Explicit encoding

Changelog.rst

@@ -6,6 +6,8 @@ Next release
 
 Default development branch is now called `main`.
 
+Always use `utf-8` encoding explicitly
+
 6.3.2 (2021-04-19)
 ------------------
 

pyproject.toml

@@ -37,6 +37,7 @@ pytest = "^6.2.1"
 pytest-cov = "^2.8.1"
 pytest-mock = "^2.0.0"
 safety = "^1.9.0"
+semgrep = "^0.52.0"
 
 [tool.poetry.scripts]
 tbump = "tbump.main:main"

semgrep.yml

@@ -0,0 +1,13 @@
+rules:
+  - id: python-no-implicit-encoding-when-reading
+    languages:
+      - python
+    message: Specify encoding when using read_text()
+    pattern: $OBJECT.read_text()
+    severity: ERROR
+  - id: python-no-implicit-encoding-when-writing
+    languages:
+      - python
+    message: Specify encoding when using write_text()
+    pattern: $OBJECT.write_text($X)
+    severity: ERROR

tasks.py

@@ -39,6 +39,12 @@ def mypy(c, machine_readable=False):
     c.run(cmd)
 
 
+@task
+def semgrep(c):
+    print("Running semgrep")
+    c.run("semgrep -c semgrep.yml")
+
+
 @task
 def test(c):
     print("Running pytest")
@@ -47,6 +53,7 @@ def test(c):
 
 @task(
     pre=[
+        call(semgrep),
         call(black, check=True),
         call(isort, check=True),
         call(flake8),

tbump/config.py

@@ -191,12 +191,12 @@ def get_config_file(project_path: Path) -> ConfigFile:
 def _get_config_file(project_path: Path) -> ConfigFile:
     toml_path = project_path / "tbump.toml"
     if toml_path.exists():
-        doc = tomlkit.loads(toml_path.read_text())
+        doc = tomlkit.loads(toml_path.read_text(encoding="utf-8"))
         return TbumpTomlConfig(toml_path, doc)
 
     pyproject_path = project_path / "pyproject.toml"
     if pyproject_path.exists():
-        doc = tomlkit.loads(pyproject_path.read_text())
+        doc = tomlkit.loads(pyproject_path.read_text(encoding="utf-8"))
         try:
             doc["tool"]["tbump"]
         except KeyError:

tbump/file_bumper.py

@@ -216,7 +216,7 @@ def compute_patches_for_change_request(
         for file_path_str in glob.glob(str(file_path_glob), recursive=True):
             file_path = Path(file_path_str)
             expanded_src = file_path.relative_to(self.working_path)
-            old_lines = file_path.read_text().splitlines(keepends=False)
+            old_lines = file_path.read_text(encoding="utf-8").splitlines(keepends=False)
 
             for i, old_line in enumerate(old_lines):
                 if should_replace(old_line, old_string, search):

tbump/test/conftest.py

@@ -36,7 +36,7 @@ def restore_cwd() -> Iterator[None]:
 
 
 def file_contains(path: Path, text: str) -> bool:
-    for line in path.read_text().splitlines():
+    for line in path.read_text(encoding="utf-8").splitlines():
         if text in line:
             return True
     return False

tbump/test/data/after.py

@@ -7,7 +7,7 @@
 
 
 def main() -> None:
-    Path("after-hook.stamp").write_text("")
+    Path("after-hook.stamp").write_text("", encoding="utf-8")
 
 
 if __name__ == "__main__":

tbump/test/data/before.py

@@ -9,7 +9,7 @@
 
 def main() -> None:
     current, new = sys.argv[1:]
-    Path("before-hook.stamp").write_text(current + " -> " + new)
+    Path("before-hook.stamp").write_text(current + " -> " + new, encoding="utf-8")
 
 
 if __name__ == "__main__":

tbump/test/test_config.py

@@ -57,7 +57,7 @@ def test_uses_pyproject_if_tbump_toml_is_missing(
     to_write = tomlkit.dumps(pyproject_config)
 
     pyproject_toml = tmp_path / "pyproject.toml"
-    pyproject_toml.write_text(to_write)
+    pyproject_toml.write_text(to_write, encoding="utf-8")
 
     actual_file = tbump.config.get_config_file(tmp_path)
     assert actual_file.get_config() == expected_file.get_config()
@@ -71,7 +71,7 @@ def test_complain_if_pyproject_does_not_contain_tbump_config(tmp_path: Path) ->
     profile = "black"
     """
     )
-    pyproject_toml.write_text(to_write)
+    pyproject_toml.write_text(to_write, encoding="utf-8")
 
     with pytest.raises(tbump.config.ConfigNotFound):
         tbump.config.get_config_file(tmp_path)
@@ -100,7 +100,7 @@ def test_validate_schema_in_pyrpoject_toml(tmp_path: Path) -> None:
           '''
        """
     )
-    pyproject_toml.write_text(to_write)
+    pyproject_toml.write_text(to_write, encoding="utf-8")
 
     with pytest.raises(tbump.config.InvalidConfig) as e:
         tbump.config.get_config_file(tmp_path)