Moved creation of resolve errors list to inside the resolve function

Author: spookylukey

fluent.runtime/fluent/runtime/__init__.py

@@ -53,9 +53,7 @@ def format(self, message_id, args=None):
         message = self._get_message(message_id)
         if args is None:
             args = {}
-        errors = []
-        resolved = resolve(self, message, args, errors=errors)
-        return resolved, errors
+        return resolve(self, message, args)
 
     def _get_message(self, message_id):
         if message_id.startswith('-'):

fluent.runtime/fluent/runtime/resolver.py

@@ -46,19 +46,18 @@ class ResolverEnvironment(object):
     part_count = attr.ib(default=0)
 
 
-def resolve(context, message, args, errors=None):
+def resolve(context, message, args):
     """
     Given a FluentBundle, a Message instance and some arguments,
     resolve the message to a string.
 
     This is the normal entry point for this module.
     """
-    if errors is None:
-        errors = []
+    errors = []
     env = ResolverEnvironment(context=context,
                               args=args,
                               errors=errors)
-    return fully_resolve(message, env)
+    return fully_resolve(message, env), errors
 
 
 def fully_resolve(expr, env):