Parser/ASPO/CSV_to_RDF/datini/CSV_to_RDF_onomastica_datini.py

#Parser to convert the Datini onomastics CSV file into TTL format

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

import json
import re


# 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/DallASPO/'
export_dir = '/Users/federicaspinelli/TEAMOVI/Parser/Data/DallASPO/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
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:')
owlCoords = RDFcoords('<http://www.w3.org/2002/07/owl#>', 'owl:')

# 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('@prefix ' + owlCoords.prefix + ' ' + owlCoords.uri + closeLine)
    output.write('\n')


filePrefix = 'onomastica_'
fileType = 'datini'
max_entries = 10000000000000

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']
            aspoPlaceHolder = aspoCoords.prefix + id_aspo
            line = triple(aspoPlaceHolder, 
                          nsCoords.prefix + 'type', 
                          cidocCoords.prefix + 'E21_Person') + closeLine
            output.write(line)

            line = triple(aspoPlaceHolder, 
                          nsCoords.prefix + 'type', 
                          personCoords.prefix + 'Person') + closeLine
            output.write(line)

            line = triple(aspoPlaceHolder,
                          nsCoords.prefix + 'type',
                          foafCoords.prefix + 'person') + closeLine
            output.write(line)

            line = triple(aspoPlaceHolder,
                          cidocCoords.prefix + 'P1_is_identified_by',
                          aspoPlaceHolder + "_E42") + closeLine
            output.write(line)

            line = triple(aspoPlaceHolder + "_E42",
                          nsCoords.prefix + 'type',
                          cidocCoords.prefix + 'E42_Identifier') + closeLine
            output.write(line)
            line = triple(aspoPlaceHolder + "_E42",
                          rdfsCoords.prefix + 'label',
                          '\"' + id_aspo + '\"') + closeLine
            output.write(line)

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

            line = triple(aspoPlaceHolder,
                          rdfsCoords.prefix + 'label',
                          '\"' + row['nameEntry@normal'] + '\"') + closeLine
            output.write(line)

            if row['nome proprio'] != '':
                #Remove all white-space characters:
                txt = row['nome proprio']
                x = re.sub(" \n", "", txt)
                y = re.sub("\s\s", "", x)
                name = re.sub("\n", "", y)
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'givenName',
                              '\"' + name + '\"') + closeLine
                output.write(line)

            if row['nome di famiglia'] != '':
                #Remove all white-space characters:
                txt = row['nome di famiglia']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", "", x)
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'familyName',
                              '\"' + y + '\"') + closeLine
                output.write(line)
            
            if row['Alias'] != '' and row['Alias'] != ' ':
                #Remove all white-space characters:
                txt = row['Alias']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", "", x)
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'alternateName',
                              '\"' + y + '\"') + closeLine
                output.write(line)

            if row['genere'] != '':
                #Remove all white-space characters:
                txt = row['genere']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", "", x)
                line = triple(aspoPlaceHolder,
                              foafCoords.prefix + 'gender',
                              '\"' + y + '\"') + closeLine
                output.write(line)

            if row['patronimico/matronimico'] != '':
                #Remove all white-space characters:
                txt = row['patronimico/matronimico']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", "", x)
                line = triple(aspoPlaceHolder,
                              personCoords.prefix + 'patronymicName',
                              '\"' + y + '\"') + closeLine
                output.write(line)

            if row['occupation'] != '' and row['occupation'] != ' ' :
                occupazioni = []
                pipe = "|"
                if pipe in row['occupation']:
                    occupazioni = row['occupation'].split('|') 
                    for occupazione in occupazioni:
                        #Remove all white-space characters:
                        txt = occupazione
                        x = re.sub("\n", " ", txt)
                        y = re.sub("\s\s", "", x)
                        occ = re.sub(r'[^A-Za-z]','', y)
                        occupationPlaceHolder = '<http://www.archiviodistato.prato.it/' + occ.replace(" ","_") + '>'
                        line = triple(aspoPlaceHolder,
                                    schemaCoords.prefix + 'hasOccupation',
                                    occupationPlaceHolder) + closeLine
                        output.write(line)
                        line = triple(occupationPlaceHolder,
                                    nsCoords.prefix + 'type',
                                    schemaCoords.prefix + 'Occupation') + closeLine
                        output.write(line)
                        line = triple(occupationPlaceHolder,
                                    rdfsCoords.prefix + 'label',
                                    '\"' + y + '\"') + closeLine
                        output.write(line)
                else:
                    #Remove all white-space characters:
                    txt = row['occupation']
                    x = re.sub("\n", " ", txt)
                    y = re.sub("\s\s", "", x)
                    occ = re.sub(r'[^A-Za-z]','', y)
                    occupationPlaceHolder = '<http://www.archiviodistato.prato.it/' + occ.replace(" ","_") + '>'
                    line = triple(aspoPlaceHolder,
                                schemaCoords.prefix + 'hasOccupation',
                                occupationPlaceHolder) + closeLine
                    output.write(line)
                    line = triple(occupationPlaceHolder,
                                nsCoords.prefix + 'type',
                                schemaCoords.prefix + 'Occupation') + closeLine
                    output.write(line)
                    line = triple(occupationPlaceHolder,
                                rdfsCoords.prefix + 'label',
                                '\"' + y + '\"') + closeLine
                    output.write(line)

            if row['avo 1'] != '':
                avo1 = '<http://www.archiviodistato.prato.it/accedi-e-consulta/aspoMV001/scheda/' + id_aspo + "/avo1>"
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'relatedTo',
                              avo1) + closeLine
                output.write(line)
                line = triple(avo1,
                              nsCoords.prefix + 'type',
                              foafCoords.prefix + 'Person') + closeLine
                output.write(line)
                line = triple(avo1,
                              rdfsCoords.prefix + 'label',
                              '\"' + row['avo 1'] + '\"') + closeLine
                output.write(line)

            if row['avo 2'] != '':
                avo2 = '<http://www.archiviodistato.prato.it/accedi-e-consulta/aspoMV001/scheda/' + id_aspo + "/avo2>"
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'relatedTo',
                              avo2) + closeLine
                output.write(line)
                line = triple(avo2,
                              nsCoords.prefix + 'type',
                              foafCoords.prefix + 'Person') + closeLine
                output.write(line)
                line = triple(avo2,
                              rdfsCoords.prefix + 'label',
                              '\"' + row['avo 2'] + '\"') + closeLine
                output.write(line)

            if row['Qualifica'] != '':
                qualifiche = []
                pipe = "|"
                if pipe in row['Qualifica']:
                    qualifiche = row['Qualifica'].split('|') 
                    for qualifica in qualifiche:
                        #Remove all white-space characters:
                        txt = qualifica
                        x = re.sub("\n", " ", txt)
                        y = re.sub("\s\s", " ", x)
                        line = triple(aspoPlaceHolder, schemaCoords.prefix + 'honorificPrefix', '\"' + str(y) + '\"') + closeLine
                        output.write(line)
                else:
                #Remove all white-space characters:
                    txt = row['Qualifica']
                    x = re.sub("\n", " ", txt)
                    y = re.sub("\s\s", " ", x)
                    line = triple(aspoPlaceHolder, schemaCoords.prefix + 'honorificPrefix', '\"' + y + '\"') + closeLine
                    output.write(line)

            if row['place_occupation_Qualifica'] != '':
                #Remove all white-space characters:
                txt = row['place_occupation_Qualifica']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", "", x)
                line = triple(aspoPlaceHolder,
                              schemaCoords.prefix + 'workLocation',
                              '\"' + row['place_occupation_Qualifica'].replace('\\','\\\\').replace('"','\\"') + '\"') + closeLine
                output.write(line)

            if row['biogHist p'] != '':
                #Remove all white-space characters:
                txt = row['biogHist p']
                x = re.sub("\n", " ", txt)
                y = re.sub("\s\s", " ", x)
                note = re.sub("\"", "", x)
                line = triple(aspoPlaceHolder,
                              cidocCoords.prefix + 'P3_has_note',
                              '\"' + note + '\"') + closeLine
                output.write(line)
            
            if row['Variante'] != '': 
                varianti = []
                pipe = "|"
                if pipe in row['Variante']:
                    varianti = row['Variante'].split('|')
                    for variante in varianti: 
                        line = triple(aspoPlaceHolder,
                        owlCoords.prefix + 'sameAs',
                        aspoCoords.prefix + str(variante)) + closeLine
                        output.write(line)
                else:
                    line = triple(aspoPlaceHolder,
                    owlCoords.prefix + 'sameAs',
                    aspoCoords.prefix + row['Variante']) + closeLine
                    output.write(line)
            
            output.write('\n')
        #
        # Limit number of entries processed (if desired)
        if (ii > max_entries):
            break