Nusnus/ElasticsearchPythonExample

Elasticsearch Python Example

This is a simple python script for running simple queries against the example Elasticsearch Shakespeare DB.

Make sure you set up Elasticsearch and load up the example dataset before running this.

How to run

Just run run_locally.sh and you should see something like that:

How many lines are in Shakespeare's plays: 111396
Again, as SQL: 111396
Again, using scan(): 111396
Which are Shakespeare's longest plays (top 3): ['Hamlet', 'Coriolanus', 'Cymbeline']
Again, as SQL: [['Hamlet'], ['Coriolanus'], ['Cymbeline']]
The line number where Hamlet says "To be, or not to be: that is the question" (full line information): {
  "_index": "shakespeare",
  "_type": "_doc",
  "_id": "34229",
  "_score": 25.40203,
  "_source": {
    "type": "line",
    "line_id": 34230,
    "play_name": "Hamlet",
    "speech_number": 19,
    "line_number": "3.1.64",
    "speaker": "HAMLET",
    "text_entry": "To be, or not to be: that is the question:"
  }
}
The 3 most talkative characters in Shakespeare's plays: ['GLOUCESTER', 'HAMLET', 'IAGO']
The most talkative character (GLOUCESTER) was in Shakespeare's plays: ['Richard III', 'King Lear', 'Henry VI Part 2', 'Henry VI Part 3', 'Henry VI Part 1', 'Henry V']

Original queries

The following are the original queries implemented here with python. Once you have the Elasticsearch all set up, you should be able to use Kibana (Dev Tools) - usually found at http://localhost:5601/, to test these queries and make sure the database is set correctly.

GET /shakespeare/_count

GET /shakespeare/_search
{
  "size": 0, 
  "aggs": {
    "unique_play_name": {
      "terms": {
        "field": "play_name",
        "size": 3
      }
    }
  }
}

GET /shakespeare/_search
{
  "query": {
    "match_phrase": {
      "text_entry": "To be, or not to be: that is the question"
    }
  }
}

GET /shakespeare/_search
{
  "size": 0, 
  "aggs": {
    "unique_play_name": {
      "terms": {
        "field": "speaker",
        "size": 3
      }
    }
  }
}

GET /shakespeare/_search
{
  "query": {
    "term": {
      "speaker": {
        "value": "GLOUCESTER"
      }
    }
  },
  "size": 0,
  "aggs": {
    "unique_play_name": {
      "terms": {
        "field": "play_name"
      }
    }
  }
}

POST /_sql?format=txt
{
    "query": "SELECT COUNT(*) FROM shakespeare"
}

POST /_sql?format=txt
{
    "query": "SELECT play_name FROM shakespeare GROUP BY play_name ORDER BY Count(*) DESC"
}