Well, the plot comes out exactly as expected, as one line + error bars per item of y, so I do think this is the way to do it (the docs are really sparse on how each plot type takes it's data, especially the things taking more than one vector per line). The issue is the colors are not in the normal order. Normally they go from blue, red, green, purple, magenta, dirty yellow, teal, yellow, beige, cyan (as can be seen in the key) but the color of the lines plotted takes one, then skips two, takes the third for the next line...
I think there is a tiny mistake in the logic giving a line it's color, since there is 3 sets of data per item (one for each of high, mid and low). 2014-09-05 2:57 GMT+02:00 bill lam <[email protected]>: > Reporting bugs in j general forum should be ok. see > http://www.jsoftware.com/jwiki/System/BugReports > > Unfortunately I know very little of plot. Are you sure your test data is > correct? > On Sep 5, 2014 2:39 AM, "Jan-Pieter Jacobs" <[email protected]> > wrote: > >> I just noticed a slight bug in the colors assigned to different items >> in "errorbar" and "hilo" plots: >> >> d =: 0 10 20 30 +"0 2 ] 1 2 3 +"0 1 i. 10 >> 'errorbar;key 0 1 2 3 4 5 6 7 8 9 10 11' plot d >> >> the colors assigned to each subsequent item in d jumps with 3, causing >> the plot key not to correspond with the plot colors. The same happens >> for the "hilo" plot... >> >> Is there actually a bugtracker somewhere for this kind of issues, or >> is the mailing list indeed the appropriate place for reporting bugs? >> ---------------------------------------------------------------------- >> For information about J forums see http://www.jsoftware.com/forums.htm >> > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
