Command Reference

Data Wrapper

CREATE SERVER

Create new server objects.

Syntax

CREATE SERVER [IF NOT EXISTS] <server_name>
    FOREIGN DATA WRAPPER <foreign_data_wrapper_name>
     WITH ( <option> = <value> [, ... ] )

Description

A foreign server object represents a named external datasource, along with configuration information required to access the datasource. Server objects are referenced during foreign table creation in order to indicate the backing datasource for the created table.

Parameters

server_name	The name of the foreign server to be created.
foreign_data_wrapper_name	The foreign data wrapper to use. Available data wrappers are: delimited_file, parquet_file, regex_parsed_file, odbc (beta), raster_file (beta).

Foreign Server Options for File Data Wrappers (delimited_file, parquet_file, and regex_parsed_file)

STORAGE_TYPE	Available types are LOCAL_FILE, AWS_S3.
BASE_PATH	Base file path for the data files or directory.
S3_BUCKET	S3 bucket containing files. Only applies to AWS_S3 storage type
AWS_REGION	AWS region where files/objects are stored. Only applies to AWS_S3 storage type.
S3_ENDPOINT	Custom S3 endpoint, which allows data access from S3-compatible services, such as Google Cloud Storage. Similar in function to the COPY FROM command.

Foreign Server Options for ODBC Data Wrapper

ODBC HeavyConnect is currently in beta.

DATA_SOURCE_NAME	Data source name (DSN) configured in the odbc.ini file. Only one of data_source_name or connection_string is required.
CONNECTION_STRING	A set of semicolon-separated key=value pairs, which define the connection parameters for an RDMS. For example, Driver=DriverName;Database=DatabaseName;Servername=HostName;Port=1234. Only one of data_source_name or connection_string is required.

Examples

Create a local server that uses the delimited_file data wrapper and the file path /home/my_user/data/csv/:

CREATE SERVER my_csv_server FOREIGN DATA WRAPPER delimited_file WITH (
  storage_type = 'LOCAL_FILE', 
  base_path = '/home/my_user/data/csv/'
);

Create an AWS Server that uses the delimited_file data wrapper with an AWS S3 bucket "fsi-s3-bucket" in the us-west-1 AWS region:

CREATE SERVER my_aws_server FOREIGN DATA WRAPPER delimited_file WITH (
  storage_type = 'AWS_S3', 
  s3_bucket = 'fsi-s3-bucket', 
  aws_region = 'us-west-1'
);

Create an ODBC server using a data source name (DSN):

CREATE SERVER my_postgres_server FOREIGN DATA WRAPPER odbc WITH (
  data_source_name = 'postgres_db_1'
);

Create an ODBC server using a connection string:

CREATE SERVER my_postgres_server FOREIGN DATA WRAPPER odbc WITH (
  connection_string = 'Driver=PostgreSQL;Database=my_postgres_db;Servername=my_postgres.example.com;Port=1234'
);

To prevent collisions with available HeavyDB default servers, server names with default*, system*, and internal* (case insensitive) prefixes are not allowed.

ALTER SERVER

Update server objects.

Syntax

ALTER SERVER <server_name>
    [ SET (<option> = <value> [, ... ] ) ]
ALTER SERVER <server_name> OWNER TO <new_owner>
ALTER SERVER <server_name> RENAME TO <new_server_name>
ALTER SERVER <server_name> SET FOREIGN DATA WRAPPER <foreign_data_wrapper_name>

Description

Allow users to update properties of existing server objects.

Parameters

Parameter	Usage
Parameter	Usage
server_name	The name of the foreign server to be updated.
new_server_name	The new name that should be assigned to the foreign server.
option	Clause specific to the option to be updated.
new_owner	Username of the new foreign server owner.
foreign_data_wrapper_name	The name of the foreign data wrapper that should be associated with the server object.

Example: Update server property

Update the my_csv_server object's base path “/home/my_user/data/csv/” to “/home/my_user/data/new_csv/”:

ALTER SERVER my_csv_server WITH (base_path = '/home/my_user/data/new-csv/');

Example: Change server name

Change the my_csv_server object's name to my_new_csv_server:

ALTER SERVER my_csv_server RENAME TO my_new_csv_server;

DROP SERVER

Delete an existing server object

Syntax

DROP SERVER [ IF EXISTS ] <server_name>

Parameters

Parameter	Usage
server_name	The name of the foreign server to be deleted.

Example

Delete the my_csv_server server object:

DROP SERVER my_csv_server;

SHOW SERVERS

View information about existing server objects.

Syntax

SHOW SERVERS 
    [WHERE <metadata_name> { = | LIKE } <metadata_value>]

Column Results

Result	Definition
server_name	Server name used in the CREATE/ALTER SERVER command.
owner	User who created the server object or who has been assigned as the owner of the server object via an ALTER SERVER command.
data_wrapper	Name of data wrapper associated with server object.
options	JSON string representation of options used in CREATE/ALTER SERVER command.
created_at	Timestamp when the server object was created.

Example - View all server objects in the current database

SHOW SERVERS;
Result:
server_name       | owner | data_wrapper   | options                                                                                | created_at
------------------+-------+----------------+----------------------------------------------------------------------------------------+----------------------
my_csv_server     | joe   | delimited_file | {"storage_type": "LOCAL_FILE", "base_path": "/home/my_user/data/csv/"}                 | 2019-11-15T23:54:35Z
my_parquet_server | admin | parquet_file   | {"storage_type": "AWS_S3", "s3_bucket": "my_parquet_files", "aws_region": "us-east-1"} | 2019-11-15T23:54:35Z

Example - View all server objects with the delimited_file data wrapper type

SHOW SERVERS WHERE data_wrapper = 'delimited_file';
Result:
server_name       | owner | data_wrapper   | options                                                                                | created_at
------------------+-------+----------------+----------------------------------------------------------------------------------------+----------------------
my_csv_server     | joe   | delimited_file | {"storage_type": "LOCAL_FILE", "base_path": "/home/my_user/data/csv/"}                 | 2019-11-15T23:54:35Z

Example - View all server objects with the name containing “csv”

SHOW SERVERS WHERE server_name LIKE '%csv%';
Result:
server_name       | owner | data_wrapper   | options                                                                                | created_at
------------------+-------+----------------+----------------------------------------------------------------------------------------+----------------------
my_csv_server     | joe   | delimited_file | {"storage_type": "LOCAL_FILE", "base_path": "/home/my_user/data/csv/"}                 | 2019-11-15T23:54:35Z

CREATE FOREIGN TABLE

Creates the foreign table.

Syntax

CREATE FOREIGN TABLE [ IF NOT EXISTS ] <table_name> (
  { <column_name> <data_type> [NOT NULL] [ENCODING <encoding_spec>] }
  [, ... ]
)
SERVER <server_name>
[ WITH ( <option> = <value> [, ... ] ) ]

Description

A foreign table acts as a proxy to an external data source for which access details are provided by a server object. A foreign table provides an interface for querying data from an external datasource in a similar way to a local table.

Parameters

Parameter	Usage
table_name	The name of the foreign table to be created.
column_name	The name of the column to be created.
data_type	Data type of the column to be created.
NOT NULL	Specify that NULL Is not allowed for this column type.
ENCODING	Specify the encoding of the data type.
SERVER {custom_server_name	default_local_delimited

Foreign Tables Refresh Options

Parameter	Usage
ALLOW_FILE_ROLL_OFF	Allow the oldest files (based on file path sort order) in a directory of files to be rolled off when a refresh occurs. Only applies when REFRESH_UPDATE_TYPE is set to APPEND.
REFRESH_UPDATE_TYPE={APPEND | ALL}	Specifies what portion of the data is refreshed. APPEND specifies that new data is added to existing. ALL specifies that updated data can exist anywhere within the data set.
REFRESH_TIMING_TYPE={MANUAL | SCHEDULED}	Identifies if data refresh activities are manual or automated.
REFRESH_START_DATE_TIME 'YYYY-MM-DDTHH:MM:SSZ'	Identifies the time to start data refreshes. Used with REFRESH_TIMING_TYPE=’SCHEDULED’. Time is specified in UTC time in ISO 8601 format and cannot be in the past.
REFRESH_INTERVAL interval_count{D | M}	Used with REFRESH_TIMING_TYPE=’SCHEDULED’. Available periods: D(ays), M(onth). To specify 5 days, use ‘5D’.

Foreign Table Options for File Data Wrappers (delimited_file, parquet_file, and regex_parsed_file)

Parameter	Usage
FILE_PATH	Name or path to the file that serves as the datasource for the foreign table. This file path is appended to the BASE_PATH specified in the foreign server.
REGEX_PATH_FILTER	See Globbing, Filtering, and Sorting Parquet and CSV Files
FILE_SORT_ORDER_BY	See Globbing, Filtering, and Sorting Parquet and CSV Files
FILE_SORT_REGEX	See Globbing, Filtering, and Sorting Parquet and CSV Files

Additional Foreign Table Options for the delimited_file Data Wrapper

Parameter

Usage

S3_ACCESS_TYPE

Used to identify the S3 CSV access type. Supported values are S3_Direct and S3_Select. This value is not required if using Parquet data formats. The S3_Direct access type uses HeavyConnects's standard CSV data wrapper and HeavyDB supported datatypes. Choosing the S3_Select access type uses the S3 Select capability which can improve performance. S3 Select does not support GEO and ARRAY types and TEXT types with embedded row delimiters such as newlines. If your data contains these types, use the S3_Direct access type. When using S3_Select with public buckets, the CREATE USER MAPPING FOR PUBLIC SERVER is required to set credentials but is not required when using public buckets and S3_Direct.

All delimited file parser options mentioned in https://docs.heavy.ai/loading-and-exporting-data/command-line/load-data#csv-tsv-import

Additional Foreign Table Options for the regex_parsed_file Data Wrapper

Parameter	Usage
LINE_REGEX	Regular expression that matches an line in the given file. Each capturing group in the regular expression should correspond to a column in the foreign table definition. Capturing groups are expected to match columns in number and order specified in the foreign table definition.
LINE_START_REGEX	Regular expression pattern that should match the start of a new entry in the given file. This option is useful for use cases where entries span multiple lines e.g. when a message contains line breaks. If this option is specified, lines with beginnings that do not match this regular expression are appended to the previous line. This parameter is optional.

Foreign Table Options for ODBC Data Wrapper

ODBC HeavyConnect is currently in beta.

Parameter	Usage
SQL_SELECT	SQL statement used to issue a query to an RDMS database.
SQL_ORDER_BY	Comma separated list of column names that provide a unique ordering for the result set returned by the SQL statement that is set in the SQL_SELECT option.

Example - Using the default local file server and default property

Create a foreign table that accesses a local CSV file at /home/my_user/data/csv/my_file.csv using this default server object.

CREATE FOREIGN TABLE my_csv_table (first_name TEXT, last_name TEXT ENCODING NONE) 
  SERVER default_local_delimited 
  WITH (file_path = '/home/my_user/data/csv/my_file.csv');

Example 2 - Using a custom local file server

Create a foreign table that accesses a local CSV file at path /home/my_user/data/csv/my_other_file.csv using the above my_csv_server object.

CREATE FOREIGN TABLE my_csv_table (first_name TEXT, last_name TEXT ENCODING NONE) 
  SERVER my_csv_server 
  WITH (file_path = 'my_other_file.csv');

The default "default_local_delimited" server object can also be used to accomplish the above. However, if you use a custom server object, you do not have to specify the full path when creating foreign tables.

Example 3 - Using the custom server and duration property

Create a foreign table that accesses a remote Parquet file named my_parquet_file, referenced by my_parquet_server object, for which content could change every 2 days.

CREATE FOREIGN TABLE my_parquet_table (first_name TEXT, last_name TEXT ENCODING NONE) 
  SERVER my_parquet_server 
  WITH (file_path = 'my_parquet_file', refresh_interval = '2D');

Example 4 - Using S3, custom S3 server, and the S3_Select option

CREATE FOREIGN TABLE my_s3_csv_table (first_name TEXT, last_name TEXT ENCODING NONE) 
  SERVER my_aws_server 
  WITH (s3_access_type = 's3_select', 
        file_path = 'my_csv_file', 
        refresh_update_type = 'append', 
        refresh_interval = '2D');

Use the CREATE USER MAPPING command to set access credentials when using S3. S3_Select requires set credentials even when using public buckets.

Additional Foreign Table Options for the default_local_raster Data Wrapper

Raster HeavyConnect is currently in beta.

Parameter	Usage
RASTER_TILE_WIDTH	Specifies the file/block width by which the raster data should be grouped. If none is specified, this will default to the block size provided by the file.
RASTER_TILE_HEIGHT	Specifies the file/block height by which the raster data should be grouped. If none is specified, this will default to the block size provided by the file.
RASTER_FILTER_BANDS	Equivalent to the raster_import_bands option used during raster import. Allows specification of one or more band names to selectively import; useful in the context of large raster files where not all the bands are relevant. Bands are imported in the order provided, regardless of order in the file. You can rename bands using <bandname>=<newname>[,<bandname>=<newname,...>] Names must be those discovered by the detection process, including any suffixes for de-duplication.
RASTER_POINT_TRANSFORM	Equivalent to the raster_point_transform option used during raster import. Specifies the processing for floating-point coordinate values: auto - Transform based on raster file type (world for geo, none for non-geo). none - No affine or world-space conversion. Values will be equivalent to the integer pixel coordinates. file - File-space affine transform only. Values will be in the file's coordinate system, if any (e.g. geospatial). world - World-space geospatial transform. Values will be projected to WGS84 lon/lat (if the file has a geospatial SRID). Note that this option requires correct column types to be specified when creating the table schema i.e. if raster_point_transform='none' then the point columns must be specified with integer types.

ALTER FOREIGN TABLE

Update properties of existing foreign tables.

Syntax

ALTER FOREIGN TABLE <table_name>
    RENAME TO <new_table_name>

ALTER FOREIGN TABLE <table_name>
    RENAME COLUMN <column_name> TO <new_column_name>    

ALTER FOREIGN TABLE <table_name>
    [ SET (<option> = <value> [, ... ] ) ]

The following table options can be altered:

• ALLOW_FILE_ROLL_OFF

• BUFFER_SIZE

• REFRESH_INTERVAL

• REFRESH_START_DATE_TIME

• REFRESH_TIMING_TYPE

• REFRESH_UPDATE_TYPE

• SQL_ORDER___BY

• SQL_SELECT

Description

Allow users to update properties of existing foreign tables.

Parameters

Parameter	Usage
table_name	The name of the foreign table to be updated.
new_table_name	The new name of the foreign table.
column_name	The column name to be updated.
new_column_name	The new name of the column.

Example - Change foreign table name

ALTER FOREIGN TABLE my_csv_table RENAME TO my_new_csv_table;

Example - Change foreign table column name

ALTER FOREIGN TABLE my_csv_table RENAME COLUMN first_name TO full_name; 

Example - Change the foreign table update type

ALTER FOREIGN TABLE my_csv_table SET (REFRESH_UPDATE_TYPE=‘APPEND’);

Example - Change the foreign table update schedule

ALTER FOREIGN TABLE my_csv_table 
  SET (REFRESH_TIMING_TYPE=‘SCHEDULED’, 
       REFRESH_START_DATE_TIME=‘2020-12-20 10:30’, 
       REFRESH_INTERVAL=‘1D’) ;

DROP FOREIGN TABLE

Delete a foreign table.

Syntax

DROP FOREIGN TABLE [ IF EXISTS ] <table_name>;

Description

Allows users to delete an existing foreign table.

Parameters

Parameters	Usage
table_name	The name of the foreign table to be deleted.

Example: Delete my_csv_table

DROP FOREIGN TABLE my_csv_table;

REFRESH FOREIGN TABLES

Refresh a foreign table.

Syntax

REFRESH FOREIGN TABLES <table_name>, <table_name> [ WITH ( <option> = <value> [, ... ] ) ];

Description

Allows users to do an on-demand refresh of one or more foreign tables.

Options

Parameter	Usage
EVICT	Indicate whether to evict all cached foreign table data while refreshing (TRUE) or not (FALSE).

Parameters

Parameters	Usage
table_name	The name of the foreign table(s) to be refreshed.

User Mappings

A user mapping is an object that contains access credentials to an external datasource/foreign server. User mappings provide secure handling of user credentials when access to restricted external datasources or foreign servers is required. User mapping DDL commands can only be executed by super users or owners of referenced foreign servers. The following DDL commands are supported.

CREATE USER MAPPING

Create new user mappings.

Syntax

CREATE USER MAPPING [IF NOT EXISTS] FOR PUBLIC SERVER <server_name>
     WITH ( <option> = <value> [, ... ] )

Description

Allows users to create new user mappings. See related AWS documentation for information about AWS credentials.

Parameters

Parameter	Usage
server_name	Name of the server the user mapping applies to.

S3 User Mapping Options

Parameter	Usage
s3_access_key	AWS access key id for accessing S3 objects (see related documentation for more details). This is provided as an option and only applies to AWS S3 based servers (i.e. delimited_file_,_ parquet_file_,_ or regex_parsed_file data wrappers with storage_type of AWS_S3).
s3_secret_key	AWS secret access key for accessing S3 objects (see related documentation for more details). This is provided as an option and only applies to AWS S3 based servers (i.e. delimited_file_,_ parquet_file_,_ or regex_parsed_file data wrappers with storage_type of AWS_S3).
s3_session_token	AWS session token generated for accessing S3 objects using temporary credentials; see AWS documentation for more information. This option applies only to AWS S3-based servers (for example, delimited_file_,_ parquet_file_,_ or regex_parsed_file data wrappers with storage type AWS_S3).

AWS credentials that are configured on the server, such as IAM roles, can be used in place of user mappings. Enable this option with the allow-s3-server-privileges server configuration.

ODBC User Mapping Options

ODBC HeavyConnect is currently in beta.

Parameter	Usage
username	Username on the RDMS. This option only applies to user mappings that reference an odbc foreign server, which uses data_source_name.
password	Password credential for the RDMS. This option only applies to user mappings that reference an odbc foreign server, which uses data_source_name.
credential_string	A set of semicolon separated key=value pairs, which define the access credential parameters for an RDMS. For example, Username=username;Password=password. This option only applies to user mappings that reference an odbc foreign server, which uses connection_string.

Examples

Set up credentials for users to be able to query a foreign table (see related documentation for more details) that gets data from a previously created AWS S3 backed my_parquet_server,

CREATE USER MAPPING FOR PUBLIC SERVER my_parquet_server 
  WITH (aws_access_key_id = '1234', aws_secret_access_key = 'abcd');

DELETE USER MAPPING

Delete existing user mappings.

Syntax

DROP USER MAPPING [IF NOT EXISTS] FOR PUBLIC SERVER <server_name>

Description

Allows users to delete existing user mappings.

Parameters

Parameter	Usage
server_name	Name of the server the user mapping applies to.

Examples

Delete user mapping that was setup for my_parquet_server.

DROP USER MAPPING FOR PUBLIC SERVER my_parquet_server;