Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.graphics.apps.gnuplot > #2294 > unrolled thread

spurious horizontal lines in svg output (impulses plot)

Started bymartin.kraska@googlemail.com
First post2014-02-05 13:00 -0800
Last post2014-02-05 16:29 -0800
Articles 3 — 2 participants

Back to article view | Back to comp.graphics.apps.gnuplot


Contents

  spurious horizontal lines in svg output (impulses plot) martin.kraska@googlemail.com - 2014-02-05 13:00 -0800
    Re: spurious horizontal lines in svg output (impulses plot) Ethan A Merritt <merritt@u.washington.edu> - 2014-02-05 14:47 -0800
      Re: spurious horizontal lines in svg output (impulses plot) martin.kraska@googlemail.com - 2014-02-05 16:29 -0800

#2294 — spurious horizontal lines in svg output (impulses plot)

Frommartin.kraska@googlemail.com
Date2014-02-05 13:00 -0800
Subjectspurious horizontal lines in svg output (impulses plot)
Message-ID<4b39dad2-b266-4cbd-9989-229c46388b82@googlegroups.com>
Hello,

we get rendering problems with svg output of the file below. See the problem description, the input files and the svg file here:

http://en.smath.info/forum/yaf_postsm12035_Image-Region.aspx#post12035

The following part from the svg file is meant to create the vertical lines (impulses). Possibly, due to a gnuplot bug there is a small horizontal line in the end (h0.01). If that line is removed from the file, then the plots are correct in whatever viewer. Is there a way to avoid this spurious line?


<g style="fill:none; color:red; stroke:currentColor; stroke-width:8.00; stroke-linecap:butt; stroke-linejoin:miter">
	<path stroke='rgb(  0,   0, 255)'  d='M41.9,126.0 L41.9,24.5 M54.4,126.0 L54.4,24.7 M66.8,126.0 L66.8,25.0 M79.3,126.0 L79.3,25.5 
		M91.7,126.0 L91.7,26.2 M104.2,126.0 L104.2,27.1 M116.6,126.0 L116.6,28.1 M129.1,126.0 L129.1,29.4 
		M141.5,126.0 L141.5,31.0 M154.0,126.0 L154.0,32.7 M166.4,126.0 L166.4,34.7 M178.8,126.0 L178.8,37.0 
		M191.3,126.0 L191.3,39.5 M203.7,126.0 L203.7,42.3 M216.2,126.0 L216.2,45.3 M228.6,126.0 L228.6,48.5 
		M241.1,126.0 L241.1,52.0 M253.5,126.0 L253.5,55.7 M266.0,126.0 L266.0,59.5 M278.4,126.0 L278.4,63.5 
		M290.9,126.0 L290.9,67.6 M303.3,126.0 L303.3,71.7 M315.8,126.0 L315.8,75.8 M328.2,126.0 L328.2,80.0 
		M340.7,126.0 L340.7,84.1 M353.1,126.0 L353.1,88.0 M365.6,126.0 L365.6,91.9 M378.0,126.0 L378.0,95.5 
		M390.5,126.0 L390.5,99.0 M402.9,126.0 L402.9,102.3 M415.4,126.0 L415.4,105.3 M427.8,126.0 L427.8,108.0 
		M440.3,126.0 L440.3,110.6 M452.7,126.0 L452.7,112.8 M465.1,126.0 L465.1,114.8 M477.6,126.0 L477.6,116.6 
		M490.0,126.0 L490.0,118.1 M502.5,126.0 L502.5,119.4 M514.9,126.0 L514.9,120.5 M527.4,126.0 L527.4,121.4 
		M539.8,126.0 L539.8,122.0 M552.3,126.0 L552.3,122.6 M564.7,126.0 L564.7,122.9 M577.2,126.0 L577.2,123.0 
		 h0.01'/></g>

set origin 0.0, 0.0
set obj 1 rectangle behind from screen 0.0,0.0 to screen 1.0,1.0
set style rectangle fillcolor rgb '#fefefe' fs solid 1.0 noborder
set size noratio
set xrange [-22.0:22.0]
set yrange [0.0:1.1]
set cbrange [*:*]
unset logscale x
unset logscale y
unset logscale cb
unset grid
set title ''
set xlabel ''
set ylabel ''
set border 15
unset xzeroaxis
unset yzeroaxis
set xtics norotate border autofreq
unset x2tics
set ytics norotate border autofreq
unset y2tics
set cbtics autofreq
set colorbox
set cblabel ''
set palette rgbformulae 7,5,15
set term svg fsize 8 fname 'Arial' noenhanced dynamic size 600,150
plot 'C:/Users/Kraska/data.gnuplot' index 0 t '' w i lw 8.0 lt 1 lc rgb '#0000ff' axis x1y1
unset output

contents of the data file:
-21.5 0.97173054266333 
-20.5 0.97015458346886 
-19.5 0.96697480282461 
-18.5 0.96213701490613 
-17.5 0.95556374637968 
-16.5 0.94715868521459 
-15.5 0.9368124108865 
-14.5 0.92440922403982 
-13.5 0.90983484025496 
-12.5 0.89298465876465 
-11.5 0.87377226666184 
-10.5 0.85213779728338 
-9.5 0.8280557335159 
-8.5 0.80154173808751 
-7.5 0.77265810792432 
-6.5 0.74151749120938 
-5.5 0.70828457453844 
-4.5 0.67317554162008 
-3.5 0.6364552197632 
-2.5 0.59843195894586 
-1.5 0.55945042167504 
-0.5 0.51988259011269 
0.5 0.48011740988736 
1.5 0.44054957832502 
2.5 0.4015680410542 
3.5 0.36354478023686 
4.5 0.32682445837998 
5.5 0.29171542546161 
6.5 0.25848250879067 
7.5 0.22734189207574 
8.5 0.19845826191254 
9.5 0.17194426648415 
10.5 0.14786220271667 
11.5 0.12622773333821 
12.5 0.10701534123539 
13.5 0.090165159745083 
14.5 0.07559077596023 
15.5 0.063187589113547 
16.5 0.052841314785452 
17.5 0.044436253620367 
18.5 0.037862985093908 
19.5 0.033025197175423 
20.5 0.029845416531165 
21.5 0.028269457336693 

[toc] | [next] | [standalone]


#2295

FromEthan A Merritt <merritt@u.washington.edu>
Date2014-02-05 14:47 -0800
Message-ID<lcuf1a$k4a$1@dont-email.me>
In reply to#2294
martin.kraska@googlemail.com wrote:

> Hello,
> 
> we get rendering problems with svg output of the file below. See the
> problem description, the input files and the svg file here:
> 
> http://en.smath.info/forum/yaf_postsm12035_Image-Region.aspx#post12035
> 
> The following part from the svg file is meant to create the vertical lines
> (impulses). Possibly, due to a gnuplot bug there is a small horizontal
> line in the end (h0.01). If that line is removed from the file, then the
> plots are correct in whatever viewer. Is there a way to avoid this
> spurious line?

Sometimes you can't win.  That extra step (h0.01) was added to work around a
bug in earlier svg toolkits, most widely seen by users of Firefox versions
3.X.    At the time, this magic workaround fixed the rendering in Firefox
and did not seem to break anything else.  Now it seems that a new generation
of svg viewers exhibits a contrary bug; the tiny horizontal increment
should be invisible, but apparently some viewers handle it badly.

Since Firefox version 3.x and its ilk are long gone, I will see to it
that the old work-around is removed from future gnuplot releases.
I regret that I cannot offer a workaround for your currently problematic
viewers.  The gnuplot output displays correctly in current Firefox, 
for whatever that is worth.

	Ethan (gnuplot development team)


> 
> 
> <g style="fill:none; color:red; stroke:currentColor; stroke-width:8.00;
> stroke-linecap:butt; stroke-linejoin:miter">
> <path stroke='rgb(  0,   0, 255)'  d='M41.9,126.0 L41.9,24.5 M54.4,126.0
> L54.4,24.7 M66.8,126.0 L66.8,25.0 M79.3,126.0 L79.3,25.5 M91.7,126.0
> L91.7,26.2 M104.2,126.0 L104.2,27.1 M116.6,126.0 L116.6,28.1 M129.1,126.0
> L129.1,29.4 M141.5,126.0 L141.5,31.0 M154.0,126.0 L154.0,32.7 M166.4,126.0
> L166.4,34.7 M178.8,126.0 L178.8,37.0 M191.3,126.0 L191.3,39.5 M203.7,126.0
> L203.7,42.3 M216.2,126.0 L216.2,45.3 M228.6,126.0 L228.6,48.5 M241.1,126.0
> L241.1,52.0 M253.5,126.0 L253.5,55.7 M266.0,126.0 L266.0,59.5 M278.4,126.0
> L278.4,63.5 M290.9,126.0 L290.9,67.6 M303.3,126.0 L303.3,71.7 M315.8,126.0
> L315.8,75.8 M328.2,126.0 L328.2,80.0 M340.7,126.0 L340.7,84.1 M353.1,126.0
> L353.1,88.0 M365.6,126.0 L365.6,91.9 M378.0,126.0 L378.0,95.5 M390.5,126.0
> L390.5,99.0 M402.9,126.0 L402.9,102.3 M415.4,126.0 L415.4,105.3
> M427.8,126.0 L427.8,108.0 M440.3,126.0 L440.3,110.6 M452.7,126.0
> L452.7,112.8 M465.1,126.0 L465.1,114.8 M477.6,126.0 L477.6,116.6
> M490.0,126.0 L490.0,118.1 M502.5,126.0 L502.5,119.4 M514.9,126.0
> L514.9,120.5 M527.4,126.0 L527.4,121.4 M539.8,126.0 L539.8,122.0
> M552.3,126.0 L552.3,122.6 M564.7,126.0 L564.7,122.9 M577.2,126.0
> L577.2,123.0 h0.01'/></g>

[toc] | [prev] | [next] | [standalone]


#2296

Frommartin.kraska@googlemail.com
Date2014-02-05 16:29 -0800
Message-ID<d3ce0982-3c4f-4c88-b9a1-653dcd277ca7@googlegroups.com>
In reply to#2295
Thank you for the immediate response. Obviously we cannot do anything except waiting for a new version. I could produce eps or pdf images but don't know yet if the image display programmer can display them in a scalable way. If not, we can just use pixel graphics.

Martin


Am Mittwoch, 5. Februar 2014 23:47:04 UTC+1 schrieb Ethan A Merritt:

> Sometimes you can't win.  That extra step (h0.01) was added to work around a
> 
> bug in earlier svg toolkits, most widely seen by users of Firefox versions
> 
> 3.X.    At the time, this magic workaround fixed the rendering in Firefox
> 
> and did not seem to break anything else.  Now it seems that a new generation
> 
> of svg viewers exhibits a contrary bug; the tiny horizontal increment
> 
> should be invisible, but apparently some viewers handle it badly.
> 
> 
> 
> Since Firefox version 3.x and its ilk are long gone, I will see to it
> 
> that the old work-around is removed from future gnuplot releases.
> 
> I regret that I cannot offer a workaround for your currently problematic
> 
> viewers.  The gnuplot output displays correctly in current Firefox, 
> 
> for whatever that is worth.
> 
> 
> 
> 	Ethan (gnuplot development team)

[toc] | [prev] | [standalone]


Back to top | Article view | comp.graphics.apps.gnuplot


csiph-web