Parser/CSV_to_RDF/CSV_to_RDF_ASPOPeople.py

# Utilities to read/write csv files
import csv
import re
# 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/alessiaspadi/Documents/RESTORE/temp_ASPO/'
export_dir = '/Users/alessiaspadi/Documents/RESTORE/temp_ASPO/'


# 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
aspoCoords = RDFcoords('<http://www.archiviodistato.prato.it/accedi-e-consulta/aspoMV001/scheda/>', 'aspo:')
foafCoords = RDFcoords('<http://xmlns.com/foaf/0.1/>', 'foaf:')
cidocCoords = RDFcoords('<http://www.cidoc-crm.org/cidoc-crm/>', 'crm:')
schemaCoords = RDFcoords('<http://schema.org/>', 'schema:')
personCoords = RDFcoords('<http://www.w3.org/ns/person#>', 'person:')
nsCoords = RDFcoords('<http://www.w3.org/1999/02/22-rdf-syntax-ns#>', 'rdf:')
rdfsCoords = 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 ' + aspoCoords.prefix + ' ' + aspoCoords.uri + closeLine)
    output.write('@prefix ' + foafCoords.prefix + ' ' + foafCoords.uri + closeLine)
    output.write('@prefix ' + cidocCoords.prefix + ' ' + cidocCoords.uri + closeLine)
    output.write('@prefix ' + personCoords.prefix + ' ' + personCoords.uri + closeLine)
    output.write('@prefix ' + schemaCoords.prefix + ' ' + schemaCoords.uri + closeLine)
    output.write('@prefix ' + nsCoords.prefix + ' ' + nsCoords.uri + closeLine)
    output.write('@prefix ' + rdfsCoords.prefix + ' ' + rdfsCoords.uri + closeLine)

    output.write('\n')


filePrefix = 'Onomastica_'
fileType = 'Datini'
max_entries = 1000000000

with open(import_dir + filePrefix + fileType + '.csv', newline="") as csv_file, open(
        export_dir + filePrefix + fileType + '.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
        if row['entityType'] == 'person':

            id_aspo = row['recordId']

            #placeHolders
            aspoPlaceHolder = aspoCoords.prefix + id_aspo

            line = triple(aspoPlaceHolder,
                          nsCoords.prefix + 'type',
                          foafCoords.prefix + 'person') + closeLine
            output.write(line)
            line = triple(aspoPlaceHolder,
                          foafCoords.prefix + 'name',
                          '\"' + row['nameEntry@normal'] + '\"') + closeLine
            output.write(line)

            if row['nome proprio'] != '':
                name = row['nome proprio'].replace("\n", "")
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'givenName',
                              '\"' + name + '\"') + closeLine
                output.write(line)

            if row['nome di famiglia'] != '':
                familyName = row['nome di famiglia'].replace("\n", "")
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'familyName',
                              '\"' + familyName + '\"') + closeLine
                output.write(line)

            if row['Alias'] != '':
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'alternateName',
                              '\"' + row['Alias'] + '\"') + closeLine
                output.write(line)

            if row['genere'] != '':
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'gender',
                              '\"' + row['genere'] + '\"') + closeLine
                output.write(line)

            if row['patronimico/matronimico'] != '':
                pp = row['patronimico/matronimico']
                patronimyc = pp.replace("\n", "")
                line = triple(aspoPlaceHolder,
                              personCoords.prefix + 'patronymicName',
                              '\"' + patronimyc + '\"') + closeLine
                output.write(line)

            if row['occupation'] != '' and row['occupation'] != '\n':
                occ = row['occupation']
                occupation = re.sub(r'[^A-Za-z]', '', occ)
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'hasOccupation',
                              aspoCoords.prefix + occupation) + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + occupation,
                              nsCoords.prefix + 'type',
                              schemaCoords.prefix + 'Occupation') + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + occupation,
                              rdfsCoords.prefix + 'label',
                              '\"' + row['occupation'] + '\"') + closeLine
                output.write(line)

            if row['avo 1'] != '':
                avo1 = row['avo 1'].replace('di ', '')
                avo1card = re.sub(r'[^A-Za-z]', '', avo1)
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'relatedTo',
                              aspoCoords.prefix + avo1card) + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + avo1card,
                              nsCoords.prefix + 'type',
                              foafCoords.prefix + 'Person') + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + avo1card,
                              rdfsCoords.prefix + 'label',
                              '\"' + avo1 + '\"') + closeLine
                output.write(line)

            if row['avo 2'] != '':
                avo2 = row['avo 2'].replace('di ', '')
                avo2card = re.sub(r'[^A-Za-z]', '', avo2)
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'relatedTo',
                              aspoCoords.prefix + avo2card) + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + avo2card,
                              nsCoords.prefix + 'type',
                              foafCoords.prefix + 'Person') + closeLine
                output.write(line)
                line = triple(aspoCoords.prefix + avo2card,
                              rdfsCoords.prefix + 'label',
                              '\"' + avo2 + '\"') + closeLine
                output.write(line)

            if row['Qualifica'] != '':
                qq = row['Qualifica']
                qualifiche = []
                if '|' in qq:
                    qualifiche = qq.split('|')
                else:
                    qualifiche.append(qq)
                for qualifica in qualifiche:
                    honorific = qualifica.replace("\n", "")
                    line = triple(aspoPlaceHolder,
                                  schemaCoords.prefix + 'honorificPrefix',
                                  '\"' + honorific + '\"') + closeLine
                    output.write(line)

            if row['place_occupation_Qualifica'] != '':
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'workLocation',
                              '\"' + row['place_occupation_Qualifica'] + '\"') + closeLine
                output.write(line)

            if row['biogHist p'] != '':
                bio = row['biogHist p']
                biog = bio.replace("\n", " ")
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'description',
                              '\"' + biog + '\"') + closeLine
                output.write(line)


        output.write('\n')
        #
        #
        # Limit number of entries processed (if desired)
        if (ii > max_entries):
            break