Graph Datatypes

DSE Graph supports a variety of datatypes. Ruby driver for DSE transparently maps each of those datatypes to a specific Ruby type.

Background

Given

a running dse cluster with graph enabled

And

an existing graph called “datatypes” with schema:

schema.propertyKey('bigint').Bigint().ifNotExists().create();
schema.propertyKey('blob').Blob().ifNotExists().create();
schema.propertyKey('boolean').Boolean().ifNotExists().create();
schema.propertyKey('decimal').Decimal().ifNotExists().create();
schema.propertyKey('double').Double().ifNotExists().create();
schema.propertyKey('duration').Duration().ifNotExists().create();
schema.propertyKey('float').Float().ifNotExists().create();
schema.propertyKey('inet').Inet().ifNotExists().create();
schema.propertyKey('int').Int().ifNotExists().create();
schema.propertyKey('text').Text().ifNotExists().create();
schema.propertyKey('timestamp').Timestamp().ifNotExists().create();
schema.propertyKey('uuid').Uuid().ifNotExists().create();
schema.propertyKey('varint').Varint().ifNotExists().create();
schema.propertyKey('smallint').Smallint().ifNotExists().create();
schema.vertexLabel('datatypes').properties('bigint', 'blob', 'boolean', 'decimal', 'double', 'duration',
                                           'float', 'inet', 'int', 'text', 'timestamp', 'uuid', 'varint',
                                           'smallint').ifNotExists().create();

Using strings

Given

the following example:

require 'dse'

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

blob = 'YmxvYg=='
text = 'text'
session.execute_graph("graph.addVertex(label, 'datatypes', 'blob', blob_value);
                       graph.addVertex(label, 'datatypes', 'text', text_value)",
                       arguments: {blob_value: blob,
                                   text_value: text
                       }
)

vertex = session.execute_graph("g.V().hasLabel('datatypes').has('blob')").first
puts "Blob: #{vertex.properties['blob'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('text')").first
puts "Text: #{vertex.properties['text'].first.value}"

When

it is executed

Then

its output should contain:

Blob: YmxvYg==
Text: text

Using numbers

Given

the following example:

require 'dse'

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

bigint  = 765438000
decimal = BigDecimal('1313123123.234234234234234234123')
double  = 3.141592653589793
float   = 1.25
int     = 4
varint  = 67890656781923123918798273492834712837198237
session.execute_graph("graph.addVertex(label, 'datatypes', 'bigint', bigint_value);
                       graph.addVertex(label, 'datatypes', 'decimal', decimal_value);
                       graph.addVertex(label, 'datatypes', 'double', double_value);
                       graph.addVertex(label, 'datatypes', 'float', float_value);
                       graph.addVertex(label, 'datatypes', 'int', int_value);
                       graph.addVertex(label, 'datatypes', 'varint', varint_value)",
                       arguments: {bigint_value:  bigint,
                                   decimal_value: decimal,
                                   double_value:  double,
                                   float_value:   float,
                                   int_value:     int,
                                   varint_value:  varint
                       }
)

vertex = session.execute_graph("g.V().hasLabel('datatypes').has('bigint')").first
puts "Bigint: #{vertex.properties['bigint'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('decimal')").first
puts "Decimal: #{vertex.properties['decimal'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('double')").first
puts "Double: #{vertex.properties['double'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('float')").first
puts "Float: #{vertex.properties['float'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('int')").first
puts "Int: #{vertex.properties['int'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('varint')").first
puts "Varint: #{vertex.properties['varint'].first.value}"

When

it is executed

Then

its output should contain:

Bigint: 765438000
Decimal: 1313123123.2342343
Double: 3.141592653589793
Float: 1.25
Int: 4
Varint: 67890656781923123918798273492834712837198237

Using identifiers, booleans and ip addresses

Given

the following example:

require 'dse'

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

boolean   = true
inet      = ::IPAddr.new('200.199.198.197')
timestamp = ::Time.at(1358013521, 123000)
uuid      = Cassandra::Uuid.new('00b69180-d0e1-11e2-8b8b-0800200c9a66')
session.execute_graph("graph.addVertex(label, 'datatypes', 'boolean', boolean_value);
                       graph.addVertex(label, 'datatypes', 'inet', inet_value);
                       graph.addVertex(label, 'datatypes', 'timestamp', timestamp_value);
                       graph.addVertex(label, 'datatypes', 'uuid', uuid_value)",
                       arguments: {boolean_value:   boolean,
                                   inet_value:      inet,
                                   timestamp_value: timestamp,
                                   uuid_value:      uuid
                       }
)

vertex = session.execute_graph("g.V().hasLabel('datatypes').has('boolean')").first
puts "Boolean: #{vertex.properties['boolean'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('inet')").first
puts "Inet: #{vertex.properties['inet'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('timestamp')").first
puts "Timestamp: #{vertex.properties['timestamp'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('datatypes').has('uuid')").first
puts "Uuid: #{vertex.properties['uuid'].first.value}"

When

it is executed

Then

its output should contain:

Boolean: true
Inet: 200.199.198.197
Timestamp: 2013-01-12T17:58:41Z
Uuid: 00b69180-d0e1-11e2-8b8b-0800200c9a66

Using Duration datatype

Given

the following example:

require 'dse'

duration = Dse::Graph::Duration.new(2, 3, 1, 4.528)
puts "Days: #{duration.days}"
puts "Hours: #{duration.hours}"
puts "Minutes: #{duration.minutes}"
puts "Seconds: #{duration.seconds}"
puts "Duration as seconds: #{duration.as_seconds}"
puts "String form: #{duration.to_s}"
puts ""

duration2 = Dse::Graph::Duration.parse('P2DT3H1M4.528S')
puts "String form: #{duration2.to_s}"
duration2.days -= 2
duration2.hours += 48
duration2.minutes = -1
duration2.seconds += 120
puts "Duration as seconds: #{duration.as_seconds}"
puts "String form: #{duration2.to_s}"
puts "Are duration values equivalent? #{duration == duration2}"
puts ""

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

session.execute_graph("graph.addVertex(label, 'datatypes', 'duration', duration_value)",
                       arguments: {duration_value: duration}
)

vertex = session.execute_graph("g.V().hasLabel('datatypes').has('duration')").first
puts "Duration: #{vertex.properties['duration'].first.value}"

When

it is executed

Then

its output should contain:

Days: 2
Hours: 3
Minutes: 1
Seconds: 4.528
Duration as seconds: 183664.528
String form: P2DT3H1M4.528S

String form: P2DT3H1M4.528S
Duration as seconds: 183664.528
String form: P0DT51H-1M124.528S
Are duration values equivalent? true

Duration: PT51H1M4.528S

Using geospatial types in DSE 5.0

Given

the following graph schema for “datatypes”:

schema.propertyKey('point').Point().ifNotExists().create();
schema.propertyKey('linestring').Linestring().ifNotExists().create();
schema.propertyKey('polygon').Polygon().ifNotExists().create();
schema.vertexLabel('geotypes').properties('point', 'linestring', 'polygon').ifNotExists().create();

And

the following example:

require 'dse'

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

point       = Dse::Geometry::Point.new(38.0, 21.0)
line_string = Dse::Geometry::LineString.new('LINESTRING (30 10, 10 30, 40 40)')
polygon     = Dse::Geometry::Polygon.new('POLYGON ((0.0 0.0, 10.0 0.0, 10.0 10.0, 0.0 10.0, 0.0 0.0),
                                                   (1.0 1.0, 4.0 9.0, 9.0 1.0, 1.0 1.0))'
)

session.execute_graph("graph.addVertex(label, 'geotypes', 'point', point_value);
                       graph.addVertex(label, 'geotypes', 'linestring', line_string_value);
                       graph.addVertex(label, 'geotypes', 'polygon', polygon_value)",
                       arguments: {point_value:       point,
                                   line_string_value: line_string,
                                   polygon_value:     polygon
                       }
)

vertex = session.execute_graph("g.V().hasLabel('geotypes').has('point')").first
puts "Point: #{vertex.properties['point'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('geotypes').has('linestring')").first
puts "Line_string: #{vertex.properties['linestring'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('geotypes').has('polygon')").first
puts "Polygon: #{vertex.properties['polygon'].first.value}"

When

it is executed

Then

its output should contain:

Point: POINT (38 21)
Line_string: LINESTRING (30 10, 10 30, 40 40)
Polygon: POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0), (1 1, 4 9, 9 1, 1 1))

Using geospatial types in DSE > 5.0

Given

the following graph schema for “datatypes”:

schema.propertyKey('point').Point().withGeoBounds().ifNotExists().create();
schema.propertyKey('linestring').Linestring().withGeoBounds().ifNotExists().create();
schema.propertyKey('polygon').Polygon().withGeoBounds().ifNotExists().create();
schema.vertexLabel('geotypes').properties('point', 'linestring', 'polygon').ifNotExists().create();

And

the following example:

require 'dse'

cluster = Dse.cluster(graph_name: 'datatypes')
session = cluster.connect

point       = Dse::Geometry::Point.new(38.0, 21.0)
line_string = Dse::Geometry::LineString.new('LINESTRING (30 10, 10 30, 40 40)')
polygon     = Dse::Geometry::Polygon.new('POLYGON ((0.0 0.0, 10.0 0.0, 10.0 10.0, 0.0 10.0, 0.0 0.0),
                                                   (1.0 1.0, 4.0 9.0, 9.0 1.0, 1.0 1.0))'
)

session.execute_graph("graph.addVertex(label, 'geotypes', 'point', point_value);
                       graph.addVertex(label, 'geotypes', 'linestring', line_string_value);
                       graph.addVertex(label, 'geotypes', 'polygon', polygon_value)",
                       arguments: {point_value:       point,
                                   line_string_value: line_string,
                                   polygon_value:     polygon
                       }
)

vertex = session.execute_graph("g.V().hasLabel('geotypes').has('point')").first
puts "Point: #{vertex.properties['point'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('geotypes').has('linestring')").first
puts "Line_string: #{vertex.properties['linestring'].first.value}"
vertex = session.execute_graph("g.V().hasLabel('geotypes').has('polygon')").first
puts "Polygon: #{vertex.properties['polygon'].first.value}"

When

it is executed

Then

its output should contain:

Point: POINT (38 21)
Line_string: LINESTRING (30 10, 10 30, 40 40)
Polygon: POLYGON ((0 0, 10 0, 10 10, 0 10, 0 0), (1 1, 4 9, 9 1, 1 1))