Prepare data

This topic describes how to use the data demo package to generate data that simulates data in a real environment and use the data in subsequent data analytics.

1. Download the data demo package.
   You can download one of the following data demo packages based on your operating system. In this topic, the Windows 7 64-bit operating system is used.

   • Data demo package for macOS
   • Data demo package for Linux
   • Data demo package for Windows 7 64-bit
2. Configure the data demo.
   Decompress the data demo package and edit the app.conf file in the conf folder.

   The following content is an example of the app.conf file:

   endpoint = "https://workshop-bj-001.cn-beijing.ots.aliyuncs.com"
   instanceName = "workshop-bj-001"
   accessKeyId = "LTAIF24u7g******"
   accessKeySecret = "CcwFeF3sWTPy0wsKULMw34Px******"
   usercount = "200"
   daysCount = "7"

   You must modify the following parameters:

   • endpoint: the endpoint that is used to access the Tablestore instance. We recommend that you use the public endpoint. You can log on to the Tablestore console. On the Overview page, click the instance in the Instance Name column. On the Instance Management page, view the endpoint from the Instance Access URL section on the Instance Details tab.
   • instanceName: the name of the Tablestore instance. You can obtain the instance name from the Overview page in the Tablestore console.
   • accessKeyId and accessKeySecret: the AccessKey pair that is used to access Alibaba Cloud APIs.
3. Start the data demo to test data.
   1. Start Command Prompt in Windows, go to the directory in which the data demo package is saved, and run the following command to view the usage of commands that are related to the data demo:

      workshop_demo.exe -h

      The command returns the following results:

      ...\workshop_demo>workshop_demo.exe -h 
      prepare will prepare the data
      raw 00005 "2017-12-19" will query user data by user id and time
      new/day_active/month_active/day_pv/month_pv will query metrics data 

      • prepare: prepares test data, creates data tables, and generates weekly user behavior logs based on the number of users specified in the app.conf file.
      • raw and raw ${userid} ${date} ${Number of top logs}: queries a specified number of logs of the specified user on the specified date.
      • new/day_active/month_active/day_pv/month_pv: queries the report data in result tables based on the following report types:
        • new
        • day_active
        • month_active
        • day_pv
        • month_pv
   2. Run the following command to generate test data:

      ...\workshop_demo>workshop_demo.exe prepare

      The following figure shows the command output.

   In this process, the data demo package automatically creates two tables in Tablestore. The following tables describe the columns in the created tables.

   • Raw log table: user_trace_log

     Column	Data type	Description
     md5	STRING	The MD5 value of the user ID. This column is a primary key column.
     uid	STRING	The user ID. This column is a primary key column.
     ts	BIGINT	The timestamp from when the user performed operations. This column is a primary key column.
     ip	STRING	The IP address of the client that sends requests.
     status	BIGINT	The status code returned by the server.
     bytes	BIGINT	The number of bytes sent to the client.
     device	STRING	The model of the terminal used by the user.
     system	STRING	The version of the operating system used by the user, in the format of iosxxx or androidxxx.
     customize_event	STRING	The custom event, which is logon, exit, purchase, register, click, run at the background, switch user, or browse.
     use_time	BIGINT	The duration for which an app is used at one time. This column is available when the custom event is exit, background running, or user switch.
     customize_event_content	STRING	The content of the custom event.

   • Analysis result table: analysis_result

     Column	Data type	Description
     metric	STRING	The report type. Valid values: new, day_active, month_active, day_pv, and month_pv. This column is a primary key column.
     ds	STRING	The data timestamp, in the format of yyyy-mm-dd or yyyy-mm. This column is a primary key column.
     num	BIGINT	The value of the specified metric.

4. Verify data.
   • Query log details of a specified user.
     Run the following command to query a specified number of logs of a specified user on a specified date. In the command, set the date to when the logs are generated.

     raw ${userid} ${date} ${Number of top logs}

     In the preceding command, ${userid} indicates the user ID, ${date} indicates the date when the logs are generated, and ${Number of top logs} indicates the number of logs to query. For example, if a table is created on June 15, 2019, you can run the workshop_demo.exe raw 00010 "2019-06-15" 20 command to query 20 logs for the user whose ID is 00010.
     Note Tablestore is schema-free. Therefore, you do not need to predefine attribute columns. Different events in the customize_event column have different event content. Therefore, the data demo generates both a custom event and its content in a data record.
   • Query data in the analysis result table.
     You can run the workshop_demo.exe day_active command to query the number of daily active users.