メインコンテンツまでスキップ
バージョン: 3.12
Features and pricing

Get Started with ScalarDL TableStore

ScalarDL TableStore is a high-level abstraction on top of the low-level ledger abstraction. It offers an SQL interface instead of primitive CRUD interfaces like get and put, enabling you to build versatile, tamper-evident applications with the familiar data model and commands quickly and easily.

This getting started tutorial explains how to configure TableStore on your preferred database and manage tables and records in a tamper-evident manner.

What is ScalarDL TableStore?​

TableStore provides table-based data management through an SQL interface. You can create tables in a flexible schemaless manner, perform SQL operations like SELECT, INSERT, and UPDATE, and maintain complete audit trails of all data modifications.

Prerequisites​

警告

Since ScalarDL is built with JDK 8, contracts must be a JDK 8–compatible binary. If you use a version other than JDK 8, you must configure your build tool to build the JDK 8–compatible binary. There are several ways to specify binary compatibility, including using the --release 8 option for javac or setting Gradle or Maven configurations to use the JDK 8 toolchain. The following shows the configuration for Gradle:

java {
    toolchain {
        languageVersion.set(JavaLanguageVersion.of(8))
    }
}

For more details about the Gradle and Maven configurations, see Toolchains for JVM projects for Gradle and Guide to Using Toolchains for Maven.

Clone the ScalarDL samples repository​

Open Terminal, then clone the ScalarDL samples repository by running the following command:

git clone https://github.com/scalar-labs/scalardl-samples

Then, go to the directory that contains the sample configuration by running the following command:

cd scalardl-samples

Start up ScalarDL with your preferred database​

Select your database, and follow the instructions to deploy ScalarDL Ledger with it. For a list of databases that ScalarDL supports, see Databases.

Set up your license (Enterprise edition only)

If you're using the ScalarDL Enterprise edition, set up your license as follows. If you're using the Community edition, skip to the next section to start up ScalarDL.

See here to set up your license

  1. Enable the Docker image for the Enterprise edition in the mysql/docker-compose-ledger.yml file as follows:

    • Before changing the image (default configuration):

      services:
        scalardl-ledger:
          image: ghcr.io/scalar-labs/scalardl-ledger:${SCALARDL_VERSION}
          # image: ghcr.io/scalar-labs/scalardl-ledger-byol:${SCALARDL_VERSION}

    • After changing the image:

      services:
        scalardl-ledger:
          # image: ghcr.io/scalar-labs/scalardl-ledger:${SCALARDL_VERSION}
          image: ghcr.io/scalar-labs/scalardl-ledger-byol:${SCALARDL_VERSION}

  2. Set your license key for ScalarDL Ledger. In the mysql/ledger.properties file, replace <SET_YOUR_LICENSE_KEY> with your license key. For example:

    ##### PLEASE REPLACE THIS VALUE WITH YOUR LICENSE KEY (ENTERPRISE EDITION ONLY) #####
    scalar.dl.licensing.license_key={"organization_name":"XXXXXXXX","expiration_date_time":"YYYY-MM-DDTHH:mm:SS+TIMEZONE","product_name":"ScalarDL Ledger","product_version":N,"license_type":"trial","signature":"XXXXXXXX"}
    ##### PLEASE REPLACE THIS VALUE WITH YOUR LICENSE KEY (ENTERPRISE EDITION ONLY) #####

  3. To check the license, update the mysql/docker-compose-ledger.yml file as follows. If you're using a trial license, skip this step.

    • Before changing the certificate file path (default configuration):

      services:
        scalardl-ledger:
          volumes:
            - ./ledger.properties:/scalar/ledger/ledger.properties.tmpl
            - ../fixture/ledger-key.pem:/scalar/ledger-key.pem
            - ../fixture/trial-license-cert.pem:/scalar/license-cert.pem
            # If you have a commercial license key, you must use `commercial-license-cert.pem` instead of `trial-license-cert.pem`.
            # - ../fixture/commercial-license-cert.pem:/scalar/license-cert.pem

    • After changing the certificate file path:

      services:
        scalardl-ledger:
          volumes:
            - ./ledger.properties:/scalar/ledger/ledger.properties.tmpl
            - ../fixture/ledger-key.pem:/scalar/ledger-key.pem
            # - ../fixture/trial-license-cert.pem:/scalar/license-cert.pem
            # If you have a commercial license key, you must use `commercial-license-cert.pem` instead of `trial-license-cert.pem`.
            - ../fixture/commercial-license-cert.pem:/scalar/license-cert.pem

Start up ScalarDL

You can start ScalarDL Ledger by following the steps below:

  1. Run MySQL locally by running the following command:

    docker compose -f mysql/docker-compose-ledger.yml up -d mysql

  2. Load the database schema for ScalarDL Ledger by running the following command:

    docker compose -f mysql/docker-compose-ledger.yml up -d scalardl-ledger-schema-loader

  3. Run ScalarDL Ledger and its dependent components by running the following command:

    docker compose -f mysql/docker-compose-ledger.yml up -d

Download the Client SDK​

Next, you'll use the TableStore client tools. Specify a version that is the same as the deployed ScalarDL version and is used for downloading the tools by running the following command:

VERSION=$(grep SCALARDL_VERSION .env | awk -F= '{print $2}')

Then, download the tools by running the following command:

curl -OL https://github.com/scalar-labs/scalardl/releases/download/v$VERSION/scalardl-tablestore-java-client-sdk-$VERSION.zip
unzip scalardl-tablestore-java-client-sdk-$VERSION.zip
mv scalardl-tablestore-java-client-sdk-$VERSION client

Configure the client properties​

Before interacting with TableStore, you need to configure the client. To create a configuration file with the minimum required properties for the client, run the following command:

cat << 'EOF' > client.properties
# A host name for ScalarDL Ledger.
scalar.dl.client.server.host=localhost

# An ID for the certificate holder. This must be configured for each private key and must be unique in the system.
scalar.dl.client.cert_holder_id=foo

# A path to the certificate file.
scalar.dl.client.cert_path=./fixture/client.pem

# A path to the private key file. 
scalar.dl.client.private_key_path=./fixture/client-key.pem
EOF

You can use localhost for the ScalarDL Ledger host name in this tutorial. For the private key and certificate, you can use the ones provided in the fixture directory of the scalardl-samples repository (client-key.pem and client.pem, respectively). For the certificate holder, any unique ID can be specified.

警告

Do not use the sample private key and certificate in production environments. For details about getting your own certificate, see How to Get a Certificate.

Bootstrap​

Next, you can bootstrap TableStore by running the following command:

client/bin/scalardl-tablestore bootstrap --properties client.properties

The bootstrap command internally registers identity information (a certificate or secret) and predefined contracts necessary to use TableStore.

Interact with TableStore​

Now you can execute SQL statements with TableStore. In this section, you'll try the following functionalities through two sample tables (employee and department) that can be joined through the department IDs of employees:

Create and show tables​

You can create the sample table by running the following commands:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "CREATE TABLE employee (id STRING PRIMARY KEY, department STRING)"
client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "CREATE TABLE department (id STRING PRIMARY KEY)"

When creating a table, you need to specify the name and the primary key. You can create secondary indexes by specifying additional columns. Because ScalarDL TableStore treats a JSON object as a record in tables, you don't have to specify a strict schema when creating a table.

You can show the created tables by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT * FROM information_schema.tables"

You should get a result like the following:

Result:
[ {
  "name" : "employee",
  "key" : "id",
  "type" : "string",
  "indexes" : [ {
    "key" : "department",
    "type" : "string"
  } ]
}, {
  "name" : "department",
  "key" : "id",
  "type" : "string",
  "indexes" : [ ]
} ]

Insert records​

Next, insert several employee records by running the following commands:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "INSERT INTO employee VALUES {'id': '1001', 'name': 'Alice', 'department': 'sales', 'salary': 654.3}"
client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "INSERT INTO employee VALUES {'id': '1002', 'name': 'Bob', 'department': 'sales', 'salary': 543.2}"
client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "INSERT INTO employee VALUES {'id': '1003', 'name': 'Carol', 'department': 'engineering', 'salary': 654.3}"

Insert the corresponding department records as well by running the following commands:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "INSERT INTO department VALUES {'id': 'sales', 'location': 'Shinjuku', 'phone': '000-1234'}"
client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "INSERT INTO department VALUES {'id': 'engineering', 'location': 'Shibuya', 'phone': '000-4321'}"

Select records​

Then, check the inserted records. You need to specify at least a primary key or index key to select records. For example, you can get an employee record by specifying the primary key by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT id, name, department FROM employee WHERE id = '1001'"

You can optionally project the columns by specifying top-level fields in the JSON record object. You should get a result like the following:

Result:
[ {
  "id" : "1001",
  "name" : "Alice",
  "department" : "sales"
} ]

You can also specify an index key to select records by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT id, name, department FROM employee WHERE department = 'sales'"

You should get a result like the following:

Result:
[ {
  "id" : "1001",
  "name" : "Alice",
  "department" : "sales"
}, {
  "id" : "1002",
  "name" : "Bob",
  "department" : "sales"
} ]

If you want to filter records, specify additional conditions by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT id, name, department FROM employee WHERE department = 'sales' AND salary < 600"

You should get a result like the following:

Result:
[ {
  "id" : "1002",
  "name" : "Bob",
  "department" : "sales",
  "salary" : 543.2
} ]

You can also join the two tables by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT * FROM employee JOIN department ON employee.department = department.id WHERE employee.department = 'engineering'"

You should get a result like the following:

Result:
[ {
  "employee.id" : "1003",
  "employee.name" : "Carol",
  "employee.department" : "engineering",
  "employee.salary" : 654.3,
  "department.id" : "engineering",
  "department.location" : "Shibuya",
  "department.phone" : "000-4321"
} ]

Update records​

You can update the employee records by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "UPDATE employee SET salary = 754.3 WHERE department = 'engineering'"

Make sure to specify at least a primary key or an index key to update the records, in the same way as using the SELECT statement.

Get record histories​

You can get the update history of a record by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT history() FROM employee WHERE id = '1003'"

You should get a result like the following:

Result:
[ {
  "age" : 1,
  "values" : {
    "id" : "1003",
    "name" : "Carol",
    "department" : "engineering",
    "salary" : 754.3
  }
}, {
  "age" : 0,
  "values" : {
    "id" : "1003",
    "name" : "Carol",
    "department" : "engineering",
    "salary" : 654.3
  }
} ]

If you want to limit the number of versions (ages), specify the LIMIT clause by running the following command:

client/bin/scalardl-tablestore execute-statement --properties client.properties \
--statement "SELECT history() FROM employee WHERE id = '1003' LIMIT 1"

You should get the specified number of the latest records like the following:

Result:
[ {
  "age" : 1,
  "values" : {
    "id" : "1003",
    "name" : "Carol",
    "department" : "engineering",
    "salary" : 754.3
  }
} ]

Validate data managed by TableStore​

In ScalarDL, you occasionally need to validate your data to make sure all the data is in a valid state. You can use the validate-ledger command to validate data managed by TableStore.

You can validate the table schema by running the following command:

client/bin/scalardl-tablestore validate-ledger --properties client.properties \
--table-name employee

You should get a result like the following:

{
  "status_code" : "OK",
  "Ledger" : {
    "id" : "tbl_employee",
    "age" : 0,
    "nonce" : "26af1229-1c1f-4b89-86e2-ec011da3b313",
    "hash" : "ZA9yFzjIg1qeHAd7Sub8uFvt2JrTb6XSzGUktPEITr0=",
    "signature" : "MEUCIAh4Xj93J/jldqbQor7AVM4ii9+suxQrZlCFnKWWDIo0AiEAiM6Yi6GO4bQ2VZg2GnqKmOFPEANrTU4g7pjBMcaX6TQ="
  },
  "Auditor" : null
}

You can validate the record by running the following command:

client/bin/scalardl-tablestore validate-ledger --properties client.properties \
--table-name employee --primary-key-column-name id --column-value '"1001"'
注記

The --column-value option expects a JSON value; thus, you need to put double quotes for a string value.

You should get a result like the following:

{
  "status_code" : "OK",
  "Ledger" : {
    "id" : "rec_employee_id_1001",
    "age" : 0,
    "nonce" : "41a18e7f-314f-4aec-8984-62bf6cd355d0",
    "hash" : "n7KJLuC/KOzFZLnGKEs6pOQvCbl4WSF+xplOUd9MrSo=",
    "signature" : "MEUCIEHafCsSXWWtZnDbSpAwFQk4qjW1B7cXjEgdwVF8uKQeAiEAsvzEMKyuNFozAbLC/E8FEviCMLCqo9DPRQe4tVBFwIk="
  },
  "Auditor" : null
}

You can validate the index record by running the following command:

client/bin/scalardl-tablestore validate-ledger --properties client.properties \
--table-name employee --index-key-column-name department --column-value '"sales"'

You should get a result like the following:

{
  "status_code" : "OK",
  "Ledger" : {
    "id" : "idx_employee_department_sales",
    "age" : 0,
    "nonce" : "41a18e7f-314f-4aec-8984-62bf6cd355d0",
    "hash" : "n7KJLuC/KOzFZLnGKEs6pOQvCbl4WSF+xplOUd9MrSo=",
    "signature" : "MEUCIEHafCsSXWWtZnDbSpAwFQk4qjW1B7cXjEgdwVF8uKQeAiEAsvzEMKyuNFozAbLC/E8FEviCMLCqo9DPRQe4tVBFwIk="
  },
  "Auditor" : null
}
注記

ScalarDL TableStore internally assigns a dedicated asset ID to an asset record, which is an object in the primitive data model of ScalarDL. The asset ID consists of a prefix for the asset type and keys for identification; for example, a prefix rec_, table name, primary key column name, and column value are used for the asset ID of a record. You will see such raw asset IDs in the result of validate-ledger.

See also​

To interact with ScalarDL TableStore in your Java applications, see the following:

References​