextIOWrapper(buffer)
+        monkeypatch.setattr(sys, "stdin", fobj)
+
+    return set_stdin

tests/test_cli.py

@@ -1,4 +1,3 @@
-from io import StringIO
 import os
 import sys
 from unittest.mock import patch
@@ -143,8 +142,8 @@ def test_formatter_plugin(tmp_path, monkeypatch):
     assert file_path.read_text() == "```lang\ndummy\n```\n"
 
 
-def test_dash_stdin(capfd, monkeypatch):
-    monkeypatch.setattr(sys, "stdin", StringIO(UNFORMATTED_MARKDOWN))
+def test_dash_stdin(capfd, patch_stdin):
+    patch_stdin(UNFORMATTED_MARKDOWN)
     assert run(("-",)) == 0
     captured = capfd.readouterr()
     assert captured.out == FORMATTED_MARKDOWN
@@ -298,8 +297,8 @@ def test_eol__keep_crlf(tmp_path):
     assert file_path.read_bytes() == b"Oi\r\n"
 
 
-def test_eol__crlf_stdin(capfd, monkeypatch):
-    monkeypatch.setattr(sys, "stdin", StringIO("Oi\n"))
+def test_eol__crlf_stdin(capfd, patch_stdin):
+    patch_stdin("Oi\n")
     assert run(["-", "--end-of-line=crlf"]) == 0
     captured = capfd.readouterr()
     assert captured.out == "Oi\r\n"

tests/test_config_file.py

@@ -1,4 +1,3 @@
-from io import StringIO
 import sys
 from unittest import mock
 
@@ -90,11 +89,11 @@ def test_invalid_conf_value(bad_conf, conf_key, tmp_path, capsys):
     assert f"Invalid '{conf_key}' value" in captured.err
 
 
-def test_conf_with_stdin(tmp_path, capfd, monkeypatch):
+def test_conf_with_stdin(tmp_path, capfd, patch_stdin):
     config_path = tmp_path / ".mdformat.toml"
     config_path.write_text("number = true")
 
-    monkeypatch.setattr(sys, "stdin", StringIO("1. one\n1. two\n1. three"))
+    patch_stdin("1. one\n1. two\n1. three")
 
     with mock.patch("mdformat._cli.Path.cwd", return_value=tmp_path):
         assert run(("-",), cache_toml=False) == 0