Explain the "dispose" method appropriately

[rohit2sharma95]

Closes #27957

Updated the document for the alert component.

In my opinion, it should be done for all the components that have dispose method.

[XhmikosR]

That's not how you rebase :P

[rohit2sharma95]

My bad @XhmikosR. I actually merged a pull request to update my branch.

[XhmikosR]

@rohit2sharma95 can you please update all the other instance too?

[rohit2sharma95]

Sure @XhmikosR
Let me do it 😊

[XhmikosR]

@mdo can you have a quick look regarding capitalization et all?

[mdo]

If you let me know which one does what in a list here, I can happily update. I'm already riffing on this PR to fix the inconsistency around our methods and options documentation (going all in on the tables instead of a mix of tables and text).

[mdo]

@Johann-S Friendly ping—can you list out what is removed for each component? <3

[rohit2sharma95]

@mdo These are my findings after reading the code
@Johann-S Please correct me where I am wrong :)

When a constructor is called for any component, it stores the data (the actual instance being created from that constructor) on a private variable (not on the DOM element in real). The association between the private variable and DOM element is done by a key and id, those are registered on the DOM Element.
This data is stored so that the user can get the actual instance of the component by the DOM element and it is being used in the jQueryInterface as well.

So when calling the dispose method for any component's instance:

• Alert:
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Button:
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Carousel:
  • Removes event listeners from the DOM Element
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Collapse:
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Dropdowns:
  • Removes event listeners from the DOM Element
  • Removes this instance from the DOM Element
  • Destroys popper's instance if created for the dropdown
  • Sets the instance's private properties to null
• Modals:
  • Removes event listeners from the DOM Elements (from window and document as well)
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Toast:
  • Clears timeout and removes show class from the DOM Element if exists
  • Removes event listeners from the DOM Element
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null
• Tooltips and popovers:
  • Removes event listeners from the DOM Elements
  • Removes this instance from the DOM Element
  • Destroys popper's instance if created for the tooltip or popover
  • Sets the instance's private properties to null
• Scrollspy:
  • Removes event listeners from the DOM Element
  • Removes this instance from the DOM Element
  • Sets the instance's private properties to null

Removes this instance from the DOM Element means removing this instance from the private variable where it was stored and removing the key and id from the actual DOM element.

[Johann-S]

it looks good to me @rohit2sharma95 👍

[XhmikosR]

@Johann-S what's left here?

[Johann-S]

Nothing to me, we're ready to merge 👍

[rohit2sharma95]

These commits are there in PR #31337 though

[XhmikosR]

Not sure what you mean?

…

On Fri, Oct 2, 2020, 12:42 Rohit Sharma ***@***.***> wrote: These commits are there in PR #31780 <#31780> though — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#30838 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AACVLNOQAXAU7WLXAX6TIWTSIWOAHANCNFSM4NBYHZGA> .

[rohit2sharma95]

Sorry, @XhmikosR I mentioned the wrong PR in the comment above. I mean the changes suggested in this PR are kept in the new PR #31337 created by @mdo that is still in progress.

[XhmikosR]

True, let's close this one then.

[XhmikosR]

Actually wait, I'd rather merge this one and rebase #31337.