elasticsearch-sense-examples

###################################################
###################################################
# Crash course
###################################################
###################################################
En
###################################################
# Quickstart
###################################################

# Delete the "books" index, just in case it is still there from previous experiments.
DELETE /books

# Create an empty "books" index.
PUT /books

# Add a book of type "tech" to the index.
PUT /books/tech/1
{
  "author": "Joshua Bloch",
  "title": "Effective Java",
  "date": "2008-05-08",
  "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"
}

# Retrieve the book again by ID.
GET /books/tech/1

# Add another book.
PUT /books/tech/2
{
  "author": "Robert C. Martin",
  "title": "Clean Code",
  "date": "2008-08-01",
  "text": "Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."
}

# Retrieve both books by ID.
GET /books/tech/_mget
{
  "ids": ["1", "2"]
}

# Add one more book.
PUT /books/tech/3
{
  "author": "Brian Goetz",
  "title": "Java Concurrency in Practice",
  "date": "2006-05-09",
  "text": "Writing correct programs is hard; writing correct concurrent programs is harder."
}

# Search for books with "clean" in the title.
GET /books/tech/_search
{
  "query": {
    "match": {"title": "clean"}
  }
}

# Search for books with "code" in the text.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "code"}
  }
}

# Only retrieve some fields.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "code"}
  },
  "fields": ["author", "title"]
}

### Talk about scoring.

# Search for multiple fields. Here: Books with "code" in the text and "java" in the title.
GET /books/tech/_search
{
  "query": {
    "bool": {
      "must": [
        {"match": {"text": "code"}},
        {"match": {"title": "java"}}
      ]
    }
  }
}

# Search for books with "resource" in the text. Won't match because no stemming is applied.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "resource"}
  }
}





###################################################
# Analysis
###################################################

### Talk about the concept of inverted index and what it means for indexing and querying.

# Whitespace analyzer.
GET /books/_analyze?analyzer=whitespace&text=I like good books about BigData

# Standard analyzer. Also lowercases.
GET /books/_analyze?analyzer=standard&text=I like good books about BigData

# English analyzer. Stems words. Here: books -> book.
GET /books/_analyze?analyzer=english&text=I like good books about BigData

# Custom analyzer.
GET /books/_analyze?tokenizer=keyword&filters=lowercase&text=I like good books about BigData

# Custom analyzer with word delimiter filter (wrong order).
GET /books/_analyze?tokenizer=keyword&filters=lowercase,word_delimiter&text=I like good books about BigData

# Custom analyzer with word delimiter filter (correct order).
GET /books/_analyze?tokenizer=keyword&filters=word_delimiter,lowercase&text=I like good books about BigData

# Analyzer used by a field of the index.
GET /books/_analyze?field=text&text=I like good books about BigData

### Show inquisitor plugin.

### Show reference guide to analysis. As an example maybe Word Delimiter filter.





###################################################
# Mapping
###################################################

# Get the current mappings of the books index.
GET /books/_mapping

# Close the index in order to update a non-dynamic setting.
POST /books/_close

# Create a custom analyzer.
PUT /books/_settings
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        }
      }
    }
  }
}

# Open the index again.
POST /books/_open

# Test the analyzer.
GET /books/_analyze?analyzer=my_title_analyzer&text=I like good books about BigData

# Try to update the analyzer of the title field. Won't work because the field is already mapped.
PUT /books/tech/_mapping
{
  "tech": {
    "properties": {
      "title": {
        "type": "string",
        "analyzer": "my_title_analyzer"
      }
    }
  }
}

# Delete the old mapping (and with it, the documents).
DELETE /books/tech/_mapping

# Try again.
PUT /books/tech/_mapping
{
  "tech": {
    "properties": {
      "title": {
        "type": "string",
        "analyzer": "my_title_analyzer"
      }
    }
  }
}

# Check the mapping.
GET /books/_mapping

# Add the books again. Note: Bulk action.
POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

# Check the mapping again.
GET /books/_mapping

# See if it works.
GET /books/_search
{
  "query": {
    "match": {"title": "effec"}
  }
}

# Umm, why does this also match? We had a maxgram size of 6, but this is 7.
GET /books/_search
{
  "query": {
    "match": {"title": "effecti"}
  }
}

### Check with inquisitor.

# NGrams are created for the query, too. We don't want this. See how, as it is now, "clean" will also match "Concurrency" in a book title.
GET /books/_search
{
  "query": {
    "match": {"title": "clean"}
  }
}

# So let's split up the analyzers for indexing and querying.
DELETE /books
PUT /books
POST /books/_close
PUT /books/_settings
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_index_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        },
        "my_title_search_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "lowercase"]
        }
      }
    }
  }
}
POST /books/_open

# And now configure separate index and search analyzers for the title field.
PUT /books/tech/_mapping
{
  "tech": {
    "properties": {
      "title": {
        "type": "string",
        "index_analyzer": "my_title_index_analyzer",
        "search_analyzer": "my_title_search_analyzer"
      }
    }
  }
}
POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

# And let's check again.
GET /books/_search
{
  "query": {
    "match": {"title": "effecti"}
  }
}

### Confirm with Inquisitor.

### Note: Usually, this is the way you evolve your mapping, and when you are done you store it on index creation.

# A "final" mapping might be:
DELETE /books
PUT /books
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_index_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        },
        "my_title_search_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "lowercase"]
        }
      }
    }
  },
  "mappings": {
    "tech": {
      "properties": {
        "author": {
          "type": "string",
          "analyzer": "standard"
        },
        "title": {
          "type": "string",
          "index_analyzer": "my_title_index_analyzer",
          "search_analyzer": "my_title_search_analyzer"
        },
        "date": {
          "type": "date"
        },
        "text": {
          "type": "string",
          "analyzer": "english"
        }
      }
    }
  }
}

POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

# Example with stemming, as we have now configured the English analyzer for the text field.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "resource"}
  }
}

### Confirm with Inquisitor.





###################################################
# Search features
###################################################

# Highlighting.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "resource"}
  },
  "highlight": {
    "fields": { "text": {}}
  }
}

# Suggestions. Here: May helo to fix a typo.
GET /books/tech/_search
{
  "query": {
    "match": { "text": "claen"}
  },
  "suggest": {
    "my-suggestion-1": {
      "text": "claen",
      "term": { "field": "text"}
    }
  }
}

# More like this. Here: Title "Effective Java".
GET /books/tech/_search
{
  "query": {
    "more_like_this": {
      "fields": ["title"],
      "ids": ["1"],
      "min_term_freq": 1,
      "min_doc_freq": 1
    }
  }
}

# Percolator.
GET /books/.percolator/1
{
  "query" : {
        "match": {"title" : "java"}
    }
}

# Percolate a new document. Note: Indexing needs to be done separately.
GET /books/tech/_percolate
{
  "doc": {
    "author": ["Charlie Hunt", "Binu John"],
    "title": "Java Performance",
    "date": "2011-10-14",
    "text": "Improvements in the Java platform and new multicore/multiprocessor hardware have made it possible to dramatically improve the performance and scalability of Java software."
  }
}

# Aggregation.
GET /books/tech/_search?search_type=count
{
  "aggs": {
    "year_of_publication": {
      "date_histogram": {
        "field": "date",
        "interval": "year"
      }
    }
  }
}





###################################################
# Sharding + Replication
###################################################

### Show that documents live in different shards. Introduce "routing" to assign documents to shards.

### Explain the advantages of sharding and replication.

### Start up three more cluster nodes. Show what happens in Marvel or Head plugin.

### Discuss number of shards/replicas.

# Add more replicas.
PUT /books/_settings
{
  "index": {"number_of_replicas": 2}
}

### Shut down a node. See what happens.

### Talk about master role.





###################################################
###################################################
# The road to expertise
###################################################
###################################################

###################################################
# Map a field multiple times
###################################################

# Start all over.
DELETE /books
PUT /books
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_index_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        },
        "my_title_search_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "lowercase"]
        }
      }
    }
  },
  "mappings": {
    "tech": {
      "properties": {
        "author": {
          "type": "string",
          "analyzer": "standard"
        },
        "title": {
          "type": "string",
          "index_analyzer": "my_title_index_analyzer",
          "search_analyzer": "my_title_search_analyzer"
        },
        "date": {
          "type": "date"
        },
        "text": {
          "type": "string",
          "analyzer": "english"
        }
      }
    }
  }
}

POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

# See what a search for "programming" brings.
GET /books/tech/_search
{
  "query": {
    "match": {"text": "programming"}
  },
  "highlight": {
    "fields": {"text": {}}
  }
}

# Map a field twice.
DELETE /books
PUT /books
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_index_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        },
        "my_title_search_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "lowercase"]
        }
      }
    }
  },
  "mappings": {
    "tech": {
      "properties": {
        "author": {
          "type": "string",
          "analyzer": "standard"
        },
        "title": {
          "type": "string",
          "index_analyzer": "my_title_index_analyzer",
          "search_analyzer": "my_title_search_analyzer"
        },
        "date": {
          "type": "date"
        },
        "text": {
          "type": "string",
          "analyzer": "english",
          "fields": {
            "standard": {
              "type": "string",
              "analyzer": "standard"
            }
          }
        }
      }
    }
  }
}

POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

# Example with higher score for documents that match both fields.
GET /books/tech/_search
{
  "query": {
    "bool": {
      "should": [
        {"match": {"text": "programming"}},
        {"match": {"text.standard": "programming"}}
      ]
    }
  },
  "highlight": {
    "fields": {"text": {}}
  }
}





###################################################
# Explain
###################################################

# Same query as before but with score explanation.
GET /books/tech/_search?explain
{
  "query": {
    "bool": {
      "should": [
        {"match": {"text": "programming"}},
        {"match": {"text.standard": "programming"}}
      ]
    }
  },
  "highlight": {
    "fields": {"text": {}}
  }
}

# Query validation.
GET /books/tech/_validate/query?explain
{
  "query": {
    "bool": {
      "should": [
        {"match": {"text": "programming"}},
        {"match": {"text.standard": "programming"}}
      ]
    }
  }
}





###################################################
# Filters
###################################################

# Let's say we have a publisher field.
PUT /books/tech/_mapping
{
  "tech": {
    "properties": {
      "publisher": {
        "type": "string",
        "index": "not_analyzed"
      }
    }
  }
}

# Re-index a book with publisher information.
PUT /books/tech/1
{
  "author": "Joshua Bloch",
  "title": "Effective Java",
  "date": "2008-05-08",
  "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!",
  "publisher": "Addison Wesley"
}

# Combine a query with a filter for publisher and a date filter.
GET /books/tech/_search
{
  "query": {
    "filtered": {
      "query": {
        "match": {"text": "program"}
      },
      "filter": {
        "bool": {
          "must": [
            {"term": {"publisher": "Addison Wesley"}},
            {"range": {"date": {"gte": "2008-01-01"}}}
          ]
        }
      }
    }
  }
}





###################################################
# Updates
###################################################

# Update the title of a book.
POST /books/tech/2/_update
{
  "doc": {
    "title": "Clean Code: A Handbook of Agile Software Craftsmanship"
  }
}

GET books/tech/2

# Add two comments to a book.
POST /books/tech/1/_update
{
  "doc": {
    "comments": [
      {
        "author": "Patrick Peschlow",
        "text": "Great book!"
      },
      {
        "author": "Daniel Schneller",
        "text": "I don't like it."
      }
    ]
  }
}

GET books/tech/1





###################################################
# Relations
###################################################

# See what search for those comments might bring.
GET /books/tech/_search
{
  "query": {
    "bool": {
      "must": [
        {"match": {"comments.author": "Patrick Peschlow"}},
        {"match": {"comments.text": "I don't like it."}
        }
      ]
    }
  }
}

# Let's create a new mapping with type "nested" for comments.
DELETE /books
PUT /books
{
  "settings": {
    "analysis": {
      "type": "custom",
      "filter": {
        "my_edge_ngram_filter": {
          "type": "edgeNGram",
          "min_gram": "1",
          "max_gram": "6"
        }
      },
      "analyzer": {
        "my_title_index_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "my_edge_ngram_filter", "lowercase"]
        },
        "my_title_search_analyzer": {
          "tokenizer": "whitespace",
          "filter": ["word_delimiter", "lowercase"]
        }
      }
    }
  },
  "mappings": {
    "tech": {
      "properties": {
        "author": {
          "type": "string",
          "analyzer": "standard"
        },
        "title": {
          "type": "string",
          "index_analyzer": "my_title_index_analyzer",
          "search_analyzer": "my_title_search_analyzer"
        },
        "date": {
          "type": "date"
        },
        "text": {
          "type": "string",
          "analyzer": "english"
        },
        "comments": {
          "type": "nested"
        }
      }
    }
  }
}

POST /books/tech/_bulk
{"index": {"_id": "1"}}
{"author": "Joshua Bloch", "title": "Effective Java", "date": "2008-05-08", "text": "Are you looking for a deeper understanding of the Java programming language so that you can write code that is clearer, more correct, more robust, and more reusable? Look no further!"}
{"index": {"_id": "2"}}
{"author": "Robert C. Martin", "title":"Clean Code", "date":"2008-08-01", "text":"Even bad code can function. But if code isn't clean, it can bring a development organization to its knees. Every year, countless hours and significant resources are lost because of poorly written code. But it doesn't have to be that way."}
{"index": {"_id": "3"}}
{"author": "Brian Goetz", "title": "Java Concurrency in Practice", "date": "2006-05-09", "text":"Writing correct programs is hard; writing correct concurrent programs is harder."}

POST /books/tech/1/_update
{
  "doc": {
    "comments": [
      {
        "author": "Patrick Peschlow",
        "text": "Great book!"
      },
      {
        "author": "Daniel Schneller",
        "text": "I don't like it."
      }
    ]
  }
}

GET books/tech/1

# Now search again. No result anymore!
GET /books/tech/_search
{
  "query": {
    "bool": {
      "must": [
        {"match": {"comments.author": "Patrick Peschlow"}},
        {"match": {"comments.text": "I don't like it."}}
      ]
    }
  }
}

# Do a nested search.
GET /books/tech/_search
{
  "query": {
    "nested": {
      "path": "comments",
      "query": {
        "bool": {
          "must": [
            {"match": {"comments.author": "Patrick Peschlow"}},
            {"match": {"comments.text": "Great book!"}}
          ]
        }
      }
    }
  }
}

# Double-check. Shouldn't match the bad combination.
GET /books/tech/_search
{
  "query": {
    "nested": {
      "path": "comments",
      "query": {
        "bool": {
          "must": [
            {"match": {"comments.author": "Patrick Peschlow"}},
            {"match": {"comments.text": "I don't like it."}}
          ]
        }
      }
    }
  }
}

### Talk about parent/child.





###################################################
# Alias
###################################################

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "books",
        "alias": "literature"
      }
    }
  ]
}

GET _aliases

GET /literature/tech/_search
{
  "query": {
    "match_all": {}
  }
}

POST /_aliases
{
  "actions": [
    {
      "add": {
        "index": "books",
        "alias": "thatday",
        "filter": {
          "term": {"date": "2006-05-09"}
        }
      }
    }
  ]
}

GET /thatday/_search
{
  "query": {
    "match_all": {}
  }
}