Add longoptions support for gmtspatial. #8065
Conversation
Use factor instead of fact for clarity. Added missing +s description for -D documentation and added alias limits to range since latter is not good for PyGMT. Rest looks fine to me.
|
Well, I agree with your -S |
src/longopt/gmtspatial_inc.h
Outdated
| { 0, 'T', "", | ||
| "", "", | ||
| { 0, 'S', "spatial", | ||
| "b,h,i,j,s,u", "buffer,hole,intersection,join,split,union", |
There was a problem hiding this comment.
The directives (some) are also different in Julia. GMT.jl has holes instead of hole and dateline instead of split. I think dateline is better because it clearly indicates where the split occurs.
There was a problem hiding this comment.
Tend to agree with this. Could do alias holes here but I like dateline.
There was a problem hiding this comment.
We can add alias holes but not that all the others are singular nouns so holes stand out.
There was a problem hiding this comment.
Best if GMT.jl now adds the hole alias to match CLI. I did the dateline change and added alias holes.
Use dateline and hole with holes as alias
|
Right, but what it detects are |
|
I see it is unclear. I will rephrase. |
Improve documentation for -S directives.
|
I updated the -S documentation. Le the know if this is clearer: 
|
|
Yes, much better. But now you raised me another doubt. What is meant by reverses them? You mean make them of opposite wise of the parent polygon (clockwise or counter-clockwise)? |
Yes, any holes will get the opposite handedness as the parent (coco vs counter-clock). Il'' fix the docs for h to say
Any other things before I add this? |
|
No, sounds good. |
|
A bit dumb to list -Si, -Sj, -Su when not implemented (after many years). Leaves two options:
Those directives were clearly aspirational but here we are... I think there is a C++ library out there but perhaps there is a C library somewhere we could use? |
Description of proposed changes
Python differences:
No python module.
Julia differences:
Used in_polygons for -N instead of in_polyg (awkward abbreviation
and inconsistent with -F force_polygons).
Used spatial for -S instead of polygons. Almost all options are
related to polygons, so using 'polygons' alone is not helpful in
distinguishing between them.
Fixes #8064
Reminders