Data Types

This document provides a comprehensive overview of all data types supported by NeuG.

Data Types Summary Table

The following table showcases all data types supported by NeuG and their differences with Neo4j. The System Default Value column indicates the value automatically assigned by the system during data import when the user does not explicitly define a default value in the schema and the corresponding data field is not given (or given as Null value) in the raw data. This mechanism prevents Null values, ensures data consistency, and provides stable defaults for subsequent queries and computations.

Detailed Introduction

Primitive Types

INT32

• Description: 32-bit signed integer type
• Range: [2,147,483,648, 2,147,483,647]
• Query Example: Return CAST(42, 'INT32') as int32_value;

UINT32

• Description: 32-bit unsigned integer type
• Range: [0, 4,294,967,295]
• Query Example: RETURN CAST(42, 'UINT32') AS uint32_value;

INT64

• Description: 64-bit signed integer type, default type of integer values
• Range: [-9,223,372,036,854,775,808, 9,223,372,036,854,775,807]
• Query Example: RETURN 9223372036854775807 AS int64_value;

UINT64

• Description: 64-bit unsigned integer type
• Range: [0, 18,446,744,073,709,551,615]
• Query Example: RETURN CAST(18446744073709551615, 'UINT64') AS uint64_value;

FLOAT

• Description: Single-precision floating-point number
• Precision: ~7 decimal digits
• Query Example: RETURN CAST(3.14, 'FLOAT') AS float_value;

DOUBLE

• Description: Double-precision floating-point number, default type of float values
• Precision: ~15-17 decimal digits
• Query Example: RETURN 3.14159265359 AS double_value;

BOOL

• Description: Boolean type representing true or false values
• Values: true, false
• Query Example: RETURN true AS bool_value;

NULL

• Description: Represents missing or undefined values
• Query Example: RETURN null AS null_value;

String Types

VARCHAR

• Description: Variable-length character string with UTF-8 encoding
• Query Example: RETURN 'Hello World' AS string_value;
• Length: Variable, limited by system constraints, default is 65536

Temporal Types

DATE

• Description: Date type for storing calendar dates
• Format: YYYY-MM-DD
• Query Example: RETURN date('2022-06-06') AS date_value;

DATETIME

• Description: Combination of date and time type
• Format: YYYY-MM-DD HH:MM:SS
• Query Example: RETURN timestamp('2022-06-06 12:00:00') AS datetime_value;

INTERVAL

• Description: The INTERVAL type represents a duration or time interval and consists of the following fields: year, month, day, hour, minute, second, millisecond, and microsecond . The INTERVAL type supports two primary formats for specifying values:
  • Date-based components (year, month, day): Specified using a natural language format. Example: 1 year 2 month 3 day.
  • Time-based components (hour, minute, second, millisecond, microsecond): Specified using a natural language format. Example: 12 hour 12 minute 2 second - represents 12 hours, 12 minutes, and 2 seconds.
• Query Example: RETURN interval('1 year 2 month 3 day 12 hour 12 minute 2 second') AS interval_value;

Composite Types

LIST

• Description: Ordered collection of values with heterogeneous types
• Query Example: RETURN [1, 2, 3] AS list_value;

The following table shows all Component Types that LIST can support:

Important Note on LIST Component Types:

NeuG supports lists through tuple data types, meaning composite types can be heterogeneous. Here are some examples:

Mixing different primitive types in a single list:

RETURN ['marko', 2];

Combining different property types from nodes in a list:

MATCH (n:person) RETURN [n.name, n.age];

Supporting nested list structures:

MATCH (n:person) RETURN [["name", n.name], ["age", n.age]];

Key Technical Details:

• Lists in NeuG can contain elements of different data types (heterogeneous lists)
• This is achieved through internal tuple data type support
• Type conversion is handled automatically when possible
• Nested lists are fully supported for complex data structures
• The system maintains type safety while allowing flexibility in list composition

Graph Types

NODE

• Description: Represents a node in the graph
• Internal Structure (order is insignificant): _ID (internal identifier), _LABEL (indication of node type) and property fields
• Query Example: MATCH (n:person) RETURN n AS node_value;
• NeuG Format: {_ID: 0, _LABEL: person, id: 1, name: marko, age: 29}

REL (Edge)

• Description: Represents an edge in the graph
• Internal Structure (order is insignificant): _ID (edge internal identifier), _LABEL (indication of edge type), _SRC_ID (internal identifier of source node), _SRC_LABEL (label of source node), _DST_ID (internal identifier of destination node), _DST_LABEL (label of destination node), and property fields
• Query Example: MATCH ()-[r:knows]->() RETURN r AS rel_value;
• NeuG Format: {_ID: 2, _LABEL: knows, _SRC_ID: 0, _SRC_LABEL: person, _DST_ID: 2, _DST_LABEL: person, weight: 1.0}

PATH

• Description: Represents a graph path formed by alternating nodes and edges.
• Internal Structure: An ordered sequence of nodes and edges along the path, including the starting and ending nodes.
• Query Example: MATCH (a:person)-[p*1..2]->(c) RETURN p AS path_value;
• NeuG Format: {_ID: 0, _LABEL: person}, {_ID: 4294967298, _LABEL: created, _SRC_LABEL: person, _DST_LABEL: person, _SRC_ID: 0, _DST_ID: 2}, {_ID: 2, _LABEL: person}, {_ID: 4297064449, _LABEL: created, _SRC_LABEL: person, _DST_LABEL: software, _SRC_ID: 2, _DST_ID: 72057594037927937}, {_ID: 72057594037927937, _LABEL: software}