OpenEuphoria

2. Re: Confusing Coding Conundrum

• Posted by Undetermined origin c/o LISTSERV administrator Aug 17, 2000
• 377 views

--------------1349434434EE8BA264FD3604

Travis Beaty wrote:

> Hello All!
>
> As some of you may be aware, I've been working an
> IDE for Euphoria.  Since the project is finally
> starting to come together somewhat, I suppose it
> is time to announce its development publicly.
> The name that I have chosen for it is "Vision
> Euphoria" -- I would really liked to have called
> it "Visual Euphoria," but that name has been
> taken.  It will be a fairly large and complex
> program, and it will not be completed until some
> time in November, thanks to my hectic schedule;
> however, at least now the program is not 99%
> vaporware.  The completed package will be
> shareware, and the price will be under $30 --
> completely honor based (no nag screens or broken
> code).
>
> Vision Euphoria will consist of a core program,
> assisted by various support programs that
> (hopefully) will make programming in win32lib.ew
> and Euphoria in general very easy.  One
> such "support program" is the Euphoria Database
> Editor, which assists in the formation and
> editing of files created using database.e.  It is
> now 80% operational, and should be available for
> download (all support programs will be freeware)
> within the next two weeks.  I have a little bit
> of coding left, as well as some brief
> documentation ... its comprehensive documentation
> will be packaged with Vision Euphoria.  The
> program is very straightforward and simple, and
> even without a 150 page manual, it should be
> understandable to anyone vaguely familiar with
> Euphoria.
>
> I have a confusing problem at this point,
> however, and I'm hoping that somebody out there
> in Euphorialand can point me in the right
> direction.
>
> Within EDB, I have a find feature, which allows
> the user to find either Euphoria strings or
> objects (integers, sequences, atoms, etc.)  I've
> gotten the program to find any object as long as
> it's not nested in a layer of sequences, but I'm
> having a problem accurately displaying its
> location.
>
> The object displayed in the data mle has been
> sprint'ed so that it can be shown.  When the
> program finds a match for the search object, it
> needs to be highlighted (selected).  Thanks to
> information graciously supplied by Mr. Fritz, I
> have a mechanism to highlight.  But I can't
> figure out how to convert the position of the
> match into its position within the sprint'ed
> object.
>
> Example:
>
> TheDarnObject: {23,45,65,71,33,14,19,
> {30,30,32},12,16,4,0}
>
> User wants to find:  16 --------------------------
> --^
>
> Now then, my first approach would be to get the
> index of the matched element, in this case 10,
> then count the commas in the sprinted version.
> One I got to the ninth one, just one position
> over and there it is.  But the nested sequence
> bites me in the backside, as I would need to
> factor in the nested sequence's length.  Not a
> problem for this example.  But what if there are
> dozens of layers of nested sequences?  No good.
>
> If someone can help me out, it would be very
> appreciated!  The answer is more than likely
> right there in front of me, and I'm too flustered
> to see it.  At any rate, as soon as I get this
> problem solved, and throw in some printing code,
> she'll be ready to go!
>
> Travis Beaty
> Claude, Texas
>
> -----
>
> "The fastest line is the one that you're not in."
>
>                               -- Murphy's Law
>
> ______________________________________________________
> Get your free web-based email at http://www.xoom.com
> Birthday? Anniversary? Send FREE animated greeting
> cards for any occasion at http://greetings.xoom.com

You could use a modified version of sprint which returns the starting
and ending indexes of a particular element, or you could search through
the resulting string and keep track of the depth of  nested sequences.
I've attached a modified version of sprint. It probably could be a
little faster or cleaner, but it works (included are a few tests). Note
that it still requires sprint.

Jeff Fielding
JJProg at cyberbury.net

--------------1349434434EE8BA264FD3604
 name="sprint2.e"
Content-Disposition: inline;
 filename="sprint2.e"

include misc.e
global function sprint2(object x, object index)
-- Return the string representation of any Euphoria data object.
-- This is the same as the output from print(1, x) or '?', but it's
-- returned as a string sequence rather than printed.
-- modified by Jeff Fielding to return the indexes of the object too
    sequence s, temp
    integer startIndex, endIndex

    if atom(x) then
        s = sprintf("%.10g",x)
        return {s, 1, length(s)}
    else
        s = "{"
        if sequence(index) then
                for i = 1 to length(x) do
                        if index[1] = i then
                                temp = index[2..length(index)]
                                if length(temp) then
                                    temp = sprint2(x[i],temp)
                                    startIndex = length(s)+temp[2]
                                    endIndex = length(s)+temp[3]
                                    s &= temp[1]
                                else
                                    startIndex = length(s)+1
                                    s &= sprint(x[i])
                                    endIndex = length(s)
                                end if
                        else
                                s &= sprint(x[i])
                        end if
                        if i < length(x) then
                                s &= ','
                        end if
                end for
        else
                for i = 1 to length(x) do
                        if index = i then
                                startIndex = length(s)+1
                                temp = sprint(x[i])
                                endIndex = length(s)+length(temp)
                                s &= temp
                        else
                                s &= sprint(x[i])
                        end if
                        if i < length(x) then
                                s &= ','
                        end if
                end for
        end if
        s &= "}"
        return {s, startIndex, endIndex}
    end if
end function
constant C = {1,2,{3,{4}}}
procedure p(object index)
        sequence t
        puts(1,"index: ")
        print(1,index)
        t = sprint2(C,index)
        puts(1,"\n" & t[1] & "\n")
        for i = 1 to t[2]-1 do
                puts(1,' ')
        end for
        for i = t[2] to t[3] do
                puts(1,'^')
        end for
        puts(1,'\n')
end procedure
p(1)
p(2)
p(3)
p({1})
p({3,1})
p({3,2})
p({3,2,1})

--------------1349434434EE8BA264FD3604--