GMT figures are not fully isolated in one single session

[seisman]

A bash script to reproduce the issue:

gmt begin

gmt figure makecpt1 pdf
gmt makecpt -Ccategorical -W
gmt colorbar -C -R10/70/-5/10 -B -DJBC

gmt figure other pdf
gmt coast -R0/10/-20/-10 -JM6i -B -Gblack

gmt figure makecpt2 pdf
gmt makecpt -Ccategorical -W
gmt colorbar -C -R10/70/-5/10 -B -DJBC

gmt end

In this example, three figures are generated. The 1st and the 3rd figure uses the same code, so we expect makecpt1.pdf and makecpt2.pdf are the same, but actually, these two images are slightly different. However, if we remove the code for the 2nd figure, makecpt1.pdf and makecpt2.pdf are the same again.

I think it may mean that the "settings" for each "figure" are not fully isolated and it looks like a GMT bug.

Originally posted by @seisman in GenericMappingTools/pygmt#1289 (comment)

[PaulWessel]

Sory, can you help me see the difference between 1 and 3? Looks the same to my old eye.

[seisman]

The size of the annotations is slightly different.

[seisman]

Here is the diff image from

gm compare -density 200 -highlight-color magenta -highlight-style assign -metric rmse -file diff.png makecpt1.pdf makecpt2.pdf

[PaulWessel]

Can you confirm that this seems related to slightly different pens and font values?

[seisman]

I have no idea. I tried to add -Vd to the two colorbar commands separately and check the debugging messages.

Here are the main differences:

The first figure shows:

colorbar [DEBUG]: Anchor point specified via justification code: BC
colorbar [DEBUG]: Computed automatic parameters using dimension scaling: 1
colorbar [DEBUG]: Determined colorbar side = S and axis = x
colorbar [INFORMATION]: No file /Users/seisman/.gmt/sessions/gmt_session.79282/gmt.frame with frame information - no adjustments made
colorbar [DEBUG]: Bar settings: justify = 10, dx = 0 dy = 0.125
colorbar [DEBUG]: gmtlib_get_graphics_item: Fig: 1 Subplot: 2 Panel: () Inset: 0
colorbar [INFORMATION]: Selected current /Users/seisman/.gmt/sessions/gmt_session.79282/gmt.1.cpt file cpt

but the second figure show:

colorbar [DEBUG]: Anchor point specified via justification code: BC
colorbar [DEBUG]: Computed automatic parameters using dimension scaling: 1
colorbar [DEBUG]: Determined colorbar side = S and axis = x
colorbar [DEBUG]: B item = 

colorbar [DEBUG]: Adding annotation space
colorbar [DEBUG]: gmtlib_get_graphics_item: Fig: 3 Subplot: 2 Panel: () Inset: 0
colorbar [DEBUG]: Reading GMT Default parameters from file: /Users/seisman/.gmt/sessions/gmt_session.79328/gmt.conf
colorbar [DEBUG]: Bar settings: justify = 10, dx = 0 dy = 0.324444
colorbar [DEBUG]: gmtlib_get_graphics_item: Fig: 3 Subplot: 2 Panel: () Inset: 0
colorbar [INFORMATION]: Selected current /Users/seisman/.gmt/sessions/gmt_session.79328/gmt.3.cpt file cpt

[PaulWessel]

Thanks, that gmt.frame message is the information I need: Looks like that file needs to get the same treatment as other files and be per-figure, per-subset, etc. It is using that file from the 2nd plot which is unrelated. I will start a separate fix.