Re: bug(s) in 2.5.19 skel.c

[Bruce Lilly]

Bruce Lilly wrote:
> Having code which is known to work on all existing
> implementations *and* which is robust w.r.t. maintenance is
> far preferable to code designed to accomodate some unspecified
> hypothetical architecture which nobody can provide an example
> of, but which is quite fragile as the code evolves (or w.r.t.
> options and pattern specifics).

And it is noteworthy that the bug which started this thread
involved explicit initialization (as suggested by Hans) where
the initialization did not account for changes in the structure
definition.