You must be logged in to view saved presets
This template sets up an AWS Glue ML Transform for finding matches, along with a Glue Catalog Database and a Glue Catalog Table configured with various properties including storage, serialization, and partitioning.
resource "aws_glue_catalog_database" "test" {
name = "example"
}
resource "aws_glue_catalog_table" "test" {
database_name = aws_glue_catalog_database.test.name
name = "example"
owner = "my_owner"
parameters = {
param1 = "param1_val"
}
partition_keys {
comment = "my_column_1_comment"
name = "my_column_1"
type = "int"
}
partition_keys {
comment = "my_column_2_comment"
name = "my_column_2"
type = "string"
}
retention = 1
storage_descriptor {
bucket_columns = ["bucket_column_1"]
columns {
comment = "my_column1_comment"
name = "my_column_1"
type = "int"
}
columns {
comment = "my_column2_comment"
name = "my_column_2"
type = "string"
}
compressed = false
input_format = "SequenceFileInputFormat"
location = "my_location"
number_of_buckets = 1
output_format = "SequenceFileInputFormat"
parameters = {
param1 = "param1_val"
}
ser_de_info {
name = "ser_de_name"
parameters = {
param1 = "param_val_1"
}
serialization_library = "org.apache.hadoop.hive.serde2.columnar.ColumnarSerDe"
}
skewed_info {
skewed_column_names = ["my_column_1"]
skewed_column_value_location_maps = ["my_column_1_val_loc_map"]
skewed_column_values = {
0 = "skewed_val_1"
}
}
sort_columns {
column = "my_column_1"
sort_order = 1
}
stored_as_sub_directories = false
}
table_type = "VIRTUAL_VIEW"
view_expanded_text = "view_expanded_text_1"
view_original_text = "view_original_text_1"
}
resource "aws_glue_ml_transform" "test" {
depends_on = ["aws_iam_role_policy_attachment.test"]
input_record_tables = [aws_glue_catalog_table.test.database_name, aws_glue_catalog_table.test.name]
name = "example"
parameters {
find_matches_parameters {
primary_key_column_name = "my_column_1"
}
transform_type = "FIND_MATCHES"
}
role_arn = "aws_iam_role.test.arn"
}