Re: pfDeleteFunc(3pf)

New Message Reply Date view Thread view Subject view Author view

Marcus Barnes (marcus++at++multigen.com)
Mon, 21 Sep 1998 13:16:08 -0700

On Sep 18, 7:49am, Steve Baker wrote:
>I agree with you in that there may be many other reasons for needing a
>pfDeleteFunc that don't entail userdata. (There must be or you and I
>would never have tripped over this problem!)

The problem is resource ownership. Performer has a few holes in this regard.
One that is problematic for loaders is that pfNodeTravData is not owned by
pfNode. Any data attached as TravData must find other means to insure deletion.

>The same problem popped up with some of the other per-node user callbacks
>like maybe pfCopyFunc.

Yes ... for example, TravData is shallow copied only.

>I *thought* that these problems had been addressed - but my memory is
>kinda fuzzy on this.

Nope. However, pfuAutoList was added to Performer 2.2 ... It helps enforce and
extend ownership wherever Performer already has proper ownership semantics.

Regards.

--
+ Marcus Barnes, Technical Staff        mailto:marcus++at++multigen.com +
+ Multigen Inc.                         http://www.multigen.com    +
+ 550 S. Winchester Blvd.               phoneto:1-408-367-2654     +
+ Suite 500 San Jose CA 95128           faxto:1-408-261-4103       +

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Mon Sep 21 1998 - 13:13:39 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.