Create a large amount of dev data

[ncounter]

An already set up database must exists ahead (rake dev:test_data:create)

Run:

• bundle exec rake dev:test_data:create_more_data[100] if inside the container
• docker compose run --rm frontend rake dev:test_data:create_more_data[100] if outside of the container

where [100] can be any number you desire to flood your db with (the number represents the repetitions the create objects tasks are executed).

[danidoni]

Why don't we use this as the default data creation?

[ncounter]

Why don't we use this as the default data creation?

That's a valid point to discuss further: initially my attempt was meant to be a follow-up step, just in case there is a need to test the app performance wise, without affecting the existing setup. Basically I see some points to adjust if we want to squash the create and create_more_data into a default one task:

• it may take a lot of time depending on the repetitions value
  • what to do with the special corner cases like Rake::Task['dev:requests:request_with_multiple_submit_actions_builds_and_diffs'].invoke(1, 1000), it's not a default case we want to set up all the time, but it may be useful to create on purpose
• how to handle the initialization of the database and all the sub steps that are not meant to be repeated
• at the moment the default create destroys and sets up a new db on every call; the create_more_data is something like and idempotent task instead, it always does the same change without changing the existing data

Ideally, we should split better what is an initial db setup that is required in order to play with the app (the minimal data), and what is a pure data_creation that can be called as much as we want to increase the db size.

[danidoni]

Why don't we use this as the default data creation?

That's a valid point to discuss further: initially my attempt was meant to be a follow-up step, just in case there is a need to test the app performance wise, without affecting the existing setup. Basically I see some points to adjust if we want to squash the create and create_more_data into a default one task:

* it may take a lot of time depending on the repetitions value
  
  * what to do with the special corner cases like `Rake::Task['dev:requests:request_with_multiple_submit_actions_builds_and_diffs'].invoke(1, 1000)`, it's not a default case we want to set up all the time, but it may be useful to create on purpose

* how to handle the initialization of the database and all the sub steps that are not meant to be repeated

* at the moment the default `create` destroys and sets up a new db on every call; the `create_more_data` is something like and idempotent task instead, it always does the same change without changing the existing data

Ideally, we should split better what is an initial db setup that is required in order to play with the app (the minimal data), and what is a pure data_creation that can be called as much as we want to increase the db size.

I see

[danidoni]

If it's not valid, what happens with this action?

[ncounter]

I think it doesn't get saved thus it doesn't get into the request that has being created. It's the previous already existing logic implementation.

[danidoni]

Why it cannot be valid?