Home
Arenadata Hadoop
How to
Zeppelin
Work with interpreters
Add a custom interpreter to a group

Add a custom interpreter to a group

Daria Barysheva

Contents

Preparation
Method 1. Using prefixes
Method 2. Adding an interpreter from scratch

The list of built-in Zeppelin interpreters is available in the web UI after selecting the Interpreter item in the account menu. You can also add your own interpreters within the existing groups. This section shows two methods of creating a new interpreter, using the example of adding Phoenix interpreter to the jdbc group.

NOTE

You can find more information about interpreters in the Zeppelin documentation. Zeppelin also allows to create your own interpreters, which are not based on the existing groups, and to install third-party interpreters.

Preparation

As described in the Connect to Phoenix section, you can access Phoenix databases via JDBC. In the examples throughout this section, we use the Phoenix Thin Driver for this purpose. Before using any of two methods, you should perform preparation steps:

Get the corresponding version of the JAR file phoenix-<phoenix_version>-HBase-<hbase_version>-thin-client.jar. For example, for HBase 2.0 and Phoenix 5.0.0, this file is called phoenix-5.0.0-HBase-2.0-thin-client.jar. You can copy it from any host of your HBase cluster (from the /usr/lib/phoenix directory). The other way is to download it from Phoenix repository.
Copy the prepared JAR file from the source folder to the interpreter directory /usr/lib/zeppelin/interpreter/jdbc/:
```
$ sudo cp /usr/lib/phoenix/phoenix-5.0.0-HBase-2.0-thin-client.jar /usr/lib/zeppelin/interpreter/jdbc/
```

Verify that the file is copied successfully using the command:

$ ls -la /usr/lib/zeppelin/interpreter/jdbc/

The output looks as follows:

total 35412
drwxr-xr-x.  3 root root     4096 Dec  7 17:46 .
drwxr-xr-x. 28 root root     4096 Dec  7 08:36 ..
-rw-r--r--.  1 root root     2314 Apr 15  2021 ansi.sql.keywords
-rw-r--r--.  1 root root   167962 Apr 15  2021 commons-dbcp2-2.0.1.jar
-rw-r--r--.  1 root root    60686 Apr 15  2021 commons-logging-1.1.1.jar
-rw-r--r--.  1 root root   109568 Apr 15  2021 commons-pool2-2.3.jar
-rw-r--r--.  1 root root     4800 Apr 15  2021 interpreter-setting.json
-rw-r--r--.  1 root root   489884 Apr 15  2021 log4j-1.2.17.jar
drwxr-xr-x.  2 root root       55 Dec  6 09:10 META-INF
-rw-r--r--.  1 root root 34651233 Dec  7 17:46 phoenix-5.0.0-HBase-2.0-thin-client.jar
-rw-r--r--.  1 root root   648487 Apr 15  2021 postgresql-9.4-1201-jdbc41.jar
-rw-r--r--.  1 root root     6573 Apr 15  2021 postgresql-native-driver-sql.keywords
-rw-r--r--.  1 root root    32119 Apr 15  2021 slf4j-api-1.7.10.jar
-rw-r--r--.  1 root root     8866 Apr 15  2021 slf4j-log4j12-1.7.10.jar
-rw-r--r--.  1 root root    46253 Apr 15  2021 zeppelin-jdbc-0.8.1.jar

TIP

Starting with Zeppelin 0.9.0, you can skip the preparation steps and use the Apache Phoenix guide to create a new Phoenix interpreter exclusively via the web UI. In older versions, errors may occur while accessing the Maven repository for resolving dependencies. That is why we recommend placing the necessary jars manually.

Method 1. Using prefixes

This method is based on adding new properties with a specific prefix to the existing interpreter. Thus, when working with notes, you can use the source interpreter not only with the default settings, but with the settings that have the special prefix. To call such an interpreter from the note, you should append the prefix to the main interpreter name, for example: %jdbc(phoenix).

Select the Interpreter item in the account menu placed in the top right corner of the Zeppelin home page.

Switching to the list of interpreters

Switching to the list of interpreters
Using the search bar, find the jdbc interpreter, and click edit.

Editing the JDBC interpreter

Editing the JDBC interpreter

Add new properties to the interpreter settings list with the phoenix prefix.

Name Value Description

phoenix.driver

org.apache.phoenix.queryserver.client.Driver

The driver used to connect to Phoenix

phoenix.url

jdbc:phoenix:thin:url=http://<IP>:8765;serialization=PROTOBUF

The url used for connecting to Phoenix. <IP> means the IP-address of your Phoenix Query Server

phoenix.user

—

The login used for connecting to Phoenix

phoenix.password

—

The password used for connecting to Phoenix

To add a property, you should enter its name, value, and then click + to add the next field.

Adding new fields

Adding new fields

After adding all the necessary fields, click Save.

Saving the JDBC interpreter

Saving the JDBC interpreter
Confirm the operation in the opened dialog window. Your interpreter is ready.

Confirming saving the interpreter

Confirming saving the interpreter

Method 2. Adding an interpreter from scratch

This method assumes adding a new interpreter based on the existing one. For this, select the current interpreter as a group (in this example — jdbc) and change its default properties to your own. All the libraries available for the source interpreter are also available for the newly created one. To call such an interpreter from a note, you should use the new interpreter name: %phoenix.

Select the Interpreter item in the account menu placed in the top right corner of the Zeppelin home page.

Switching to the list of interpreters

Switching to the list of interpreters
Click Create.

Creating a new interpreter

Creating a new interpreter
Enter phoenix in the Interpreter Name field and select jdbc in the Interpreter group.

Filling a name and a group

Filling a name and a group
Clear the values of the following fields: default.driver, default.url, default.user, default.password. Fill them with the new values according to the table from the previous section.
Click Save. The new interpreter is ready.

Saving the Phoenix interpreter

Saving the Phoenix interpreter

NOTE

The example of creating a note using the described interpreters can be found in the article Create a note using Phoenix.

Found a mistake? Seleсt text and press Ctrl+Enter to report it