Daily Elastic Byte - Interesting field types

data.json

#############################
## unsigned long
#############################
DELETE unsigned_long_test

PUT unsigned_long_test
{
  "mappings": {
    "properties": {
      "my_counter": {
        "type": "unsigned_long"
      }
    }
  }
}

# Long.MAX_VALUE = 9223372036854775807 (signed)
PUT unsigned_long_test/_doc/1
{"my_counter": 9223372036854775808}

GET unsigned_long_test/_search

GET unsigned_long_test/_search
{
  "query": {
    "range": {
      "my_counter": {
        "gt": 9223372036854775807
      }
    }
  }
}

# ensure your JSON parsing library can deal
# with scientific notation
GET unsigned_long_test/_search
{
  "size": 0, 
  "aggs": {
    "by_histogram": {
      "histogram": {
        "field": "my_counter",
        "interval": 50
      }
    }
  }
}


#############################
## version
##
## See https://semver.org
#############################
DELETE version_index

PUT version_index
{
  "mappings": {
    "properties": {
      "my_version": {
        "type": "version"
      }
    }
  }
}

PUT version_index/_bulk?refresh
{ "index" : {} }
{"my_version": "1.5.0-alpha"}
{ "index" : {} }
{"my_version": "1.2.5"}
{ "index" : {} }
{"my_version": "1.2.5-alpha2"}
{ "index" : {} }
{"my_version": "1.2.5-alpha2.beta"}
{ "index" : {} }
{"my_version": "1.2.5-alpha1"}
{ "index" : {} }
{"my_version": "1.2.5-beta1"}

GET version_index/_search?filter_path=**.my_version
{
  "sort": [
    {
      "my_version": {
        "order": "desc"
      }
    }
  ]
}




#############################
## flattened
##
## https://www.elastic.co/guide/en/elasticsearch/reference/7.11/flattened.html
#############################
DELETE flattened

PUT flattened
{
  "mappings": {
    "properties": {
      "labels": {
        "type": "flattened"
      }
    }
  }
}

POST flattened/_doc/1
{
  "labels": {
    "size": "xl",
    "color" : "red",
    "type" : "v-neck",
    "price": 25,
    "last_updated" : 1614594536173
  }
}

GET flattened/_mapping

POST flattened/_doc/2?refresh
{
  "labels": {
    "last_updated" : 2
  }
}

POST flattened/_search
{
  "query": {
    "term": {"labels": "red"}
  }
}

POST flattened/_search
{
  "query": {
    "term": {"labels.color": "red"}
  }
}

# everything is stored as a keyword
# so the timestamp of 2 is greater than the other tone
POST flattened/_search
{
  "sort": [
    {
      "labels.last_updated": {
        "order": "desc"
      }
    }
  ]
}





#############################
## aggregate_metric
##
## https://www.elastic.co/guide/en/elasticsearch/reference/7.11/aggregate-metric-double.html
#############################
DELETE sales_per_day

PUT sales_per_day
{
  "mappings": {
    "properties": {
      "agg_metric": {
        "type": "aggregate_metric_double",
        "metrics": [ "min", "max", "value_count", "sum" ],
        "default_metric": "value_count"
      }
    }
  }
}

PUT sales_per_day/_doc/january
{
  "agg_metric": {
    "min": 13,
    "max": 34,
    "sum" : 4653.00, 
    "value_count": 234
  }
}

PUT sales_per_day/_doc/february?refresh
{
  "agg_metric": {
    "min": 5,
    "max": 28,
    "sum" : 3124.34, 
    "value_count": 260
  }
}

POST sales_per_day/_search?size=0
{
  "aggs": {
    "metric_min": { "min": { "field": "agg_metric" } },
    "metric_max": { "max": { "field": "agg_metric" } },
    "metric_value_count": { "value_count": { "field": "agg_metric" } },
    "metric_avg": { "avg": { "field": "agg_metric" } }
  }
}





#############################
## dense_vector
##
## https://www.elastic.co/guide/en/elasticsearch/reference/7.11/dense-vector.html
#############################
DELETE books

# add vector field
PUT books
{
  "mappings": {
    "properties": {
      "vector_recommendation": {
        "type": "dense_vector",
        "dims": 3
      }
    }
  }
}

# Add a vector for each document
# Features are the following
# First dimension is the number of pages divided by 100 (0-10)
# Second dimension is the rating (0-5)
# Third dimension is the price (0-10) 0 means cost > 100 , 10 cost < 10
PUT books/_bulk
{ "index" : { "_id" : "database-internals" } }
{ "vector_recommendation" : [3.5, 4.5, 5.2] } 
{ "index" : { "_id" : "designing-data-intensive-applications" } }
{ "vector_recommendation" : [5.9, 4.4, 6.8] } 
{ "index" : { "_id" : "kafka-the-definitive-guide" } }
{ "vector_recommendation" : [2.97, 3.9, 6.2] }
{ "index" : { "_id" : "effective-java" } }
{ "vector_recommendation" : [4.12, 4.2, 7.2] }
{ "index" : { "_id" : "daemon" } }
{ "vector_recommendation" : [4.48, 4.0, 8.7] }
{ "index" : { "_id" : "cryptonomicon" } }
{ "vector_recommendation" : [10.0, 4.0, 9.3] }
{ "index" : { "_id" : "garbage-collection-handbook" } }
{ "vector_recommendation" : [5.1, 5.0, 1.3] } 
{ "index" : { "_id" : "radical-candor" } }
{ "vector_recommendation" : [4.0, 4.0, 9.2] } 
{ "index" : { "_id" : "never-split-the-difference" } }
{ "vector_recommendation" : [4.0, 4.3, 8.9] } 
{ "index" : { "_id" : "not-giving-a-fsck" } }
{ "vector_recommendation" : [2.8, 4.4, 8.9] } 
{ "index" : { "_id" : "permanent-record" } }
{ "vector_recommendation" : [3.3, 4.7, 8.7] } 

GET books/_search
{
  "query": {
    "exists": { "field": "vector_recommendation"}
  }
}

# short, good rating, cheap
GET books/_search
{
  "query": {
    "script_score": {
      "query" : {
        "match_all" : {}
      },
      "script": {
        "source": "cosineSimilarity(params.query_vector, 'vector_recommendation') + 1.0", 
        "params": {
          "query_vector": [1.0, 5.0, 10.0]  
        }
      }
    }
  }
}


# long, good rating, any price
GET books/_search
{
  "query": {
    "script_score": {
      "query" : {
        "match_all" : {}
      },
      "script": {
        "source": "cosineSimilarity(params.query_vector, 'vector_recommendation') + 1.0", 
        "params": {
          "query_vector": [10.0, 5.0, 5.0]  
        }
      }
    }
  }
}



#############################
## Do you want to know more?
##
## https://www.elastic.co/blog/aggregate-all-the-things-new-aggregations-in-elasticsearch-7
#############################