Re: list strawman

[Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov>]

Florian Weimer <fw@deneb.enyo.de> writes:

> Stephen Leake <stephen.a.leake.1@gsfc.nasa.gov> writes:
> 
> > I've done a priliminary implementation of
> > http://www.telepath.com/dennison/Ted/Containers-Lists-Unbounded.ads.html
> > using SAL lists as the base.
> 
> While working on test cases for a similar list implementation, I
> discovered that it isn't clear what should happen if one tries to
> insert the same list into itself using the iterator-based Insert
> procedure.

Yes, this could easily lead to an infinite loop. I suggest testing for
this case, and raising some exception.

> The specification seems to allow it, but to me, it seems that this
> feature is not useful in practice and just adds unnecessary
> overhead.

Which feature? The whole idea of inserting a list using an iterator?
or just inserting a list into itself?

What overhead? Testing for the special case? That's really small.

-- 
-- Stephe