Plot coordinate pairs: Difference between revisions
(add gnuplot example) |
m (→{{header|C}}: link) |
||
Line 131: | Line 131: | ||
}</lang> |
}</lang> |
||
No one would use the previous code to produce a plot (that looks [http://i40.tinypic.com/f2t0l0.png this way]; instead, normally we produce data through a program, then we plot the data using e.g. gnuplot or other powerful tools; the result (with GNUplot and without enhancement) could look [http://i41.tinypic.com/2qivbsn.png like this] instead. |
No one would use the previous code to produce a plot (that looks [http://i40.tinypic.com/f2t0l0.png this way]; instead, normally we produce data through a program, then we plot the data using e.g. [[Plot x, y arrays#gnuplot|gnuplot]] or other powerful tools; the result (with GNUplot and without enhancement) could look [http://i41.tinypic.com/2qivbsn.png like this] instead. |
||
=={{header|gnuplot}}== |
=={{header|gnuplot}}== |
Revision as of 17:41, 15 May 2009
You are encouraged to solve this task according to the task description, using any language you may know.
Plot a function represented as `x', `y' numerical arrays.
Post link to your resulting image for input arrays (see Example section for Python language on Query Performance page):
x = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}; y = {2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0};
This task is intended as a subtask for Measure relative performance of sorting algorithms implementations.
C
We could use the suite provided by Raster graphics operations, but those functions lack a facility to draw text.
<lang c>#include <stdio.h>
- include <stdlib.h>
- include <math.h>
- include <plot.h>
- define NP 10
double x[NP] = {0, 1, 2, 3, 4, 5, 6, 7, 8, 9}; double y[NP] = {2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0};
void minmax(double *x, double *y, double *minx, double *maxx, double *miny, double *maxy, int n) {
int i;
*minx = *maxx = x[0]; *miny = *maxy = y[0]; for(i=1; i < n; i++) { if ( x[i] < *minx ) *minx = x[i]; if ( x[i] > *maxx ) *maxx = x[i]; if ( y[i] < *miny ) *miny = y[i]; if ( y[i] > *maxy ) *maxy = y[i]; }
}
/* likely we must play with this parameter to make the plot looks better
when using different set of data */
- define YLAB_HEIGHT_F 0.1
- define XLAB_WIDTH_F 0.2
- define XDIV (NP*1.0)
- define YDIV (NP*1.0)
- define EXTRA_W 0.01
- define EXTRA_H 0.01
- define DOTSCALE (1.0/150.0)
- define MAXLABLEN 32
- define PUSHSCALE(X,Y) pl_fscale((X),(Y))
- define POPSCALE(X,Y) pl_fscale(1.0/(X), 1.0/(Y))
- define FMOVESCALE(X,Y) pl_fmove((X)/sx, (Y)/sy)
int main() {
int plotter, i; double minx, miny, maxx, maxy; double lx, ly; double xticstep, yticstep, nx, ny; double sx, sy; char labs[MAXLABLEN+1];
plotter = pl_newpl("png", NULL, stdout, NULL); if ( plotter < 0 ) exit(1); pl_selectpl(plotter); if ( pl_openpl() < 0 ) exit(1);
/* determines minx, miny, maxx, maxy */ minmax(x, y, &minx, &maxx, &miny, &maxy, NP);
lx = maxx - minx; ly = maxy - miny; pl_fspace(floor(minx) - XLAB_WIDTH_F * lx, floor(miny) - YLAB_HEIGHT_F * ly,
ceil(maxx) + EXTRA_W * lx, ceil(maxy) + EXTRA_H * ly);
/* compute x,y-ticstep */ xticstep = (ceil(maxx) - floor(minx)) / XDIV; yticstep = (ceil(maxy) - floor(miny)) / YDIV;
pl_flinewidth(0.25);
/* compute scale factors to adjust aspect */ if ( lx < ly ) { sx = lx/ly; sy = 1.0; } else { sx = 1.0; sy = ly/lx; }
pl_erase();
/* a frame... */ pl_fbox(floor(minx), floor(miny),
ceil(maxx), ceil(maxy));
/* labels and "tics" */ pl_fontname("HersheySerif"); for(ny=floor(miny); ny < ceil(maxy); ny += yticstep) { pl_fline(floor(minx), ny, ceil(maxx), ny); snprintf(labs, MAXLABLEN, "%6.2lf", ny); FMOVESCALE(floor(minx) - XLAB_WIDTH_F * lx, ny); PUSHSCALE(sx,sy); pl_label(labs); POPSCALE(sx,sy); } for(nx=floor(minx); nx < ceil(maxx); nx += xticstep) { pl_fline(nx, floor(miny), nx, ceil(maxy)); snprintf(labs, MAXLABLEN, "%6.2lf", nx); FMOVESCALE(nx, floor(miny)); PUSHSCALE(sx,sy); pl_ftextangle(-90); pl_alabel('l', 'b', labs); POPSCALE(sx,sy); }
/* plot data "point" */ pl_fillcolorname("red"); pl_filltype(1); for(i=0; i < NP; i++) { pl_fbox(x[i] - lx * DOTSCALE, y[i] - ly * DOTSCALE, x[i] + lx * DOTSCALE, y[i] + ly * DOTSCALE); }
pl_flushpl(); pl_closepl();
}</lang>
No one would use the previous code to produce a plot (that looks this way; instead, normally we produce data through a program, then we plot the data using e.g. gnuplot or other powerful tools; the result (with GNUplot and without enhancement) could look like this instead.
gnuplot
<lang gnuplot>unset key # Only one data set, so the key is uninformative
plot '-' # '-' can be replaced with a filename, to read data from that file.
0 2.7 1 2.8 2 31.4 3 38.1 4 68.0 5 76.2 6 100.5 7 130.0 8 149.3 9 180.0
e</lang>
J
load 'plot' magnitudes =: 2.7 2.8 31.4 38.1 58.0 76.2 100.5 130.0 149.3 180.0 'dot; pensize 2.4' plot magnitudes
Maxima
(%i1) ".." (m, n) := makelist (i, i, m, n); infix ("..")$ (%i2) x: 0 .. 9$ y:[2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0]$ (%i3) plot2d(['discrete, x, y], [style, [points,5,1,1]], [gnuplot_term, png], [gnuplot_out_file, "qsort-range-10-9.png"])$
OCaml
<lang ocaml>#load "graphics.cma" open Graphics
let round x = int_of_float (floor(x +. 0.5))
let x = [0; 1; 2; 3; 4; 5; 6; 7; 8; 9] and y = [2.7; 2.8; 31.4; 38.1; 58.0; 76.2; 100.5; 130.0; 149.3; 180.0]
let () =
open_graph ""; List.iter2 (fun x y -> (* scale to fit in the window *) let _x = x * 30 and _y = round(y *. 2.0) in plot _x _y) x y; ignore(wait_next_event [Key_pressed]); close_graph();
- </lang>
Octave
<lang octave>x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]; y = [2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0]; plot(x,y,"o"); pause;</lang>
Perl
<lang perl>
use GD::Graph::points; @data = ( [0, 1, 2, 3, 4, 5, 6, 7, 8, 9], [2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0], ); $graph = GD::Graph::points->new(400, 300); $gd = $graph->plot(\@data) or die $graph->error; # Save as image. open(OUF, ">qsort-range-10-9.png"); binmode OUF; print OUF $gd->png; close(OUF);
</lang>
<lang perl>
use Imager; use Imager::Plot; @x = (0, 1, 2, 3, 4, 5, 6, 7, 8, 9); @y = (2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0); $plot = Imager::Plot->new( 'Width' => 400, 'Height' => 300, 'GlobalFont' => 'PATH_TO_TTF_FONT', ); $plot->AddDataSet( 'X' => \@x, 'Y' => \@y, 'style' => { 'marker' => { 'size' => 2, 'symbol' => 'circle', 'color' => Imager::Color->new('red'), }, }, ); $img = Imager->new( 'xsize' => 500, 'ysize' => 400, ); $img->box('filled' => 1, 'color' => 'white'); $plot->Render('Image' => $img, 'Xoff' => 50, 'Yoff' => 350); $img->write('file' => 'qsort-range-10-9.png');
</lang>
Python
>>> x = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] >>> y = [2.7, 2.8, 31.4, 38.1, 58.0, 76.2, 100.5, 130.0, 149.3, 180.0]
>>> import pylab >>> pylab.plot(x, y, 'bo') >>> pylab.savefig('qsort-range-10-9.png')
qsort-range-10-9.png (23 KiB)
Tcl
This solution does not use existing plotting packages, but constructs the graphics from bare-metal Tk code.
<lang Tcl>package require Tk
- The actual plotting engine
proc plotxy {canvas xs ys} {
global xfac yfac set maxx [tcl::mathfunc::max {*}$xs] set maxy [tcl::mathfunc::max {*}$ys] set xfac [expr {[winfo width $canvas] * 0.8/$maxx}] set yfac [expr {[winfo height $canvas] * 0.8/$maxy}] scale $canvas x 0 $maxx $xfac scale $canvas y 0 $maxy $yfac foreach x $xs y $ys { dot $canvas [expr {$x*$xfac}] [expr {$y*$yfac}] -fill red }
}
- Rescales the contents of the given canvas
proc scale {canvas direction from to fac} {
set f [expr {$from*$fac}] set t [expr {$to*$fac}] switch -- $direction { x { set f [expr {$from * $fac}] set t [expr {$to * $fac}] $canvas create line $f 0 $t 0 $canvas create text $f 0 -anchor nw -text $from $canvas create text $t 0 -anchor n -text $to } y { set f [expr {$from * -$fac}] set t [expr {$to * -$fac}] $canvas create line 0 $f 0 $t $canvas create text 0 $f -anchor se -text $from $canvas create text 0 $t -anchor e -text $to } }
}
- Helper to make points, which are otherwise not a native item type
proc dot {canvas x y args} {
set id [$canvas create oval [expr {$x-3}] [expr {-$y-3}] \ [expr {$x+3}] [expr {-$y+3}]] $canvas itemconfigure $id {*}$args
}
pack [canvas .c -background white] update set xs {0 1 2 3 4 5 6 7 8 9} set ys {2.7 2.8 31.4 38.1 58.0 76.2 100.5 130.0 149.3 180.0} plotxy .c $xs $ys .c config -scrollregion [.c bbox all] .c move all 20 20
- Save image (this is the only part that requires an external library)
package require Img set im [image create photo -data .c] $im write plotxy.png -format PNG</lang> Of course, if we were generating an encapsulated postscript version, we would be able to do that directly.