Minor bug fix in writeg and parse_gfile_header

[mgyoo86]

Minor bug fixes in

• runtest.jl
• parse_gfile_header
  • New default time in writeg with postfix, (e.g., "1ms") is parsed as Expr, which causes an error.
  • If time is Expr, convert it to String first, then the error is gone.

@bclyons12 could you please review the changes?
I cannot find the reviewers button for this PR..

Suggestion

If the GitHub CI action is added, it would be useful to automatically test runtest.jl.
The current Travis CI seems not to be triggered automatically.

[bclyons12]

@eldond Do you know why the "####ms" time style isn't being parsed properly? That's what all the sample gEQDSK files in OMFIT have, so I assume it's the default EFIT output format.

[mgyoo86]

Additional Fixes in imas2geqdsk

1. Correct 2D Flux Map Reading
   The code now uses findfirst(:rectangular, eqt.profiles_2d) to correctly read the 2D flux map from dd.

2. Interpolation of 1D Profiles on Rectangular Grids
   GEQDSK requires the number of 1D profiles to match nw (the number of horizontal R grid points). Since IMAS does not impose this restriction, 1D profiles are now interpolated when the grid is rectangular (i.e. when nw != nh) to satisfy the GEQDSK format.

3. New Test Dataset and Test Added
   A new test dataset and corresponding tests have been added to validate these changes.

@bclyons12 @eldond Could you please review the changes? Thanks!

[mgyoo86]

Minor fix in writeg

• Check the length of description before writing a file, which could prevent accidental overwriting of the file.

[orso82]

@mgyoo86 @bclyons12 @lstagner should this be merged and a new version released?

There's also the question of where to add documentation like @mgyoo86 example on Discord:

import EFIT
import EFIT.IMASdd

filename = joinpath(dirname(dirname(pathof(EFIT))), "test", "test_dd_eq.json")
dd = IMASdd.json2imas(filename);

# Convert dd to GEQDSK struct (Note: GEQDSK is not a file)
ggs = EFIT.imas2geqdsk(dd)

# Write into file
for gg in ggs
    file_name = gg.file  # default file name has "g00000.xxxxx" format
    EFIT.writeg(gg, file_name)
end

# You can specify a file name
EFIT.writeg(ggs[1], "my_equilibrium.geqdsk")

# specify description, shot, and time as keyword arguments
# NOTE: time must be in `ms` unit (You can just put a number in `ms` without postfix `ms`) 
EFIT.writeg(ggs[1], "my_equilibrium.geqdsk"; desc="my description", shot="my shot 12345", time="1000ms")

# If the description is too long, it will cause an error.
EFIT.writeg(ggs[1], "my_equilibrium.geqdsk"; desc="This too long description will cause error.")

[bclyons12]

Overall looks good, but I'd make the simplifying change I suggest and maybe add a test.

[bclyons12]

I have confirmed that the format produced by this PR can be opened in OMFIT, which for me, is the minimal requirement :-)

[mgyoo86]

@bclyons12
The followings are updated:

• Reduction of duplication for the interpolation of 1D profiles
• Added tests for the keywords in writeg

@orso82
Maybe, we can add some examples into README.md file?

[orso82]

@mgyoo86 I'd be in favor of that, just the basic functionality. We can then add a link to the regression tests for people to see any extra/optional features.

[bclyons12]

@mgyoo86 Great work. And I agree. In the Readme, just add a section on reading and writing gEQDSK files, including from IMAS.

[mgyoo86]

@orso82 @bclyons12
I updated README.md by adding an example of writing GEQDSK files from dd.
It seems this PR is now ready to be merged.

[orso82]

Fantastic! Thank you @mgyoo86
I'll release a new version of EFIT.jl