Re: [Changeset] isa function with float and numeric arguments

[David Bateman]

John W. Eaton wrote:
> On 12-May-2008, Bill Denney wrote:
>
> | David Bateman wrote:
> | > David Bateman wrote:
> | >   
> | >> The isa function should allow the "float" and "numeric" arguments. The
> | >> attached patch adds this.
> | >>
> | >> D.
> | > Please use this version instead.
> | >
> | > D.
> | >   
> | Wouldn't this be easier to read (and marginally faster) if written as:
> | 
> | if (strcmp (cname, "float"))
> |   retval = any (strcmp (class (x), {"double" "single"}));
> | elseif (strcmp (cname, "fnumeric"))
> |   classes = {"double" "single" "uint8" "uint16" "uint32" "uint64" "int8" 
> "int16" "int32" "int64"};
> |   retval = any (strcmp (class (x), classes));
> | else
> |   retval = strcmp (class (x), cname);
> | endif
>
> I made this change, and used persistent variables for the cell arrays
> to avoid the need for reinitializing them on each call.
>   
The version in the repository seems to be mine.. You don't seem to have
committed the changes that make the classes variable persisent..

D.

-- 
David Bateman                                address@hidden
Motorola Labs - Paris                        +33 1 69 35 48 04 (Ph) 
Parc Les Algorithmes, Commune de St Aubin    +33 6 72 01 06 33 (Mob) 
91193 Gif-Sur-Yvette FRANCE                  +33 1 69 35 77 01 (Fax) 

The information contained in this communication has been classified as: 

[x] General Business Information 
[ ] Motorola Internal Use Only 
[ ] Motorola Confidential Proprietary