Create an external table

This topic describes how to create an external table by using SQL statements. It also describes the prerequisites, overview, and considerations for creating an external table, and provides examples.

Overview

An external table is a logical table object. Its data is stored in an external storage system instead of the database.

For more information about external tables, see Overview.

Prerequisites

Before you create an external table, make sure that:

• You have deployed an OceanBase cluster and created an Oracle tenant. For more information about how to deploy an OceanBase cluster, see Deployment overview.

• You have connected to an Oracle tenant of OceanBase Database. For more information about how to connect to the database, see Overview of connection methods.

• The current user has the CREATE TABLE privilege. For more information about how to view the privileges of the current user, see View user privileges. If you do not have this privilege, contact the administrator to grant it to you. For more information about how to grant privileges to users, see Grant privileges directly.

Considerations

• An external table can only be queried, and DML operations are not supported.

• When you query an external table, if the external file accessed by the table is deleted, the system does not return an error, but instead returns empty rows.

• When the external storage system that manages the file accessed by the external table is unavailable, an error is returned when you query the external table.

• Data in an external table is stored in an external data source. Therefore, factors such as network latency and file system performance affect the query performance. When you create an external table, select an appropriate data source and optimization strategy to improve query efficiency.

Create an external table by using the command line

You can execute the CREATE EXTERNAL TABLE statement to create an external table.

Define External Name

When you create an external table, you must first define a name for it. To avoid confusion and ambiguity, we recommend that you use specific naming rules or prefixes to distinguish external tables from regular tables. For example, you can add a suffix such as _csv to the name of an external table.

Here is an example:

Create an external table that stores student information. You can name it students_csv.

CREATE EXTERNAL TABLE students_csv external_options

Define columns

You cannot define constraints such as DEFAULT, NOT NULL, UNIQUE, CHECK, PRIMARY KEY, and FOREIGN KEY for columns of an external table.

The column types supported for an external table are the same as those for a regular table. For more information about the data types supported in the Oracle mode of OceanBase Database, see Overview of built-in data types.

Define LOCATION

The LOCATION option specifies the path where the files of the external table are stored. Generally, the data files of an external table are stored in a dedicated directory, which can contain subdirectories. When you create an external table, the system automatically collects all files in the specified directory.

OceanBase Database supports the following two path formats:

• Local path: LOCATION = '[file://] local_file_path'

  Notice

  For scenarios that use local paths, you must set the system variable secure_file_priv to specify an accessible path. For more information, see secure_file_priv.

• Remote path: LOCATION = '{oss|cos}://$ACCESS_ID:$ACCESS_KEY@$HOST/remote_file_path'

  $ACCESS_ID, $ACCESS_KEY, and $HOST are required for accessing Alibaba Cloud OSS or Tencent Cloud COS. These sensitive access information is stored in the system tables of the database in an encrypted form.

Define FORMAT

• FORMAT = ( TYPE = 'CSV'... ) specifies the CSV format for external files. The parameters are as follows:
  • TYPE: the type of the external file.
  • LINE_DELIMITER: the line delimiter for the CSV file. The default value is LINE_DELIMITER='\n'.
  • FIELD_DELIMITER: the field delimiter for the CSV file. The default value is FIELD_DELIMITER='\t'.
  • ESCAPE: the escape character for the CSV file, which can be only 1 byte in length. The default value is ESCAPE ='\'.
  • FIELD_OPTIONALLY_ENCLOSED_BY: the characters that enclose the field values in the CSV file. The default value is an empty string.
  • ENCODING: the character set encoding format of the file. For more information about the character sets supported in MySQL mode, see Character set and collation. If this parameter is not specified, the default value UTF8MB4 is used.
  • NULL_IF: the strings to be treated as NULL values.
  • SKIP_HEADER: the number of lines to be skipped at the beginning of the file.
  • SKIP_BLANK_LINES: specifies whether to skip blank lines. The default value is FALSE, which specifies not to skip blank lines.
  • TRIM_SPACE: specifies whether to remove leading and trailing spaces from fields in the file. The default value is FALSE, which specifies not to remove leading and trailing spaces from fields in the file.
  • EMPTY_FIELD_AS_NULL: specifies whether to treat empty strings as NULL values. The default value is FALSE, which specifies not to treat empty strings as NULL values.
• FORMAT = ( TYPE = 'PARQUET'... ) specifies the Parquet format for external files.

(Optional) Define PATTERN

You can specify the PATTERN option to specify a regular pattern string to filter files in the LOCATION directory. For each file under the LOCATION directory, the external table accesses the file if the file path matches the pattern string, and skips the file otherwise. If you do not specify this parameter, the external table accesses all files under the LOCATION directory by default. The external table stores the list of files that match the LOCATION specified path and the PATTERN pattern in the system table of the database. During a scan, the external table accesses external files based on this list.

(Optional) Define partitions of an external table

Partitions are automatically defined

The external table calculates and adds partitions based on the partitioning key expression. You can specify the value or range of the partitioning key in a query. In this case, the partitions are pruned, and the external table reads files only in the specified partitions.

Partitions are manually defined

If you want to manually add and drop partitions instead of letting the external table automatically manage partitions, specify the PARTITION_TYPE = USER_SPECIFIED option.

Examples

The following example describes how to create an external table in the Oracle mode of OceanBase Database, given that the external file is located both locally and in the Oracle mode. The steps are as follows:

1. Create an external file.

   Execute the following command to create a file named test_tbl1.csv in the /home/admin/external_csv directory on the server that you will log in to.

   [admin@xxx /home/admin/external_csv]# vi test_tbl1.csv
   

   The content of the file is as follows:

   1,'Emma'
   2,'William'
   3,'Olivia'
   

2. Set the path of the imported file.

   Notice

   For security reasons, you can only connect to the database through a local socket to execute the SQL statement that changes the global variable secure_file_priv. For more information, see secure_file_priv.

   1. Execute the following command to log in to the server where the OBServer node resides.

      ssh admin@10.10.10.1
      

   2. Execute the following command to connect to the oracle001 tenant through a local Unix socket.

      obclient -S /home/admin/oceanbase/run/sql.sock -usys@oracle001 -p******
      

   3. Execute the following SQL command to set the import path to /home/admin/external_csv.

      SET GLOBAL secure_file_priv = "/home/admin/external_csv";
      

3. Reconnect to the oracle001 tenant.

   Here is an example:

   obclient -h10.10.10.1 -P2881 -usys@oracle001 -p****** -A
   

4. Execute the following SQL command to create an external table named test_tbl1_csv.

   CREATE EXTERNAL TABLE test_tbl1_csv ( 
       id INT, 
       name VARCHAR(50)
       )
       LOCATION = '/home/admin/external_csv'
       FORMAT = (
         TYPE = 'CSV'
         FIELD_DELIMITER = ','
         FIELD_OPTIONALLY_ENCLOSED_BY =''''
         )
       PATTERN = 'test_tbl1.csv';
   

5. Execute the following SQL command to view the data of the external table named test_tbl1_csv.

   SELECT * FROM test_tbl1_csv;
   

   The return result is as follows:

   +------+---------+
   | ID   | NAME    |
   +------+---------+
   |    1 | Emma    |
   |    2 | William |
   |    3 | Olivia  |
   +------+---------+
   3 rows in set
   

References

For more information about how to view and modify the schema of an external file, see Manage external files.