Re: [Factor-talk] Macros needed?

[Björn Lindqvist]

I think the key is to use the nth word instead of first and second.
Then do it like this:

  { "a" 2 3 } { "b" 5 6 } 2 random 0 = 0 1 ? rot over [ swap nth ] 2bi@

In your code it would be:

  dim chart chart-axes vert? 0 1 ? rot over [ swap nth ] 2bi@

Or

  dim chart chart-axes vert? 0 1 ? [ swap nth ] keep rot nth

Or even:

  dim chart chart-axes vert? 0 1 ? '[ _ swap nth ] bi@


2017-02-04 2:31 GMT+01:00 Alexander Ilin :
> That's great!
>
>   I did this:
>
> : ?[x/y] ( ? -- quot )
> [ x ] [ y ] ? [ call( a -- b ) ] curry ; inline
>
> ...
> dim chart chart-axes vert? ?[x/y] bi@
>
> 04.02.2017, 04:29, "John Benediktsson" :
>> Well I'm not sure about the compiler effect declaration, but you could do 
>> this:
>>
>> vert? [ x ] [ y ] ? [ call( a -- b ) ] curry
>>
>>>  On Feb 3, 2017, at 5:04 PM, Alexander Ilin  wrote:
>>>
>>>  vert? [ x ] [ y ] ? check-effect( a -- b )
>>
>> --
>> Check out the vibrant tech community on one of the world's most
>> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
>> ___
>> Factor-talk mailing list
>> Factor-talk@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/factor-talk
>
> ---=---
>  Александр
>
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> ___
> Factor-talk mailing list
> Factor-talk@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/factor-talk



-- 
mvh/best regards Björn Lindqvist

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

Re: [Factor-talk] Macros needed?

[Alexander Ilin]

That's great!

  I did this:

: ?[x/y] ( ? -- quot )
[ x ] [ y ] ? [ call( a -- b ) ] curry ; inline

...
dim chart chart-axes vert? ?[x/y] bi@

04.02.2017, 04:29, "John Benediktsson" :
> Well I'm not sure about the compiler effect declaration, but you could do 
> this:
>
> vert? [ x ] [ y ] ? [ call( a -- b ) ] curry
>
>>  On Feb 3, 2017, at 5:04 PM, Alexander Ilin  wrote:
>>
>>  vert? [ x ] [ y ] ? check-effect( a -- b )
>
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> ___
> Factor-talk mailing list
> Factor-talk@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/factor-talk

---=--- 
 Александр

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

Re: [Factor-talk] Macros needed?

[John Benediktsson]

Well I'm not sure about the compiler effect declaration, but you could do this:

vert? [ x ] [ y ] ? [ call( a -- b ) ] curry



> On Feb 3, 2017, at 5:04 PM, Alexander Ilin  wrote:
> 
> vert? [ x ] [ y ] ? check-effect( a -- b )

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

Re: [Factor-talk] Macros needed?

[Alexander Ilin]

Hi, John! The trouble in my case is -- I'd have to specify the stack effect for a call that's hidden deep inside the bi@ implementation. I'm trying to refrain from inventing my own implementation of `bi@(`. Is there a way to tell the compiler about the effect of a quotation on stack, so it would perform the run-time check, and take that into consideration for the further analysis? Something like this: vert? [ x ] [ y ] ? check-effect( a -- b ) bi@ Is there a way to teach the compiler that both input quotations to the `?` have the same stack effect, therefore it wouldn't matter for the analysis which one is selected. Bjourne? 04.02.2017, 03:42, "John Benediktsson" :You just need to tell Factor the expected stack effect. See, this fails to compile, because the quotation is only selected at run-time and the compiler isn't smart enough to examine both:     : foo ( seq -- seq' )        dup length even? [ 1 - ] [ 2 + ] ? map ; But, you can always do something like this where you tell Factor to call the quotation with the expected stack effect:     : foo ( seq -- seq' )        dup length even? [ 1 - ] [ 2 + ] ?        '[ _ call( elt -- elt' ) ] map ; I think that might incur some run-time performance penalties because it would check the stack effect on every call, which doesn't matter for your use-case, but might if it was inside a hot loop. Best,John.  On Fri, Feb 3, 2017 at 4:25 PM, Alexander Ilin  wrote:Hello!  I'm developing my chart gadget. I want to achieve this:ALIAS: x firstALIAS: y second: chart-axes ( chart -- seq )    [ dim>> ] [ axes>> ] bi [        nip    ] [        [ 0 swap 2array ] map    ] if* ;M: axis draw-gadget*    dup parent>> dup chart? [| axis chart |        axis vertical?>> :> vert?        chart dim>> :> dim        dim chart chart-axes vert? [ [ x ] bi@ ] [ [ y ] bi@ ] if        ! etc...    ] [ 2drop ] if ;  I thought I found a clever way to simplify that `if` there:        dim chart chart-axes vert? [ x ] [ y ] ? bi@  But the compiler says that I can't use `call` on runtime-computed quotations. Is there a way to work around that, like, by using a MACRO: or something?  Or is this a hard limitation on the available abstraction of the computation?  Basically, depending on a boolean flag I need to either take the first or the second element of the two arrays on the stack, and place the taken elements on the stack in the same order.---=--- Александр--Check out the vibrant tech community on one of the world's mostengaging tech sites, SlashDot.org! http://sdm.link/slashdot___Factor-talk mailing listFactor-talk@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/factor-talk,--Check out the vibrant tech community on one of the world's mostengaging tech sites, SlashDot.org! http://sdm.link/slashdot,___Factor-talk mailing listFactor-talk@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/factor-talk  ---=---Александр --
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

Re: [Factor-talk] Macros needed?

[John Benediktsson]

You just need to tell Factor the expected stack effect.

See, this fails to compile, because the quotation is only selected at
run-time and the compiler isn't smart enough to examine both:

: foo ( seq -- seq' )
dup length even? [ 1 - ] [ 2 + ] ? map ;

But, you can always do something like this where you tell Factor to call
the quotation with the expected stack effect:

: foo ( seq -- seq' )
dup length even? [ 1 - ] [ 2 + ] ?
'[ _ call( elt -- elt' ) ] map ;

I think that might incur some run-time performance penalties because it
would check the stack effect on every call, which doesn't matter for your
use-case, but might if it was inside a hot loop.

Best,
John.


On Fri, Feb 3, 2017 at 4:25 PM, Alexander Ilin  wrote:

> Hello!
>
>   I'm developing my chart gadget. I want to achieve this:
>
> ALIAS: x first
> ALIAS: y second
>
> : chart-axes ( chart -- seq )
> [ dim>> ] [ axes>> ] bi [
> nip
> ] [
> [ 0 swap 2array ] map
> ] if* ;
>
>
> M: axis draw-gadget*
> dup parent>> dup chart? [| axis chart |
> axis vertical?>> :> vert?
> chart dim>> :> dim
> dim chart chart-axes vert? [ [ x ] bi@ ] [ [ y ] bi@ ] if
> ! etc...
> ] [ 2drop ] if ;
>
>   I thought I found a clever way to simplify that `if` there:
>
> dim chart chart-axes vert? [ x ] [ y ] ? bi@
>
>   But the compiler says that I can't use `call` on runtime-computed
> quotations. Is there a way to work around that, like, by using a MACRO: or
> something?
>
>   Or is this a hard limitation on the available abstraction of the
> computation?
>
>   Basically, depending on a boolean flag I need to either take the first
> or the second element of the two arrays on the stack, and place the taken
> elements on the stack in the same order.
>
> ---=---
>  Александр
>
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot
> ___
> Factor-talk mailing list
> Factor-talk@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/factor-talk
>
--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

[Factor-talk] Macros needed?

[Alexander Ilin]

Hello!

  I'm developing my chart gadget. I want to achieve this:

ALIAS: x first
ALIAS: y second

: chart-axes ( chart -- seq )
[ dim>> ] [ axes>> ] bi [
nip
] [
[ 0 swap 2array ] map
] if* ;


M: axis draw-gadget*
dup parent>> dup chart? [| axis chart |
axis vertical?>> :> vert?
chart dim>> :> dim
dim chart chart-axes vert? [ [ x ] bi@ ] [ [ y ] bi@ ] if
! etc...
] [ 2drop ] if ;

  I thought I found a clever way to simplify that `if` there:

dim chart chart-axes vert? [ x ] [ y ] ? bi@

  But the compiler says that I can't use `call` on runtime-computed quotations. 
Is there a way to work around that, like, by using a MACRO: or something?

  Or is this a hard limitation on the available abstraction of the computation?

  Basically, depending on a boolean flag I need to either take the first or the 
second element of the two arrays on the stack, and place the taken elements on 
the stack in the same order.

---=--- 
 Александр

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk