Parser/Museo/CSV_to_RDF/CSV_to_RDF_mpp.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/alessiaspadi/Documents/RESTORE/temp_MPP/tabelle/Ospedale/mod/'
export_dir = '/Users/alessiaspadi/Documents/RESTORE/temp_MPP/tabelle/Ospedale/mod/'


# 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('<http://palazzopretorio.comune.prato.it/it/le-opere/alcuni-capolavori/>', 'mpp:')

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('\n')


filePrefix = 'SR20OA_'
fileType = 'Ospedale'
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['RVEL'] == '' or row['RVEL'] == '0':

            sb = ''
            subj = ''
            pp = row['OGTD'] + ' (' + row['ACC'] + ') '
            if row['SGTI'] != '':
                sb = pp + row['SGTI']
            if row['LDCN'] != '':
                subj = sb + ' in ' + row['LDCN']
            else:
                subj = sb

            # Triplify the 'codice' -- should exist for every entry
            codice = ''
            if (row['NCTR'] != '' and row['NCTN'] != ''):
                codice = row['NCTR'] + row['NCTN']
            '''if (row['RVEL'] != ' '):
                codice = codice + "-" + row['RVEL']'''

            url = row['URL']

            # placeHolders
            datplaceHolder = museoCoords.prefix + url
            e42placeHolder = museoCoords.prefix + url + '_E42'
            e42CplaceHolder = museoCoords.prefix + url + '_E42_MPP'
            e73placeHolder = museoCoords.prefix + url + '_E73'
            e55placeHolder = museoCoords.prefix + url + '_E55'
            e35placeHolder1 = museoCoords.prefix + url + '_E35'
            e53placeHolder = museoCoords.prefix + url + '_E53'
            e1placeHolder = museoCoords.prefix + url + '_E1'
            e74placeHolder = museoCoords.prefix + url + '_E74'

            if (codice != ''):
                line = triple(datplaceHolder, cidocCoords.prefix + 'P1_is_identified_by', e42placeHolder) + closeLine
                output.write(line)
                line = triple(e42placeHolder, nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E42_Identifier') + closeLine
                output.write(line)
                '''line = triple(e42placeHolder, cidocCoords.prefix + 'P2_has_type',
                              aatCoords.prefix + '300404626') + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + '300404626', schemaCoords.prefix + 'label',
                              "identifier") + closeLine
                output.write(line)'''
                line = triple(e42placeHolder, schemaCoords.prefix + 'label',
                              '\"Codice univoco del bene: ' + codice + '\"') + closeLine
                output.write(line)

                '''
                # AS
                e55placeHolder = "<http://www.museodipalazzopretorio.it/" + codice + '/' + identifierCoords.code + ">"
                line = triple(e42placeHolder, hasTypePCoords.prefix, e55placeHolder) + closeLine
                output.write(line)
                line = triple(e55placeHolder, hasTypeCoords.prefix, typeCoords.prefix) + closeLine
                output.write(line)
                line = triple(e55placeHolder, labelCoords.prefix, '\"Codice univoco del bene\"') + closeLine
                output.write(line)
                # Fine AS
                '''

            # Write E22 Man Made Object & E73 Information Object -- should exist for every entry?
            line = triple(datplaceHolder, nsCoords.prefix + 'type',
                          cidocCoords.prefix + 'E22_Man-Made_Object') + closeLine
            output.write(line)
            # Added by AS
            line = triple(datplaceHolder, schemaCoords.prefix + 'label', '\"' + subj + '\"') + closeLine
            output.write(line)
            # End AS
            line = triple(datplaceHolder, cidocCoords.prefix + 'P128_carries', e73placeHolder) + closeLine
            output.write(line)
            line = triple(e73placeHolder, nsCoords.prefix + 'type',
                          cidocCoords.prefix + 'E73_Information_Object') + closeLine
            output.write(line)

            # AS
            ss = ''
            if row['SGTI'] != '':
                ss = row['SGTI']
            else:
                ss = 'senza titolo'

            line = triple(e73placeHolder, schemaCoords.prefix + 'label',
                          '\"Opera d\'arte raffigurante ' + ss + '\"') + closeLine
            output.write(line)

            # E73 - P2 - E55

            tt = ''
            typeLabel = ''

            if row['OGTD'] == 'dipinto':
                tt = aatCoords.prefix + "300033618"
            elif row['OGTD'] == 'rilievo':
                tt = aatCoords.prefix + "300047230"
            elif row['OGTD'] == 'polittico':
                tt = aatCoords.prefix + "300178235"
            elif row['OGTD'] == 'predella':
                tt = aatCoords.prefix + "300003745"
            else:
                tt = e55placeHolder

            line = triple(e73placeHolder,
                          cidocCoords.prefix + 'P2_has_type',
                          tt) + closeLine
            output.write(line)
            line = triple(tt, schemaCoords.prefix + 'label',
                          '\"' + row['OGTD'] + '\"') + closeLine
            output.write(line)

            # E73 - P1 - E35

            if row['SGTT'] != '':
                line = triple(e73placeHolder, cidocCoords.prefix + 'P1_is_identified_by', e35placeHolder1) + closeLine
                output.write(line)
                line = triple(e35placeHolder1, nsCoords.prefix + 'type', cidocCoords.prefix + 'E35_Title') + closeLine
                output.write(line)
                line = triple(e35placeHolder1, schemaCoords.prefix + 'label', '\"' + row['SGTT'] + '\"') + closeLine
                output.write(line)

                line = triple(e35placeHolder1, cidocCoords.prefix + 'P2_has_type',
                              aatCoords.prefix + "300417193") + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + "300417193", schemaCoords.prefix + 'label',
                              '\"titolo\"') + closeLine
                output.write(line)

            # E22 - P62 - E1

            if row['SGTI'] != '':
                line = triple(datplaceHolder,
                              cidocCoords.prefix + 'P62_depicts',
                              e1placeHolder) + closeLine
                output.write(line)
                line = triple(e1placeHolder,
                              nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E1_CRM_Entity') + closeLine
                output.write(line)
                line = triple(e1placeHolder,
                              schemaCoords.prefix + 'label', '\"' +
                              row['SGTI'] + '\"') + closeLine
                output.write(line)

            # doppio titolo - se usato cancellare E73 -E35

            '''if row['SGTI'] != 'None' and row['SGTI'] != ' ':

                line = triple(e73placeHolder, identifiedByCoords.prefix, "<http://www.museodipalazzopretorio.it/" + codice + "/sgti>") + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/sgti>",
                              hasTypePCoords.prefix,
                              "<http://www.museodipalazzopretorio.it/" + codice + "/ico>") + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/sgti>",
                              labelCoords.prefix,
                              '\"' + row['SGTI'] + '\"') + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/ico>",
                              labelCoords.prefix,
                              '\"' + 'Identificazione Iconografica' + '\"') + closeLine
                output.write(line)


            if row['SGTT'] != ' ':

                line = triple(e73placeHolder,
                              identifiedByCoords.prefix,
                              "<http://www.museodipalazzopretorio.it/" + codice + "/sgtt>") + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/sgtt>",
                              hasTypePCoords.prefix,
                              "<http://www.museodipalazzopretorio.it/" + codice + "/titolo>") + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/sgtt>",
                              labelCoords.prefix,
                              '\"' + row['SGTT'] + '\"') + closeLine
                output.write(line)
                line = triple("<http://www.museodipalazzopretorio.it/" + codice + "/titolo>",
                              labelCoords.prefix,
                              '\"' + 'Titolo' + '\"') + closeLine
                output.write(line)'''

            # Fine doppio titolo

            # Attention: these triples are identified only for C100005 Museo di Palazzo Pretorio

            if row['ESC'] == 'C100005':
                line = triple(datplaceHolder,
                              cidocCoords.prefix + 'P52_has_current_owner',
                              e74placeHolder) + closeLine
                output.write(line)
                line = triple(e74placeHolder,
                              nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E74_Group') + closeLine
                output.write(line)
                line = triple(e74placeHolder,
                              schemaCoords.prefix + 'label',
                              '\"Museo di Palazzo Pretorio\"') + closeLine
                output.write(line)

                line = triple(e74placeHolder,
                              cidocCoords.prefix + 'P1_is_identified_by',
                              e42CplaceHolder) + closeLine
                output.write(line)
                line = triple(e42CplaceHolder,
                              nsCoords.prefix + 'type',
                              cidocCoords.prefix + 'E42_Identifier') + closeLine
                output.write(line)
                line = triple(e42CplaceHolder,
                              schemaCoords.prefix + 'label',
                              '\"' + row['ESC'] + '\"') + closeLine
                output.write(line)

                line = triple(e42CplaceHolder,
                              cidocCoords.prefix + 'P2_has_type',
                              aatCoords.prefix + '300404626') + closeLine
                output.write(line)
                line = triple(aatCoords.prefix + '300404626',
                              schemaCoords.prefix + 'label',
                              '\"identificatore numerico\"') + closeLine
                output.write(line)


            currentLocation = ''

            # E22 - P54 - E53
            if row['LDCN'] != '':
                if row['LDCS'] != '':
                    currentLocation = row['LDCS']
                else:
                    currentLocation = currentLocation
                if row['LDCM'] != '':
                    currentLocation = currentLocation + ', ' + row['LDCM']
                else:
                    currentLocation = currentLocation
                if row['LDCN'] != '':
                    currentLocation = currentLocation + ', ' + row['LDCN']
                else:
                    currentLocation = currentLocation

                currentLocation = currentLocation + ', ' + row['PVCC'] + ' (' + row['PVCP'] + ')'

                line = triple(datplaceHolder, cidocCoords.prefix + 'P54_has_current_permanent_location',
                              e53placeHolder) + closeLine
                output.write(line)
                line = triple(e53placeHolder, nsCoords.prefix + 'type', cidocCoords.prefix + 'E53_Place') + closeLine
                output.write(line)
                line = triple(e53placeHolder, schemaCoords.prefix + 'label',
                              '\"' + currentLocation + '\"') + closeLine
                output.write(line)
            # End AS

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