Fix EncodedFile.writelines

This is implemented by the underlying stream already, which additionally
checks if the stream is not closed, and calls `write` per line.

Ref/via: #6558 (comment)

Author: blueyed

src/_pytest/capture.py

@@ -426,10 +426,6 @@ def write(self, obj):
             )
         return self.buffer.write(obj)
 
-    def writelines(self, linelist):
-        data = "".join(linelist)
-        self.write(data)
-
     @property
     def name(self):
         """Ensure that file.name is a string."""

testing/test_capture.py

@@ -1497,3 +1497,15 @@ def test_fails():
 def test_stderr_write_returns_len(capsys):
     """Write on Encoded files, namely captured stderr, should return number of characters written."""
     assert sys.stderr.write("Foo") == 3
+
+
+def test_encodedfile_writelines(tmpfile):
+    ef = capture.EncodedFile(tmpfile, None)
+    with pytest.raises(TypeError):
+        ef.writelines(["line1", "line2"])
+    assert ef.writelines([b"line1", b"line2"]) is None
+    tmpfile.seek(0)
+    assert tmpfile.read() == b"line1line2"
+    tmpfile.close()
+    with pytest.raises(ValueError):
+        tmpfile.read()