Re: [Gcl-devel] string is not vector of character

[Paul F. Dietz]

Camm Maguire wrote:

> Greetings!  cmucl and clisp both seem to fail this too.  Is string
> supposed to be `(or (vector standard-char) (vector base-char) (vector
> extended-char) (vector character))?

The three tests string-is-not-vector-of-character.[135]
have the note :nil-vectors-are-strings.  This note indicates
the test is assuming that (vector nil) is a valid type,
that (upgraded-array-element-type nil) == nil, and that
(vector nil) is a subtype of string.

What this means is that you can't assume that every string
has the same underlying element type, since there are at
least two cases that must differ: (vector character)
and (vector nil).

This is not terribly important to implement.  If you want
to disable these tests, add

#+gcl (rt::disable-note :nil-vectors-are-strings)

somewhere (the end of notes.lsp, say).


        Paul