Windbarbs supplement

[joa-quim]

This PR contains the code to plot windbarbs either from table data as well as two grids, à lá grdvector.

The code was a little reworked from that of https://sites.google.com/site/masahigaki/gmt/gmt-windbarb?authuser=0 with changes to make it compile with GMT master version and drop the complicated plugin recipe.

[PaulWessel]

Things to do:

Compare with psxy and note there needs to be changes to:

1. How the usage function is not using the new way with the GMT_Usage function (to handle variable terminal widths)
2. Probably something related to -C if no argument (modern mode stuff)
3. Maybe KEYS - did not look very carefully.
4. Perhaps RST stuff using our SYN-* macros.
5. ???

[joa-quim]

How the usage function is not using the new way with the GMT_Usage function (to handle variable terminal widths)

It does

Perhaps RST stuff using our SYN-* macros

It does, and I made several changes too.

It now builds without need to set anything in ConfigUser.cmake. That is, it lands directly in the supplements shared lib.
Docs are also visible if you do ninja docs_html; ninja install

[PaulWessel]

Ok I just browsed the code on GitHub and that used GMT_Message. Will try tomorrow

[joa-quim]

When you try, change the mode to +x of the two sh examples to stop that Code Validator failure. It ignores what I do on Windows.

[PaulWessel]

Done, but unless you have uncommitted changes, there is no use of GMT_Usage in these codes etc.

[PaulWessel]

And since this is a branch: Given how all the other supplements are (all "required" together) you should places those tests and rst in their final places now, i.e, test/windbarb/yourwtoscripts and in rst/source/supplements/windbarg and so on. It is OK to not have PS files from the test so far so no worries about dvc.

[joa-quim]

test/windbarb/yourwtoscripts

I just copied his 2 shell examples. They are not tests.

rst/source/supplements/windbarg

They are already in doc\rst\source\supplements\windbarbs

[joa-quim]

there is no use of GMT_Usage in these codes etc.

I don't even know how to use them. I find it to much pain for no gain.

[masahigaki]

Dear all,
It would be my great pleasure if the windbarb extension would be included in the GMT Supplement.

you should places those tests and rst in their final places now, i.e, test/windbarb/yourwtoscripts

Regarding test scripts, I rewrite the example scripts *barb_example.sh, and put them into test/windbarbs/ subdirectory on my private repository masahigaki/gmt@e6cb3f6
The shbang of the scripts is fixed to "#!/usr/bin/env bash" to pass the code validator.

I would appreciate if you could examine the changes.
Masakazu

[joa-quim]

Hi Masakazu,

Thanks for that and sorry for not having been more clear. When I said some examples, I was thinking 3 or 4. Just the enough to test the different capabilities (~ps vs grd, with and without a CPT, a 3D case). Not all of them because our test set already takes quite some time to run. The other thing is that we want to avoid having to store the data in our cache system, so we need to generate it for the tests, which your examples already do. I added just one test in this branch to show what I mean. If you could add a minimal set of other representative cases, that would be nice.

Joaquim

[masahigaki]

Dear Joaquim,

Thank you for the clarification and I am sorry for my delayed response.
As per your suggestion, I chose four scripts for windbarbs test, removed data files in the test directory, and updated the scripts to generate data.
Please look at masahigaki/gmt@9e9133aa

[masahigaki]

I have rewrited {grd,ps,wind}barb.c to use GMT_Usage instead of GMT_Message at masahigaki/gmt@7519afbf .
Your confirmation would be appreciated.

[joa-quim]

Dear Mazakazu,

Thank you very much for these contributions. I've integrated them (plus the extra tests) in the official branch.

One further thing. We could drop the -T option from psbarb, at least from the documentation. The rational is that psxy already offers that option (also called -T) and in grdbarb -T means another thing.

[joa-quim]

@PaulWessel I think this is good for merging.

[github-actions]

Summary of changed images

This is an auto-generated report of images that have changed on the DVC remote

Status	Path
added	test/baseline/windbarbs/
added	test/baseline/windbarbs/grdbarb_01.ps
added	test/baseline/windbarbs/grdbarb_02.ps
added	test/baseline/windbarbs/psbarb_01.ps
added	test/baseline/windbarbs/psbarb_02.ps

Image diff(s)

Added images

• test/baseline/windbarbs/grdbarb_01.png

• test/baseline/windbarbs/grdbarb_02.png

• test/baseline/windbarbs/psbarb_01.png

• test/baseline/windbarbs/psbarb_02.png

Modified images

Path	Old	New

Report last updated at commit 14e78f7

[PaulWessel]

Think it is ready now.

[PaulWessel]

Tests work, docs look good, -T gone from sight, C-fixes seems fine. Have a look too.

[joa-quim]

I see that the psbarb.rst includes barb.rst but no manual is included. The page has synopsis and jumps to Examples.

[PaulWessel]

I forgot - will look in < 1 hour

[PaulWessel]

Had forgotten to do what is also done for psmeca.rst and meca.rst.

[PaulWessel]

psbarb is fixed

[joa-quim]

Ok, let's give this light.

[masahigaki]

Thank you very much for fixing the windbarbs code and merging it into master.

[joa-quim]

Hi Masakazu, If you have a nice windbarbs map, kind of official looking with meteorological agency logos and so, it would be a nice post in the Forum Showcase

[PaulWessel]

Hi Masahigaki-san. Thanks for your contribution - we have update some of the code, scripts, and docs to fit with the rest of the supplements. Because of such changes, please be on the lookout for any odd behavior so we can fix it quickly before the upcoming 6.5 release.

[masahigaki]

Paul-san,
Thank you again for adding the windbarbs code into GMT Supplement.

I think "2-D" should be added to

gmt/doc/rst/source/supplements/module_supplements_purpose.rst_

Line 103 in 74a5caa

.. |barb_purpose| replace:: Plot wind barbs in 3-D

as well as in

gmt/src/windbarbs/psbarb.c

Line 33 in 74a5caa

#define THIS_MODULE_PURPOSE "Plot wind barbs in 2-D and 3-D"

Regards,
Masakazu

[PaulWessel]

Yes, I had already done the update (it is a define statement in the C code) but forgot to run the script that builds those rst includes. Thanks, now updated in master