amforth-devel

[Amforth] lib/bitnames.frt

[Enoch <ix...@ho...>]

Hi,

Isn't the following more sensible?

 : is_low? ( pinmask portaddr -- f)
-    c@ invert and
+    c@ and 0=
 ;
 
 : is_high? ( pinmask portaddr -- f)
-    c@ and
+    c@ and 0>
 ;

Thanks, Enoch.

1) the writeup says they return "true if..." which means return -1 and
not a bitmask.

2) true/false values are also more convenient when analyzing multiple
port inputs.

Re: [Amforth] lib/bitnames.frt

[Matthias T. <mt...@we...>]

Hi Enoch,

> Isn't the following more sensible?
> 
>  : is_low? ( pinmask portaddr -- f)
> -    c@ invert and
> +    c@ and 0=
>  ;
>  
>  : is_high? ( pinmask portaddr -- f)
> -    c@ and
> +    c@ and 0>
>  ;
> 
> Thanks, Enoch.
> 
> 1) the writeup says they return "true if..." which means return -1 and
> not a bitmask.

technically only FALSE is specified: 0. TRUE is everything else. But
I agree, that  a flag should be wellformed. Do your changes work in all
cases?

Matthias

Re: [Amforth] lib/bitnames.frt

[Enoch <ix...@ho...>]

Hello Matthias,

Matthias Trute <mt...@we...> writes:
> Hi Enoch,
>
>> Isn't the following more sensible?
>> 
>>  : is_low? ( pinmask portaddr -- f)
>> -    c@ invert and
>> +    c@ and 0=
>>  ;
>>  
>>  : is_high? ( pinmask portaddr -- f)
>> -    c@ and
>> +    c@ and 0>
>>  ;
>> 
>> Thanks, Enoch.
>> 
>> 1) the writeup says they return "true if..." which means return -1 and
>> not a bitmask.
>
> technically only FALSE is specified: 0. TRUE is everything else. But
> I agree, that  a flag should be wellformed. Do your changes work in all
> cases?

Yes, so far :-)

According to "Forth 2012 RC1" 3.1.3.1 Flags, standard words should
return -1 or 0 only. I was surprised to find that emit? returns the
bitmask 32 rather than -1 (what caused me trouble with bitnames.frt
implementation).

Thanks, Enoch.