Transform dic_lipids.py into json files

[HubLot]

After thorough discussion, it will be better to transform the dic_lipids.py into json files.
This dic_lipids.py is used to reconstructed hydrogens from a united atom bilayer.

This will allow the users to add their own .json files for their own trajectories.
A regular database of json will be created in the buildH source code into a directory called lipids to provide a set of supported lipids.
We will add an option to the program for user to provided their own json.

Steps:

• Convert the current dic_lipids.py into a series of .json files into buildh/lipids directory
• Transform the code to read those json files
• Provide a list of supported lipids when using the -h of the program (by parsing the directory)
• Add an argument for users to provide their own .json files.
• Integrate the lipids from NMRLipids@d0a2bc1 (issue Add json files for other UA force fields and lipids #44)

[patrickfuchs]

I reopen this issue because we're not done yet. I checked what's been done above. I think the most urgent thing is to implement Provide a list of supported lipids when using the -h. We can discuss it this afternoon.

[HubLot]

It's done ;)
It's within the epilog of the ArgumentParser constructor:

buildH/buildh/cli.py

Lines 132 to 134 in 6c79661

	epilog = f"The list of supported lipids (-l option) are: {lipids_supported_str}."
	parser = argparse.ArgumentParser(description=message,
	epilog=epilog)

$ buildH -h
[...]
  -e END, --end END     The last frame (ps) to read from the trajectory.
  -pi PICKLE, --pickle PICKLE
                        Output pickle filename. The structure pickled is a dictonnary containing for each Order
                        parameter, the value of each lipid and each frame as a matric

The list of supported lipids (-l option) are: CHARMM_POPC, Berger_POP, Berger_POPC, Berger_PLA.

[HubLot]

but the issue #44 is not done.

[patrickfuchs]

Oops sorry 😁... Great, so I can incorporate that into the first notebook. I checked the corresponding checkboxes up there.
So indeed, what is left to do is adding new lipids (#44). Almost there!

[patrickfuchs]

We close this issue since it's redundant with #44 at this stage.