API Question DetachChild

[GLS]

	 /**
	 * \brief Break an existing attachment to a child.
	 * \param attachment attachment handle
	 * \param vel separation velocity [m/s]
	 * \return \e true when detachment is successful, \e false if no child was
	 *   attached, or [COLOR="Red"]if child refuses to detach[/COLOR].
	 * \sa CreateAttachment, SetAttachmentParams, GetAttachmentParams,
	 *   GetAttachmentId, GetAttachmentStatus, AttachmentCount, GetAttachmentIndex,
	 *   GetAttachmentHandle, AttachChild
	 */
	bool DetachChild (ATTACHMENTHANDLE attachment, double vel = 0.0) const;

How would that be achieved? I looked in all of the API and there appears to be nothing to do this with.
There is a clbkDockEvent(), which (an "attachment version" of that callback) would probably be a good way to allow the child to do the check, but that callback returns void instead of bool, so this can't even be done to the docking ports...

 

[kuddel]

Not sure how this was meant to be, but what happens in this scenario:

A (parent) and B(child) are docked,
then B deletes the docking port [via DelAttachment(...)]
and A tries to detach [via DetachChild(...)] using a previously[*] stored ATTACHMENTHANDLE (which it shouldn't do).


[*] "previously" as in before the detaching attempt of child B
​

Could that lead to a negative (false) return?

On the other hand, I agree that there doesn't seem to be a clean interface to reject/refuse detaching of a child or parent.

 

[GLS]

Not sure how this was meant to be, but what happens in this scenario:

A (parent) and B(child) are docked,
then B deletes the docking port [via DelAttachment(...)]
and A tries to detach [via DetachChild(...)] using a previously[*] stored ATTACHMENTHANDLE (which it shouldn't do).


[*] "previously" as in before the detaching attempt of child B
​

Could that lead to a negative (false) return?

Without testing, I'd say that would either CTD or return false.

 

[kuddel]

I would guess that the "child refuse to detach" part is a left-over from an idea Martin had...
Current HEAD version of Orbiter BETA repository also doesn't have any additional API regarding attachments, so I guess it's just a "misleading" comment.

 

[GLS]

I would guess that the "child refuse to detach" part is a left-over from an idea Martin had...

... or never finished... :shrug: