#!/bin/env python3
import datetime

from rdflib import ConjunctiveGraph, Graph
from rdflib.namespace import RDF, Namespace, DCTERMS

DCAT = Namespace('http://www.w3.org/ns/dcat#')
GDP = Namespace('http://gss-data.org.uk/def/gdp#')

ds = ConjunctiveGraph('Sleepycat')
ds.open('datasets.db')
print(f'Datasets store has {len(ds)} triples')
datasets = set(ds.subjects(RDF.type, DCAT.Dataset))
datasets.update(ds.subjects(RDF.type, GDP.NationalStatistics))
datasets.update(ds.subjects(RDF.type, GDP.OfficialStatistics))

print(f' and {len(datasets)} datasets.')

gss = Graph('SPARQLStore', identifier='http://gss-data.org.uk')
gss.open("http://gss-data.org.uk/sparql")

for dataset in gss.query("""
PREFIX dcat: <http://www.w3.org/ns/dcat#>
PREFIX dct: <http://purl.org/dc/terms/>

SELECT ?id ?url ?issued WHERE {
    ?id a dcat:Dataset ;
      dcat:landingPage ?url ;
      dct:issued ?issued .
}"""):
    latest_pub_date = ds.value(subject=dataset.url, predicate=DCTERMS.issued, any=False)
    gss_issued = dataset.issued.value
    if type(gss_issued) == datetime.datetime:
        gss_issued = gss_issued.date()
    if latest_pub_date is None:
        print(f"Dataset {dataset.url} not listed in gov.uk statistical datasets.")
    elif latest_pub_date.value.date() > gss_issued:
        print(f"Dataset {dataset.url} has more recent update.")
    else:
        print(f"Dataset {dataset.url} is up to date.")

gss.close()
ds.close()