Crazy Formula

[Ackley Bing]

I made this script which resulted in me having to make a wrap() function for numbers that exceed a maximum. I'm not a mathematician, but I wrote this crazy formula myself. It seems to work, but now that I almost finished with my script, it seems to be giving incorrect results. Does anyone know of a standard math equation to do the same thing? Or how I can fix this formula?

// List2DialogButtons() for llDialog use
// example: llDialog(id,"",List2DialogButtons(Gdummylist,Gindex),1);
// Uses "<<" & ">>" dialog buttons for navigation
list Gdummylist=["a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y","z"];
//list Gdummylist=["1","2","3","4","5","6","7","8","9","10","11","12"];
integer Gindex;
list List2DialogButtons(list chain, integer i)
{
    list buttons;
    integer j=llGetListLength(chain);
    if (j>12)
        buttons=
        ["<<"]
        +llList2List(chain,wrap(i+9,j),wrap(i+9,j))
        +[">>"]
        +llList2List(chain,wrap(i+6,j),wrap(i+8,j))
        +llList2List(chain,wrap(i+3,j),wrap(i+5,j))
        +llList2List(chain,wrap(i,j),wrap(i+2,j));
    else
    {
        integer k;
        do buttons+=llList2List(chain,k,k);
        while(++k<j);
    }
    return buttons;
}
integer wrap(integer x,integer max){return (((x<0)*(x+max))+((x>=0&&x<max)*x)+((x>=max)*(x-max)));}
default
{
    state_entry()
    {
        llListen(1,"",llGetOwner(),"");
    }
    touch_start(integer index)
    {
        Gindex=0;
        llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);
    }
    listen(integer channel, string name, key id, string message)
    {
        if(message=="<<"|message==">>")
        {
            integer total=llGetListLength(Gdummylist);
            Gindex=Gindex+((message==">>")*9)-((message=="<<")*9);
            if(Gindex>=total)Gindex=Gindex-total;
            else if(Gindex<0)Gindex=total-Gindex-2;
            llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);
        }
        else
        {
            integer find_index=llListFindList(Gdummylist,[message]);
            if(~find_index)
            {
                //Gindex=find_index;
                //Action(Gindex);
            }
        }
    }
}

[Charlotte Holmeforth]

Is it to get buttons into the right order? Once I have the buttons I want, I use this:

lDialogButtons = llList2List(lDialogButtons,-3,-1)
+ llList2List(lDialogButtons,-6,-4)
+ llList2List(lDialogButtons,-9,-7)
+ llList2List(lDialogButtons,-12,-10);

it makes a new list by re-ordering the buttons in groups of three, starting from the end of the list, so that the last three buttons end up at the top and the first three at the bottom.

Usually I put page navigation buttons on the last row -  << OK >> - then the rest of the buttons line up.

 

[Ackley Bing]

Yes. Also I made the script seamlessly connect the end with the beginning, with no gaps
For example: x,y,z,a,b,c,d,e... instead of just x,y,z.

Ordering is fine:

[a] [b] [c][d] [e] [f][g] [h] [i][<] [j] [>][j] [k] [l][m] [n] [o][p] [q] [r][<] [s] [>]

Up until the end,
where it does this:

[s] [t] [u][v] [w] [x][a] [y] [z][<] [b] [>]

Notice the buttons, a,y,z. The problem is in the wrap(). I will fix it with enough trial and error... I am no mathematician. I was hoping there might be some standard math equation to do this instead of my crazy formula.

integer wrap(integer x,integer max){return (((x<0)*(x+max))+((x>=0&&x<max)*x)+((x>=max)*(x-max)));}

[Charlotte Holmeforth]

I would do the layout like this, so 9 buttons per page, and the last row is all navigation. There's a spacer button after z, to keep the rest of the buttons aligned. If you wanted to wrap to the beginning you could go back to a on the next page.

[a] [c]

[d] [e] [f]

[g] [h]

[|] [ok] [>]

 

[j] [k] [l]

[m] [n] [o]

[p] [q] [r]

[<] [ok] [>]

 

[t]

[v] [w] [x]

[y] [z] [-]

[<] [ok] [|]

 

[Nova Convair]

For this kind of overflow the modulo operation % becomes handy.

list Gdummylist=["a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y","z"];integer Gindex;list List2DialogButtons(list chain, integer i){    list buttons;    integer j=llGetListLength(chain);    if (j>12)        buttons=        ["<<",llList2String(chain,(i+9)%j),">>"]        +llList2List(chain,(i+6)%j,(i+8)%j)        +llList2List(chain,(i+3)%j,(i+5)%j)        +llList2List(chain,i,(i+2)%j);    else    {        integer k;        do buttons+=llList2List(chain,k,k);        while(++k<j);    }    return buttons;}default{    state_entry()    {        llListen(1,"",llGetOwner(),"");    }    touch_start(integer index)    {        Gindex=0;        llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);    }    listen(integer channel, string name, key id, string message)    {        if(message=="<<"|message==">>")        {            integer j=llGetListLength(Gdummylist);            if (message=="<<") Gindex=(Gindex-10)%j;            else if (message==">>") Gindex=(Gindex+10)%j;                        llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);        }        else        {            llOwnerSay(message);            Gindex=0;        }    }}

No need for crazy formulas.

 

[wherorangi]

Nova's use of modulus is the way to go

another issue OP has is when a 3 button bracket spans the bounds. In this case [max-1,0,1] and [max-2,max-1,0]. To catch this a way to mod Nova's code is:

 

list wrap(list c, integer x, integer y){    if (x <= y)        return llList2List(c, x, y);     return llList2List(c, x, -1) + llList2List(c, 0, y);       } ...

list List2DialogButtons(list chain, integer i) {    ... 
    if (j>12)        buttons =            ["<<",llList2String(chain,(i+9)%j),">>"]            + wrap(chain,(i+6)%j,(i+8)%j)            + wrap(chain,(i+3)%j,(i+5)%j)            + wrap(chain,i,(i+2)%j);    else ...
}
...listen(integer channel, string name, key id, string message){   ...

   if (message == "<<")    {   	Gindex -= 10; if (Gindex < 0) Gindex += j;   }   else if (message == ">>") 
        Gindex = (Gindex + 10) % j;

   ...
}

 

[Nova Convair]

---

wherorangi wrote:

Nova's use of modulus is the way to go

another issue OP has is when a 3 button bracket spans the bounds. In this case [max-1,0,1] and [max-2,max-1,0]. To catch this a way to mod Nova's code is:

 

 

---

Nope. llList2List does the wrap. No extra code needed.

llList2List(chain,25,1) will return the list elements: 25,0,1

The extra wrap function is obsolete.

[wherorangi]

try stepping back

[Ackley Bing]

Thanks Nova This a lot less crazy yet i'm still getting button ordering like [a] [y] [z].

[Ackley Bing]

Your wrap does the trick.

Thanks to everyone here is what I finally came up with.

list Gdummylist=["a","b","c","d","e","f","g","h","i","j","k","l","m","n","o","p","q","r","s","t","u","v","w","x","y","z"];integer Gindex;list List2DialogButtons(list chain, integer i){    list buttons;    integer j=llGetListLength(chain);    if (j>12)        buttons =            ["<<",llList2String(chain,(i+9)%j),">>"]            + wrap(chain,(i+6)%j,(i+8)%j)            + wrap(chain,(i+3)%j,(i+5)%j)            + wrap(chain,i,(i+2)%j);    else    {        integer k;        do buttons+=llList2List(chain,k,k);        while(++k<j);    }    return buttons;}list wrap(list c, integer x, integer y){    if(x<=y) return llList2List(c, x, y);    return llList2List(c,x,-1)+llList2List(c,0,y);}default{    state_entry()    {        llListen(1,"",llGetOwner(),"");    }    touch_start(integer index)    {        Gindex=0;        llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);    }    listen(integer channel, string name, key id, string message)    {        if(message=="<<"|message==">>")        {            integer j=llGetListLength(Gdummylist);            if (message=="<<") Gindex=(Gindex-10)%j;            else if (message==">>") Gindex=(Gindex+10)%j;            llDialog(llGetOwner()," ",List2DialogButtons(Gdummylist,Gindex),1);        }        else        {            integer find_index=llListFindList(Gdummylist,[message]);            if(~find_index)            {                llOwnerSay(message);                //actions...            }        }    }}

[Ackley Bing]

The Formula was 100% correct!!

But not needed because SL has % !!!

:catlol:

[wherorangi]

---

Ackley Bing wrote:

i'm still getting button ordering like [a] [y] [z].

---

yes. That was my point

we do have to account for this when ordering brackets/sets of 3, when 10 is not a factor of max (24)

certainly yes tho, modulus like Nova says is the way to do this. [is the base building block]

another way to do the accounting is to swap the list parameter indices (X with Y) when required. Doing this tho requires a test also

to keep the main body as inline code (as wrote originally) then the test has to be done in a external function (typically minmax), and if we are going to conform to this inline coding style then it makes sense for the test function to return the set as well 

 

 eta: base

 

[wherorangi]

---

Ackley Bing wrote:

BTW, your listen event gave me an error on compile :smileywink: j is not declared in the event.

---

correct. Is just p-code that shows the edits. The actual editing of your codes is for you to do (: 

[Ackley Bing]

Thanks again

I gave both of you credit in my final posting on the wiki

https://wiki.secondlife.com/wiki/User:Ackley_Bing/List2DialogButtons

[wherorangi]

is a pretty cool idea this. A circular button menu dialog

i like it

 

[Rolig Loon]

Lovely work. :smileyvery-happy:  I'll add that to my collection of menu managers. 

Here are a couple of others that take different approaches to the same problem.

https://community.secondlife.com/t5/LSL-Library/A-Simple-MultiPage-Dialog-Menu-System/m-p/2126747

https://community.secondlife.com/t5/LSL-Library/Dynamic-Multi-page-Dialog-AKA-Pagination/m-p/708119

 

[Ackley Bing]

Yes. Circular is what i'm looking for because my of Next Idea, which, btw, is sacret, for now.

There is still a bug in this here. When you click thru... it's producing [a][z] and [a][y][z].

[Ackley Bing]

Aw Thank You Rolig

Hey, with enough invention we'll be getting vehicles with 4 different wheels!

[wherorangi]

---

Ackley Bing wrote:

Yes. Circular is what i'm looking for because my of Next Idea, which, btw, is sacret, for now.

There is still a bug in this here. When you click thru... it's producing [a]

[z] and [a][y][z].

---

is that when you go backwards ?

eta:

ps. I let you off ok (:

look again at the edits pcode and work thru why the last edit treats going backwards different to going forwards ?