4.3.8.4 Use PAR to Load Data

Before You Begin

Review the requirements and recommendations to Access Object Storage with Pre-Authenticated Requests.
Review How To Load Data From Object Storage Using Auto Parallel Load.

Types of PARs

You can create the following types of PARs:

Bucket or prefix PAR: This type of PAR ends with /o/. You can load multiple files with this PAR. You can also use name, prefix, or pattern parameters to specify files in the bucket or Object Storage folder.
Object PAR: This type of PAR ends with /o/object_name. This PAR specifies individual files. You cannot use name, prefix, or pattern parameters with this PAR.

PAR Recommendations

When creating PARs consider the following recommendations:

Only use read-only PARs.
Set a short expiration date for the PAR URL that matches the data loading plan.
Do not make a PAR URL publicly accessible.
If the target defines a bucket or uses a prefix or pattern:
- Use Enable Object Listing when creating the PAR in the Oracle Cloud Infrastructure (OCI) console.
- When creating the PAR from the command line, include the --access-type AnyObjectRead parameter.

Note

Use a resource principal for access to more sensitive data in Object Storage as it is more secure. See Access Object Storage with Resource Principals.

PAR Parameters

To set up PARs, you configure the following parameters:

par: Provide the PAR URL.
name: Use this to specify a file for an object or bucket PAR.
pattern: Use this to set a regular expression that defines a set of Object Storage files. The pattern follows the modified Modified ECMAScript regular expression grammar.
prefix: Use this to define a set of Object Storage folders and files.

PAR Examples

The following examples use these parameters and commands to set up the loading of data with PARs. Replace the values in the examples with your own.

SET @input_list creates a session variable that stores all the parameters for the loading of the file.
db_name identifies the database name to store the table. MySQL HeatWave automatically creates the database if it does not exist.
table_name sets the table name to store the data. MySQL HeatWave automatically creates the table if it does not exist.
engine_attribute defines the parameters of the external file.
dialect defines the format options of the file.
file defines the file or files to load.
CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL); runs Auto Parallel Load to load the external tables from the parameters set in @input_list and with no additional options.

Note

If you are on MySQL 9.1.2 and earlier, you need to update dialect with the field delimiter and record delimiter parameters. As of MySQL 9.2.0, MySQL HeatWave Lakehouse can automatically detect these values. See Lakehouse External Table Syntax to learn more.

To review all syntax options for loading external files, see Auto Parallel Load Syntax and Lakehouse External Table Syntax.

Load a Single File

To load a single file with a pre-authenticated request, you can create a pre-authenticated request for that individual file or specify the file in the command.

The following example loads a single file by using a PAR for that file.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/data_file_1.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

The following example loads a single file by using a PAR for a bucket and specifying the file in the name parameter. The file is in the data_files Object Storage folder.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/", "name": "data_files/data_file_1.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

Load Multiple Files into One External Table

If you want to load multiple files into one external table, you can do one of the following:

Create a folder in the Object Storage bucket and upload the required files into that folder. See Managing Folders in an Object Storage Bucket in Oracle Cloud Infrastructure Documentation.
Create a PAR for a bucket or folder, and then specify each file to load with the name parameter.
Create a PAR for each file to load, and then specify the files as separate items in the ENGINE_ATTRIBUTE parameter.

To load multiple external files into one external table, you can create a folder in the Object Storage bucket and upload the required files into that folder. See Managing Folders in an Object Storage Bucket in Oracle Cloud Infrastructure Documentation.

If you create a folder with the files to load, create a PAR for that folder and do the following:

For the Pre-Authenticated Request Target, select Objects with prefix.
Use Enable Object Listing.

The following example loads all the files uploaded to the folder in the Object Storage bucket into one table.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                           "engine_attribute": {"dialect": {"format": "csv"},
                                           "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

The following example uses one PAR for a bucket and specifies two files to load with the name parameter.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/", "name": "data_files/data_file_1.csv"},
                                                  {"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/", "name": "data_files/data_file_2.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

The following example loads two separate PARs to load two files.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/data_files/data_file_1.csv"},
                                                  {"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/data_files/data_file_2.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

Load Multiple Files into Multiple Tables

To load multiple external files into multiple tables with one command, create a pre-authenticated for each required file and create the necessary number of tables items.

The following example loads two files into two tables.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {
                                         "dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/data_files/data_file_1.csv"}]}},
                                        {"table_name": "table_2",
                                         "engine_attribute": {
                                         "dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenant_1/b/bucket_1/o/data_files/data_file_2.csv"}]
}}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

Load Files with a Prefix

You can add a prefix parameter to a PAR to load specific files. To do this, create a PAR for the folder that stores the files to load.

The following example loads all files in the data_files folder that begin with data_file_. For example, MySQL HeatWave loads the files data_file_1, data_file_2, and data_file_3 into table_1.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/", 
                                                          "prefix": "data_files/data_file_"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

Load Files with a Pattern

You can add a pattern parameter to a PAR to use regular expression to load specific files.

The regular expression syntax requires certain characters to have an escape character.

The escape character is the backslash character, and it is a reserved character in both JSON and MySQL. Therefore, it is necessary to escape the backslash character twice, and specify \\ for both JSON and MySQL.

However, the regular expression escape sequence depends upon the NO_BACKSLASH_ESCAPES SQL mode:

Use \\. to escape a period if NO_BACKSLASH_ESCAPES is enabled.
Use \\\\. to escape a period if NO_BACKSLASH_ESCAPES is not enabled. The following examples use this sequence because it is the default mode.

See the following to learn more:

The following example loads all files in the data_files folder that have a numerical suffix of one or more digits. For example, MySQL HeatWave loads the files data_file_1.csv, data_file_2.csv, and data_file_3.csv into table_1.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/", 
                                                          "pattern": "data_files/data_file_\\\\d+\\\\.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

The following examples load all files in the bank_data folder that have an alphabetical suffix of one or more lowercase characters. For example, MySQL HeatWave loads the files data_file_a.csv, data_file_b.csv, and data_file_c.csv.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/", 
                                                          "pattern": "data_files/data_file_[a-z]+\\\\.csv"}]}
}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

To update the previous example to include uppercase and lowercase characters, replace [a-z] with [A-Za-z].

The following example loads all files in the bank_data folder that have a numerical suffix that end in 0 with one preceding digit. For example, MySQL HeatWave loads the files data_file_10.csv, data_file_20.csv, and data_file_30.csv.

mysql> SET @input_list = '[{"db_name": "lakehouse_db",
                             "tables": [{"table_name": "table_1",
                                         "engine_attribute": {"dialect": {"format": "csv"},
                                         "file": [{"par": "https://objectstorage.us-ashburn-1.oraclecloud.com/p/.../n/tenancy_1/b/bucket_1/o/", 
                                                          "pattern": "data_files/data_file_\\\\d0\\\\.csv"}]
}}]}]';
mysql> CALL sys.HEATWAVE_LOAD(CAST(@input_list AS JSON), NULL);

To update the previous example to load files with one or more digits preceding the 0, update \\\\d0\\\\ with \\\\d+0\\\\.

What's Next

After successfully loading data and creating external tables, learn how to do the following: