RailsCursorPagination

Gem Version License Tests

This library allows to paginate through an ActiveRecord relation using cursor pagination. It also supports ordering by any column on the relation in either ascending or descending order.

Cursor pagination allows to paginate results and gracefully deal with deletions / additions on previous pages. Where a regular limit / offset pagination would jump in results if a record on a previous page gets deleted or added while requesting the next page, cursor pagination just returns the records following the one identified in the request.

To learn more about cursor pagination, check out the "How does it work?" section below.

Installation

Add this line to your application's Gemfile:

gem 'rails_cursor_pagination'

And then execute:

$ bundle install

Or install it yourself as:

$ gem install rails_cursor_pagination

Usage

Using it is very straight forward by just interfacing with the RailsCursorPagination::Paginator class.

Let's assume we have an ActiveRecord model called Post of which we want to fetch some data and then paginate through it. Therefore, we first apply our scopes, where clauses or other functionality as usual:

posts = Post.where(author: 'Jane')

And then we pass these posts to our paginator to fetch the first response page:

RailsCursorPagination::Paginator.new(posts).fetch(with_total: true)

This will return a data structure similar to the following:

{
  total: 42,
  page_info: {
    has_previous_page: false,
    has_next_page: true,
    start_cursor: "MQ==",
    end_cursor: "MTA="
  },
  page: [
    { cursor: "MQ==", data: #<Post:0x00007fd7071b2ea8 @id=1> },
    { cursor: "Mg==", data: #<Post:0x00007fd7071bb738 @id=2> },
    ...,
    { cursor: "MTA=", data: #<Post:0x00007fd707238260 @id=10> }
  ]
}

Note that any ordering of the relation at this stage will be ignored by the gem. Take a look at the next section "Ordering" to see how you can have an order different than ascending IDs. Read the "The passed relation" to learn more about the relation that can be passed to the paginator.

As you saw in the request, with_total is an option. If omitted, or set to false, the resulting hash will lack the :total key, but this will also cause one DB query less. It is therefore recommended to only pass with_total: true when requested by the user. So in the next examples we will also leave it away.

To then get the next result page, you simply need to pass the last cursor of the returned page item via:

RailsCursorPagination::Paginator
  .new(posts, after: 'MTA=')
  .fetch

This will then fetch the next result page. You can also just as easily paginate to previous pages by using before instead of after and using the first cursor of the current page.

RailsCursorPagination::Paginator
  .new(posts, before: "MTE=")
  .fetch

By default, this will always return up to 10 results. But you can also specify how many records should be returned. You can pass first: 2 to get the very first 2 records of the relation:

RailsCursorPagination::Paginator
  .new(posts, first: 2)
  .fetch

Then, you can also combine first with after to get the first X records after a given one:

RailsCursorPagination::Paginator
  .new(posts, first: 2, after: 'MTA=')
  .fetch

Or you can combine before with last to get the last X records before a given one:

RailsCursorPagination::Paginator
  .new(posts, last: 2, before: 'MTA=')
  .fetch

Alternatively, you can use the limit column with either after or before. This will behave like either first or last respectively and fetch X records.

RailsCursorPagination::Paginator
  .new(posts, limit: 2, after: 'MTA=')
  .fetch

RailsCursorPagination::Paginator
  .new(posts, limit: 2, before: 'MTA=')
  .fetch

Ordering

As said, this gem ignores any previous ordering added to the passed relation. But you can still paginate through relations with an order different than by ascending IDs.

The order parameter

The first option you can pass is the order parameter. It allows you to order the relation in reverse, descending.

RailsCursorPagination::Paginator
  .new(posts, order: :desc)
  .fetch

The default is :asc, therefore this doesn't need to be passed.

The order_by parameter

However, you can also specify a different column to order the results by. Therefore, the order_by parameter needs to be passed.

RailsCursorPagination::Paginator
  .new(posts, order_by: :author)
  .fetch

This will now order the records ascending by the :author field. You can also combine the two:

RailsCursorPagination::Paginator
  .new(posts, order_by: :author, order: :desc)
  .fetch

This will then sort the results by the author field in a descending order. Of course, this can both be combined with first, last, before, and after.

Important: If your app regularly orders by another column, you might want to add a database index for this. Say that your order column is author then you'll want to add a compound index on (author, id). If your table is called posts you can use a query like this in MySQL or Postgres:

CREATE INDEX index_posts_on_author_and_id ON posts (author, id);

Or you can just do it via an ActiveRecord::Migration:

class AddAuthorAndIdIndexToPosts < ActiveRecord::Migration
  def change
    add_index :posts, %i[author id]
  end
end

Please take a look at the "How does it work?" to find out more why this is necessary.

Order by more complex logic

Sometimes you might not only want to oder by a column ascending or descending, but need more complex logic. Imagine you would also store the post's category on the posts table (as a plain string for simplicity's sake). And the category could be pinned, announcement, or general. Then you might want to show all pinned posts first, followed by the announcement ones and lastly show the general posts.

In MySQL you could e.g. use a FIELD(category, 'pinned', 'announcement', 'general') query in the ORDER BY clause to achieve this. However, you cannot add an index to such a statement. And therefore, the performance of this is – especially when using cursor pagination where we not only have an ORDER BY clause but also need it twice in the WHERE clauses – is pretty dismal.

For this reason, the gem currently only supports ordering by natural columns of the relation. You cannot pass a generic SQL query to the order_by parameter.

Implementing support for arbitrary SQL queries would also be fairly complex to handle in this gem. We would have to ensure that SQL injection attacks aren't possible by passing malicious code to the oder_by parameter. And we would need to return the data produced by the statement so that it can be encoded in the cursor. This is, for now, out of scope of the functionality of this gem.

What is recommended if you do need to order by more complex logic is to have a separate column that you only use for ordering. You can use ActiveRecord hooks to automatically update this column whenever you change your data. Or, for example in MySQL, you can also use a generated column that is automatically being updated by the database based on some stored logic.

Configuration options

You can also change the default page size to a value that better fits the needs of your application. So if a user doesn't request a given first or last value, the default amount of records is being returned.

To change the default, simply add an initializer to your app that does the following:

RailsCursorPagination.configure do |config|
  config.default_page_size = 50
end

This would set the default page size to 50.

You can also select a global max_page_size to prevent a client from requesting too large a page.

RailsCursorPagination.configure do |config|
  config.max_page_size = 100
end

The passed relation

The relation passed to the RailsCursorPagination::Paginator needs to be an instance of an ActiveRecord::Relation. So if you e.g. have a Post model that inherits from ActiveRecord::Base, you can initialize your paginator like this:

RailsCursorPagination::Paginator
  .new(Post.all)

This would then paginate over all post records in your database.

Limiting the paginated records

As shown above, you can also apply .where clauses to filter your records before pagination:

RailsCursorPagination::Paginator
  .new(Post.where(author: 'Jane'))

This would only paginate over Jane's records.

Limiting the queried fields

You can also use .select to limit the fields that are requested from the database. If, for example, your post contains a very big content field that you don't want to return on the paginated index endpoint, you can select to only get the fields relevant to you:

RailsCursorPagination::Paginator
  .new(Post.select(:id, :author))

One important thing to note is that the ID of the record will always be returned, whether you selected it or not. This is due to how the cursor is generated. It requires the record's ID to always be present. Therefore, even if it is not selected by you, it will be added to the query.

The same goes for any field that is specified via order_by:, this field is also required for building the cursor and will therefore automatically be requested from the database.

How does it work?

The cursor that we use for the before or after query encodes a value that uniquely identifies a given row for the requested order. Then, based on this cursor, you can request the "n first records after the cursor" (forward-pagination) or the "n last records before the cursor" (backward-pagination).

As an example, assume we have a table called "posts" with this data:

id author
1 Jane
2 John
3 John
4 Jane
5 Jane
6 John
7 John

Now if we make a basic request without any first/after, last/before, custom order or order_by column, this will just request the first page of this relation.

RailsCursorPagination::Paginator
  .new(relation)
  .fetch

Assume that our default page size here is 2 and we would get a query like this:

SELECT *
FROM "posts"
ORDER BY "posts"."id" ASC
LIMIT 2

This will return the first page of results, containing post #1 and #2. Since no custom order is defined, each item in the returned collection will have a cursor that only encodes the record's ID.

If we want to now request the next page, we can pass in the cursor of record #2 which would be "Mg==". So now we can request the next page by calling:

RailsCursorPagination::Paginator
  .new(relation, first: 2, after: "Mg==")
  .fetch

And this will decode the given cursor and issue a query like:

SELECT *
FROM "posts"
WHERE "posts"."id" > 2
ORDER BY "posts"."id" ASC
LIMIT 2

Which would return posts #3 and #4. If we now want to paginate back, we can request the posts that came before the first post, whose cursor would be "Mw==":

RailsCursorPagination::Paginator
  .new(relation, last: 2, before: "Mw==")
  .fetch

Since we now paginate backward, the resulting SQL query needs to be flipped around to get the last two records that have an ID smaller than the given one:

SELECT *
FROM "posts"
WHERE "posts"."id" < 3
ORDER BY "posts"."id" DESC
LIMIT 2

This would return posts #2 and #1. Since we still requested them in ascending order, the result will be reversed before it is returned.

Now, in case that the user wants to order by a column different than the ID, we require this information in our cursor. Therefore, when requesting the first page like this:

RailsCursorPagination::Paginator
  .new(relation, order_by: :author)
  .fetch

This will issue the following SQL query:

SELECT *
FROM "posts"
ORDER BY "posts"."author" ASC, "posts"."id" ASC
LIMIT 2

As you can see, this will now order by the author first, and if two records have the same author it will order them by ID. Ordering only the author is not enough since we cannot know if the custom column only has unique values. And we need to guarantee the correct order of ambiguous records independent of the direction of ordering. This unique order is the basis of being able to paginate forward and backward repeatedly and getting the correct records.

The query will then return records #1 and #4. But the cursor for these records will also be different to the previous query where we ordered by ID only. It is important that the cursor encodes all the data we need to uniquely identify a row and filter based upon it. Therefore, we need to encode the same information as we used for the ordering in our SQL query. Hence, the cursor for pagination with a custom column contains a tuple of data, the first record being the custom order column followed by the record's ID.

Therefore, the cursor of record #4 will encode ['Jane', 4], which yields this cursor: "WyJKYW5lIiw0XQ==".

If we now want to request the next page via:

RailsCursorPagination::Paginator
  .new(relation, order_by: :author, first: 2, after: "WyJKYW5lIiw0XQ==")
  .fetch

We get this SQL query:

SELECT *
FROM "posts"
WHERE (author > 'Jane' OR (author = 'Jane') AND ("posts"."id" > 4))
ORDER BY "posts"."author" ASC, "posts"."id" ASC
LIMIT 2

You can see how the cursor is being used by the WHERE clause to uniquely identify the row and properly filter based on this. We only want to get records that either have a name that is alphabetically after "Jane" or another "Jane" record with an ID that is higher than 4. We will get the records #5 and #2 as response.

When using a custom order_by, this affects both filtering as well as ordering. Therefore, it is recommended to add an index for columns that are frequently used for ordering. In our test case we would want to add a compound index for the (author, id) column combination. Databases like MySQL and Postgres are able to then use the leftmost part of the index, in our case author, by its own or can use it combined with the id index.

Development

Make sure you have MySQL installed on your machine and create a database with the name rails_cursor_pagination_testing.

After checking out the repo, run bin/setup to install dependencies. Then, run rake spec to run the tests. You can also run bin/console for an interactive prompt that will allow you to experiment.

To install this gem onto your local machine, run bundle exec rake install. To release a new version, update the version number in version.rb, and then run bundle exec rake release, which will create a git tag for the version, push git commits and the created tag, and push the .gem file to rubygems.org.

Supported environments

This gem should run in any project that uses:

  • Ruby
  • ActiveRecord
  • Postgres or MySQL

We aim to support all versions that are still actively maintained and extend support until one year past the version's EOL. While we think it's important to stay up-to-date with versions and update as soon as an EOL is reached, we know that this is not always immediately possible. This way, we hope to strike a balance between being usable by most projects without forcing them to upgrade, but also keeping the supported version combinations manageable.

This project is tested against different permutations of Ruby versions and DB versions, both Postgres and MySQL. Please check the test automation file under ./.github/workflows/test.yml to see all officially supported combinations.

Contributing

Bug reports and pull requests are welcome on GitHub at https://github.com/xing/rails_cursor_pagination. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the code of conduct.

If you open a pull request, please make sure to also document your changes in the CHANGELOG.md. This way, your change can be properly announced in the next release.

License

The gem is available as open source under the terms of the MIT License.

Code of Conduct

Everyone interacting in the RailsCursorPagination project's codebases, issue trackers, chat rooms and mailing lists is expected to follow the code of conduct.