Re: [PATCH] Marking weak alist vectors

[Han-Wen Nienhuys]

Marius Vollmer wrote:
> > I may have struck that or something similar a while back.  Mikael
> > explained it was a reference held in the print data of the output
> > port, or something.  Used to detect cyclic structures, or something.
> > I don't think I understood why such a reference ought to persist once
> > outside the print, I'd suspect it'd be better if it didn't, if that
> > could be arranged.
>
>
> If I remember right, I have been in exactly this spot as well, but I
> couldn't really fix this.  I don't remember why, unfortunately...

Isn't this a matter of clearing the print state at the end of scm_prin1() ?

--
 Han-Wen Nienhuys - address@hidden - http://www.xs4all.nl/~hanwen