diff --git a/commands/preprocess.py b/commands/preprocess.py
index e745b8e1..5172fa6f 100644
--- a/commands/preprocess.py
+++ b/commands/preprocess.py
@@ -386,7 +386,7 @@ def remove_unused_external(html):
 
 
 def preprocess_html_file(root, fn, rename_map):
-    parser = etree.HTMLParser()
+    parser = etree.HTMLParser(encoding='utf-8')
     html = etree.parse(fn, parser)
     output = io.StringIO()
 
diff --git a/commands/preprocess_cssless.py b/commands/preprocess_cssless.py
index e6f9e022..d377f940 100644
--- a/commands/preprocess_cssless.py
+++ b/commands/preprocess_cssless.py
@@ -29,9 +29,9 @@
 
 
 def preprocess_html_merge_cssless(src_path, dst_path):
-    with open(src_path, 'r') as a_file:
+    with open(src_path, 'r', encoding='utf-8') as a_file:
         content = a_file.read()
-        parser = etree.HTMLParser()
+        parser = etree.HTMLParser(encoding='utf-8')
         stripped = content.strip()
         root = etree.fromstring(stripped, parser)
 
diff --git a/gadgets/replace_tests_base.py b/gadgets/replace_tests_base.py
index 353d927b..e3158ea7 100755
--- a/gadgets/replace_tests_base.py
+++ b/gadgets/replace_tests_base.py
@@ -46,14 +46,14 @@ def main():
 
     for path in paths:
         print('Processing {0}'.format(path))
-        with open(path, 'r') as file:
+        with open(path, 'r', encoding='utf-8') as file:
             text = file.read()
 
         # TODO user proper XML parser, not this hack
         text = re.sub('<link rel="selenium.base" href="(.*)" />',
                       '<link rel="selenium.base" href="' + args.url + '" />', text)
 
-        with open(path, 'w') as file:
+        with open(path, 'w', encoding='utf-8') as file:
             file.write(text)
 
 if __name__ == '__main__':
diff --git a/gadgets/sync_tests_mwiki.py b/gadgets/sync_tests_mwiki.py
index b9e13374..dab44701 100755
--- a/gadgets/sync_tests_mwiki.py
+++ b/gadgets/sync_tests_mwiki.py
@@ -71,7 +71,7 @@ def sync_single_page(page, direction, dest_root):
     if direction == SYNC_DIRECTION_UPLOAD:
         if not os.path.exists(dest_path):
             return
-        with open(dest_path, 'r') as file:
+        with open(dest_path, 'r', encoding='utf-8') as file:
             new_text = file.read()
         if fix_whitespace(text) != fix_whitespace(new_text):
             page.put(new_text, 'sync with git')
@@ -82,7 +82,7 @@ def sync_single_page(page, direction, dest_root):
         if not os.path.exists(dest_dir):
             os.makedirs(dest_dir)
 
-        with open(dest_path, 'w') as file:
+        with open(dest_path, 'w', encoding='utf-8') as file:
             file.write(fix_whitespace(text))
         print('Downloaded {0}'.format(dest_path))
 
diff --git a/index2ddg.py b/index2ddg.py
index 71d15ed8..c4249b99 100755
--- a/index2ddg.py
+++ b/index2ddg.py
@@ -582,7 +582,7 @@ def main():
         # i+=1
 
         root = e.parse(os.path.join(args.reference, fn),
-                       parser=html.HTMLParser())
+                       parser=html.HTMLParser(encoding='utf-8'))
 
         for ident in idents:
 
diff --git a/tests/test_preprocess.py b/tests/test_preprocess.py
index 3c0c9525..1d1b0cb5 100644
--- a/tests/test_preprocess.py
+++ b/tests/test_preprocess.py
@@ -258,7 +258,7 @@ def setUp(self):
         self.testdata = os.path.join(os.path.dirname(__file__),
                                      'preprocess_data')
         infile = os.path.join(self.testdata, "fabs.html")
-        self.parser = etree.HTMLParser()
+        self.parser = etree.HTMLParser(encoding='utf-8')
         self.html = etree.parse(infile, self.parser)
 
     # Check whether the HTML matches the contents of the specified test data
diff --git a/tests/test_preprocess_cssless.py b/tests/test_preprocess_cssless.py
index 64864ccf..b62e7f2c 100644
--- a/tests/test_preprocess_cssless.py
+++ b/tests/test_preprocess_cssless.py
@@ -43,10 +43,10 @@ def test_preprocess_html_merge_cssless(self):
 
         preprocess_html_merge_cssless(src_path, dst_path)
 
-        with open(dst_path, 'r') as a_file:
+        with open(dst_path, 'r', encoding='utf-8') as a_file:
             test = a_file.read()
 
-        with open(expected_path, 'r') as a_file:
+        with open(expected_path, 'r', encoding='utf-8') as a_file:
             expected = a_file.read()
 
         self.assertEqual(test, expected)
@@ -63,10 +63,10 @@ def test_preprocess_html_merge_cssless2(self):
 
         preprocess_html_merge_cssless(src_path, dst_path)
 
-        with open(dst_path, 'r') as a_file:
+        with open(dst_path, 'r', encoding='utf-8') as a_file:
             test = a_file.read()
 
-        with open(expected_path, 'r') as a_file:
+        with open(expected_path, 'r', encoding='utf-8') as a_file:
             expected = a_file.read()
 
         self.assertEqual(test, expected)
@@ -83,7 +83,7 @@ def assert_converts_html(self, input, expected_output, function):
         expected_output = \
             '<html><body>{0}</body></html>'.format(expected_output)
 
-        parser = etree.HTMLParser()
+        parser = etree.HTMLParser(encoding='utf-8')
         root = etree.fromstring(input, parser)
 
         root = function(root)