Parser/MPP/CSV_to_RDF/CSV_to_RDF_Ruoli.py

# Utilities to read/write csv files
import csv
# Utilities to handle character encodings
import unicodedata
# Ordered Dicts
from collections import OrderedDict

import json

# OPZIONAL IMPORTS

# For timestamping/simple speed tests
from datetime import datetime
# Random number generator
from random import *
# System & command line utilities
import sys
# Json for the dictionary
import json

import_dir = '/Users/federicaspinelli/TEAMOVI/Parser/DATA/MPP/CSV/corretti/'
export_dir = '/Users/federicaspinelli/TEAMOVI/Parser/DATA/MPP/RDF/'
# Custom class to store URIs + related infos for the ontologies/repositories

class RDFcoords:
    def __init__(self, uri, prefix, code=None):
        self.uri = uri
        self.prefix = prefix
        self.code = code


# Repositories
museoCoords = RDFcoords('<https://palazzopretorio.prato.it/it/le-opere/alcuni-capolavori/>', 'mpp:')
autCoords = RDFcoords('<https://palazzopretorio.prato.it/it/opere/autori/>', 'aut:')
cidocCoords = RDFcoords('<http://www.cidoc-crm.org/cidoc-crm/>', 'crm:')
aatCoords = RDFcoords('<http://vocab.getty.edu/aat/>', 'aat:')
nsCoords = RDFcoords('<http://www.w3.org/1999/02/22-rdf-syntax-ns#>', 'rdf:')
schemaCoords = RDFcoords('<http://www.w3.org/2000/01/rdf-schema#>', 'rdfs:')


# Basic functions for triples / shortened triples in TTL format

def triple(subject, predicate, object1):
    line = subject + ' ' + predicate + ' ' + object1
    return line


def doublet(predicate, object1):
    line = '    ' + predicate + ' ' + object1
    return line


def singlet(object1):
    line = '        ' + object1
    return line


# Line endings in TTL format
continueLine1 = ' ;\n'
continueLine2 = ' ,\n'
closeLine = ' .\n'


def writeTTLHeader(output):
    output.write('@prefix ' + museoCoords.prefix + ' ' + museoCoords.uri + closeLine)
    output.write('@prefix ' + cidocCoords.prefix + ' ' + cidocCoords.uri + closeLine)
    output.write('@prefix ' + aatCoords.prefix + ' ' + aatCoords.uri + closeLine)
    output.write('@prefix ' + schemaCoords.prefix + ' ' + schemaCoords.uri + closeLine)
    output.write('@prefix ' + nsCoords.prefix + ' ' + nsCoords.uri + closeLine)
    output.write('@prefix ' + autCoords.prefix + ' ' + autCoords.uri + closeLine)

    output.write('\n')


filePrefix = 'AR20AUT_'
fileType = 'Ospedale'
max_entries = 1000000000


def get_role(role):
    role_file = open('/Users/federicaspinelli/TEAMOVI/Parser/DATA/MPP/CSV/AAT_RUOLI.csv', newline="")
    reader = csv.DictReader(role_file)
    for row in reader:
        if row['Label'] == role:
            return row['AAT']


with open(import_dir + filePrefix + fileType + '.csv', newline="") as csv_file, open(
        export_dir + filePrefix + fileType + '_roles.ttl', 'w') as output:
    reader = csv.DictReader(csv_file)
    writeTTLHeader(output)
    first = True
    ii = 0
    for row in reader:
        # The index ii is used to process a limited number of entries for testing purposes
        ii = ii + 1

        url = row['URL']

        # placeHolders
        datplaceHolder = museoCoords.prefix + url

        # E12 - PC14 - E21
        if row['AUTH'] != '':
            aut_role = ''
            if row['AUTQ'] != '':
                aut_role = row['AUTQ']
                ll = row['AUTN'] + '_' + aut_role
                lab = ll.replace(' ', '')
                label = lab.replace(',', '')
                line = triple(museoCoords.prefix + '_' + label,
                          schemaCoords.prefix + 'label',
                          '\"' + row['AUTN'] + ' nel ruolo di artista' + '\"') + closeLine
                output.write(line)
            else:
                aut_role = 'artista'
                ll = row['AUTN'] + '_' + 'artista'
                lab = ll.replace(' ', '')
                label = lab.replace(',', '')
                line = triple(museoCoords.prefix + '_' + label,
                          schemaCoords.prefix + 'label',
                          '\"' + row['AUTN'] + ' nel ruolo di artista' + '\"') + closeLine
                output.write(line)

            AuthorPlaceholder = autCoords.prefix + row['URL']

            line = triple(museoCoords.prefix + '_' + label,
                          nsCoords.prefix + 'type',
                          cidocCoords.prefix + 'PC14_carried_out_by') + closeLine
            output.write(line)

            line = triple(museoCoords.prefix + '_' + label,
                          cidocCoords.prefix + 'P02_has_range',
                          AuthorPlaceholder) + closeLine
            output.write(line)

            if aut_role != '' and aut_role != 'artista':
                role = get_role(aut_role)
                line = triple(museoCoords.prefix + '_' + label,
                              cidocCoords.prefix + 'P14.1_in_the_role_of',
                              aatCoords.prefix + role) + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + role,
                              nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E55_Type') + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + role,
                              schemaCoords.prefix + 'label',
                              '\"' + aut_role + '\"') + closeLine
                output.write(line)

            else:
                role = 'artista'
                line = triple(museoCoords.prefix + '_' + label,
                              cidocCoords.prefix + 'P14.1_in_the_role_of',
                              aatCoords.prefix + role) + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + role,
                              nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E55_Type') + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + role,
                              schemaCoords.prefix + 'label',
                              '\"' + aut_role + '\"') + closeLine
                output.write(line)


        output.write('\n')

        #
        #
        # Limit number of entries processed (if desired)
        if (ii > max_entries):
            break