{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# How to Check Referential Integrity\n",
    "\n",
    "This example uses MongoDB"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "fc4794dd116ed21"
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "outputs": [],
   "source": [
    "from linkml_store import Client\n",
    "\n",
    "client = Client()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.760981Z",
     "start_time": "2024-05-04T19:51:08.378243Z"
    }
   },
   "id": "initial_id"
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [],
   "source": [
    "db = client.attach_database(\"mongodb://localhost:27017\", \"test\")\n",
    "db.metadata.ensure_referential_integrity = True\n",
    "db.set_schema_view(\"../../tests/input/countries/countries.linkml.yaml\")\n",
    "countries_coll = db.create_collection(\"Country\", alias=\"countries\", recreate_if_exists=True)\n",
    "routes_coll = db.create_collection(\"Route\", alias=\"routes\", recreate_if_exists=True)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.788932Z",
     "start_time": "2024-05-04T19:51:09.771112Z"
    }
   },
   "id": "cc164c0acbe4c39d"
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [],
   "source": [
    "COUNTRIES = \"../../tests/input/countries/countries.jsonl\"\n",
    "ROUTES = \"../../tests/input/countries/routes.csv\""
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.789681Z",
     "start_time": "2024-05-04T19:51:09.786454Z"
    }
   },
   "id": "5286ef4e9dd0f316"
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "outputs": [
    {
     "data": {
      "text/plain": "[{'origin': 'DE', 'destination': 'FR', 'method': 'rail'}]"
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from linkml_store.utils.format_utils import load_objects\n",
    "\n",
    "countries = load_objects(COUNTRIES)\n",
    "routes = load_objects(ROUTES)\n",
    "routes"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.795894Z",
     "start_time": "2024-05-04T19:51:09.790413Z"
    }
   },
   "id": "2e21988e4fc13f58"
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "outputs": [],
   "source": [
    "countries_coll.insert(countries)\n",
    "routes_coll.insert(routes)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.803272Z",
     "start_time": "2024-05-04T19:51:09.798758Z"
    }
   },
   "id": "668e59a8f28e7bfe"
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "outputs": [
    {
     "data": {
      "text/plain": "[{'origin': 'DE', 'destination': 'FR', 'method': 'rail'}]"
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "routes_coll.find().rows"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.810617Z",
     "start_time": "2024-05-04T19:51:09.804004Z"
    }
   },
   "id": "995e63f873ea9353"
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "outputs": [],
   "source": [
    "for result in db.iter_validate_database():\n",
    "    print(result)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.956191Z",
     "start_time": "2024-05-04T19:51:09.809082Z"
    }
   },
   "id": "a8ef16a3fbc6bfe6"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Inserting invalid data\n",
    "\n",
    "We will intentionally insert an invalid row"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "24fb15bce092c2d1"
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "outputs": [],
   "source": [
    "routes_coll.insert({\"origin\": \"ZZZ\", \"destination\": \"YYY\", \"method\": \"rail\"})"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.961815Z",
     "start_time": "2024-05-04T19:51:09.956721Z"
    }
   },
   "id": "f712a82be775f413"
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "outputs": [
    {
     "data": {
      "text/plain": "  origin destination method\n0     DE          FR   rail\n1    ZZZ         YYY   rail",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>origin</th>\n      <th>destination</th>\n      <th>method</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>DE</td>\n      <td>FR</td>\n      <td>rail</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>ZZZ</td>\n      <td>YYY</td>\n      <td>rail</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "routes_coll.find().rows_dataframe"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:51:09.974226Z",
     "start_time": "2024-05-04T19:51:09.961675Z"
    }
   },
   "id": "18ffa996e3893b96"
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "type='ReferentialIntegrity' severity=<Severity.ERROR: 'ERROR'> message='Referential integrity error: Country not found' instance='ZZZ' instance_index=None instantiates='Country'\n",
      "type='ReferentialIntegrity' severity=<Severity.ERROR: 'ERROR'> message='Referential integrity error: Country not found' instance='YYY' instance_index=None instantiates='Country'\n"
     ]
    }
   ],
   "source": [
    "results = list(db.iter_validate_database())\n",
    "for result in results:\n",
    "    print(result)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-04T19:52:20.044928Z",
     "start_time": "2024-05-04T19:52:19.996008Z"
    }
   },
   "id": "c67517aece5d47c5"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}