MathGroup Archive 1992

[Date Index] [Thread Index] [Author Index]

Search the Archive

Finding effective directive of a graphics primitive

  • To: Mathematica user's group <MathGroup at yoda.physics.unc.edu>
  • Subject: Finding effective directive of a graphics primitive
  • From: Robby Villegas <Villegas at knox.bitnet>
  • Date: Wednesday, May 20 1992

Hello,

     I have a nested list of graphics primitives and directives, of the kind
that Graphics and Graphics3D can accept in the first position, and I would
like a function to determine for me what color directive (GrayLevel, RGBColor,
Hue, CMYKColor, or whatever) will apply to a particular Line object, which can
be buried anywhere in the nested list.  As I understand directives (correct me
if I am wrong), if there are directives preceding the Line object in the list
where the Line sits, then the "most recent" one will determine the style of
the line, where "most recent" means closest to the Line in the list ordering,
and preceding the Line.  If there are no directives for a Line primitive in
the list, then if the list itself is an element of a higher list, the elements
preceding it in the higher list are checked to see if any apply to a Line.  If
so, the most recent, again, is the effective one.  So if we had

     {RGBColor[1, 0, 1], GrayLevel[0.8], {Hue[1/3], Line[somepoints]}}

the color Hue[1/3] would override, but if we had

     {RGBColor[1, 0, 1], GrayLevel[0.8], {PointSize[0.02], Line[somepoints]}}

there would be none at the same level as the Line, so the GrayLevel from one
level up would apply, overriding the RGBColor before it.

     And so on, up the levels of the nested expression until something is
found.  I am interested to know an efficient and/or elegant way to locate this
"most recent" color directive (or any aspect of style, for that matter), so
that given any particular Line in the expression (say defined by its
part-specification) I can ask "How will it be rendered, color-wise?" and get
the right answer.  I've thought about using a loop (maybe Until or For) to
backtrack through the elements at each level and ascend the levels until the
first matching directive is found.  But I've wondered if it would be more
efficient or elegant to just generate all the predecessor elements at all
levels and extract the valid Line directives from it (probably with Cases),
taking the last of these to be the one.  Any suggestions on the best
implementation of either of these, or better yet a really simple and elegant
way to locate what I want?

                                              Robby Villegas
                                              Knox College
                                              (Villegas at Knox.Bitnet)





  • Prev by Date: Re:Inputting integer data in binary form
  • Next by Date: Inputting integer data in binary form
  • Previous by thread: Fortran output
  • Next by thread: Mathematica conference schedule