fix https://stackoverflow.com/a/45281922/13079050.15.21

8 files changed, 43 insertions, 8 deletions

diff --git a/CHANGES b/CHANGES
index d689dae..47c2a4a 100644
--- a/CHANGES
+++ b/CHANGES
@@ -1,3 +1,6 @@
+[0, 15, 21]: 2017-07-25
+  - fix for writing unicode in new API, https://stackoverflow.com/a/45281922/1307905
+
 [0, 15, 20]: 2017-07-23
   - wheels for windows including C extensions
 
diff --git a/README.rst b/README.rst
index 652f826..a2afacf 100644
--- a/README.rst
+++ b/README.rst
@@ -35,6 +35,9 @@ ChangeLog
 
 .. should insert NEXT: at the beginning of line for next key
 
+0.15.21 (2017-07-25):
+  - fix for writing unicode in new API, https://stackoverflow.com/a/45281922/1307905
+
 0.15.20 (2017-07-23):
   - wheels for windows including C extensions
 
diff --git a/__init__.py b/__init__.py
index 372a1be..ddd115d 100644
--- a/__init__.py
+++ b/__init__.py
@@ -7,8 +7,8 @@ if False:  # MYPY
 
 _package_data = dict(
     full_package_name='ruamel.yaml',
-    version_info=(0, 15, 20),
-    __version__='0.15.20',
+    version_info=(0, 15, 21),
+    __version__='0.15.21',
     author='Anthon van der Neut',
     author_email='a.van.der.neut@ruamel.eu',
     description='ruamel.yaml is a YAML parser/emitter that supports roundtrip preservation of comments, seq/map flow style, and map key order',  # NOQA
diff --git a/_doc/conf.py b/_doc/conf.py
index 1e1acb9..19a19a3 100644
--- a/_doc/conf.py
+++ b/_doc/conf.py
@@ -74,7 +74,7 @@ try:
         sys.argv = ['ryd', 'convert', '--no-pdf', str(fn)]
         rc = RYDCmd()
         rc.parse_args()
-        print(*sys.argv, '->', rc.run())
+        print(sys.argv, '->', rc.run())
     sys.argv = oldargv
 
 except Exception as e:
diff --git a/_test/roundtrip.py b/_test/roundtrip.py
index 53bc667..7ce51ae 100644
--- a/_test/roundtrip.py
+++ b/_test/roundtrip.py
@@ -119,7 +119,7 @@ class YAML(ruamel.yaml.YAML):
         unordered_lines = lkw.pop('unordered_lines', False)
         if expected and expected[0] == '\n':
             expected = expected[1:]
-        lkw['stream'] = st = StringIO() if self.encoding is None else BytesIO()
+        lkw['stream'] = st = StringIO()
         ruamel.yaml.YAML.dump(self, data, **lkw)
         res = st.getvalue()
         print(res)
diff --git a/_test/test_api_change.py b/_test/test_api_change.py
index 7152eda..2b8ad40 100644
--- a/_test/test_api_change.py
+++ b/_test/test_api_change.py
@@ -125,3 +125,22 @@ class TestDuplSet:
             ? c
             ? a
             """))
+
+
+class TestDumpLoadUnicode:
+    # test triggered by SamH on stackoverflow (https://stackoverflow.com/q/45281596/1307905)
+    # and answer by randomir (https://stackoverflow.com/a/45281922/1307905)
+    def test_write_unicode(self, tmpdir):
+        yaml = YAML()
+        text_dict = {"text": u"HELLO_WORLD©"}
+        file_name = str(tmpdir) + '/tstFile.yaml'
+        yaml.dump(text_dict, open(file_name, 'w'))
+        assert open(file_name, 'rb').read().decode('utf-8') == u'text: HELLO_WORLD©\n'
+
+    def test_read_unicode(self, tmpdir):
+        yaml = YAML()
+        file_name = str(tmpdir) + '/tstFile.yaml'
+        with open(file_name, 'wb') as fp:
+            fp.write(u'text: HELLO_WORLD©\n'.encode('utf-8'))
+        text_dict = yaml.load(open(file_name, 'r'))
+        assert text_dict["text"] == u"HELLO_WORLD©"
diff --git a/cyaml.py b/cyaml.py
index 78438ff..7f504a5 100644
--- a/cyaml.py
+++ b/cyaml.py
@@ -92,6 +92,7 @@ class CSafeDumper(CEmitter, SafeRepresenter, Resolver):  # type: ignore
                           explicit_start=explicit_start,
                           explicit_end=explicit_end,
                           version=version, tags=tags)
+        self._emitter = self._serializer = self._representer = self
         SafeRepresenter.__init__(self, default_style=default_style,
                                  default_flow_style=default_flow_style)
         Resolver.__init__(self)
diff --git a/main.py b/main.py
index 58b1c81..35bcdcf 100644
--- a/main.py
+++ b/main.py
@@ -94,7 +94,8 @@ class YAML(object):
             self.Composer = ruamel.yaml.composer.Composer
             self.Constructor = ruamel.yaml.constructor.BaseConstructor
         elif self.typ == 'unsafe':
-            self.Emitter = ruamel.yaml.emitter.Emitter
+            self.Emitter = ruamel.yaml.emitter.Emitter if pure or CEmitter is None \
+                else CEmitter
             self.Representer = ruamel.yaml.representer.Representer
             self.Parser = ruamel.yaml.parser.Parser if pure or CParser is None else CParser
             self.Composer = ruamel.yaml.composer.Composer
@@ -118,7 +119,7 @@ class YAML(object):
         self.version = None
         self.preserve_quotes = None
         self.allow_duplicate_keys = False  # duplicate keys in map, set
-        self.encoding = None
+        self.encoding = 'utf-8'
         self.explicit_start = None
         self.explicit_end = None
         self.tags = None
@@ -382,7 +383,10 @@ class YAML(object):
             delattr(self, "_serializer")
             delattr(self, "_emitter")
         if transform:
-            fstream.write(transform(stream.getvalue()))   # type: ignore
+            val = stream.getvalue()
+            if self.encoding:
+                val = val.decode(self.encoding)
+            fstream.write(transform(val))   # type: ignore
         return None
 
     def get_serializer_representer_emitter(self, stream, tlca):
@@ -425,7 +429,12 @@ class YAML(object):
                                           default_flow_style=default_flow_style)
                 rslvr.__init__(selfx)
         self._stream = stream
-        dumper = XDumper(stream)
+        dumper = XDumper(stream, default_style=self.default_style,
+                         default_flow_style=self.default_flow_style,
+                         allow_unicode=self.allow_unicode, line_break=self.line_break,
+                         explicit_start=self.explicit_start,
+                         explicit_end=self.explicit_end,
+                         version=self.version, tags=self.tags)
         self._emitter = self._serializer = dumper
         return dumper, dumper, dumper