Menu
▾
▴
gnuplot-beta
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 00:52:48
|
In the latest CVS the "fillbetween.dem" demo for set term postscript color does not fill in the cross hatch pattern of the second example. (I think this worked at one time.) The PDF terminal produces the desired output. If this sounds like a familiar bug, let me know. Otherwise I will put it on SourceForge. Dan |
|
From: Ethan M. <merritt@u.washington.edu> - 2007-01-26 00:56:55
|
On Thursday 25 January 2007 17:04, Daniel J Sebald wrote: > In the latest CVS the "fillbetween.dem" demo for > > set term postscript color > does not fill in the cross hatch pattern of the second example. Yes, it does. You have rediscovered the longstanding bug in ghostview/gv that fails to display pattern fill if anti-aliasing is enabled. If you toggle the antialiasing setting in your viewer, you will see the pattern. |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 02:14:30
|
Ethan Merritt wrote: > On Thursday 25 January 2007 17:04, Daniel J Sebald wrote: > >>In the latest CVS the "fillbetween.dem" demo for >> >>set term postscript color >>does not fill in the cross hatch pattern of the second example. > > > Yes, it does. > > You have rediscovered the longstanding bug in ghostview/gv that fails to > display pattern fill if anti-aliasing is enabled. > > If you toggle the antialiasing setting in your viewer, you will see the pattern. Oh... OK thanks. I just built the latest version of ghostscript. Rather than not show fill, 8.54 issues error messages. (Turning off antialiasing works correctly.) So still unresolved, unless it is in their SVN repository somewhere. Dan |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 05:47:53
|
Here's one thing to consider. I noticed that in this fill between example where the antialiasing filter filters out the pattern leaving the area between curves blank there is the key sample which appears even though the fill between curves doesn't. The sample and the fill are done two different ways. The sample looks like: 6479 4452 399 70 40 45 2 PatternFill I replaced that with something looking more like what is used for the fill: gsave 6479 4452 N 0 70 V 399 0 V 0 -70 V Pattern2 fill grestore which makes the fill behave and appear similar. Is it worth changing this? Dan |
|
From: Ethan A M. <merritt@u.washington.edu> - 2007-01-26 06:05:52
|
On Thursday 25 January 2007 21:59, Daniel J Sebald wrote: > Here's one thing to consider. I noticed that in this fill between example where the antialiasing filter filters out the pattern leaving the area between curves blank there is the key sample which appears even though the fill between curves doesn't. > The sample and the fill are done two different ways. The sample looks like: > > 6479 4452 399 70 40 45 2 PatternFill > > I replaced that with something looking more like what is used for the fill: > > gsave 6479 4452 N 0 70 V 399 0 V 0 -70 V Pattern2 fill grestore > > which makes the fill behave and appear similar. Similar as in "both work" or similar as in "neither works"? -- Ethan A Merritt Biomolecular Structure Center University of Washington, Seattle 98195-7742 |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 06:24:05
|
Ethan A Merritt wrote: > Similar as in "both work" or similar as in "neither works"? Similar as in neither works, until antialiasing is turned off. |
|
From: Ethan A M. <merritt@u.washington.edu> - 2007-01-26 06:26:09
|
On Thursday 25 January 2007 22:35, Daniel J Sebald wrote: > Ethan A Merritt wrote: > > which makes the fill behave and appear similar. > > Is it worth changing this? > > > Similar as in "both work" or similar as in "neither works"? > > Similar as in neither works, until antialiasing is turned off. Well then no, I don't see much worth in breaking the part that's working :-) -- Ethan A Merritt Biomolecular Structure Center University of Washington, Seattle 98195-7742 |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 06:33:35
|
Ethan A Merritt wrote: > Well then no, I don't see much worth in breaking the part that's working :-) Suppose not. |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 09:15:18
Attachments:
fb2.ps.gz
|
There may still be a problem with pattern fill. I ran fillbetween.dem to the second plot, then hit Cntrl-C, then set term postscript color set output 'fb2.ps' replot set output to generate the attached file. When I look at this with gv (antialiasing off) things are fine. But when I send it to an HP2200D printer (lpr -P simplex fb2.ps) it prints the fill between line except the very first column (left most... which actually may be the last strip if things are drawn as they are popped off the stack). The left most column comes out white and there is no annotation on the plot. I do the same process for the first plot in fillbetween.dem and everything prints fine. Dan |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 09:24:00
|
Daniel J Sebald wrote: > There may still be a problem with pattern fill. I ran fillbetween.dem ...but I don't think what I described in the last email has anything to do with that antialiasing issue in gv. If I generate the pattern fill example using "set term pdf", gv behaves the same way on the resulting PDF. Dan |
|
From: Daniel J S. <dan...@ie...> - 2007-01-26 18:04:08
|
One small issue, for someone familiar with prologue.ps. In this definition for Pat3 at the end of the paint process it says "fill" whereas all the others say "stroke". Selecting Pat3 results in a solid red fill. Is that desired? Or should that be "stroke"?
<< Tile8x8
/PaintProc {0.5 setlinewidth pop 0 0 M 0 8 L
8 8 L 8 0 L 0 0 L fill}
>> matrix makepattern
/Pat3 exch def
Daniel J Sebald wrote:
> Daniel J Sebald wrote:
>
>>There may still be a problem with pattern fill. I ran fillbetween.dem
>
>
> ...but I don't think what I described in the last email has anything to do with that antialiasing issue in gv. If I generate the pattern fill example using "set term pdf", gv behaves the same way on the resulting PDF.
>
> Dan
|
|
From: Ethan M. <merritt@u.washington.edu> - 2007-01-26 18:46:16
|
On Friday 26 January 2007 10:15, Daniel J Sebald wrote: > One small issue, for someone familiar with prologue.ps. > In this definition for Pat3 at the end of the paint process it says "fill" whereas > all the others say "stroke". Selecting Pat3 results in a solid red fill. Correct. Pattern 3 is solid fill on most terminal types. |
