Sync data through a Hive connection

After using ThoughtSpot DataFlow to establish a connection to a Hive database, you can create automatic data updates, to seamlessly refresh your data.

To sync your data, perform these tasks:

Select table

  1. Immediately after creating a new connection, the connection detail page appears.

    You can fill out this information immediately, or return to it at a later stage, by clicking on the connection name in the list of connections.

  2. On the connection detail page, click Add table sync.

  3. In the Add table sync interface, on the database selector, scroll to select and open the source database.

    Note that you can search for the database by name.

  4. From the database, select the table you plan to sync. You can also search for the table by name.

    Note that if you use multiple tables, you have to repeat these steps for each table.

    Before confirming that you plan to sync a particular table, examine it to ensure that it has the right information: the correct data types, and reasonable sample data.

    For large tables, use the search bar to search column names.

  5. Click Setup sync.

Specify sync schedule

Immediately after creating a new connection, the connection detail page appears.

When you schedule you create a sync, you have the following scheduling options:

After you specify the sync schedule, click Save and sync now. This action saves the schedule, and starts the data sync immediately.

Alternatively, if you click Save, the system saves the schedule, but does not sync the data. This gives you the opportunity to fine-tune the column mapping between external data sources and tables inside ThoughtSpot, such as naming, visibility, data type conversion, conditions, and so on. The first sync starts on the specified schedule.

Hourly sync

Repeat

Select hourly.

Mandatory field.

Starts

Accept the defaults, or set your own starting date and time for syncing.

Note that the timezone is in UTC+00:00.

Mandatory fields.

Run the task every X hour(s)

Select the frequency of the sync.

The valid range is from 1 (default) to 12 hours.

Mandatory field.

Alert notifications

Specify when to notify of the sync status: Failures (default), Success and Failures, or Never.

Daily sync

Repeat

Select daily.

Mandatory field.

Starts

Accept the defaults, or set your own starting date and time for syncing.

Note that the timezone is in UTC+00:00.

Mandatory fields.

Alert notifications

Specify when to notify of the sync status: Failures (default), Success and Failures, or Never.

Weekly sync

Repeat

Select weekly.

Mandatory field.

Starts

Accept the defaults, or set your own starting date and time for syncing.

Note that the timezone is in UTC+00:00.

Mandatory fields.

Days of the week

Select the days of the week when you want to sync.

For example, you may select only Monday, or only Monday through Friday.

Mandatory field.

Alert notifications

Specify when to notify of the sync status: Failures (default), Success and Failures, or Never.

Monthly sync

Repeat

Select monthly.

Mandatory field.

Starts

Accept the defaults, or set your own starting date and time for syncing.

Note that the timezone is in UTC+00:00.

Mandatory fields.

Day of the Month

Select the days of the month when you want to sync.

Mandatory field.

There are two basic approaches:

Cardinal day

Click the Day selector, and choose by date of the month from drop down menu.

For example, select 15 to run sync on 15th of each month.

Ordinal day

Click the The selector, an choose one of First(default), Second, Third, Fourth, or Last. Then choose one of the days of the week, Sunday through Saturday.

Alert notifications

Specify when to notify of the sync status: Failures (default), Success and Failures, or Never.

Does not repeat

Repeat

Select Does not repeat.

Mandatory field.

Alert notifications

Specify when to notify of the sync status: Failures (default), Success and Failures, or Never.

Map tables

To map the external tables to ThoughtSpot’s internal database, follow these steps:

  1. Open the Advanced setup interface by clicking the toggle to open.

  2. Choose the Map tables and columns tab.

  3. Notice that the external database and table already appear, under External data source.

  4. Specify the following information for internal ThoughtSpot storage:

    ThoughtSpot database

    Select an existing ThoughtSpot database from the drop-down menu. If you wish to create a new database in ThoughtSpot, click TQL Editor, enter the CREATE DATABASE my_database; command, and click Execute.
    Mandatory field.

    ThoughtSpot schema

    Select an existing ThoughtSpot schema from the drop-down menu. If you wish to create a new schema in your ThoughtSpot database, click TQL Editor, enter the CREATE SCHEMA my_schema; command, and click Execute.
    Mandatory field.

    New or Existing table

    Mandatory field.

    Create a new table

    Choose this option when you want to load data into a new table.
    The system creates a new table automatically.
    If you want the new table to have some special properties, create the table using the TQL Editor, proceed to Choose existing table, and then select the table you just created. Click TQL Editor, enter the CREATE TABLE my_table ...; command, and click Execute.
    Specify the table name.

    Choose existing table

    Choose this option to load data into a table that already exists inside ThoughtSpot.
    Select the table name from the drop-down menu.

    ThoughtSpot table

    The name of the target table for data sync, inside ThoughtSpot.
    Mandatory field.

Map columns

To map the columns of the external tables to columns in ThoughtSpot’s internal tables, follow these steps:

  1. Open the Advanced setup interface by clicking the toggle to open.

  2. Choose the Map tables and columns tab.

  3. Scroll down, to Map the columns…​ section.

  4. Specify the following information for columns:

    1. Select (or deselect) columns for syncing into ThoughtSpot.
      By default, all columns are selected.

    2. Search for columns by name; this is very useful for very wide tables.

    3. Rename columns in the ThoughtSpot table, for easier search.
      To make this change, click the pencil (edit) icon next to the name of the column.

    4. Change the data type of the column inside the ThoughtSpot table.
      Click the down chevron icon to open the drop-down menu, and select a new data type.
      For example, if you know you have integer data, change the default DOUBLE datatype to INT32.

    5. Set Primary keys of the table by toggling the selector to the 'on' position.
      Note that several columns may be primary keys.

    6. Set the Sharding keys of the table by toggling the selector to the 'on' position.

    7. Specify the Number of shards in the table.

    8. Use Add new formula to transform your data before loading it into ThoughtSpot. Add logic to incorporate a frequently-used or complex formula at data ingestion, so ThoughtSpot doesn’t re-calculate it in every Answer and Pinboard.
      Formulas support all row-level native database functions; we do not support aggregate functions.

      • Click Add new formula.

      • In the Add new formula interface, name the formula and then build it by specifying the functions, columns, and parameters. Note that the support for functions depends on the support by the native database.

      • Click Add to finalize the formula. Notice this adds it to the list of mapped columns.

  5. Save your work by clicking Save.
    Alternatively, click Save and sync now to save your work and sync data at the same time.

Set sync properties

Basic sync properties

To set the sync properties, follow these steps: . Open the Advanced setup interface by clicking the toggle to open. . Choose the Sync properties tab. . Under Set sync properties, specify the following information:

Condition

Add the condition that restricts the import of new rows into the ThoughtSpot table.
Optional field.
For example, to import rows of data from yesterday, enter the expression DATE > DATEADD(TODAY(),-1).
To find the correct functions for the condition, click Expression editor, use it to create a valid expression, and click OK.

Sync mode

Choose the sync mode, either Append or Overwrite.
Mandatory field.

For Append or Overwrite, when the table has a primary key, ThoughtSpot uses UPSERT to update the new rows to the table.
Append

This option adds new rows to the table.

Overwrite

This option removes all existing rows, and then adds new rows to the table.

  1. Save your work by clicking Save.
    Alternatively, click Save and sync now to save your work and sync data at the same time. primary key.

Scripting for sync

To use scripting before or after the sync, follow these steps:

  1. Open the Advanced setup interface by clicking the toggle to open.

  2. Choose the Sync properties tab.

  3. Scroll down to Advanced configuration, and click the toggle to expand.

  4. Specify the following information:

    Pre-script

    Add the script that you want to run before syncing.
    Optional field.
    For example, to drop data that is over 1 year old, enter the following expression:
    DELETE from FACT_TABLE Where DATE < ADD_DAYS(TODAY(),-365).
    To find the correct functions for the script, click Expression editor, use it to create a valid command (or several commands), and click OK.

    Post-script

    Add the script that you want to run after syncing.
    Optional field.
    For example, to drop data that is over 1 year old, enter the following expression:
    DELETE from FACT_TABLE Where DATE < ADD_DAYS(TODAY(),-365).
    To find the correct functions for the script, click Expression editor, use it to create a valid command (or several commands), and click OK.

  5. Save your work by clicking Save.
    Alternatively, click Save and sync now to save your work and sync data at the same time.

Connector-specific sync properties

To modify connection-specific properties for sync, follow these steps:

  1. Open the Advanced setup interface by clicking the toggle to open.

  2. Choose the Sync properties tab.

  3. Scroll down to Sync connector properties, and click the toggle to expand.

  4. Specify the sync properties for Hive:

    See Sync properties for details, defaults, and examples.

    Data extraction mode

    Specify the extraction type.

    Null value

    Specifies the string literal that should indicate the null value in the extracted data. During the data load the column value matching this string will be loaded as null in the target. For Hadoop Extract only.

    Enclosing character

    Specify if the text columns in the source data needs to be enclosed in quotes.

    Escape character

    Specify the escape character if using a text qualifier in the source data.

    Max ignored rows

    Abort the transaction after encountering 'n' ignored rows.

    TS load options

    Specify the parameters passed with the tsload command, in addition to the commands already included by the application.

  5. Save your work by clicking Save.

    Alternatively, click Save and sync now to save your work and sync data at the same time.

Related information

Was this page helpful?
×