[TEP007] Add cmfgen2tardis script

[vg3095]

This script takes a CMFGEN file as an input , and an output path to save the converted CSV file in new TARDIS file format.
An example of input CMFGEN file (Link) and its converted CSV file (Link)
Example, for how to use-
Format - cmfgen2tardis /path/to/input_file path/to/output/
cmfgen2tardis tardis_example/DDC15_SN_HYDRO_DATA_0.976d tardis_example/

---

• Now velocity, density, electron_density and temperature are also saved.

• Values are stored in reverse order, to match how Tardis stores things. For ex - velocity values in Tardis are stored in increasing order, while in Cmfgen files , it is in decreasing order .

[vg3095]

@wkerzendorf @yeganer When I leave __init__.py file empty, travis fails, because it skips this file.
(warning: build_scripts: scripts/__init__.py is an empty file (skipping)). Travis Report.
Can this be avoided ?.

[wkerzendorf]

I don't think it works that way. look at the link I sent you

[vg3095]

@wkerzendorf I think, you are saying to create a __main__.py file under scripts, as in the link you sent. If we do , as in the link you sent, we may be able to do something like this -

tardis_scripts cmfgen2tardis --input_file=/path/file --output_file=/tmp/
tardis_scripts get_spectrum --gdb

So, first argument tardis_scripts will be fixed, meaning we want to run scripts, second argument will be the script we want to run, and rest will be arguments required for individual scripts.

By my current approach
cmfgen2tardis /path/file /tmp/
will work, and for this we do not require a __main__.py file.Stackoverflow

So do you want the 1st way or 2nd way ?

[unoebauer]

Would be good to have a more detailed description of the scope of this PR... Is it taking CMFGEN config files and then generating appropriate TARDIS inputs or is it supposed to do something else?

[vg3095]

@unoebauer I have updated the description for PR now.

[wkerzendorf]

@vg3095 I don't think these should live within scripts outside of the package. I think you also misunderstood the entrypoints idea. Essentially it will generate a script based on calling a "main" function that takes args. so you wrote it like a script but that is not the right approach. Write is as a function that takes args.

[wkerzendorf]

@vg3095 can you also link a CMFGEN model?

[vg3095]

@wkerzendorf I have now added a CMFGEN file in the description

[wkerzendorf]

@vg3095 once this works - we also need to read the velocity, temperature, atom density and electron density.

[wkerzendorf]

I think they entry point actually gives args to the function. So we might want to use that.

[vg3095]

@wkerzendorf See these links

• Stackoverflow. First and second answer. Here it is mentioned

setuptools console_scripts entry point wants a function of no arguments.

• Blog. Here also entry point function is used without args.

[vg3095]

If I am missing something very obvious here, which I can't see, then please point it out.
I also saw runner script , here run method is used as an entry point function and is used without passing args.

[wkerzendorf]

@vg3095 actually, I think its close. Let's do some documentation and then I think we can merge and work on getting the velocity and other quantities in there.

[vg3095]

@wkerzendorf I have now added doc for the script.

[vg3095]

@wkerzendorf As you said in PR #772 , if we are reversing velocity, then all other columns should also be reversed, so that means , while saving abundances from CMFGEN file, we should also reverse them, -- right?

[vg3095]

• Now velocity, density, electron_density and temperature are also saved as DAT file.
• Since, we are reversing velocity, so reversing all columns along with it.

[vg3095]

Changed script, to store all quantities in a single CSV file. Example of a converted file --> Link