params: support unicode characters (#5363)

* params: support unicode characters

* open remove() calls with utf-8

dvc/dvcfile.py

@@ -131,7 +131,7 @@ def _load(self):
             raise StageFileIsNotDvcFileError(self.path)
         if self._is_git_ignored():
             raise FileIsGitIgnored(self.path)
-        with self.repo.tree.open(self.path) as fd:
+        with self.repo.tree.open(self.path, encoding="utf-8") as fd:
             stage_text = fd.read()
         d = parse_yaml(stage_text, self.path)
         return self.validate(d, self.relpath), stage_text
@@ -276,7 +276,7 @@ def remove_stage(self, stage):
         if not self.exists():
             return
 
-        with open(self.path, "r") as f:
+        with open(self.path, "r", encoding="utf-8") as f:
             d = parse_yaml_for_update(f.read(), self.path)
 
         self.validate(d, self.path)
@@ -380,7 +380,7 @@ def remove_stage(self, stage):
         if not self.exists():
             return
 
-        with open(self.path) as f:
+        with open(self.path, encoding="utf-8") as f:
             d = parse_yaml_for_update(f.read(), self.path)
         self.validate(d, self.path)
 

tests/unit/test_params.py

@@ -20,3 +20,15 @@ def test_params_order(tmp_dir, dvc, dummy_stage):
 
     # params are sorted during dumping, therefore p1 is first
     assert list(dvc.params.show()[""]) == ["p1.yaml", p2_path, "params.yaml"]
+
+
+def test_repro_unicode(tmp_dir, dvc, dummy_stage):
+    tmp_dir.gen({"settings.json": '{"Ω_value": 1}'})
+    dummy_stage(params=[{"settings.json": ["Ω_value"]}])
+    (stage,) = dvc.reproduce(dry=True)
+    stage.cmd = "foo"
+    stage.dvcfile._lockfile.dump(stage)
+
+    dvc.remove(stage.name)
+    assert not (tmp_dir / "dvc.yaml").exists()
+    assert not (tmp_dir / "dvc.lock").exists()