Would changing the value of the dcc.Store component's clear_data

Dash Python
1

This one was not addressed in the Sharing Data Between Callbacks video - a minor oversight in a great video,

And it is defined, but not discussed, in the documentation:

clear_data ( boolean ; default False ): Set to True to remove the data contained in data_key .

But it has become important to my attempts to resolve a major issue, notably that the previous session’s data remains cached somewhere in my browser or app, making it impossible to analyze new data.

I spent hours experimenting with gc.collect, clearing lists, deleting variables, closing files, etc…and now I am trying to clear this data by using dcc.Store to contain it and then change the clear_value parameter to True when the Reset button is clicked. The code below - my latest attempt - is not successful.

May I ask, what is wrong with it? Or is my approach misguided here?

Do you know of a solution to this kind of problem?

Here is the code for dcc.Store at the tail end of an Accordion component:

 dcc.Store(id='store-data', data=[], storage_type='memory', clear_data=False) # 'local' or 'session'
]) # Later use callback to chang clear_data to True when final analysis is completed.

Here is the callback in question:

@callback(
    Output('store-data', 'clear_data'),
    [Input("recipe-textarea-reset-button", "n_clicks")],
    prevent_initial_call=True
)
def clear_memory(n_clicks):
    if n_clicks > 0:
        return True

Thank you,

Robert

2

If the intent of your callback is to clear the data in your store-data dcc.Store, I would do the following:

@callback(
    Output('store-data', 'data'),
    [Input("recipe-textarea-reset-button", "n_clicks")],
    prevent_initial_call=True
)
def clear_memory(n_clicks):
    if n_clicks > 0:
        return []

I am a little perplexed on what the data_key property is used for. In my usage, I set the storage_type to session. I would assume the above callback would work if set to memory

3

Thank you for your response.

I was also perplexed by the documentation on the data_key.

I have only been writing in Dash for a few weeks, and coding for about eight months; but I ultimately decided that documentation needs editing. It seems to imply there is a property called ‘data_key’ like there is a property called ‘clear_data’. But now I think it just refers to the key “data”, where the data is stored. I think you’ve written a much better callback to control that action, and I am going to try it out now.

Thank you again.

Robert

4

This was the initial reaction to a cut-and-paste job.

I’ve never heard of `dash.callback_context’ and need to look it up:

In the callback for output(s):
  store-data.data
Output 0 (store-data.data) is already in use.
Any given output can only have one callback that sets it.
To resolve this situation, try combining these into
one callback function, distinguishing the trigger
by using `dash.callback_context` if necessary.

Without my morning coffee, I am a little fuzzy, but I think it means 1) you wrote the right callback, but its 2) conflicting with another callback pasted below. I need to wake up and experiment with it.

Do you think it mean there is a redundancy and the “data” from the callback below is already stored in dcc.Store and I don’t need another one to control it? That I just need to incorporate the clear_data property into the callback/function below?

Like I said, I need to wake up!

If you are interested (I don’t want to impose), here is only other callback that uses data:

def create_nutritional_profile(n_clicks, data):
    if n_clicks > 0:

        finalize_ing_dicts(data)
        create_match_idslist()
        create_nutrient_dicts()
        merge_nutrient_to_ing_dicts()
        results = final_conversions()

        for each_dict in results:
            for k,v in each_dict.items():
                print(k, v)

        df = pd.read_csv('final_charts_data.csv')
        # round numbers to two decimal places
        dff = df.round(decimals = 2)
        # supress/hide certain columns
        dfff = dff.drop(['Gender', 'Ages', 'Benchmarks'], axis=1)
        # unique values only
        final_data = dfff.drop_duplicates(subset=['Nutrients'])

        return dbc.Table.from_dataframe(final_data, id='review-final-analysis-datatable',
            class_name="me-2", 
            striped=True, 
            bordered=True, 
            hover=True, 
            style={"width": "1200px"} )

Thanks again,

Robert

5

That error means you have another callback that sets store-data.data parameter. Dash only allows one callback to set the output of a component’s parameter. So in your case, you have a callback that stores the results in store-data.data…that same function must also be used to clear the data.

1 Like