- From: David Robinson <drtr1@cam.ac.uk>
- Date: Wed, 22 Mar 95 17:08 GMT
- To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Somewhat of a detail, this, but anyway:
In section 5.4.1 Accept, you say:
float = < ANSI-C floating point text representation,
where (0.0 < float < 1.0) >
What, precisely, do you mean by this? You don't give a reference, so I could
imagine the following:
1. The representation of a floating-point number (float, double or
long double) as a token in a conforming ANSI C program, or
2. the representation of a floating-point number produced by the ANSI printf
routine - but which format? %g, %e, %f??
3. the representation of a floating-point number accepted by an ANSI scanf
routine.
I even wonder whether 'floating point text representation' means the text
representation of a floating-point number, or the floating-point (text)
representation of (any) number.
In particular, does it allow:
q=1e-4
q=0.4f
q=1.343e-1l
?
And, perhaps more seriously, it currently forbids q=0. and q=1.
David Robinson. (drtr1@cam.ac.uk)
Received on Wednesday, 22 March 1995 09:28:14 UTC