Onkar Ravgan 7fdab21a72
Glue Topology Refactor (#6231)
* rebase with main

* refactored glue for topology

* added glue unit tests

* resolved review comments

Co-authored-by: Onkar Ravgan <onkarravgan@Onkars-MacBook-Pro.local>
2022-07-28 15:04:43 +02:00

417 lines
12 KiB
JSON

{
"mock_database_paginator": {
"DatabaseList": [
{
"Name": "default",
"CreateTime": "2022-05-13 17:02:17",
"CreateTableDefaultPermissions": [
{
"Principal": {
"DataLakePrincipalIdentifier": "IAM_ALLOWED_PRINCIPALS"
},
"Permissions": [
"ALL"
]
}
],
"CatalogId": "118146679784"
},
{
"Name": "mydatabase",
"CreateTime": "2022-05-13 17:02:17",
"CreateTableDefaultPermissions": [
{
"Principal": {
"DataLakePrincipalIdentifier": "IAM_ALLOWED_PRINCIPALS"
},
"Permissions": [
"ALL"
]
}
],
"CatalogId": "118146679784"
},
{
"Name": "testdatalake_db",
"LocationUri": "s3://awsdatalake-testing/data",
"CreateTime": "2022-05-13 17:02:17",
"CreateTableDefaultPermissions": [
{
"Principal": {
"DataLakePrincipalIdentifier": "IAM_ALLOWED_PRINCIPALS"
},
"Permissions": [
"ALL"
]
}
],
"CatalogId": "118146679784"
},
{
"Name": "zipcode-db",
"LocationUri": "s3://datalake-openmetadata-ohio/zipcode",
"CreateTime": "2022-05-13 17:02:17",
"CreateTableDefaultPermissions": [
{
"Principal": {
"DataLakePrincipalIdentifier": "IAM_ALLOWED_PRINCIPALS"
},
"Permissions": [
"ALL"
]
}
],
"CatalogId": "118146679784"
}
],
"ResponseMetadata": {
"RequestId": "21a5c41c-da65-4531-83cb-e6d3425c4ec7",
"HTTPStatusCode": 200,
"HTTPHeaders": {
"date": "Tue, 26 Jul 2022 16:28:52 GMT",
"content-type": "application/x-amz-json-1.1",
"content-length": "1062",
"connection": "keep-alive",
"x-amzn-requestid": "21a5c41c-da65-4531-83cb-e6d3425c4ec7"
},
"RetryAttempts": 0
}
},
"mock_table_paginator": {
"TableList": [
{
"Name": "cloudfront_logs",
"DatabaseName": "default",
"Owner": "hadoop",
"CreateTime": "2022-05-13 17:02:17",
"UpdateTime": "2022-05-13 17:02:17",
"LastAccessTime": "2022-05-13 17:02:17",
"Retention": 0,
"StorageDescriptor": {
"Columns": [
{
"Name": "date",
"Type": "date"
},
{
"Name": "time",
"Type": "string"
},
{
"Name": "location",
"Type": "string"
},
{
"Name": "bytes",
"Type": "int"
},
{
"Name": "requestip",
"Type": "string"
},
{
"Name": "method",
"Type": "string"
},
{
"Name": "host1",
"Type": "string"
},
{
"Name": "uri1",
"Type": "string"
},
{
"Name": "status",
"Type": "int"
},
{
"Name": "referrer",
"Type": "string"
},
{
"Name": "os",
"Type": "string"
},
{
"Name": "browser",
"Type": "string"
},
{
"Name": "browserversion",
"Type": "string"
}
],
"Location": "s3://athena-examples-MyRegion/cloudfront/plaintext",
"InputFormat": "org.apache.hadoop.mapred.TextInputFormat",
"OutputFormat": "org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat",
"Compressed": false,
"NumberOfBuckets": -1,
"SerdeInfo": {
"SerializationLibrary": "org.apache.hadoop.hive.serde2.RegexSerDe",
"Parameters": {
"input.regex": "^(?!#)([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s ([^ ] )\\s [^(] [(]([^;] ).*\\ ([^/] )[/](.*)$",
"serialization.format": "1"
}
},
"BucketColumns": [],
"SortColumns": [],
"Parameters": {},
"SkewedInfo": {
"SkewedColumnNames": [],
"SkewedColumnValues": [],
"SkewedColumnValueLocationMaps": {}
},
"StoredAsSubDirectories": false
},
"PartitionKeys": [],
"TableType": "EXTERNAL_TABLE",
"Parameters": {
"EXTERNAL": "TRUE",
"transient_lastDdlTime": "1652441537"
},
"CreatedBy": "arn:aws:iam::118146679784:user/abhishek-pandey",
"IsRegisteredWithLakeFormation": false,
"CatalogId": "118146679784"
},
{
"Name": "cloudfront_logs2",
"DatabaseName": "default",
"Description": "TEST DESCRIPTION",
"Owner": "hadoop",
"CreateTime": "2022-05-13 17:02:17",
"UpdateTime": "2022-05-13 17:02:17",
"LastAccessTime": "2022-05-13 17:02:17",
"Retention": 0,
"StorageDescriptor": {
"Columns": [
{
"Name": "date",
"Type": "date",
"Comment": "TEST COMMENT 1"
},
{
"Name": "time",
"Type": "string"
},
{
"Name": "location",
"Type": "string",
"Comment": "NEW LOC"
},
{
"Name": "bytes",
"Type": "bigint",
"Comment": "TEST BYTES"
},
{
"Name": "request_ip",
"Type": "string"
},
{
"Name": "method",
"Type": "string"
},
{
"Name": "host",
"Type": "string"
},
{
"Name": "uri",
"Type": "string"
},
{
"Name": "status",
"Type": "int"
},
{
"Name": "referrer",
"Type": "string"
},
{
"Name": "user_agent",
"Type": "string"
},
{
"Name": "query_string",
"Type": "string"
},
{
"Name": "cookie",
"Type": "string"
},
{
"Name": "result_type",
"Type": "string"
},
{
"Name": "request_id",
"Type": "string"
},
{
"Name": "host_header",
"Type": "string"
},
{
"Name": "request_protocol",
"Type": "string"
},
{
"Name": "request_bytes",
"Type": "bigint"
},
{
"Name": "time_taken",
"Type": "float"
},
{
"Name": "xforwarded_for",
"Type": "string"
},
{
"Name": "ssl_protocol",
"Type": "string"
},
{
"Name": "ssl_cipher",
"Type": "string"
},
{
"Name": "response_result_type",
"Type": "string"
},
{
"Name": "http_version",
"Type": "string"
},
{
"Name": "fle_status",
"Type": "string"
},
{
"Name": "fle_encrypted_fields",
"Type": "int"
},
{
"Name": "c_port",
"Type": "int"
},
{
"Name": "time_to_first_byte",
"Type": "float"
},
{
"Name": "x_edge_detailed_result_type",
"Type": "string"
},
{
"Name": "sc_content_type",
"Type": "string"
},
{
"Name": "sc_content_len",
"Type": "bigint"
},
{
"Name": "sc_range_start",
"Type": "bigint"
},
{
"Name": "sc_range_end",
"Type": "bigint"
}
],
"Location": "s3://athena-postgres/",
"InputFormat": "org.apache.hadoop.mapred.TextInputFormat",
"OutputFormat": "org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat",
"Compressed": false,
"NumberOfBuckets": -1,
"SerdeInfo": {
"SerializationLibrary": "org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe",
"Parameters": {
"field.delim": "\t",
"serialization.format": "\t"
}
},
"BucketColumns": [],
"SortColumns": [],
"Parameters": {},
"SkewedInfo": {
"SkewedColumnNames": [],
"SkewedColumnValues": [],
"SkewedColumnValueLocationMaps": {}
},
"StoredAsSubDirectories": false
},
"PartitionKeys": [],
"TableType": "EXTERNAL_TABLE",
"Parameters": {
"EXTERNAL": "TRUE",
"skip.header.line.count": "2",
"transient_lastDdlTime": "1652446944"
},
"CreatedBy": "arn:aws:iam::118146679784:user/abhishek-pandey",
"IsRegisteredWithLakeFormation": false,
"CatalogId": "118146679784"
},
{
"Name": "map_table",
"DatabaseName": "default",
"Owner": "hadoop",
"CreateTime": "2022-05-13 17:02:17",
"UpdateTime": "2022-05-13 17:02:17",
"LastAccessTime": "2022-05-13 17:02:17",
"Retention": 0,
"StorageDescriptor": {
"Columns": [
{
"Name": "c1",
"Type": "map<string,int>"
}
],
"Location": "s3://athena-postgres/map-test",
"InputFormat": "org.apache.hadoop.mapred.TextInputFormat",
"OutputFormat": "org.apache.hadoop.hive.ql.io.HiveIgnoreKeyTextOutputFormat",
"Compressed": false,
"NumberOfBuckets": -1,
"SerdeInfo": {
"SerializationLibrary": "org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe",
"Parameters": {
"serialization.format": "1"
}
},
"BucketColumns": [],
"SortColumns": [],
"Parameters": {},
"SkewedInfo": {
"SkewedColumnNames": [],
"SkewedColumnValues": [],
"SkewedColumnValueLocationMaps": {}
},
"StoredAsSubDirectories": false
},
"PartitionKeys": [],
"TableType": "EXTERNAL_TABLE",
"Parameters": {
"EXTERNAL": "TRUE",
"transient_lastDdlTime": "1653641000"
},
"CreatedBy": "arn:aws:iam::118146679784:user/pere-brull",
"IsRegisteredWithLakeFormation": false,
"CatalogId": "118146679784"
}
],
"ResponseMetadata": {
"RequestId": "c0524fd8-cba1-44c3-915d-fbeccc9b3334",
"HTTPStatusCode": 200,
"HTTPHeaders": {
"date": "Tue, 26 Jul 2022 17:55:29 GMT",
"content-type": "application/x-amz-json-1.1",
"content-length": "5247",
"connection": "keep-alive",
"x-amzn-requestid": "c0524fd8-cba1-44c3-915d-fbeccc9b3334"
},
"RetryAttempts": 0
}
}
}