run.py

"""
Input: disease_catalog.csv and disease.csv

Output: 

AntDesign multilevel selector

[{
  "value": "A00-B99",
  "label": "某些传染病和寄生虫病",
  "children": [{
    "value": "A00-A09",
    "label": "肠道传染病",
    "children": [{
      "value": "A00",
      "label": "霍乱",
    }],
    ...
  }],
  ...
}]
"""

import csv
import json

cat1_children = {}  # {cat1_code: children_list}
cat2_children = {}  # {cat2_code: children_list}
res = []            # final array for jsonify


def in_code_range(code, code_range):
    start, end = code_range.split('-')
    return start <= code <= end

with open('disease_catalog.csv') as file:
    reader = csv.DictReader(file, delimiter='\t')
    for row in reader:
        code = row['code_lower_bound'] + '-' + row['code_upper_bound']
        if row['level'] == "1":
            _children = []
            cat1_children[code] = _children
            res.append({
                'value': code,
                'label': row['catalog'],
                'children': _children,
            })
            continue
        cat2_children[code] = []
        _children.append({
            'value': code,
            'label': row['catalog'],
            'children': cat2_children[code],
        })

with open('disease.csv') as file:
    reader = csv.DictReader(file, delimiter='\t')
    for d in reader:
        for k, v in cat2_children.items():
            if in_code_range(d['code'], k):
                v.append({
                    'value': d['code'],
                    'label': d['disease'],
                })
                break
        else:
            # not child of any cat2, add it to matched cat1
            for k, v in cat1_children.items():
                if in_code_range(d['code'], k):
                    v.append({
                        'value': d['code'],
                        'label': d['disease'],
                    })
                    break

with open('diseases.json', 'w', encoding='utf-8') as file:
    file.write(u'\uFEFF')  # add UTF-8 BOM
    json.dump(res, file, ensure_ascii=False, separators=(',', ':'))