Skip to content

[Merged by Bors] - Remove EntityCommands::add_children#6942

Closed
tim-blackbird wants to merge 1 commit intobevyengine:mainfrom
tim-blackbird:remove-add_child
Closed

[Merged by Bors] - Remove EntityCommands::add_children#6942
tim-blackbird wants to merge 1 commit intobevyengine:mainfrom
tim-blackbird:remove-add_child

Conversation

@tim-blackbird
Copy link
Contributor

@tim-blackbird tim-blackbird commented Dec 13, 2022

Objective

Remove a method with an unfortunate name and questionable usefulness.
Added in #4708

It doesn't make sense to me for us to provide a method to work around a limitation of closures when we can simply, not use a closure.
The limitation in this case is not being able to initialize a variable from inside a closure:

let child_id;
commands.spawn_empty().with_children(|parent| {
    // Error: passing uninitalized variable to a closure.
    child_id = parent.spawn_empty().id();
});

// Do something with child_id

The docs for add_children suggest the following:

let child_id = commands
    .spawn_empty()
    .add_children(|parent| parent.spawn_empty().id());

I would instead suggest using the following snippet.

let parent_id = commands.spawn_empty().id();
let child_id = commands.spawn_empty().set_parent(parent_id).id();

// To be fair, at the time of #4708 this would have been a bit more cumbersome since `set_parent` did not exist.

Using add_children gets more unwieldy when you also want the parent_id.

let parent_commands = commands.spawn_empty();
let parent_id = parent_commands.id();
let child_id = parent_commands.add_children(|parent| parent.spawn_empty().id());

The name

I see why add_children is named that way, it's the non-builder variant of with_children so it kinda makes sense,
but now the method name situation for add_child, add_children and push_children is rather unfortunate.

Removing add_children and renaming push_children to add_children in one go is kinda bleh, but that way we end up with the matching methods add_child and add_children.

Another reason to rename push_children is that it's trying to mimick the Vec api naming but fails because push is for single elements. I guess it should have been extend_children_from_slice, but lets not name it that :)

Questions

Should push_children be renamed in this PR? This would make the migration guide easier to deal with.
Let's do that later.

Does anyone know of a way to do a simple text/regex search through all the github repos for usage of add_children?
That way we can have a better idea of how this will affect users. My guess is that usage of add_children is quite rare.

Migration Guide

The method add_children on EntityCommands was removed.
If you were using add_children over with_children to return data out of the closure you can use set_parent or add_child to avoid the closure instead.

@alice-i-cecile alice-i-cecile added C-Usability A targeted quality-of-life change that makes Bevy easier to use A-Hierarchy labels Dec 13, 2022
Copy link
Member

@alice-i-cecile alice-i-cecile left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I agree with removing this. I'm neutral on the rename: I think it's a better name and should be done eventually, but migrating directly may lead to confusing compiler messages.

@james7132 james7132 added the S-Ready-For-Final-Review This PR has been approved by the community. It's ready for a maintainer to consider merging it label Dec 14, 2022
@james7132
Copy link
Member

Agreed that the API duplication here is definitely undesirable. The recent API additions definitely obviates the need for such an API.

@james7132 james7132 added this to the 0.10 milestone Dec 14, 2022
@cart
Copy link
Member

cart commented Dec 16, 2022

bors r+

bors bot pushed a commit that referenced this pull request Dec 16, 2022
# Objective
Remove a method with an unfortunate name and questionable usefulness.
Added in #4708

It doesn't make sense to me for us to provide a method to work around a limitation of closures when we can simply, *not* use a closure.
The limitation in this case is not being able to initialize a variable from inside a closure:

```rust
let child_id;
commands.spawn_empty().with_children(|parent| {
    // Error: passing uninitalized variable to a closure.
    child_id = parent.spawn_empty().id();
});

// Do something with child_id
```
The docs for `add_children` suggest the following:
```rust
let child_id = commands
    .spawn_empty()
    .add_children(|parent| parent.spawn_empty().id());
```
I would instead suggest using the following snippet.
```rust
let parent_id = commands.spawn_empty().id();
let child_id = commands.spawn_empty().set_parent(parent_id).id();

// To be fair, at the time of #4708 this would have been a bit more cumbersome since `set_parent` did not exist.
```

Using `add_children` gets more unwieldy when you also want the `parent_id`.
```rust
let parent_commands = commands.spawn_empty();
let parent_id = parent_commands.id();
let child_id = parent_commands.add_children(|parent| parent.spawn_empty().id());
```
### The name
I see why `add_children` is named that way, it's the non-builder variant of `with_children` so it kinda makes sense,
but now the method name situation for `add_child`, `add_children` and `push_children` is *rather* unfortunate.

Removing `add_children` and renaming `push_children` to `add_children` in one go is kinda bleh, but that way we end up with the matching methods `add_child` and `add_children`. 

Another reason to rename `push_children` is that it's trying to mimick the `Vec` api naming but fails because `push` is for single elements. I guess it should have been `extend_children_from_slice`, but lets not name it that :)

### Questions
~~Should `push_children` be renamed in this PR? This would make the migration guide easier to deal with.~~
Let's do that later.

Does anyone know of a way to do a simple text/regex search through all the github repos for usage of `add_children`?
That way we can have a better idea of how this will affect users. My guess is that usage of `add_children` is quite rare.

## Migration Guide
The method `add_children` on `EntityCommands` was removed.
If you were using `add_children` over `with_children` to return data out of the closure you can use `set_parent` or `add_child` to avoid the closure instead.

Co-authored-by: devil-ira <justthecooldude@gmail.com>
@bors bors bot changed the title Remove EntityCommands::add_children [Merged by Bors] - Remove EntityCommands::add_children Dec 16, 2022
@bors bors bot closed this Dec 16, 2022
@tim-blackbird tim-blackbird deleted the remove-add_child branch December 16, 2022 20:21
bors bot pushed a commit that referenced this pull request Dec 25, 2022
# Objective
Align the hierarchy API between `EntityCommands` and `EntityMut`.

Added missing methods to `EntityMut`.
Replaced the duplicate `Command` implementations with the ones on `EntityMut` (e.g. The `AddChild` command is now just `world.entity_mut(..).add_child(..)`)

Fixed `update_old_parents` not sending `ChildAdded` events.

This PR does not add `add_children` to `EntityMut` as I would like to remove it from `EntityCommands` instead in #6942.

## Changelog
* Added `add_child`, `set_parent` and `remove_parent` to `EntityMut`
* Fixed missing `ChildAdded` events


Co-authored-by: devil-ira <justthecooldude@gmail.com>
bors bot pushed a commit that referenced this pull request Dec 25, 2022
# Objective
Align the hierarchy API between `EntityCommands` and `EntityMut`.

Added missing methods to `EntityMut`.
Replaced the duplicate `Command` implementations with the ones on `EntityMut` (e.g. The `AddChild` command is now just `world.entity_mut(..).add_child(..)`)

Fixed `update_old_parents` not sending `ChildAdded` events.

This PR does not add `add_children` to `EntityMut` as I would like to remove it from `EntityCommands` instead in #6942.

## Changelog
* Added `add_child`, `set_parent` and `remove_parent` to `EntityMut`
* Fixed missing `ChildAdded` events


Co-authored-by: devil-ira <justthecooldude@gmail.com>
alradish pushed a commit to alradish/bevy that referenced this pull request Jan 22, 2023
# Objective
Remove a method with an unfortunate name and questionable usefulness.
Added in bevyengine#4708

It doesn't make sense to me for us to provide a method to work around a limitation of closures when we can simply, *not* use a closure.
The limitation in this case is not being able to initialize a variable from inside a closure:

```rust
let child_id;
commands.spawn_empty().with_children(|parent| {
    // Error: passing uninitalized variable to a closure.
    child_id = parent.spawn_empty().id();
});

// Do something with child_id
```
The docs for `add_children` suggest the following:
```rust
let child_id = commands
    .spawn_empty()
    .add_children(|parent| parent.spawn_empty().id());
```
I would instead suggest using the following snippet.
```rust
let parent_id = commands.spawn_empty().id();
let child_id = commands.spawn_empty().set_parent(parent_id).id();

// To be fair, at the time of bevyengine#4708 this would have been a bit more cumbersome since `set_parent` did not exist.
```

Using `add_children` gets more unwieldy when you also want the `parent_id`.
```rust
let parent_commands = commands.spawn_empty();
let parent_id = parent_commands.id();
let child_id = parent_commands.add_children(|parent| parent.spawn_empty().id());
```
### The name
I see why `add_children` is named that way, it's the non-builder variant of `with_children` so it kinda makes sense,
but now the method name situation for `add_child`, `add_children` and `push_children` is *rather* unfortunate.

Removing `add_children` and renaming `push_children` to `add_children` in one go is kinda bleh, but that way we end up with the matching methods `add_child` and `add_children`. 

Another reason to rename `push_children` is that it's trying to mimick the `Vec` api naming but fails because `push` is for single elements. I guess it should have been `extend_children_from_slice`, but lets not name it that :)

### Questions
~~Should `push_children` be renamed in this PR? This would make the migration guide easier to deal with.~~
Let's do that later.

Does anyone know of a way to do a simple text/regex search through all the github repos for usage of `add_children`?
That way we can have a better idea of how this will affect users. My guess is that usage of `add_children` is quite rare.

## Migration Guide
The method `add_children` on `EntityCommands` was removed.
If you were using `add_children` over `with_children` to return data out of the closure you can use `set_parent` or `add_child` to avoid the closure instead.

Co-authored-by: devil-ira <justthecooldude@gmail.com>
alradish pushed a commit to alradish/bevy that referenced this pull request Jan 22, 2023
…yengine#6926)

# Objective
Align the hierarchy API between `EntityCommands` and `EntityMut`.

Added missing methods to `EntityMut`.
Replaced the duplicate `Command` implementations with the ones on `EntityMut` (e.g. The `AddChild` command is now just `world.entity_mut(..).add_child(..)`)

Fixed `update_old_parents` not sending `ChildAdded` events.

This PR does not add `add_children` to `EntityMut` as I would like to remove it from `EntityCommands` instead in bevyengine#6942.

## Changelog
* Added `add_child`, `set_parent` and `remove_parent` to `EntityMut`
* Fixed missing `ChildAdded` events


Co-authored-by: devil-ira <justthecooldude@gmail.com>
ItsDoot pushed a commit to ItsDoot/bevy that referenced this pull request Feb 1, 2023
# Objective
Remove a method with an unfortunate name and questionable usefulness.
Added in bevyengine#4708

It doesn't make sense to me for us to provide a method to work around a limitation of closures when we can simply, *not* use a closure.
The limitation in this case is not being able to initialize a variable from inside a closure:

```rust
let child_id;
commands.spawn_empty().with_children(|parent| {
    // Error: passing uninitalized variable to a closure.
    child_id = parent.spawn_empty().id();
});

// Do something with child_id
```
The docs for `add_children` suggest the following:
```rust
let child_id = commands
    .spawn_empty()
    .add_children(|parent| parent.spawn_empty().id());
```
I would instead suggest using the following snippet.
```rust
let parent_id = commands.spawn_empty().id();
let child_id = commands.spawn_empty().set_parent(parent_id).id();

// To be fair, at the time of bevyengine#4708 this would have been a bit more cumbersome since `set_parent` did not exist.
```

Using `add_children` gets more unwieldy when you also want the `parent_id`.
```rust
let parent_commands = commands.spawn_empty();
let parent_id = parent_commands.id();
let child_id = parent_commands.add_children(|parent| parent.spawn_empty().id());
```
### The name
I see why `add_children` is named that way, it's the non-builder variant of `with_children` so it kinda makes sense,
but now the method name situation for `add_child`, `add_children` and `push_children` is *rather* unfortunate.

Removing `add_children` and renaming `push_children` to `add_children` in one go is kinda bleh, but that way we end up with the matching methods `add_child` and `add_children`. 

Another reason to rename `push_children` is that it's trying to mimick the `Vec` api naming but fails because `push` is for single elements. I guess it should have been `extend_children_from_slice`, but lets not name it that :)

### Questions
~~Should `push_children` be renamed in this PR? This would make the migration guide easier to deal with.~~
Let's do that later.

Does anyone know of a way to do a simple text/regex search through all the github repos for usage of `add_children`?
That way we can have a better idea of how this will affect users. My guess is that usage of `add_children` is quite rare.

## Migration Guide
The method `add_children` on `EntityCommands` was removed.
If you were using `add_children` over `with_children` to return data out of the closure you can use `set_parent` or `add_child` to avoid the closure instead.

Co-authored-by: devil-ira <justthecooldude@gmail.com>
ItsDoot pushed a commit to ItsDoot/bevy that referenced this pull request Feb 1, 2023
…yengine#6926)

# Objective
Align the hierarchy API between `EntityCommands` and `EntityMut`.

Added missing methods to `EntityMut`.
Replaced the duplicate `Command` implementations with the ones on `EntityMut` (e.g. The `AddChild` command is now just `world.entity_mut(..).add_child(..)`)

Fixed `update_old_parents` not sending `ChildAdded` events.

This PR does not add `add_children` to `EntityMut` as I would like to remove it from `EntityCommands` instead in bevyengine#6942.

## Changelog
* Added `add_child`, `set_parent` and `remove_parent` to `EntityMut`
* Fixed missing `ChildAdded` events


Co-authored-by: devil-ira <justthecooldude@gmail.com>
@alice-i-cecile alice-i-cecile added A-ECS Entities, components, systems, and events and removed A-Hierarchy labels Jan 28, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

A-ECS Entities, components, systems, and events C-Usability A targeted quality-of-life change that makes Bevy easier to use S-Ready-For-Final-Review This PR has been approved by the community. It's ready for a maintainer to consider merging it

Projects

None yet

Development

Successfully merging this pull request may close these issues.

6 participants