hatvp

_stats

pays    -    90 values, 4 distincts
----
95.6% (86) FRANCE
2.2% (2) France
1.1% (1) ROYAUME-UNI
1.1% (1) 


categorieOrganisation.label    -    90 values, 7 distincts
----
42.2% (38) Organisation professionnelle
32.2% (29) Société commerciale
11.1% (10) Association
7.8% (7) Cabinet de conseil
3.3% (3) Chambre consulaire
2.2% (2) Syndicat
1.1% (1) Autres organisations


codePostal    -    90 values, 59 distincts
----
18.9% (17) 75008
4.4% (4) 75009
4.4% (4) 75010
3.3% (3) 75012
3.3% (3) 75017
3.3% (3) 75011
2.2% (2) 92100
2.2% (2) 75002
2.2% (2) 92506
1.1% (1) 97400


departement    -    90 values, 20 distincts
----
58.9% (53) 75
15.6% (14) 92
3.3% (3) 97
3.3% (3) 59
2.2% (2) 69
1.1% (1) 14
1.1% (1) 44
1.1% (1) 91
1.1% (1) 13
1.1% (1) 34


listSecteursActivites.label    -    319 values, 29 distincts
----
13.2% (42) Santé, sécurité sociale
9.4% (30) Travail, emploi, solidarité
7.8% (25) Numérique
7.2% (23) Banques, assurances, secteur financier
7.2% (23) Environnement
6.3% (20) Economie
6.0% (19) Fiscalité, finances publiques
5.6% (18) Concurrence, consommation
5.6% (18) Recherche, innovation
5.0% (16) Energie


listNiveauIntervention.label    -    216 values, 4 distincts
----
38.9% (84) National
28.2% (61) Local
23.1% (50) Européen
9.7% (21) Mondial

stats.py

import json
from collections import Counter

DATA = json.load(open('stock.json'))

def attrget(item, key):
    keys = key.split('.')
    for key in keys:
        item = item.get(key,'')
        if item == None:
            return
    return item

def stats(key=None, attrget=attrget, limit=10, inception=False, data=None, label=None):
    data = data if data else DATA
    def flat(arr):
        for x in arr:
            if type(x) == type([]):
                yield from flat(x)
            elif type(x) == str:
                yield x.strip()
            else:
                yield x

    c = Counter(flat([attrget(el,key) for el in data]))

    count_all = len(data)
    count_distinct = len(c)
    print()
    print(label if label else key,"   -   ",count_all,"values,",count_distinct,"distincts")
    print('----')

    for el,n in c.most_common(limit):
        p = n/count_all*100
        print("{:.1f}% ({}) {}".format(p,n,el))
    print()

    if inception:
        stats(key="---> "+key+'-ception', attrget=lambda i, k: i, data=c.values())

stats("pays")
stats("categorieOrganisation.label")
stats("codePostal")
stats("departement", lambda item, key: item['codePostal'][:2])

secteurs = []
for item in DATA:
    secteurs += item['activites']['listSecteursActivites']

stats('label', data=secteurs, label='listSecteursActivites.label')

interventions = []
for item in DATA:
    interventions += item['activites']['listNiveauIntervention']

stats('label', data=interventions, label='listNiveauIntervention.label')