From c84833bab1d81dfdf9d89157b2ce41ae085f3781 Mon Sep 17 00:00:00 2001
From: Batuhan Taskaya <batuhanosmantaskaya@gmail.com>
Date: Mon, 1 Feb 2021 16:56:53 +0300
Subject: [PATCH] params: support unicode characters (#5363)

* params: support unicode characters

* open remove() calls with utf-8
---
 dvc/dvcfile.py            |  6 +++---
 tests/unit/test_params.py | 12 ++++++++++++
 2 files changed, 15 insertions(+), 3 deletions(-)

diff --git a/dvc/dvcfile.py b/dvc/dvcfile.py
index 5092668ab..459f82510 100644
--- a/dvc/dvcfile.py
+++ b/dvc/dvcfile.py
@@ -131,7 +131,7 @@ class FileMixin:
             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 @@ class PipelineFile(FileMixin):
         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 @@ class Lockfile(FileMixin):
         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)
 
diff --git a/tests/unit/test_params.py b/tests/unit/test_params.py
index c7ba4fb59..9b777caf1 100644
--- a/tests/unit/test_params.py
+++ b/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()