Thanks for this contribution! I have picked up both files and they are now in the lastest pacman updates.
I did not pick up chapgo and lchap as I am not sure where to doc or put them. My feeling now is that lab chapters were a mistake and we'd be better to break them into separate labs that are tied together by a common title prefix. On Mon, Jan 17, 2011 at 11:16 AM, David Mitchell <[email protected]>wrote: > Here are rough and ready versions of jhs plot and the j.ijt lab. > > For me, with these, the "Lab: A J Introduction" lab will now run to the end > and > displays some not unreasonable charts and graphs. > > > http://www.jsoftware.com/jwiki/DavidMitchell?action=AttachFile&do=get&target=jgcp.ijs > > > http://www.jsoftware.com/jwiki/DavidMitchell?action=AttachFile&do=get&target=j.ijt > > I also used these to help in getting around in the labs without the J6 gui: > > chapgo=:3 : 0 > labopenchapter_jlab_ <:y > ) > > lchap=:3 : 0 > (<"0 >: i.#CHAPTERS_jlab_),.CHAPTERS_jlab_ > ) > -- > > On 1/16/2011 13:23, Eric Iverson wrote: > > There is a lot of work to be done on GC plot. I'm hoping this will be > done > > by the community after the release is out. With relatively little effort > GC > > plot could do a much better mapping for plot options and data. I think > this > > effort is better delayed until after the release. > > > > I have made the following changes as a result of your comments: > > > > 1. xline flipped axis as you suggest (embarassing mistake) > > 2. added type xsticklinexy > > 3. kludge plot to do xsticklinexy for case: > > 'stick,line' plot (i.10);10?10 > > 4. kept default height 100 (rather than your change to 300) > > > > Note that with jgc you can get your 300 height. E.g., > > 'reset xsticklinexy&chs=300x300 show' jgc (i.10);10?.10 > > > > On Sun, Jan 16, 2011 at 12:03 PM, David Mitchell<[email protected] > >wrote: > > > >> On further reflection, it looks like the axes display is reversed. > >> > >> I get better plot results with: > >> > >> xline=: > '&cht=lc&chs=300x300&chds=<MIN>,<MAX>&chxt=y,x&chxr=1,<MIN>,<MAX>' > >> > >> where chtxt is changed from x,y to y,x. > >> > >> Also, this seems to work for the 'stick,line' plot: > >> > >> > >> > xlinexy=:'&cht=lxy&chs=300x300&chds=<MIN>,<MAX>&chxt=y,x&chxr=0,<YMIN>,<YMAX>|1,<MIN>,<MAX>&chm=v,FF0000,0,::1,2' > >> -- > >> > >> On 1/16/2011 11:07, Eric Iverson wrote: > >>> I agree that this is an important lab to get to run clean and will make > >> some > >>> minor adjustments. > >>> > >>> Having a mix of jhs GC plot and jgtk plot in the same session probably > >>> shouldn't be allowed to happen in the lab. > >>> > >>> For now my goal is to get that lab to run clean in JHS. I will probably > >>> adjust it so that only JHS viewmat and plot is used. > >>> > >>> For the first release gcplot won't be able to cope with all the > >>> requirements of that lab. For now I think the reports of 'unsupported > >>> options and complexity' are acceptable. The eventual fix is to extend > the > >>> capabilities of gcplot, not to revert to jgtk plot. > >>> > >>> On Sun, Jan 16, 2011 at 7:07 AM, David Mitchell<[email protected] > >>> wrote: > >>> > >>>> I tried running this and ran afoul of viewmat and plot. > >>>> > >>>> I changed the first > >>>> > >>>> require 'plot graph' > >>>> > >>>> to > >>>> > >>>> require 'viewmat' > >>>> > >>>> I found this rather nice, as the viewmat output was now displayed > >> in-line > >>>> in the > >>>> browser. > >>>> > >>>> But, non-viewmat plots were a thornier problem. > >>>> > >>>> Removing > >>>> > >>>> require'plot graph' > >>>> > >>>> invoked the new Google plot, which was fine for a few plots, but > failed > >> on > >>>> most > >>>> with errors like this: > >>>> > >>>> unsupported plot options: stick,line > >>>> unsupported data complexity > >>>> > >>>> Changing > >>>> > >>>> require'plot graph' > >>>> > >>>> to > >>>> > >>>> require 'plot' > >>>> > >>>> now got what looks like the new GTK plot. This successfully displayed > >> all > >>>> the > >>>> plots, but the plot window opened behind my browser window and I had > to > >>>> hunt for > >>>> it. Also, the plot window appeared before the text in the lab > >> describing > >>>> the > >>>> plot. Finally, the plot window had to be closed before I could > continue > >>>> with > >>>> the lab. > >>>> > >>>> And, this would probably would not work well if the J lab browser were > >> not > >>>> on > >>>> the same system as jhs. > >>>> ---------------------------------------------------------------------- > >>>> 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 > >> > > ---------------------------------------------------------------------- > > 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
