code/separate_posneg.py

import json
from tqdm import tqdm

if __name__ == '__main__':
    import sys
    fn = sys.argv[1]
    d = json.load(open(fn))

    lpos = []
    lneg = []

    for n, it in enumerate(d):
        print(f'processing {n+1}th...')

        try:
            if it['Ion_Mode'].strip().lower() == 'negative':
                lneg.append(it)
            else:
                lpos.append(it)
        except:
            if it['species'].strip().endswith('-'):
                lneg.append(it)
            else:
                lpos.append(it)

    print(f'Len lpos = {len(lpos)}, len lneg = {len(lneg)}, sum = {len(lpos)+len(lneg)}')

    json.dump(lpos, open(fn.replace('.json', '-pos.json'), 'w'), indent=2)
    json.dump(lneg, open(fn.replace('.json', '-neg.json'), 'w'), indent=2)