Loading ...
Sorry, an error occurred while loading the content.

Re: 3.6 extra lines nohidden3d

Expand Messages
  • crawford richard
    ... Would it not be better in the long run to make this sort of thing an option -- probably on the set hidden3d command -- rather than requiring
    Message 1 of 4 , Apr 16 9:54 AM
    View Source
    • 0 Attachment
      David Denholm wrote:

      > Anderson Mills (nodog@...) wrote:
      > > Is there a way to get rid of the diagonal lines on each grid square when
      > > using nohidden3d?
      >
      > > I'm using...
      > > unix version 3.5 (pre 3.6)
      > > patchlevel beta 328
      >
      > I think it's a compile-time option : look in hidden3d.c
      >
      > /* HBB 961212: this #define lets you choose if the diagonals that
      > * divide each original quadrangle in two triangles will be drawn
      > * visible or not: do draw them, define it to be 7L, otherwise let be
      > * 3L (for the FOUR_TRIANGLES method, the values are 7L and 1L) */
      > #ifndef TRIANGLE_LINES_TO_DRAW
      > #define TRIANGLE_LINES_TO_DRAW 7L
      > #endif


      Would it not be better in the long run to make this sort of thing
      an option -- probably on the "set hidden3d" command -- rather than
      requiring recompilation? I realize that for testing and debugging
      the #ifdef scheme is handy, but ...


      Dick Crawford, aka rccrawford@...

      [[[[ unsubscribe from info-gnuplot-beta via majordomo@... ]]]]
    • Hans-Bernhard Broeker
      ... Of course, I considered that as well, but kept away from it. I didn t want to break the feature freeze we, errr..., used to have ... , but I did want to
      Message 2 of 4 , Apr 17 9:55 AM
      View Source
      • 0 Attachment
        On Wed, 16 Apr 1997, crawford richard wrote:

        > Would it not be better in the long run to make this sort of thing
        > an option -- probably on the "set hidden3d" command -- rather than
        > requiring recompilation? I realize that for testing and debugging
        > the #ifdef scheme is handy, but ...

        Of course, I considered that as well, but kept away from it. I didn't want
        to break the feature freeze we, errr..., used to have ... , but I did want
        to test as many new features as I could. And most of these options aren't
        really very old yet, so even if had changed the 'set hidden3d' syntax,
        they might not have been included in that change yet.

        With the #defines, I could do whatever I wanted without changing
        anything outside of hidden3d.c, so I could send a modified hidden3d.c
        to anyone who wanted to have it, at any time (not that too many asked
        for it ... :-)

        And I also wasn't too sure I could hack at the command line parsing code
        without breaking something. I kept hoping someone else would want to do
        that for me, I think.

        But you're correct, of course, Dick: sooner or later, 'set hidden3d'
        will need to get a syntax like:

        set hidden3d [[no]diag] [[no]clip] [{bottomstyle|bottomadd} {linestyle}]

        where the standard beta330 version would correspond to:

        set hidden3d diag clip bottomadd 1

        Opinions, anyone? Maybe even a volunteer?

        David, if you think this to be a must-have for your announced release,
        please tell me as early as possible: although I'm not at my University
        currently (shift duty at the H1 detector in Hamburg), I'ld then implement
        at least the parts of these 'set' options inside hidden3d.c in as short a
        time as I can manage, and maybe even hack the parser as well, if I can.

        Bye for now

        HBB

        [[[[ unsubscribe from info-gnuplot-beta via majordomo@... ]]]]
      • crawford richard
        ... I understand your motivation for operating this way -- and it s probably what I would have done myself had I the courage to tackle a project as large as
        Message 3 of 4 , Apr 17 11:02 AM
        View Source
        • 0 Attachment
          On Thu, 17 Apr 1997, Hans-Bernhard Broeker wrote:

          > On Wed, 16 Apr 1997, crawford richard wrote:
          >
          > > Would it not be better in the long run to make this sort of thing
          > > an option -- probably on the "set hidden3d" command -- rather than
          > > requiring recompilation? I realize that for testing and debugging
          > > the #ifdef scheme is handy, but ...
          >
          > Of course, I considered that as well, but kept away from it. I didn't want
          > to break the feature freeze we, errr..., used to have ... , but I did want
          > to test as many new features as I could. And most of these options aren't
          > really very old yet, so even if had changed the 'set hidden3d' syntax,
          > they might not have been included in that change yet.
          >
          > With the #defines, I could do whatever I wanted without changing
          > anything outside of hidden3d.c, so I could send a modified hidden3d.c
          > to anyone who wanted to have it, at any time (not that too many asked
          > for it ... :-)

          I understand your motivation for operating this way -- and it's probably
          what I would have done myself had I the courage to tackle a project as
          large as rewriting "hidden3d.c". I wasn't complaining -- I was just
          interested in your long-term intentions.

          > And I also wasn't too sure I could hack at the command line parsing code
          > without breaking something. I kept hoping someone else would want to do
          > that for me, I think.
          >
          > But you're correct, of course, Dick: sooner or later, 'set hidden3d'
          > will need to get a syntax like:
          >
          > set hidden3d [[no]diag] [[no]clip] [{bottomstyle|bottomadd} {linestyle}]
          >
          > where the standard beta330 version would correspond to:
          >
          > set hidden3d diag clip bottomadd 1
          >
          > Opinions, anyone? Maybe even a volunteer?
          >

          Well, I've already had one user express a dislike of the diagonals...

          Do we need a 'linestyle' option on "set hidden3d"? (And, if we have it,
          should it have the same type/thickness options as does "splot"?) How would
          that be distinct from the same option on "splot" itself? In particular, how
          would the sequences

          set hidden3d 7
          splot "file1" lt 3, "file2" lt 5

          and

          splot "file1" lt 3, "file2" lt 5

          be different? Or am I missing something here?

          BTW, for those of us who want to play with the options using the current
          #ifdef scheme, what are the compiler flags and what do they do?


          Dick Crawford, aka rccrawford@...

          [[[[ unsubscribe from info-gnuplot-beta via majordomo@... ]]]]
        • David Denholm
          ... Perhaps the out-of-the-box #defines should look like 3.5 as much as possible, except for the handling of undefined and perhaps out-of-range points. ...
          Message 4 of 4 , Apr 17 12:01 PM
          View Source
          • 0 Attachment
            > On Thu, 17 Apr 1997, Hans-Bernhard Broeker wrote:
            >
            > > On Wed, 16 Apr 1997, crawford richard wrote:
            > >
            > > > Would it not be better in the long run to make this sort of thing
            > > > an option -- probably on the "set hidden3d" command -- rather than
            > > > requiring recompilation? I realize that for testing and debugging
            > > > the #ifdef scheme is handy, but ...
            > >
            > > Of course, I considered that as well, but kept away from it. I didn't want
            > > to break the feature freeze we, errr..., used to have ... , but I did want
            > > to test as many new features as I could. And most of these options aren't
            > > really very old yet, so even if had changed the 'set hidden3d' syntax,
            > > they might not have been included in that change yet.
            > >
            > > With the #defines, I could do whatever I wanted without changing
            > > anything outside of hidden3d.c, so I could send a modified hidden3d.c
            > > to anyone who wanted to have it, at any time (not that too many asked
            > > for it ... :-)
            >
            > I understand your motivation for operating this way -- and it's probably
            > what I would have done myself had I the courage to tackle a project as
            > large as rewriting "hidden3d.c". I wasn't complaining -- I was just
            > interested in your long-term intentions.
            >


            > Well, I've already had one user express a dislike of the diagonals...

            Perhaps the out-of-the-box #defines should look like 3.5 as much as possible,
            except for the handling of undefined and perhaps out-of-range points.


            >
            > BTW, for those of us who want to play with the options using the current
            > #ifdef scheme, what are the compiler flags and what do they do?

            There are some comments in hidden3d.c



            dd
            --
            david.denholm@...
            Tel +44 (0)1494 453376 (work) +44 (0)1494 459742 (home)



            [[[[ unsubscribe from info-gnuplot-beta via majordomo@... ]]]]
          Your message has been successfully submitted and would be delivered to recipients shortly.