sshoc-skosmapping/sshoc_lessico_panlatino.ipynb

Mapping Pan-Latin Textile Fibres Vocabulary from spreadsheet to SKOS resources¶

This Notebook implements a simple parser used to transform the Pan-Latin Textile Fibres Vocabulary, developed within the Realiter network, and published as spreadsheets, into SKOS resources. The parser reads the spreadsheets and transforms the content in SKOS data following a set of mapping rules, the result is stored in two Turtle files.

In [ ]:

import pandas as pd
import rdflib
import itertools
import yaml
import datetime

The file config.yaml contains the external information used in the parsing, including the position of the spreadsheets. Set the correct values before running the Notebook.

In [ ]:

try:
    with open("config-lessico.yaml", 'r') as stream:
        try:
           conf=yaml.safe_load(stream)
        except yaml.YAMLError as exc:
            print(exc)
except FileNotFoundError:
    print('Warning config.yaml file not present! Please store it in the same directory as the notebook')
#print (conf)

The following cells defines the Namespaces used in the parsing

In [ ]:

from rdflib.namespace import DC, DCAT, DCTERMS, OWL, \
                            RDF, RDFS, SKOS,  \
                           XMLNS, XSD, XMLNS
from rdflib import Namespace
from rdflib import URIRef, BNode, Literal

pltextile = Namespace(conf['Namespaces']['TEXTILETERM'])
dc11=Namespace("http://purl.org/dc/elements/1.1/");
dct = Namespace("http://purl.org/dc/terms/")
iso369=Namespace("http://id.loc.gov/vocabulary/iso639-3");

Download Lessico spreadsheet and show it to check if the operation has been executed correctly

In [ ]:

url=conf['Source']['LESSICOSOURCE']
df_data=pd.read_csv(url)

In [ ]:

df_data.head()

In [ ]:

df_data.rename(columns = {'es [ARG]': 'es-arg', 'es [MEX]': 'es-mex', 'fr [CA]': 'fr-ca'}, inplace = True)
#df_data.head()

In [ ]:

df_data.iloc[0].it.split('\n')[0].split(' ')[0]

Create a graph for the SKOS data and bind the namespaces to it

In [ ]:

c1rdf = rdflib.Graph()
c1rdf.bind("pltextile", pltextile)
c1rdf.bind("dc11", dc11)
c1rdf.bind("dct", dct)
c1rdf.bind("iso369-3", iso369)
c1rdf.bind("skos", SKOS)
c1rdf.bind("dc", DC)
c1rdf.bind("rdf", RDF)
c1rdf.bind("rdfs", RDFS)
c1rdf.bind("owl", OWL)
c1rdf.bind("xsd", XSD)

Insert in the graph the SKOS.ConceptScheme

In [ ]:

now = datetime.datetime.today()
today_date=now.date()
title=Literal(conf['Texts']['LESSICOTITLE'], lang=conf['Texts']['LANG'])
description=Literal(conf['Texts']['LESSICODESCRIPTION'], lang=conf['Texts']['LANG'])
description_it=Literal(conf['Texts']['LESSICODESCRIPTION_IT'], lang='it')
identifier=Literal(conf['Texts']['LESSICOID'], lang=conf['Texts']['LANG'])
#identifier=URIRef(conf['Texts']['VOCABULARYID'])
createddate= Literal(conf['Texts']['LESSICOCREATEDATE'],datatype=XSD.date)
moddate= Literal(today_date,datatype=XSD.date)
version= Literal(conf['Texts']['LESSICOVERSION'],datatype=XSD.string)

c1rdf.add((pltextile[''], RDF.type, SKOS.ConceptScheme))
c1rdf.add((pltextile[''], DC.title, title))
c1rdf.add((pltextile[''], DC.identifier, identifier))
c1rdf.add((pltextile[''], DC.description, description))
c1rdf.add((pltextile[''], DC.description, description_it))
c1rdf.add((pltextile[''], dct.created, createddate))
c1rdf.add((pltextile[''], dct.modified, moddate))
c1rdf.add((pltextile[''], OWL.versionInfo, version))
c1rdf.add((pltextile[''], dct.language, iso369.eng))
c1rdf.add((pltextile[''], dct.language, iso369.es))
c1rdf.add((pltextile[''], dct.language, iso369.fra))
c1rdf.add((pltextile[''], dct.language, iso369.gl))
c1rdf.add((pltextile[''], dct.language, iso369.ita))
c1rdf.add((pltextile[''], dct.language, iso369.ro))
c1rdf.add((pltextile[''], dct.language, iso369.pt))
c1rdf.add((pltextile[''], dct.language, iso369.ca))

In [ ]:

#c1rdf.serialize(destination='data/skostest.rdf', format="n3");#format="pretty-xml")
#comrdf.serialize(destination='data/parsed_rdf/prima_cantica_forme_com.rdf', format="n3");
df_data.fillna('', inplace=True)
df_data.head()

The following cell implements the mapping rules for creating SKOS resources.

In [ ]:

#df_data.iloc[0].it.split('\n')[0].split(' ')[0]
for index, row in df_data.iterrows():
    
    strlabel=row.it.split('\n')[0].split(' (')[0].strip()
    label=strlabel.replace(" ", "_")
    #label=URIRef(row.it.split('\n')[0].split(' (')[0].strip())
    c1rdf.add((pltextile[''], SKOS.hasTopConcept, pltextile[label]))    
    frlabel=Literal(row["fr"].split('\n')[0].strip(), lang='fr')
    fraltlabels=row["fr"].split('\n')[1:]
    itlabel=Literal(row['it'].split('\n')[0].strip(), lang='it')
    italtlabels=row["it"].split('\n')[1:]                    
    calabel=Literal(row['ca'].split('\n')[0].strip(), lang='ca')
    caaltlabels=row["ca"].split('\n')[1:]
    eslabel=Literal(row['es'].split('\n')[0].strip(), lang='es')
    esaltlabels=row["es"].split('\n')[1:]
    gllabel=Literal(row['gl'].split('\n')[0].strip(), lang='gl')
    glaltlabels=row["gl"].split('\n')[1:]
    ptlabel=Literal(row['pt'].split('\n')[0].strip(), lang='pt')
    ptaltlabels=row["pt"].split('\n')[1:]
    rolabel=Literal(row['ro'].split('\n')[0].strip(), lang='ro')
    roaltlabels=row["ro"].split('\n')[1:]
    enlabel=Literal(row['en'].split('\n')[0].strip(), lang='en')
    enaltlabels=row["en"].split('\n')[1:]
    
    esarglabel=Literal(row['es-arg'].split('\n')[0].strip(), lang='es-ar')
    esargaltlabels=row["es-arg"].split('\n')[1:]
    #es-arg-mex
#     esargmexarglabel=Literal(row['es-arg-mex'].split('\n')[0].strip(), lang='es-ar')
#     esargmexmexlabel=Literal(row['es-arg-mex'].split('\n')[0].strip(), lang='es-mx')
#     esargmexaltlabels=row["es-arg-mex"].split('\n')[1:]
    
    esmexlabel=Literal(row['es-mex'].split('\n')[0].strip(), lang='es-mx')
    esmexaltlabels=row["es-mex"].split('\n')[1:]
    frcalabel=Literal(row['fr-ca'].split('\n')[0].strip(), lang='fr-ca')
    frcaaltlabels=row["fr-ca"].split('\n')[1:]
    
    #definition
    itdef=Literal(row["DEF"].strip(), lang='it')
    
 
    
    c1rdf.add((pltextile[label], RDF.type, SKOS.Concept))
    c1rdf.add((pltextile[label], SKOS.inScheme, pltextile['']))
    c1rdf.add((pltextile[label], SKOS.topConceptOf, pltextile['']))
    
    for alab in esargaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-ar')))
    
#     for alab in esargmexaltlabels:
#         c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-ar')))
#         c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-mx')))
     
    for alab in esmexaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-mx')))
        
    for alab in frcaaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='fr-ca')))
    
    for alab in esaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es')))
    
    for alab in glaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='gl')))
    
    for alab in ptaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='pt')))
    
    for alab in roaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ro')))
    
    for alab in enaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='en')))
        
    for alab in caaltlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ca')))
        
    for alab in fraltlabels:
        #print ("tt "+alab)
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='fr')))
    for alab in italtlabels:
        c1rdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='it')))
   
    
    if(frlabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, frlabel))
    if(itlabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, itlabel))
    if(gllabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, gllabel))
    
    if(ptlabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, ptlabel))
    if(rolabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, rolabel))
    if(enlabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, enlabel))
        
    if(calabel):    
        c1rdf.add((pltextile[label], SKOS.prefLabel, calabel))
    if(eslabel):  
        c1rdf.add((pltextile[label], SKOS.prefLabel, eslabel))
    if(esarglabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, esarglabel))
    
#     if(esargmexarglabel):
#         c1rdf.add((pltextile[label], SKOS.prefLabel, esargmexarglabel))
#         c1rdf.add((pltextile[label], SKOS.prefLabel, esargmexmexlabel))
        
    if(esmexlabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, esmexlabel))
    if(frcalabel):
        c1rdf.add((pltextile[label], SKOS.prefLabel, frcalabel))
    
    if (itdef):
        c1rdf.add((pltextile[label], SKOS.definition, itdef))

print(len(c1rdf))

In [ ]:

# for s, p, o in c1rdf.triples((None,  None, None)):
#    print("{}  {}".format(s, o.n3))

Create a Turtle file in the /data directory with the SKOS resources for Data Stewardship terminology

In [ ]:

c1rdf.serialize(destination='data/lexpanlatskos_11.ttl', format="n3");#format="pretty-xml")
c1rdf.serialize(destination='data/lexpanlatskos_11.rdf', format="pretty-xml");#format="pretty-xml")

Lessico panlatino delle Maniche¶

In [ ]:

urlma=conf['Source']['LESSICOMANICHESOURCE']
df_data_maniche=pd.read_csv(urlma)
df_data_maniche.rename(columns = {'es [ARG]': 'es-arg', 'es [MEX]': 'es-mex', 'pt [BR]': 'pt-br'}, inplace = True)
df_data_maniche.fillna('', inplace=True)
#df_data_maniche.info()

In [ ]:

cl_manicherdf = rdflib.Graph()
cl_manicherdf.bind("pltextile", pltextile)
cl_manicherdf.bind("dc11", dc11)
cl_manicherdf.bind("dct", dct)
cl_manicherdf.bind("iso369-3", iso369)
cl_manicherdf.bind("skos", SKOS)
cl_manicherdf.bind("dc", DC)
cl_manicherdf.bind("rdf", RDF)
cl_manicherdf.bind("rdfs", RDFS)
cl_manicherdf.bind("owl", OWL)
cl_manicherdf.bind("xsd", XSD)
now = datetime.datetime.today()
today_date=now.date()
title=Literal(conf['Texts']['LESSICOMANICHETITLE'], lang=conf['Texts']['LANG'])
description=Literal(conf['Texts']['LESSICOMANICHEDESCRIPTION'], lang=conf['Texts']['LANG'])
description_it=Literal(conf['Texts']['LESSICOMANICHEDESCRIPTION_IT'], lang='it')
identifier=Literal(conf['Texts']['LESSICOMANICHEID'], lang=conf['Texts']['LANG'])
#identifier=URIRef(conf['Texts']['VOCABULARYID'])
createddate= Literal(conf['Texts']['LESSICOCREATEDATE'],datatype=XSD.date)
moddate= Literal(today_date,datatype=XSD.date)
version= Literal(conf['Texts']['LESSICOVERSION'],datatype=XSD.string)

cl_manicherdf.add((pltextile[''], RDF.type, SKOS.ConceptScheme))
cl_manicherdf.add((pltextile[''], DC.title, title))
cl_manicherdf.add((pltextile[''], DC.identifier, identifier))
cl_manicherdf.add((pltextile[''], DC.description, description))
cl_manicherdf.add((pltextile[''], DC.description, description_it))
cl_manicherdf.add((pltextile[''], dct.created, createddate))
cl_manicherdf.add((pltextile[''], dct.modified, moddate))
cl_manicherdf.add((pltextile[''], OWL.versionInfo, version))
cl_manicherdf.add((pltextile[''], dct.language, iso369.eng))
cl_manicherdf.add((pltextile[''], dct.language, iso369.es))
cl_manicherdf.add((pltextile[''], dct.language, iso369.fra))
cl_manicherdf.add((pltextile[''], dct.language, iso369.ca))
cl_manicherdf.add((pltextile[''], dct.language, iso369.ita))
cl_manicherdf.add((pltextile[''], dct.language, iso369.pt))

In [ ]:

# Mapping
for index, row in df_data_maniche.iterrows():
    
    strlabel=row.it.split('\n')[0].split('(')[0].strip()
    label=strlabel.replace(" ", "_").replace("’","").replace("'","").strip()
    #label=URIRef(row.it.split('\n')[0].split(' (')[0].strip())
    cl_manicherdf.add((pltextile[''], SKOS.hasTopConcept, pltextile[label]))    
    frlabel=Literal(row["fr"].split('\n')[0].strip(), lang='fr')
    fraltlabels=row["fr"].split('\n')[1:]
    itlabel=Literal(row['it'].split('\n')[0].strip(), lang='it')
    italtlabels=row["it"].split('\n')[1:]                    
    calabel=Literal(row['ca'].split('\n')[0].strip(), lang='ca')
    caaltlabels=row["ca"].split('\n')[1:]
    eslabel=Literal(row['es'].split('\n')[0].strip(), lang='es')
    esaltlabels=row["es"].split('\n')[1:]
    #gllabel=Literal(row['gl'].split('\n')[0].strip(), lang='gl')
    #glaltlabels=row["gl"].split('\n')[1:]
    ptlabel=Literal(row['pt'].split('\n')[0].strip(), lang='pt')
    ptaltlabels=row["pt"].split('\n')[1:]
#     rolabel=Literal(row['ro'].split('\n')[0].strip(), lang='ro')
#     roaltlabels=row["ro"].split('\n')[1:]
    enlabel=Literal(row['en'].split('\n')[0].strip(), lang='en')
    enaltlabels=row["en"].split('\n')[1:]
    
    esarglabel=Literal(row['es-arg'].split('\n')[0].strip(), lang='es-ar')
    esargaltlabels=row["es-arg"].split('\n')[1:]
  

    esmexlabel=Literal(row['es-mex'].split('\n')[0].strip(), lang='es-mx')
    esmexaltlabels=row["es-mex"].split('\n')[1:]
    ptbrlabel=Literal(row['pt-br'].split('\n')[0].strip(), lang='pt-br')
    ptbraltlabels=row["pt-br"].split('\n')[1:]
    
    #definition
    itdef=Literal(row["DEF"].strip(), lang='it')
    #DOI as rdfs:seeAlso
    imagedoi=row["DOI"].strip()
    
    cl_manicherdf.add((pltextile[label], RDF.type, SKOS.Concept))
    cl_manicherdf.add((pltextile[label], SKOS.inScheme, pltextile['']))
    cl_manicherdf.add((pltextile[label], SKOS.topConceptOf, pltextile['']))
    if (imagedoi!=''):
        cl_manicherdf.add((pltextile[label], RDFS.seeAlso, URIRef('https://doi.org/'+imagedoi)))
    
    for alab in esargaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-ar')))
    
     
    for alab in esmexaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-mx')))
        
    for alab in ptbraltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='pt-br')))
    
    for alab in esaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es')))
    
#     for alab in glaltlabels:
#         cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='gl')))
    
    for alab in ptaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='pt')))
    
#     for alab in roaltlabels:
#         cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ro')))
    
    for alab in enaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='en')))
        
    for alab in caaltlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ca')))
        
    for alab in fraltlabels:
        #print ("tt "+alab)
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='fr')))
    for alab in italtlabels:
        cl_manicherdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='it')))
   
    
    if(frlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, frlabel))
    if(itlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, itlabel))
#     if(gllabel):
#         cl_manicherdf.add((pltextile[label], SKOS.prefLabel, gllabel))
    
    if(ptlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, ptlabel))
#     if(rolabel):
#         cl_manicherdf.add((pltextile[label], SKOS.prefLabel, rolabel))
    if(enlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, enlabel))
        
    if(calabel):    
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, calabel))
    if(eslabel):  
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, eslabel))
    if(esarglabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, esarglabel))
    

    if(esmexlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, esmexlabel))
    if(ptbrlabel):
        cl_manicherdf.add((pltextile[label], SKOS.prefLabel, ptbrlabel))
    
    if (itdef):
        cl_manicherdf.add((pltextile[label], SKOS.definition, itdef))

print(len(cl_manicherdf))

In [ ]:

cl_manicherdf.serialize(destination='data/lexpanlatmanicheskos_11.ttl', format="n3");#format="pretty-xml")
cl_manicherdf.serialize(destination='data/lexpanlatmanicheskos_11.rdf', format="pretty-xml");#format="pretty-xml")

Lessico panlatino dei Colli¶

In [ ]:

urlco=conf['Source']['LESSICOCOLLISOURCE']
df_data_colli=pd.read_csv(urlco)

In [ ]:

df_data_colli.rename(columns = {'es [ARG]': 'es-arg', 'es [MEX]': 'es-mex', 'pt [BR]': 'pt-br'}, inplace = True)
df_data_colli.fillna('', inplace=True)
#df_data_colli.head()

In [ ]:

cl_collirdf = rdflib.Graph()
cl_collirdf.bind("pltextile", pltextile)
cl_collirdf.bind("dc11", dc11)
cl_collirdf.bind("dct", dct)
cl_collirdf.bind("iso369-3", iso369)
cl_collirdf.bind("skos", SKOS)
cl_collirdf.bind("dc", DC)
cl_collirdf.bind("rdf", RDF)
cl_collirdf.bind("owl", OWL)
cl_collirdf.bind("xsd", XSD)

SKOS concept scheme

In [ ]:

now = datetime.datetime.today()
today_date=now.date()
title=Literal(conf['Texts']['LESSICOCOLLITITLE'], lang=conf['Texts']['LANG'])
description=Literal(conf['Texts']['LESSICOCOLLIDESCRIPTION'], lang=conf['Texts']['LANG'])
description_it=Literal(conf['Texts']['LESSICOCOLLIDESCRIPTION_IT'], lang='it')
identifier=Literal(conf['Texts']['LESSICOCOLLIID'], lang=conf['Texts']['LANG'])
#identifier=URIRef(conf['Texts']['VOCABULARYID'])
createddate= Literal(conf['Texts']['LESSICOCREATEDATE'],datatype=XSD.date)
moddate= Literal(today_date,datatype=XSD.date)
version= Literal(conf['Texts']['LESSICOVERSION'],datatype=XSD.string)

cl_collirdf.add((pltextile[''], RDF.type, SKOS.ConceptScheme))
cl_collirdf.add((pltextile[''], DC.title, title))
cl_collirdf.add((pltextile[''], DC.identifier, identifier))
cl_collirdf.add((pltextile[''], DC.description, description))
cl_collirdf.add((pltextile[''], DC.description, description_it))
cl_collirdf.add((pltextile[''], dct.created, createddate))
cl_collirdf.add((pltextile[''], dct.modified, moddate))
cl_collirdf.add((pltextile[''], OWL.versionInfo, version))
cl_collirdf.add((pltextile[''], dct.language, iso369.eng))
cl_collirdf.add((pltextile[''], dct.language, iso369.es))
cl_collirdf.add((pltextile[''], dct.language, iso369.fra))
cl_collirdf.add((pltextile[''], dct.language, iso369.ita))
cl_collirdf.add((pltextile[''], dct.language, iso369.pt))
cl_collirdf.add((pltextile[''], dct.language, iso369.ca))

In [ ]:

# Mapping
for index, row in df_data_colli.iterrows():
    
    strlabel=row.it.split('\n')[0].split(' (')[0].strip()
    label=strlabel.replace(" ", "_").replace("’","")
    #label=URIRef(row.it.split('\n')[0].split(' (')[0].strip())
    cl_collirdf.add((pltextile[''], SKOS.hasTopConcept, pltextile[label]))    
    frlabel=Literal(row["fr"].split('\n')[0].strip(), lang='fr')
    fraltlabels=row["fr"].split('\n')[1:]
    itlabel=Literal(row['it'].split('\n')[0].strip(), lang='it')
    italtlabels=row["it"].split('\n')[1:]                    
    calabel=Literal(row['ca'].split('\n')[0].strip(), lang='ca')
    caaltlabels=row["ca"].split('\n')[1:]
    eslabel=Literal(row['es'].split('\n')[0].strip(), lang='es')
    esaltlabels=row["es"].split('\n')[1:]
    #gllabel=Literal(row['gl'].split('\n')[0].strip(), lang='gl')
    #glaltlabels=row["gl"].split('\n')[1:]
    ptlabel=Literal(row['pt'].split('\n')[0].strip(), lang='pt')
    ptaltlabels=row["pt"].split('\n')[1:]
#     rolabel=Literal(row['ro'].split('\n')[0].strip(), lang='ro')
#     roaltlabels=row["ro"].split('\n')[1:]
    enlabel=Literal(row['en'].split('\n')[0].strip(), lang='en')
    enaltlabels=row["en"].split('\n')[1:]
    
    esarglabel=Literal(row['es-arg'].split('\n')[0].strip(), lang='es-ar')
    esargaltlabels=row["es-arg"].split('\n')[1:]
  

    esmexlabel=Literal(row['es-mex'].split('\n')[0].strip(), lang='es-mx')
    esmexaltlabels=row["es-mex"].split('\n')[1:]
    ptbrlabel=Literal(row['pt-br'].split('\n')[0].strip(), lang='pt-br')
    ptbraltlabels=row["pt-br"].split('\n')[1:]
    
    #definition
    itdef=Literal(row["DEF"].strip(), lang='it')
    #DOI as rdfs:seeAlso
    imagedoi=row["DOI"].strip()
    
    cl_collirdf.add((pltextile[label], RDF.type, SKOS.Concept))
    cl_collirdf.add((pltextile[label], SKOS.inScheme, pltextile['']))
    cl_collirdf.add((pltextile[label], SKOS.topConceptOf, pltextile['']))
    if (imagedoi!=''):
        cl_collirdf.add((pltextile[label], RDFS.seeAlso, URIRef('https://doi.org/'+imagedoi)))
    
    for alab in esargaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-ar')))
    
     
    for alab in esmexaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es-mx')))
        
    for alab in ptbraltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='pt-br')))
    
    for alab in esaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='es')))
    
#     for alab in glaltlabels:
#         cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='gl')))
    
    for alab in ptaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='pt')))
    
#     for alab in roaltlabels:
#         cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ro')))
    
    for alab in enaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='en')))
        
    for alab in caaltlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='ca')))
        
    for alab in fraltlabels:
        #print ("tt "+alab)
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='fr')))
    for alab in italtlabels:
        cl_collirdf.add((pltextile[label], SKOS.altLabel, Literal(alab, lang='it')))
   
    
    if(frlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, frlabel))
    if(itlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, itlabel))
#     if(gllabel):
#         cl_collirdf.add((pltextile[label], SKOS.prefLabel, gllabel))
    
    if(ptlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, ptlabel))
#     if(rolabel):
#         cl_collirdf.add((pltextile[label], SKOS.prefLabel, rolabel))
    if(enlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, enlabel))
        
    if(calabel):    
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, calabel))
    if(eslabel):  
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, eslabel))
    if(esarglabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, esarglabel))
    

    if(esmexlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, esmexlabel))
    if(ptbrlabel):
        cl_collirdf.add((pltextile[label], SKOS.prefLabel, ptbrlabel))
    
    if (itdef):
        cl_collirdf.add((pltextile[label], SKOS.definition, itdef))

print(len(cl_collirdf))

In [ ]:

cl_collirdf.serialize(destination='data/lexpanlatcolliskos_11.ttl', format="n3");#format="pretty-xml")
cl_collirdf.serialize(destination='data/lexpanlatcolliskos_11.rdf', format="pretty-xml");#format="pretty-xml")