support streamlit-feedback

README.md

@@ -181,6 +181,7 @@ if show_history:
 	- [x] streaming output message
 	- [x] show message in expander
 	- [ ] style the output message
+    - [x] feedback by user
 
 - [x] export & import chat history
 	- [x] export to markdown

example.py

@@ -31,6 +31,22 @@
 chat_box.init_session()
 chat_box.output_messages()
 
+def on_feedback(
+    feedback,
+    chat_history_id: str = "",
+    history_index: int = -1,
+):
+    reason = feedback["text"]
+    score_int = chat_box.set_feedback(feedback=feedback, history_index=history_index) # convert emoji to integer
+    # do something
+    st.session_state["need_rerun"] = True
+
+
+feedback_kwargs = {
+    "feedback_type": "thumbs",
+    "optional_text_label": "欢迎反馈您打分的理由",
+}
+
 if query := st.chat_input('input your question here'):
     chat_box.user_say(query)
     if streaming:
@@ -51,6 +67,11 @@
         # update the element without focus
         chat_box.update_msg(text, element_index=0, streaming=False, state="complete")
         chat_box.update_msg("\n\n".join(docs), element_index=1, streaming=False, state="complete")
+        chat_history_id = "some id"
+        chat_box.show_feedback(**feedback_kwargs,
+                                key=chat_history_id,
+                                on_submit=on_feedback,
+                                kwargs={"chat_history_id": chat_history_id, "history_index": len(chat_box.history) - 1})
     else:
         text, docs = llm.chat(query)
         chat_box.ai_say(

setup.py

@@ -9,7 +9,7 @@ def readme():
 
 setuptools.setup(
     name='streamlit-chatbox',
-    version='1.1.10',
+    version='1.1.11',
     author='liunux',
     author_email='[email protected]',
     description='A chat box and some helpful tools used to build chatbot app with streamlit',
@@ -23,5 +23,6 @@ def readme():
     install_requires=[
         'streamlit>=1.26.0',
         'simplejson',
+        'streamlit-feedback',
     ]
 )

streamlit_chatbox/__init__.py

@@ -4,7 +4,7 @@
 from .messages import *
 
 
-__version__ = "1.1.9"
+__version__ = "1.1.11"
 
 
 __all__ = [

streamlit_chatbox/messages.py

@@ -1,4 +1,5 @@
 from streamlit_chatbox.elements import *
+from streamlit_feedback import streamlit_feedback
 import time
 import inspect
 import simplejson as json
@@ -15,6 +16,12 @@
 #           - if in expander: element is rendered in a st.empty in the st.status
 
 
+POSSIBLE_SCORES = {
+    "thumbs": ["👍", "👎"],
+    "faces": ["😞", "🙁", "😐", "🙂", "😀"],
+}
+
+
 class ChatBox:
     def __init__(
         self,
@@ -290,20 +297,45 @@ def ai_say(
         for element in elements:
             element(render_to=container)
 
-        self.history.append({"role": "assistant", "elements": elements, "metadata": metadata})
+        self.history.append({"role": "assistant", "elements": elements, "metadata": metadata.copy()})
         return elements
 
+    def show_feedback(self, history_index=-1, **kwargs):
+        '''
+        render feedback component
+        '''
+        with self._chat_containers[history_index]:
+            self.history[history_index]["metadata"]["feedback_kwargs"] = kwargs
+            return streamlit_feedback(**kwargs)
+
+    def set_feedback(self, feedback: Dict, history_index=-1) -> int:
+        '''
+        set the feedback state for msg with a index of history_index
+        return the index of streamlit_feedback's emoji score
+        '''
+        self.history[history_index]["metadata"]["feedback"] = feedback
+        score = feedback.get("score")
+        for v in POSSIBLE_SCORES.values():
+            if score in v:
+                return v.index(score)
+
     def output_messages(self):
         self.init_session()
         self._chat_containers = []
-        for msg in self.history:
+        for i, msg in enumerate(self.history):
             avatar = self._user_avatar if msg["role"] == "user" else self._assistant_avatar
             chat_ele = st.chat_message(msg["role"], avatar=avatar)
             container = chat_ele.container()
             self._chat_containers.append(container)
             for element in msg["elements"]:
                 element(render_to=container)
 
+            feedback_kwargs = msg["metadata"].get("feedback_kwargs", {})
+            if feedback_kwargs:
+                if feedback := msg["metadata"].get("feedback"):
+                    feedback_kwargs["disable_with_score"] = feedback["score"]
+                self.show_feedback(history_index=i, **feedback_kwargs)
+
     def update_msg(
         self,
         element: Union["OutputElement", str] = None,
@@ -314,6 +346,7 @@ def update_msg(
         title: str = None,
         expanded: bool = None,
         state: bool = None,
+        metadata: Dict = {},
     ) -> st._DeltaGenerator:
         self.init_session()
         if not self.history or not self.history[history_index]["elements"]:
@@ -331,6 +364,8 @@ def update_msg(
             element.status_from(old_element)
             self.history[history_index]["elements"][element_index] = element
 
+        self.history[history_index]["metadata"].update(metadata)
+
         dg = old_element.update_element(
             element,
             title=title,