Top Tweets for #UIAction

11 months ago

One big discussion that we have in #Android is between passing individual functions to composables that need to make action, or passing #uiAction with a sealed class of possible actions. Using individual functions for actions is: 👉 Simple, as we need no extra classes. 👉 Safer, as each composable receives only what is needs. 😨 Can badly scale, as every new action might require passing it through many components, and every argument change might require changes in many signatures. On the other hand using uiAction: 👉 Scales better, as we do not need to pass all functions separately, and maintain their definitions. 👉 Aggregates all actions in one place. 😨 Adds a lot of new classes to our project. 😨 Reduces safety, as many composables receive functions that allow making actions they do not need. That problem can be mitigated by using sealed interfaces to define groups of actions, but that adds even more actions, what makes project harder to understand and reason about. 💡 It is often recommended to choose the first option when we expect simple screens, and the second one for complex, evolving screens. Which one do you prefer❓ Which one do you use❓ Neither of those two approaches deals well with screens that have a lot of state-dependent actions.

ktdotacademy's tweet photo. One big discussion that we have in #Android is between passing individual functions to composables that need to make action, or passing #uiAction with a sealed class of possible actions.

Using individual functions for actions is:

👉 Simple, as we need no extra classes.
👉 Safer, as each composable receives only what is needs.
😨 Can badly scale, as every new action might require passing it through many components, and every argument change might require changes in many signatures.

On the other hand using uiAction:

👉 Scales better, as we do not need to pass all functions separately, and maintain their definitions.
👉 Aggregates all actions in one place.
😨 Adds a lot of new classes to our project.
😨 Reduces safety, as many composables receive functions that allow making actions they do not need. That problem can be mitigated by using sealed interfaces to define groups of actions, but that adds even more actions, what makes project harder to understand and reason about.

💡 It is often recommended to choose the first option when we expect simple screens, and the second one for complex, evolving screens. Which one do you prefer❓ Which one do you use❓

Neither of those two approaches deals well with screens that have a lot of state-dependent actions.

609

WomenNowSN @WomenNowSN

over 3 years ago

Check out Jesalyn's first blog post on #UIAction redirects in #ServicePortal! https://t.co/9BP55HUeR9 #womenNowDev #ServiceNow #ServiceNowDeveloper #SNDevs

Top Tweets for #UIAction

Last Seen Hashtags on Sotwe

Trends for you

Most Popular Users