[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- To: email@example.com.COM
- Subject: code-char bug?
- From: firstname.lastname@example.org
- Date: Mon, 24 Jun 91 13:27:29 -0700
I think I found a small bug in code-char in MCL 2.0b1p3. code-char
takes a positive integer and converts it into a char. In CLtL2 (pg.
382) it says if a corresponding char can't be found, nil is returned.
MCL gives the following instead:
? (code-char 256)
> Error: Argument 256 is not of type (INTEGER 0 255).
> While executing: CODE-CHAR
This may have been missed since the entire old definition of code-char
has a dotted-line beside it. The part with the solid-line just says
that the old bits and font arguments are now defunct. On page xiii it
says that you should be 'wary' of information with a dotted-line
beside it, but given that the solid-line part says nothing about the
behavior when the code argument is out of range, it seems (code-char
256) should still return nil. I just checked MCL 1.3.2 and it also
signals an error, which is definitely wrong. The fix is obvious.