fix: add explicit UTF-8 encoding to all file read/write operations

On Windows, Path.read_text() and Path.write_text() use the system
default encoding (e.g. CP1251) instead of UTF-8. This causes
UnicodeDecodeError when config files contain non-ASCII characters
such as Cyrillic text in commitizen customization options.

Fixes #1636

Author: veeceey

commitizen/commands/changelog.py

@@ -183,8 +183,8 @@ def _export_template(self, dist: str) -> None:
         if filename is None:
             raise NotAllowed("Template filename is not set")
 
-        text = Path(filename).read_text()
-        Path(dist).write_text(text)
+        text = Path(filename).read_text(encoding="utf-8")
+        Path(dist).write_text(text, encoding="utf-8")
 
     def __call__(self) -> None:
         commit_parser = self.cz.commit_parser

commitizen/project_info.py

@@ -14,7 +14,7 @@ def get_default_version_provider() -> Literal[
 ]:
     pyproject_path = Path("pyproject.toml")
     if pyproject_path.is_file():
-        if "[tool.poetry]" in pyproject_path.read_text():
+        if "[tool.poetry]" in pyproject_path.read_text(encoding="utf-8"):
             return "poetry"
         if Path("uv.lock").is_file():
             return "uv"

commitizen/providers/base_provider.py

@@ -58,13 +58,15 @@ class JsonProvider(FileProvider):
     indent: ClassVar[int] = 2
 
     def get_version(self) -> str:
-        document = json.loads(self.file.read_text())
+        document = json.loads(self.file.read_text(encoding="utf-8"))
         return self.get(document)
 
     def set_version(self, version: str) -> None:
-        document = json.loads(self.file.read_text())
+        document = json.loads(self.file.read_text(encoding="utf-8"))
         self.set(document, version)
-        self.file.write_text(json.dumps(document, indent=self.indent) + "\n")
+        self.file.write_text(
+            json.dumps(document, indent=self.indent) + "\n", encoding="utf-8"
+        )
 
     def get(self, document: Mapping[str, str]) -> str:
         return document["version"]
@@ -79,13 +81,13 @@ class TomlProvider(FileProvider):
     """
 
     def get_version(self) -> str:
-        document = tomlkit.parse(self.file.read_text())
+        document = tomlkit.parse(self.file.read_text(encoding="utf-8"))
         return self.get(document)
 
     def set_version(self, version: str) -> None:
-        document = tomlkit.parse(self.file.read_text())
+        document = tomlkit.parse(self.file.read_text(encoding="utf-8"))
         self.set(document, version)
-        self.file.write_text(tomlkit.dumps(document))
+        self.file.write_text(tomlkit.dumps(document), encoding="utf-8")
 
     def get(self, document: tomlkit.TOMLDocument) -> str:
         return document["project"]["version"]  # type: ignore[index,return-value]

commitizen/providers/cargo_provider.py

@@ -43,8 +43,8 @@ def set_version(self, version: str) -> None:
             self.set_lock_version(version)
 
     def set_lock_version(self, version: str) -> None:
-        cargo_toml_content = parse(self.file.read_text())
-        cargo_lock_content = parse(self.lock_file.read_text())
+        cargo_toml_content = parse(self.file.read_text(encoding="utf-8"))
+        cargo_lock_content = parse(self.lock_file.read_text(encoding="utf-8"))
         packages = cargo_lock_content["package"]
 
         if TYPE_CHECKING:
@@ -75,7 +75,9 @@ def set_lock_version(self, version: str) -> None:
                         continue
 
                     cargo_file = Path(path) / "Cargo.toml"
-                    package_content = parse(cargo_file.read_text()).get("package", {})
+                    package_content = parse(
+                        cargo_file.read_text(encoding="utf-8")
+                    ).get("package", {})
                     if TYPE_CHECKING:
                         assert isinstance(package_content, dict)
                     try:
@@ -92,7 +94,7 @@ def set_lock_version(self, version: str) -> None:
                 if package["name"] in members_inheriting:
                     cargo_lock_content["package"][i]["version"] = version  # type: ignore[index]
 
-        self.lock_file.write_text(dumps(cargo_lock_content))
+        self.lock_file.write_text(dumps(cargo_lock_content), encoding="utf-8")
 
 
 def _try_get_workspace(document: TOMLDocument) -> dict:

commitizen/providers/npm_provider.py

@@ -36,29 +36,37 @@ def get_version(self) -> str:
         """
         Get the current version from package.json
         """
-        package_document = json.loads(self.package_file.read_text())
+        package_document = json.loads(
+            self.package_file.read_text(encoding="utf-8")
+        )
         return self.get_package_version(package_document)
 
     def set_version(self, version: str) -> None:
         package_document = self.set_package_version(
-            json.loads(self.package_file.read_text()), version
+            json.loads(self.package_file.read_text(encoding="utf-8")), version
         )
         self.package_file.write_text(
-            json.dumps(package_document, indent=self.indent) + "\n"
+            json.dumps(package_document, indent=self.indent) + "\n",
+            encoding="utf-8",
         )
         if self.lock_file.is_file():
             lock_document = self.set_lock_version(
-                json.loads(self.lock_file.read_text()), version
+                json.loads(self.lock_file.read_text(encoding="utf-8")), version
             )
             self.lock_file.write_text(
-                json.dumps(lock_document, indent=self.indent) + "\n"
+                json.dumps(lock_document, indent=self.indent) + "\n",
+                encoding="utf-8",
             )
         if self.shrinkwrap_file.is_file():
             shrinkwrap_document = self.set_shrinkwrap_version(
-                json.loads(self.shrinkwrap_file.read_text()), version
+                json.loads(
+                    self.shrinkwrap_file.read_text(encoding="utf-8")
+                ),
+                version,
             )
             self.shrinkwrap_file.write_text(
-                json.dumps(shrinkwrap_document, indent=self.indent) + "\n"
+                json.dumps(shrinkwrap_document, indent=self.indent) + "\n",
+                encoding="utf-8",
             )
 
     def get_package_version(self, document: Mapping[str, str]) -> str:

commitizen/providers/uv_provider.py

@@ -26,15 +26,17 @@ def set_version(self, version: str) -> None:
         self.set_lock_version(version)
 
     def set_lock_version(self, version: str) -> None:
-        pyproject_toml_content = tomlkit.parse(self.file.read_text())
+        pyproject_toml_content = tomlkit.parse(
+            self.file.read_text(encoding="utf-8")
+        )
         project_name = pyproject_toml_content["project"]["name"]  # type: ignore[index]
         normalized_project_name = canonicalize_name(str(project_name))
 
-        document = tomlkit.parse(self.lock_file.read_text())
+        document = tomlkit.parse(self.lock_file.read_text(encoding="utf-8"))
 
         packages: tomlkit.items.AoT = document["package"]  # type: ignore[assignment]
         for i, package in enumerate(packages):
             if package["name"] == normalized_project_name:
                 document["package"][i]["version"] = version  # type: ignore[index]
                 break
-        self.lock_file.write_text(tomlkit.dumps(document))
+        self.lock_file.write_text(tomlkit.dumps(document), encoding="utf-8")