{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-23T18:50:19.036357Z",
     "start_time": "2019-09-23T18:50:19.031896Z"
    }
   },
   "source": [
    "# RDF Conversions\n",
    "\n",
    "This notebook demonstrates how to [convert](https://nexus-forge.readthedocs.io/en/latest/interaction.html#converting) data in any RDF format (n3, Turtle, ...) to Resources and vice-versa. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2019-09-23T18:50:20.068658Z",
     "start_time": "2019-09-23T18:50:19.054054Z"
    }
   },
   "outputs": [],
   "source": [
    "from kgforge.core import KnowledgeGraphForge"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A configuration file is needed in order to create a KnowledgeGraphForge session. A configuration can be generated using the notebook [00-Initialization.ipynb](00%20-%20Initialization.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "forge = KnowledgeGraphForge(\"../../configurations/forge.yml\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from kgforge.core import Resource"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Resource to rdflib.Graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "jane = Resource(type=\"Person\", name=\"Jane Doe\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<action> _register_one\n",
      "<succeeded> True\n"
     ]
    }
   ],
   "source": [
    "forge.register(jane)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "graph = forge.as_graph(jane)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(graph) == 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('http://schema.org/name'), rdflib.term.Literal('Jane Doe'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#type'), rdflib.term.URIRef('http://schema.org/Person'))\n"
     ]
    }
   ],
   "source": [
    "for triple in graph:\n",
    "    print(triple)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "graph_store_metadata = forge.as_graph(jane, store_metadata=True)\n",
    "len(graph_store_metadata)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('http://schema.org/name'), rdflib.term.Literal('Jane Doe'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/updatedAt'), rdflib.term.Literal('2022-04-12T22:47:28.984000+00:00', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#dateTime')))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/rev'), rdflib.term.Literal('1', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#integer')))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/createdAt'), rdflib.term.Literal('2022-04-12T22:47:28.984000+00:00', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#dateTime')))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/self'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/constrainedBy'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/schemas/unconstrained.json'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/schemaProject'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/projects/dke/kgforge'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/createdBy'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/realms/bbp/users/sy'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/deprecated'), rdflib.term.Literal('false', datatype=rdflib.term.URIRef('http://www.w3.org/2001/XMLSchema#boolean')))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/updatedBy'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/realms/bbp/users/sy'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/outgoing'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57/outgoing'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('http://www.w3.org/1999/02/22-rdf-syntax-ns#type'), rdflib.term.URIRef('http://schema.org/Person'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/project'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/projects/dke/kgforge'))\n",
      "(rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57'), rdflib.term.URIRef('https://bluebrain.github.io/nexus/vocabulary/incoming'), rdflib.term.URIRef('https://bbp.epfl.ch/nexus/v1/resources/dke/kgforge/_/22ec9959-ab76-419a-9864-131b0dc12d57/incoming'))\n"
     ]
    }
   ],
   "source": [
    "for triple in graph_store_metadata:\n",
    "    print(triple)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## rdflib.Graph to Resource"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "#data in n3 format. Other RDF formats can be used\n",
    "data = '''\n",
    "@prefix schema: <http://schema.org/> .\n",
    "[] a schema:Person;\n",
    "     schema:name \"Jane Doe\" ;\n",
    "     schema:knows <http://example.org/JohnDoe>;\n",
    "     schema:affiliation <https://www.grid.ac/institutes/grid.5333.6>.\n",
    "\n",
    "<http://example.org/JohnDoe> a schema:Person;\n",
    "     schema:name \"John Doe\" .\n",
    "     \n",
    "<https://www.grid.ac/institutes/grid.5333.6> a schema:Organization;\n",
    "     schema:name \"EPFL\" .\n",
    "'''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Graph identifier=Nabbd6999596d410b9d9760651f8b8bcd (<class 'rdflib.graph.Graph'>)>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import rdflib\n",
    "graph = rdflib.Graph()\n",
    "graph.parse(data=data, format=\"n3\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(graph)==8"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "resources = forge.from_graph(graph)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(resources) == 3"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Apply a JSON-LD Frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "frame = {\n",
    "        \"@type\": [ 'http://schema.org/Person'],\n",
    "        \"@embed\": True\n",
    "}  \n",
    "resources = forge.from_graph(data=graph, frame= frame)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(resources) == 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[{'@context': {'brick': 'https://brickschema.org/schema/Brick#',\n",
       "   'csvw': 'http://www.w3.org/ns/csvw#',\n",
       "   'dc': 'http://purl.org/dc/elements/1.1/',\n",
       "   'dcam': 'http://purl.org/dc/dcam/',\n",
       "   'dcat': 'http://www.w3.org/ns/dcat#',\n",
       "   'dcmitype': 'http://purl.org/dc/dcmitype/',\n",
       "   'dcterms': 'http://purl.org/dc/terms/',\n",
       "   'doap': 'http://usefulinc.com/ns/doap#',\n",
       "   'foaf': 'http://xmlns.com/foaf/0.1/',\n",
       "   'odrl': 'http://www.w3.org/ns/odrl/2/',\n",
       "   'org': 'http://www.w3.org/ns/org#',\n",
       "   'owl': 'http://www.w3.org/2002/07/owl#',\n",
       "   'prof': 'http://www.w3.org/ns/dx/prof/',\n",
       "   'prov': 'http://www.w3.org/ns/prov#',\n",
       "   'qb': 'http://purl.org/linked-data/cube#',\n",
       "   'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',\n",
       "   'rdfs': 'http://www.w3.org/2000/01/rdf-schema#',\n",
       "   'schema': 'https://schema.org/',\n",
       "   'schema1': 'http://schema.org/',\n",
       "   'sh': 'http://www.w3.org/ns/shacl#',\n",
       "   'skos': 'http://www.w3.org/2004/02/skos/core#',\n",
       "   'sosa': 'http://www.w3.org/ns/sosa/',\n",
       "   'ssn': 'http://www.w3.org/ns/ssn/',\n",
       "   'time': 'http://www.w3.org/2006/time#',\n",
       "   'vann': 'http://purl.org/vocab/vann/',\n",
       "   'void': 'http://rdfs.org/ns/void#',\n",
       "   'xsd': 'http://www.w3.org/2001/XMLSchema#'},\n",
       "  '@type': 'schema1:Person',\n",
       "  'schema1:affiliation': {'@id': 'https://www.grid.ac/institutes/grid.5333.6',\n",
       "   '@type': 'schema1:Organization',\n",
       "   'schema1:name': 'EPFL'},\n",
       "  'schema1:knows': {'@id': 'http://example.org/JohnDoe',\n",
       "   '@type': 'schema1:Person',\n",
       "   'schema1:name': 'John Doe'},\n",
       "  'schema1:name': 'Jane Doe'},\n",
       " {'@context': {'brick': 'https://brickschema.org/schema/Brick#',\n",
       "   'csvw': 'http://www.w3.org/ns/csvw#',\n",
       "   'dc': 'http://purl.org/dc/elements/1.1/',\n",
       "   'dcam': 'http://purl.org/dc/dcam/',\n",
       "   'dcat': 'http://www.w3.org/ns/dcat#',\n",
       "   'dcmitype': 'http://purl.org/dc/dcmitype/',\n",
       "   'dcterms': 'http://purl.org/dc/terms/',\n",
       "   'doap': 'http://usefulinc.com/ns/doap#',\n",
       "   'foaf': 'http://xmlns.com/foaf/0.1/',\n",
       "   'odrl': 'http://www.w3.org/ns/odrl/2/',\n",
       "   'org': 'http://www.w3.org/ns/org#',\n",
       "   'owl': 'http://www.w3.org/2002/07/owl#',\n",
       "   'prof': 'http://www.w3.org/ns/dx/prof/',\n",
       "   'prov': 'http://www.w3.org/ns/prov#',\n",
       "   'qb': 'http://purl.org/linked-data/cube#',\n",
       "   'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',\n",
       "   'rdfs': 'http://www.w3.org/2000/01/rdf-schema#',\n",
       "   'schema': 'https://schema.org/',\n",
       "   'schema1': 'http://schema.org/',\n",
       "   'sh': 'http://www.w3.org/ns/shacl#',\n",
       "   'skos': 'http://www.w3.org/2004/02/skos/core#',\n",
       "   'sosa': 'http://www.w3.org/ns/sosa/',\n",
       "   'ssn': 'http://www.w3.org/ns/ssn/',\n",
       "   'time': 'http://www.w3.org/2006/time#',\n",
       "   'vann': 'http://purl.org/vocab/vann/',\n",
       "   'void': 'http://rdfs.org/ns/void#',\n",
       "   'xsd': 'http://www.w3.org/2001/XMLSchema#'},\n",
       "  '@id': 'http://example.org/JohnDoe',\n",
       "  '@type': 'schema1:Person',\n",
       "  'schema1:name': 'John Doe'}]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "forge.as_jsonld(resources)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Select a Type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "_type= [\"http://schema.org/Organization\"]\n",
    "resources = forge.from_graph(data=graph, type= _type,use_model_context=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "isinstance(resources, Resource)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(resources)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "isinstance(resources, Resource)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "epfl_json = {'id': 'https://www.grid.ac/institutes/grid.5333.6',\n",
    "  'type': 'Organization',\n",
    "  'name': 'EPFL'}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "epfl_json == forge.as_json(resources)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Use the forge Model context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "_type= [\"http://schema.org/Organization\"]\n",
    "resources = forge.from_graph(data=graph, type= _type, use_model_context=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "isinstance(resources, Resource)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.7 (nexusforgelatest)",
   "language": "python",
   "name": "nexusforgelatest"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}