Re: Uninitialized SCM variables

[Reinhold Kainhofer]

Am Thursday, 18. August 2011, 11:45:25 schrieb David Kastrup:
> Dan Eble <address@hidden> writes:
> > Backing up… I believe the compiler will initialize the bits in the
> > aforementioned variables to zero, but is zero a desirable default for
> > SCM variables in general, and these in particular?
> > 
> > It also just sank in that in another thread there was a statement that
> > treating a SCM as a boolean is "very wrong".  That would include a
> > number of lines in ly_property_lookup_stats and note_property_access
> > that use these variables.
> 
> I have not looked at the respective code yet (short of time right now)
> but the names of those functions suggests that they might be looking at
> those variables not as much as Scheme values but rather as internals
> related to the operation of the Scheme interpreter.  

Yes, the checks are not checks whether the scheme value is true or not, but 
they are checks whether the SCM variable has been set to some value or is 
still uninitialized... It's basically the same as the !cached we had a few 
days ago.

Cheers,
Reinhold

-- 
------------------------------------------------------------------
Reinhold Kainhofer, address@hidden, http://reinhold.kainhofer.com/
 * Financial & Actuarial Math., Vienna Univ. of Technology, Austria
 * http://www.fam.tuwien.ac.at/, DVR: 0005886
 * LilyPond, Music typesetting, http://www.lilypond.org