Project

pursuit

0.0
No release in over a year
Advanced key-based searching for ActiveRecord objects.
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
 Dependencies

Runtime

>= 5.2.0, <= 8.0.0
>= 5.2.0, <= 8.0.0
~> 2.0
 Project Readme

Pursuit

Search your ActiveRecord objects with ease!

Installation

You can install Pursuit using the following command:

$ gem install pursuit

Or, by adding the following to your Gemfile:

gem 'pursuit'

Usage

Pursuit comes with three different strategies for interpreting queries:

  • Simple
  • Term
  • Predicate

Simple Search

Simple takes the entire query and generates a SQL LIKE (or ILIKE for PostgreSQL) statement for each attribute added to the search instance. Here's an example of how you might use simple to search a hypothetical Product record:

search = Pursuit::SimpleSearch.new(default_table: Product.arel_table)
search.search_attribute(:title)
search.search_attribute(:subtitle)
search.apply('Green Shirt', Product.all)

Which results in the following SQL query:

SELECT
  "products".*
FROM
  "products"
WHERE
  "products"."title" LIKE '%Green Shirt%'
  OR "products"."subtitle" LIKE '%Green Shirt%'

The initializer method also accepts a block, which is evaluated within the instance's context. This can make it cleaner when declaring the searchable attributes:

search = Pursuit::SimpleSearch.new(default_table: Product.arel_table) do
  search_attribute :title
  search_attribute :subtitle
end

search.apply('Green Shirt', Product.all)

You can also pass custom Arel::Attribute::Attribute objects, which are especially useful when using joins:

search = Pursuit::SimpleSearch.new(default_table: Product.arel_table) do
  search_attribute :title
  search_attribute ProductVariation.arel_table[:title]
end

search.apply('Green Shirt', Product.left_outer_joins(:variations).group(:id))

Which results in the following SQL query:

SELECT
  "products".*
FROM
  "products"
  LEFT OUTER JOIN "product_variations" ON "product_variations"."product_id" = "products"."id"
WHERE
  "products"."title" LIKE '%Green Shirt%'
  OR "product_variations"."title" LIKE '%Green Shirt%'
GROUP BY
  "products"."id"

Term Search

Term searches break a query into individual terms on spaces, while providing double and single quoted strings as a means to include spaces. Here's an example of using term searches on the same Product record from earlier:

search = Pursuit::TermSearch.new(default_table: Product.arel_table) do
  search_attribute :title
  search_attribute :subtitle
end

search.apply('Green "Luxury Shirt"', Product.all)

Which results in a SQL query similar to the following:

SELECT
  "products".*
FROM
  "products"
WHERE
  (
    "products"."title" LIKE '%Green%'
    OR "products"."subtitle" LIKE '%Green%'
  ) AND (
    "products"."title" LIKE '%Luxury Shirt%'
    OR "products"."subtitle" LIKE '%Luxury Shirt%'
  )

Predicate Search

Predicate searches use a parser (implemented with the parslet gem) to provide a minimal query language. This syntax is similar to the WHERE and HAVING clauses in SQL, but uses only symbols for operators and joins.

Attributes can only be used in predicate searches when they have been added to the list of permitted attributes. You can also rename attributes, and add attributes for joined records.

Here's a more complex example of using predicate-based searches with joins on the Product record from earlier:

search = Pursuit::PredicateSearch.new(default_table: Product.arel_table) do
  # Product Attributes
  permit_attribute :title

  # Product Category Attributes
  permit_attribute :category_name, ProductCategory.arel_table[:name]

  # Product Variation Attributes
  permit_attribute :variation_title, ProductVariation.arel_table[:title]
  permit_attribute :variation_currency, ProductVariation.arel_table[:currency]
  permit_attribute :variation_amount, ProductVariation.arel_table[:amount]
end

search.apply(
  'title = "Luxury Shirt" & (variation_amount = 0 | variation_amount > 1000)',
  Product.left_outer_join(:category, :variations).group(:id)
)

This translates to "a product whose title is 'Luxury Shirt' and has at least one variation with either an amount of 0, or an amount greater than 1000", which could be expressed in SQL as:

SELECT
  "products".*
FROM
  "products"
  LEFT OUTER JOIN "product_categories" ON "product_categories"."id" = "products"."category_id"
  LEFT OUTER JOIN "product_variations" ON "product_variations"."product_id" = "products"."id"
WHERE
  "products"."title" = 'Luxury Shirt'
  AND (
    "product_variations"."amount" = 0
    OR "product_variations"."amount" > 1000
  )
GROUP BY
  "products"."id"

You can use any of the following operators in comparisons:

  • = checks if the attribute is equal to the value.
  • != checks if the attributes is not equal to the value.
  • > checks if the attribute is greater than the value.
  • < checks if the attribute is less than the value.
  • >= checks if the attribute is greater than or equal to the value.
  • <= checks if the attribute is less than or equal to the value.
  • ~ checks if the attribute matches the value (using LIKE or ILIKE).
  • !~ checks if the attribute does not match the value (using LIKE or ILIKE).

Predicate searches also support "aggregate modifiers" which enable the use of aggregate functions, however this feature must be explicitly enabled and requires you to use a GROUP BY clause:

search = Pursuit::PredicateSearch.new(default_table: Product.arel_table, permit_aggregate_modifiers: true) do
  # Product Attributes
  permit_attribute :title

  # Product Category Attributes
  permit_attribute :category, ProductCategory.arel_table[:id]
  permit_attribute :category_name, ProductCategory.arel_table[:name]

  # Product Variation Attributes
  permit_attribute :variation, ProductVariation.arel_table[:id]
  permit_attribute :variation_title, ProductVariation.arel_table[:title]
  permit_attribute :variation_currency, ProductVariation.arel_table[:currency]
  permit_attribute :variation_amount, ProductVariation.arel_table[:amount]
end

search.apply(
  'title = "Luxury Shirt" & #variation > 5',
  Product.left_outer_join(:category, :variations).group(:id)
)

And the resulting SQL from this query:

SELECT
  "products".*
FROM
  "products"
  LEFT OUTER JOIN "product_categories" ON "product_categories"."id" = "products"."category_id"
  LEFT OUTER JOIN "product_variations" ON "product_variations"."product_id" = "products"."id"
WHERE
  "products"."title" = 'Luxury Shirt'
GROUP BY
  "products"."id"
HAVING
  COUNT("product_variations"."id") > 5

There's no distinction between the WHERE and HAVING clause in the predicate syntax, as it's intended to be easy to use, but this does come with a caveat.

The query must have all aggregate-modified comparisons before or after non-aggregate-modified comparisons, you can't mix both.

For example, this query would result in a parsing error: title ~ Shirt & #variation > 5 & category_name = Shirts

You can preceed any attribute with one of these aggregate modifier symbols:

  • # uses the COUNT aggregate function
  • + uses the MAX aggregate function
  • - uses the MIN aggregate function
  • * uses the SUM aggregate function
  • ~ uses the AVG aggregate function

Development

After checking out the repo, run bundle exec rake spec to run the tests.

To install this gem onto your machine, run bundle exec rake install.