Wikidata : Add the family name property of a person if missing

create_family_name.py

"""Add the family name property of a person if missing

For a person, check if the family name exists.
If not, create it and link it to the person
"""

import sys
import pywikibot
from pywikibot import pagegenerators

_VERBOSE = True

# Iterate over several persons (or itemsID)
itemsID = (
           "Q8023",  # Nelson Mandela
           )

INSTANCE_OF_PROP = "P31"
FAMILY_NAME_PROP = "P734"
HUMAN_ITEM = "Q5"
FAMILY_NAME_ITEM = "Q101352"


def get_family_name_item(repo, itemtitle):
    """get family name item from string

    "repo" is the repo for pywikibot.
    "itemtitle" is the string to match.

    Returns ID if found.
    """
    sparql = "SELECT ?item WHERE {{ ?item rdfs:label '{}'@en .\
              ?item wdt:{} wd:{} }}".format(itemtitle, INSTANCE_OF_PROP, FAMILY_NAME_ITEM)
    entities = pagegenerators.WikidataSPARQLPageGenerator(sparql, site=repo)
    entities = list(entities)
    if len(entities) > 1:
        raise ValueError("Multiple family name instance with the same label in Wikidata")
    elif len(entities) == 1:
        return entities[0].getID()
    else:
        return


def query_yes_no_or_skip(question, default="yes"):
    """Ask a yes/no question via raw_input() and return their answer.

    "question" is a string that is presented to the user.
    "default" is the presumed answer if the user just hits <Enter>.
        It must be "yes" (the default), "no" or None (meaning
        an answer is required of the user).

    The "answer" return value is True for "yes" or False for "no".
    """
    valid = {"yes": 'y', "y": 'y', "ye": 'y',
             "no": 'n', "n": 'n',
             "skip": 's', "s": 's'}
    if default is None:
        prompt = " [y/n/s/?] " # Question mark means 'show me more explanation'
    elif default == "yes":
        prompt = " [Y/n/s/?] "
    elif default == "no":
        prompt = " [y/N/s/?] "
    else:
        raise ValueError("invalid default answer: '%s'" % default)

    while True:
        sys.stdout.write(question + prompt)
        choice = input().lower()
        if default is not None and choice == '':
            return valid[default]
        elif choice in valid:
            return valid[choice]
        else:
            sys.stdout.write("Please respond with 'yes' or 'no' or 'skip'"
                             "(or 'y' or 'n' or 's').\n")


def query_yes_no(question, default="yes"):
    """Ask a yes/no question via raw_input() and return their answer.

    "question" is a string that is presented to the user.
    "default" is the presumed answer if the user just hits <Enter>.
        It must be "yes" (the default), "no" or None (meaning
        an answer is required of the user).

    The "answer" return value is True for "yes" or False for "no".
    """
    valid = {"yes": True, "y": True, "ye": True,
             "no": False, "n": False}
    if default is None:
        prompt = " [y/n] "
    elif default == "yes":
        prompt = " [Y/n] "
    elif default == "no":
        prompt = " [y/N] "
    else:
        raise ValueError("invalid default answer: '%s'" % default)

    while True:
        sys.stdout.write(question + prompt)
        choice = input().lower()
        if default is not None and choice == '':
            return valid[default]
        elif choice in valid:
            return valid[choice]
        else:
            sys.stdout.write("Please respond with 'yes' or 'no' "
                             "(or 'y' or 'n').\n")


site = pywikibot.Site("wikidata", "wikidata")
repo = site.data_repository()

for itemID in itemsID:
    person_item = pywikibot.ItemPage(repo, itemID)

    person_dict = person_item.get()
    person_label = person_dict['labels']['en']
    clm_dict = person_dict["claims"]
    if FAMILY_NAME_PROP in clm_dict:
        # Family name exists, nothing to do
        clm_list = clm_dict[FAMILY_NAME_PROP]

        for clm in clm_list:
            clm_trgt = clm.getTarget()
            fn_label = clm_trgt.text['labels']['en']
            print(u"'{}' is the family name indicated for '{}'\n".format(fn_label, person_label))
    else:
        # Check first that the item is an instance of 'human'
        humanity_validated = False
        if INSTANCE_OF_PROP in clm_dict:
            for clm in clm_dict[INSTANCE_OF_PROP]:
                if HUMAN_ITEM == clm.getTarget().getID():
                    humanity_validated = True
        if humanity_validated:
            full_name = person_item.text['labels']['en']
            print(u"'{}' ({}) has no family name".format(full_name, itemID))
            # Take the last part of the name to guess the family name (e.g. 'Doe' from 'John Doe')
            proposed_family_name = full_name.split()[-1]
            # Check (by a human) that family name is correct
            user_input = query_yes_no_or_skip(u"Is '{}' OK ?".format(proposed_family_name))
            if user_input == 'y':
                print(u"You validate '{}'".format(proposed_family_name))
                family_name_str = proposed_family_name
            elif user_input == 's':
                print('') # Add a new line
                continue
            else:
                while True:
                    # proposed_family_name was not accepted, enter a new one
                    new_name = input("Enter a new name: ")
                    if query_yes_no(u"You entered '{}', do you validate ?".format(new_name)):
                        family_name_str = new_name
                        break

            # Create the item for his/her family name if it doesn't exist
            name_item_id = get_family_name_item(repo, family_name_str)

            if name_item_id :
                print(u"Name '{}' already exists!".format(family_name_str))
            else:
                print(u"'{}' doesn't exist yet!".format(family_name_str))
                # name_item is not defined, create it!
                name_item = pywikibot.ItemPage(site)

                # create the labels
                name_labels = {}
                for language in ('en', 'fr', 'nl', 'de'):
                    name_labels[language] = family_name_str
                name_item.editLabels(name_labels, summary='Set labels')

                # create the descriptions
                name_descriptions = {'en': 'family name',
                                     'fr': 'nom de famille',
                                     'nl': 'achternaam',
                                     'de': 'Familienname'}
                name_item.editDescriptions(name_descriptions, summary='Setting descriptions.')

                # 'name_item' is an instance of 'FAMILY_NAME_ITEM'
                claim = pywikibot.Claim(repo, INSTANCE_OF_PROP)
                target = pywikibot.ItemPage(repo, FAMILY_NAME_ITEM)
                claim.setTarget(target)
                name_item.addClaim(claim, summary=u'Adding claim "instance of family name"')

                name_item_id = name_item.getID()

                print(u"'{}' created!".format(family_name_str))

            # family name of 'person_item' is 'name_item'
            claim = pywikibot.Claim(repo, FAMILY_NAME_PROP)
            target = pywikibot.ItemPage(repo, name_item_id)
            claim.setTarget(target)
            person_item.addClaim(claim, summary=u'Adding claim "has family name"')
            print(u"'{}' has been linked to '{}'\n".format(family_name_str, person_label))
        else:
            print(u"'{}' is not an instance of 'human'\n".format(person_label))