Error in colorbar position with Julia?

Q&A GMT.jl Q&A
1

I am beginning with Julia so I am not very sure if I am doing something wrong. I try to reproduce the Color Image example.

In the original image, the colorbar is on top of the map. I got it within the map.

(I reduce its quality to upload it here).

index
index1850×2169 785 KB

Full script run in Pluto Notebook.

begin
	using GMT
	topo = makecpt(color=:rainbow, range=(1000,5000,500), continuous=true);
	grdimage("@tut_relief.nc", shade=(azimuth=100, norm="e0.8"), proj=:Mercator, frame=:a, color=topo)
	colorbar!(pos=(anchor=:TC,length=(12.5,0.6), horizontal=true, offset=(0,1.0)),
	          color=topo, frame=(ylabel=:m,), fmt=:jpg, show=true)
end
2

I’ll try to look at this later but in very quick try I even got a strange error

colorbar!(pos=(anchor=:TC,length=(12.5,0.6), horizontal=true, offset=(0,1.0)), color=topo, frame=(ylabel=:m,), show=true)
psscale [ERROR]: North is outside -90 to +90 degree range
psscale [ERROR]: General map projection error
ERROR: Something went wrong when calling the module. GMT error number = 74

however, this more concentrated form works

topo = makecpt(color=:rainbow, range=(1000,5000,500), continuous=true);
imshow("@tut_relief.nc", shade=(azimuth=100, norm="e0.8"), proj=:Mercator, frame=:a, color=topo, colorbar=(pos=(anchor=:BC,length=(12.5,0.6), horizontal=true, offset=(0,1.0)), ylabel=:m))

Ah, and this also works

grdimage("@tut_relief.nc", shade=(azimuth=100, norm="e0.8"), proj=:Mercator, frame=:a, color=topo)

colorbar!(pos=(anchor=:BC,length=(12.5,0.6), horizontal=true, offset=(0,1.0)), color=topo, ylabel=:m, show=true)

I dropped the fmt=:jpg because it really brings nothing when one do not specify also the figure name.

3

Don’t understand how you got your example map. Either with Julia or plain GMT CLI I get the colorbar inside but on the top side.

One thing very useful to know about the Julia wrapper is that the option Vd=1 in any command prints the corresponding GMT command. So, using it I reproduced the bellow CLI commands that renders the same figure as the Julia ones.

julia> gmtwrite("topo.cpt", topo)

gmt grdimage @tut_relief.nc  -JM14c -Ba -n+a -Ctopo.cpt -I+a100+ne0.8 -P -K > lixo.ps
gmt psscale  -R -J -Baf -By+lm -BWSen -DJTC+w12.5/0.6+h -Ctopo.cpt -K -O >> lixo.ps
4

Joaquim, How exactly do you work with Julia? Do you write directly on the terminal? Use a file?

I managed to get the figure. I saw that I got this message:
From worker 4: psscale [ERROR]: Found no history for option -R

So I add region="@tut_relief.nc" in my code within grdimage.

begin
	topo2 = makecpt(color=:rainbow, range=(1000,5000,500), continuous=true);
	grdimage("@tut_relief.nc", shade=(azimuth=100, norm="e0.8"), proj=:Mercator, frame=:a, color=topo2, region="@tut_relief.nc")
	colorbar!(pos=(anchor=:TC,length=(12.5,0.6), horizontal=true, offset=(0,1.0)),
	color=topo2, frame=(ylabel=:m,), fmt=:jpg, show=true)
end
5

Hmm, don’t know what Pluto is doing here. I don’t like that Pluto obliges to put multi-commands inside a begin/end block so I never used that. I either put one command per cell or use the alternative package Neptune that does not have this constrain and is not reactive.

I always use the REPL or scripts in VSCode and Neptune or Jupyter for notebooks.

6

Thanks Joaquim.

I try the original script on Neptune and it works fine. So there is a problem with Pluto.

7

The thing that occurs to me is it runs those two commands in parallel and hence in different sessions, so common options are not visible to each other.