Streamlit script gets stuck on @st.cache

Question

I'm using streamlit to create a dashboard to analyse Indian Gov data using this python package, [datagovindia][1]. Everything works fine till I use @st.cache to cache the python object returned from above library.

1. @st.cache() with no args gets stuck on the very first load.

2. @st.cache(allow_output_mutation=True) lets the script load for first time but gives this error second time: [![enter image description here][2]][2]

3. @st.experimental_singleton works same as 2, i.e. works for first time and same error for next load. Here is the code:

   import streamlit as st
   
   @st.cache(suppress_st_warning=True)
   def getDataGovAdaptor(): 
       return DataGovIndia(API_KEY)
   
   datagovin = getDataGovAdaptor()
   result = datagovin.search(description="Wheat",max_results=1,print_results=True)
   

  [1]: https://github.com/addypy/datagovindia
  [2]: https://i.stack.imgur.com/CSGC1.png

Answer 1

I did try your script using API_KEY = '579b464db66ec23bdd000001cdd3946e44ce4aad7209ff7b23ac571b' and it did not gets stuck it just takes more time around 118s with cache and 18s without cache.

Try to understand the cache idea at https://docs.streamlit.io/library/api-reference/performance/st.cache.

Then compare the function of DataGovIndia()

Note:
=====
Initializing this class may take a few seconds, depending on the speed
of your internet connection.
Initialization performs two key tasks:
    1) Tests server
    - Tests data.gov.in server to check if APIs are functional.
    2) Validates the API-key provided
    - Once validated, the API-key is stored and does not need to be entered again.
    3) Loads latest API meta-data.
    - Downloads and loads data containing the latest details of available APIs.

Can the 1) Tests server be cached? No.

So it is better not to use the cache.