adding post processing to stop ans in first '\n'

bothub_nlp_api/utils.py

@@ -300,9 +300,14 @@ def request_wenigpt(context, question):
     response_json = response.json()
     text_answers = response_json["output"].get("text")
 
+    answers = []
     if text_answers:
-        text_answers = [{"text": answer.strip()} for answer in text_answers]
-    else:
-        text_answers = []
-
-    return {"answers": text_answers, "id": "0"}
+        for answer in text_answers:
+            answer = answer.strip()
+            ans = ""
+            for ch in answer:
+                if ch == '\n':
+                    break
+                ans += ch
+            answer.append({"text": ans})
+    return {"answers": answers, "id": "0"}