adding encoding

fextorbis.py

@@ -6,6 +6,7 @@ from __future__ import absolute_import, unicode_literals, division
 import json
 import logging
 import subprocess
+import io
 
 import nlp_ws
 from desifex.docextractor import DocExtractor
@@ -57,8 +58,14 @@ class FextorbisWorker(nlp_ws.NLPWorker):
         )
         results = docex.extract_from_document(input_path)
 
-        with open(output_path, 'w') as json_ofs:
-            json.dump(results, json_ofs)
+        encode_output = task_options.get('encoding', False)
+        if encode_output == False:
+            with open(output_path, 'w') as json_ofs:
+                json.dump(results, json_ofs)
+        else: 
+            with io.open(output_path, 'w',encoding='utf8') as json_ofs:
+                data = json.dumps(results, ensure_ascii=False, encoding='utf8')
+                json_ofs.write(unicode(data))
 
     @staticmethod
     def __add_filter(cfgdict, feature, filterdef):