POLYGON ignores line thickness in PostScript output

[karlmsmith]

Reported by @AndrewWittenberg on 28 Aug 2010 21:41 UTC Specifying thicker lines with POLYGON has no effect on the PostScript (metafile) output, despite showing up correctly on the display and in GIF output.

The example below uses PLOT/VS and POLYGON to generate two lines that both ought be thick; however, the POLYGON lines are not thickened properly in the Postscript output (my.ps). I am attaching the my.gif and my.ps that are generated.

        NOAA/PMEL TMAP
        FERRET v6.62  
        Linux rh5 (gfortran) 2.6.18-164.11.1.el5 - 06/11/10
        28-Aug-10 17:27     

yes? set mode meta:my.plt
yes? poly/line=13 {0,1},{0,1}
yes? plot/ov/vs/line=13 {0,1},{1,0}
yes? frame/file=my.gif
yes? can mode meta; ppl clsplt
yes? sp Fprint -o my.ps -p portrait -l cps -R my.plt
yes? sp rm my.plt

Migrated-From: http://dunkel.pmel.noaa.gov/trac/ferret/ticket/1751

[karlmsmith]

Comment by @AndrewWittenberg on 28 Aug 2010 21:45 UTC Could this be related to ticket 844?

[karlmsmith]

Comment by steven.c.hankin on 30 Aug 2010 16:46 UTC Just fyi --

In the context of Karl's work on "Pfyrret" (Python Ferret) we have been discussing the replacement of XGKS with some other graphical primitive library. Doing so would mean that the PostScript generation (which lies in these libraries, not in Ferret code per se) would be altogether new and presumably much more robust.

We are weighing the priority of this task, as it is a considerable effort (likely to have lots of side effects that would need to be tracked down). We are weighing it against the competing priorities of linking Ferret to Python-based 3-D visualizations (a much simpler task). Our current discussions have favored the 3D viz tasks on a cost/benefit basis, but the floor is open for discussions.

[karlmsmith]

Comment by @AnsleyManke on 23 Feb 2011 19:12 UTC Just a bit more information. I did some simple checking and the call is being made to set the polygon outline to the thick setting. The polygon command is putting the code for line-width into the postscript file. Drawing just the PLOT and comparing the ouput, and then drawing just the POLY and comparing, in both cases a thin line has

1.000000 lw

and the thick line has

3.000000 lw

in the postscript, which sets the thickness to 1 for thin or 3 for thick. So this is not just a bug in the implementation of the POLYGON call, but something deeper in the graphics.

[karlmsmith]

Attachment from @AndrewWittenberg on 28 Aug 2010 21:42 UTC my.gif

[karlmsmith]

Attachment from @AndrewWittenberg on 28 Aug 2010 21:42 UTC my.ps my.ps.gz