Lark Sheets connector

    The connector requires a Lark app in order to access the Lark Sheets API.

    1. Grant View, comment, and export Sheets permission to the app at Permission & Scopes page. And publish a new release for the app at App Release page to make the permissions take effect.

    Create etc/catalog/larksheets.properties to mount the Lark Sheets connector as the larksheets catalog, replacing the properties as appropriate:

    Create app-secret.json as the example below:

    1. {"app-secret": "abcdefghijklmnopqrstuvwxyzabcdef"}

    The following configuration properties are available:

    In the connector, a Lark spreadsheet is mapped to a Presto schema, and each sheet inside the spreadsheet is mapped to a Presto table.

    The token property is taken from the last path segment of the spreadsheet url, using the pattern https://{lark_site_domain}/sheets/{token}?sheet={sheet_id}. For example, shtcnBf5pg4BNSkwV2Ku5xwW9Pf is the token of the spreadsheet at

    After that, all sheets (tabs) of the spreadsheet are automatically mapped to tables in the schema.

    1. SHOW TABLES FROM my_ss

    For each table, its name is taken from the sheet title; its columns are inferred from the head row (the first row of the sheet). Sheets metadata could be listed through the query as:

    In a query statement, a table can be specified in such ways:

    1. -- by sheet title
    3. -- by sheet index
    4. SELECT * FROM my_ss."$0"
    6. SELECT * FROM my_ss."@MT1p4I"

    As for column infer, columns that have blank value in the head row are ignored. A sheet with duplicate column names can not be queried.

    Only the spreadsheet that enables Link Sharing can be queried.

    The schema is visible and manipulative only to its creator by default. To make the schema visible to others, create it as: