compile vs runtime

[Programmingkid]

I am trying to make a word that does one thing at compile time, and does another thing at runtime. I think this is how you do it:

: myword 
." one"
does>  ." two"
cr ;

When I first use this word, I see "one". Then when I use this word again, I see "two", but it causes OpenBIOS to panic because of a segmentation violation at 3. Is this an OpenBIOS bug, or did I do something wrong?

[Tarl Neustaedter]

On 2010-3-31 4:07 PM, Programmingkid wrote:
> I am trying to make a word that does one thing at compile time, and does another thing at runtime. I think this is how you do it:
>
> : myword
> ." one"
> does>   ." two"
> cr ;
>
> When I first use this word, I see "one". Then when I use this word again, I see "two", but it causes OpenBIOS to panic because of a segmentation violation at 3. Is this an OpenBIOS bug, or did I do something wrong?

That's not how you want to be using does> (in my opinion, you don't want 
to be using does> at all, but that's just my dislike of the construct).

If you want to play games with what's done at compile vs runtime, you 
should be using the "compile" and "[compile]" tokens.

[Stefan Reinauer]

On 3/31/10 10:07 PM, Programmingkid wrote:
> I am trying to make a word that does one thing at compile time, and does another thing at runtime. I think this is how you do it:
>
> : myword 
> ." one"
> does>  ." two"
> cr ;
>
> When I first use this word, I see "one". Then when I use this word again, I see "two", but it causes OpenBIOS to panic because of a segmentation violation at 3. Is this an OpenBIOS bug, or did I do something wrong?

you can also do

: mywork
state @ if
." compile"
else
." interpret"
then
; immediate

or so

[Segher Boessenkool]

> I am trying to make a word that does one thing at compile time, and  
> does another thing at runtime.

: my-weird-word  [ do-stuff-at-compile-time ] do-stuff-at-run-time ;

Segher