Streamlit Feedback Component

Common errors

• streamlit.errors.StreamlitAPIException: st.text_input() was called with a key='...' that is already in use. Each widget must have a unique key.

  cause You have multiple `st_feedback` components (or other Streamlit widgets) on the same page without unique `key` arguments, causing a conflict.
  fix Assign a unique string to the `key` parameter for each `st_feedback` call, e.g., `st_feedback(..., key="feedback_section_1")`.

• ModuleNotFoundError: No module named 'streamlit_feedback'

  cause The `streamlit-feedback` library is not installed in your current Python environment.
  fix Install the library using pip: `pip install streamlit-feedback`.

• My feedback component doesn't appear in my Streamlit app when I run it.

  cause This often occurs if the Streamlit app is not being run correctly, if the component call is inside an un-met conditional block, or if there's an earlier error preventing rendering. It also might not appear until a certain interaction happens.
  fix Ensure your script is launched with `streamlit run your_script_name.py`. Verify that the `st_feedback` call is executed unconditionally, or that its conditional logic is met. Check the terminal for any Python or Streamlit-specific errors that might be preventing rendering.

Warnings

• gotcha The `st_feedback` component must be executed within a Streamlit application (`streamlit run your_app.py`). Running the script as a standalone Python file will not display the UI component and may result in errors related to missing Streamlit context.
  Fix: Always launch your application using the `streamlit run` command.
• gotcha When using `st_feedback` multiple times on the same Streamlit page, each instance *must* be assigned a unique `key` parameter. Failing to do so will cause a `StreamlitAPIException` due to key conflicts.
  Fix: Provide a unique string for the `key` parameter for every `st_feedback` call, e.g., `st_feedback(..., key='unique_id_1')`.
• gotcha The structure of the dictionary returned by `st_feedback` when feedback is submitted varies based on the `feedback_type` (e.g., 'thumbs', 'faces', 'text'). Developers must inspect the `feedback_type` field and conditionally access fields like `score` or `text`.
  Fix: Always check `feedback_result['feedback_type']` before trying to access `feedback_result['score']` or `feedback_result['text']` to ensure correct data parsing, as not all fields are present for all types.

Install

• pip install streamlit-feedback Install library

Imports

• st_feedback

  from streamlit_feedback import st_feedback

Quickstart

This quickstart demonstrates how to integrate `streamlit-feedback` into a basic Streamlit application. It shows both 'thumbs' and 'faces' feedback types, including how to handle the returned feedback and the essential use of the `key` parameter for multiple components. The app must be run using `streamlit run`.

import streamlit as st
from streamlit_feedback import st_feedback

st.set_page_config(layout="wide")

st.header("User Feedback Example")

# Basic thumbs up/down feedback
feedback_thumbs = st_feedback(
    feedback_type="thumbs",
    optional_text_label="Tell us more (optional):"
)

if feedback_thumbs:
    st.success(f"Thumbs Feedback received: {feedback_thumbs}")
    # In a real application, you would store this feedback (e.g., to a database).
else:
    st.info("Please provide your thumbs feedback.")

st.markdown("--- ")

# Faces feedback with a custom key
st.subheader("Rate your experience")
feedback_faces = st_feedback(
    feedback_type="faces",
    feedback_options=["😀 Excellent", "🙂 Good", "😐 Neutral", "🙁 Poor", "😞 Very Poor"],
    key="experience_feedback" # Unique key is crucial for multiple widgets
)

if feedback_faces:
    st.success(f"Faces Feedback received: {feedback_faces}")
else:
    st.info("Please rate your experience.")

st.caption("To run this app, save it as `app.py` and execute: `streamlit run app.py`")