#
# A basic "module" to translate a RDF resource to a Python dict
# Usage: 
#   import rdf2dict
#   rdf2dict.get_dict(node_uri, {document_uri})
#
# By Alexandre Passant <alex@passant.org>
# This code is public domain

import RDF

prefixes = {
    'rdf'   : 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',
    'rdfs'  : 'http://www.w3.org/2000/01/rdf-schema#',
    'owl'   : 'http://www.w3.org/2002/07/owl#',
    'foaf'  : 'http://xmlns.com/foaf/0.1/',
    'bio'   : 'http://purl.org/vocab/bio/0.1/',
    'rel'   : 'http://purl.org/vocab/relationship/',
    'doap'  : 'http://usefulinc.com/ns/doap#',
    'geo'   : 'http://www.w3.org/2003/01/geo/wgs84_pos#',
    'dc'    : 'http://purl.org/dc/elements/1.1/',
    'dcterms' : 'http://purl.org/dc/terms/',
    'skos'  : 'http://www.w3.org/2004/02/skos/core#',
    'sioc'  : 'http://rdfs.org/sioc/ns#',
    'sioct' : 'http://rdfs.org/sioc/types#',
    'rss'   : 'http://purl.org/rss/1.0/',
    'content' : 'http://purl.org/rss/1.0/modules/content/',
    'cc'    : 'http://web.resource.org/cc/',
    'admin' : 'http://webns.net/mvcb/'
}

def get_dict(node_uri, document_uri = ''):
    model = RDF.Model()
    parser = RDF.Parser(name="rdfxml")
    if not document_uri:
        document_uri = node_uri[0:node_uri.find('#')]
    parser.parse_into_model(model, document_uri)
    return _node_dict(RDF.Node(uri_string = node_uri), model)

def _node_dict(uri, model, visited = [], next = 1):
    ## Save URI for resources
    if uri.is_resource():
        dict = {'uri': uri.uri.__str__()}
    else:
        dict = {}
    ## Find related statements
    statements = model.find_statements(RDF.Statement(uri, None, None))
    if not statements.end():
        visited.append(uri)
        literals = {}
        for s in statements:
            p = s.predicate.uri.__str__() 
            ## Replace full URIs by prefix_thing
            for prefix, url in prefixes.items():
                if p[0:len(url)] == url:
                    p = "%s_%s" %(prefix, p[len(url):])
            ## Create array if not exists
            if not p in dict.keys():
                dict[p] = []
            ## Saving literals, strings, and strings by lang
            if s.object.is_literal():
                if not p in literals.keys():
                    literals[p] = {'literals':[], 'strings':[]}
                    dict[p].append(literals[p])
                literal = s.object.literal_value
                literals[p]['literals'].append(literal)
                literals[p]['strings'].append(literal['string'].strip())
                lang = literal['language']
                if lang:
                    if not lang in literals[p].keys():
                        literals[p]["strings_%s" %lang] = []
                    literals[p]["strings_%s" %lang].append(literal['string'].strip())
            ## Loop if not visited URI, else do only a last step (needed to get rdfs:label for a visited node)
            else:
                if not s.object in visited:
                    dict[p].append(_node_dict(s.object, model, visited, 1))
                else:
                    if next:
                        dict[p].append(_node_dict(s.object, model, visited, 0))
                    elif not s.object.is_blank():
                        dict[p] = [{'uri': s.object.uri.__str__()}]
    return dict