Setup Server-Side Pagination on Table
This page shows you how to set up server-side pagination on a Table widget, which allows you to manage and display large datasets within your application. It involves fetching and displaying only a portion of data from the server at a time, enhancing performance.
If you are using the one-click binding feature to connect data, Appsmith automatically generates server-side pagination queries for you. However, if you prefer to manually configure the server-side setup, you can do so by following the instructions in this guide.
Configure query
Most databases and APIs support server-side pagination, although the methods of implementation can vary.
Appsmith can handle query responses of up to 5 MB. Server-side pagination can be implemented using Offset-based pagination or Cursor-based pagination.
- Offset-based pagination
- Cursor-based-pagination
Offset-based pagination works by using the page number and size to calculate the offset of records to fetch from a database or API.
Create a query to fetch data from the database/API using pageSize
, pageNo
, and pageOffset
reference properties to implement pagination.
Example:
- For PostgreSQL, you can configure the query as follows:
SELECT * FROM users LIMIT {{ Table1.pageSize }} OFFSET {{ Table1.pageOffset }};
This SQL query retrieves data from the users
table with pagination based on the specified pagesize and offset values.
- For the REST API, the page number can be passed as a query parameter to retrieve the corresponding subset of data, as shown in the URL:
https://mock-api.appsmith.com/users?page={{Table1.pageNo}}
You can refer to the datasource reference for specific instructions on setting up pagination for your selected datasource.
Cursor-based pagination is a method that uses unique identifiers (cursors) to navigate and fetch data in large datasets. The cursor itself serves as a reference point, and you query for a specific number of records that come after or before that cursor.
Create a query to fetch data from the database/API using previousPageVisited
and nextPageVisited
reference properties to implement pagination.
Example:
- For PostgreSQL, you can configure the query as follows:
SELECT * FROM users {{Table1.nextPageVisited ? "WHERE id > "+ " "+ Table1.tableData[Table1.tableData.length-1]["id"] : Table1.previousPageVisited ? "WHERE id <"+ " "+ Table1.tableData[0]["id"] : "" }} ORDER BY id LIMIT {{Table1.pageSize}} ;
This SQL query selects all columns from the users
table and applies cursor-based pagination to limit the number of results returned. The WHERE
clause is dynamically generated based on whether the user has already visited the next or previous page, and orders the results by ID
.
Please ensure that you turn off prepared statements in the query editor for this configuration.
- For the REST API, you can make use of the URL's query parameter to retrieve data under specific conditions using
next
andprevious
:
https://api.site.com/users/?pageDirection={{Table1.nextPageVisited ? "next" : Table1.previousPageVisited? "previous":"default"}}
//The "pageDirection" serves as a query parameter within the API
You can refer to the datasource reference for specific instructions on setting up pagination for your selected datasource.
Configure Table widget
Follow these steps to configure the Table widget to display fetched data, and implement server-side pagination:
- Bind the query data into the Table data property of the Table widget.
Example:
{{fetchData.data}}
-
Enable the Server-side pagination property in the table.
-
Set the Table widget's onPageChange event to run the pagination query.
With this setup, users can paginate through data, ensuring an efficient browsing experience.
Configure total records
To provide the user with information about the number of records in the table, you can configure the Total records property to be displayed in the table header.
PostgreSQL Example:
SELECT COUNT(*) from users where name ilike '%{{Table1.searchText}}%';
This SQL query uses the ilike
condition on the name
column, pinpointing relevant data rather than performing a blanket count of all records.
To display the count, add the following code to the Total records property:
{{fetch_users_count.data[0].count}}