FakeSource

FakeSource connector

Support Those Engines

Spark
Flink
SeaTunnel Zeta

Description

The FakeSource is a virtual data source, which randomly generates the number of rows according to the data structure of the user-defined schema, just for some test cases such as type conversion or connector new feature testing

Key Features

• batch
• stream
• exactly-once
• column projection
• parallelism
• support user-defined split

Source Options

Name	Type	Required	Default	Description
tables_configs	list	no	-	Define Multiple FakeSource, each item can contains the whole fake source config description below
schema	config	yes	-	Define Schema information
rows	config	no	-	The row list of fake data output per degree of parallelism see title Options rows Case.
row.num	int	no	5	The total number of data generated per degree of parallelism
split.num	int	no	1	the number of splits generated by the enumerator for each degree of parallelism
split.read-interval	long	no	1	The interval(mills) between two split reads in a reader
map.size	int	no	5	The size of map type that connector generated
array.size	int	no	5	The size of array type that connector generated
bytes.length	int	no	5	The length of bytes type that connector generated
string.length	int	no	5	The length of string type that connector generated
string.fake.mode	string	no	range	The fake mode of generating string data, support range and template, default range，if use configured it to template, user should also configured string.template option
string.template	list	no	-	The template list of string type that connector generated, if user configured it, connector will randomly select an item from the template list
tinyint.fake.mode	string	no	range	The fake mode of generating tinyint data, support range and template, default range，if use configured it to template, user should also configured tinyint.template option
tinyint.min	tinyint	no	0	The min value of tinyint data that connector generated
tinyint.max	tinyint	no	127	The max value of tinyint data that connector generated
tinyint.template	list	no	-	The template list of tinyint type that connector generated, if user configured it, connector will randomly select an item from the template list
smallint.fake.mode	string	no	range	The fake mode of generating smallint data, support range and template, default range，if use configured it to template, user should also configured smallint.template option
smallint.min	smallint	no	0	The min value of smallint data that connector generated
smallint.max	smallint	no	32767	The max value of smallint data that connector generated
smallint.template	list	no	-	The template list of smallint type that connector generated, if user configured it, connector will randomly select an item from the template list
int.fake.template	string	no	range	The fake mode of generating int data, support range and template, default range，if use configured it to template, user should also configured int.template option
int.min	int	no	0	The min value of int data that connector generated
int.max	int	no	0x7fffffff	The max value of int data that connector generated
int.template	list	no	-	The template list of int type that connector generated, if user configured it, connector will randomly select an item from the template list
bigint.fake.mode	string	no	range	The fake mode of generating bigint data, support range and template, default range，if use configured it to template, user should also configured bigint.template option
bigint.min	bigint	no	0	The min value of bigint data that connector generated
bigint.max	bigint	no	0x7fffffffffffffff	The max value of bigint data that connector generated
bigint.template	list	no	-	The template list of bigint type that connector generated, if user configured it, connector will randomly select an item from the template list
float.fake.mode	string	no	range	The fake mode of generating float data, support range and template, default range，if use configured it to template, user should also configured float.template option
float.min	float	no	0	The min value of float data that connector generated
float.max	float	no	0x1.fffffeP+127	The max value of float data that connector generated
float.template	list	no	-	The template list of float type that connector generated, if user configured it, connector will randomly select an item from the template list
double.fake.mode	string	no	range	The fake mode of generating float data, support range and template, default range，if use configured it to template, user should also configured double.template option
double.min	double	no	0	The min value of double data that connector generated
double.max	double	no	0x1.fffffffffffffP+1023	The max value of double data that connector generated
double.template	list	no	-	The template list of double type that connector generated, if user configured it, connector will randomly select an item from the template list
vector.dimension	int	no	4	Dimension of the generated vector, excluding binary vectors
binary.vector.dimension	int	no	8	Dimension of the generated binary vector
vector.float.min	float	no	0	The min value of float data in vector that connector generated
vector.float.max	float	no	0x1.fffffeP+127	The max value of float data in vector that connector generated
common-options		no	-	Source plugin common parameters, please refer to Source Common Options for details

Task Example

Simple:

This example Randomly generates data of a specified type. If you want to learn how to declare field types, click here.

schema = {
  fields {
    c_map = "map<string, array<int>>"
    c_map_nest = "map<string, {c_int = int, c_string = string}>"
    c_array = "array<int>"
    c_string = string
    c_boolean = boolean
    c_tinyint = tinyint
    c_smallint = smallint
    c_int = int
    c_bigint = bigint
    c_float = float
    c_double = double
    c_decimal = "decimal(30, 8)"
    c_null = "null"
    c_bytes = bytes
    c_date = date
    c_timestamp = timestamp
    c_row = {
      c_map = "map<string, map<string, string>>"
      c_array = "array<int>"
      c_string = string
      c_boolean = boolean
      c_tinyint = tinyint
      c_smallint = smallint
      c_int = int
      c_bigint = bigint
      c_float = float
      c_double = double
      c_decimal = "decimal(30, 8)"
      c_null = "null"
      c_bytes = bytes
      c_date = date
      c_timestamp = timestamp
    }
  }
}

Random Generation

16 data matching the type are randomly generated

source {
  # This is a example input plugin **only for test and demonstrate the feature input plugin**
  FakeSource {
    row.num = 16
    schema = {
      fields {
        c_map = "map<string, string>"
        c_array = "array<int>"
        c_string = string
        c_boolean = boolean
        c_tinyint = tinyint
        c_smallint = smallint
        c_int = int
        c_bigint = bigint
        c_float = float
        c_double = double
        c_decimal = "decimal(30, 8)"
        c_null = "null"
        c_bytes = bytes
        c_date = date
        c_timestamp = timestamp
      }
    }
    plugin_output = "fake"
  }
}

Customize the data content Simple:

This is a self-defining data source information, defining whether each piece of data is an add or delete modification operation, and defining what each field stores

source {
  FakeSource {
    schema = {
      fields {
        c_map = "map<string, string>"
        c_array = "array<int>"
        c_string = string
        c_boolean = boolean
        c_tinyint = tinyint
        c_smallint = smallint
        c_int = int
        c_bigint = bigint
        c_float = float
        c_double = double
        c_decimal = "decimal(30, 8)"
        c_null = "null"
        c_bytes = bytes
        c_date = date
        c_timestamp = timestamp
      }
    }
    rows = [
      {
        kind = INSERT
        fields = [{"a": "b"}, [101], "c_string", true, 117, 15987, 56387395, 7084913402530365000, 1.23, 1.23, "2924137191386439303744.39292216", null, "bWlJWmo=", "2023-04-22", "2023-04-22T23:20:58"]
      }
      {
        kind = UPDATE_BEFORE
        fields = [{"a": "c"}, [102], "c_string", true, 117, 15987, 56387395, 7084913402530365000, 1.23, 1.23, "2924137191386439303744.39292216", null, "bWlJWmo=", "2023-04-22", "2023-04-22T23:20:58"]
      }
      {
        kind = UPDATE_AFTER
        fields = [{"a": "e"}, [103], "c_string", true, 117, 15987, 56387395, 7084913402530365000, 1.23, 1.23, "2924137191386439303744.39292216", null, "bWlJWmo=", "2023-04-22", "2023-04-22T23:20:58"]
      }
      {
        kind = DELETE
        fields = [{"a": "f"}, [104], "c_string", true, 117, 15987, 56387395, 7084913402530365000, 1.23, 1.23, "2924137191386439303744.39292216", null, "bWlJWmo=", "2023-04-22", "2023-04-22T23:20:58"]
      }
    ]
  }
}

Due to the constraints of the HOCON specification, users cannot directly create byte sequence objects. FakeSource uses strings to assign bytes type values. In the example above, the bytes type field is assigned "bWlJWmo=", which is encoded from "miIZj" with base64. Hence, when assigning values to bytes type fields, please use strings encoded with base64.

Specified Data number Simple:

This case specifies the number of data generated and the length of the generated value

FakeSource {
  row.num = 10
  map.size = 10
  array.size = 10
  bytes.length = 10
  string.length = 10
  schema = {
    fields {
      c_map = "map<string, array<int>>"
      c_array = "array<int>"
      c_string = string
      c_boolean = boolean
      c_tinyint = tinyint
      c_smallint = smallint
      c_int = int
      c_bigint = bigint
      c_float = float
      c_double = double
      c_decimal = "decimal(30, 8)"
      c_null = "null"
      c_bytes = bytes
      c_date = date
      c_timestamp = timestamp
      c_row = {
        c_map = "map<string, map<string, string>>"
        c_array = "array<int>"
        c_string = string
        c_boolean = boolean
        c_tinyint = tinyint
        c_smallint = smallint
        c_int = int
        c_bigint = bigint
        c_float = float
        c_double = double
        c_decimal = "decimal(30, 8)"
        c_null = "null"
        c_bytes = bytes
        c_date = date
        c_timestamp = timestamp
      }
    }
  }
}

Template data Simple:

Randomly generated according to the specified template

Using template

FakeSource {
  row.num = 5
  string.fake.mode = "template"
  string.template = ["tyrantlucifer", "hailin", "kris", "fanjia", "zongwen", "gaojun"]
  tinyint.fake.mode = "template"
  tinyint.template = [1, 2, 3, 4, 5, 6, 7, 8, 9]
  smalling.fake.mode = "template"
  smallint.template = [10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
  int.fake.mode = "template"
  int.template = [20, 21, 22, 23, 24, 25, 26, 27, 28, 29]
  bigint.fake.mode = "template"
  bigint.template = [30, 31, 32, 33, 34, 35, 36, 37, 38, 39]
  float.fake.mode = "template"
  float.template = [40.0, 41.0, 42.0, 43.0]
  double.fake.mode = "template"
  double.template = [44.0, 45.0, 46.0, 47.0]
  schema {
    fields {
      c_string = string
      c_tinyint = tinyint
      c_smallint = smallint
      c_int = int
      c_bigint = bigint
      c_float = float
      c_double = double
    }
  }
}

Range data Simple:

The specified data generation range is randomly generated

FakeSource {
  row.num = 5
  string.template = ["tyrantlucifer", "hailin", "kris", "fanjia", "zongwen", "gaojun"]
  tinyint.min = 1
  tinyint.max = 9
  smallint.min = 10
  smallint.max = 19
  int.min = 20
  int.max = 29
  bigint.min = 30
  bigint.max = 39
  float.min = 40.0
  float.max = 43.0
  double.min = 44.0
  double.max = 47.0
  schema {
    fields {
      c_string = string
      c_tinyint = tinyint
      c_smallint = smallint
      c_int = int
      c_bigint = bigint
      c_float = float
      c_double = double
    }
  }
}

Generate Multiple tables

This is a case of generating a multi-data source test.table1 and test.table2

FakeSource {
  tables_configs = [
    {
      row.num = 16
      schema {
        table = "test.table1"
        fields {
          c_string = string
          c_tinyint = tinyint
          c_smallint = smallint
          c_int = int
          c_bigint = bigint
          c_float = float
          c_double = double
        }
      }
    },
    {
      row.num = 17
      schema {
        table = "test.table2"
        fields {
          c_string = string
          c_tinyint = tinyint
          c_smallint = smallint
          c_int = int
          c_bigint = bigint
          c_float = float
          c_double = double
        }
      }
    }
  ]
}

Options rows Case

rows = [
  {
    kind = INSERT
    fields = [1, "A", 100]
  },
  {
    kind = UPDATE_BEFORE
    fields = [1, "A", 100]
  },
  {
    kind = UPDATE_AFTER
    fields = [1, "A_1", 100]
  },
  {
    kind = DELETE
    fields = [1, "A_1", 100]
  }
]

Options table-names Case

source {
  # This is a example source plugin **only for test and demonstrate the feature source plugin**
  FakeSource {
    table-names = ["test.table1", "test.table2", "test.table3"]
    parallelism = 1
    schema = {
      fields {
        name = "string"
        age = "int"
      }
    }
  }
}

Options defaultValue Case

Custom data can be generated by row and columns. For the time type, obtain the current time by CURRENT_TIMESTAMP 、CURRENT_TIME 、 CURRENT_DATE

    schema = {
        fields {
            pk_id = bigint
            name = string
            score = int
            time1 = timestamp
            time2 = time
            time3 = date
        }
    }
    # use rows
    rows = [
        {
            kind = INSERT
            fields = [1, "A", 100, CURRENT_TIMESTAMP, CURRENT_TIME, CURRENT_DATE]
        }
    ]

      schema = {
          # use columns
           columns = [
           {
              name = book_publication_time
              type = timestamp
              defaultValue = "2024-09-12 15:45:30"
              comment = "book publication time"
           },
           {
              name = book_publication_time2
              type = timestamp
              defaultValue = CURRENT_TIMESTAMP
              comment = "book publication time2"
           },
           {
              name = book_publication_time3
              type = time
              defaultValue = "15:45:30"
              comment = "book publication time3"
           },
           {
              name = book_publication_time4
              type = time
              defaultValue = CURRENT_TIME
              comment = "book publication time4"
           },
           {
              name = book_publication_time5
              type = date
              defaultValue = "2024-09-12"
              comment = "book publication time5"
           },
           {
              name = book_publication_time6
              type = date
              defaultValue = CURRENT_DATE
              comment = "book publication time6"
           }
       ]
      }

Use Vector Example

source {
  FakeSource {
      row.num = 10
      # Low priority 
      vector.dimension= 4
      binary.vector.dimension = 8
      # Low priority 
      schema = {
           table = "simple_example"
           columns = [
           {
              name = book_id
              type = bigint
              nullable = false
              defaultValue = 0
              comment = "primary key id"
           },
            {
              name = book_intro_1
              type = binary_vector
              columnScale =8
              comment = "vector"
           },
           {
              name = book_intro_2
              type = float16_vector
              columnScale =4
              comment = "vector"
           },
           {
              name = book_intro_3
              type = bfloat16_vector
              columnScale =4
              comment = "vector"
           },
           {
              name = book_intro_4
              type = sparse_float_vector
              columnScale =4
              comment = "vector"
           }
       ]
     }
  }
}

Changelog

2.2.0-beta 2022-09-26

• Add FakeSource Source Connector

2.3.0-beta 2022-10-20

• [Improve] Supports direct definition of data values(row) (2839)
• [Improve] Improve fake source connector: (2944)
  • Support user-defined map size
  • Support user-defined array size
  • Support user-defined string length
  • Support user-defined bytes length
• [Improve] Support multiple splits for fake source connector (2974)
• [Improve] Supports setting the number of splits per parallelism and the reading interval between two splits (3098)

next version

• [Feature] Support config fake data rows 3865
• [Feature] Support config template or range for fake data 3932