lexeme-maker.py

# License: GPL2+
import pywikibot
import json
import random
import string
import numpy as np
import math

site = pywikibot.Site('en', 'lexeme')
distro = np.random.normal(10, 5, 550)
distro_list = [math.ceil(abs(i)) for i in distro]

def request(params):
    if not site.logged_in():
        site.login()
    req = pywikibot.data.api.Request(site=site, **params)
    return req.submit()

def get_token():
    if not site.logged_in():
        site.login()
    return site.tokens['csrf']

def make_lexeme():
    params = {
        'action': 'wbeditentity',
        'new': 'lexeme',
        'data':  json.dumps({"lemmas": {"en": {"language": "en", "value": ''.join(random.choice(string.ascii_uppercase + string.digits + string.ascii_lowercase + ' ') for _ in range(20))}  },"lexicalCategory": "Q2", "language": "Q3" }),
        'token': get_token()
    }
    res = request(params)
    if 'success' in res:
        print(res['entity']['id'])

def new_guid():
    return '{:04X}{:04X}-{:04X}-{:04X}-{:04X}-{:04X}{:04X}{:04X}'.format(
        random.randint(0, 65535),
        random.randint(0, 65535),
        random.randint(0, 65535),
        random.randint(16384, 20479),
        random.randint(32768, 49151),
        random.randint(0, 65535),
        random.randint(0, 65535),
        random.randint(0, 65535)
    )

def add_statements_to_lexeme(lexeme_id):
    params = {
        'action': 'wbsetclaim',
        'claim': json.dumps(
            {"type":"statement","mainsnak":{"snaktype":"value","property":"P17","datavalue":{"type":"wikibase-entityid","value":{"id":"L%d" % random.randint(55,545)}}},"id":"%s$%s" % (lexeme_id, new_guid()),"rank":"normal"}
        ),
        'token': get_token()
    }
    res = request(params)
    if 'success' in res:
        print(lexeme_id, res['claim']['id'])




#for i in range(460):
#    try:
#        make_lexeme()
#    except:
#        continue

for i in range(55, 550):
    for j in range(distro_list[i-54]):
        try:
            add_statements_to_lexeme('L%d' % i)
        except:
            try:
                add_statements_to_lexeme('L%d' % i)
            except:
                continue