[augeas-devel] Towards a stable interface for libfa

[David Lutterkort]

Hi Nate,

On Sun, 2009-03-22 at 13:38 -0400, Nate Foster wrote:
> Have you considered generalizing the code to elimiate this
> precondition for fa_ambig_example?
> 
> > * Neither the language of FA1 or of FA2 may contain words with the
> > * characters '\001' and '\002', as they are used during construction of
> > * the ambiguous word.

I would love to, but don't have any good ideas how. I _think_ that the
precondition is overly cautious - Anders doesn't have anything similar
in his implementation (he uses \u0000 and \u0001 as the markers)

Augeas violates this precondition all over the place (it's quite common
to have lenses whose ctype uses all 8bit characters in its alphabet),
without any adverse effects. I haven't completely convinced myself
though that the precondition is unnecessary.

any thoughts ?
David